Design, synthesis, and structure-activity relationship studies of a potent PACE4 inhibitor

Article abstract of DOI:10.1021/jm401457n

PACE4 plays an important role in the progression of prostate cancer and is an attractive target for the development of novel inhibitor-based tumor therapies. We previously reported the design and synthesis of a novel, potent, and relatively selective PACE4 inhibitor known as a Multi-Leu (ML) peptide. In the present work, we examined the ML peptide through detailed structure-activity relationship studies. A variety of ML-peptide analogues modified at the P8-P5 positions with leucine isomers (Nle, DLeu, and DNle) or substituted at the P1 position with arginine mimetics were tested for their inhibitory activity, specificity, stability, and antiproliferative effect. By incorporating d isomers at the P8 position or a decarboxylated arginine mimetic, we obtained analogues with an improved stability profile and excellent antiproliferative properties. The DLeu or DNle residue also has improved specificity toward PACE4, whereas specificity was reduced for a peptide modified with the arginine mimetic, such as 4-amidinobenzylamide.

Full text of DOI:10.1021/jm401457n

Article
pubs.acs.org/jmc
Design, Synthesis, and Structure−Activity Relationship Studies of a
Potent PACE4 Inhibitor
Anna Kwiatkowska,† Frederic Couture,† Christine Levesque,† Kevin Ly,† Roxane Desjardins,†
́
́
́
Sophie Beauchemin,‡ Adam Prahl,§ Bernard Lammek,§ Witold Neugebauer,† Yves L. Dory,*,‡
and Robert Day*,†
†Institut de Pharmacologie de Sherbrooke, Department of Surgery/Urology Division, Universite
North, Sherbrooke, Quebec, J1H 5N4, Canada,
‡Institut de Pharmacologie de Sherbrooke (IPS), Department of Chemistry, Faculty of Science, Universite
́
de Sherbrooke, 3001 12th Avenue
́
́
de Sherbrooke, 3001 12th
sk, Gdansk 80-952,
́
Avenue North, Sherbrooke, Quebec J1H 5N4, Canada
§Institute of Organic Synthesis, Department of Organic Chemistry, Faculty of Chemistry, University of Gdan
Poland
́
́
S
* Supporting Information
ABSTRACT: PACE4 plays an important role in the
progression of prostate cancer and is an attractive target for
the development of novel inhibitor-based tumor therapies. We
previously reported the design and synthesis of a novel, potent,
and relatively selective PACE4 inhibitor known as a Multi-Leu
(ML) peptide. In the present work, we examined the ML
peptide through detailed structure−activity relationship
studies. A variety of ML-peptide analogues modified at the
P8−P5 positions with leucine isomers (Nle, DLeu, and DNle)
or substituted at the P1 position with arginine mimetics were
tested for their inhibitory activity, specificity, stability, and
antiproliferative effect. By incorporating D isomers at the P8
position or a decarboxylated arginine mimetic, we obtained
analogues with an improved stability profile and excellent antiproliferative properties. The DLeu or DNle residue also has
improved specificity toward PACE4, whereas specificity was reduced for a peptide modified with the arginine mimetic, such as 4-
amidinobenzylamide.
Therefore, PCs have emerged as potential targets for the
development of novel therapeutics.5−7
INTRODUCTION
■
The proprotein convertases (PCs) are a family of Ca2+-
dependent serine endoproteases involved in many physiological
processes, including the activation of a wide array of regulatory
proteins (e.g., receptor pro-forms, hormones, growth factors,
zymogens, or cell membrane proteins).1 Nine human PCs have
been identified, namely, furin, PC1/3, PC2, PACE4, PC4,
PC5/6, PC7, PCSK9, and SKI-1/S1P. The first seven members
of this family have a cleavage preference for substrates
containing multibasic cleavage motifs, for which a consensus
sequence has been established, namely, Arg-X-Arg/Lys-Arg↓.2
The other two enzymes exhibit different substrate specificities.
SKI-1/S1P cleaves its substrates at nonbasic residues,3 whereas
PCSK9 cleaves itself once and acts in a nonenzymatic fashion as
a binding protein to the low-density lipoprotein receptor, thus
regulating blood cholesterol levels.4 In addition to their normal
physiological function, there is a large body of literature
describing the implication of PCs in numerous pathological
states, for example, viral and bacterial infections, tumorigenesis,
neurodegenerative disorders, diabetes, and dyslipidemia.5
A wide variety of PC inhibitors have been reported, and
among these are proteins,8 peptides,9,10 peptidomimetics,11,12
small molecules,13 antisense RNA,14 and monoclonal antibodies
for PCSK9.15 The earliest studies included the use of furin
inhibitors for the treatment of cancer as well as viral and
pathogenic infections.5,7 As examples, the injection of cancer
cells expressing the protein-based furin inhibitor α1-Antitrypsin
Portland demonstrated decreased tumorigenesis and metastasis
in transgenic mice,16 and the irreversible synthetic inhibitor
decanoyl-Arg-Val-Lys-Arg-chloromethyl ketone (dec-RVKR-
CMK) has been shown to prevent tumor growth through the
inhibition of matrix metalloproteinase 2 processing.17 However,
these inhibitors are not expected to become therapeutic agents
because of their size, toxicity, lack of specificity, and
stability.5,12,18 Therefore, they have been mainly used to
establish proof of concept for the role of PCs in various
Received: September 20, 2013
Published: December 10, 2013
© 2013 American Chemical Society
98
dx.doi.org/10.1021/jm401457n | J. Med. Chem. 2014, 57, 98−109

Journal of Medicinal Chemistry
Article
diseases. In contrast, peptidomimetic furin inhibitors appear to
be promising compounds because of their antiviral activity and
their protective effect against bacterial toxins, thereby offering
possibilities to generate more efficient treatments for infectious
diseases.11,12 Moreover, on the basis of the crystal structure of
furin, several nonpeptide small molecules have been
developed;13,19,20 however, among them, only tetrabasic 2,5-
dideoxystreptamine derivatives exhibit potent activity against
furin, with inhibition constants (Ki) values <10 nM, and have a
protective effect against intoxication of macrophages by anthrax
protective antigen.13 Despite their excellent potency, none of
these compounds have reached the clinical development stage
so far.
Figure 1. Proposed modifications of the Multi-Leu peptide inhibitor.
position of a furin inhibitor significantly improved its potency.
Therefore, in a second approach, the P1 position of the ML-
peptide inhibitor was replaced by arginine mimetics with similar
chemical character (Figure 2). In addition, we designed a
Our recent studies revealed that another PC, namely,
PACE4, is a viable therapeutic target for prostate cancer
(PCa). We showed that PACE4 is overexpressed in PCa and
plays a major role in the progression of the disease to more
aggressive forms. In addition, we demonstrated that inhibition
of this PC results in a decreased cellular proliferation rate and
reduced tumorigenic growth in nude mice.21 Moreover, our
studies also showed that PACE4 has an essential role that is not
compensated by other cancer-associated PCs, such as furin or
PC7.22 On the basis of these findings, we proposed that PACE4
is a novel target for the management of PCa and that
compounds that target PACE4 would make good PCa
therapeutic agents. However, the high degree of sequence
and structural homology between the catalytic domains of PCs
complicates the development of PACE4-specific inhibitors.
Nevertheless, our group has successfully generated a potent and
relatively selective PACE4 inhibitor (20-fold specificity over
furin) known as the Multi-Leu (ML) peptide.23 This peptide
inhibitor exhibits potent antiproliferative effects on PCa cell
lines DU145 and LNCaP. The pharmacological profile of the
ML peptide makes it a promising lead compound for the
development of novel anti-PCa agents.
Peptides have a number of advantages over small organic
molecules, proteins, and antibodies as drug candidates because
of their high biological activity, specificity, and low toxicity.
However, the utility of peptides is limited by their poor
metabolic stability, which results in a short duration of in vivo
activity and low bioavailability. In the present study, we set out
to overcome these limitations by introducing various chemical
modifications into the structure of the ML-peptide inhibitor.
Our strategy consisted of identifying the amino acids that could
be substituted in combination with an understanding of the
specificity and stability profile of the ML peptide. We used a
peptidomimetic strategy that involves unnatural or D amino
acid substitutions and arginine mimetics to create more stable
and potent analogues of the ML peptide. The best compounds
were then tested for their continued efficacy for anti-PCa
activity.
Figure 2. Structure of arginine mimetics used in the present study.
compound with azaβ3-Arg (azaβ3-R) in the P1 position, which
is formed by replacing the α-C-atom with a hydrazine moiety.
This hydrazinopeptoid was chosen to determine how the loss
of chirality and the nitrogen enrichment of the peptide
backbone influence the activity of the resulting analogue.
All peptides were obtained via solid-phase peptide synthesis
(SPPS) or a combination of SPPS and solution synthesis. The
inhibitors containing Arg, Gpa, or the azaβ3-Arg modification
at the P1 position (peptides 1−16, 19, and 20) and a control-
scrambled peptide (Ac-RLRLLKVL-NH2) were prepared on
polystyrene resin (TentaGel S RAM) using standard Fmoc
SPPS.
Inhibitors modified with the decarboxylated derivatives
ΔAgm or Amba (17 and 18) were prepared as described in
Scheme 1. First, the P8−P2 segments of these derivatives were
synthesized by standard Fmoc SPPS on a 2-chlorotrityl-chloride
resin. Next, the protected peptides were cleaved from the resin
followed by coupling of Amba·2HCl or ΔAgm(Boc)2 in
solution. Side-chain deprotection by TFA was performed in the
final step.
Fluorescent versions of the selected analogues were prepared
by incorporation of a fluorescein isothiocyanate isomer I
(FITC) at the N terminus (βAla was used as a spacer).
All inhibitors were purified by reverse-phase high-perform-
ance liquid chromatography (RP-HPLC) and obtained as
lyophilized TFA salts. The identity and purity of the
compounds were determined by HPLC and matrix-assisted
laser desorption/ionization (MALDI) or surface-enhanced laser
desorption/ionization (SELDI) mass spectrometry (MS) and
by high-resolution mass spectrometry (HRMS). The desired
products were obtained in good yield. Information regarding
RESULTS
■
Synthesis. Our peptidomimetic strategy is presented in
Figure 1. On the basis of our previous results showing that the
leucine core plays a key role in the specificity of a ML-peptide
inhibitor for PACE4,23 only leucine isomers (unusual or/and D
amino acids) were used. We designed two series of ML-peptide
analogues modified by single or multiple amino acid
substitutions with L-norleucine or D-leucine/D-norleucine at
the P8, P7, P6, and P5 positions. A recent study by Becker et
al.11 demonstrated that the introduction of a decarboxylated
and conformationally restricted arginine mimetic in the P1
99
dx.doi.org/10.1021/jm401457n | J. Med. Chem. 2014, 57, 98−109

Journal of Medicinal Chemistry
Article
a
Scheme 1. Synthesis of Inhibitors 17, 18
a
Using the example of the peptide 18. Reagents and conditions: (a) 2-chlorotrityl chloride resin, 1.2 equiv of Fmoc-Lys(Boc)-OH, 4 equiv of
DIPEA, dry DCM, 2.5 h; (b) Fmoc SPPS, coupling with 3 equiv of amino acid, 6Cl-HOBt, and HATU or PyBOP, respectively, and 9 equiv of NMM
or DIPEA; (c) HFIP/DCM (1:4, v/v), 2 h; (d) 2 equiv of Amba·2HCl, 2 equiv of COMU, and 4 equiv of NMM, DMF, 12 h; (e) TFA/TIS/H2O
(95:2.5:2.5, v/v/v), 3 h.
the analytical characterization of the inhibitors is provided in
the Supporting Information (Tables S1 and S2).
Scheme 3. Synthesis of ΔAgm(Boc)2 (N-(2-butenyl)
a
guanidine-(Boc)2; IX)
All noncommercially available derivatives (Amba·2HCl,
ΔAgm(Boc)2, and Fmoc-azaβ3-Arg(Boc)2) used in the present
study were synthesized according to the literature and/or
methods developed in our laboratory. Amba·2HCl was
prepared as described24,25 with a slight modification of the
acylation step (Scheme 2). To obtain ΔAgm(Boc)2, a novel
Scheme 2. Synthesis of Amba·2HCl (4-Amidinobenzylamine·
a
2HCl; IV)
a
Reagents and conditions: (a) DMF, 48 h; (b) hexamethylenetetr-
amine, CHCl3, 48 h; (c) EtOH, HCl, reflux 2 h; (d) Et3N, DCM,
overnight; (e) MeOH, CHCl3, hydrazine, 4 h.
Kinetic Measurements. Our prior studies that led to the
ML-peptide inhibitor development established that the leucine
core is essential for its specificity toward PACE4 over furin.23
To assess the impact of the proposed modifications on the
specificity and potency of our leading compound, the inhibition
constants (Ki) of the newly synthesized compounds were
determined via competitive kinetic assays using recombinant
human furin and PACE4. For an analogue modified with Amba,
the values of the apparent Ki were calculated according to the
equation for the kinetics of reversible tight-binding inhib-
itors.28,29 The Ki values of all of the novel compounds together
with those of the ML peptide are presented in Figures 3 and 4.
Analogues Modified with Norleucine (Nle). The substitu-
tion of Leu by Nle at the P8, P7, and P5 positions had little
effect on the inhibitory potency of the resulting analogues
(peptides 1, 2, and 4, respectively) against PACE4 (Figure 3A).
However, a single modification of the P6 position with Nle
yielded compound 3, which had substantially reduced activity
(19-fold less potent than the ML peptide). In the case of
peptides with multiple Nle substitutions, all analogues modified
at the P6 position (5, 8, and 9) showed significantly reduced
inhibitory activity (8.5−25-fold reduced activity). In contrast,
peptides that were not modified at this position (6, 7, and 10)
remained potent PACE4 inhibitors. One of the novel
compounds, analogue 10 (Ac-[Nle]LL[Nle]RVKR-NH2), was
a
Reagents and conditions: (a) H2O, 1 N NaOH, overnight; (b) NH2-
OH·HCl, DIPEA, MeOH, reflux 12 h; (c) bis(trichloromethyl)
carbonate, pyridine/THF, 0 °C, 1 h, 60 °C, 12 h; (d) 50% AcOH(aq),
H2 (50 psi), Pd/C (10%), 2 h; (e) H2O, HCl, 1.5 h.
synthetic strategy was developed (Scheme 3) on the basis of
work by Jeon et al.26 Briefly, the desired protected ΔAgm
compound was obtained in four steps starting from 1,4-
dibromo-trans-2-butene. First, the mono-N-phthalimide deriv-
ative was prepared (V). Then, the remaining bromine was
converted with hexamethylenetetramine to a quaternary
ammonium salt (VI), which was hydrolyzed under acidic
conditions to a primary amine salt (VII). In the final steps, the
guanidine fragment was introduced into the molecule (VIII)
using N,N′-di-Boc-N″-triflylguanidine as the guanidylation
reagent followed by the deprotection of the phthalimide
group with hydrazine hydrate to give the desired compound,
ΔAgm(Boc)2 (IX). Full experimental details and analytical data
1
for the compounds V−IX (including copies of the H NMR
spectra) are provided in the Supporting Information. Fmoc-
azaβ3-Arg(Boc)2 was obtained following the procedure
reported by Busnel et al.27
even more potent than the ML peptide (Ki = 14 2 vs 20
2
nM for the ML-peptide). Regarding the inhibitory potency
100
dx.doi.org/10.1021/jm401457n | J. Med. Chem. 2014, 57, 98−109

Journal of Medicinal Chemistry
Article
these results, we designed a single analogue modified with D-
norleucine (peptide 16), whose activity was similar to that of its
DLeu counterpart (Figure 3B). Three of the novel peptides
(11, 13, and 16) modified at the P8 or P6 position were 1.7−
2.3-fold less effective against furin than the ML peptide (Figure
4B). The inhibitory potency of compounds with DLeu at
position P5 or with the entire leucine core with an inverted
configuration (Ac-LLL[DLeu]RVKR-NH2 and Ac-[DLeu1−4]-
RVKR-NH2) was strongly decreased (17−22-fold reduction).
Moreover, analogue 12, which was modified at position P7, was
6.6-fold less potent toward furin when compared with the ML-
peptide inhibitor. The inhibitory potency of two analogues
from this group modified at position P8 (11 and 16) against
PACE4 was comparable to that of our leading compound;
however, their selectivity was greatly improved (≤32-fold,
Figure 4B).
Analogues Substituted at Position P1. The incorporation
of the decarboxylated and conformationally restricted arginine
mimetics ΔAgm and Amba in the structure of the ML peptide
(17 and 18) yielded analogues with improved inhibitory
potency toward PACE4 (Ki = 5.3
1.5 and 3.1
0.8 nM,
respectively). In contrast, as shown in Figure 3C, the analogue
containing an aromatic amino acid derivative (19) had
dramatically reduced potency (48-fold reduction). Modification
of the structure with an azaβ3-Arg substitution had no
beneficial effect on PACE4 inhibition (20-fold diminished
activity). Analogues 17 and 18 were also strong furin inhibitors;
therefore, these modifications greatly reduced the specificity
toward PACE4. For example, the incorporation of Amba
drastically improved (100-fold) activity toward furin, which
resulted in comparable Ki values for PACE4 and furin (Ki = 3.1
0.8 and 4.3 0.8 nM, respectively). The ΔAgm-substituted
inhibitor (17) binds to furin with a Ki value of 27 2 nM,
resulting in 4-fold diminished specificity toward PACE4 when
compared with the ML peptide. The other P1-modified
analogues (19 and 20) were significantly less potent furin
inhibitors, with Ki values of 11.0 and 1.12 μM, respectively.
Although the substrate specificities of PCs differ, they all
possess a strong preference for similar multibasic cleavage
sequences. Therefore, we selected a few potent PACE4
inhibitors (peptides 1, 4, 10, 11, and 16−18) and measured
their selectivity toward related PCs, hPC2 and hPC7 (Table S2,
Supporting Information). The inhibitory activity of the novel
compounds against PC7 is similar to that of our leading
compound, with the exception of decarboxylated peptides 17
and 18, which exhibited higher potency (Ki = 72 2 and 13.8
0.4 nM, respectively). In the case of PC2, analogues with
substitutions in the multileucine core (1, 4, 10, 11, and 16)
were stronger inhibitors (1.7−4-fold increased inhibition) when
compared to the ML peptide. The incorporation of the ΔAgm
and Amba modifications also significantly increased the
inhibitory potency of the resulting analogues toward PC2 (Ki
= 1.7 0.6 and 0.24 0.03 nM, respectively).
Figure 3. Inhibition constant (Ki) against PACE4 for analogues
modified at Multi-Leu core with (A) L-norleucine (Nle), (B) D-amino
acids, or (C) arginine mimetics. The results are presented on a
logarithmic scale to facilitate comparison. The data for the ML-peptide
inhibitor was included as a control and is identical to a previous
publication.23
toward furin, all analogues with single or multiple Nle
substitutions were less potent than the original ML peptide
(Figure 4A).
Inhibitors 1 and 10 exhibited significantly improved
selectivity toward PACE4 when compared with the ML peptide
(31- or 48-fold improvement, respectively, Figure 4A). The
selectivity of the two other peptides, analogues 2 and 4, was
slightly improved or maintained. In contrast, all of the newly
prepared compounds with Nle at position P6 or with multiple
Nle substitutions were poor PACE4 inhibitors with reduced
selectivity, with the exception of above-mentioned peptide 10.
Analogues Modified with D-Leucine (DLeu) and D-
Norleucine (DNle). The introduction of the DLeu residue in
the structure of the ML-peptide inhibitor yielded analogues
with reduced (peptide 12; 7-fold reduction) or drastically
diminished (compounds 13, 14, and 15; 300−700-fold
reduction) inhibitory potency toward PACE4. However,
peptide 11 (obtained by DLeu substitution at the P8 position)
appears to be an exception to the rule because it remained a
potent PACE4 inhibitor (Ki = 24 1 nM). On the basis of
In Vitro Plasma Stability Studies. Multibasic peptides are
potential substrates for a wide variety of trypsin-like proteases
that are present in blood; therefore, the plasma stability of our
ML-peptide inhibitor and its analogues was a significant
parameter to evaluate. We determined the stability profiles of
the selected analogues in mouse plasma using RP-HPLC
analyses and calculated their half-life (t1/2) values. Furthermore,
we investigated the degradation pattern of these inhibitors via
MS analyses (MALDI-TOF MS). The results of these studies
are presented in Figure 5.
101
dx.doi.org/10.1021/jm401457n | J. Med. Chem. 2014, 57, 98−109

Journal of Medicinal Chemistry
Article
Figure 4. Specificity ratio for PACE4/furin of Multi-Leu analogues modified at the Multi-Leu core with (A) L-norleucine (Nle), (B) D amino acids,
and (C) arginine mimetics at the P1 position. Peptides 13, 14, and 15 were more specific toward furin. The data for the ML-peptide inhibitor was
included as a control and is identical to a previous publication.23
The half-life of the ML-peptide inhibitor was 5.1 h. No
benefits were observed after the incorporation of norleucines in
its structure, and the plasma stabilities of the resulting
analogues were reduced in some cases (Figure 5A). Analogues
containing D amino acids (DLeu and DNle) had greatly
improved proteolytic resistance, with t1/2 values of 8.6 and 7.1
h, respectively. Among inhibitors modified at the P1 position,
the compound substituted with Amba was 2.4-fold more stable
(t1/2 = 12 h) than the ML peptide. In contrast, ΔAgm and
azaβ3-Arg did not have such a strong impact on the plasma
stability, and the half-life of the resulting analogues was 7.2 and
5.6 h, respectively. The first degradation products of the ML
inhibitor were detectable after 30 min of incubation at 37 °C in
murine plasma, having molecular ions of m/z 897, 769, and 670
(Figure 5C). These molecular ions correspond to seven-, six-,
and five-amino acid sequences that lack the C-terminal amino
acids (-R-NH2, -KR-NH2, and -VKR-NH2, respectively). As
shown in Figure 5C, all of the selected peptides having Arg at
the P1 position also followed the same degradation pattern and
were cleaved at those specific positions (after basic amino acid
residues). The incorporation of decarboxylated arginine
mimetics (Amba and ΔAgm) prevented C-terminal degrada-
tion after the P3 and P2 positions. However, for all peptides, a
102
dx.doi.org/10.1021/jm401457n | J. Med. Chem. 2014, 57, 98−109

Journal of Medicinal Chemistry
Article
Figure 5. Stability profile of the selected Multi-Leu analogues. The peptides were incubated with the plasma at 37 °C for different time periods. (A)
Plasma half-life (t1/2) of the ML analogues. (B) Percentage of the peptide remaining at each degradation time point. (C) C-terminal degradation
pattern.
molecular ion corresponding to a fragment cleaved after Arg at
position P4 was detected.
based cell proliferation assays. PACE4 inhibitors that were
potent in the enzymatic assay exhibited comparable (1 and 4)
or improved (peptides 10, 11, and 16−18) antiproliferative
activities relative to the ML-peptide inhibitor (Figure 6A).
Among all of the analogues tested, two peptides, 10 and 18,
showed the highest activity against both DU145 (IC50 = 40
10 and 25 10 μM, respectively) and LNCaP (IC50 = 90 30
Antiproliferative Activity and Cytotoxicity against
PCa Cell Lines. We previously observed a strong correlation
between cellular PACE4 levels and ML-peptide activity in three
human PCa cell lines (LNCaP, DU145, and PC3) with
different PACE4 expression levels.23 The ML peptide
significantly inhibited the proliferation of DU145 and LNCaP
cells (IC50 values of 100 10 and 180 60 μM, respectively)
and had little effect on PC3 cells, which were used as a negative
control because PACE4 is almost undetectable in this cell line.
To confirm the specific activity of the ML peptide, we used a
scrambled version of this inhibitor (the same amino acid
composition but in a different order, a sequence without a PC
recognition motif) with the following structure Ac-RLRLLKVL-
NH2. A scrambled peptide had no inhibitory activity against
either PACE4 or furin and did not possess antiproliferative
properties (Figure S1, Supporting Information).
and 40
15 μM, respectively) cells (Figure 6B). The
antiproliferative activity of analogue 18 substituted with
Amba increased 4- and 4.5-fold with DU145 and LNCaP
cells, respectively, in comparison to the ML-peptide inhibitor.
In contrast, the analogues modified with D amino acids
(peptides 11 and 16) were only slightly more potent than
our control peptide (Figure 6A,B) and also displayed cytotoxic
activity toward PC3 cells (IC50 = 190 50 and 150 20 μM,
respectively). As expected, the analogues with poor inhibitory
potency toward PACE4 (peptides 3, 13, 15, and 20) also
exhibited weak antiproliferative activity. The relatively high IC50
values of peptides 3 and 13 (in DU145 or DU145 and LNCaP
cells, respectively) are most likely linked to the nonspecific
cytotoxicity of these compounds. Lactate dehydrogenase
(LDH) assays were performed with DU145 cells to determine
To determine whether the inhibitory activity of the new ML-
peptide analogues against recombinant enzyme could be
translated to an antiproliferative effect, we selected several
analogues with different inhibitory profiles to perform MTT-
103
dx.doi.org/10.1021/jm401457n | J. Med. Chem. 2014, 57, 98−109

Journal of Medicinal Chemistry
Article
Figure 6. Antiproliferative activity and cytotoxicity of the selected Multi-Leu inhibitors on prostate cancer cell lines. (A) Comparison of the IC50
values and lactate dehydrogenase (LDH) levels. The concentration of peptides that inhibited cell growth by 50% (IC50) was determined by MTT
cell survival assays on DU145, LNCaP, and PC3 cell lines. N.C. indicates that the curve did not converged to 50% with doses up to 350 μM. Acute
cytotoxicity of the analogue at the indicated concentration (100 μM) was measured by lactate dehydrogenase (LDH) assay after 4 h of incubation
with DU145 cells. (B, C) Dose−response curves for the antiproliferative effect of the ML-peptide inhibitor and its most potent analogues (peptides
10, 11, and 18) in DU145 and LNCaP cells.
whether the effective concentration of ML-peptide analogues
for inhibiting cell proliferation indicates growth inhibition
rather than acute cytotoxicity. The activity of LDH, which is an
enzyme released into the culture medium from cells with
damaged membranes, was determined as an indicator of cell
death/lysis severity. After 4 h of incubation with 100 μM of any
of the inhibitors, the LDH measurements indicated <5% acute
toxicity in DU145 cells (Figure 6A).
Cell Penetration Studies. We previously suggested that
the ML-peptide inhibitor targets intracellular PACE4 to inhibit
DU145 proliferation, and its N-terminal modification via
incorporation of the PEG8 moiety altered this activity by
reducing cell permeability.23 To investigate if the cell
penetration efficiency of the ML peptide would be affected
Figure 7. Cellular uptake of the fluorescent ML analogues by DU145
by the proposed modifications, the cellular uptake of selected
cells. The cells were treated with 10 μM FITC-labeled compounds,
compounds 11 and 17, which possess the most suitable overall
and the spectra were collected after 1 h incubation. Untreated cells
profile (high inhibitory potency in both enzymatic and cell-
were used as a control, and their autofluorescence intensity is
based assays and improved plasma stability), were compared.
presented as a gray curve. Numbers indicate the geometric mean
This test was performed with FITC-labeled versions of these
peptides (FITC-βAla-[DLeu]LLLRVKR-NH2 and FITC-βAla-
LLLLRVK-Amba) after 1 h of incubation with DU145 cells and
was analyzed using fluorescence-activated cell sorting (FACS)
analyses as previously described.23 False-positive signals
resulting from nonspecific interactions of peptides with the
cell membrane were eliminated after trypsinization. The results
of these assays are presented in Figure 7 together with the
spectra for the ML-peptide inhibitor. In general, the fluorescent
version of the ML peptide yielded the highest geometric mean
fluorescence intensity (GMFI = 63.92), and the incorporation
fluorescence intensity for peptide-treated cells and autofluorescence
intensity of the cells.
of DLeu or Amba into its structure resulted in decreased
fluorescence intensities (GMFI = 8.43 and 18.86, respectively),
revealing a lower accumulation of those analogues in DU145
cells after 1 h.
DISCUSSION
■
We previously demonstrated that PACE4 plays a critical role in
PCa tumor progression.21,22 Therefore, the inhibition of this
104
dx.doi.org/10.1021/jm401457n | J. Med. Chem. 2014, 57, 98−109

Journal of Medicinal Chemistry
Article
protease may offer novel and more effective treatment
strategies. These encouraging results prompted us to generate
a potent and specific PACE4 inhibitor, which led to the
development of the ML peptide.23 The interesting pharmaco-
logical profile of this inhibitor (i.e., specificity toward PACE4,
potent antiproliferative effects on PCa cell lines, and good cell
permeability) makes the ML peptide an attractive candidate for
development as a novel anticancer agent. However, because this
molecule is a peptide, the ML-peptide inhibitor encounters
additional challenges in the drug-development process,
including poor metabolic stability and low bioavailability,
which are mainly the result of rapid in vivo degradation by
proteases. Therefore, the main goal of this study was to
improve the pharmacological profile of the ML peptide through
modifications of its structure using a peptidomimetic approach.
The first group of compounds was designed with single or
multiple Nle substitutions in the tetraleucine tail of the ML
inhibitor. The enzyme kinetic analyses results of these
analogues (Figure 4A) demonstrated that the Leu residue in
the P6 position plays a significant role in the inhibitory activity
against PACE4, and its substitution by the Nle residue resulted
in analogues with substantially reduced activity (8.5−19-fold
reduction). Interestingly, this relatively small modification of
the side chain (replacing the branched alkyl side chain of Leu
with the linear side chain of Nle) can cause relatively large
changes in the Ki value. In contrast, Nle substitutions at the
other positions (alone or in combination) had little effect
(peptides 1, 4, and 10) or slightly diminished (compounds 2, 6,
and 7) the inhibitory potency toward PACE4. Moreover, two
analogues, Ac-[Nle]LLLRVKR-NH2 (1) and Ac-[Nle]LL[Nle]-
RVKR-NH2 (10), exhibited improved selectivity profiles when
compared with the ML peptide. In the case of peptide 10, the
selectivity toward PACE4 was enhanced nearly 48-fold.
Unfortunately, the Nle modifications, despite their beneficial
effect on the specificity of the resulting analogues, did not
increase their stability in plasma (Figure 5A). However, the
potent in vitro activities of the novel Nle-modified inhibitors
(analogues 1, 4, and 10) were translated into antiproliferative
effects on PCa cells, whereas poor PACE4 inhibitors (peptides
3 and 5) had significantly reduced efficacy in the cellular assays.
The next group of compounds studied consisted of ML-
peptide analogues modified at the tetraleucine region with
DLeu or DNle residues (peptides 11−16). The incorporation
of D amino acids into biologically active peptides is a well-
known strategy to improve their resistance to enzymatic
degradation.30 For PC inhibition, inversion of the configuration
of all of the amino acid residues in a potent furin inhibitor,
nona-L-arginine (Ki = 40 nM), resulted in a more potent and
stable compound, nona-D-arginine, which inhibited furin with a
Ki value of 1.3 nM.31 In contrast, the replacement of one amino
acid by its D isomer at the P4−P1 region of PCs substrate in a
series of peptidomimetic furin inhibitors destroyed their
activity.32 The same effects were observed for the ML-peptide
inhibitor after incorporation of DArg at the P1 position, and the
resulting peptide, Ac-LLLLRVK[DArg]-NH2, was used as a
negative control in our previous study.23 The results presented
here clearly demonstrate that the incorporation of D amino
acids at the P5, P6, and P7 positions of the ML peptide yielded
analogues with strongly diminished inhibitory activity toward
PACE4 (peptides 12−15). This effect was stronger when the
DLeu residue was closer to the P4−P1 recognition motif
(Figure 4B). Only the replacement of Leu at the P8 position by
its D isomer was tolerated and did not alter the inhibitory
potency toward PACE4 (compounds 11 and 16), suggesting
that the L configuration of the seven amino acid residues (from
the P7 to P1 position) is required for proper interaction with
the enzyme catalytic pocket. The presence of a DLeu or DNle
residue at the P8 position significantly improved the specificity
of the resulting analogues toward PACE4 when compared to
the ML inhibitor (Figure 4B). However, the major advantage of
these two analogues is their enhanced stability in plasma
(Figure 5) and improved antiproliferative activity (Figure 6).
Interestingly, we observed a considerable difference between
the penetration of the fluorescently labeled version of the
analogue 11 and the ML peptide into PCa cells (Figure 7).
These findings suggest that the incorporation of the DLeu
residue at the N terminus significantly affected the cellular
uptake of the resulting peptide.
A third group of inhibitors was prepared by incorporating
arginine mimetics in the P1 position (17−20). Among these
analogues, only peptides 17 and 18 (with decarboxylated
derivatives) exhibited improved activity in enzyme kinetic
studies and bound to PACE4 with inhibition constants of 5.3
1.5 and 3.1 0.8 nM, respectively (Figure 4C). The enhanced
binding affinities of these inhibitors were most likely induced by
the reduction of the conformational freedom at the P1 position
with the ΔAgm or Amba modification when compared with the
more flexible arginine side chain of the ML inhibitor. Both the
presence of an (E) double bond in the side chain of an
agmatine derivative and the aromatic ring of 4-amidinobenzy-
lamide stabilized the conformation of the P1 position of the
resulting analogues, which presumably facilitated their inter-
actions with the enzyme’s catalytic pocket. Interestingly, the
other analogue containing rigid aromatic P1 residue (peptide
19) exhibited poor inhibitory potency, indicating that not only
the flexibility reduction but also the proper length and side-
chain orientation of the selected modification affects activity.
The incorporation of azaβ3-Arg in inhibitor 20 provided no
advantage and resulted in an analogue with significantly
reduced (20-fold reduction) potency against PACE4. The
compounds modified with the ΔAgm and Amba residues,
despite their excellent potency against PACE4, suffered from
reduced selectivity. For instance, the incorporation of the
ΔAgm residue also greatly improved (16-fold) the inhibitory
potency of the resulting analogue toward furin and thereby
reduced its specificity for PACE4 (to ∼5-fold, compared to the
20-fold specificity of the ML inhibitor against furin; Figure 4C).
In the case of peptide 18, the specificity was even more strongly
diminished, and the inhibition constants toward PACE4 and
furin were comparable (Ki = 3.1
0.8 and 4.3
0.8 nM,
respectively; Figure 4C). Both characterized analogues also
displayed limited selectivity toward related PCs and were
strong inhibitors of hPC7 and hPC2, which were not observed
in the case of the ML inhibitor and its other analogues (Table
S2). A similar inhibition profile toward furin and PACE4 was
also observed for a series of peptidomimetic furin inhibitors
containing the Amba modification at their C termini; however,
for those compounds, hPC7 and hPC2 were less affected than
furin and PACE4.12 Predicting the effects of the reduced
selectivity of our novel inhibitors in practical use remains
challenging. Considering the redundancy among the PCs and
their ability to compensate for each other in vivo, it might be
beneficial to inhibit more than one enzyme in some
pathophysiological situations. However, our recent studies
demonstrated that PACE4 plays a critical role in the prostatic
neoplastic process that is not duplicated in vivo by other
105
dx.doi.org/10.1021/jm401457n | J. Med. Chem. 2014, 57, 98−109

Journal of Medicinal Chemistry
Article
cancer-associated PCs, such as furin or PC7.22 Therefore, a
highly specific inhibitor should be also exceptionally effective in
treating PCa.
Synthesis of 4-Amidinobenzylamine·2HCl (Amba·2HCl). This
compound was prepared as described in the previously11,24,25 with a
slight modification concerning the acylation step (Scheme 2).
A mixture of Boc-4-cyanobenzyloamine (I) (50 mmol, 11.6 g),
NH2-OH·HCl (75 mmol, 5.12 g), and DIPEA (75 mmol, 13.06 mL)
in methanol was gently stirred for 10 min and then heated under reflux
for 12 h. The mixture was cooled, and the solvent was removed in
vacuo. Then, the residue was partitioned between ethyl acetate and
water. The organic layer was washed with saturated NaHCO3 and
brine and dried over anhydrous MgSO4 before being evaporated under
reduced pressure to afford Boc-4-(aminomethyl)-N-hydroxybenzimi-
damide (II) as a white solid (11.8 g, 89%).
To the solution of product II (40 mmol, 10.6 g) in pyridine/THF
(40 mL) at 0 °C was added bis(trichloromethyl) carbonate (14 mmol,
4.14 g). The resulting solution was stirred at 0 °C for 1 h and then at
60 °C for 12 h. The mixture was cooled, and then ethyl acetate and 0.5
M HCl were added. The organic layer was successively washed with
0.5 M HCl and brine. The combined organic layers were dried over
anhydrous MgSO4, filtered, and concentrated to afford 3-(Boc-4-
(aminomethyl)phenyl)-1,2,4-oxadiazol-5(4H)-one (III) as a yellow
solid (9.3 g, 80%).
In addition to an excellent inhibitory potency against PACE4
and reduced specificity, both compounds (17 and 18) exhibited
improved stability profile in plasma (Figure 5). It should be
noted that the peptide modified with Amba (18) exhibited the
highest stability of all of the analogues tested, with t1/2 = 12 h.
Furthermore, the incorporation of the ΔAgm or Amba
modification led to a significant improvement in the
antiproliferative activity of the resulting analogues against
PCa cell lines (Figure 6). Therefore, both compounds might
represent a good starting point for developing a new generation
of antiprostate cancer agents. However, bearing in mind that
high target selectivity is a primary objective in the drug
development process, we are undertaking further SAR studies
on our Amba- and ΔAgm-substituted inhibitors to improve
their selectivities.
It is interesting to note that the incorporation of the non-
natural residue, the Amba modification, also reduced the cell
penetration properties of the resulting FITC-labeled peptide, as
was seen from a fluorescent version of analogue 11 modified
with the DLeu residue (Figure 7).
Next, product III was dissolved in 50% aqueous acetic acid (300
mL) and treated with catalyst (10% Pd/C) under 50 psi hydrogen for
2 h. The catalyst was removed by filtration, and the solvent was
removed in vacuo. The remaining Boc-4-amidinobenzylamide·acetate
was dissolved in water (200 mL) and treated with concentrated HCl
(40 mL). The resulting solution was stirred for 1.5 h at rt, the solvent
was removed in vacuo, and the residue was lyophilized to afford 4-
amidinobenzylamine·2HCl (IV) as a white solid (6.5 g). MS: [M +
In conclusion, structure−activity studies of our recently
developed ML-peptide inhibitor resulted in several analogues
with an improved overall pharmacological profile. The most
potent PACE4 inhibitors were obtained by the incorporation of
the Amba or ΔAgm residues at the P1 position. These
substitutions had the additional benefit of preventing C-
terminal proteolytic degradation. Similarly, the replacement of
Leu at the P8 position with a DLeu or DNle benefits the
stability of the resulting peptides from N-terminal proteases.
Overall, these peptides exhibited excellent antiproliferative
activities and improved stability profiles. Additional modifica-
tions are being sought to provide a compound with unique anti-
PCa properties.
1
H]+calcd, 149.09; found, 150.0. H NMR (300 MHz, D2O) δ (ppm)
4.36 (s, 2 H), 7.60−7.80 (2d, 4 H, aromatic). 13C NMR (75.5 MHz,
CDCl3) δ (ppm) 42 (CH2), 128, 129, 130, 138 aromatics), 166
(amidino group).
Synthesis of ΔAgm(Boc)2. Full experimental details and the
characterization data for this compound are presented in the
Supporting Information.
Synthesis of Fmoc-azaβ3-Arg(Boc)2. This modification was
prepared following a previously described procedure.27 Briefly, a
substituted hydrazone was synthesized by the condensation between
the aldehyde (3-(Boc-amino)-1-propanal) and commercially available
Fmoc-hydrazine. The hydrazone was then reduced by NaBH3CN to
afford the Fmoc-substituted hydrazine. The -CH2-COOH moiety was
introduced by nucleophilic substitution of benzyl bromoacetate.
Cleavage of the Boc protecting group and guanidinylation with
Goodman’s reagent, (BocNH)2NTf, yielded the protected analogue of
arginine [Fmoc-azaβ3-Arg(Boc)2].
Peptide Synthesis. The inhibitors containing Arg, Gpa, or azaβ3-
Arg modification at position P1 (peptides 1−16, 19, and 20) were
prepared on polystyrene resin (TentaGel S RAM; Rapp Polymere,
capacity 0.23 mmol/g) on a scale of 100 μmol with a Pioneer Peptide
Synthesizer (Applied Biosystems) or manually according to standard
coupling procedures and Fmoc/But strategy with HATU as a coupling
reagent.33 Fmoc-Arg(Pbf)-OH, Fmoc-Lys(Boc)-OH, Fmoc-Gpa-
(Boc)2, and Fmoc-azaβ3-Arg(Boc)2 were used for the protection of
side-chain functionalities. After final Fmoc deprotection, N-terminal
acetylation was carried out in DMF with acetic anhydride (0.5%) and
2,6-lutidine (0.6%). After completion of the synthesis, the protected
peptidyl resins were treated with a cocktail of TFA/H2O/TIS
(95:2.5:2.5 v/v/v) at room temperature for 3 h. The solutions of
the released peptides were filtered and evaporated in vacuo to a
volume of about 1 mL. Then, the peptides were precipitated with
diethyl ether to afford crude products.
EXPERIMENTAL SECTION
■
Synthesis. General Procedures. All solvents and reagents were
obtained from commercial suppliers and were used without further
purification. All amino acid derivatives and coupling reagents were
purchased from Matrix Innovation, Inc. (Montreal, QC, Canada) and
ChemImpex International (Wood Dale, IL, USA). TentaGel S RAM
resin and 2-chlorotrityl chloride resin were purchased from Rapp
Polymere (Tubingen, Germany) and Matrix Innovation, Inc.
̈
(Montreal, QC, Canada), respectively. High-performance liquid
chromatography was carried out on an Agilent Technologies 1100
system (analytical and semipreparative) equipped with a diode array
detector (λ = 210, 230, and 254 nm). The purity of the peptides was
determined with an Agilent Eclipse XDB C18 column (5 μm, 4.6 ×
250 mm). Semipreparative HPLC was carried out using a Zorbax
Eclipse XDB-C18 column (9.4 × 250 mm, 80 Å). The following
solvent systems were used: [A] 0.1% aqueous trifluoroacetic acid
(TFA) and [B] acetonitrile: 0.1% aqueous TFA. All inhibitors were
obtained as TFA salts after lyophilization. SELDI-TOF mass
spectrometry (ProteinChip, Bio-Rad Laboratories, Inc., Hercules,
CA, USA) or MALDI-TOF mass spectrometry and HRMS
(TripleTOF 5600, ABSciex; Foster City, CA, USA) were used to
Analogues modified with Amba or ΔAgm were obtained manually
by a combination of solid-phase peptide synthesis and solution
synthesis. Fmoc-Lys(Boc)-OH (1.2 equiv) was dissolved in dry DCM
(15 mL) and DIPEA (4 equiv) and was immediately added to the 2-
chlorotrityl chloride resin (1 g, 1 equiv, resin loading: 0.8 mmol/g).
The mixture was shaken for 2.5 h at rt followed by treatment with
DCM/MeOH/DIPEA (17:2:1 v/v/v), washed several times with
DCM, DMF, and DCM, and dried in vacuo. The loading of the resin
1
confirm the identity of the pure products. The H and 13C NMR
spectra were recorded on a Bruker Spectrospin 300 NMR instrument
with the chemical shifts reported as δ in ppm. The following
abbreviations were used to describe spin multiplicity: s, singlet; d,
doublet; t, triplet; q, quintet; dd, doublet of doublets; and m, multiplet.
106
dx.doi.org/10.1021/jm401457n | J. Med. Chem. 2014, 57, 98−109

Journal of Medicinal Chemistry
Article
was ∼0.4 mmol/g. The synthesis (which was performed on a scale of
200 μmol for each analogue) proceeded via manual Fmoc SPPS
according to standard coupling procedures using 3-fold excess of Fmoc
amino acids, HATU, or PyBOP (with 6-Cl-HOBt) as coupling agent
and DIPEA or NMM (9 equiv) in DMF. The Fmoc groups were
removed by treatment with 20% piperidine in DMF. After final Fmoc
deprotection, N-terminal acetylation was performed in acetic
anhydride/DIPEA/DCM (15:15:70 v/v/v) for 20 min. The protected
peptide was cleaved from resin by HFIP/DCM (1:4 v/v) for 2 h at rt.
The solvent was removed in vacuo, and the peptide was lyophilized
from 50% t-butanol in water. Protected peptide (1 equiv) and Amba·
2HCl (2 equiv) or ΔAgm(Boc2) (2 equiv), COMU (2 equiv), and
NMM (4 equiv) were dissolved in DMF and stirred for 12 h at rt. The
solvent was removed in vacuo to give a brownish oil. This was
dissolved in TFA/TIS/H2O (95:2.5:2.5 v/v/v), stirred for 3 h at rt,
precipitated by cold ether, washed two times with ether, and
lyophilized.
The FITC peptides were labeled with fluorescein isothiocyanate
isomer I (FITC) through their N terminus via the β-Ala spacer. FITC
(1.5 equiv) in pyridine−DCM (1:4 v/v) was added to the resin and
allowed to couple overnight. Then, peptidyl resins were treated as their
counterparts without the FITC moiety.
All peptides were purified by semipreparative HPLC on a C18
column. The appropriate fractions were pooled and lyophilized. The
purity of the peptides was controlled using analytical HPLC (Agilent
Technologies 1100 system, Waldbronn, Germany) with an Agilent
Elipse XDB C18 column. SELDI-TOF MS or MALDI-TOF MS and
high-resolution mass spectrometry (TripleTOF 5600, ABSciex, Foster
City, CA) were used to confirm the identity of the pure products.
According to HPLC and MS analysis, the purity of the peptides
exceeded 95%. Their physicochemical properties are presented in
Supporting Information Tables S1 and S2.
Biology. Enzyme Kinetics. The human enzymes furin, PC2,
PACE4, and PC7 were expressed in Drosophila Schneider 2 cells and
purified as previously described.34 All measurements were performed
on a Gemini EM 96-well spectrofluorometer (Molecular Devices;
Sunnyvale, CA, USA) (λem, 370 nm; λex, 460 nm; cutoff, 435 nm) with
pyroGlu-Arg-Thr-Lys-Arg-AMC as substrate (Bachem, Switzerland) at
a fixed concentration (Table S4, Supporting Information) in the
presence of various inhibitor concentrations at 37 °C over a period of
1 h. Enzyme inhibition assays for furin were determined in 100 mM
Hepes, pH 7.5, containing 1 mM CaCl2, 1 mM β-mercaptoethanol,
and 1.8 mg/mL of BSA. All activity measurements with PACE4 and
PC7 were performed in 20 mM Bis-Tris, pH 6.5, containing 1 mM
CaCl2 and 1.8 mg/mL of BSA, whereas PC2 assays were performed in
the same buffer at pH 5.7. Kinetics assays were analyzed using
SoftMaxPro5, and Ki values were determined from IC50 using Cheng
and Prusoff’s equation.35 For analogues modified with Amba, the
values for apparent Ki were calculated according to the Morrisson
equation for kinetics of reversible tight-binding inhibitors.28,29
In Vitro Plasma Stability Studies. Stability measurements of
inhibitors were performed by incubating the compounds (0.5 μg/μL)
in CD1 mouse plasma from mixed-sex animals collected with heparin
sodium (Innovative Research; Novi, MI, USA) at 37 °C. At each time
point, the reaction was stopped by adding 150 μL of 1 M guanidium-
HCl solution followed by 300 μL of acetonitrile. Proteins were
removed by centrifugation, and supernatants were analyzed by reverse-
phase HPLC. Peptides AUC were compared to nonincubated samples,
and the percentage remaining was plotted to calculate the half-life
from one-phase decay curves. Each experiment was performed at least
three times in duplicate for each time point. For MALDI-TOF
characterization of peptide fragmentation in murine plasma, 100 μg of
peptide was incubated at 37 °C. After each incubation at the indicated
time points, 5 μL of 3% TFA was added to 10 μL of plasma to stop the
enzymatic reaction. Proteins were precipitated by the addition of 2
volumes of acetonitrile followed by 17 000g centrifugation. Super-
natants were collected, lyophilized, and resolubilized in water with
0.06% TFA. Samples were desalted and partially purified using C18
ZipTip extraction (Millipore) before MALDI-TOF analysis was done
using 1 μL of the desalted sample on the ProteinChip Gold Array
(Bio-Rad) with 1 μL of α-cyano-4-hydroxycinnamic acid (HCCA; 10
mg/mL; 30:70 acetonitrile/water 0.06% TFA) upon analysis on a
ProteinChip SELDI system (Bio-Rad).
Cell Culture. Human prostate cancer cell lines DU145, PC3, and
LNCaP were obtained from ATCC (Manassas, VA, USA) and
maintained in Roswell Park Memorial Institute medium (RPMI 1640)
supplemented with either 5% fetal bovine serum (FBS; Wisent
Bioproducts, St-Bruno, Canada) for DU145 or 10% FBS for PC3 and
LNCaP. Cells were grown at 37 °C in a water-saturated atmosphere
with 5% CO2.
MTT Proliferation Assays. DU145, LNCaP, and PC3 prostate
cancer cells were treated for 72 h with various concentrations of the
selected compounds (peptides 1, 3, 4, 10, 11, 13, 15−18, and 20) to
investigate their growth inhibitory activity against those cells. After 24
h, media was changed, and peptides were added to the cells. Cell
viability was then assessed using the [3-(4,5-dimethylthiazol-2-yl)-2,5-
diphenyltetrazolium] bromide (MTT) assay, as described previously.23
LDH Cytotoxicity Assays. The cytotoxicity of the selected
compounds was determined using the LDH Cytotoxicity Detection
Kit (Promega; Madison, WI, USA), which quantifies the release of
lactate deshydrogenase after membrane disruption by an enzymatic
assay. Briefly, 24 h after cell plating, peptides were added, and LDH
release was measured after 4 h of incubation.
Cellular Uptake. To evaluate the internalization of FITC-labeled
inhibitors, DU145 cells were incubated for 1 h in serum-free RPMI
media with 10 μM of FITC-βAla-[DLeu]LLLRVKR-NH2 and FITC-
βAla-LLLLRVK-Amba and collected by centrifugation. Following the
incubation, cells were washed twice with trypsin (0.05% v/v) for 5 min
at 37 °C to remove nonspecific interactions with the membrane
proteins. Then, cells were incubated for 2 min with propidium iodine
(10 μg/mL) to exclude cells with altered membrane and were
immediately analyzed (10 000 events/sample) by fluorescence
activated cell sorting (FACS; Becton Dickinson; Mountain View,
CA, USA). GeoMeans were determined using CellQuest Software
(Becton Dickinson).
Statistical Analysis. All measurements were made in triplicate, and
all values are expressed as mean SEM (the antiproliferative activity)
and mean SD values (the inhibitory activity).
ASSOCIATED CONTENT
* Supporting Information
■
S
Analytical data (HPLC and MS analysis) for all inhibitors,
general synthetic procedures and analytical data for compounds
V−IX, information for enzyme kinetics, additional inhibition
constants, and MTT assays. This material is available free of
charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Authors
*(Y.L.D.) Telephone: (819) 564-5299. E-mail: yves.dory@
usherbrooke.ca.
■
*(R.D.) Telephone: (819) 564-5428. Fax: (819) 820-6886. E-
mail: robert.day@usherbrooke.ca.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was awarded by Prostate Cancer Canada and is
proudly funded by the Movember Foundation, grant nos. 2012-
951 and D2013-8, the Canadian Cancer Society, grant no.
́
701590, and the “Fondation Mon Etoile” for cancer research.
F.C., C.L., K.L. acknowledge the Fonds de Recherche du
Queb
Franco
́
ec-Sante
́
(FRQS) for studentship support. We thank
̧
is D’Anjou for helpful discussions, Sophie Routhier for
the preparation of enzymes, and Xue Wen Yuan for assistance
in peptide synthesis and purification. We also thank Dr. Leonid
107
dx.doi.org/10.1021/jm401457n | J. Med. Chem. 2014, 57, 98−109

Journal of Medicinal Chemistry
Article
PC7: Use of positional scanning-synthetic peptide combinatorial
libraries as a tool for the optimization of inhibitory sequences. Mol.
Pharmacol. 2007, 71, 323−332.
(11) Becker, G. L.; Sielaff, F.; Than, M. E.; Lindberg, I.; Routhier, S.;
Day, R.; Lu, Y.; Garten, W.; Steinmetzer, T. Potent inhibitors of furin
and furin-like proprotein convertases containing decarboxylated P1
arginine mimetics. J. Med. Chem. 2010, 53, 1067−1075.
(12) Becker, G. L.; Lu, Y.; Hardes, K.; Strehlow, B.; Levesque, C.;
Lindberg, I.; Sandvig, K.; Bakowsky, U.; Day, R.; Garten, W.;
Steinmetzer, T. Highly potent inhibitors of proprotein convertase furin
as potential drugs for treatment of infectious diseases. J. Biol. Chem.
2012, 287, 21992−22003.
(13) Jiao, G. S.; Cregar, L.; Wang, J.; Millis, S. Z.; Tang, C.; O’Malley,
S.; Johnson, A. T.; Sareth, S.; Larson, J.; Thomas, G. Synthetic small
molecule furin inhibitors derived from 2,5-dideoxystreptamine. Proc.
Natl. Acad. Sci. U.S.A. 2006, 103, 19707−19712.
(14) Senzer, N.; Barve, M.; Kuhn, J.; Melnyk, A.; Beitsch, P.; Lazar,
M.; Lifshitz, S.; Magee, M.; Oh, J.; Mill, S. W.; Bedell, C.; Higgs, C.;
Kumar, P.; Yu, Y.; Norvell, F.; Phalon, C.; Taquet, N.; Rao, D. D.;
Wang, Z.; Jay, C. M.; Pappen, B. O.; Wallraven, G.; Brunicardi, F. C.;
Shanahan, D. M.; Maples, P. B.; Nemunaitis, J. Phase I trial of “bi-
shRNAi(furin)/GMCSF DNA/autologous tumor cell” vaccine
(FANG) in advanced cancer. Mol. Ther. 2012, 20, 679−686.
(15) Crunkhorn, S. Trial watch: PCSK9 antibody reduces LDL
cholesterol. Nat. Rev. Drug Discovery 2012, 11, 11.
(16) Khatib, A. M.; Siegfried, G.; Prat, A.; Luis, J.; Chretien, M.;
Metrakos, P.; Seidah, N. G. Inhibition of proprotein convertases is
associated with loss of growth and tumorigenicity of HT-29 human
colon carcinoma cells: Importance of insulin-like growth factor-1
(IGF-1) receptor processing in IGF-1-mediated functions. J. Biol.
Chem. 2001, 276, 30686−30693.
(17) Maquoi, E.; Noel, A.; Frankenne, F.; Angliker, H.; Murphy, G.;
Foidart, J. M. Inhibition of matrix metalloproteinase 2 maturation and
HT1080 invasiveness by a synthetic furin inhibitor. FEBS Lett. 1998,
424, 262−266.
(18) Remacle, A. G.; Gawlik, K.; Golubkov, V. S.; Cadwell, G. W.;
Liddington, R. C.; Cieplak, P.; Millis, S. Z.; Desjardins, R.; Routhier,
S.; Yuan, X. W.; Neugebauer, W. A.; Day, R.; Strongin, A. Y. Selective
and potent furin inhibitors protect cells from anthrax without
significant toxicity. Int. J. Biochem. Cell Biol. 2010, 42, 987−995.
(19) Coppola, J. M.; Bhojani, M. S.; Ross, B. D.; Rehemtulla, A. A
small-molecule furin inhibitor inhibits cancer cell motility and
invasiveness. Neoplasia 2008, 10, 363−370.
(20) Komiyama, T.; Coppola, J. M.; Larsen, M. J.; van Dort, M. E.;
Ross, B. D.; Day, R.; Rehemtulla, A.; Fuller, R. S. Inhibition of furin/
proprotein convertase-catalyzed surface and intracellular processing by
small molecules. J. Biol. Chem. 2009, 284, 15729−15738.
(21) D’Anjou, F.; Routhier, S.; Perreault, J. P.; Latil, A.; Bonnel, D.;
Fournier, I.; Salzet, M.; Day, R. Molecular validation of PACE4 as a
target in prostate cancer. Transl. Oncol. 2011, 4, 157−172.
(22) Couture, F.; D’Anjou, F.; Desjardins, R.; Boudreau, F.; Day, R.
Role of proprotein convertases in prostate cancer progression.
Neoplasia 2012, 14, 1032−1042.
(23) Levesque, C.; Fugere, M.; Kwiatkowska, A.; Couture, F.;
Desjardins, R.; Routhier, S.; Moussette, P.; Prahl, A.; Lammek, B.;
Appel, J. R.; Houghten, R. A.; D’Anjou, F.; Dory, Y. L.; Neugebauer,
W.; Day, R. The Multi-Leu peptide inhibitor discriminates between
PACE4 and furin and exhibits antiproliferative effects on prostate
cancer cells. J. Med. Chem. 2012, 55, 10501−10511.
(24) Judkins, B. D.; Allen, D. G.; Cook, T. A.; Evans, B.; Sardharwala,
T. E. A versatile synthesis of amidines from nitriles via amidoximes.
Synth. Commun. 1996, 26, 4351−4367.
Volkov for his help with flow cytometry experiments and Dr.
Hugo Gagnon (PhenoSwitch Biosciences Inc.) for high-
resolution mass spectrometry analysis.
ABBREVIATIONS USED
■
6-Cl-HOBt, 1-hydroxy-6-chloro-benzotriazole; Ac, acetyl;
ΔAgm, 2,3-dehydroagmatine; Amba, 4-amidinobenzylamide;
azaβ3-R, azaβ3-arginine; CMK, chloromethyl ketone; COMU,
1-cyano-2-ethoxy-2-oxoethylidenaminooxy)dimethylamino-
morpholino-carbenium hexafluorophosphate; DCM, dichloro-
methane; dec, decanoyl; DIPEA, N,N-diisopropylethylamine;
FITC, fluorescein isothiocyanate isomer I; Fmoc, 9-fluorenyl-
methoxycarbonyl; DMF, dimethylformamide; Gpa, 4-guanidi-
no-L-phenylalanine; HATU, 2-(1H-7-azabenzotriazol-1-yl)-
1,1,3,3-tetramethyluronium hexafluorophosphate; HFIP, hexa-
fluoro-2-propanol; HRMS, high-resolution mass spectrometry;
IC50, concentration of peptides required to inhibit 50% of the
cell growth; Ki, inhibition constant; Nle, norleucine; NMM, N-
methylmorpholine; MALDI-TOF MS, matrix-assisted laser
desorption/ionization time-of-flight mass spectrometry; ML,
Multi-Leu; PCa, prostate cancer; PCs, proprotein convertases;
PyBOP, benzotriazole-1-yl-oxy-tris-pyrrolidino-phosphonium
hexafluorophosphate; SELDI-TOF MS, surface-enhanced laser
desorption/ionization time-of-flight mass spectrometry; SPPS,
solid-phase peptide synthesis; t1/2, half-life; TFA, trifluoror-
acetic acid; TIS, triisopropylsilane
REFERENCES
■
(1) Fugere, M.; Day, R. Cutting back on pro-protein convertases:
The latest approaches to pharmacological inhibition. Trends
Pharmacol. Sci. 2005, 26, 294−301.
(2) Hosaka, M.; Nagahama, M.; Kim, W. S.; Watanabe, T.;
Hatsuzawa, K.; Ikemizu, J.; Murakami, K.; Nakayama, K. Arg-X-Lys/
Arg-Arg motif as a signal for precursor cleavage catalyzed by furin
within the constitutive secretory pathway. J. Biol. Chem. 1991, 266,
12127−12130.
(3) Pasquato, A.; Pullikotil, P.; Asselin, M. C.; Vacatello, M.; Paolillo,
L.; Ghezzo, F.; Basso, F.; Di Bello, C.; Dettin, M.; Seidah, N. G. The
proprotein convertase SKI-1/S1P. In vitro analysis of Lassa virus
glycoprotein-derived substrates and ex vivo validation of irreversible
peptide inhibitors. J. Biol. Chem. 2006, 281, 23471−23481.
(4) Benjannet, S.; Rhainds, D.; Essalmani, R.; Mayne, J.; Wickham,
L.; Jin, W.; Asselin, M. C.; Hamelin, J.; Varret, M.; Allard, D.; Trillard,
M.; Abifadel, M.; Tebon, A.; Attie, A. D.; Rader, D. J.; Boileau, C.;
Brissette, L.; Chretien, M.; Prat, A.; Seidah, N. G. NARC-1/PCSK9
and its natural mutants: zymogen cleavage and effects on the low
density lipoprotein (LDL) receptor and LDL cholesterol. J. Biol. Chem.
2004, 279, 48865−48875.
(5) Couture, F.; D’Anjou, F.; Day, R. On the cutting edge of
proprotein convertase pharmacology: From molecular concepts to
clinical applications. Biomol. Concepts 2011, 2, 421−438.
(6) Taylor, N. A.; Van De Ven, W. J.; Creemers, J. W. Curbing
activation: Proprotein convertases in homeostasis and pathology.
FASEB J 2003, 17, 1215−1227.
(7) Seidah, N. G.; Prat, A. The biology and therapeutic targeting of
the proprotein convertases. Nat. Rev. Drug Discovery 2012, 11, 367−
383.
(8) Anderson, E. D.; Thomas, L.; Hayflick, J. S.; Thomas, G.
Inhibition of HIV-1 gp160-dependent membrane fusion by a furin-
directed alpha 1-antitrypsin variant. J. Biol. Chem. 1993, 268, 24887−
24891.
(9) Cameron, A.; Appel, J.; Houghten, R. A.; Lindberg, I.
Polyarginines are potent furin inhibitors. J. Biol. Chem. 2000, 275,
36741−36749.
(10) Fugere, M.; Appel, J.; Houghten, R. A.; Lindberg, I.; Day, R.
Short polybasic peptide sequences are potent inhibitors of PC5/6 and
(25) Schweinitz, A.; Sturzebecher, A.; Sturzebecher, U.; Schuster, O.;
Sturzebecher, J.; Steinmetzer, T. New substrate analogue inhibitors of
factor Xa containing 4-amidinobenzylamide as P1 residue: Part 1. Med.
Chem. 2006, 2, 349−361.
(26) Jeon, H. B.; Lee, Y.; Qiao, C.; Huang, H.; Sayre, L. M. Inhibition
of bovine plasma amine oxidase by 1,4-diamino-2-butenes and -2-
butynes. Bioorg. Med. Chem. 2003, 11, 4631−4641.
108
dx.doi.org/10.1021/jm401457n | J. Med. Chem. 2014, 57, 98−109

Journal of Medicinal Chemistry
Article
(27) Busnel, O.; Bi, L.; Dali, H.; Cheguillaume, A.; Chevance, S.;
Bondon, A.; Muller, S.; Baudy-Floc’h, M. Solid-phase synthesis of
“mixed” peptidomimetics using Fmoc-protected aza-beta3-amino acids
and alpha-amino acids. J. Org. Chem. 2005, 70, 10701−10708.
(28) Morrison, J. F. Kinetics of the reversible inhibition of enzyme-
catalysed reactions by tight-binding inhibitors. Biochim. Biophys. Acta
1969, 185, 269−286.
(29) Williams, J. W.; Morrison, J. F. The kinetics of reversible tight-
binding inhibition. Methods Enzymol. 1979, 63, 437−467.
(30) Pauletti, G. M.; Gangwar, S.; Siahaan, T. J.; Aube, J.; Borchardt,
R. T. Improvement of oral peptide bioavailability: Peptidomimetics
and prodrug strategies. Adv. Drug Delivery Rev. 1997, 27, 235−256.
(31) Kacprzak, M. M.; Peinado, J. R.; Than, M. E.; Appel, J.; Henrich,
S.; Lipkind, G.; Houghten, R. A.; Bode, W.; Lindberg, I. Inhibition of
furin by polyarginine-containing peptides: Nanomolar inhibition by
nona-D-arginine. J. Biol. Chem. 2004, 279, 36788−36794.
(32) Becker, G. L.; Hardes, K.; Steinmetzer, T. New substrate
analogue furin inhibitors derived from 4-amidinobenzylamide. Bioorg.
Med. Chem. Lett. 2011, 21, 4695−4697.
(33) Fields, G. B.; Noble, R. L. Solid phase peptide synthesis utilizing
9-fluorenylmethoxycarbonyl amino acids. Int. J. Pept. Protein Res. 1990,
35, 161−214.
(34) Fugere, M.; Limperis, P. C.; Beaulieu-Audy, V.; Gagnon, F.;
Lavigne, P.; Klarskov, K.; Leduc, R.; Day, R. Inhibitory potency and
specificity of subtilase-like pro-protein convertase (SPC) prodomains.
J. Biol. Chem. 2002, 277, 7648−7656.
(35) Cheng, Y.; Prusoff, W. H. Relationship between the inhibition
constant (KI) and the concentration of inhibitor which causes 50%
inhibition (I50) of an enzymatic reaction. Biochem. Pharmacol. 1973,
22, 3099−3108.
109
dx.doi.org/10.1021/jm401457n | J. Med. Chem. 2014, 57, 98−109