Efficient Entropy-Driven Inhibition of Dipeptidyl Peptidase III by Hydroxyethylene Transition-State Peptidomimetics

Article abstract of DOI:10.1002/chem.202102204

Dipeptidyl peptidase III (DPP3) is a ubiquitously expressed Zn-dependent protease, which plays an important role in regulating endogenous peptide hormones, such as enkephalins or angiotensins. In previous biophysical studies, it could be shown that substr

Full text of DOI:10.1002/chem.202102204

Accepted Article
Accepted Article
Title: Efficient entropy-driven inhibition of dipeptidyl peptidase III by
hydroxyethylene transition state peptidomimetics
Authors: Rolf Breinbauer, Jakov Ivkovic, Shalinee Jha, Christian
Lembacher-Fadum, Johannes Puschnig, Prashant Kumar,
Viktoria Reithofer, Karl Gruber, and Peter Macheroux
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.202102204
Link to VoR: https://doi.org/10.1002/chem.202102204
01/2020

10.1002/chem.202102204
Chemistry - A European Journal
FULL PAPER
Efficient Entropy-Driven Inhibition of Dipeptidyl Peptidase III by
Hydroxyethylene Transition-State Peptidomimetics
Jakov Ivkovic,[a] Shalinee Jha,[b] Christian Lembacher-Fadum,[a] Johannes Puschnig,[a] Prashant
Kumar,[c] Viktoria Reithofer,[c] Karl Gruber,[c] Peter Macheroux,*[b] and Rolf Breinbauer*[a]
Dedicated to the memory of Prof. Francois Diederich
Abstract: Dipeptidyl peptidase III (DPP3) is a ubiquitously expressed
Zn-dependent protease, which plays an important role in regulating
endogenous peptide hormones, such as enkephalins or angiotensins.
In previous biophysical studies, we could show that substrate binding
is driven by a large entropic contribution due to the release of water
molecules from the closing binding cleft. Here, we report the design,
synthesis and biophysical characterization of peptidomimetic
inhibitors using for the first time an hydroxyethylene transition-state
mimetic for a metalloprotease. Efficient routes for the synthesis of
both stereoisomers of the pseudopeptide core were developed, which
allowed the synthesis of peptidomimetic inhibitors mimicking the
VVYPW-motif of tynorphin. The best inhibitors inhibit DPP3 in the low
µM range. Biophysical characterization by means of ITC
measurement and X-ray crystallography confirm the unusual entropy-
driven mode of binding. Stability assays demonstrated the desired
stability of these inhibitors, which efficiently inhibited DPP3 in mouse
brain homogenate.
glioblastoma cells,[5] and ovarian malignant tissue.[6] It has been
shown that DPP3 is able to block the ubiquitination of NRF2
(nuclear factor erythroid 2-related factor 2) by competing to
interact with KEAP1 (kelch-like ECH-associated protein 1)
ubiquitin ligase.[7] Recently, Deniau et al. could show that
circulating DPP3 (cDPP3) was elevated in cardiogenic shock
patients and that high levels of cDPP3 were associated with
altered hemodynamics and poor outcomes.[8] Elevated levels of
cDPP3 were found in a cohort study of critically ill sepsis patients
with the concentrations corresponding to the severity of the
disease. Septic shock patients showed significantly higher levels
of cDPP3 compared to patients with severe sepsis.[9] In an
experimental model of sepsis it could be shown, that the inhibitory
cDPP3-antibody Procizumab restored altered cardiac function
during sepsis in rats.[10]
These biological observations indicate that DPP3 plays an
important role in several pathophysiological disease states. In
order to study its chemical biology and validate it as a drug target
a specific small molecule inhibitor would be highly desirable.[11]
Mammalian DPP3 is significantly inhibited by different small
molecules that are potent covalently binding cysteine peptidase
inhibitors, such as various organomercury compounds,[12-13] and
multiple covalently binding serine peptidase inhibitors like
Introduction
Dipeptidyl peptidase III (DPP3), also known as
enkephalinase B, is a Zn-dependent metalloprotease, which is
ubiquitously expressed by human, bovine, porcine, monkey, rat,
insect, yeast and other organisms. DPP3 cleaves the first two
amino acids at the N-terminus of biologically important
oligopeptides ranging from 3-10 amino acids.
Over the last decades, several reports have been published
characterizing the biological functions of DPP3. DPP3 was found
to have a very high affinity towards angiotensins and enkephalins,
suggesting a role in regulating enkephalin and angiotensin
signaling.[1-2] In fact, patients suffering pain show a lower activity
of DPP3 in cerebrospinal fluid (CSF) compared to patients without
pain.[3] Literature reports an unusually high concentration of DPP3
in cancer cells, such as squamous cell lung carcinoma,[4]
phenylmethanesulfonylfluoride
phosphate (DFP), 3,4-dichloroisocoumarin (DCI) or tosyl-
phenylalanyl chloromethyl ketone The
(TPCK).[13-15]
(PMSF),
diisopropylfluoro-
naphthoquinone natural products fluostatins A and B isolated
from Streptomyces sp. TA-3391 are very potent competitive
inhibitors of placental DPP3. With arginyl-arginine-2-
naphthylamide (H-Arg-Arg-NA) as a synthetic substrate, the
naphthoquinone natural products fluostatins A and B exhibited
IC50 values of 1.44 and 74.0 μM, respectively.[16] However the
quinone moiety is considered as a “pan-assay-interference
compounds” (PAINS) fragment.[17] To date, the strongest
described “inhibitors” of DPP3 are oligopeptides. Yamamoto et
al.[18] and Nishimura et al.[19] reported the finding and isolation of
very potent and selective “inhibitors”, spinorphin (LVVYPWT) and
its truncated form tynorphin (VVYPW), respectively. Tynorphin,
which exhibited selectivity for enkephalin-degrading enzymes, is
slowly hydrolyzed by the enzyme, resulting in complete
degradation of tynorphin in human serum at 37 °C within 4 h.[18]
Therefore, these peptides should be better regarded as slowly-
converted (poor) substrates instead of true inhibitors. Twelve
synthesized amidino benzimidazole compounds were tested as
inhibitors of DPP3, solely on the basis of bioisosterism of amidino
groups to arginine guanidine residues from the DPP3 test
substrate H-Arg-Arg-βNA and were found to inactivate the
enzyme by an irreversible inhibition mechanism.[20] In summary,
none of the above described substances, fulfills the criteria
[a]
Dr. Jakov Ivkovic, Dipl.-Ing. Christian Lembacher-Fadum, Dipl.-Ing.
Johannes Puschnig, Prof. Dr. Rolf Breinbauer
Institute of Organic Chemistry
Graz University of Technology
Stremayrgasse 9, A-8010 Graz, Austria
E-mail: breinbauer@tugraz.at
[b]
[c]
Dr. Shalinee Jha, Prof. Dr. Peter Macheroux
Institute of Biochemistry
Graz University of Technology
Petersgasse 12/2, A-8010 Graz, Austria
peter.macheroux@tugraz.at
Dr. Prashant Kumar, Dr. Viktoria Reithofer, Prof. Dr. Karl Gruber
Institute of Molecular Biosciences
University of Graz
Humboldtstr. 50, A-8010 Graz, Austria
Supporting information for this article is given via a link at the end of
the document.((Please delete this text if not appropriate))
This article is protected by copyright. All rights reserved.

10.1002/chem.202102204
Chemistry - A European Journal
FULL PAPER
expected from a chemical probe,[21] which would allow to validate
DPP3 as a drug target.
The first three amino-terminal amino acid units of the
peptide substrate bind in the form of an extended β-sheet
hydrogen bond network. Importantly, valorphin, tynorphin,
Herein, we report how we used available information of
preferred peptide substrates and from crystal structures of the
open and substrate-bound form of DPP3, produced in our
laboratory, to design a transition-state based inhibitor of DPP3,
which should overcome limitations of previous attempts of DPP3
inhibitors as described above.
tynorphin-like pentapeptides, and angiotensin III share
a
structural motif in their first three amino acid residues. The first
amino acid is variable, but the second and the third are the same
(XVY). This common structure-activity relationship feature could
be an indication of evolutionary training of DPP3 to recognize
such sequences with higher selectivity.
The N-terminus of the peptide substrate is charged and
forms a salt bridge to the side chain of Glu316 of the enzyme
(Figure 1C). It also forms two more hydrogen bonds to the side
chains of Asn394 and Asn391. This very tight interaction most
probably represents an ammonium cation recognition site. The C-
terminal tryptophan residue, along with two hydrogen bonds, adds
also a pincer-like cation-π interaction with Lys670 and Arg669
residues within the enzyme. The cation-π interaction is well
characterized as one of the strongest noncovalent interactions in
protein environments.[24] This additional localized set of tight
interactions in the tynorphin-DPP3 complex is the probable cause
of the higher affinity of binding and the ability to act as an inhibitor,
compared to the other peptide substrates of DPP3 (e.g.
enkephalins, which have Leu or Met as a C-terminal residue, or
endomorphins, which are shorter by one amino acid unit and thus
can hardly interact with this site).[25]
Results and Discussion
Previously, we had solved the structure of an inactive
E451A-variant of hDPP3 in complex with tynorphin (PDB code:
3T6B), which reveals how the peptide substrate binds to this
enzyme (Figure 1).[22] DPP3 has two big lobe-like domains mostly
composed of -helices with a smaller -sheet portion in the lower
domain. Upon ligand binding, the lobes close encapsulating the
substrate completely. This process is accompanied by a large
loss of solvent-accessible area of approximately 3,500 Å2, thereby
releasing approximately 60 structured water molecules, which
represents the entropic driving force of binding overcompensating
the unfavourable, i.e. positive, binding enthalpy of the peptide
substrate (Figure 1A).[22]
The second peptide bond (between P1 and P1’ according to
the Schechter-Berger-nomenclature) is positioned for cleavage
by nucleophilic attack in the wild type enzyme. It is surrounded by
the catalytic apparatus, consisting of four ligands complexing the
Zn-ion (a molecule of water, Glu508, His450 and His 455) and the
Glu451, as well as two additional residues (Tyr318 and His568)
involved in precise substrate positioning and stabilization of the
transition state. Tyr318 has been reported as an important,
conserved residue in the family of DPP3 enzymes. Replacement
of this residue leads to a decrease of the kcat-value by two orders
of magnitude.[22,26] Tyr318 forms hydrogen bonds to the first amide
bond of the peptide ligand and to Glu508, thus bringing together
the catalytic apparatus and the peptide substrate backbone.
His568 apparently has the major role in stabilization of the
transition state. From the X-ray structure of the complex, it can be
predicted to be within hydrogen bonding distance to the carbonyl
oxygen of the cleavable peptide bond, but the interaction is weak,
because lone electron pair orbitals are positioned orthogonally to
the N–H donor of the His568.
A
Asn394
Asn391
The binding mode of substrates to DPP3 occurs through
backbone interactions of the peptide via hydrogen bonds, which
is a common binding mode in proteases and enables a broad
substrate specificity. The domain motion is responsible for
positioning the scissile amide bond correctly to the catalytic centre.
The zinc ion of hDPP3 is tetrahedrally coordinated by His450,
His455 (both from HELLGH motif), Glu508 (from EECRAE motif)
and a water molecule. A glutamate from HELLGH motif (Glu451)
has been proposed to deprotonate a water molecule, which
attacks the peptide bond coordinated to the zinc ion and His568.
After a tetrahedral transition state, the peptide bond is cleaved
(Scheme 1).[22,27]
Glu31
B
C
Figure 1. (A) X-ray structure of hDPP3 without a ligand (PDB: 3FVY) (left). Upon
substrate binding (PDB: 3T6B) (right) large conformational changes occur and
structured H2O molecules (red balls) are released. (B) Binding mode of
tynorphin showing an extended β-sheet. The Zn-ion and the catalytic Glu451
residue were added and force field optimized with molecular modelling software
MOLOC.[23] (C) Amino-terminal ammonium group of the ligand is bound very
tightly via three hydrogen bonds and a salt bridge to Glu316.
This article is protected by copyright. All rights reserved.

10.1002/chem.202102204
Chemistry - A European Journal
FULL PAPER
Scheme 1. Hydrolysis mechanism of DPP3.
In a recent study, we analyzed the structures of complexes
of hDPP3 with enkephalin (Leu/Met), angiotensin II,
endomorphin-2 as well as IVYPW via X-ray crystallography and
observed a difference in coordination of the carbonyl group of the
scissile bond. In the complexes with Leu- and Met-enkephalin, the
zinc ion does not interact directly with the carbonyl group; instead,
a water molecule is coordinated to it. However, in the complex
with IVYPW, the water molecule is missing, but Glu451 has a
smaller distance to the scissile peptide bond. Therefore, Glu451
may also act directly to the bond as a nucleophile forming an acyl-
enzyme-like intermediate (Scheme 1).[28] These two distinct
mechanisms may be the explanation for the open question, why
some peptides are good substrates whereas other peptides act
as “inhibitors”.[28]
From the analysis of our crystallographic structures and
biophysical characterization of the binding of peptide substrates,
we framed the following tenets for inhibitor design: 1) the inhibitor
should be structurally very similar to the best substrates in order
to snuggly fit into the binding pockets inducing the release of
bound water molecules to support entropically driven binding; 2)
in order to take advantage of the enthalpic contribution of
backbone binding, the inhibitor should be peptide based, 3) to be
a true inhibitor the scissile bond between the second and third
amino acid should be replaced by a stable entity, 4) in order to
promote specificity, typical Zn binding motifs known to bind to
metalloproteases such as hydroxamic acids, sulfonamides, etc.
should be avoided,[29] and 5) a transition-state mimetic might
increase the binding affinity.[30] While transition-state mimetics
have been widely and very successfully used for the development
of Ser-, Cys- and Asp-proteases,[31] they have so far only been
applied to thermolysin, as the sole representative of
metalloprotease.[32]
a
Not discouraged by the lack of precedence, we proposed for
our DPP3 inhibitor design to incorporate a hydroxyethylene
moiety instead of the cleavable peptide bond, as a noncleavable
isostere resembling the transition state in peptide bond hydrolysis.
[33b] Hydroxyethylene has a tetrahedral geometry, equivalent to the
geometry of the transition state. Moreover, it has a stable chiral
configuration, and it can be obtained in two different
configurations, both viable for synthesis (Figure 2).
It could be predicted from the representations of both
inhibitors in the binding site, that the (S)-hydroxyethylene could
coordinate to the zinc ion with a lone electron pair from the
hydroxyl substituent. On the other hand, (R)-hydroxyethylene was
expected to form both a coordinative bond with the zinc ion and a
hydrogen bond to His568, very similar to the transition state
This article is protected by copyright. All rights reserved.

10.1002/chem.202102204
Chemistry - A European Journal
FULL PAPER
configuration occurring during peptide hydrolysis. With one
additional major noncovalent interaction, (R)-hydroxyethylene
was expected to have a more favorable enthalpy of binding and
thus a stronger inhibitory effect.
valine (1) was converted to the corresponding chiral aldehyde 2
using our recently developed convenient one-pot two-step
methodology in which the amino acid is activated with Staab´s
reagent (1,1´-carbonyldiimidazole, CDI),[35] and the resulting
intermediate imidazolide is selectively reduced to the aldehyde
using diisobutylaluminium hydride (DIBAL-H).[36] The aldehyde
was isolated in 84% yield and >99% ee by an extractive workup,
requiring no further purification. The amino aldehyde 2 was
reacted with lithiated ethyl propiolate to obtain a mixture of two
diastereomers of propargylic alcohols 3, which were inseparable
at this stage.
The alcohols 3 were catalytically hydrogenated with Pd/C,
and the saturated intermediates were lactonized with catalytic
amounts of p-toluenesulfonic acid. The resulting lactones 4 and 5
were readily separated by flash chromatography. The major
diastereomer was lactone 4 with the desired (S)-configuration at
the stereogenic centre at the lactone. 4 was enolized by
deprotonation with lithium hexamethyldisilazide (LiHMDS)
at -78 °C and alkylated with benzyl bromide. As the approach to
the Si-face of the ring of the enolate lactone is hindered by the
bulky substituent, the electrophile attacked preferentially from the
Re-face. Accordingly, the benzylated lactone 6 was isolated in
61% yield without observation of significant amounts of the
diastereomer that would be formed from the attack of the opposite
face.
With the stereoselectively alkylated lactone 6 in hand, the
pseudodipeptide could be produced by lactone opening for which
several experimental issues had to be very carefully addressed.
Although the lactone was completely opened by stirring for 1 h
with at least 4.0 eq of LiOH in tetrahydrofuran/water (THF/H2O),
the 4-hydroxyacid intermediate appears to be very prone to
spontaneous lactonization. The degree of unwanted lactonization
was found to be highly dependent on the temperature and the
acidity of the workup. The acidification had to be performed
carefully at 0 °C using a 25% aqueous citric acid to adjust the pH
value to 4. Also, it was found that if the temperature of the water
bath used for the evaporation of solvents was >30 °C, the majority
of the intermediate lactonized to the starting material. Further
problems in this reaction were faced due to known issues with the
tert-butyldimethylsilyl (TBS) protection of sterically hindered
secondary alcohols like the 4-hydroxyacid intermediate. Initially,
our efforts to utilize the standard TBSCl/imidazole method[37] were
unsuccessful. When using a large excess of reagents according
to the protocol of Ghosh,[34,38] the reactions were extremely slow,
very low yielding, and irreproducible. Fortunately, a silylation
protocol developed by Stawinski[39] for sterically demanding
Figure 2. Proposed hydroxyethylene transition state mimetics as tynorphin
derived inhibitors of hDPP3.
It could be predicted from the representations of both
inhibitors in the binding site, that the (S)-hydroxyethylene could
coordinate to the zinc ion with a lone electron pair from the
hydroxyl substituent. On the other hand, (R)-hydroxyethylene was
expected to form both a coordinative bond with the zinc ion and a
hydrogen bond to His568, very similar to the transition state
configuration occurring during peptide hydrolysis. With one
additional major noncovalent interaction, (R)-hydroxyethylene
was expected to have a more favorable enthalpy of binding and
thus a stronger inhibitory effect.
We set out to produce both (S)- and (R)-epimers of
hydroxyethylene transition state mimetics of tynorphin. To
facilitate our initial synthetic efforts, we left out the hydroxy residue
of the tyrosine side chain and decided to produce the molecules
containing a pseudo-phenylalanine instead (Figure 2). Using the
main tynorphin scaffold was expected to provide selectivity over
other enkephalinases, as demonstrated in the research on
endogenous peptides inhibiting DPP3.
substrates
using
the
combination
of
TBSCl/N-
methylimidazole/iodine worked successfully for our substrate. In
[34,38]
contrast to literature precedence,
we observed that the
simultaneously formed TBS-ester could not be readily
deprotected by short methanolysis. This could be solved by
performing the methanolysis reaction in the presence of a
catalytic amount of citric acid. The selective cleavage was
completed within 6 h without harming the TBS-ether and furnished
protected pseudopeptide building block 7 in 55% yield.
The synthesis of the central pseudopeptide building block
followed a homoenolate strategy, which had been established by
Ghosh for a slightly different substrate leading to a SARS
protease inhibitor.[34] N-tert-Butyloxycarbonyl (Boc) protected
This article is protected by copyright. All rights reserved.

10.1002/chem.202102204
Chemistry - A European Journal
FULL PAPER
Scheme 2. Synthesis of the (S)-hydroxyethylene transition state mimetic “SHE” (10).
In order to complete the synthesis, the protected
intermediate 7 was coupled with the H-Pro-Trp-OMe dipeptide
fragment using (2-(1H-benzotriazol-1-yl)-1,1,3,3-tetramethyl-
uronium hexafluorophosphate (HBTU) as a coupling reagent
delivering pseudotetrapeptide 8 in 74% isolated yield. As the
hydroxyethylene acid cannot form an oxazolone intermediate, no
racemization was observed in this coupling step.[40] In order to
complete the required pentapeptide-like scaffold, N-Boc
deprotection of the peptide and coupling with Boc-L-valine were
necessary. While the peptide coupling was straightforward the
orchestration of the deprotection strategy turned out to be a
delicate matter. Treatment of 8 with TFA/ethanethiol resulted in
neat and selective N-Boc deprotection, whereas anhydrous HCl
in the protic solvent resulted in rapid dual deprotection of both the
Boc and TBS function, causing also a thermodynamically favored,
acid-catalyzed lactone cyclization “backbite” in the peptide, which
has already been observed by Rich in earlier studies.[33b]
Ultimately, we could solve this issue by the use of ZnBr2 in
This article is protected by copyright. All rights reserved.

10.1002/chem.202102204
Chemistry - A European Journal
FULL PAPER
2,2,2-trifluoroethanol[41] with EtSH as an additive, which led to dual
deprotection of both Boc as well as TBS without generating
byproducts.
Taking advantage of the higher nucleophilicity of amines
compared to alcohol nucleophiles,[42] the N,O-deprotected
pseudotetrapeptide was coupled with Boc-Val-OH to afford 9 in
52% isolated yield (2 steps) with no observed ester coupling
byproduct. The target molecule 10 was easily obtained after
saponification of the C-terminal methyl ester with LiOH and N-Boc
deprotection with ZnBr2 and EtSH. Since the (S)-hydroxyethylene
transition state isostere is contained in 10 we assigned an
arbitrary abbreviation “SHE” to this final molecule.
Scheme 3. Synthesis of (R)-hydroxyethylene pseudopeptide 18.
The synthesis route for the opposite hydroxyethylene
diastereomer used methodology developed by Rich,[33a] who
reported the synthesis of a Gln-Phe hydroxyethylene dipeptide
isostere as a precursor for proposed BoNT metalloprotease
inhibitors. Our synthetic route started from N-Boc-protected L-
valine (1), which was converted into methyl ester 11 by using
methyl iodide and potassium bicarbonate in N,N-
dimethylformamide (DMF) in 99% yield and excellent purity
without requiring further purification.[43-44] In the next step,
dimethyl methylphosphonate was lithiated with n-butyllithium to
generate lithiated compound 11a, which was allowed to react with
11 to β-keto phosphonate 12 at -78 °C.[45] Deprotonation of 12 was
Stereoselective reduction of 15 with lithium tri-tert-
butoxyaluminum hydride
at
a
temperature optimum
between -40 °C to -30 °C to minimize formation of the unwanted
(S,S)-diastereomer and required 20 h for completion. The
resulting amino alcohol intermediate 15a could be easily
lactonized by addition of a catalytic amount of p-toluenesulfonic
acid in toluene and stirring at 60 °C for 12 h. The major part of the
desired (R,S)-γ-lactone 16 could be obtained in a very pure form
just by precipitation induced by addition of n-hexane onto the
crude oil. An additionally performed flash chromatography of the
concentrated filtrate increased the yield of 16 to an acceptable
55%. The absolute configuration of the stereogenic centers could
be unambiguously assigned via X-ray crystallography.
achieved with NaH and underwent a HWE reaction with freshly
[33a,46-47]
prepared methyl glyoxylate (13)
at -30 °C. The crude
With γ-lactone 16 in hand the O-protected pseudodipeptide
core molecule was synthesized by a multi-step sequence leading
to the -alkylation product, followed by opening of the lactone and
mixture of cis- and trans-isomers 14 was then hydrogenated with
Pd/C to produce keto ester 15 in 80% yield over two steps.
This article is protected by copyright. All rights reserved.

10.1002/chem.202102204
Chemistry - A European Journal
FULL PAPER
immediate protection of the hydroxy moiety. For the
stereoselective alkylation a four-step sequence according to
Nadin was used.[48] In the first step, γ-lactone 16 was enolized by
deprotonation with lithium diisopropylamide (LDA) at -78 °C in
THF, and allowed to react with freshly distilled benzaldehyde to
form the corresponding β-hydroxy lactone 16a as a complex
mixture of diastereomers in 71% yield. Subsequent treatment with
stereoselectivity for the production of 17.[33a] For the delicate ring
opening of lactone 17, we could build on the experiences made in
the synthesis of SHE (10) in order to avoid degradative
lactonization. After opening of the lactone with LiOH in THF/H2O
= 2:3 (v/v) at rt, the reaction mixture was cooled to 0 °C and
acidified to pH=4 by the careful addition of 25% citric acid. After
aqueous workup the solvents were removed carefully at a
temperature below 30 °C to avoid degradative lactonization.
Immediately after drying of the crude product the hydroxy moiety
methanesulfonic anhydride, followed by
a
Et3N-induced
elimination resulted in an α,β-unsaturated lactone 16b. The crude
16b was then hydrogenated with Raney®-Nickel as hydrogenation
in previous attempts with palladium on charcoal had failed.
Adsorption of the lactone substrate to the metal catalyst occurs
with the less hindered face explaining the observed high
was
TBS-protected
with
TBSCl/I2/N-methylimidazole.[39]
Methanolysis with a catalytic amount of citric acid converted the
resulting TBS-ester[34] into the desired pseudodipeptide core
molecule 18 in 85% yield over 3 steps.
Scheme 4. Synthesis of (R)-hydroxyethylene inhibitor “HER” (23).
Now the core structure was ready to be elongated at both
the C- and N-terminus using similar steps as developed for the
synthesis of SHE. 18 was coupled with dipeptide H-Pro-Trp-OMe
with HBTU/i-Pr2NEt (DIPEA) as coupling agent. After purification
of the reaction mixture via flash chromatography, a significant
amount of tetramethylurea from peptide coupling remained, which
could be removed by extensively washing with H2O.
Pseudopentapeptide 19 was isolated as a white solid in 73% yield
in sufficient quantities, which gave us the opportunity to split the
bulk for the synthesis of additional inhibitors. For the next peptide
coupling the Boc-group of 19 was first removed with TFA and then
Boc-Val-OH was attached with O-(7-azabenzotriazol-1-yl)-
N,N,N',N'-tetramethyluronium hexafluorophosphate (HATU) as a
coupling reagent to avoid the risk of epimerization. Again,
tetramethylurea had to be removed after flash chromatography by
several H2O washing steps delivering 20 in 64% yield.
Unfortunately, the final deprotection steps turned out to be more
challenging than expected. The outcome of the first attempt,
saponification of the methyl ester and subsequent simultaneous
deprotection of both the Boc and TBS group with TFA, resulted in
acid catalyzed γ-lactone “backbite” and cleavage of the
molecule.[33b] Hence, an alternative deprotection strategy had to
be found. HF/pyridine[49-51] was used as a milder deprotecting
agent, which afforded the selective cleavage of TBS-ether in a
first step to obtain 21 in 36% yield after flash chromatography.
Despite careful reaction control, partial cleavage of the molecule
due to the formation of the lactone could not be avoided.
Saponification of 21 with lithium hydroxide and subsequent
treatment of crude 22 with zinc bromide and ethanethiol in
This article is protected by copyright. All rights reserved.

10.1002/chem.202102204
Chemistry - A European Journal
FULL PAPER
2,2,2-trifluoroethanol (TFE) provided the final target molecule
“HER” (23) in 44% yield (2 steps) after preparative HPLC.
signatures of the HIV-protease inhibitors Indinavir and
Nelfinavir.[54]
We were fortunate to crystallize a SHE/hDPP3 (E451A-
variant) complex and determine its structure with X-ray
crystallography (Figure 4A and 4B). The binding mode of SHE can
be almost perfectly aligned with the binding mode of tynorphin in
the previously obtained structure of tynorphin-hDPP3 complex,
the main differences being around the catalytic zinc-complex. The
C-terminal part of the ligand is bound in a cation-π complex of the
indole moiety of SHE with Lys670 and Arg669 of hDPP3. Arg669
also forms a salt bridge to the carboxylate group of SHE. The
equivalence to the tynorphin-hDPP3 complex is also apparent at
the N-terminus of SHE, which is also tightly bound by the same
three hydrogen bonds (Glu316, Asn394, Asn391) and the salt
bridge to the Glu316.
A
B
C
D
Figure 3. ITC measurement curves of DPP3 with (A) HER with Kd = 11 ± 1 µM
and (B) SHE with Kd = 23 ± 4 µM. The bottom panel shows the thermodynamic
signatures for (C) HER and (D) SHE derived from the ITC experiment. The
mean ΔG, ΔH, -TΔS values for HER are -5.3, 10.7 and -16.0 kcal/mol
respectively, whereas the ΔG, ΔH, -TΔS values for SHE are -5.8, 2.3 and -8.1
kcal/mol respectively. The data represents mean of three independent
measurements.
Glu508
His450
His455
His568
H2O
Glu451
Tyr318
E
With the two epimers of our target structures in hand, we
could set out to characterize their potential as inhibitors of DPP3.
Inhibition potencies of both SHE (10) and HER (23) were
investigated via fluorescence-based competitive inhibition assay
of degradation of the H-Arg-Arg-βNA substrate. IC50 values were
calculated based on the resulting dose response curves. Both
transition state mimetics inhibited hDPP3. SHE inhibited the
enzyme with an IC50 of 98.5 µM, while inhibition with HER resulted
in an IC50 of 13.8 µM, making it 7-fold more potent than SHE.
Importantly, HER could also efficiently inhibit DPP3 in mouse
brain homogenate without losing much of its potency (Figure SI1).
The inhibitory capacities of these compounds were corroborated
by ITC measurements, where an endothermic binding was
observed for both SHE and HER, similar to tynorphin. While SHE
could bind to hDPP3 in-vitro with an affinity of Kd = 23 µM, the
binding of HER to hDPP3 was stronger with a Kd = 11 µM (Figure
3). The thermodynamic signatures depicted in Figure 3 show that
both SHE and HER are compounds, which show that the
unfavorable enthalpy of binding is overcompensated by strongly
entropy-driven binding.[52-53] Such behavior is very rarely observed,
but has been reported previously for the thermodynamic
F
2.7 Å
3.2 Å
2.0 Å
2.7 Å
H2O
Figure 4. (A) Overall structure of the complex of human DPP3 (E451A-variant)
with the (S)-hydroxyethylene-peptidomimetic SHE. The two lobes of the enzyme
are shown in light blue and brown, the bound ligand is depicted in a balls-and-
sticks representation with carbon atoms colored yellow. The zinc ion in the
active site is highlighted as a blue sphere. (B) Polder omit-map[56] around the
bound inhibitor contoured at 3. The ligand is shown as a sticks-representation
and the electron density map is depicted a blue mesh. (C) Binding mode of SHE
in the binding site of the inactive E451A mutant of hDPP3. (D) Binding mode of
tynorphin in the binding site of the inactive E451A mutant of hDPP3.[22] Glu451,
Zn-ion and the water molecule, missing out from the tynorphin-hDPP3 structure,
were computationally added and force field-optimized using MOLOC
software.[23] (E) Initially proposed scheme of binding mode of SHE to the active
site of the wild type hDPP3, having the catalytic Glu451 residue. (F) Measured
hydrogen bonding and zinc coordinating interactions around the
hydroxyethylene moiety in the SHE-hDPP3 complex.
This article is protected by copyright. All rights reserved.

10.1002/chem.202102204
Chemistry - A European Journal
FULL PAPER
In contrast to the binding mode of tynorphin, the
hydroxyethylene group in SHE does not interact with His568,
which most probably presents a significant penalty to the enthalpy
of binding (Figure 4C and 4D). On the other hand, based on the
initial rough model of binding of (S)-hydroxyethylene type of
inhibitor (Figure 4E), the hydroxyl substituent complexes as
expected to the zinc ion, which was confirmed by measurement
of the bond length from the crystallographic data. The Zn–O bond
was found to be 1.9 Å (Figure 4F), which falls into the range of
1.9–2.4 Å for values typically observed for the length of Zn–O
coordinating bonds in literature.[55] Another interesting feature is a
water molecule (Figure 4C and 4F) hydrogen bonded to the
hydroxyethylene, which occupies the space where normally the
carboxylate from Glu451 would be positioned within the active
hDPP3 (Figure 4D). Unfortunately, we could not yet achieve an
X-ray structure of the HER/DPP3 complex, as we were not able
to obtain crystals of this complex. However, from the good
concordance of our models with the experimental structures of the
complex with SHE, we are quite confident that also the proposed
structure of the complex with HER in Figure 2 explains the
observed higher affinity of this ligand to DPP3.
In order to address the question, that the hydroxyethylene
isostere also leads to a stable inhibitor of DPP3 we performed a
comparison of time-dependent inhibition of hDPP3 by HER and
tynorphin, which revealed that HER is not degraded by the
peptidase and therefore can efficiently inhibit the activity over a
period of 24 h. In contrast, tynorphin being a slow converting
substrate was purportedly cleaved by hDPP3 and thereby lost its
efficacy within the first hour (Figure 5 top). This was also
confirmed by a thermal stability assay where protein stabilization
due to binding of HER, indicated by an increase in the protein
melting temperature, was retained even after 24 h (Figure 5
bottom).
With these experiments we have shown that HER (23) is a
suitable inhibitor of DPP3, which fulfills the criteria of selectivity
and stability nicely confirming our design rationale. In a next step
we sought out to use the underlying (R)-hydroxyethylene scaffold
for the synthesis of additional variants in which we varied the Phe-
analog of the pseudopeptide core to include the Tyr-motif of the
tynorphin lead structure, as well as F- and CH3-substitutents in
para-position of the Phe-group. The synthesis followed the route
established for HER (30) starting with the aldol condensation of γ-
lactone 11 with the respective benzaldehydes (Scheme 5). While
performing the TBS-protection of compound 26c, an unexpected
side reaction was observed using the silylation procedure of
Stawinski.[39] The presence of iodine during the protection
procedure allowed the iodination of the aromatic ring leading to
an almost 1:1 mixture of 26c and diiodinated isomer 26c*, which
could not be separated by flash chromatography. Fortunately,
when the mixture of compounds 26c* and 27c was treated with
H2, Pd/C in the presence of triethylamine in methanol, the desired
compound 27c was isolated in acceptable 45% yield after an
ultimate four-step sequence (Scheme 5).
Figure 5. Top: Time-dependent inhibition of DPP3 by tynorphin (black
rectangles) and HER (blue circles). The x-axis represents incubation time in
hours and the y-axis represents the percentage of DPP3 activity compared to
control. Bottom: Thermal transitions (upper panel) and melting temperatures of
hDPP3 in the presence and absence of HER (taken at two time points: 0 and
24 h of incubation) determined by Thermofluor™.
With all the pseudodipeptide fragments in hand, the core
structures were now ready for elongation at the C-terminus with
the corresponding dipeptides under standard coupling conditions
leading to the tetrapeptides 28a-c. The subsequent N-terminal
elongation with the characteristic Val residue started with the Boc-
deprotection of the tetrapeptide using TFA, followed by coupling
of the Val residue under standard conditions with HATU as
coupling additive and DIPEA as base in DMF leading to
pentapeptides 30a-b. In the case of the Tyr-pseudotetrapeptide
28c, the aryl silyl ether had first to be cleaved off. Following the
protocol for selective cleavage of phenol silylethers of Lakshman
[57]
deprotection of 28c with KHF2 in methanol led to Tyr-analog
28c* in excellent yield. The already established treatment with
ZnBr2/EtSH in TFE led to the successful tandem deprotection to
compound 29c. Subsequent coupling with Boc-Val-OH provided
desired compound 30c in 36% overall yield. Finally, using the
This article is protected by copyright. All rights reserved.

10.1002/chem.202102204
Chemistry - A European Journal
FULL PAPER
established twofold deprotection strategy saponification and N-
Boc deprotection delivered the target compounds 31a-c, which
showed similar activity in the DPP3-inhibition assay (IC50 values
of 12-15 µM) as has been observed for HER, confirming the
structural significance of the (R)-hydroxyethylene subunit, while
the para-phenyl substituent in this moiety plays only a minor role.
In addition to the already desired compounds, the methyl ester 32
was synthesized, in which only the amino group of 30c was
deprotected under already established conditions. 32 still showed
remarkable inhibition activity of DPP3 (IC50 = 17.9 µM), offering
opportunities for improving bioavailability for this less polar
peptide.
.
Scheme 5. Synthesis of various pseudopeptide analogs and their IC50 values.
In order to study the influence of the Pro on the conformation
and activity of the peptide inhibitors, we synthesized a set of
compounds using the larger pipecolic acid instead of Pro
(Scheme 6). For the coupling of 18 with the Pip-Trp-OMe
dipeptide fragment, the more powerful coupling agent HATU[58]
had to be used instead of HBTU, as the pipecolic acid with its
piperidine moiety is less nucleophilic than proline.[59] This - what
could be believed to be only a – minor change in structure, also
made adaptations in the deprotection strategy necessary, as the
conditions established for the Pro-containing peptide (Scheme 5)
resulted in amide cleavage for the pipecolic containing peptide.
The observed degradative lactonization only led to the isolation of
the lactone and the dipeptide fragment H-Pip-Trp-OMe. In the
need of a new deprotection strategy, we tried different additives
for selective deprotection. Interestingly, HF/pyridine in THF, which
showed being not successful for the silyl ether cleavage of the
Pro-containing compounds 26a and 26b, emerged for the
pipecolic substrate 33 as the reagent of choice and selective and
quick deprotection of the hydroxyl moiety of 33 could be observed
after 1 h, according to NMR. N-Boc-deprotection of 34 was
This article is protected by copyright. All rights reserved.

10.1002/chem.202102204
Chemistry - A European Journal
FULL PAPER
achieved by using TFA in ethanethiol and subsequent coupling
with either Boc-Val-OH or Boc-tert-leucine-OH provided
compounds 36a and 36b in 31% and 42% overall yield
respectively. Saponification of the methyl ester with LiOHxH2O
and removal of the N-Boc group with ZnBr2/EtSH in TFE produced
the pipecolic acid derivatives 37a-b (Scheme 6). In the DPP3
inhibition assay, compound 37a, which can be considered as a
Pip-analog of HER, showed a strongly increased IC50 of 34.4 µM.
Replacing the N-terminal Val with the bulkier tert-butyl leucine
resulted in an even more diminished inhibitory activity (IC50 = 95.0
µM). Together these results suggest that the original design of
HER which was inspired by the amino acid sequence of tynorphin
represents already a good starting point for further optimization
which would profit from a high resolution crystal structure of the
HER/DPP3 complex, where new interactions for improving the
enthalpy of binding could be found or where conformational
restrictions could be introduced which would increase the
favorable entropic contribution of binding[60] even further.
Scheme 6. Synthesis of (R)-hydroxyethylene inhibitor with pipecolic acid and its IC50 values.
DPP3, such as tynorphin, which are degraded by this enzyme
within a short time. These are the first hydroxyethylene transition
state mimetic inhibitors that demonstrably inhibit
Conclusions
a
Starting from the pentapeptide tynorphin (VVYPW), which is
metalloprotease and might provide a guidance that this type of
transition state mimetics should be considered for the future
design of inhibitors of other metalloproteases and -hydrolases.
The insights gained by the characterization of this new class
of inhibitors will provide a starting point for the design of molecular
tools specific for inhibiting hDPP3 and pave the way to exploit this
enzyme as a potential drug target for pain intervention strategies,
control of blood pressure, and treatment of septic shock.
a substrate “inhibitor” slowly converted by DPP3, we designed
pseudopeptide inhibitors of DPP3, directed by observations of
preferred non-covalent interactions in the crystal structure of the
tynorphin-hDPP3 complex. In order to convert tynorphin from
being a slowly converted substrate to a true inhibitor of DPP3, a
non-cleavable hydroxyethylene isostere was used to replace the
scissile peptide bond. We could show that among the two possible
epimers the (R)-hydroxyethylene pseudopeptide showed
significantly better inhibition activity supporting our model about
the active site interaction. Kinetic and thermodynamic
characterization of the resulting inhibitors confirmed their
inhibitory properties on hDPP3 and revealed that these inhibitors
exhibit a strongly entropy driven binding behavior compensating
for a positive binding enthalpy. Such a thermodynamic signature
of binding has only been very rarely been observed before, most
notable for the HIV-protease inhibitors Indinavir and Nelfinavir.
We could also demonstrate the long-term stability of these
peptidomimetics in assays in contrast to peptide “inhibitors” of
Experimental Section
Experimental details describing the synthesis and spectroscopic
characterization of the compounds and details about the biophysical
characterization are described in the Supporting Information. All animal
experiments were approved by the Austrian Federal Ministry for Science,
Research, and Economy (protocol number BMWF-66.007/7-ll/3b/), the
ethics committee of the University of Graz, and conducted in compliance
with the Council of Europe Convention (ETS 123).
This article is protected by copyright. All rights reserved.

10.1002/chem.202102204
Chemistry - A European Journal
FULL PAPER
Deposition
Number
<url
25, 1202-1214; f) C. Diederich, M. Leypold, M. Culka, H. Weber, R.
Breinbauer, G. M. Ullmann, W. Blankenfeldt, Sci. Rep. 2017, 7, 6272.
[12] M. Abramić, D. Schleuder, L. Dolovcak, W. Schröder, K. Strupat, D. Sagi,
J. Peter-Katalini, L. Vitale, Biol. Chem. 2000, 381, 1233-1243.
[13] M. Smyth, G. O'Cuinn, J. Neurochem. 1994, 63, 1439-1445.
[14] K. Fukasawa, K. M. Fukasawa, M. Kanai, S. Fujii, J. Hirose, M. Harada,
Biochem. J. 1998, 329, 275-282.
href="https://www.ccdc.cam.ac.uk/services/structures?id=doi:10.1002/ch
em.20210XXX"> 2098559 (for 16)url> contains the supplementary
crystallographic data for this paper. These data are provided free of charge
by the joint Cambridge Crystallographic Data Centre and
Fachinformationszentrum
http://www.ccdc.cam.ac.uk/structures ">Access Structures service</url>.
Karlsruhe
<url
href="
[15] M. A. Sentandreu, F. Toldrá, J. Agric. Food Chem. 1998, 46, 3977-3984.
[16] T. Akiyama, S. Harada, F. Kojima, Y. Takahashi, C. Imada, Y. Okami, Y.
Muraoka, T. Aoyagi, T. Takeuchi, J. Antibiot. 1998, 51, 553-559.
[17] a) J. B. Baell, G. A. Holloway, J. Med. Chem. 2010, 53, 2719−2740; b) J.
B. Baell, J. W. M. Nissink, ACS Chem. Biol. 2018, 13, 36−44.
Acknowledgements
This work was supported by the doctoral program W901 DK
Molecular Enzymology, funded by the Austrian Science Fund
(FWF) and by NAWI Graz. We thank Ing. Carina Illaszewicz-
Trattner and Prof. Dr. Hansjörg Weber for assistance in NMR
measurements, Prof. Dr. Roland Fischer for providing the X-ray
dataset of γ-lactone 16, and Christian Leypold, Marina Toplak,
Ivana Urošević, and Jovana Trifunović for skillful assistance in the
lab.
[18] Y. Yamamoto, J.-i. Hashimoto, M. Shimamura, T. Yamaguchi, T. Hazato,
Peptides 2000, 21, 503-508.
[19] K. Nishimura, T. Hazato, Biochem. Biophysi. Res. Co. 1993, 194, 713-
719.
[20] D. Agić, M. Hranjec, N. Jajčanin, K. Starčević, G. Karminski-Zamola, M.
Abramić, Bioorg. Chem. 2007, 35, 153-169.
[21] C. H. Arrowsmith, J. E. Audia, C. Austin, J. Baell, J. Bennett, J. Blagg, C.
Bountra, P. E. Brennan, P. J. Brown, M. E. Bunnage, C. Buser-Doepner,
R. M. Campbell, A. J. Carter, P. Cohen, R. A. Copeland, B. Cravatt, J. L.
Dahlin, D. Dhanak, A. M. Edwards, S. V. Frye, N. Gray, C. E. Grimshaw,
D. Hepworth, T. Howe, K. V. Huber, J. Jin, S. Knapp, J. D. Kotz, R. G.
Kruger, D. Lowe, M. M. Mader, B. Marsden, A. Mueller-Fahrnow, S.
Müller, R. C. O'Hagan, J. P. Overington, D. R. Owen, S. H. Rosenberg,
B. Roth, R. Ross, M. Schapira, S. L. Schreiber, B. Shoichet, M.
Sundström, G. Superti-Furga, J. Taunton, L. Toledo-Sherman, C.
Walpole, M. A. Walters, T. M. Willson, P. Workman, R. N. Young, W. J.,
Zuercher, Nat. Chem. Biol. 2015, 11, 536-541.
Keywords: Drug Design • Inhibitor • Medicinal Chemistry •
Metalloprotein • Peptidomimetics
[1]
[2]
C. M. Lee, S. H. Snyder, Biol. Chem. 1982, 257, 12043-12050.
S. Jha, U. Taschler, O. Domenig, M. Poglitsch, B. Bourgeois, M.
Pollheimer, L. M. Pusch, G. Malovan, S. Frank, T. Madl, K. Gruber, R.
Zimmermann, P. Macheroux, J. Biol. Chem. 2020, 295, 13711-13723.
H. Sato, K. Kimura, Y. Yamamoto, T. Hazato, Masui. 2003, 52, 257-263.
B. E. Hast, D. Goldfarb, K. M. Mulvaney, M. A. Hast, P. F. Siesser, F.
Yan, D. N. Hayes, M. B. Major, Cancer Res. 2013, 73, 2199.
A. A. Shukla, M. Jain, S. S. Chauhan, FEBS J. 2010, 277, 1861.
Š. Šimaga, D. Babić, M. Osmak, J. Ilić-Forko, L. Vitale, D. Miličić, M.
Abramić, Eur. J. Cancer 1998, 34, 399.
[22] G. A. Bezerra, E. Dobrovetsky, R. Viertlmayr, A. Dong, A. Binter, M.
Abramic, P. Macheroux, S. Dhe-Paganon, K. Gruber, Proc. Natl. Acad.
Sci. U.S.A. 2012, 109, 6525-6530.
[3]
[4]
[23] P. R. Gerber, K. Moeller, J. Comput. Aided. Mol. Des. 1995, 9, 251–268.
[24] J. C. Ma, D. A. Dougherty, Chem. Rev. 1997, 97, 1303-1324.
[25] S. C. Prajapati, S. S. Chauhan, FEBS Journal 2011, 278, 3256-3276.
[26] B. Salopek-Sondi, B. Vukelić, J. Špoljarić, Š. Šimaga, D. D. Vujaklija, J.
Makarević, N. Jajčanin, M. Abramić, J. Spoljarić, S. Simaga, et al., Biol.
Chem. 2008, 389, 163-167.
[5]
[6]
[7]
a) A. Kobayashi, M.-I. Kang, H. Okawa, M. Ohtsuji, Y. Zenke, T. Chiba,
K. Igarashi, M. Yamamoto, Mol. Cell. Biol. 2004, 24, 7130; b) K. Lu, A. L
Alcivar, J. Ma, T. K. Foo, S. Zywea, A. Mahdi, Y. Huo, T. W. Kensler, M.
L. Gatza, B. Xia, Cancer Res. 2017, 77, 2881-2892; c) for a review of the
KEAP1-NRF2-pathway, see: L. Baird, M. Yamamoto, Mol. Cell. Biol.
2020, 40, e00099-20.
[27] P. K. Baral, N. Jajčanin-Jozić, S. Deller, P. Macheroux, M. Abramić, K.
Gruber, J. Biol. Chem. 2008, 283, 22316-22324.
[28] P. Kumar, V. Reithofer, M. Reisinger, S. Wallner, T. Pavkov-Keller, P.
Macheroux, K. Gruber, Sci. Rep. 2016, 6, 23787.
[29] Reviews: a) J. A. Jacobsen, J. L. M. Jourden, M. T.Miller, S. M.Cohen,
Biochim. Biophys. Acta – Mol. Cell Res. 2010, 1803, 72-94; b) F. E.
Jacobsen, J. A. Lewis, S. M. Cohen, ChemMedChem 2007, 2, 152-171.
[30] Reviews: a) V. L. Schramm, Chem. Rev. 2018, 118, 11194-11258; b) R.
Wolfenden, Bioorg. Med. Chem. 1999, 7, 647-652.
[8]
[9]
B. Deniau, L. Rehfeld, K. Santos, A. Dienelt, F. Azibani, M. Sadoune, P.
R. Kounde, J. L. Samuel, H. Tolpannen, J. Lassus, V.-P. Harjola, N.
Vodovar, A. Bergmann, O. Hartmann, A. Mebazaa, A. Blet, Eur. J. Heart
Fail. 2019, 22, 290-299.
L. Rehfeld, E. Funk, S. Jha, P. Macheroux, O. Melander, A. Bergmann,
J. Appl. Lab. Med. 2019, 3, 943-953.
[31] Reviews about peptidomimetics: a) E. Lenci, A. Trabocchi, Chem. Soc.
Rev. 2020, 49, 3262-3277; b) N. Qvit, S. J. S. Rubin, T. J. Urban, D.
Mochly-Rosen, E. R. Gross, Drug Disc. Today 2017, 22, 454-462; c) I.
Avan, D. C. Hall, A. R. Katritzky, Chem. Soc. Rev. 2014, 43, 3575-3594;
d) A. Grauer, B. Koenig, Eur. J. Org. Chem. 2009, 5099-5111; e) J.
Deska, U. Kazmaier, Curr. Org. Chem. 2008, 12, 355-385; f) J. Vagner,
H. Qu, V. J. Hruby, Curr. Opin. Chem. Biol. 2008, 12, 292-296; g) A. S.
Ripka, D. H. Rich, Curr. Opin. Chem. Biol. 1998, 2, 441-452.
[10] B. Deniau, A. Blet, K. Santos, P. V. Ayar, M. Genest, M. Kästorf, M.
Sadoune, A de Sousa Jorge, J. L. Samuel, N. Vodovar, A. Bergmann,
A. Mebazza, F. Azibani, PLoS One 2020, 15, e0238039.
[11] For tool compounds developed for other protein targets in our laboratory,
see: a) E. G. Ahuja, P. Janning, M. Mentel, A. Graebsch, R. Breinbauer,
W. Hiller, B. Costisella, L. S. Thomashow, D. V. Mavrodi, W. Blankenfeldt,
J. Am. Chem. Soc. 2008, 130, 17053-17061; b) M. Mentel, W.
Blankenfeldt, R. Breinbauer, Angew. Chem. 2009, 121, 9248-9251;
Angew. Chem. Int. Ed. 2009, 48, 9084-9087; c) J. M. Krysiak, J. Kreuzer,
P. Macheroux, A. Hermetter, S.A. Sieber, R. Breinbauer, Angew. Chem.
2012, 124, 7142-7147; Angew. Chem. Int. Ed. 2012, 51, 7035-7040; d)
N. Mayer, M. Schweiger, M. Romauch, G. F. Grabner, T. O. Eichmann,
E. Fuchs, J. Ivkovic, C. Heier, I. Mrak, A. Lass, G. Höfler, C. Fledelius, R.
Zechner, R. Zimmermann, R. Breinbauer, Nat. Chem. Biol. 2013, 9, 785-
787; e) M. Trobe, M. Peters, S. H. Grimm, R. Breinbauer, Synlett 2014,
[32] Transition-state based inhibitors of thermolysin: Phosphonamides: a) J.
Cramer, S. G. Krimmer, A. Heine, G. Klebe, J. Med. Chem. 2017, 60,
5791-5799; b) J. Cramer, S. G. Krimmer, V. Fridh, Veronica, T. Wulsdorf,
R. Karlsson, A. Heine, G. Klebe, ACS Chem. Biol. 2017, 12, 225-233; c)
P.A. Bartlett, N. Yusuff, A. C. Rico, M. K. Lindvall, J. Am. Chem. Soc.
2002, 124, 3853-3857; d) D. Grobelny, U. B. Goli, R. E. Galardy,
Biochemistry 1989, 28, 4948-4951; e) H. M. Holden, D. E. Tronrud, A. F.
Monzingo, L. H. Weaver, B. W. Matthews, Biochemistry 1987, 26, 8542-
8553; f) D. E. Tronrud, H. M. Holden, B. W. Matthews, Science 1987,
235, 571-574; g) P.A. Bartlett, C. K. Marlowe, Science 1987, 235, 569-
This article is protected by copyright. All rights reserved.

10.1002/chem.202102204
Chemistry - A European Journal
FULL PAPER
571; Phosphonates: h) D. A. Campbell, J. C. Bermak, T. S. Burkoth, D.
V. Patel, J. Am. Chem. Soc. 1995, 117, 5381-5382; Silandiol: i) J. Kim,
S. M. Sieburth, J. Org. Chem. 2004, 69, 3008-3014; j) J. Kim, A. Glekas,
S. M. Sieburth, Bioorg. Med. Chem. Lett. 2002, 12, 3625-3627.
[33] Synthesis of hydroxyethylene bioisosteres: a) B. E. Haug, D. H. Rich,
Org. Lett. 2004, 6, 4783-4786; b) B. E. Haug, M. Brewer, D. H. Rich, J.
Pept. Res. 2005, 65, 77-83 (In this publication it was announced that the
hydroxyethylene bioisosteres could be used for the inhibition of
metalloproteases, but no data were published).
[47] T. R. Kelly, T. E. Schmidt, J. G. Haggerty, Synthesis 1972, 1972, 544-
545.
[48] A. Nadin, J. M. Sánchez López, J. G. Neduvelil, S. R. Thomas,
Tetrahedron 2001, 57, 1861-1864.
[49] J. Lenard, A. B. Robinson, J. Am. Chem. Soc. 1967, 89, 181-182. DOI:
10.1021/ja00977a057.
[50] S. Matsuura, C.-H. Niu, J. S. Cohen, J. Chem. Soc., Chem. Commun.
1976, 451. DOI: 10.1039/c39760000451.
[51] A. Toró, P. Nowak, P. Deslongchamps, J. Am. Chem. Soc. 2000, 122,
4526-4527. DOI: 10.1021/ja000728f.
[34] A. K. Ghosh, K. Xi, K. Ratia, B. D. Santarsiero, W. Fu, B. H. Harcourt, P.
A. Rota, S. C. Baker, M. E. Johnson, A. D. Mesecar, J. Med. Chem. 2005,
48, 6767-6771.
[52] a) E. Freire, in Thermodynamics and Kinetics of Drug Binding (Methods
and Principles in Medicinal Chemistry, Vol. 65) (Eds.: G. M. Keserü, D.
C. Swinney), Wiley-VCH, Weinheim, 2015, pp. 3-13; b) Y. Kawasaki, E.
Freire, Drug Disc. Today 2011, 16, 985-990.
[35] H. A. Staab, K. Wendel, Org. Synth. 1968, 48, 44.
[36] J. Ivkovic, C. Lembacher-Fadum, R. Breinbauer, Org. Biomol. Chem.
2015, 13, 10456-10460.
[53] A. R. Blaazer, I. J. P. de Esch, in Thermodynamics and Kinetics of Drug
Binding (Methods and Principles in Medicinal Chemistry, Vol. 65) (Eds.:
G. M. Keserü, D. C. Swinney), Wiley-VCH, Weinheim, 2015, pp. 81-105.
[54] a) H. Ohtaka, E. Freire, Prog. Biophys. Mol. Biol. 2005, 88, 193-208; b)
A. Velazquez-Campoy, M. J. Todd, E. Freire, Biochemistry 2000, 39,
2201-2207.
[37] a) E. J. Corey, A. Venkateswrlu, J. Am. Chem. Soc. 1972, 94, 6190-6191;
b) E. J. Corey, T. K. Schaaf, W. Huber, U. Koelliker, N. M. Weinshenker,
J. Am. Chem. Soc. 1970, 92, 397-398.
[38] A. K. Ghosh, G. Bilcer, C. Harwood, R. Kawahama, D. Shin, K. A.
Hussain, L. Hong, J. a. Loy, C. Nguyen, G. Koelsch, et al., J. Med. Chem.
2001, 44, 2865-2868.
[55] a) X.-M. Chen, X.-C. Huang, M.-L. Tong, Y.-X. Tong, S. W. Ng, Aust. J.
Chem. 1997, 50, 865; b) T. Hori, T. Kumasaka, M. Yamamoto, T. Nonaka,
N. Tanaka, Y. Hashimoto, T. Ueki, K. Takio, Acta Crystallogr. Sect. D
Biol. Crystallogr. 2001, 57, 361-368; c) M. J. Loferer, C. S. Tautermann,
H. H. Loeffler, K. R. Liedl, J. Am. Chem. Soc. 2003, 125, 8921-8927; d)
J. Li, Y. Deng, S. Jie, B.-G. Li, J. Organomet. Chem. 2015, 797, 76-82.
[56] D. Liebschner, P.V. Afonine, N. W. Moriarty, B. K. Poon, O. V. Sobolev,
T. C. Terwilliger, P. D. Adams, Acta Cryst. D 2017, 73, 148-157.
[57] M. K. Lakshman, F. A. Tine, T. A. Khandaker, V. Basava, N. B.
Agyemang, M. S. A. Benavidez, M. Gaši, L. Guerrera, B. Zajc, Synlett
2017, 28, 381-385.
[39] a) A. Bartoszewicz, M. Kalek, J. Nilsson, R. Hiresova, J. Stawinski,
Synlett 2008, 37–40; b) A. Bartoszewicz, M. Kalek, J. Stawinski,
Tetrahedron 2008, 64, 8843-8850.
[40] a) A. El-Faham, F. Albericio, Chem. Rev. 2011, 111, 6557-6602; b) H.
Schröder, G. A. Strohmeier, M. Leypold, T. Nuijens, P. J. L. M.
Quaedflieg, R. Breinbauer, Adv. Synth. Catal. 2013, 355, 1799-1807.
[41] J. Choy, S. Jaime-Figueroa, L. Jiang, P. Wagner, Synth. Commun. 2008,
38, 3840-3853.
[42] a) S. Minegishi, H. Mayr, J. Am. Chem. Soc. 2003, 125, 286-295; b) T.
Kanzian, T. A. Nigst, A. Maier, S. Pichl, H. Mayr, Eur. J. Org. Chem. 2009,
6379–6385.
[58] L. A. Carpino, H. Imazumi, B. M. Foxman, M. J. Vela, P. Henklein, A. El-
Faham, J. Klose, M. Bienert, Org. Lett. 2000, 2, 2253-2256.
[59] T. Kanzian, T. A. Nigst, A. Maier, S. Pichl, H. Mayr, Eur. J. Org. Chem.
2009, 36, 6379-6385.
[43] M. Brenner, L. La Vecchia, T. Leutert, D. Seebach, Org. Synth. 2003, 80,
57.
[44] Y. Hamada, M. Shibata, T. Sugiura, S. Kato, T. Shioiri, J. Org. Chem.
1987, 52, 1252-1255.
[60] A. R. Khan, J. C. Parrish, M. E. Fraser, W. W. Smith, P. A. Bartlett, M. N.
G. James, Biochemistry 1998, 37, 16839-16845.
[45] K. Shikata, H. Azuma, T. Tachibana, K. Ogino, Tetrahedron 2002, 58,
5803-5809.
[46] P. F. Schuda, C. B. Ebner, S. J. Potlock, Synthesis 1987, 1987, 1055-
1057.
This article is protected by copyright. All rights reserved.

10.1002/chem.202102204
Chemistry - A European Journal
FULL PAPER
Entry for the Table of Contents (Please choose one layout)
Layout 1:
FULL PAPER
A transition state-based inhibitor has
been developed which efficiently
inhibits the mealloprotease
dipeptidylpeptidase 3 (DPP3) by
releasing structured water molecules
from the binding cleft.
Jakov Ivkovic, Shalinee Jha, Christian
Lembacher-Fadum, Johannes Puschnig,
Prashant Kumar, Viktoria Reithofer, Karl
Gruber, Peter Macheroux*, Rolf
Breinbauer*
1 – 14
Efficient Entropy-Driven Inhibition of
Dipeptidyl Peptidase III by
Hydroxyethylene Transition-State
Peptidomimetics
This article is protected by copyright. All rights reserved.