Heteroaromatic Inhibitors of the Astacin Proteinases Meprin α, Meprin β and Ovastacin Discovered by a Scaffold-Hopping Approach

Article abstract of DOI:10.1002/cmdc.202000822

Astacin metalloproteinases, in particular meprins α and β, as well as ovastacin, are emerging drug targets. Drug-discovery efforts have led to the development of the first potent and selective inhibitors in the last few years. However, the most recent compounds are based on a highly flexible tertiary amine scaffold that could cause metabolic liabilities or decreased potency due to the entropic penalty upon binding to the target. Thus, the aim of this study was to discover novel conformationally constrained scaffolds as starting points for further inhibitor optimization. Shifting from flexible tertiary amines to rigid heteroaromatic cores resulted in a boost in inhibitory activity. Moreover, some compounds already exhibited higher activity against individual astacin proteinases compared to recently reported inhibitors and also a favorable off-target selectivity profile, thus qualifying them as very suitable chemical probes for target validation.

Full text of DOI:10.1002/cmdc.202000822

Full Papers
doi.org/10.1002/cmdc.202000822
ChemMedChem
Very Important Paper
1
2
Heteroaromatic Inhibitors of the Astacin Proteinases
Meprin α, Meprin β and Ovastacin Discovered by a
Scaffold-Hopping Approach
3
4
5
6
7
8
9
Kathrin Tan,[a] Christian Jäger,[a, b] Hagen Körschgen,[c] Stefanie Geissler,[a]
Dagmar Schlenzig,[a] Mirko Buchholz,[a] Walter Stöcker,[c] and Daniel Ramsbeck*[a]
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
Astacin metalloproteinases, in particular meprins α and β, as
well as ovastacin, are emerging drug targets. Drug-discovery
efforts have led to the development of the first potent and
selective inhibitors in the last few years. However, the most
recent compounds are based on a highly flexible tertiary amine
scaffold that could cause metabolic liabilities or decreased
potency due to the entropic penalty upon binding to the target.
Thus, the aim of this study was to discover novel conformation-
ally constrained scaffolds as starting points for further inhibitor
optimization. Shifting from flexible tertiary amines to rigid
heteroaromatic cores resulted in a boost in inhibitory activity.
Moreover, some compounds already exhibited higher activity
against individual astacin proteinases compared to recently
reported inhibitors and also a favorable off-target selectivity
profile, thus qualifying them as very suitable chemical probes
for target validation.
Introduction
the proteolytic processing of several cytokines and was also
discussed as a potential drug target for inflammatory bowel
disease.[6] Furthermore, meprin β is able to cleave the amyloid
precursor protein, resulting in the release of neurotoxic amyloid
β peptides. Hence, it was revealed as an alternative β-secretase
in Alzheimer’s disease.[7] Ovastacin was found to be involved in
the regulation of female fertility.[8] Small amounts of ovastacin
seeping out of the oocytes lead to premature hardening of the
zona pellucida, which is physiologically controlled by the
endogenous inhibitor fetuin-B.[9] Fetuin deficiency causes zona
pellucida hardening, thus preventing sperm cells from entering
the oocyte and consequently leading to infertility. Hence,
inhibition of ovastacin by small molecules in vivo could be a
novel approach addressing female infertility or a novel supple-
ment to facilitate in vitro fertilization. Moreover, ovastacin was
also found to be expressed by several types of tumors as
tumor-oocyte-neoantigen.[10] A potential involvement in tumor
migration and invasion is currently under investigation and
might render ovastacin as anticancer target in the future.
Recently, we reported the development of selective meprin
β inhibitors starting from the broad-spectrum metalloproteinase
inhibitor NNGH.[11] Further structural modification and simplifi-
cation of the sulfonamide backbone by SO2!CH2 switching led
to a tertiary amine hydroxamate scaffold. This modification
significantly improved the inhibitor selectivity against off-target
metalloproteinases, that is, MMPs and ADAMs. The derivatiza-
tion of this scaffold revealed potent and selective inhibitors of
either meprin α or β.[12,13] More recently, selected compounds of
this chemotype were also found to be potent inhibitors of
ovastacin (Figure 1), rendering them as pan-selective astacin
inhibitors.[14] Nevertheless, the tertiary amine motif exhibits a
high conformational freedom, potentially leading to entropic
penalty upon binding to the active site of the enzyme.
Furthermore, the benzylic CH2-N bonds represent hotspots that
might lead to reduced metabolic stability due to oxidative
Proteinases of the astacin family as potential drug targets
recently moved into the focus of drug development. In humans,
this family comprises the meprins α and β, ovastacin and BMP-
1.[1] The last, together with its variants, that is, tolloids, have
been in the focus of drug development for the last two decades
as targets for the development of antifibrotic drugs.[2] In
contrast, the remaining human astacins only emerged as drug
targets recently. Due to their procollagenase activity, meprin α
and β are involved in the assembly of collagen fibrils like BMP-1
and are also potentially involved in fibrotic disorders, such as
keloids or lung fibrosis.[3] Furthermore, they have been linked to
cancer[4] and kidney injury[5], as they are able to cleave several
components of the extracellular matrix and thus contribute to
ECM remodeling. Hence, meprins could promote the migration
and invasion of cancer cells into healthy tissue or contribute to
the damage of renal tissue. In particular meprin β is involved in
[a] K. Tan, C. Jäger, Dr. S. Geissler, Dr. D. Schlenzig, Dr. M. Buchholz,
Dr. D. Ramsbeck
Department of Drug Design and Target Validation MWT
Fraunhofer Institute for Cell Therapy and Immunology IZI
Biocenter, Weinbergweg 22, 06120 Halle (Saale) (Germany)
E-mail: daniel.ramsbeck@izi.fraunhofer.de
[b] C. Jäger
present address: Vivoryon Therapeutics N.V.,
Weinbergweg 22, 06120 Halle (Saale) (Germany)
[c] Dr. H. Körschgen, Prof. Dr. W. Stöcker
Institute of Molecular Physiology, Cell and Matrix Biology
Johannes Gutenberg-University Mainz
Johann-Joachim-Becher-Weg 7, 55128 Mainz (Germany)
Supporting information for this article is available on the WWW under
https://doi.org/10.1002/cmdc.202000822
© 2020 The Authors. ChemMedChem published by Wiley-VCH GmbH. This is
an open access article under the terms of the Creative Commons Attribution
Non-Commercial NoDerivs License, which permits use and distribution in
any medium, provided the original work is properly cited, the use is non-
commercial and no modifications or adaptations are made.
ChemMedChem 2020, 15, 1–14
1
© 2020 The Authors. ChemMedChem published by Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cmdc.202000822
ChemMedChem
rather than a simple aliphatic cyclization of the amine and a
locked conformation (Figure 2).
1
2
We formally assumed three possible heteroaromatic cycliza-
tions for the rigidification of the tertiary amine, a “1,2,3” and
“1,2,4” cyclization yielding scaffolds 2 and 3, or a ring fusion
yielding scaffold 4, respectively. As either glycine and β-alanine
derivatives were found to be suitable spacers between the
tertiary amine and the hydroxamic acid and could affect the
inhibitor activity against meprin α or β and ovastacin,[12–14] both
spacers between hydroxamic acid and the heteroaromatic cores
were investigated, as well. Prototypic, synthetically tractable
inhibitors of these three scaffolds are represented by 2,5-
diphenyl-1-yl-pyrroles 2a, b, 3,5-diphenyl-1-yl-pyrazoles 3a, b
and 2-phenyl-1-yl-indoles 4a, b.
To investigate if these compounds could potentially be able
to mimic the bioactive conformation of the tertiary amine
scaffold, we utilized a flexible alignment approach. Recently,
the structure of meprin β in complex with compound 1a was
elucidated (internal communication). Thus, we intended to use
the conformation of the inhibitor bound to the active site of
meprin β as fixed template for the flexible alignment of the
scaffold prototypes.
However, the crystal structure of meprin β in complex with
compound 1a and further molecular dynamics simulations
revealed a high flexibility of one of the two benzylic residues.
While the moiety addressing the S1’ pocket and the hydroxamic
acid were quite fixed in these complexes, the benzyl residue
targeting the S1 pocket, that is, R184, exhibited a high degree of
flexibility. This is also true for R184 itself, rendering the S1 pocket
of meprin β rather flexible. Hence, we only fixed the respective
substructures in the template for the flexible alignments, i.e.
the hydroxamic acid and the benzyl moiety addressing S1’,
while the second benzyl residue was allowed to be flexible
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
Figure 1. Tertiary amine-based inhibitors of meprin α/β and ovastacin with
different selectivity profiles.
dealkylation, albeit the compounds exhibited acceptable plas-
ma half-lives in rats.[12] However, these putative shortcomings of
the tertiary amines prompted us to investigate possible replace-
ments to generate bioisosteric scaffolds suitable for the
development of novel astacin inhibitors.
Results and Discussion
Scaffold design
As mentioned above, the tertiary amine based compounds
exhibit a high degree of conformational freedom, due to the
high flexibility of the aliphatic amine and a high number of
rotatable bonds. Hence, this flexibility might lead to entropic
penalty, caused by the loss of rotational freedom upon binding
to the target. A general strategy to overcome this issue in lead
optimization is the structural preorganization or rigidification of
a scaffold, for example by cyclization or macrocyclization of the
respective compounds. Hence, we assumed that the replace-
ment of the central tertiary amine by a suitable heterocycle
could lead to more rigid compounds and potentially improve
binding of the inhibitors to meprin α and β and maybe also
other astacin proteinases, for example, ovastacin. However,
initial experiments that replaced the acyclic tertiary amine 1 by
an alicyclic amine, i.e., a pyrrolidine moiety, just led to
equipotent inhibitors without any improved activity towards
meprin α or β (not shown). Thus, we next focused on
heteroaromatic cores as potential isosteres of the tertiary
amine, concurrently also causing deeper changes in chemical-
and physicochemical properties, that is, a scaffold hopping,
Figure 2. Possible strategies for rigidification of the tertiary amine scaffold:
A) “1,2,3” cyclization leading to prototypic pyrroles 2a, b; B) “1,2,4”-
cyclization leading to prototypic pyrazoles 3a, b; C) “ring-fusion” leading to
prototypic indoles 4a, b as potential novel inhibitors of astacin proteinases.
ChemMedChem 2020, 15, 1–14
www.chemmedchem.org
2
© 2020 The Authors. ChemMedChem published by Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cmdc.202000822
ChemMedChem
during the simulations (Figure 3A and B). This setup provided
alignment results with reasonable conformations and geo-
metries of the respective central heteroaromatic cores (Fig-
ure 3C–H). Although all compounds match the hydroxamic acid
and the benzyl moiety addressing the S1’ pocket due to the
fixed template, no complete match of the second benzyl
residue of the actual starting conformation of the template,
that is, the conformation of enzyme bound ligand 1a, could be
observed, regardless of the spacer between heterocycle and
hydroxamic acid. Nevertheless, the alignments of the pyrrole
(2a, b, Figure 3C and F) and indole derivatives (4a, b, Figure 3E
and H) suggested a better conformational fit of the compounds
compared to the tertiary amine template, as the respective
pyrazole-derivatives (3a, b, Figure 3D and G), indicating a more
suitable replacement of the amine scaffold. Based on this, we
assumed that compounds 2a and b as well as 4a and b could
be putative novel astacin inhibitors based on hitherto unex-
plored scaffolds in the context of metzincin proteinases.
To further corroborate these findings, in particular integrat-
ing also the environment of the target proteins, docking studies
were performed. For this purpose, a constraint was used,
tethering the hydroxamic acid substructure to ensure a proper
placement and zinc-binding of the ligands within the active site
cleft of the receptor proteins. The resulting docking solutions
revealed poses for all compounds within the active site cleft of
meprin β that are slightly diverging from the X-ray conforma-
tion of 1a (Figure 4). Although lacking functional groups, i.e.,
carboxylates, may rule out attractive interactions with R184 and
R238, thus leading to a different orientation of the scaffold within
the active site cleft. For the compounds with C1 spacer between
heteroaromatic core and hydroxamic acid, that is, 2a, and 4a
(Figure 4A, G), the aryl moiety addressing the S1 site is in a quite
similar orientation like for 1a, suggesting the possibility of the
formation of interactions with R184 upon suitable functionaliza-
tion and decoration. However, the docking solutions for the
pyrazole derivative 3a suggest no interaction with the S1
residue (Figure 4D). On the other hand, the moiety addressing
the S1’ pocket is shifted deeper into the subpocket compared to
1a. While for 3a (Figure 4D) the deviation from the conforma-
tion of 1a is marginal, the aryl moiety of 2a (Figure 4A) and 4a
(Figure 4G) is moved deeper in the S1’ pocket and thus closer to
R238 compared to 1a.
However, the docking solutions of 2a and 4a suggest
possible interactions with Y211 at the bottom of the active site
cleft and also F216 shaping the lower rim of the cleft of meprin
β. The docking solutions of 2a, 3a and 4a within the active site
of meprin α exhibit a slightly different orientation compared to
the poses found for meprin β (Figure 4B, E, H). This might be
due to a more narrow active site cleft in the homology model
of meprin α, created by the bulky Y187 shaping the S1 subpocket.
However, this could enable possible πÀ π interactions of Y187
and the aryl residue or even the heteroaromatic cores of 2a, 3a
and 4a, respectively, while the second aryl-moiety addresses
the S1’ subpocket.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
The docking solutions of the compounds within the active
site of ovastacin also suggest potential πÀ π interactions of the
heteroaromatic cores, although with F243, rather than the
residue shaping the S1 pocket, that is, F214 (Figure 4C, F, I). Like
for the meprins, the docking poses also revealed a proper
positioning of the putative inhibitors within the S1’ pocket.
Figure 3. Flexible alignments of putative novel inhibitors. A) Conformations of 1a from meprin β co-crystal; B) template for flexible alignment derived from
(A), red – fixed atoms, green – free atoms; C)–H) flexible alignments of compounds 2a, b, 3a, b, 4a, b (coloured), template start conformation (light grey),
template aligned with individual compounds (dark grey).
ChemMedChem 2020, 15, 1–14
www.chemmedchem.org
3
© 2020 The Authors. ChemMedChem published by Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cmdc.202000822
ChemMedChem
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
Figure 4. Docking of compounds with GOLD (Chemscore®, 10 runs, top-ranked solutions) within the active site of meprin β: A) 2a, D) 3a, G) 4a, X-ray
conformation of 1a in light grey, meprin α: B) 2a, E) 3a, H) 4a and ovastacin: C) 2a, F) 3a, I) 4a.
Taken together, the visual inspection of the docking experi-
ments revealed in general a good fit of the putative inhibitors
2a, 3a and 4a within the active sites of meprin α, β and
ovastacin, albeit compounds 2a and 4a seem to be able to
address the S1 and S1’ pockets slightly better than 3a, in
particular based on the docking solutions found for meprin β.
Hence, this would be in line with the results from the flexible
alignments, that also suggested a better fit of 2a and 4a (vide
supra). However, the docking experiments of the compounds
with C2 spacer (2b, 3b and 4b) yielded comparable poses (see
the Supporting Information), although in this case the solutions
of pyrazole 3b suggested a better fit within the active sites of
meprin α and β, compared to the solutions of the respective
derivative with shorter spacer (3a). Although none of the novel
compounds completely matched the bioactive conformation of
1a, the in silico experiments indicated additional interactions,
enabled by the introduction of the heteroaromatic core itself,
that is, electrostatic interactions with Y211 or F216 in meprin β or
the corresponding amino acids in meprin α and ovastacin,
respectively. Hence, this might also contribute to improved
ligand binding, rather than solely the fixation or rigidification of
the bioactive conformation of 1a. In particular the S1 pockets of
meprin α and ovastacin exhibit a lipophilic environment, since
they are formed by Y187 and F214, respectively. Thus, this creates
a favorable environment for hydrophobic aryl moieties of the
corresponding inhibitors, also corroborated by the docking
solutions.
Summarized, the docking solutions and the results from the
flexible alignments supported the idea of replacing the tertiary
amine with heteroaromatic cores and we assumed that all of
the proposed heteroaromatic inhibitors might be suitable
replacements of the amine scaffold. Moreover, the in silico
experiments indicated suitable binding modes for all the
putative compounds, either with C1 or C2 spacer. Hence, we
supposed that all of the aforementioned compounds could
yield suitable scaffolds for the development of meprin α,
meprin β and ovastacin inhibitors.
Due to the putative electrostatic interactions we also aimed
at the exploration of the influence of the electrostatic properties
of the heteroaromatic cores. Moreover, the introduction of
nitrogen at different positions of the heteroaromatic cores, that
is, “the necessary nitrogen”, might have a huge impact on
ChemMedChem 2020, 15, 1–14
www.chemmedchem.org
4
© 2020 The Authors. ChemMedChem published by Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cmdc.202000822
ChemMedChem
pharmacodynamic and pharmacokinetic properties that might
influence the further development of novel astacin proteinase
inhibitors.[15] Thus, we intended to introduce further five-
membered heterocycles, i.e., isoxazole, pyrazole, 1,2,3-triazole,
1,2,4-triazole and imidazole, and further fused heteroaromatic
cores, i.e. benzimidazole, imidazo[4,5-b]pyridines and indazole
as novel scaffolds. This enabled the exploration of different
electrostatic properties and also introduces different synthetic
linchpins for further derivatization of the core structures as part
of further optimization efforts.
tives 9c and d, respectively. Final aminolysis by means of
hydroxyl amine hydrochloride furnished the hydroxamic acids
2c–f.
1
2
3
4
5
6
7
8
9
The 3,5-diphenyl-1,2,4-triazoles (2g and h) were synthesized
starting either from N-benzoyl glycine or N-benzoyl β-alanine,
respectively.[16] Reaction with 2-fluoropyridine, trifluorome-
thane-sulfonic acid anhydride and benzhydrazide under micro-
wave conditions yielded the 1,2,4-triazole esters 11a and b that
were converted to this corresponding hydroxamic acids 2g and
h by means of hydroxyl amine hydrochloride. The imidazole 2i
was prepared starting from ethyl 4-amino-3-oxo-4-phenylbuta-
noate hydrochloride (12) that was acylated with phenylisothio-
cyanate followed by subsequent cyclization using pyridinium
para-toluene sulfonate (PPTS), yielding thioimidazolidinone
13.[17] Desulfurization by means of hydrogen peroxide furnished
imidazole ester 14, that was finally converted to the corre-
sponding hydroxamic acid 2i by aminolysis using hydroxyl-
amine hydrochloride. The 1,2,3-triazole derivative 2j was
synthesized from methyl 4-oxo-4-phenyl-butanoate (15) by
copper mediated reaction with tosylhydrazide and aniline.[18]
The resulting methyl 1,4-diphenyl-1,2,3-triazol ester (16) was
further treated with hydroxyl amine hydrochloride yielding
compound 2j.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
Chemistry
The synthesis of the monocyclic heteroaromatic compounds
was accomplished as depicted in Scheme 1. The 2,5-diphenyl
pyrrole derivatives 2a and b were synthesized starting from 1,4-
diphenylbutane-1,4-dione 5 by Paal–Knorr synthesis followed
by deprotection of the benzylhydroxamic acid derivatives 6a
and b. The 3,5-diphenyl isoxazole- (2c and d) and the 3,5-
diphenyl pyrazole-derivatives (2e and f) were synthesized from
1,3-diphenylpropane-1,3-dione 7. After alkylation by either
methyl bromoacetate or methyl acrylate, the isoxazoles 9a and
b were furnished by cyclocondensation of 8a and b and
hydroxylamine hydrochloride. Condensation of 8a and b with
hydrazine hydrate yielded the corresponding pyrazole deriva-
The two 3,5-diphenyl pyrazoles 3a&b were synthesized
from 1,3-diphenylpropane-1,3-dione 7, that was converted to
pyrazole 17 by means of hydrazine hydrate. Alkylation with
Scheme 1. Synthesis of monocyclic heteroaromatic inhibitors. a) pTSA, THF/toluene (1:1, v/v), reflux, 23–50%; b) H2, 4 bar, Pd/C, MeOH/THF (1:1, v/v), RT,
°
°
°
4–37%; c) NaH, DMF, 0 C to RT, 73–100%; d) DBU, CH2Cl2, 0 C to RT, 35%; e) NH2OH·HCl, EtOH/water (3:2, v/v), μW, 110 C, 15–31%; f) N2H4 ·H2O, EtOH/THF
°
°
°
°
(2:1, v/v), RT, 58–99%; g) NH2OH·HCl, NaOCH3, MeOH, μW, 80 C, 19–66%; h) i: 2-fluoropyridine, Tf2O, CH2Cl2, 0 C, ii: μW, 140 C, 29–34%; i) i: TEA, EtOH, 50 C,
°
°
°
ii: PPTS, toluene, 120 C, 48%; j) H2O2, AcOH, RT, 42%; k) i: tosylhydrazide, toluene, 80 C, ii: Cu(OAc)2, pivalic acid, 100 C, 24%; l) DBU, MeCN, RT, 96%.
ChemMedChem 2020, 15, 1–14
www.chemmedchem.org
5
© 2020 The Authors. ChemMedChem published by Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cmdc.202000822
ChemMedChem
either methyl bromoacetate or methyl acrylate, respectively,
yielded pyrazole esters 18a and b, that were finally converted
to corresponding hydroxamic acids 3a and b by means of
hydroxylamine hydrochloride.
Structure-activity relationships
1
2
3
4
5
6
7
8
9
The synthesized inhibitors were first evaluated with regard to
their ability to modulate meprin α and β activity (Table 1). The
prototypic inhibitors of the “1,2,3-cyclization”, that is, pyrroles
2a and 2b, exhibited a rather unexpected high activity
compared to the reference inhibitors, based on the tertiary
amine (1d and 1e). The transition from the aliphatic tertiary
amine to the heteroaromatic core led to an almost 400 fold
increased Kiapp against meprin α of 2a compared to 1d.
The synthesis of the fused heteroaromatic inhibitors is
depicted in scheme 2. The two N-alkyl indole derivatives 4a and
b were synthesized from 2-phenylindole (19), that was alkylated
using either methyl bromoacetate or methyl acrylate, followed
by aminolysis of the intermediate esters 20. Benzimidazoles 4c
and d were synthesized starting from phenyl-1,2-diamine
(21).[19] Alkylation of the resulting 2-phenylbenzimidazole (22)
with methyl bromoacetate or methyl acrylate led to benzimida-
zole esters 23a and b, that were converted to the correspond-
ing hydroxamic acids 4c and d by means of hydroxyl amine
hydrochloride. The indole derivative 4e was synthesized by
Fischer indole synthesis from benzoylpropionic acid (24) and
subsequent esterification, followed by aminolysis of the ester
intermediate 25. Imidazo[4,5-b]pyridine 4f was synthesized by
SNAr reaction starting from 2-chloro-3-nitropyridine 26.[20]
Reductive cyclization of 27 with benzaldehyde and sodium
dithionite yielded ester intermediate 28[21], that was finally
converted to hydroxamic acid 4f by hydroxyaminolysis. The
respective regioisomer 4g was synthesized starting from 2,3-
diaminopyridine 29.[22] Cyclization with benzaldehyde yielded
imidazopyridine 30. Alkylation with methyl bromoacetate,
followed by separation of the isomers furnished ester inter-
mediate 31 that was treated with hydroxylamine, yielding
imidazo[4,5-b]pyridine 4g. Finally, indazole 4h was prepared
from azobenzene 32.[23] Rhodium catalyzed reaction with meth-
yl acrylate yielded 33 that was converted to the respective
hydroxamate 4h by hydroxylaminolysis.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
Although the activity of 2a against meprin
β is less
pronounced, the scaffold-hopping yielded a compound with an
activity in the micromolar range compared to an almost inactive
tertiary amine counterpart (1e). The inhibitory activity of the
corresponding compound bearing a C2 spacer between hydrox-
amate and scaffold (2b) against meprin α and β is also
increased compared to the respective tertiary amine derivative
1e. However, the gain of activity for this compound is just
twofold for meprin α and about fourfold for meprin β. Of note,
the activity of 2a, bearing a shorter spacer, is significantly
higher compared to 2b. This is contrary to the SAR found for
the tertiary amine based inhibitors, where a longer spacer
resulted in an increased activity against meprin α.[13] This might
also indicate a different binding mode, as already suggested by
the in silico experiments (vide supra). The exploration of
different heteroaromatic cores led to further improved poten-
cies against meprin α and also meprin β, corroborating the
potential πÀ π interactions of the heterocycles revealed by the
docking studies.
The replacement of the central pyrrole of 2a by an isoxazole
(2c) led to a further 5-fold increase in activity against both
proteinases, exhibiting a Kiapp in the lower one-digit nanomolar
range against meprin α and a sub-micromolar Kiapp of 813 nM
against meprin β, yielding an unprecedented 2000-fold im-
°
°
Scheme 2. Synthesis of fused heteroaromatic inhibitors. a) NaH, DMF, 0 C to RT, 92–96%; b) DBU, MeCN, RT, 67%; c) NH2OH·HCl, NaOCH3, μW, 80 C, 15–66%;
°
°
d) H2O2, CAN, MeCN, μW, 50 C, 53%; e) K2CO3, MeCN, reflux, 66%; f) pTSA, ZnCl2, AcOH, μW, 180 C, 47%; g) MeOH, H2SO4, reflux, 71%; h) TEA, DMF, μW,
°
°
°
120 C, 94%; i) Na2S2O4, DMF/EtOH (1:1, v/v), 80 C, 36%; j) H2O, reflux, 26%; k) [Cp*RhCl2]2, Cu(OAc)2, DMF, 130 C, 13%.
ChemMedChem 2020, 15, 1–14
www.chemmedchem.org
6
© 2020 The Authors. ChemMedChem published by Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cmdc.202000822
ChemMedChem
Table 1. Inhibition of meprin α and β by five-membered and fused heteroaromatic hydroxamates.
1
2
scaffold
No
Het
n
Kiapp [nM]
SF[b]
Meprin α[a]
Meprin β[a]
3
4
5
6
1d[c]
1e[c]
–
–
1
2
7868 (1.04)
1647 (1.04)
n.d.[d]
74361 (1.04)
–
45
7
8
9
2a
2b
2c
1
2
1
20 (1.25)
736 (1.07)
4 (1.06)
4311 (1.09)
15026 (1.08)
813 (1.08)
216
20
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
203
2d
2
31 (1.09)
13491 (1.05)
435
2e
2f
1
2
3 (1.04)
76 (1.12)
199 (1.05)
8034 (1.07)
66
106
2g
2h
1
2
66 (1.02)
306 (1.12)
8680 (1.04)
42320 (1.09)
132
138
2i
2j
1
1
2 (1.15)
9 (1.01)
443 (1.22)
222
161
1448 (1.03)
3a
3b
1
2
256 (1.08)
406 (1.09)
10440 (1.06)
15411 (1.13)
41
38
4a
4b
1
2
69 (1.08)
414 (1.02)
1076 (1.05)
16990 (1.21)
16
41
4c
4d
1
2
163 (1.14)
1031 (1.09)
1625 (1.08)
27915 (1.05)
10
27
4e
4f
1
1
82 (1.20)
1184 (1.02)
4556 (1.12)
14
17
262 (1.01)
4g
1
1483 (1.35)
4170 (1.08)
3
4h
1
–
273 (1.13)
8 (1.21)
3364 (1.08)
219 (1.28)
12
27
Actinonin
–
–
[a] Geometric mean of three independent experiments with standard deviation factor. [b] SF: selectivity factor (Kiapp meprin β/Kiapp meprin α). [c] Data from
ref. [12]. [d] Not determined, relative activity 82% @ 100 μM
provement in potency against meprin α compared to the
respective amine 1d. Moreover, 2c exhibits already a quite
favorable selectivity for meprin α over meprin β, i.e., by factor
200. The introduction of the C2 spacer again led to a slightly
ChemMedChem 2020, 15, 1–14
www.chemmedchem.org
7
© 2020 The Authors. ChemMedChem published by Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cmdc.202000822
ChemMedChem
decreased activity of 2d compared to 2c. However, that
decrease is even more pronounced with regard to meprin β
inhibition. Thus, the selectivity for meprin α over β inhibition is
even higher for 2d, about factor 400. Although these com-
pounds were originally designed just as novel scaffolds, the
high activity of 2c and 2d against meprin α without further
scaffold decoration in combination with the favorable selectivity
over meprin β renders them as suitable probes for target
validation, rather than just a scaffold for further derivatization.
The same SAR-trend was also observed for the respective
pyrazole derivatives 2e and f, albeit the activity against meprin
β was slightly higher than for the isoxazoles. However, the
inhibitory activity of the respective 1,2,4-triazoles 2g and h was
slightly decreased compared to the isoxazoles, by factor 10, but
still in the same range as the corresponding pyrroles 2a and b.
This might underpin the influence of the electrostatic properties
of the heteroaromatic core on the inhibitory activity by
affecting the putative πÀ π interactions found by the docking
experiments (vide supra). Nevertheless, also 2g and h are also
far more potent compared to the tertiary amines 1d and e. The
activity of the imidazole 2i and 1,2,3-triazole 2j is again in the
same activity range as the isoxazole 2c or pyrazole 2e,
exemplifying for 2i that also a basic heterocycle is tolerated,
resembling also the basic properties of the tertiary amines.
The prototypic compounds for the “1,2,4-cyclization”, that
is, the pyrazoles 3a and b, exhibited also increased activity
compared to the corresponding tertiary amines 1d and e, albeit
to a smaller extend as the respective 1,2,3-cyclization com-
pounds 2a–i. This is in line with the results of the in silico
experiments (vide supra), which also led to the conclusion that
pyrazoles 3a and b are less likely to meet the active
conformation of tertiary amine 1a and are less likely to address
the S1 and S1’ subpocket within the active site. A different
binding mode is further corroborated by the missing influence
of the spacer, since the corresponding inhibitors with either C1
or C2 spacer exhibit virtually the same activities. Finally, the
compounds that were rigidified by means of ring fusion, i.e.,
indoles 4a and 4b, also exhibited increased activities compared
to the tertiary amines 1d and e. The gain of activity is in a
comparable range as found for the pyrroles 2a and b,
respectively. The introduction of the C2 spacer (4b) led to a
reduced activity compared to the compound with C1 spacer
(4a), which is again contrary to the SAR of the tertiary amines
but in line with the SAR of the 1,2,3-cylized inhibitors 2a–h. The
introduction of an additional nitrogen (4c, d and h) or an
altered indole substitution pattern (4e) had only marginal
influence on the activity. However, the introduction of
Table 2. Inhibition of ovastacin by selected inhibitors.
1
2
3
4
5
6
7
8
9
Cmpd.
Kiapp [nM]
SF
Meprin α[a]
Meprin β[b]
2c
2e
2f
66 (1.09)
17
65
32
74
104
4
0.1
1
196 (1.19)
2423 (1.04)
4877 (1.06)
31799 (1.01)
1045 (1.01)
1096 (1.07)
93 (1.19)
1300 (1.09)
2090 (1.02)
9798 (1.06)
0.3
0.6
0.8
0.1
0.1
0.1
0.1
1.3
0.4
2g
2h
3a
3b
4a
4b
4c
4d
3
1.3
3.1
13
9.5
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
[a] Selectivity factor (Kiapp ovastacin/Kiapp meprin α); [b] Selectivity factor
(Kiapp ovastacin/Kiapp meprin β)
compared to meprin α, they are in the same range or even
better than found for meprin β, rendering these compounds
also as highly suitable scaffolds for the further development of
ovastacin inhibitors. In particular, the compounds obtained by
“1,2,3-cylization” exhibit favorable inhibition properties. Of note,
isoxazole 2c, pyrazole 2e and indole 4a exhibit even higher
activities without any scaffold decoration as the recently
reported tertiary amine based ovastacin inhibitors.[14] Thus,
these compounds are the most potent small molecule ovastacin
inhibitors reported to date.
Finally, the inhibition of exemplary proteases belonging to
the metzincin target family, that is, ADAMs and MMPs, was
evaluated (Table 3). The selected inhibitors exhibited a slightly
less favorable selectivity profile compared to the tertiary amines
reported recently,[12] showing some inhibition of individual
proteases at a concentration of 200 μM. Nevertheless, all tested
compounds exhibited only little inhibition of the off-target
proteinases at a lower concentration of 10 μM. Thus, these
heteroaromatic cores could be considered as suitable scaffolds
for the development of astacin inhibitors, without significant
off-target inhibition liabilities, per se. With regard to the high
potency of compounds 2c and e against meprin α in the low
nanomolar range, the low off-target inhibition at 10 μM results
in a selectivity factor >10000, supporting their suitability as
chemical probes for meprin α target validation. Moreover, none
of the tested compounds suffered from cytotoxicity issues,
exemplified by the cell viability of liver and neuronal cell lines
(Table 3).
imidazopyridines (4f and g) affects in particular the activity Conclusion
dependent on the substitution pattern, underpinning a poten-
tial influence of the electrostatic properties of the heterocyclic
cores.
Following the inhibitor evaluation with meprin α and β,
selected compounds were investigated regarding their inhib-
itory activity against ovastacin (Table 2). Except triazole 2h, all
of the tested compounds exhibited also activities in the lower
micromolar (2f, g; 3a, b; 4b, c, d) or even nanomolar range (2c,
Proteinases of the astacin family are increasingly getting into
the focus of drug discovery. In particular, the inhibition of
meprin α and β, as well as ovastacin offers novel treatment
options for kidney diseases, fibrosis, cancer or infertility,
respectively. Recently, the first potent and selective inhibitors of
meprins and ovastacin have been reported, based on a tertiary
amine scaffold. Albeit this scaffold exhibits favorable in vitro
properties, it also exhibits several potential liabilities, such as
e
and 4a). Although the inhibitory activities are lower
ChemMedChem 2020, 15, 1–14
www.chemmedchem.org
8
© 2020 The Authors. ChemMedChem published by Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cmdc.202000822
ChemMedChem
Table 3. Inhibition of off-target proteinases and in vitro toxicity of selected compounds.
compound [μM] 2a 2c 2d
1
2
3
c
2e
Relative enzyme activity [%]
2f
2g
4a
4c
4
5
6
7
8
9
MMP2
10
200
10
200
10
200
10
200
10
200
83
1
92
1
79
5
82
29
82
25
87
98
74
89
39
79
66
75
21
75
24
104
61
81
27
75
67
73
23
61
7
97
95
70
99
54
93
76
89
83
90
56
85
89
34
86
16
80
31
80
22
62
9
77
93
65
93
68
66
12
76
27
41
90
53
77
54
89
68
88
56
46
96
68
102
72
87
76
92
56
MMP9
MMP13
ADAM10
ADAM17
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
Cell viability [%]
HepG2
SY-5Y
30
30
n.t.
n.t.
85
96
90
123
100
92
93
93
94
99
n.t.
n.t.
85
94
n.t.: not tested
high-resolution positive ion ESI mass spectra were obtained from a
LTQ Orbitrap XL (Thermo Fisher Scientific). The NMR spectra were
recorded at a DDR2 400 spectrometer (Agilent) and an AvanceIII-
System (Bruker Biospin GmbH). [D6]DMSO was used as solvent
unless otherwise specified. Chemical shifts are expressed as parts
per million (ppm). The solvent was used as internal standard.
Splitting patterns have been designated as follows: s (singlet), d
(doublet), dd (doublet of doublet), t (triplet), m (multiplet), and br
(broad signal). Semi preparative HPLC was performed on a Prepstar
device (Varian) equipped with a Phenomenex Luna 10 μM C18(2)
column (250 mm×21 mm). The compounds were eluted using the
same solvent system as described above, applying a flow rate of
21 mL/min.
metabolic instability or also entropic unfavorable binding to the
target due to the high conformational flexibility. The latter issue
was addressed by a rigidification and scaffold-hopping ap-
proach within this study, yielding novel astacin inhibitors with
heteroaromatic scaffolds.
Thus, the formal cyclization of the tertiary amine resulted in
an unprecedented increase in activity against meprin α and β.
Moreover, selected compounds also exhibited favorable inhib-
ition of ovastacin. The novel heteroaromatic scaffolds did not
show marked inhibition of related metalloproteinases, nor
cytotoxic effects, rendering them very suitable starting points
for further medicinal chemistry exploration and optimization.
However, already at this stage – without any functionalization
or decoration – some compounds exhibited remarkable high
potency against meprin α accompanied with a favorable
selectivity over meprin β. Hence, these inhibitors could be
directly used as chemical probes for target validation purposes
of meprin α, for example, in the context of cancer, as a more
suitable replacement for the commonly utilized less selective
broadspectrum metallo-proteinase inhibitor actinonin.
General method for deprotection of hydroxamic acids by hydro-
genation: The respective benzyl-protected hydroxamic acid deriva-
tive was dissolved in MeOH/THF (1:1, v/v, 10 mL). Palladium on
charcoal was added and the vial was purged with hydrogen. After
4 h at 4 bar, the mixture was filtered through celite and evaporated.
The residue was purified by semi-preparative HPLC.
2-(2,5-Diphenylpyrrol-1-yl)ethanehydroxamic acid (2a): The com-
pound was synthesized from O-benzyl-2-(2,5-diphenylpyrrol-1-yl)
ethane-hydroxamic acid (6a, 90 mg, 0.24 mmol, 1 equiv) as de-
scribed above. Yield: 25 mg (37%); ESI-MS: m/z 293.2 [M+H]+;
HPLC: tR =15.84 min (97.7%); 1H NMR (400 MHz, [D6]DMSO): δ=
4.34 (s, 1.5H), 4.65 (s, 0.5H), 6.27 (s, 2H), 7.33–7.36 (m, 2H), 7.41–7.47
(m, 8H), 8.99 (br s, 0.6H), 9.19 (br s, 0.2H), 10.26 (br s, 0.2H), 10.54
(br s, 0.8H) mixture of cis-trans isomers; 13C NMR (176 MHz, [D6]
Experimental Section
DMSO): δ=46.2, 109.6, 127.6, 128.7, 128.9, 129.0, 133.5, 137.1,
Chemistry
166.4; HRMS m/z: 315.1114 [M+Na]+, calcd for C18H16N2NaO2
:
+
General. Starting materials and solvents were purchased from
Aldrich, Activate Scientific, Alfa Aesar and Merck Millipore. The
purity of the compounds was assessed by HPLC and confirmed to
be �95%. The analytical HPLC-system consisted of a MerckÀ
Hitachi device (model LaChrom) utilizing a Phenomenex Luna 5 μM
C18(2) column (125 mm×4.0 mm) with λ=214 nm as the reporting
wavelength. The compounds were analyzed using a gradient at a
flow rate of 1 mL/min, whereby eluent (A) was acetonitrile, eluent
(B) was water, both containing 0.04% (v/v) trifluoro acetic acid
applying the following gradient: 0–5 min: 5% (A), 5–15 min: 5!
60% (A), 15–20 min: 60!95% (A), 20–30 min: 95% (A); gradient 2:
0–15 min: 5!50% (A), 15–20 min: 50!95% (A), 20–23 min: 95%
(A). The purities of all reported compounds were determined by the
percentage of the peak area at 214 nm. ESI or APCI mass spectra
were obtained with an Expression CMS spectrometer (Advion). The
315.1104.
3-(2,5-Diphenylpyrrol-1-yl)propanehydroxamic acid (2b): The
compound was synthesized from O-benzyl-3-(2,5-diphenylpyrrol-1-
yl)-propanehydroxamic acid (6b, 300 mg, 0.76 mmol, 1 equiv) as
described above. Yield: 9 mg (4%); ESI-MS: m/z 307.4 [M+H]+,
329.4 [M+Na]+; HPLC: tR =16.27 min (>99%); 1H NMR (400 MHz,
[D6]DMSO): δ=1.83–1.87 (m, 1.8H), 2.09–2.13 (m, 0.2H), 4.28–4.32
(m, 2H), 6.23 (s, 2H), 7.34–7.38 (m, 2H), 7.44–7.51 (m, 8H), 8.57 (br s,
0.6H), 8.81 (br s, 0.1H), 9.76 (br s, 0.1H), 10.16 (br s, 0.9H) mixture of
cis-trans isomers; HRMS m/z: 329.1273 [M+Na]+, calcd for
C19H18N2NaO2+: 329.1260.
General method for the synthesis of hydroxamic acids from
carboxylic acid esters: The respective ester derivative (1 equiv) was
dissolved in MeOH (5 mL), treated with NaOCH3 (6 equiv) and
ChemMedChem 2020, 15, 1–14
www.chemmedchem.org
9
© 2020 The Authors. ChemMedChem published by Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cmdc.202000822
ChemMedChem
hydroxylamine hydrochloride (3 equiv). The mixture was heated in
2-(3,5-Diphenylimidazol-4-yl)ethanehydroxamic acid (2i): The
compound was synthesized from ethyl 2-(3,5-diphenylimidazol-4-yl)
acetate (14, 89 mg, 0.29 mmol, 1 equiv) as described above. Yield:
34 mg (39%); ESI-MS: m/z 294.1 [M+H]+; HPLC: tR =6.51 min (>
99%); 1H NMR (400 MHz, [D6]DMSO): δ=3.47 (s, 1.7H), 3.77 (s, 0.3H),
7.44-7.47 (m, 1H), 7.51–7.69 (m, 9H), 8.88 (br s, 0.1H), 9.32 (br s,
0.1H), 10.25 (s, 0.2H), 10.61 (s, 0.8H) mixture of cis-trans isomers; 13C
NMR (176 MHz, [D6]DMSO): δ=28.1, 124.4, 126.9, 128.2, 129.2,
129.5, 129.6, 130.3, 130.5, 134.6, 137.1, 158.6, 165.2; HRMS m/z:
294.1231 [M+H]+, calcd for C17H16N3O2+: 294.1237.
1
2
3
4
5
6
7
8
9
°
a microwave at 80 C for 10 minutes. The volatiles were evaporated.
The remains were taken up in water, acidified by means of diluted
aqueous HCl and extracted with EtOAc (3×25 mL). The combined
organic layers were dried over Na2SO4 and evaporated. The residue
was purified by semi-preparative HPLC.
2-(3,5-Diphenylisoxazol-4-yl)ethanehydroxamic acid (2c): The
compound was synthesized from methyl 2-(3,5-diphenylisoxazol-4-
yl)acetate (9a, 92 mg, 0.31 mmol, 1 equiv) as described above. Yield:
32 mg (35%); ESI-MS: m/z 295.1 [M+H]+, 317.2 [M+Na]+; HPLC:
tR =14.56 min (>99%); 1H NMR (400 MHz, [D6]DMSO): δ=3.41 (s,
2H), 7.55–7.62 (m, 6H), 7.68–7.70 (m, 2H), 7.79–7.82 (m, 2H), 10.77
(s, 1H); 13C NMR (176 MHz, [D6]DMSO): δ=27.4, 108.2, 127.5, 127.8,
128.5, 128.7, 129.1, 129.4, 129.7, 130.3, 130.8, 164.2, 166.6, 167.6;
HRMS m/z: 317.0907 [M+Na]+, calcd for C17H14N2NaO3+: 317.0897.
2-(3,5-Diphenyl-1,2,3-triazol-4-yl)ethanehydroxamic acid (2j): The
compound was synthesized from methyl 2-(3,5-diphenyl-1,2,3-
triazol-4-yl)acetate (16, 70 mg, 0.24 mmol, 1 equiv) as described
above. Yield: 23 mg (32%); ESI-MS: m/z 295.1 [M+H]+; HPLC: tR =
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
1
10.43 min (>99%); H NMR (400 MHz, [D6]DMSO): δ=3.63 (s, 1.7H),
3
3.92 (s, 0.3H), 7.42–7.45 (m, 1H), 7.52 (t, 2H, J=7.8 Hz), 7.57–7.69
3-(3,5-Diphenylisoxazol-4-yl)propanehydroxamic acid (2d): The
compound was synthesized from methyl 3-(3,5-diphenylisoxazol-4-
yl)propanoate (9b, 49 mg, 0.16 mmol)) as described above. Yield:
11 mg (22%); ESI-MS: m/z 309.4 [M+H]+, 331.4 [M+Na]+; HPLC:
(m, 5H), 7.75–7.77 (m, 2H), 9.04 (br s, 0.6H), 9.39 (br s, 0.2H), 10.33
(s, 0.1H), 10.76 (s, 0.9H) mixture of cis-trans isomers; 13C NMR
(176 MHz, [D6]DMSO): δ=28.1, 125.9, 126.1, 127.2, 127.4, 128.6,
129.2, 129.3, 130.2, 130.4, 131.3, 136.3, 145.5, 164.9; HRMS m/z:
295.1207 [M+H]+, calcd for C16H15N4O2+: 295.1190.
1
tR =14.93 min (>99%); H NMR (400 MHz, [D6]DMSO): δ=2.09–2.13
(m, 2H), 2.95–2.99 (m, 2H), 7.57–7.63 (m, 2H), 7.69–7.71 (m, 2H),
7.81–7.84 (m, 2H), 10.36 (br s, 1H); 13C NMR (176 MHz, [D6]DMSO):
δ=28.42, 108.45, 127.79, 128.03, 128.29, 129.08, 168.31; HRMS m/z:
294.1251 [M+H]+, calcd for C17H16N3O2+: 294.1237.
2-(3,5-Diphenylpyrazol-1-yl)ethanehydroxamic acid (3a): The
compound was synthesized from methyl 2-(3,5-diphenyl-1H-pyra-
zol-1-yl)acetate (18a, 247 mg, 0.84 mmol, 1 equiv) as described
above. Yield: 141 mg (57%); ESI-MS: m/z 294.4 [M+H]+; HPLC: tR =
2-(3,5-Diphenyl-1H-pyrazol-4-yl)ethanehydroxamic acid (2e): The
compound was synthesized from methyl 2-(3,5-diphenyl-1H-pyra-
zol-4-yl)acetate (9c, 209 mg, 0.72 mmol, 1 equiv) as described
above. Yield: 83 mg (40%); ESI-MS: m/z 294.5 [M+H]+; HPLC: tR =
1
15.63 min (97.5%); H NMR (400 MHz, [D6]DMSO): δ=4.68 (s, 1.8H),
5.04 (br s, 0.2H), 6.91 (s, 1H), 7.31–7.35 (m, 1H), 7.41–7.55 (m, 5H),
7.65–7.68 (m, 2H), 7.83–7.85 (m, 2H), 9.12 (br s, 0.9H), 9.36 (br s,
0.1H), 10.35 (br s, 0.1H), 10.93 (br s, 0.9H) mixture of cis-trans
isomers; 13C NMR (176 MHz, [D6]DMSO): δ=50.4, 103.7, 125.6, 128.2,
129.1, 129.2, 129.3, 130.4, 133.5, 146.4, 150.3, 164.4; HRMS m/z:
294.1251 [M+H]+, calcd for C17H16N3O2+: 294.1237.
1
12.03 min (>99%); H NMR (400 MHz, [D6]DMSO): δ=3.32 (s, 1.8H),
3.63 (s, 0.2H), 7.38–7.42 (m, 2H), 7.46–7.49 (m, 4H), 7.64–7.66 (m,
4H), 10.09 (br s, 0.1H), 10.62 (br s, 0.9H) mixture of cis-trans isomers;
13C NMR (176 MHz, [D6]DMSO): δ=28.4, 108.5, 127.8, 128.0, 128.3,
+
129.1, 168.3; HRMS m/z: 294.1251 [M+H]+, calcd for C17H16N3O2
:
3-(3,5-Diphenylpyrazol-1-yl)propanehydroxamic acid (3b): The
compound was synthesized from methyl 3-(3,5-diphenyl-1H-pyra-
zol-1-yl)propanoate (18b, 259 mg, 0.84 mmol, 1 equiv) as described
above. Yield: 171 mg (66%); ESI-MS: m/z 308.4 [M+H]+; HPLC: tR =
17.68 min (>99%); 1H NMR (400 MHz, [D6]DMSO): δ=2.66 (t, 2H,
294.1237.
3-(3,5-Diphenyl-1H-pyrazol-4-yl)propanehydroxamic acid (2f):
The compound was synthesized from methyl 3-(3,5-diphenyl-1H-
pyrazol-4-yl)propanoate (9d, 306 mg, 1 mmol, 1 equiv) as described
above. Yield: 58 mg (19%); ESI-MS: m/z 308.4 [M+H]+; HPLC: tR =
3
3J=7.2 Hz), 4.32 (t, 2H, J=7.2 Hz), 6.84 (s, 1H), 7.30–7.34 (m, 1H),
7.41–7.45 (m, 2H), 7.47-7.57 (m, 3H), 7.60–7.63 (m, 2H), 7.84–7.86
(m, 2H), 9.97 (br s, 0.1H), 10.53 (br s, 0.9H) mixture of cis-trans
isomers; 13C NMR (176 MHz, [D6]DMSO): δ=33.0, 45.6, 130.8, 125.6,
128.0, 129.1, 129.1, 129.2, 129.3, 130.5, 133.7, 145.2, 149.9, 166.8;
HRMS m/z: 330.1223 [M+Na]+, calcd for C18H17N3NaO2+: 330.1213.
1
15.52 min (96.8%); H NMR (400 MHz, [D6]DMSO): δ=2.06–2.11 (m,
2H), 2.93–2.97 (m, 2H), 7.41 (t, 2H, 3J=7.3 Hz), 7.50 (t, 4H, 3J=
3
7.6 Hz), 7.64 (d, 4H, J=7.8 Hz), 10.33 (br s, 1H); 13C NMR (176 MHz,
[D6]DMSO): δ=20.0, 33.5, 114.1, 128.0, 128.2, 129.2, 168.7; HRMS m/
z: 308.1409 [M+H]+, calcd for C18H18N3O2+: 308.1394.
2-(2-Phenylindol-1-yl)ethanehydroxamic acid (4a): The compound
was synthesized from methyl 2-(2-phenyl-1H-indol-1-yl)acetate
(20a, 265 mg, 1 mmol, 1 equiv) as described above. Yield: 91 mg
2-(3,5-Diphenyl-1,2,4-triazol-4-yl)ethanehydroxamic acid (2g):
The compound was synthesized from methyl 2-(3,5-diphenyl-4H-
1,2,4-triazol-4-yl)acetate (11a, 86 mg, 0.29 mmol, 1 equiv) as de-
scribed above. Yield: 45 mg (53%); ESI-MS: m/z 295.3 [M+H]+;
1
(34%); ESI-MS: m/z 267.3 [M+H]+; HPLC: tR =15.81 min (>99%); H
NMR (400 MHz, [D6]DMSO): δ=4.66 (s, 1.7H), 4.99 (s, 0.3H), 6.58
(s,1H), 7.08-7.12 (m, 1H), 7.16–7.20 (m, 1H), 7.36–7.38 (m, 1H), 7.43–
7.47 (m, 1H), 7.49-7.53 (m, 2H), 7.58–7.64 (m, 3H), 10.34 (s, 0.1H),
10.93 (s, 0.9H) mixture of cis-trans isomers; 13C NMR (176 MHz, [D6]
DMSO): δ=44.9, 102.2, 110.7, 120.4, 120.6, 122.0, 128.1, 128.6,
129.2, 129.6, 132.5, 138.5, 142.0, 165.3; HRMS m/z: 289.0960 [M+
Na]+, calcd for C16H14N2NaO2+: 289.0947.
1
HPLC: tR =7.73 min (98.2%); H NMR (400 MHz, [D6]DMSO): δ=4.56
(s, 1.5 H), 4.85 (s, 0.5 H), 7.58–7.70 (m, 10 H), 10.50 (br s, 0.3 H),
10.81 (br s, 0.7 H) mixture of cis-trans isomers; 13C NMR (176 MHz,
[D6]DMSO): δ=45.5, 127.6, 129.0, 129.2, 129.4, 129.5, 130.6, 155.9,
+
164.0; HRMS m/z: 295.1204 [M+H]+, calcd for C16H15N4O2
:
295.1190.
3-(3,5-Diphenyl-1,2,4-triazol-4-yl)propanehydroxamic acid (2h):
The compound was synthesized from methyl 3-(3,5-diphenyl-4H-
1,2,4-triazol-4-yl)propanoate (11b, 39 mg, 0.34 mmol, 1 equiv) as
described above. Yield: 39 mg (37%); ESI-MS: m/z 309.4 [M+H]+;
3-(2-Phenylindol-1-yl)propanehydroxamic acid (4b): The com-
pound was synthesized from methyl 2-(2-phenyl-1H-indol-1-yl)
acetate (20b, 188 mg, 0.67 mmol, 1 equiv) as described above.
Yield: 33 mg (18%); ESI-MS: m/z 281.2 [M+H]+; HPLC: tR =
1
HPLC: tR =8.59 min (97.0%); H NMR (400 MHz, [D6]DMSO): δ=2.10
1
15.22 min (>99%); H NMR (400 MHz, [D6]DMSO): δ=2.34–2.38 (m,
3
3
(t, 2H, J=7.6 Hz), 4.39 (t, 2H, J=7.5 Hz), 7.61–7.63 (m, 6H), 7.75–
7.78 (m, 4H), 9.89 (br s, 0.1H), 10.28 (br s, 0.9H) mixture of cis-trans
isomers; 13C NMR (176 MHz, [D6]DMSO): δ=31.4, 41.5, 127.5, 129.4,
129.5, 130.8, 155.0, 165.9; HRMS m/z: 309.1358 [M+H]+, calcd for
C17H17N4O2+: 309.1346.
2H), 4.36–4.40 (m, 2H), 6.54 (s, 1H), 7.07–7.10 (m, 1H), 7.18–7.22 (m,
1H), 7.45–7.49 (m, 1H), 7.51–7.58 (m, 6H), 9.97 (br s, 0.1H), 10.46 (br
s, 0.9H) mixture of cis-trans isomers; 13C NMR (176 MHz, [D6]DMSO):
δ=33.3, 40.6, 102.6, 111.0, 120.7, 122.1, 128.2, 128.6, 129.2, 129.6,
ChemMedChem 2020, 15, 1–14
www.chemmedchem.org
10
© 2020 The Authors. ChemMedChem published by Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cmdc.202000822
ChemMedChem
132.7, 137.5, 141.1, 166.7; HRMS m/z: 281.1301 [M+H]+, calcd for
99%); 1H NMR (400 MHz, [D6]DMSO): δ=3.83 (s, 1.8H), 4.15 (s, 0.2H),
7.08–7.11 (m, 1H), 7.30–7.34 (m, 1H), 7.55–7.66 (m, 4H), 7.72–7.78
(m, 3H), 10.21 (s, 0.1H), 10.86 (s, 0.9H) mixture of cis-trans isomers;
13C NMR (176 MHz, CD3OD): δ=29.0, 116.3, 119.9, 121.6, 126.3,
127.2, 129.1, 129.4, 130.2, 139.0, 148.3, 166.3; HRMS m/z: 268.1097
[M+H]+, calcd for C15H14N3O2+, 268.1081.
1
2
3
4
5
6
7
8
9
C17H17N2O2+: 281.1285.
2-(2-Phenylbenzimidazol-1-yl)ethanehydroxamic acid (4c): The
compound was synthesized from methyl 2-(2-phenyl-1H-benzimi-
dazol-1-yl)acetate (23a, 134 mg, 0.5 mmol, 1 equiv) as described
above. Yield: 88 mg (66%); ESI-MS: m/z 268.3 [M+H]+; HPLC: tR =
1
5.96 min (>99%); H NMR (400 MHz, [D6]DMSO): δ=4.95 (s, 1.6 H),
Flexible alignment: The flexible alignments were performed with
Molecular Operating Environment (MOE), v2018.01; (Chemical
Computing Group). Compound 1a was extracted from the PDB file
of the co-crystal structure with meprin β (internal communication;
PDB ID: 7AQ1) The carboxylic acids were deleted manually and the
hydroxamic acid and the benzyl residue targeting the S1’ pocket
were constrained by fixing them. The alignments were performed
using the standard parameters and the Amber10:EHT forcefield.
5.29 (s, 0.4 H), 7.44–7.51 (m, 2H), 7.64–7.71 (m, 4H), 7.78–7.84 (m,
1H), 7.88–7.90 (m, 2H), 10.62 (br s, 0.2H), 11.10 (br s, 0.8 H) mixture
of cis-trans isomers; 13C NMR (176 MHz, [D6]DMSO): δ=45.9, 112.4,
117.4, 125.1, 129.6, 129.8, 129.9, 130.2, 132.1, 134.9, 136.0, 152.8,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
+
158.9, 163.3; HRMS m/z: 268.1092 [M+H]+, calcd for C15H14N3O2
:
268.1081.
3-(2-Phenylbenzimidazol-1-yl)propanehydroxamic acid (4d): The
compound was synthesized from methyl 2-(2-phenyl-1H-benzimi-
dazol-1-yl)propanoate (23b, 88 mg, 0.31 mmol, 1 equiv) as de-
scribed above. Yield: 33 mg (37%); ESI-MS: m/z 282.4 [M+H]+;
Docking: All dockings were performed with GOLD v5.6.1 in
combination with the HERMES visualizer. The docking target
meprin
β was extracted from the crystal structure (internal
1
HPLC: tR =6.61 min (95.7%); H NMR (400 MHz, [D6]DMSO): δ=2.59
communication; PDB ID: 7AQ1). For dockings only chain A and the
corresponding zinc ion were used (monomer). For the targets
meprin α and ovastacin the recently reported homology models
were employed.[13,14] Compounds 2a, 2b, 3a, 3b, 4a and 4b were
docked to a 20 Å radius around the catalytic zinc ion. The
hydroxamic acid was substructure constrained, based on the
orientation of the co-crystalized hydroxamate of compound 1a.
Therefore, a minimum separation (1.5 Å), a maximum separation
(3.5 Å) and a spring constant of 5 between the zinc ion and the
complexing atoms O4 and O01 of the hydroxamate were defined.
All poses were optimized within 10 GA runs and scored with
Chemscore®. The search efficacy was set to 200%.
3
3
(t, 2H, J=7.4 Hz), 4.60 (t, 2H, J=7.7 Hz), 7.49–7.57 (m, 2H), 7.67–
7.74 (m, 3H), 7.80–7.83 (m, 1H), 7.87–7.89 (m, 2H), 7.93–7.95 (m, 1H),
10.05 (br s, 0.1H), 10.50 (br s, 0.9H) mixture of cis-trans isomers; 13C
NMR (176 MHz, [D6]DMSO): δ=32.1, 42.4, 113.2, 116.7, 125.6, 129.6,
130.7, 132.2, 134.3, 151.9, 158.5, 158.9, 166.1; HRMS m/z: 282.1248
[M+H]+, calcd for C16H16N3O2+: 282.1237.
2-(2-Phenyl-1H-indol-3-yl)ethanehydroxamic acid (4e): The com-
pound was synthesized from methyl 2-(2-phenyl-1H-indol-3-yl)
acetate (25, 89 mg, 0.34 mmol, 1 equiv) as described above. Yield:
14 mg (15%); ESI-MS: m/z 267.2 [M+H]+, 206.1 [M–CH2NO2]+;
HPLC: tR =11.97 min (>99%); 1H NMR (400 MHz, [D6]DMSO): δ=
3
3
3.51 (s, 1.9H), 3.85 (s, 0.1H), 7.02 (t, 1H, J=7.6 Hz), 7.12 (t, 1H, J=
Enzymatic assays: Recombinant human meprin β was expressed in
yeast and characterized as previously described.[24] Recombinant
human meprin α was expressed and purified from insect cells (S2)
and characterized analogously. MMP2, 9, 13, ADAM10 and 17 were
purchased from a commercial vendor (R&D systems). MMPs were
activated prior to measurement by APMA (p-aminophenylmercuric
acetate) treatment according to manufacturer‘s instructions.
3
3
7.5 Hz), 7.36–7.42 (m, 2H), 7.51 (t, 2H, J=7.7 Hz), 7.61 (d, 1H, J=
3
7.8 Hz), 7.87 (d, 2H, J=7.8 Hz), 10.77 (s, 1H), 11.26 (s, 1H) mixture
of cis-trans isomers; 13C NMR (176 MHz, CD3OD: δ 29.0, 103.8, 110.7,
118.1, 119.4, 121.6, 127.4, 128.0, 128.4, 128.9, 132.8, 136.4, 136.7,
+
170.3; HRMS m/z: 289.0959 [M+Na]+, calcd for C16H14N2NaO2
:
289.0947.
2-(2-Phenylimidazo[4,5-b]pyridin-3-yl)ethanehydroxamic
acid
The determination of enzymatic activity was based on the cleavage
of internally quenched peptide substrates (see supporting informa-
tion). A typical assay of 250 μL total volume measured in black 96-
well plates consisted of 100 μL buffer, 50 μL enzyme at a final
concentration of 9×10À 10 M for meprin α and 3×10À 10 M for meprin
β, 50 μL substrate (in buffer, 0.05% DMSO) and 50 μL inhibitor
solution (in buffer, 2% DMSO). In case of 125 μL assay volume
(black 96 half area well plates) all volumes were cut in half.
Enzymatic activity of ADAMs was measured in 384 well plates with
60 μL total assay volume consisting of 20 μL inhibitor or 20 μL
buffer, 20 μL enzyme and 20 μL substrate. To ensure reproducibility,
the parameters were determined at least in duplicates independ-
ently on separate days. For Kiapp values the influence of 12 inhibitor
(4f): The compound was synthesized from methyl 2-(2-phenyl-3H-
imidazo[4,5-b]pyridin-3-yl)acetate (28, 90 mg, 0.34 mmol, 1 equiv)
as described above. Yield: 35 mg (39%); ESI-MS: m/z 269.1 [M+H]+;
1
HPLC: tR =6.75 min (>99%); H NMR (400 MHz, [D6]DMSO): δ=4.92
(s, 1.6H), 5.26 (s, 0.4H), 7.36–7.39 (m, 1H), 7.58–7.61 (m, 3H), 7.78–
7.80 (m, 0.5H), 7.87–7.89 (m, 1.5H), 8.14–8.16 (dd, 1H, 3J=7.8 Hz,
4J=1.2 Hz), 8.38–8.89 (m, 1H), 9.53 (br s, 0.2H), 10.45 (s, 0.2H), 11.01
(s, 0.8H) mixture of cis-trans isomers; 13C NMR (176 MHz, [D6]DMSO):
δ=43.3, 119.5, 127.2, 129.3, 129.6, 129.8, 130.9, 134.7, 144.6, 148.8,
+
154.9, 164.2; HRMS m/z: 269.1046 [M+H]+, calcd for C14H13N4O2
:
269.1033.
2-(2-Phenylimidazo[4,5-b]pyridin-1-yl)ethane-hydroxamic
acid
concentrations ranging from 2×10À 4 to 1×10À 10
M on the
(4g): The compound was synthesized from methyl 2-(2-phenyl-1H-
imidazo[4,5-b]pyridin-1-yl)acetate (31, 185 mg, 0.69 mmol, 1 equiv)
as described above. Yield: 38 mg (20%); ESI-MS: m/z 269.1 [M+H]+;
enzymatic activity was investigated in the presence of one standard
substrate concentration (10 μM). Initial velocities were determined
using a fluorescence plate reader (CLARIOstar, BMG Labtech) at
1
HPLC: tR =6.61 min (>99%); H NMR (400 MHz, [D6]DMSO): δ=5.54
°
30 C. The excitation/emission wavelength was 340/410 nm. The
3
(s, 1.4H), 5.86 (s, 0.6H), 7.69–7.71 (m, 3H), 7.98 (t, 1H, J=7.1 Hz),
kinetic data was evaluated using GraphPad Prism (version 5.04,
GraphPad Software). Kinetic parameters of inhibition (Kiapp) were
determined using Morrison’s equation.[25]
8.32–8.34 (m, 2H), 8.73–8.79 (m, 2H), 9.32 (br s, 0.7H), 9.84 (br s,
0.3H), 10.80 (s, 0.3H), 11.27 (s, 0.7H) mixture of cis-trans isomers; 13C
NMR (176 MHz, [D6]DMSO): δ=53.7, 118.6, 120.1, 128.6, 129.4,
130.0, 133.3, 139.7, 149.4, 158.6, 158.8, 159.1, 162.0; HRMS m/z:
269.1028 [M+H]+, calcd for C14H13N4O2+: 269.1033.
The inhibitory activity against mouse ovastacin was determined
in vitro by means of a fluorogenic enzyme activity assay as
previously described.[26] Ovastacin was expressed as previously
described[27] and activated by human plasmin (Haematologic
Technologies Inc., Essex Junction, USA) for 30 min in a molar ratio
of 10:1. Concentration of active ovastacin (1 nM) was determined
2-(2-Phenylindazol-3-yl)ethanehydroxamic acid (4h): The com-
pound was synthesized from methyl 2-(2-phenyl-2H-indazol-3-yl)
acetate (33, 32 mg, 0.12 mmol, 1 equiv) as described above. Yield:
18 mg (56%); ESI-MS: m/z 268.1 [M+H]+; HPLC: tR =9.49 min (>
ChemMedChem 2020, 15, 1–14
www.chemmedchem.org
11
© 2020 The Authors. ChemMedChem published by Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cmdc.202000822
ChemMedChem
via titration and IC50 calculation with heterologously expressed
murine fetuin-B[9]. All assays were performed as independent
Keywords: heteroaromatics
metalloproteinases · ovastacin · scaffold hopping
·
hydroxamate
·
meprin
·
1
2
3
4
5
6
7
8
9
°
double measurements in triplicate at 37 C in 100 μl final volume,
buffered with 150 mM NaCl, 50 mM Tris/HCl, pH 7.4, 0.01% Brij-35.
All hydroxamate inhibitors were dissolved in dimethyl sulfoxide.
Enzyme activity measurements were started by addition of 25 μM
[1] F. X. Gomis-Rüth, S. Trillo-Muyo, W. Stöcker, Biol. Chem. 2012, 393,
1027–104.
Ac-R-E(Edans)-D-R-Nle-V-G-D-D-P-Y-K(Dabcyl)-NH2
(Km =34�
[2] a) E. D. Turtle, W.-B. Ho, Expert Opin. Ther. Pat. 2005, 14, 1185–1197; b) E.
Turtle, N. Chow, C. Yang, S. Sosa, U. Bauer, M. Brenner, D. Solow-
Cordero, W.-B. Ho, Bioorg. Med. Chem. Lett. 2012, 22, 7397–7401; c) M.
Talantikite, P. Lécorché, F. Beau, O. Damour, C. Becker-Pauly, W.-B. Ho, V.
Dive, S. Vadon-Le Goff, C. Moali, FEBS Open Bio 2018, 8, 2011–2021.
[3] a) V. Biasin, M. Wygrecka, L. M. Marsh, C. Becker-Pauly, L. Brcic, B.
Ghanim, W. Klepetko, A. Olschewski, G. Kwapiszewska, Sci. Rep. 2017, 7,
39969; b) C. Broder, P. Arnold, S. Vadon-Le Goff, M. A. Konerding, K.
Bahr, S. Muller, C. M. Overall, J. S. Bond, T. Koudelka, A. Tholey, D. J. S.
Hulmes, C. Moali, C. Becker-Pauly, Proc. Natl. Acad. Sci. USA 2013, 110,
14219–14224; c) D. Kronenberg, B. C. Bruns, C. Moali, S. Vadon-Le Goff,
E. E. Sterchi, H. Traupe, M. Böhm, D. J. S. Hulmes, W. Stöcker, C. Becker-
Pauly, J. Invest. Dermatol. 2010, 130, 2727–2735; d) J. Prox, P. Arnold, C.
Becker-Pauly, Matrix Biol. 2015, 44–46, 7–13.
[4] a) T. Bedau, F. Peters, J. Prox, P. Arnold, F. Schmidt, M. Finkernagel, S.
Köllmann, R. Wichert, A. Otte, A. Ohler, M. Stirnberg, R. Lucius, T.
Koudelka, A. Tholey, V. Biasin, C. U. Pietrzik, G. Kwapiszewska, C. Becker-
Pauly, FASEB J. 2017, 31, 1226–1237; b) O. Breig, M. Yates, V. Neaud, G.
Couchy, A. Grigoletto, C. Lucchesi, J. Prox, J. Zucman-Rossi, C. Becker-
Pauly, J. Rosenbaum, Oncotarget 2017, 8, 7839–7851; c) D. Lottaz, C. A.
Maurer, A. Noël, S. Blacher, M. Huguenin, A. Nievergelt, V. Niggli, A.
Kern, S. Müller, F. Seibold, H. Friess, C. Becker-Pauly, W. Stöcker, E. E.
Sterchi, PLoS One 2011, 6, e26450; d) D. Lottaz, C. A. Maurer, D. Hahn,
M. W. Büchler, E. E. Sterchi, Cancer Res. 1999, 59, 1127–1133; e) P.
Minder, E. Bayha, C. Becker-Pauly, E. E. Sterchi, J. Biol. Chem. 2012, 287,
35201–35211; f) H. Schäffler, W. Li, O. Helm, S. Krüger, C. Böger, F.
Peters, C. Röcken, S. Sebens, R. Lucius, C. Becker-Pauly, P. Arnold, J. Cell
Sci. 2019, 132, jcs220665; g) X. Wang, J. Chen, J. Wang, F. Yu, S. Zhao, Y.
Zhang, H. Tang, Z. Peng, BMC Cancer 2016, 16, 383.
2.2 μM), dissolved in dimethyl sulfoxide (final concentration 1.4%).
Initial velocities were recorded for at least 1000 s (50×100 ms at
intervals of 20 s). Thereafter, 1.5 μl of proteinase K (at 20 mg/mL;
Sigma–Aldrich) were added to reach complete substrate turnover,
which was monitored and subsequently calculated using the
formula v=[S]×m/ΔF, where [S] is the substrate concentration, m
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
~
the [F/t] slope of initial linear substrate turnover, and F the
maximal fluorescence intensity corresponding to complete
turnover. Kinetic parameters of inhibition (Kiapp) were determined
using Morrison’s equation.[25]
Cell-viability assay: Cell viability was assessed in human hepatocel-
lular carcinoma cell line Hep-G2 and in human neuroblastoma cell
line SH-SY5Y. Hep-G2 cells were cultivated in RPMI1640 (Thermo-
Fisher) supplemented with 10% FBS and SH-SY5Y cells were
cultivated in DMEM (high-glucose, pyruvate; ThermoFisher) also
°
supplemented with 10% FBS in a humidified atmosphere of 37 C
and 5% CO2 (Hep-G2) or 10% (SH-SY5Y) according to standard cell
culture procedures. For the assay, cells were plated in 96-well
microtiter plates (Greiner bio-one) at densities of 50000 cells/well
(Hep-G2) and 60000 cells/well (SH-SY5Y). After 24 h, compounds
dissolved in DMSO are added to fresh medium without FBS at a
concentration of 30 μM (final concentration of DMSO: 1%, v/v) and
applied to the cells for another 24 h. On the next day, cellular
viability is determined using the CytoTox-ONE kit (Promega) based
on the viability in control wells incubated with culture medium and
1% DMSO. Cells were washed twice with PBS, followed by cell lysis
for 10 min (PBS+9%w/v Triton-X 100). Substrate mix was added
and incubated for 10 min in the dark. After stopping the reaction
the fluorescence intensity was measured (λex =544, λem =595 nm).
[5] a) S. Carmago, S. V. Shah, P. D. Walker, Kidney Int. 2002, 61, 959–966;
b) C. Herzog, R. Seth, S. V. Shah, G. P. Kaushal, Kidney Int. 2007, 71,
1009–1018; c) G. P. Kaushal, R. S. Haun, C. Herzog, S. V. Shah, Am. J.
Physiol. Ren. Physiol. 2013, 304, F1150–8; d) B. Oneda, N. Lods, D. Lottaz,
C. Becker-Pauly, W. Stöcker, J. Pippin, M. Huguenin, D. Ambort, H.-P.
Marti, E. E. Sterchi, PLoS One 2008, 3, e2278.
[6] a) S. Banerjee, J. S. Bond, J. Biol. Chem. 2008, 283, 31371–31377; b) S.
Banerjee, G. Jin, S. G. Bradley, G. L. Matters, R. D. Gailey, J. M. Crisman,
J. S. Bond, Am. J. Physiol. Gastrointest. Liver Physiol. 2011, 300, G273–82.
[7] a) J. Bien, T. Jefferson, M. Causevic, T. Jumpertz, L. Munter, G. Multhaup,
S. Weggen, C. Becker-Pauly, C. U. Pietrzik, J. Biol. Chem. 2012, 287,
33304–33313; b) C. Schönherr, J. Bien, S. Isbert, R. Wichert, J. Prox, H.
Altmeppen, S. Kumar, J. Walter, S. F. Lichtenthaler, S. Weggen, M.
Glatzel, C. Becker-Pauly, C. U. Pietrzik, Mol. Neurodegener. 2016, 11, 19;
c) C. Becker-Pauly, C. U. Pietrzik, Front. Mol. Neurosci. 2017, 9, 159.
[8] a) A. D. Burkart, B. Xiong, B. Baibakov, M. Jiménez-Movilla, J. Dean, J. Cell
Biol. 2012, 197, 37–44; b) H. Körschgen, M. Kuske, K. Karmilin, I.
Yiallouros, M. Balbach, J. Floehr, D. Wachten, W. Jahnen-Dechent, W.
Stöcker, Mol. Hum. Reprod. 2017, 23, 607–616; c) W. Stöcker, K. Karmilin,
A. Hildebrand, H. Westphal, I. Yiallouros, R. Weiskirchen, E. Dietzel, J.
Floehr, W. Jahnen-Dechent, Biol. Chem. 2014, 395, 1195–1199.
[9] K. Karmilin, C. Schmitz, M. Kuske, H. Körschgen, M. Olf, K. Meyer, A.
Hildebrand, M. Felten, S. Fridrich, I. Yiallouros, C. Becker-Pauly, R.
Weiskirchen, W. Jahnen-Dechent, J. Floehr, W. Stöcker, Sci. Rep. 2019, 9,
546.
[10] a) E. S. Pires, R. S. D’Souza, M. A. Needham, A. K. Herr, A. A. Jazaeri, H. Li,
M. H. Stoler, K. L. Anderson-Knapp, T. Thomas, A. Mandal, A. Gougeon,
C. J. Flickinger, D. E. Bruns, B. A. Pollok, J. C. Herr, Oncotarget 2015, 6,
30194–30211; b) K. A. Knapp, E. S. Pires, S. J. Adair, A. Mandal, A. M.
Mills, W. C. Olson, C. L. Slingluff, J. T. Parsons, T. W. Bauer, T. N. Bullock,
J. C. Herr, Oncotarget 2018, 9, 8972–8984.
[11] D. Ramsbeck, A. Hamann, D. Schlenzig, S. Schilling, M. Buchholz, Bioorg.
Med. Chem. Lett. 2017, 27, 2428–2431.
[12] D. Ramsbeck, A. Hamann, G. Richter, D. Schlenzig, S. Geissler, V. Nykiel,
H. Cynis, S. Schilling, M. Buchholz, J. Med. Chem. 2018, 61, 4578–4592.
[13] K. Tan, C. Jäger, D. Schlenzig, S. Schilling, M. Buchholz, D. Ramsbeck,
ChemMedChem 2018, 13, 1619–1624.
Acknowledgements
We gratefully acknowledge Antje Hamann and Mercedes Scharfe
(IZI-MWT), Dr. Andrea Porzel (Leibniz Institute of Plant Biochemis-
try, Halle), Dr. Christoph Wiedemann (Martin Luther University,
Halle–Wittenberg) and Dr. Christian Ihling (Martin Luther Univer-
sity, Halle–Wittenberg) for their excellent technical support. We
also like to thank Dr. Miriam Linnert (IZI-MWT) and Dr. Christoph
Parthier (Martin Luther University, Halle–Wittenberg) for providing
the crystal structure of meprin β, which will be published
elsewhere. Parts of this work were supported by a grant from
European Development Fund, #ZS/2019/02/97143 to D.R. Open
access funding enabled and organized by Projekt DEAL.
Conflict of Interest
C. J. is employee of Vivoryon Therapeutics N.V., Halle (Saale),
Germany. The remaining authors declare no competing inter-
ests.
[14] H. Körschgen, C. Jäger, K. Tan, M. Buchholz, W. Stöcker, D. Ramsbeck,
ChemMedChem 2020, 15, 1499–1504.
ChemMedChem 2020, 15, 1–14
www.chemmedchem.org
12
© 2020 The Authors. ChemMedChem published by Wiley-VCH GmbH
��
These are not the final page numbers!

Full Papers
doi.org/10.1002/cmdc.202000822
ChemMedChem
[15] L. D. Pennington, D. T. Moustakas, J. Med. Chem. 2017, 60, 3552–3579.
[16] W. S. Bechara, I. S. Khazhieva, E. Rodriguez, A. B. Charette, Org. Lett.
2015, 17, 1184–1187.
[17] N. Xi, S. Xu, Y. Cheng, A. S. Tasker, R. W. Hungate, P. J. Reider,
Tetrahedron Lett. 2005, 46, 7315–7319.
[18] Z. Chen, Q. Yan, Z. Liu, Y. Xu, Y. Zhang, Angew. Chem. Int. Ed. 2013, 52,
13324–13328; Angew. Chem. 2013, 125, 13566–13570.
[19] K. Bahrami, M. M. Khodaei, F. Naali, J. Org. Chem. 2008, 73, 6835–6837.
[20] S. Mikami, S. Nakamura, T. Ashizawa, I. Nomura, M. Kawasaki, S. Sasaki,
H. Oki, H. Kokubo, I. D. Hoffman, H. Zou, N. Uchiyama, K. Nakashima, N.
Kamiguchi, H. Imada, N. Suzuki, H. Iwashita, T. Taniguchi, J. Med. Chem.
2017, 60, 7677–7702.
[23] S. Cai, S. Lin, X. Yi, C. Xi, J. Org. Chem. 2017, 82, 512–520.
[24] D. Schlenzig, M. Wermann, D. Ramsbeck, T. Moenke-Wedler, S. Schilling,
Protein Expr. Purif. 2015, 116, 75–81.
[25] J. F. Morrison, Biochim. Biophys. Acta 1969, 185, 269–286.
[26] A. Cuppari, H. Körschgen, D. Fahrenkamp, C. Schmitz, T. Guevara, K.
Karmilin, M. Kuske, M. Olf, E. Dietzel, I. Yiallouros, D. de Sanctis, T.
Goulas, R. Weiskirchen, W. Jahnen-Dechent, J. Floehr, W. Stoecker, L.
Jovine, F. X. Gomis-Rüth, IUCrJ 2019, 6, 317–330.
1
2
3
4
5
6
7
8
9
[27] E. Dietzel et al, Dev. Cell 2013, 25, 106–112.
[21] D. Yang, D. Fokas, J. Li, L. Yu, C. Baldino, Synthesis 2005, 2005, 47–56.
[22] R. P. Kale, M. U. Shaikh, G. R. Jadhav, C. H. Gill, Tetrahedron Lett. 2009,
50, 1780–1782.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
Manuscript received: October 20, 2020
Version of record online: ■■■, ■■■■
ChemMedChem 2020, 15, 1–14
www.chemmedchem.org
13
© 2020 The Authors. ChemMedChem published by Wiley-VCH GmbH
��
These are not the final page numbers!

FULL PAPERS
1
2
3
4
5
6
K. Tan, C. Jäger, Dr. H. Körschgen,
Dr. S. Geissler, Dr. D. Schlenzig,
Dr. M. Buchholz, Prof. Dr. W. Stöcker,
Dr. D. Ramsbeck*
7
8
9
1 – 14
Hop–Step–Jump: Conformationally
restricted heteroaromatic astacin pro-
teinase inhibitors were designed in a
scaffold-hopping approach. This rigid-
ification led to an unexpected boost
in activity (up to a factor of 2000)
compared to recently reported
tertiary amines. Hence, these novel
heterocyclic scaffolds represent a big
jump forward in the design and devel-
opment of inhibitors for proteinases
of the astacin family.
Heteroaromatic Inhibitors of the
Astacin Proteinases Meprin α,
Meprin β and Ovastacin Discov-
ered by a Scaffold-Hopping
Approach
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57