Synthesis and enzymatic evaluation of novel partially fluorinated thiol dual ACE/NEP inhibitors

Article abstract of DOI:10.1016/j.bmcl.2009.06.064

A novel family of peptidomimetics incorporating fluoroalkyl groups, mainly a trifluoromethyl, in α-position to a zinc(II)-binding thiol function, was synthesized in solution as well as in solid-phase. These compounds showed inhibitory potency in the nanom

Full text of DOI:10.1016/j.bmcl.2009.06.064

Bioorganic & Medicinal Chemistry Letters 19 (2009) 4715–4719
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and enzymatic evaluation of novel partially fluorinated thiol dual
ACE/NEP inhibitors
Francesca Olimpieri a, Simone Tambaro b, Santos Fustero c, Paolo Lazzari b, Maria Sanchez-Roselló c,
e,f,
Luca Pani b,d, Alessandro Volonterio a,e, , Matteo Zanda
*
*
a Dipartimento di Chimica, Materiali, e Ingegneria Chimica ‘Giulio Natta’, via Mancinelli 7, 20131 Milano, Italy
b Neuroscienze PharmaNess S.c.a.r.l., c/o Parco Scientifico e Tcnologico della Sardegna, Edificio 5, Loc. Piscinamanna, 09010 Pula (CA), Italy
c Departamento di Quimica Organica, Universitad de Valencia, 46100 Burjassot, Spain
d C.N.R.-Istituto di Tecnologie Biomediche, Sezione di Cagliari, c/o Neuroscienze PharmaNess S.c.a.r.l., Loc. Piscinamanna, 09010 Pula (CA), Italy
e C.N.R.-Istituto del Riconoscimento Molecolare, via Mancinelli 7, 20131 Milano, Italy
f Institute of Medical Sciences, University of Aberdeen, Foresterhill, Aberdeen AB25 2ZD, Scotland, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
A novel family of peptidomimetics incorporating fluoroalkyl groups, mainly a trifluoromethyl, in a-posi-
Received 28 April 2009
Revised 15 June 2009
Accepted 16 June 2009
Available online 21 June 2009
tion to a zinc(II)-binding thiol function, was synthesized in solution as well as in solid-phase. These com-
pounds showed inhibitory potency in the nanomolar range against both angiotensin-converting enzyme
(ACE) and neutral endopeptidase (NEP), whereas no inhibition of endothelin-converting enzyme-1 (ECE-
1) was observed. The trifluoromethyl-derivatives were more potent than the parent unfluorinated ana-
logues in the case of ACE, and less potent in the case of NEP.
Keywords:
Ó 2009 Elsevier Ltd. All rights reserved.
ACE inhibitors
NEP inhibitors
Fluorine
Zinc metallopeptidases
Michael reaction
Angiotensin-converting enzyme (ACE), neutral endopeptidase
(NEP), and endothelin-converting enzyme-1 (ECE-1) are zinc
metallopeptidases located on the outer membrane of various cell
types (ectoenzymes). They have an important role in the metabo-
lism of a number of regulatory peptides of human nervous, cardio-
vascular, inflammatory and immune system. ACE is a dipeptidyl
carboxypeptidase which converts angiotensin I into angiotensin II
and degrades kinins, while NEP catalyses the degradation of a vari-
ety of renal and CNS-active peptides including substance P, brady-
kinin, enkephalins, atrial natriuresis.1a
ECE-1 cleaves its biologically inactive native substrate, big
endothelin-1, to generate a powerful vasoactive 21-amino acid
peptide, endothelin-1.1b Inhibition of ACE has been a highly suc-
cessful program in the pharmaceutical industry resulting in a num-
ber of effective antihypertensive drugs.2 Elegant studies relying on
structure-based design have laid a strong foundation for early pio-
neering efforts in the development of Captopril3 and its variants.4
On the basis of these positive results in a range of cardiovascular
diseases, including hypertension, diabetes and renal disease, sev-
eral investigators have started to study the potentially added ben-
eficial effects by blocking other endogenous enzymes involved in
peptide activation or degradation. In various experimental models,
combined ACE/NEP inhibition led to more potent and synergistic
hemodynamic and renal effects than selective inhibitors of the
individual enzymes.5 The synergistic effect of combined ACE and
NEP inhibition is based on similar modes of action, including block-
ade of angiotensin synthesis and simultaneously unmasking and
potentiation of the effect of peptides such as ANP, BNP, and brady-
kinin (by preventing their degradation), in turn inducing vasodila-
tation and diuresis and improving myocardial function. However,
the earliest dual metalloprotease inhibitors had limitations due
to low potency, short duration of action, or limited oral
bioavailability.
Inhibitory potencies in the nanomolar range for the NEP
enzyme were achieved with pseudo-dipeptides and tripeptides
designed from the structure of thiorphan, of general formula
HS–CH(R1)–CH(R2)–CONH–CH(R3)–COOH and HS–CH(R1)–CH(R2)–
CONH–CH(R3)–CONH–CH(R4)–COOH, respectively,6 interacting
with S10 and S20 subsites of the enzymes and bearing a thiol group
as zinc-chelating moiety. Additional interactions with the S1 bind-
ing subsite probably account for the subnanomolar inhibitory
potencies for the ACE enzyme, as evidenced by the construction
of a hypothetical model of the ACE and NEP active sites (Fig. 1).7
* Corresponding authors. Tel.: +39 02 23993139; fax: +39 02 23993080 (A.V.);
tel.: +39 02 23993084 (M.Z.).
E-mail addresses: alessandro.volonterio@polimi.it (A. Volonterio), matteo.zan
da@polimi.it (M. Zanda).
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.06.064

4716
F. Olimpieri et al. / Bioorg. Med. Chem. Lett. 19 (2009) 4715–4719
For the synthesis of the target compounds, we decided to ex-
S1'
S2'
R3
S1'
R2
S2'
O-
ploit the high Michael-type reactivity toward nucleophiles of
C@C bonds activated by an EW fluoroalkyl group in b position.12
Thus, trifluorocrotonic acid 6a and trans-b-pentafluoroethyl acrylic
acid 6b (Scheme 1) were prepared as described in the literature.13
The intermediate fluorocrotonoyl chlorides were obtained in good
yields by distillation of the refluxing mixture of the acids 6a,b and
phthaloyl dichloride, and subsequently condensed with different
S3'
O
O
R2
-S
O-
H
N
-S
N
H
O
R1
R1
O
R3
a
-aminoacid methylesters, such as H-Phe-OMe, H-tert-BuO-Tyr-
S1
OMe, H-4-F-Phe-OMe, and H-Trp-OMe affording the Michael
ACE binding site
NEP binding site
acceptors 7a–e, respectively.
Figure 1. Cartoon outlining the ACE/NEP binding sites.
The Michael addition of thiolacetic acid, which is a rather poor
nucleophile, to 7a–e afforded pseudo-dipeptides 8a–e (Scheme 1)
as equimolecular mixtures of the two diastereoisomers in very
good yields (70–89%) when performed in polar solvents such as
methanol and when the fluoroalkyl substituent is the highly EW
trifluoromethyl group. The yield dropped (33%) when the less
EW, and thus less activating, pentafluoroethyl group was present
instead. Before the final step, derivative 8b (R1 = 4-tert-BuO–Ph–
CH2–) was treated with TFA in DCM in order to deprotect the phe-
nolic hydroxy group. Finally, hydrolysis of both the ester and thio-
ester functions was accomplished in quantitative yields with a
degassed 2 N methanolic solution of NaOH, in order to avoid air
oxidation to the corresponding disulfides.
To synthesize the pseudo-tripeptide 4 (Scheme 2), the acylchlo-
ride derivative of 6a was coupled with H-Phe-Otert-Bu affording
the Michael acceptor 9 in good yield (Scheme 2). The tert-butyl
ester was then hydrolyzed (TFA/DCM), and the resulting carboxylic
acid was coupled with H-(4-tert-Bu)Tyr-OMe using EDC/HOBt pro-
viding the Michael acceptor 10.
Selective incorporation of one or more fluorine atoms in specific
points of an organic molecule can be an effective strategy for
improving and modifying its biological activity, for example by
inducing a local modification of charge or polarity or conformation,
which in turn can have a profound impact in terms of binding to
the receptor site. Alternatively introduction of fluorine can be used
to increase the metabolic stability of a bioactive molecule, for
example by preventing oxidative or hydrolytic metabolism.8 In
particular, the trifluoromethyl group is recognized in medicinal
chemistry as a substituent of distinctive qualities. It is, in fact, at
the same time highly hydrophobic, electron-withdrawing (EW),
and rather sterically demanding, moreover it can provide high
in vivo stability, and can behave as a good mimic of several natu-
rally occurring residues such as ethyl, isopropyl, phenyl, etc.9 For
example, we previously reported that the trifluoromethyl group
can replace and mimic an isobutyl group in peptidomimetic inhib-
itors of proteases.9e
Within the framework of a project aimed at the investigation of
the effect of fluoroalkyl groups on the binding process to metallo-
proteases,10 we designed novel thiol-containing pseudodi- and tri-
peptides analogues 3a–e and 4, respectively, in which the isobutyl
substituent of known inhibitors 1 and 2 is replaced by a fluoroalkyl
group (Fig. 2).
Michael addition of thiolacetic acid in MeOH led to the forma-
tion of the adduct 11 in very good yield, as an equimolecular mix-
ture of two diastereoisomers. The protected compound 11 was
then converted into the target pseudo-tripeptide 4 by cleavage of
the tert-butyl protecting group followed by hydrolysis of methyl-
and thiolesters under standard conditions (see above).
This was done with the hope of (1) better understanding the
importance of the S1 pocket in the active site of both peptidases,
(2) increasing both the ACE and NEP affinity of such inhibitors as
compared to the known inhibitors 1 and 2, thus generating new
hit compounds for further development by introducing suitable
substituents in P10 position, (3) deeply modifying the physico-
chemical properties of the inhibitors, such as local hydrophobicity,
acidity and nucleophilicity of the neighboring thiol group, pre-
ferred conformation (inducing conformational constrictions due
to the sterically demanding fluoroalkyl groups) and (4) provide
the basis for a spectroscopic study of the binding process through
19F NMR.11
In order to skip the chromatographic purification of the inter-
mediates, we decided to apply the same synthetic strategy also
in solid phase. The free aminic function of resin-bound phenylala-
nine 12 (Scheme 3) was coupled with a slight excess of acylchlo-
ride derivative of 6a leading to the solid-supported Michael
acceptor 13. The latter was reacted with thiolacetic acid in the
presence of TEA in EtOH for three days, affording the Michael
adduct 14. Both steps could be efficiently monitored by FT-IR spec-
troscopy following the appearance of the typical bands belonging
to the amide and thioester carbonyls, respectively. Finally, by treat-
ing resin 14 with a degassed 2 N methanolic solution of NaOH we
obtained, after filtration, acidification and extraction with organic
RF
O
R
O
HS
N
H
COOH
HS
N
H
COOH
3a-e
RF = CF3, CF2CF3
R = Bn, 4-HO-PhCH2, 4-F-PhCH2, indole-3-CH2
1
O
CF3 O
H
H
N
N
COOH
COOH
HS
N
HS
N
H
H
O
O
4
2
OH
OH
Figure 2. Fluorinated analogues of dual ACE/NEP inhibitors.

F. Olimpieri et al. / Bioorg. Med. Chem. Lett. 19 (2009) 4715–4719
4717
O
R1
Ref. 13
O
a,b
COOH
COOEt
RF
RF
N
H
COOMe
RF
5a,b
6a,b
7a-e
CH3COS
RF
O
R1
SH
O
R1
d,e
c
N
COOR2
RF
N
COOH
H
H
8a-e
3a-e
Scheme 1. Typical synthesis of fluorinated inhibitors. Reagents and conditions: (a) phthaloyl dichloride, 180 °C, (70–77%); (b) H-AA-OMe, TEA, DCM, 0 °C (73–96%); (c) AcSH,
EtOH, rt (rd = 1:1, 30–89%); (d) for R1 = 4-tert-BuO–Ph–CH2–, 20% TFA, DCM, rt, 1 h (98%); (e) degassed 2 N NaOH in MeOH, 0 °C, 1 h (quantitative).
O
O
H
COOH
N
COOMe
c,d
O
a,b
F3C
F3C
N
H
F3C
N
H
O
O
6a
10
9
O
CH3COS
F3C
O
SH
O
e
H
N
H
f,g
COOMe
N
COOH
N
H
F3C
N
H
O
O
11
4
O
OH
Scheme 2. Synthesis of pseudo-tripeptide 4. Reagents and conditions: (a) phtaloyl dichloride, 180 °C, (77%); (b) H-Phe-Otert-Bu, TEA, DCM, 0 °C (83%); (c) 20% TFA, DCM, rt,
1 h (98%); (d) H-(4-tert-Bu)Tyr-OMe, HOBt, EDC, TMP, DCM, rt, 3 h (77%); (e) AcSH, EtOH, rt (r.d. = 1:1, 75%); (f) 20% TFA, DCM, rt, 1 h (98%); (g) degassed 2 N NaOH in MeOH,
0 °C, 1 h (quantitative).
solvent, the final pseudo-dipeptide 3a in high purity (>98%) with-
out any further purification.
ing a Tyrosine terminal group, was the most potent, showing an
inhibitory activity towards NEP similar to those of the known ref-
erence compounds 1 and 2.
All the partially fluorinated pseudo-peptides above were as-
sayed for their capacity to inhibit ACE, NEP, and ECE (Table
1).14 The enzyme inhibitory assays were carried out according
to the general procedures reported in the literature,6b and vali-
dated through the IC50 determination of the following reference
inhibitor compounds: thiorphan, captopril, and phosphoramidon,
for ACE, NEP, and ECE, respectively.15,1b Moreover, an IC50 value
of 41 nM was extrapolated from the experimental curve obtained
with compound 1 synthesized in our laboratories. This result
was equivalent to that previously reported by Gomez-Monterrey
et al. of 37 nM.6a
In general, replacement of the isobutyl group with a trifluoro-
methyl group as the R1 substituent (see Fig. 1) produced a signifi-
cant increase of the ACE inhibitor activity of the thiol compounds
3,4, as compared with the previously reported data on the non-
fluorinated analogues, such as 1,2.6a This effect is flanked by a
clear-cut reduction of selectivity, because these CF3-compounds
showed also a slight decrease of potency towards NEP. Replace-
ment of trifluoromethyl by pentafluoroethyl group (as in 3e)
brought about a reduction of the inhibitory capacity towards both
NEP and ACE.
All the compounds showed inhibitory activity in the nanomolar
range towards both NEP and ACE. In contrast, no inhibitory activity
was evidenced on ECE-1. Among the assayed compounds, 3b, bear-
In conclusion, considering the ACE/NEP dual nanomolar inhibi-
tory activity of these novel trifluoromethyl substituted peptidom-
imetics, these results should pave the way for further biological
O
O
H
NH2
CF3
N
a
O
OH
O
O
Wang resin
13
12
O
H
H
N
CF3
b
HOOC
N
CF3
SCOCH3
c
O
SH
3a Purity > 98%
O
O
14
Scheme 3. Solid phase synthesis of 3a. Reagents and conditions: (a) trans-CF3–C@C–COCl (1.5 equiv), TMP, DCM, rt, 3 h; (b) AcSH, EtOH, rt, 3 days; (c) degassed 2 N NaOH in
MeOH, 0 °C, 1 h.

4718
F. Olimpieri et al. / Bioorg. Med. Chem. Lett. 19 (2009) 4715–4719
Table 1
IC50 values for compounds 1, 2, 3a–e, and 4
R1
O
R2
X
HS
N
H
O
Compounds
R1
R2
X
IC50 (nM)
NEP
ACE
ECE-1
1
OH
>10,000 (>10,000)a
41 (37)a
>10,000
HN
COOH
2a
2800
50
n.d.b
OH
3a
3b
CF3
CF3
OH
OH
830
160
300
80
>10,000
>10,000
OH
F
3c
CF3
CF3
OH
OH
300
400
230
580
>10,000
>10,000
NH
3d
3e
CF2CF3
OH
1010
280
1150
430
2420
HN
COOH
4
CF3
>10,000
OH
a
IC50 values taken from the ref. Gomez-Monterrey et al.6a
Not determined.
b
and pharmaceutical investigations on possible applications of
these derivatives, that is, for the treatment of congestive heart fail-
ure and chronic hypertension.16–18 Furthermore, these results con-
firm that a trifluoromethyl group can be used as a replacement of
an isobutyl group in protease inhibitors, without affecting the
affinity for the active site, but producing a significant overall phys-
ico-chemical and electronic perturbation due to the strong elec-
tronic difference (EW character, polarity, charge, etc.) between
these two groups.9e We are currently investigating structural and
electronic factors that are responsible for the influence of different
fluoroalkyl groups, particularly the trifluoromethyl, on the inhibi-
tory activity and binding affinity of these novel fluorinated metal-
loprotease inhibitors.
Supplementary data
Supplementary data (experimental procedures and 1H NMR for
the final compuonds) associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2009.06.064.
References and notes
1. (a) Skidgel, R. A.; Erdos, E. G. Peptides 2004, 25, 521; (b) Ahn, K.; Herman, S. B.;
Fahnoe, D. C. Arch. Biochem. Biophys. 1998, 359, 258.
2. (a) Lazartigues, E.; Feng, Y.; Lavoie, J. L. Curr. Pharm. Des. 2007, 13, 1231; (b)
Donnelly, R.; Manning, G. JRAAS 2007, 8, 13; (c) Ryan, M. J.; Sigmund, C. D. Circ.
Res. 2004, 94, 1; (d) White, M. C. Pharmacotherapy 1998, 18, 588; (e) Natesh, R.;
Schwager, S. L. U.; Sturrock, E. D.; Acharaya, K. R. Nature 2003, 421, 551.
3. (a) Kubo, S. H.; Cody, R. J. Clin. Pharmacokinet. 1985, 10, 377; (b) Atkinson, A. B.;
Lever, A. F.; Brown, J. J.; Robertson, J. I. S. Lancet 1980, 2, 105.
Acknowledgements
4. Song, J. S.; White, C. M. Clin. Pharmacokinet. 2002, 41, 207.
5. Jandeleit-Dam, K. A. M. J. Hum. Hypertens. 2006, 20, 478.
6. (a) Gomez-Monterrey, I.; Turcaud, S.; Lucas, E.; Bruetschy, L.; Roques, B. P.;
Fournié-Zaluski, M.-C. J. Med. Chem. 1993, 36, 87; (b) Inguimbert, N.; Coric, P.;
Poras, H.; Meudal, H.; Teffot, F.; Fournié-Zaluski, M.-C.; Roques, B. P. J. Med.
Chem. 2002, 45, 1477.
We thank MUR (project ‘nuovi peptidomimetici ad attività anal-
gesica’, protocol no. 13,378, 24/12/2003), C.N.R. and Politecnico di
Milano for economic support.

F. Olimpieri et al. / Bioorg. Med. Chem. Lett. 19 (2009) 4715–4719
4719
7. Bohacek, R.; De Lombaert, S.; McMartin, C.; Priestle, J.; Grütter, M. J. Am. Chem.
Soc. 1996, 118, 8231.
substrate o-aminobenzoic acid-Phe-Arg-Lys(2,4-dinitrophenyl)-Pro-OH [Abz-
FRK(Dnp)-P] was obtained from Biomol (PA, USA). A concentration of Abz-
8. Biomedical Frontiers of Fluorine Chemistry; Ojima, I., McCarthy, J. R., Welch, J. T.,
Eds.ACS Books; American Chemical Society: Washington, DC, 1996.
9. The bioisosterism of the trifluoromethyl group remains a controversial issue.
See for example: (a) Mikami, K.; Itoh, Y.; Yamanaka, M. Chem. Rev. 2004, 104, 1;
(b) Müller, K.; Faeh, C.; Diederich, F. Science 2007, 317, 1881; (c) Leroux, F.
ChemBioChem 2004, 5, 644; (d) Jagodzinska, M.; Huguenot, F.; Candiani, G.;
Zanda, M. ChemMedChem 2009, 4, 49; (e) Binkert, C.; Frigerio, M.; Jones, A.;
Meyer, S.; Pesenti, C.; Prade, L.; Viani, F.; Zanda, M. ChemBioChem 2006, 7, 181.
10. (a) Moreno, M.; Sani, M.; Raos, G.; Meille, S. V.; Belotti, D.; Giavazzi, R.; Bellosta,
S.; Volonterio, A.; Zanda, M. Tetrahedron 2006, 62, 10171; (b) Sani, M.;
Candiani, G.; Pecker, F.; Malpezzi, L.; Zanda, M. Tetrahedron Lett. 2005, 46,
2393; (c) Sinisi, R.; Sani, M.; Candiani, G.; Parente, R.; Pecker, F.; Bellosta, S.;
Zanda, M. Tetrahedron Lett. 2005, 46, 6515; (d) Molteni, M.; Volonterio, A.;
Fossati, G.; Lazzari, P.; Zanda, M. Tetrahedron Lett. 2007, 48, 589.
FRK(Dnp)-P of 10
lM (Km determined value) was adopted and the fluorescence
was measured at kex = 320 nm and kem = 420 nm. NEP Activity: The assay was
carried out with commercial purified NEP from porcine kidney (Calbiochem,
Germany), specific fluorescent substrate N-Dansyl-D-Ala-Gly-p-nitro-Phe-Gly
(DAGNPG) purchased from Sigma–Aldrich (St. Louis, MO). The assays were
performed at kex = 340 nm and kem = 530 nm using a DAGNPG concentration of
58 lM (Km determined value). ECE-1 Activity: Human recombinant ECE-1 and
fluorescent specific substrate 7-methoxycoumarin-4-acetyl-Arg-Pro-Pro-Gly-
Phe-Ser-Ala-Phe-Lys(2,4-dinitrophenyl)-OH [Mca-RPPGFSAFK(Dnp)-OH] was
purchased from R&D Systems (Minneapolis, MN). The assays were carried
out at kex = 340 nm and kem = 405 nm with
concentration of 16 M (Km determined value).
a Mca-RPPGFSAFK(Dnp)-OH
l
15. The IC50 values of the reference compounds we obtained were tiorphan IC50 of
6.2 nM versus NEP, captopril IC50 of 0.12 nM versus ACE and phosphoramidon
IC50 of 17.0 nM versus ECE. All the determined values were in good agreement
with previously reported data: (a) Roques, B. P.; Fournié-Zaluski, M. C.; Soroca,
E.; Lecomte, J. M.; Malfroy, B.; Llorens, C.; Schwartz, J. C. Nature 1980, 288, 286;
(b) Binevski, P. V.; Sizova, E. A.; Pozdnev, V. F.; Kost, O. A. FEBS Lett. 2003, 550,
84.
16. Tikkanen, T.; Tikkanen, I.; Rockell, M. D.; Allen, T. J.; Johnston, C. I.; Cooper, M.
E.; Burrell, L. M. Hypertension 1998, 32, 778.
17. Pu, Q.; Schiffrin, E. L. Am. J. Hypertens. 2001, 14, 1067.
11. Gerig, J. T. Fluorine Magnetic Resonance in Biochemistry. In Biological Magnetic
Resonance; Berliner, L. S., Reuben, J., Eds.; Plenum: New York, 1978; Vol. I, p
139.
12. Billard, T. Chem. Eur. J. 2006, 12, 974.
13. Jagodzinska, M.; Huguenot, F.; Zanda, M. J. Fluorine Chem. 2007, 63, 2042.
14. All the enzyme inhibitory assays were carried out with a Victor3 Fluorometer
(PerkinElmer) according to the general procedures reported in Ref. 6b. ACE
Activity: Human recombinant Angiotensin-converting Enzyme-1 (ACE) was
purchased from Calbiochem (Darmstadt, Germany), specific fluorescent
18. Johnson, A. G.; Pearce, G. L.; Danoff, T. M. J. Hum. Hypertens. 2006, 20, 496.