Incorporation of neutral C-terminal residues in 3-amidinophenylalanine-derived matriptase inhibitors

Article abstract of DOI:10.1016/j.bmcl.2009.02.047

A novel series of matriptase inhibitors based on previously identified tribasic 3-amidinophenylalanine derivatives was prepared. The C-terminal basic group was replaced by neutral residues to reduce the hydrophilicity of the inhibitors. The most potent co

Full text of DOI:10.1016/j.bmcl.2009.02.047

Bioorganic & Medicinal Chemistry Letters 19 (2009) 1960–1965
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Incorporation of neutral C-terminal residues in
3-amidinophenylalanine-derived matriptase inhibitors
Andrea Schweinitz a, Daniel Dönnecke a, Alexander Ludwig a, Peter Steinmetzer a, Alexander Schulze a,
Joscha Kotthaus b, Silvia Wein c, Bernd Clement b, Torsten Steinmetzer d,
*
a The Medicines Company (Leipzig) GmbH, Deutscher Platz 5d, D-04103 Leipzig, Germany
b Department of Pharmaceutical and Medicinal Chemistry, Pharmaceutical Institute, Christian-Albrechts-University Kiel, Gutenbergstr. 76, D-24118 Kiel, Germany
c Institute of Animal Nutrition and Physiology, Christian-Albrechts-University Kiel, Hermann-Rodewald-Strasse 9, D-24098 Kiel, Germany
d Institute of Pharmaceutical Chemistry, Philipps University Marburg, Marbacher Weg 6, D-35032 Marburg, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
A novel series of matriptase inhibitors based on previously identified tribasic 3-amidinophenylalanine
derivatives was prepared. The C-terminal basic group was replaced by neutral residues to reduce the
hydrophilicity of the inhibitors. The most potent compound 22 inhibits matriptase with a Ki value of
0.43 nM, but lacks selectivity towards factor Xa. By combination with neutral N-terminal sulfonyl resi-
dues several potent thrombin inhibitors were identified, which had reduced matriptase affinity.
Ó 2009 Elsevier Ltd. All rights reserved.
Received 6 January 2009
Revised 10 February 2009
Accepted 11 February 2009
Available online 14 February 2009
Keywords:
Matriptase
Serine protease
Protease inhibitor
Anticancer drugs
Matriptase, also named MT-SP1, TADG-15, ST14, or epithin in
mice, is a type II transmembrane serine protease that contains a
short N-terminal intracellular domain and a larger extracellular re-
gion.1 Matriptase has a complex multidomain structure, with its
trypsin-like protease domain located at the C-terminus of the pro-
tein on the cell surface. Many insights regarding the physiological
role of matriptase have been obtained with matriptase-null mutant
mice. Matriptase has a critical role in the proper development of a
number of epithelial tissues, including epidermis, oral epithelium,
thymic epithelium, and hair follicles. Matriptase knockout mice die
shortly after birth due to severe dehydration and impaired epider-
mal barrier function.2 These studies with knockout mice revealed a
defect in the proteolytic processing of the epidermal polyprotein
profilaggrin into filaggrin monomer units, which are required for
proper epidermis development.3 In addition, matriptase may also
act as an activator of prostasin, a glycophosphatidylinositol-an-
chored membrane serine protease. This was supported by the
observation that matriptase and prostasin knockout mice have
nearly identical phenotypes.
mains.4 Recently matriptase has also been identified as a potential
drug target in oncology based on many reports that have causally
linked the overexpression of matriptase or an increased matrip-
tase/HAI-1 ratio to a variety of epithelial tumors.5 However, it must
be noted that in some tumor studies also a downregulation of
matriptase and HAI-1 expression was detected.6
In addition to both profilaggrin and prostasin, which have to be
processed for normal embryogenesis and epithelial development,
other potential matriptase substrates have been identified, which
are probably involved in tumorigenicity. Among them are the pro-
form of hepatocyte growth factor, also named scatter factor (pro-
HGF/SF),7 pro-uPA, the protease activated receptor PAR-2,8 the
insulin-like growth factor binding protein-related protein-1
(IGFBP-rP1),9 and stromelysin (MMP-3).10
In a first publication on matriptase inhibitors we described the
development of tribasic secondary amides of sulfonylated 3-ami-
dinophenylalanine, such as 1 (Table 1).11 Despite excellent potency
for matriptase and high selectivity towards other trypsin-like ser-
ine proteases, such compounds have no potential for oral bioavail-
ability due to their high hydrophilicity and consequently poor
penetration. Therefore, with the assumption that it should be pos-
sible to convert the 3-amidinophenylalanine into a less basic, more
hydrophobic hydroxyamidino prodrug similar to that used for the
anti-metastatic uPA inhibitor mesupron and the thrombin inhibi-
tor prodrug ximelagatran,12 we focused our efforts on the elimina-
tion of the positively charged groups in the N- and C-terminal
Under normal conditions the activity of matriptase is tightly
regulated by the endogenous hepatocyte growth factor activator
inhibitors HAI-1 and/or HAI-2, which are type I transmembrane
proteins that contain two Kunitz-type serine protease inhibitor do-
* Corresponding author. Tel.: +49 6421 2825900; fax: +49 6421 2825901.
E-mail address: steinmetzer@staff.uni-marburg.de (T. Steinmetzer).
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.02.047

A. Schweinitz et al. / Bioorg. Med. Chem. Lett. 19 (2009) 1960–1965
1961
Table 1
Inhibition of matriptase and other trypsin-like serine proteases by inhibitors of the general formula
O
O
H
N
H2N
N
S
R
H
O2
NH2
HN
No.
1
R
Ki (lM)
Matriptase
Plasmin
uPA
Thrombin
6.0
fXa
9.5
N
NH
NH2
0.0018
5.0
1.7
N
H
N
N
O
O
2
3
0.255
3.3
2.8
17
14
18
OH
O
0.0061
2.5
0.60
6.1
N
O
O
4
5
6
0.0063
0.012
1.6
2.9
1.3
3.3
2.1
3.3
0.85
0.67
1.4
10
8.2
19
N
H
N
N
NH2
0.0075
N
O
O
O
N
N
7
8
0.013
0.017
1.2
2.8
3.3
2.2
1.25
0.35
2.5
3.4
NH2
O
9
0.031
1.3
7.4
2.6
2.1
15
N
NH2
N
N
10
11
0.033
0.037
3.3
0.27
71.7
10
2.56
0.88
2.1
NH
N
12
0.053
6.2
3.0
0.11
4.4
CH3
(continued on next page)

1962
A. Schweinitz et al. / Bioorg. Med. Chem. Lett. 19 (2009) 1960–1965
Table 1 (continued)
No.
R
Ki (lM)
Matriptase
0.053
Plasmin
3.6
uPA
Thrombin
29
fXa
44
N
N
N
O
13
16.9
9.2
N
N
N
CH3
N
H
O
O
14
0.065
0.15
0.18
13.4
2.5
5.1
14.5
2.3
26
O
CH3
15
16
8.37
2.5
OH
N
9.4
7.3
11
17
HOOC
N
17
18
0.47
0.1
4.5
4.9
34.5
76
0.395
24.2
H3C
N
N
0.10
O
H3C
O
O
19
20
2.4
69.4
79
126
10.6
127
30
O
H3C
N
11
193
159
HO
O
inhibitor segments. The results of the N-terminal modification
were previously described,13 in the present publication we report
the results of the C-terminal replacements.
introduced an oppositely charged 4-piperidylbutanoic acid. As ex-
pected, compound 2 has only poor activity (Table 1); however, a
high potency was retained for the analogous methyl ester 3, which
was prepared as a precursor of 2. Replacement of the somewhat la-
bile methyl ester by the isosteric methyl amide resulted in similar
potency (4), whereas the activity of the non-substituted amide was
slightly reduced (5). A relatively high potency was observed also
for the Cbz-protected piperazide 6, the isonepecotic acid amide
7, and the benzylpiperidide 8. Most of these analogues retained
sufficient selectivity for matriptase compared to the other exam-
ined trypsin-like serine proteases. Additional modifications re-
sulted in significantly reduced matriptase affinity (9–20).
O
O
H
N
H2N
N
H
S
N
NH
NH2
O2
N
H
NH2
1: Ki = 1.8 nM
HN
The X-ray structure of a similar inhibitor (number 8 in our pre-
vious publication)11 complexed with matriptase revealed that the
guanidinoethyl group is directed into a negatively charged cleft
of matriptase lined by the carboxylate groups of Asp96 and Asp60b
and the carbonyl groups of His57 and Ile60. However, this guanidyl
group was not directly involved in contacts to the protease and
formed only some water-mediated hydrogen bonds. To probe the
importance of the guanidyl group for matriptase inhibition we
In the further course of this work, the N-terminal b-alanyl
amide was replaced with selected groups as described in the pre-
ceding paper.13 We maintained the C-terminal 4-piperidylbutanoyl
methylamide, which should have an improved stability over the
analogous methyl ester and which has also a slightly reduced
molecular weight compared to the similar potent Cbz-protected
piperazide (Table 2). As found previously with similar analogues,13
also in this series inhibitor 22 containing an N-terminal 3-(6-ami-

A. Schweinitz et al. / Bioorg. Med. Chem. Lett. 19 (2009) 1960–1965
1963
Table 2
Inhibition of matriptase and other trypsin-like serine proteases by inhibitors of the general formula
O
H
N
R
S
N
O
O2
N
H
NH2
HN
No.
21
R
Ki (lM)
Matriptase
Plasmin
1.17
uPA
Thrombin
0.01
fXa
N
N
0.010
0.00043
0.016
0.028
0.024
0.087
0.024
0.030
0.098
9.5
0.163
0.0057
6.93
H2N
H2N
HN
22
23
24
25
26
27
28
29
0.078
0.354
0.211
0.606
0.41
7.1
4.7
11.9
13.5
17
0.02
O
0.66
N
H
0.0055
0.0036
2.05
0.641
0.371
0.834
0.012
0.704
0.85
O
O
1.0
45
0.92
N
HN
O
1.22
26
0.0011
0.0084
N
O
0.66
24
N
N
30
31
0.177
0.238
0.95
2.7
11
31
0.0013
0.053
1.8
6.7
N
N
N
no-2,3,4,5-tetrahydropyridin-3-yl)benzenesulfonyl group is the
most potent analogue, but additionally efficiently inhibits factor
Xa with a Ki value of 5.7 nM. In contrast, the 4-ethoxy- and 4-eth-
ylphenyl derivatives (24, 25) are stronger thrombin than matrip-
tase inhibitors, whereas the more bulky tert-butyl analogue (26)
is obviously not able to occupy the hydrophobic aryl binding site
of thrombin, but still maintains some affinity to matriptase, which
possesses a more open S3/S4-binding region.

1964
A. Schweinitz et al. / Bioorg. Med. Chem. Lett. 19 (2009) 1960–1965
H
N
a, b, c
OH
O
O
HN
HN
x HCl
x HCl
32
O
O
H
H
N
N
O2N
S
O2
OH
H2N
S
N
O
d, e
O2
N
H
CN
CN
33
34
O
O
H
N
H2N
N
S
N
O
f, g, h
H
O2
N
H
NH2
HN
4
Scheme 1. Reagents and conditions: (a) 1.1 equiv (Boc)2O, NaOH (pH 9) in dioxane/water 2:1, 15 min 0 °C, 3 h at room temp, (b) 10 equiv CH3–NH2ÁHCl, 1.5 equiv HBTU,
4.5 equiv DIPEA in ethyl acetate, 5 h stirring at 40 °C, (c) 1 N HCl in acetic acid, 1 h room temp., (d) 1.1 equiv of 32, PyBop, 3.0 equiv DIEA in DMF, 15 min 0 °C and 3 h room
temp, (e) zinc powder in acetic acid, 2 h, room temp, (f) Cbz-b-Ala-OH, N-methyl-morpholine, isobutyl chloroformate, À15 °C, 10 min in DMF, followed by addition of 34, 1 h,
À15 °C, overnight at room temp, (g) (i) 2 equiv hydroxylamineÁHCl and DIPEA, reflux in ethanol, 4 h, stirring overnight at room temp and evaporation of the ethanol; (ii)
2.5 equiv Ac2O in acetic acid, 30 min room temp and evaporation; (iii) H2 and Pd/C as catalyst in 90% acetic acid, stirring over night at 30 °C; (h) purification by preparative
reversed phase HPLC.1
O
O
O
H
N
H
N
N
N
S
N
O
I
S
N
O
O2
O2
N
H
a
N
H
CN
CN
5
Scheme 2. Reagents and conditions: (a) (i) 5 equiv pyridazinone, 2 equiv K3PO4, 10 mol % (1R,2R)-N,N’-dimethyl-1,2-cyclohexanediamine, 4 mol % CuI in DMF under N2,
stirring 1.5 h at 110 °C, ii: flash chromatography on silica gel using a DCM/MeOH gradient.
A third series of analogues containing a C–N-bond between
both rings within the sulfonyl residue was prepared (27–31). The
piperazinone and piperidinone inhibitors still inhibit matriptase
with Ki values 630 nM, whereas the other compounds were less
efficient. Some of the more hydrophobic analogues possess a
remarkable activity as thrombin inhibitors (e.g., 28, 30), whereas
again in this case the basic piperazin-2-one derivative 27 preferen-
tially inhibits factor Xa.
urine. These results clearly indicate that both prodrugs are not or-
ally available. In contrast, after oral application of the reference
compound ximelagatran12 (20 mg/kg, n = 3), a double prodrug of
the thrombin inhibitor melagatran that also contains a hydroxy-
amidino prodrug moiety, significant levels could be detected for
the drug melagatran (Cmax = 1.0 lmol/L) as well as for the mono
prodrug N-hydroxy-melagatran (Cmax = 1.1 lmol/L) over a period
of 120 min.
Two hydroxyamidino prodrugs12 of analogues 24 and 26 were
prepared and orally applied to rats via a gastric feeding tube at a
dose of 10 mg/kg (n = 3). These prodrugs were selected due to their
high clogP values of 3.75 and 5.02, respectively. However, the HPLC
or LC/MS-analysis of the plasma samples at different times over a
period of 180 min never revealed any concentration of the pro-
drugs or their active metabolites above the detection limit
However, although we could not demonstrate oral bioavailability
for these types of matriptaseinhibitors, we usedseveral analogues of
this and our preceding series in tumor cell-based assays to elucidate
further the role of matriptase in invasion, pro-uPA , and pro-HGF
cleavage with subsequent c-Met activation and phosphorylation.
The results of these studies will be published elsewhere.14
All inhibitors summarized in Table 1 were synthesized using
standard procedures, an example of which is described for the
preparation of 4 (Scheme 1). Briefly, commercially available 4-
(piperidin-4-yl)butanoic acid was converted into its methylamide
(100 nmol/L or 0.065
200 nmol/L or 0.13 g/mL for both prodrugs). In addition, neither
inhibitor nor prodrug could be detected in kidney, liver, or in the
lg/mL for the parent compounds, and
l

A. Schweinitz et al. / Bioorg. Med. Chem. Lett. 19 (2009) 1960–1965
1965
2. List, K.; Haudenschild, C. C.; Szabo, R.; Chen, W.; Wahl, S. M.; Swaim, W.;
Engelhol, L. H.; Behrendt, N.; Bugge, T. H. Oncogene 2002, 21, 3765.
3. List, K.; Szabo, R.; Wertz, P. W.; Segre, J.; Haudenschild, C. C.; Kim, S. Y.; Bugge,
T. H. J. Cell Biol. 2003, 163, 901.
4. (a) Lin, C.-Y.; Anders, J.; Johnson, M.; Dickson, R. B. J. Biol. Chem. 1999, 274,
18237; (b) Szabo, R.; Molinolo, A.; List, K.; Bugge, T. H. Oncogene 2007, 26,
1546; (c) Oberst, M. D.; Johnson, M. D.; Dickson, R. B.; Lin, C.-Y.; Singh, B.;
Stewart, M.; Williams, A.; al-Nafussi, A.; Smyth, J. F.; Gabra, H.; Sellar, G. C. Clin.
Cancer Res. 2002, 8, 1101; (d) Szabo, R.; Hobson, J. P.; List, K.; Molinolo, A.; Lin,
C.-Y.; Bugge, T. H. J. Biol. Chem. 2008, 283, 29495.
32. This intermediate was coupled to the 3-cyanophenylalanine
derivative 33, followed by reduction of the nitro group and cou-
pling of Cbz-b-Ala-OH using the mixed anhydride method. The
conversion of the nitrile into the amidine 4 was performed accord-
ing to the method of Judkins.11,15 Appropriate commercially avail-
able secondary amines were used as starting material for the
synthesis of other compounds given in Table 1. All inhibitors with
free carboxyl group were obtained from their analogous methyl es-
ters by saponification in the final step.
The biphenyl-3-sulfonyl derivatives (21, 24–26) were obtained
by Suzuki coupling as previously described.13 Inhibitor 22 was ob-
tained by hydrogenation of purified 21 using Pd/C as catalyst in
90% acetic acid over 2 days. The coupling of the nitrogen heterocy-
cles to the aryl iodide for inhibitors 27 to 31 was performed by a
copper-catalyzed reaction (Scheme 2).16 The final conversion of
the nitrile into the amidine was performed as described above.
5. Uhland, K. Cell. Mol. Life Sci. 2006, 63, 2968.
6. Zeng, L.; Gao, J.; Zhang, X. World J. Gastroenterol. 2005, 11, 6202.
7. Lee, S.-L.; Dickson, R. E.; Lin, C.-Y. J. Biol. Chem. 2000, 275, 36720.
8. Takeuchi, T.; Harris, J. L.; Huang, W.; Yan, K. W.; Coughlin, S. R.; Craik, C. S. J.
Biol. Chem. 2000, 275, 26333.
9. Ahmed, S.; Jin, X.; Yagi, M.; Yasuda, C.; Sato, Y.; Higashi, S.; Lin, C. Y.; Dickson, R.
B.; Miyazaki, K. FEBS J. 2006, 273, 615.
10. Jin, X.; Yagi, M.; Akiyama, N.; Hirosaki, T.; Higashi, S.; Lin, C.-Y.; Dickson, R. B.;
Kitamura, H.; Miyazaki, K. Cancer Sci. 2006, 97, 1327.
11. Steinmetzer, T.; Schweinitz, A.; Stürzebecher, A.; Dönnecke, D.; Uhland, K.;
Schuster, O.; Friedrich, R.; Than, M. E.; Bode, W.; Stürzebecher, J. J. Med. Chem.
2006, 49, 4116.
12. (a) Clement, B.; Lopian, K. Drug Metab. Dispos. 2003, 31, 645; (b) Gustafsson, D.;
Bylund, R.; Antonsson, T.; Nilsson, I.; Nyström, J-E.; Eriksson, U.; Bredberg, U.;
Teger-Nilsson, A.-C. Nat. Rev. Drug Disc. 2004, 3, 649.
Acknowledgment
13. Steinmetzer, T.; Dönnecke, D.; Korsonewski, M.; Neuwirth, C.; Steinmetzer, P.;
Schulze, A.; Saupe, M. S.; Schweinitz, A. Bioorg. Med. Chem. Lett. 2009, 19, 67.
14. Uhland, K.; Siphos, B.; Arkona, C.; Schuster, M; Petri, B.; Steinmetzer, P.; Müller,
F.; Schweinitz, A.; Steinmetzer, T., in preparation.
The authors would like to thank Robert Etges for correcting the
English version of the manuscript.
15. Judkins, B. D.; Allen, G. D.; Cook, T. A.; Evans, B.; Sardharwala, T. S. Synth.
Commun. 1996, 26, 4351.
References and notes
16. Klapars, A.; Antilla, J. C.; Huang, X.; Buchwald, S. L. J. Am. Chem. Soc. 2001, 123,
7727.
1. (a) Szabo, R.; Bugge, T. H. Int. J. Biochem. Cell Biol. 2008, 40, 1297; (b) Bugge, T.
H.; List, K.; Szabo, R. Front. Biosci. 2007, 12, 5060.