Toward the first class of suicide inhibitors of kallikreins involved in skin diseases

Article abstract of DOI:10.1021/jm500988d

The inhibition of kallikreins 5 and 7, and possibly kallikrein 14 and matriptase, (that initiates the kallikrein proteolytic cascade) constitutes an innovative way to treat some skin diseases such as Netherton syndrome. We present here the inhibitory properties of coumarin-3-carboxylate derivatives against these enzymes. Our small collection of these versatile organic compounds was enriched by newly synthesized derivatives in order to obtain molecules selective against one, two, three enzymes or acting on the four ones. We evidenced a series of compounds with IC₅₀ values in the nanomolar range. A suicide mechanism was observed against kallikrein 7 whereas the inactivation was either definitive (suicide type) or transient for kallikreins 5 and 14, and matriptase. Most of these potent inhibitors were devoid of cytotoxicity toward healthy human keratinocytes. In situ zymography investigations on skin sections from human kallikrein 5 transgenic mouse revealed significant reduction of the global proteolytic activity by several compounds.

Full text of DOI:10.1021/jm500988d

Subscriber access provided by NORTHERN ARIZONA UNIV
Article
Towards the first class of suicide inhibitors
of kallikreins involved in skin diseases
Xiao Tan, Feryel Soualmia, Laetitia Furio, Jean-François Renard, Isabelle Kempen, Lixian Qin, Maurice
Pagano, Bernard Pirotte, Chahrazade El Amri, Alain Hovnanian, and Michèle REBOUD-RAVAUX
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/jm500988d • Publication Date (Web): 09 Dec 2014
Downloaded from http://pubs.acs.org on December 12, 2014
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Towards the first class of suicide inhibitors of kallikreins involved in
skin diseases
1
a,b
1a,b
2
3
3
Xiao Tan , Feryel Soualmia , Laetitia Furio , JeanꢀFrançois Renard , Isabelle Kempen ,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
a,b
1a,b,
ꢀ
3
1a,b,*
Lixian Qin , Maurice Pagano
, Bernard Pirotte , Chahrazade El Amri
, Alain
2
,*
1a,b,*
Hovnanian , Michèle ReboudꢀRavaux
1a
Sorbonne Universités, UPMC University Paris 06, UMR 8256, B2A, Biological
Adaptation and Ageing, Integrated Cellular Ageing and Inflammation, Molecular &
Functional Enzymology, Institut de Biologie Paris Seine, 7 Quai St Bernard, Fꢀ75005
Paris, France,
1b
CNRS, UMR 8256, B2A, Biological Adaptation and Ageing, Fꢀ75005 Paris,
2
University Paris Descartes ꢀ Sorbonne Paris Cité, INSERM UMR 1163, Laboratory of
Genetic Skin Diseases, Imagine Institute, 24 boulevard du Montparnasse, Fꢀ75015 Paris,
France,
3
University of Liège, Department of Medicinal Chemistry, Center for Interdisciplinary
Research on Medicines (CIRM), 1 Avenue de l’Hôpital, Bꢀ4000 Liège, Belgium.
ꢀ
Deceased
*
Corresponding authors
KEYWORDS: Netherton syndrome, skin diseases, kallikreins, inhibitors, coumarins, suicide
substrates, in situ zymography, drug repositionning
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 52
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ABSTRACT: The inhibition of kallikreins 5 and 7, and possibly kallikrein 14 and matriptase,
(that initiates the kallikrein proteolytic cascade) constitutes an innovative way to treat some
skin diseases such as Netherton syndrome. We present here the inhibitory properties of
coumarinꢀ3ꢀcarboxylate derivatives against these enzymes. Our small collection of these
versatile organic compounds was enriched by newly synthesized derivatives in order to obtain
molecules selective against one, two, three enzymes or acting on the four ones. We evidenced
a series of compounds with IC50 values in the nanomolar range. A suicide mechanism was
observed against kallikrein 7 whereas the inactivation was either definitive (suicide type) or
transient for kallikreins 5 and 14, and matriptase. Most of these potent inhibitors were devoid
of cytotoxicity towards healthy human keratinocytes. In situ zymography investigations on
skin sections from human kallikrein 5 transgenic mouse revealed significant reduction of the
global proteolytic activity by several compounds.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
.
2
ACS Paragon Plus Environment

Page 3 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
INTRODUCTION
Kallikreinꢀrelated peptidases (KLKs) constitute a family of fifteen mammalian serine
1
proteases. A variety of endogenous inhibitors that control their activities have been
2
, 3
identified. Failure in their inhibitory capacities is implicated in several pathologies. One of
the most relevant examples is Netherton syndrome (NS), a rare autosomal recessive skin
disease due to the absence of the lymphoꢀepithelial Kazal type related inhibitor (LEKTIꢀ1)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4
, 5
expressed in stratified epithelia. Lossꢀofꢀfunction mutations of SPINK5 encoding LEKTIꢀ1
result in unopposed KLK5 activity. KLK5 is a key player of the epidermal proteolytic cascade
that also implicates KLK7 and KLK14. This unopposed proteolytic activity has profound
effects on skin barrier integrity and permeability and also induces skin inflammation and
3
allergy compromising the vital forecast of the newborn. In fact, epidermal proteases show a
high degree of cooperation leading to the activation of the kallikrein activation cascade,
KLK5 activating proꢀKLK7 and proꢀKLK14, proꢀKLK5 being activated by KLK14 and itself,
2
, 6
and matriptase being required for the activation of KLK5 and KLK7.
Recently,
mesotrypsin has also been shown to activate proꢀKLK5 and proꢀKLK7 and to degrade LEKTI
7
inhibitory fragments. Proteaseꢀinhibitor imbalances can also be involved in some forms of
8, 9
psoriasis and in common eczema.
Specific inhibitors of the implicated proteases having
appropriate pharmacokinetic and pharmacodynamic properties may potentially be of great
interest to develop new treatments of these conditions.
Currently, there are only symptomatic treatments for NS which comprise the use of
emollients, topical steroids and calcineurin inhibitors to prevent severe dehydration and to
3, 10
reduce skin inflammation and allergy.
In severe cases, intravenous immunoglobulins and
1
1, 12
antiꢀTNF alpha treatment were shown to bring clinical benefit.
The inhibition of the
KLKs by synthetic inhibitors may constitute an innovative way to treat NS by potentially
restoring the functions of the epidermis barrier and preventing deleterious immunological
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 52
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
responses since these enzymes play a critical role in corneodesmosome degradation and skin
inflammation. Relatively few inhibitors of KLKs and matriptase have been described,
especially those acting through the formation of covalent complexes with the targeted
1
3
enzymes. Peptidic inhibitors have been derived from LEKTI and sunflower trypsin inhibitor
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
4
15
(
SFTIꢀ1) as KLKs inhibitors; others were obtained by drug design approach. In addition to
1
6
17, 18
isocoumarins from Brazilian plants and isomannide derivatives
, acylꢀ1,2,4ꢀtriazoles
were described as transient inactivators of human KLK5, KLK7, KLK14 and also
19
matriptase. In silico screening of a collection of organic compounds led to the identification
20
of reversible and structurally diverse inhibitors of these enzymes. Several inhibitors of
21ꢀ24
matriptase have been described based for example on macromolecular inhibitors
or on a
2
5
structureꢀbased approach.
Our former studies on coumarinꢀ3ꢀcarboxylic acid derivatives as potent inhibitors of
2
6ꢀ28
some serine proteases
prompted us to obtain new families of nonꢀpeptidic inhibitors
susceptible to selectively target KLKs. This study was also extended to matriptase, an
6
activator of proꢀKLK5 that is considered as an initiator of the KLK cascades in the skin.
Coumarin derivatives display diverse pharmacological activities supporting their
29
30
31
32ꢀ34
35, 36
potential therapeutic use as antitumor , antibacterial , antiꢀHIV
, antiꢀinflammatory
3
7
and antioxidant agents.
6ꢀSubstituted coumarinꢀ3ꢀcarboxylate derivatives constitute
remarkable nonpeptidic protease inhibitors characterized by an easy access to organic
synthesis, a high inhibitory potency and an extreme sensitivity to structural variations at the 3ꢀ
and 6ꢀpositions, that not only influence the specificity but also the inhibition mechanism of
26, 28
the targeted enzyme.
The general structure of our collection of previously synthesized as
well as newly synthesized coumarin derivatives is presented in Figure 1. Most compounds are
aromatic esters, thioesters or amides of 6ꢀsubstituted coumarinꢀ3ꢀcarboxylic acids, thus
having an electronꢀwithdrawing group at the 3ꢀposition, which increases the electrophilicity
4
ACS Paragon Plus Environment

Page 5 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
of the conjugated lactonic carbonyl group, facilitating its nucleophilic attack by the hydroxy
38
group of the protease active serine (Ser195) (Figure 2). This attack leads to the formation of
an acylꢀenzyme. If the substituent at the 6ꢀposition is a methylene radical bearing a good
leaving group (i.e. ꢀCH
2
W), a methylene quinone intermediate may be formed displaying a
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
very high sensitivity to nucleophilic addition. The alkylation of a nucleophilic activeꢀsite
28
residue (histidine 57 in the case of αꢀchymotrypsin) leads to definitive inactivation of the
enzyme (suicide mechanism; pathway a, Figure 2). A selective and very efficient inactivation
of chymotrypsin by 6ꢀchloromethylated coumarinꢀ3ꢀcarboxylate aryl esters via this
2
7
mechanism was observed with no or poor inhibition of human leukocyte elastase and
thrombin but a major difference between chymotrypsin and HLE: the inactivation of HLE
was transient (stable acylꢀenzyme; pathway b, Figure 2). Moreover, this transient inhibition of
2
HLE was demonstrated to be specific when the latent alkylating function CH ꢀW was
26
absent.
Thus, coumarinꢀ3ꢀcarboxylic acid aryl esters appear as timeꢀdependent versatile
general inhibitors of serine proteases able to discriminate between different classes of serine
proteases, depending on the nature of the substituents, through two major distinct
mechanisms: suicide substrates (chymotrypsin for example) or alternate substrates inhibitors
3
9ꢀ41
(HLE for example). Owing to the renewal of interest for covalent drugs
, these coumarin
derivatives appeared as good candidates to obtain for the first time suicide substrates of KLKs
and matriptase. We present here the screening of our collection of coumarinꢀ3ꢀcarboxylic acid
derivatives against human KLK5, KLK7, KLK14 and matriptase and their chemical
optimization. The mechanism of enzyme inhibition and the toxicity of a selection of
compounds with respect to healthy keratinocytes were analyzed. In addition in situ
zymography experiments were performed on skinꢀspecific human KLK5 transgenic mouse
42
model as this model replicates the major cutaneous and systemic hallmarks of NS.
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 52
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
RESULTS AND DISCUSSION
Synthesis. The synthesis of several tested compounds was previously described.
2
6ꢀ28
New 2ꢀoxoꢀ2Hꢀ1ꢀbenzopyranꢀ3ꢀcarboxylic acid derivatives bearing a fluoromethyl or a
bromomethyl group at the 6ꢀposition were obtained according to the chemical pathway
described in Scheme 1. The reaction of the appropriate di(halo)phenyl ester of malonic acid 2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
27, 43
with 5ꢀhydroxymethylsalicylaldehyde
in the conditions of the Knoevenagel reaction
provided the (halo)phenyl 6ꢀ(hydroxymethyl)ꢀ2ꢀoxoꢀ2Hꢀ1ꢀbenzopyranꢀ3ꢀcarboxylates 3. The
latter 6ꢀ(hydroxymethyl)ꢀsubstituted intermediates were converted into the final 6ꢀ
(fluoromethyl)ꢀ (4a) or 6ꢀ(bromomethyl)ꢀsubstituted (halo)phenyl 2ꢀoxoꢀ2Hꢀ1ꢀbenzopyranꢀ3ꢀ
carboxylates (4bꢀe) after reaction with the Ishikawa’s reagent (56) (fluorination) or with
thionyl bromide (bromination). The naphthyl esters of 2ꢀoxoꢀ2Hꢀ1ꢀbenzopyranꢀ3ꢀcarboxylic
acid 6 and 8 were obtained after reaction of the appropriate dinaphthyl malonate (5 and 7,
respectively) with salicylaldehyde in the conditions of the Knoevenagel reaction (Scheme 2).
The isoquinolꢀ1ꢀyl ester of 2ꢀoxoꢀ2Hꢀ1ꢀbenzopyranꢀ3ꢀcarboxylic acid (56) resulted from the
reaction of 1ꢀhydroxyisoquinoline with the acid chloride of the commercially available 2ꢀoxoꢀ
2
Hꢀ1ꢀbenzopyranꢀ3ꢀcarboxylic acid. Other unpublished monosubstituted phenyl esters of 6ꢀ
chloromethyl)ꢀ2ꢀoxoꢀ2Hꢀ1ꢀbenzopyranꢀ3ꢀcarboxylic acid (compounds 18, 19, 28, 29, 30, 33)
were synthesized from the reaction of 6ꢀ(chloromethyl)ꢀ2ꢀoxoꢀ2Hꢀ1ꢀbenzopyranꢀ3ꢀcarbonyl
(
27
chloride with the appropriate phenol according to a previously described process.
Enzyme inhibition, structure-activity relationships and selectivity. Before testing
the coumarinꢀ3ꢀcarboxylic acid derivatives against the targeted enzymes, their stability in
aqueous medium was evaluated in the absence of enzyme at pH 7.5 or 8.0 by following their
UV spectrum evolution. Some compounds were stable showing no spectral evolution within 4
h such as compounds 15, 17, 23, 31, 56 and 4d. According to the chemical structure, different
halfꢀlife times were observed: for example, > 100 min (11, 24, 25), ranging from 40 to 50 min
6
ACS Paragon Plus Environment

Page 7 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(21, 45, 46) or from 25 to 30 min (53ꢀ55); some nitroꢀsubstituted compounds being less stable
1
/2
(t
= 10 min for 51). The in vitro inhibitory potency of the collection of coumarinꢀ3ꢀ
carboxylic acid derivatives towards human KLK5, KLK7, KLK14 and matriptase was
examined by measuring the hydrolysis of the appropriate fluorogenic substrate at 37 °C and
pH 8. When an inhibition was observed, the IC50 values (inhibitor concentrations giving 50 %
inhibition) were determined after 15 min incubation of the tested enzyme with various
inhibitor concentrations (Figure 2S). The results are summarized in Tables 1ꢀ5. Several
compounds belonging to different chemical series (46 out of 260) were found to be active
against at least one of the tested enzymes.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Active and potent compounds against KLKs and matriptase were found amongst aryl
esters (Z = O) and aryl thioesters (Z = S) of 6ꢀ(halomethyl)ꢀsubstituted 2ꢀoxoꢀ2Hꢀ1ꢀ
benzopyranꢀ3ꢀcarboxylic acids (i.e. 11, 4d and 14), whereas the corresponding amides such as
1
3 were less efficient inhibitors (Table 1). Looking at the nature of the substituent at the 6ꢀ
position of the coumarin ring, it was evidenced that the presence of a latent alkylating
function with a good leaving group (X = CH Cl or CH Br) was required to observe noticeable
2
2
inhibitions of the four tested enzymes (4d), or only (11) and more specifically (14) KLK7.
Comparatively, the fluoromethylꢀsubstituted compound (4a) almost devoid of an alkylating
character was found to be inefficient, while the presence of the poor acetate leaving group
induced modest inhibitory activity on KLKs (12 compared to 11 and 4d). In the absence of
3
leaving group (X = H or CH ), the inhibitory potency of esters was lost. As a result, the rank
order of potency on KLK7 was found to be directly correlated to the potential alkylating
character of the substituent at the 6ꢀposition (ꢀCH Br > ꢀCH Cl > ꢀCH OCOCH > ꢀCH F).
2
2
2
3
2
2
8
28
On the other hand, cycloalkyl (cyclododecyl; X = CH
2
Cl) ester and amide derived
from 6ꢀ(chloromethyl)ꢀ2ꢀoxoꢀ2Hꢀ1ꢀbenzopyranꢀ3ꢀcarboxylic acid were found inactive on the
27
27
selected KLKs and matriptase. It was also the case for the corresponding methyl and allyl
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 52
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
amides. Methyl27 and ethyl27 esters (X = CH
Cl) were poorly active: IC50 = 54 ± 2 µM
2
against KLK5 and 27 ± 2 µM against KLK7 for the methyl compound; IC50 = 21 ± 1 µM
against KLK7 for the ethyl compound. No inhibition was observed against KLK14 and
matriptase for both compounds, and against KLK5 for the allyl one. We previously found that
the most potent inhibitors of αꢀchymotrypsin and human leucocyte elastase possessed a
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
26, 27
phenyl or a pyridyl ester substituent at the 3ꢀposition, whereas alkyl esters were inactive.
The same structural feature seems to hold against the tested kallikreins and matriptase.
Consequently, we focused our attention on series of phenyl (Table 2), pyridꢀ2ꢀyl and pyridꢀ3ꢀ
yl (Table 3) esters bearing a chloromethyl or a bromomethyl group at the 6ꢀposition, and
bicyclic naphtyl or isoquinolyl esters devoid of substituent at this position (Table 4).
Noticeable differences in the inhibitory potency against the various targeted enzymes were
obtained according to the nature and position of substituents on the aromatic ring.
We observed that the chymoptryptic serine protease KLK7 was more efficiently
inhibited by the 6ꢀchloromethyl phenyl esters than the tryptic ones (KLK5, KLK14 and
matriptase) (Table 2) with IC values ranging from 63 nM to 8.4 µM. A single halogen atom
50
located in the ortho position (16 for Cl, 17 for I) was less favorable to inhibit KLK7 than in
the meta (11 for Cl, 23 for I) or para (31 for Cl) positions. The strongest KLK7 inhibition was
observed with Br (IC = 65 ± 3 nM for 22) and I (IC = 77 ± 3 nM for 23) in the meta
50
50
position, the corresponding chloro and fluoro compounds being less efficient by a factor of
1.4 and 3.6, respectively (rank order of potency: Br ≥ I > Cl > F). For the bulkier
methoxycarbonyl substituent (ester function), the ortho position (19) was more favorable than
the para one (33). The monosubstitution by other kinds of substituents (CH , CF
COOH, NO and N(CH groups) was usually less favorable.
≈
3
3
, CONH2,
2
3 2
)
The presence of a second substituent on the phenyl ring of the chloromethyl
derivatives did not significantly improve the inhibitory potency towards KLK7 (34, 43).
8
ACS Paragon Plus Environment

Page 9 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
However, this activity was increased with the 2,3ꢀdichloroꢀsubstituted derivative 34 (R = 2’ꢀ
1
Cl; R
ꢀchloroꢀsubstituted analogues (IC50 values were 209 ± 7 nM and 103 ± 5 nM for the
monosubstituted 16 (R = 2’ꢀCl) and 11 (R = 3’ꢀCl), respectively). As a result, chlorine
2
= 3’ꢀCl) with IC50 = 63 ± 4 nM compared to the activity of the respective 2ꢀchloroꢀ and
3
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
1
atoms at the 2,3ꢀpositions (ortho and meta positions) appeared to be the best substituent
combination in order to induce a marked inhibitory activity on KLK7. Interestingly, the
presence of chlorine atoms at the 2,5ꢀpositions (also ortho and meta positions) (42) was
responsible for a noticeable inhibitory activity on KLK5 (1.3 ± 0.1 µM) and KLK7 (0.19 ±
0.02 µM) with a moderate activity on KLK14 (54 ± 6 µM). A second substituent different
from a halogen atom was tolerated, especially when the first one was a halogen atom.
KLK5 was inhibited by a few 6ꢀhalomethylꢀsubstituted compounds with IC50 values
ranging from 0.92 µM (4d, X = CH
2 1 2 1
Br; R = 3’ꢀCl) to about 80 µM (42, X = CH Cl; R = 3’ꢀ
Cl, R = 5’ꢀCl). Some of these compounds were also found to be active on KLK14, but
2
usually with a lower efficiency, the most potent one on this enzyme being compound 4d with
an IC50 value of 2.9 ± 0.1 µM.
Only three compounds were able to inhibit the four tested enzymes in the micromolar
range: the 3’ꢀCONH
2
ꢀsubstituted 6ꢀchloromethylcoumarinꢀ3ꢀcarboxylate phenyl ester 30
(IC50 values: 21.6 µM on KLK5; 7.8 µM on KLK7; 30 µM on KLK14; 10.5 µM on
matriptase) and, with a higher efficiency, the 3’ꢀchloroꢀ and 3’ꢀbromoꢀsubstituted 6ꢀ
bromomethylcoumarinꢀ3ꢀcarboxylates phenyl esters 4d and 4e (IC50 values for 4d: 0.92 µM
on KLK5; 64 nM on KLK7; 2.9 µM on KLK14; 1.42 µM on matriptase). Matriptase was the
less sensitive enzyme to the inhibitory effect by coumarin derivatives.
The pyridꢀ3ꢀyl (44ꢀ50) and pyridꢀ2ꢀyl (51ꢀ55) esters of 6ꢀchloromethylcoumarinꢀ3ꢀ
carboxylic acid were generally inefficient or less efficient than the phenyl esters against
KLK5, KLK14 and matriptase (Table 3). The unsubstituted or halomethylꢀmonosubstituted
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 52
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
pyridꢀ3ꢀyl cycle was the best choice to inhibit KLK7 (IC50 = 0.46ꢀ12.6 µM). Only the 6ꢀ
chloroꢀsubstituted pyridꢀ2ꢀyl ester 54 was found to express an inhibitory activity on KLK5,
KLK7, KLK14 and matriptase.
The influence of a bulky aromatic group was analyzed by introducing isoquinolyl and
naphthyl ester groups at the 3ꢀposition of coumarinꢀ3ꢀcarboxylic acid (Table 4). When a
bicyclic isoquinolꢀ1ꢀyl group was introduced (56), KLK5 and KLK14 were inhibited with a
preference for KLK5 (IC50 = 7.0 ± 0.4 µM and 13 ± 1 µM, respectively). On the opposite, the
isoquinolꢀ5ꢀyl (57) or the isoquinolꢀ3ꢀyl (58) esters were totally inactive. The naphthꢀ1ꢀyl (6)
and naphthꢀ2ꢀyl (8) esters orientated against KLK7 but with a very poor efficiency.
Mechanism of inhibition. Detailed mechanistic studies were performed for the best
inhibitors acting on the various targeted proteases. All of the 6ꢀhalomethylꢀsubstituted
coumarinꢀ3ꢀcarboxylic acid phenyl esters acting on KLK7 in the nanomolar range behaved as
timeꢀdependent inhibitors indicating that the inhibitory effects are slowly elicited on the time
scale of the enzyme turnover. No spontaneous reactivations of the inhibited enzymes (> 98%)
occurred within 12 h as observed for 11, 23, 24, 34, 35, 42, 4d, 4e (against KLK7), 36, 4d, 4e
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(against KLK5), and 4d and 4e (against KLK14 and matriptase). Hydroxylamine did not
accelerate reactivations of the treated enzymes confirming the irreversible character of the
inhibitions by these inhibitors and their mechanismꢀbased behavior as previously observed for
26, 28
44
several of related derivatives against α−chymotrypsin
and thrombin. If the inactivation
was induced by the formation of a stable acylꢀenzyme, such stable acylꢀenzyme would have
been deacylated by hydroxylamine leading to enzyme activity recovery. Due to the leaving
group properties of the 6ꢀchlorine and 6ꢀbromine atoms, an electrophilic quinone methide
may be formed after the lactone ring opening induced by Ser195 nucleophilic attack as shown
in Figure 2. The progress curve method in which the inhibitor competes with the fluorogenic
10
ACS Paragon Plus Environment

Page 11 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
4
5
substrate was used to evaluate the k /K ratios , which are indexes of the inhibitory potency
i
I
for this category of inhibitors (Figure 3; Table 5).
The values obtained for k /K varied from 97 M s for the weaker inhibitor 35 to
443 M .s for the most potent inactivator 42. The presence of the CH Cl group at the 6ꢀ
position poorly influenced inactivation efficacy as shown by the results obtained with CH Br
ꢀ1 ꢀ1
i
I
ꢀ1 ꢀ1
5
2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
group located at this position (17, 11 and 22 vs 4c, 4d and 4e, respectively). For the
monosubstituted phenyl compounds (except for R =3’ꢀCOOCH ), there was little effect on
1
3
the inactivation efficiency of the substituent nature and position. The situation was more
contrasted for disubstituted phenyl molecules with the following decreasing efficiency order:
3
3’ꢀCl, 5’ꢀCl > 2’ꢀCl, 5’ꢀCl > 2’ꢀCl, 3’ꢀCl > 2’ꢀCl, 6’ꢀCl > 3’ꢀCl, 5’ꢀOCH > 2’ꢀBr, 5’ꢀF
For example, the nature of a second substituent may be critical: a 8ꢀfold decrease of activity
was observed when the 5’ꢀchlorine atom of 42 (3’ꢀCl, 5’ꢀCl) was replaced by a methoxy
3
group in 43 (3’ꢀCl, 5’ꢀOCH ).
Several identified covalent inhibitors of KLK7 were also able to inhibit irreversibly
KLK5, KLK14 and matriptase. However, some transient inactivations were observed for the
tryptic KLK5, KLK14 and matriptase (18/KLK14; 19/matriptase; 28/ KLK14, matriptase;
3
0/KLK5, matriptase; 40/KLK5). Monosubstitutions at position 2’ or 3’ by a carboxyl or ester
or amide group seemed to favor the transient inactivation of these tryptic enzymes.
Besides covalent inhibitors, we also have identified a reversible inhibitor (56) of
KLK5 and KLK14 belonging to a new subfamily. This inhibitor was characterized by the
absence of a substituent at the 6ꢀposition and the presence of a bicyclic substituent
(
isoquinoline) on the Yꢀposition (see Table 4). The mechanism of inhibition of compound 56
on KLK5 was shown to be competitive with an inhibition constant K
i
= 5.4 ± 0.6 µM (Figure
2S).
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 52
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Models inhibitor/protease complexes have been built using the CLC Drug Discovery
Workbench software to corroborate the mechanistic data. We compared the possible binding
modes of two inhibitors in the active site of the targeted proteases (Figure 4): the suicide
inhibitor 4d that was active on all proteases and the noncovalent inhibitor 56 that acted
competitively against KLK5. It was found that these molecules can adopt two preferential
conformations in agreement with optimal electronic delocalization within the plane of the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
28
intracyclic lactonic and exocyclic ester groups. All docking calculations were performed
using both conformations as starting points and revealed no differences on the resulting
binding modes. The analysis of compound 4d positioning within the active sites of KLK5,
KLK7 and matriptase showed that in all cases the aryl ester moiety at the 3ꢀposition was
accommodated within the S subsite (Figure 4AꢀC). This may explain why compound 4d
1
displayed higher efficiency for chymotryptic KLK7 than for the tryptic enzymes for which the
negatively charged Asp189 is located at the bottom of their S1 subsite instead of Asn189 for
KLK7. The distance between the oxygen atom of the OH group of Ser195 and the carbon
atom of the carbonyl group of the lactone function was compatible with a nucleophilic attack
by the active serine. The substituent at the 6ꢀposition (CH Br) pointed towards His57
2
indicating a possible formation of a covalent bond between this residue and the methylene
group formed after the bromine departure in agreement with the suicide mechanism of
inhibition identified in vitro. In the case of compound 56 bound within the active site of
KLK5, a reverse orientation of the coumarin ring compared to that of compound 4d was
observed in about ten poses (Figure 4D). The nitrogen atom of the isoquinoline ring was
found sufficiently close to the Ser195 OH group to establish a hydrogen bond. The 56ꢀKLK5
complex might potentially be stabilized by three additional putative hydrogen bonds between
the coumarin ring and Ser195 and Gln192 residues. This latter observation was in agreement
with the competitive mechanism evidenced by kinetic data. The original binding mode of the
12
ACS Paragon Plus Environment

Page 13 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
bicyclic isoquinolꢀ1ꢀyl group may constitute a valuable starting point for the elaboration of
noncovalent inhibitors with improved affinity for KLK5.
Biological evaluation
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Cytotoxicity on keratinocytes. In order to validate the effect of our coumarin inhibitors in a
biological context, we firstly tested their toxicity on normal human keratinocytes isolated
from healthy donors. In this context, compounds were selected according to their inhibitory
capacity and structural diversity and tested at two different concentrations, 1 and 10 µM
(Figure 5). In all cases, coumarinic derivatives were found to be nonꢀcytotoxic at the tested
concentrations that were 10ꢀ100 fold higher than the corresponding in vitro IC50 values. The
usual lack of toxicity of our compounds on human keratinocytes is a favorable characteristic
for a potential therapeutical use.
In situ zymography analyses. The transgenic mouse model in which human KLK5 is
overexpressed in the granular layer of the epidermis (TgꢀhKLK5) is a relevant model to study
42, 46
cutaneous and systemic phenotypic aspects of NS.
Indeed human KLK5 overexpression
in the epidermis leads to enhanced protease activity of KLK7 and KLK14, and downstream
targets of the proteolytic cascade, thus increasing the total protease activity in the skin. In our
in situ zymography experiment, the skin sections from TgꢀhKLK5 mice were treated with
selected coumarinic inhibitors prior to evaluation of total protease activity using a fluorogenic
protein substrate (casein coupled to FITC). The results obtained with compounds 4d, 11, 26,
34, 36 and 56 are shown in Figure 6. The presence of 4d (the most potent inhibitor of KLK5,
7, 14 and matriptase), 36 (preferential inhibition of KLK5 and KLK7) and 56 (selective
inhibitor of KLK5 and KLK14) induced a drastic reduction of the total protease activity of the
skin compared to the control. This reduction of the protease activity was particularly marked
in the epidermis. In this transgenic model, molecules inhibiting only KLK7 (11 and 34) or
none of the targeted enzymes (26), thus devoid of any inhibitory potency against KLK5, had
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 52
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
no effect. However 4d also appeared to exhibit its inhibitory effect in the dermis, probably
because of its broad spectrum of action on targeted proteases. These results indicate that the
inhibitory activities of coumarinꢀ3ꢀcarboxylates aryl esters on several KLKs are still effective
in a tissue context replicating phenotypic characteristics of NS.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Conclusion
In conclusion, we have identified several subfamilies of coumarin derivatives that
differentially inhibit key proteases whose activity is deregulated in NS. They act mainly as
suicide substrates. Such inhibitors are likely to be extremely selective in vitro and in vivo
since their inhibitory activity requires discrimination in the binding steps, the demasking of
the latent electrophilic function during one step of the catalytic cycle of the target enzyme,
and the irreversible modification of the active site facilitated by the formation of the acylꢀ
38
enzyme maintaining part of the inhibitor within the active site. Their inhibitory effects on
KLKs and matriptase have been validated by both in vitro enzymatic studies and in situ
zymography using TgꢀhKLK5 mouse as a model of NS. Molecular modeling was used to
better understand their binding modes to the catalytic site of different proteases. We found
that appropriate structural variations may lead to preferential inhibition of KLK7 or
orientation towards the trypsinꢀlike KLK5 and KLK14. These findings are expected to
facilitate a further rational optimization of these compounds.
EXPERIMENTAL SECTION
27
26
Compounds 11, 15ꢀ17, 23ꢀ25, 31, 32 and 42 , 45ꢀ56 , 9, 10, 12, 13, 14, 21ꢀ23, 27,
2
8
47
2
8, 34, 36, 40, 41 and 43 , 35, 37, 38 and 39 were previously reported by our laboratory.
The synthesis of new compounds 4a-e, 6, 8, 18-20, 29, 30, 33, and 56 is described hereafter.
The purity of the compounds established by elemental analysis was equal to or greater than
14
ACS Paragon Plus Environment

Page 15 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
95%. Stock solutions (10 mM) were prepared by dissolving the corresponding pure solid
compounds in DMSO and stored at ꢀ20 °C. The kallikreins (KLK5, KLK7, KLK14 and
matriptase) were purchased from R&D Systems Inc (Minneapolis, USA). The fluorogenic
substrates used in enzymatic assays were purchased from Bachem (France) and
hydroxylamine from Serva (Germany). Enzymatic studies were performed using a BMG
Fluostar microplate reader (black 96ꢀwell microplates, Costar 3915, Corning International,
USA).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Chemical synthesis
Reagents and solvents were purchased from usual commercial suppliers (SigmaꢀAldrich,
Beigium; Acros Organics, Belgium) and were used without further purification. All reactions
were routinely checked by thinꢀlayer chromatography (TLC) on silica gel 60 F254 (Merck) and
visualization was performed by UV light (254 nm). Melting points were determined on a
Stuart SMP3 apparatus in open capillary tubes and were uncorrected. IR spectra were
1
13
recorded as KBr pellets on a PerkinꢀElmer Spectrum GX FTIR spectrometer. The H and C
1
13
NMR spectra were recorded on a Bruker Avance ( H: 500 MHz; C: 125 MHz) instrument
using DMSOꢀd as solvent and tetramethylsilane (TMS) as internal standard; chemical shifts
6
were reported in δ values (ppm) relative to internal TMS. The abbreviation s = singlet, d =
doublet, t = triplet, q = quadruplet, m = multiplet and b = broad signal were used throughout.
Exact mass were obtained with a Bruker APEXꢀQe 9.4 Tesla FTꢀICR mass spectrometer
(Bruker Daltonics, Billerica, MA) in positive ion mode. Elemental analyses (C, H, N, S) were
determined on a Thermo Flash EA 1112 series elemental analyzer and were within ± 0.4% of
the theoretical values. This analytical process ensures for each target compound a purity equal
or greater than 95 %).
General synthetic pathway to di-3-halophenyl esters of malonic acid (2). The
mixture of malonic acid (2 g, 19.2 mmol), the appropriate 3ꢀhalophenol (2 eq., 38.4 mmol)
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 52
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
and phosphoryl chloride (2 eq., 38.4 mmol) was heated at 100°C for 90 min. The reaction
mixture was then carefully poured on a cooled 1% w/v aqueous solution of sodium hydroxide
(50 mL). The resulting precipitate of the title compound was collected by filtration, washed
with water, dried and recrystallized in methanol (yields: 65ꢀ80%).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The following malonate esters were obtained according to this procedure and were found to
4
8
be identical those reported in the literature (diphenyl malonate ), diꢀ3ꢀfluorophenyl
49
50
malonate , diꢀ3ꢀchlorophenyl malonate ).
Di-3-bromophenyl malonate (2d); yields: 74%; m.p.: 64ꢀ66 °C; IR (KBr) 3070,
ꢀ1 1
1
770, 1583, 1468, 1364, 1190, 1141, 1120 cm ; H NMR (DMSOꢀd ) δ (ppm): 4.12 (s, 2H,
6
CH
2
), 7.24 (dd, J = 8.1 Hz/1.6 Hz, 2H, 6ꢀH, 6’ꢀH), 7.44 (t, J = 8.1 Hz, 2H, 5ꢀH, 5’ꢀH), 7.50 (t,
13
J = 1.9 Hz, 2H, 2ꢀH, 2’ꢀH), 7.53 (d, J = 8 Hz, 2H, 4ꢀH, 4’ꢀH). C NMR (DMSOꢀd
6
) δ (ppm):
40.8; 121.0; 121.5; 124.8; 129.3; 131.5; 150.8; 164.9. Anal. (C15
H10Br O ) theoretical (%): C:
2 4
43.51; H: 2.43. Found (%): C: 43.70; H: 2.00.
General synthetic pathway to (3- halo)phenyl 6-(hydroxymethyl)-2-oxo-2H-1-
43
benzopyran-3-carboxylates (3). The solution of 5ꢀhydroxymethylsalicylaldehyde (0.4 g,
2.62 mmol) and the appropriate diꢀ(3ꢀhalo)phenyl malonate (2) (1.5 eq., 3.94 mmol) in
dioxane (12 mL) was supplemented with 8 drops of piperidine and 4 drops of glacial acetic
acid and stirred for 30 min at room temperature. At the end of the reaction, the solvent was
removed by distillation under reduced pressure and the residue was triturated with cold
methanol (10ꢀ20 mL). The resulting precipitate was collected by filtration, washed with cold
methanol, dried and recrystallized in chloroform (yields: 25ꢀ75%).
According to this method, the following compounds were obtained:
Phenyl 6-(hydroxymethyl)-2-oxo-2H-1-benzopyran-3-carboxylate (3a): yields:
25%; m.p. : 187ꢀ188 °C; IR (KBr) 3264, 3067, 1760, 1621, 1574, 1493, 1379, 1247, 1215,
ꢀ1 1
1
200 cm ; H NMR (DMSOꢀd
6
2
) δ (ppm): 4.59 (d, J = 5.1 Hz, 2H, CH ); 5.42 (t, J = 5.4 Hz,
16
ACS Paragon Plus Environment

Page 17 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1H, OH); 7.28 (d, J = 7.8 Hz, 2H, 2’ꢀH, 6’ꢀH); 7.33 (t, J = 7.4 Hz, 1H, 4’ꢀH); 7.45ꢀ7.51 (m,
13
3
H, 8ꢀH, 3’ꢀH, 5’ꢀH); 7.73 (d, J = 8.5 Hz, 1H, 7ꢀH); 7.90 (s, 1H, 5ꢀH); 9.07 (s, 1H, 4ꢀH). C
NMR (DMSOꢀd
6
) δ (ppm): 61.9; 116.0; 116.5; 117.4; 121.8; 126.2; 127.7; 129.7; 133.4;
+
1
39.5; 150.3; 150.5; 153.7; 156.0; 161.1. HRMS for C17
297.0754. Anal. (C17
.09.
H
12
O
5
+ H calcd, 297.0757; found,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
12 5
H O ) theoretical (%): C: 68.92 ; H: 4.08. Found (%): C: 68.57 ; H:
4
3-Fluorophenyl 6-(hydroxymethyl)-2-oxo-2H-1-benzopyran-3-carboxylate (3b):
yields: 30%; m.p. : 173ꢀ178°C; IR (KBr) 3253, 3057, 1763, 1621, 1606, 1574, 1486, 1377,
ꢀ
1 1
1242, 1214, 1132 cm ; H NMR (DMSOꢀd
J = 5.6 Hz, 1H, OH); 7.17 (dd, J = 8.1 Hz/1.7 Hz, 1H, 6'ꢀH), 7.21 (td, J = 8.6 Hz/2.3 Hz, 1H,
’ꢀH); 7,27 (dt, J = 9.8 Hz/2.2 Hz, 1H, 2’ꢀH); 7.46 (d, J = 8.6 Hz, 1H, 8ꢀH); 7.56 (dd, J = 15.1
Hz/8.2 Hz, 1H, 5’ꢀH);7.74 (dd, J = 8.6 Hz/1.8 Hz, 1H, 7ꢀH); 7.91 (s, 1H, 5ꢀH); 9.09 (s, 1H, 4ꢀ
6
2
) δ (ppm): 4.59 (d, J = 5.4 Hz, 2H, CH ); 5.43 (t,
4
13
H). C NMR (DMSOꢀd ) δ (ppm): 61.8; 109.9; 113.2; 116.0; 116.1; 117.4; 118.2; 127.8;
6
5
130.9; 133.6; 139.5; 150.9; 151.2; 153.8; 155.9; 160.7; 161.2ꢀ163.2. HRMS for C17H11FO +
+
H calcd, 315.0663; found, 315.0660. Anal. (C H FO ) theoretical (%): C: 64.97 ; H: 3.53.
17
11
5
Found (%): C: 64.53 ; H: 3.51.
3-Chlorophenyl 6-(hydroxymethyl)-2-oxo-2H-1-benzopyran-3-carboxylate (3c):
yields: 50%; m.p. : 162ꢀ165 °C ; IR (KBr) 3526, 3058, 1732, 1621, 1574, 1471, 1371, 1239,
ꢀ1
1
1
195 cm ; H NMR (DMSOꢀd
6
) δ (ppm) : 4.58 (d, J = 5.1 Hz, 2H, CH
2
); 5.43 (t, J = 5.5 Hz,
1H, OH); 7.30 (dd, J = 8.1 Hz/1.6 Hz, 1H, 6’ꢀH); 7.42 (dd, J = 8.1 Hz/1.1 Hz, 1H, 4’ꢀH);
7.45ꢀ7.48 (m, 2H, 8ꢀH, 2’ꢀH); 7.52 (t, J = 8.1 Hz, 1H, 5’ꢀH); 7.73 (dd, J = 8.6 Hz/1.8 Hz, 1H,
1
3
7
ꢀH); 7.90 (s, 1H, 5ꢀH); 9.10 (s, 1H, 4ꢀH). C NMR (DMSOꢀd ) δ (ppm) : 61.8; 116.0; 116.1;
6
117.4; 120.9; 122.3; 126.3; 127.8; 131.1; 133.4; 133.6; 150.9; 151.0; 153.8; 155.9; 160.7.
+
HRMS for C H ClO + H calcd, 331.0368; found, 331.0363. Anal. (C H ClO )
17 11
5
17 11
5
theoretical (%): C: 61.74; H: 3.35. Found (%): C: 62.00 ; H: 3.33.
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 52
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3-Bromophenyl 6-(hydroxymethyl)-2-oxo-2H-1-benzopyran-3-carboxylate (3d):
yields: 75%; m.p.: 157ꢀ159°C ; IR (KBr) 3504, 3059, 1748, 1621, 1571, 1473, 1397, 1376,
ꢀ1
1
1
245, 1196, 1175, 1115 cm ; H NMR (DMSOꢀd
6
2
) δ (ppm): 4.59 (d, J = 5.2 Hz, 2H, CH );
5
.42 (t, J = 5.5 Hz, 1H, OH); 7.34 (dd, J = 8.1 Hz/1.4 Hz, 1H, 6’ꢀH); 7.45ꢀ7.48 (m, 2H, 8ꢀH,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5’ꢀH); 7.55 (d, J = 8.1 Hz, 1H, 4’ꢀH); 7.61 (t, J = 1.9 Hz, 1H, 2’ꢀH); 7.74 (dd, J = 8.6 Hz/1.8
13
Hz, 1H, 7ꢀH); 7.90 (s, 1H, 5ꢀH); 9.10 (s, 1H, 4ꢀH). C NMR (DMSOꢀd
6
) δ (ppm): 61.8;
1
1
16.0; 116.1; 117.4; 121.2; 121.5; 125.1; 127.8; 129.2; 131.4; 133.6; 139.5; 150.9; 151.0;
+
53.8; 155.9; 160.7. HRMS for C17
) theoretical (%): C: 54.42 ; H: 2.96. Found (%): C: 53.98 ; H: 2.96.
-Chlorophenyl 6-(fluoromethyl)-2-oxo-2H-1-benzopyran-3-carboxylate (4a). 3ꢀ
5
H11BrO + H calcd, 374.9863; found, 374.9865. Anal.
(C17H11BrO
5
3
Chlorophenyl 6ꢀ(hydroxymethyl)ꢀ2ꢀoxoꢀ2Hꢀ1ꢀbenzopyranꢀ3ꢀcarboxylate (3c) (0.4 g, 0.68
mmol) was dissolved in methylene chloride (10 mL) and the resulting solution cooled on an
ice bath was supplemented with Ishikawa’s reagent (2.5 eq.). The reaction mixture was stirred
for 30 min at 0°C and then heated under reflux for 90 min. After cooling, the reaction medium
was poured onto water (20 mL). The organic layer was collected, dried over magnesium
sulfate, filtered and concentrated under reduced pressure. The residue of the title compound
was purified by column chromatography on silicagel (eluent: methylene chloride) and then
recrystalized in a mixture of methylene chloride and hexane: yields: 54%; m.p. : 178ꢀ179 °C;
ꢀ1 1
IR (KBr) 3081, 1777, 1720, 1625, 1578, 1378, 1247, 1221, 1205, 1174, 1112 cm ; H NMR
DMSOꢀd ) δ (ppm) : 5.52 (d, J = 47.6 Hz, 2H, CH ); 7.30 (dd, J = 8.1 Hz/1.9 Hz, 1H, 6’ꢀH);
.43 (dd, J = 7.9 Hz/1.4 Hz, 1H, 4’ꢀH); 7.48 (t, J = 2 Hz, 1H, 2’ꢀH); 7.54 (t, J = 8 Hz, 1H, 5’ꢀ
(
6
2
7
H); 7.56 (d, J = 8.5 Hz, 1H, 8ꢀH); 7.85 (d, J = 8.6 Hz, 1H, 7ꢀH); 8.05 (s, 1H, 5ꢀH); 9.12 (s,
13
1
H, 4ꢀH). C NMR (DMSOꢀd
6
) δ (ppm): 82.5ꢀ83.8; 116.5; 116.7; 117.7; 120.9; 122.3; 126.4;
1
29.8; 131.1; 132.0ꢀ133.1; 133.4; 134.6; 150.5; 150.9; 154.9; 155.6; 160.6. HRMS for
+
C H ClFO + H calcd, 333.0324; found, 333.0329. Anal. (C H ClFO ) theoretical (%): C:
17 10
4
17 10
4
18
ACS Paragon Plus Environment

Page 19 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
61.37 ; H: 3.03. Found (%): C: 60.68 ; H: 3.06.
General synthetic pathway to (3-halo)phenyl 6-(bromomethyl)-2-oxo-2H-1-
benzopyran-3-carboxylates (4b-e). The appropriate (3ꢀhalo)phenyl 6ꢀ(hydroxymethyl)ꢀ2ꢀoxoꢀ
2
Hꢀ1ꢀbenzopyranꢀ3ꢀcarboxylate (0.4 g, ~0.7 mmol) was dissolved in methylene chloride (10
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
mL) and then supplemented with pyridine (1.0 eq.) and thionyl bromide (1.15 eq.). The
reaction mixture was heated under reflux for 90 min. After cooling, the reaction medium was
poured onto water (20 mL). The organic layer was collected, dried over magnesium sulfate,
filtered and concentrated under reduced pressure. The residue of the title compound was
recrystalized in a mixture of methylene chloride and hexane (yields: 15ꢀ50%).
According to this method, the following compounds were obtained:
Phenyl 6-(bromomethyl)-2-oxo-2H-1-benzopyran-3-carboxylate (4b): yields: 15%;
ꢀ1
1
m.p. : 186ꢀ187 °C; IR (KBr) 3065, 1760, 1621, 1574, 1377, 1249, 1225, 1195 cm ; H NMR
DMSOꢀd ) δ (ppm): 4.82 (s, 2H, CH ); 7.28 (d, J = 7.7 Hz, 2H, 12’ꢀH, 6’ꢀH); 7.34 (t, J = 7.4
Hz, 1H, 4’ꢀH); 7.50 (m, 3H, 8ꢀH, 3’ꢀH, 5’ꢀH); 7.86 (dd, J = 8.6 Hz/2.1 Hz, 1H, 7ꢀH); 8.06 (d,
(
6
2
13
J = 2 Hz, 1H, 5ꢀH); 9.06 (s, 1H, 4ꢀH). C NMR (DMSOꢀd
6
) δ (ppm): 33.0; 116.9; 117.0;
117.9; 121.8; 126.2; 129.7; 130.8; 135.0; 135.9; 149.9; 150.3; 154.4; 155.7; 161.0. HRMS for
+
C
17 4 4
H11BrO + H calcd, 358.9913; found, 358.9917. Anal. (C17H11BrO ) theoretical (%): C:
56.85 ; H: 3.09. Found (%): C: 56.82 ; H: 3.15.
-Fluorophenyl 6-(bromomethyl)-2-oxo-2H-1-benzopyran-3-carboxylate (4c):
yields: 15%; m.p.: 189ꢀ190 °C ; IR (KBr) 3071, 1754, 1709, 1620, 1572, 1487, 1379, 1227
3
ꢀ1 1
cm ; H NMR (DMSOꢀd
6 2
) δ (ppm): 4.82 (s, 2H, CH ); 7.22 (m, 3H, 2’ꢀH, 4’ꢀH, 6’ꢀH); 7.51
(d, J = 8.6 Hz, 1H, 8ꢀH); 7.55 (m, 1H, 5'ꢀH); 7.87 (dd, J = 8.6 Hz/2.1 Hz, 1H, 7ꢀH); 8.06 (d, 2
13
Hz, 1H, 5ꢀH); 9.08 (s, 1H, 4ꢀH). C NMR (DMSOꢀd
6
) δ (ppm): 33.0; 109.8ꢀ110.0; 113.1ꢀ
113.3; 116.5; 116.9; 117.8; 118.2; 130.9; 131.0; 135.0; 136.1; 150.3; 151.1; 154.4; 155.6;
+
1
60.5; 161.2; 163.2. HRMS for C H BrFO + H calcd, 376.9819; found, 376.9823. Anal.
17 10
4
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 52
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(C H BrFO ) theoretical (%): C: 54.14 ; H: 2.67. Found (%): C: 53.88 ; H: 2.71.
1
7
10
4
3-Chlorophenyl 6-(bromomethyl)-2-oxo-2H-1-benzopyran-3-carboxylate (4d):
yields: 50%; m.p. : 186ꢀ187 °C ; IR (KBr) 3088, 1776, 1725, 1619, 1574, 1376, 1247, 1224,
ꢀ1 1
1
206, 1112 cm ; H NMR (DMSOꢀd ) δ (ppm): 4.83 (s, 2H, CH ); 7.30 (dd, J = 8.1 Hz/1.4
6
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Hz, 1H, 6’ꢀH); 7.43 (dd, J = 8 Hz/1.2 Hz, 1H, 4’ꢀH); 7.48 (t, J = 2.1 Hz, 1H, 2'ꢀH); 7.52 (m,
2
1
1
H, 8ꢀH, 5’ꢀH); 7.87 (dd, J = 8.6 Hz/2.1 Hz, 1H, 7ꢀH); 8.06 (d, J = 2.1 Hz, 1H, 5ꢀH); 9.08 (s,
13
H, 4ꢀH). C NMR (DMSOꢀd
6
) δ (ppm): 33.0; 116.5; 116.9; 117.8; 120.9; 122.3; 126.4;
30.9; 131.1; 133.4; 135.1; 136.1; 150.3; 150.9; 154.4; 155.6; 160.6. HRMS for
+
C
17
H
10BrClO
4
+ H calcd, 392.9524; found, 392.9526. Anal. (C17
H
10BrClO
4
) theoretical (%):
C: 51.87 ; H: 2.56. Found (%): C: 51.54 ; H: 2.55.
-Bromophenyl 6-(bromomethyl)-2-oxo-2H-1-benzopyran-3-carboxylate (4e):
yields: 16%; m.p.: 185ꢀ186 °C; IR (KBr) 3088, 1773, 1726, 1619, 1573, 1470, 1375, 1246,
3
ꢀ1
1222, 1202, 1112 cm ; 1H NMR (DMSOꢀd ) δ (ppm): 4.82 (s, 2H, CH ); 7.34 (dd, J = 8.1
6
2
Hz/1.3 Hz, 1H, 6’ꢀH); 7.47 (t, J = 8.1 Hz, 1H, 5’ꢀH); 7.51 (d, J = 8.6 Hz, 1H, 8ꢀH); 7.56 (d, J
7.9 Hz, 1H, 4’ꢀH); 7.61 (t, J = 1.9 Hz, 1H, 2'ꢀH); 7.88 (dd, J = 8.6 Hz/2.1 Hz, 1H, 7ꢀH);
=
13
8
6
.06 (d, J = 2 Hz, 1H, 5ꢀH); 9.07 (s, 1H, 4ꢀH). C NMR (DMSOꢀd ) δ (ppm): 33.0; 116.5;
116.9; 117.8; 121.2; 121.5; 125.1; 129.2; 130.9; 131.4; 135.1; 136.1; 150.3; 150.9; 154.4;
+
155.6; 160.6. HRMS for C17
H10Br
2
O
4
+ H calcd, 436.9019; found, 436.9024. Anal.
(
C H Br O ) theoretical (%): C: 46.61 ; H: 2.30. Found (%): C: 46.32 ; H: 2.32.
17 10 2 4
General synhetic pathway to dinaphthyl esters of malonic acid. The mixture of
malonic acid (2 g, 19.2 mmol), the appropriate naphthol (1ꢀnaphthol or 2ꢀnaphthol) (2 eq.,
8.4 mmol) and phosphoryl chloride (2 eq., 38.4 mmol) was heated at 100°C for 90 min. The
3
reaction mixture was then carefully poured on a cooled 1% w/v aqueous solution of sodium
hydroxide (50 mL). The resulting precipitate of the title compound was collected by filtration,
washed with water, dried and recrystallized in methanol (yields: 25ꢀ30%). The white solid of
20
ACS Paragon Plus Environment

Page 21 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
the title compound (diꢀ1ꢀnaphthyl malonate or diꢀ2ꢀnaphthyl malonate) was used in the next
step without further purification.
General synthetic pathway to naphthyl 2-oxo-2H-1-benzopyran-3-carboxylates 6
and 8. The solution of salicylaldehyde (0.4 g, 3.28 mmol) and the appropriate dinaphthyl
malonate (1.5 eq., 4.92 mmol) in dioxane (12 mL) was supplemented with 8 drops of
piperidine and 4 drops of glacial acetic acid and stirred for 30 min at room temperature. At the
end of the reaction, the solvent was removed by distillation under reduced pressure and the
residue was triturated with cold methanol (20 mL). The resulting precipitate was collected by
filtration, washed with cold methanol and dried. The white solid of the title compound was
recrystallized in a mixture of methylene chloride and hexane (yields: 75ꢀ80%).
According to this method, the following compounds were obtained:
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1-Naphthyl 2-oxo-2H-1-benzopyran-3-carboxylate (6): yields: 80ꢀ85%; m.p.: 159ꢀ160 °C;
ꢀ
1
1
IR (KBr) 3066, 1743, 1611, 1565, 1305, 1235, 1222, 1210 cm ; H NMR (DMSOꢀd
(ppm) : 7.48 (m, 2H, Harom.); 7.54 (d, J = 8.4 Hz, 1H, Harom.); 7.61 (m, 3H, Harom.); 7.83 (t, J =
.2 Hz, 1H, H arom.); 7.93 (d, J = 8.2 Hz, 1H, Harom.); 8.06 (m, 3H, Harom.); 9.22 (s, 1H, 4ꢀH).
6
) δ
7
13
C NMR (DMSOꢀd ) δ (ppm): 116.3; 116.6; 117.9; 118.6; 121.3; 125.0; 125.8; 126.3; 126.8;
6
12 4
126.9; 128.0; 130.7; 134.2; 135.1; 146.1; 150.6; 154.9; 156.1; 161.5. HRMS for C20H O +
+
H calcd, 317.0808; found, 317.0806. Anal. (C20
12 4
H O ) theoretical (%): C: 75.94 ; H: 3.82.
Found (%): C: 75.54 ; H: 3.81.
2-Naphthyl 2-oxo-2H-1-benzopyran-3-carboxylate (8): yields: 75ꢀ80%; m.p.: 166ꢀ167 °C;
ꢀ1 1
IR (KBr) 3047, 1761, 1715, 1610, 1568, 1454, 1377, 1304, 1236, 1210, 1128, 1152 cm ; H
NMR (DMSOꢀd ) δ (ppm): 7.47 (m, 2H, Harom.); 7.52 (d, J = 8.4 Hz, 1H, Harom.); 7.57 (m, 2H,
arom.); 7.82 (m, 1H, Harom.); 7.84 (d, J = 2.1 Hz, 1H, 1'ꢀH); 7.99 (m, 2H, Harom.); 8.03 (dd, J =
6
H
13
7
.8 Hz/1.2 Hz, 1H, Harom.); 8.05 (d, J = 8.9 Hz, 1H, Harom.); 9.16 (s, 1H, 4ꢀH). C NMR
(
DMSOꢀd ) δ (ppm): 116.3; 116.4; 117.9; 118.7; 121.5; 125.0; 126.0; 126.8; 127.6; 127.8;
6
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 52
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
129.5; 130.6; 131.1; 133.3; 135.1; 147.9; 150.6; 154.8; 155.9; 161.2. HRMS for C H O +
2
0
12
4
+
H calcd, 317.0808; found, 317.0804. Anal. (C H O ) theoretical (%): C: 75.94 ; H: 3.82.
20
12
4
Found (%): C: 75.53 ; H: 3.84.
Isoquinol-1-yl 2-oxo-2H-1-benzopyran-3-carboxylate (56): 1.0 g of the commercially
available 2ꢀoxoꢀ2Hꢀ1ꢀbenzopyranꢀ3ꢀcarboxylic acid and 10 mL of thionyl chloride were
refluxed for 3 h. The resulting solution was evaporated under reduced pressure. The residue
was suspended in 10 mL anhydrous toluene. The solvent was eliminated by distillation under
reduced pressure. The two last steps were repeated twice. The residue was dispersed in 10
mL dioxane. To this suspension were added 1ꢀhydroxyisoquinoline (1.1 eq.) and anhydrous
pyridine (1.1 eq.). After 90 min stirring at room temperature, the solvent was removed by
distillation under reduced pressure. The residue was solubilized in chloroform and the organic
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
phase was washed three times with HCl 0.1N, then dried over MgSO . The solvent was
4
evaporated under reduced pressure and the residue obtained was recrystallized in ethyl
acetate. Yields: 15%. White solid: m.p. 197ꢀ200°C; IR (KBr) 3076, 1726, 1706, 1676, 1641,
ꢀ1 1
6
1599, 1293, 1271, 1247, 1228, 1208, 1170, 1138 cm ; H NMR (DMSOꢀd ) d (ppm): 6.87 (d,
J = 7.7 Hz, 1H, 4ꢀHisoquin.), 7.47 (t, J = 7.2 Hz, 1H, 6ꢀHcoumar.), 7.52 (d, J = 8.2 Hz, 1H, 8ꢀ
Hcoumar.), 7.58 (t, J = 7.3 Hz, 1H, 7ꢀHisoquin.), 7.75 (m, 2H, 7ꢀHcoumar. + 5ꢀHisoquin.), 7.84 (m, 2H,
6ꢀH + 3ꢀHisoquin.), 7.93 (d, J = 7.2 Hz, 1H, 5ꢀHcoumar.), 8.16 (d, J = 7.7 Hz, 1H, 8ꢀHisoquin.), 8.56
13
(s, 1H, 4ꢀHcoumar.). C NMR (DMSOꢀd
6
) d (ppm): 108.7; 116.4; 118.2; 124.8; 125.2; 125.6;
125.7; 126.8; 128.0; 128.1; 129.9; 133.9; 134.6; 136.6; 143.7; 153.7; 157.3; 162.1; 166.4.
+
HRMS for C H NO + H calcd, 318.0761; found, 318.0547. Anal. (C H NO ) theoretical
19
11
4
19 11
4
(%): C: 71.92; H:3.49; N: 4.41. Found (%): C: 71.46; H: 3.45; N: 4.29.
General synthetic pathway to the monosubstituted phenyl esters of 2-oxo-2H-1-benzopyran-
3-carboxylic acid 18, 19, 28, 29, 30 and 33. 6ꢀ(Hydroxymethyl)ꢀ2Hꢀ1ꢀbenzopyranꢀ3ꢀ
27
carboxylic acid (1.0 g, 4.54 mmol) obtained as previously described
was suspended in
22
ACS Paragon Plus Environment

Page 23 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
thionyl chloride (10 mL) and refluxed for 3 h. The resulting solution was evaporated to
dryness under reduced pressure, and the residue of crude 6ꢀ(chloromethyl)ꢀ2Hꢀ1ꢀbenzopyranꢀ
3ꢀcarbonyl chloride was dispersed in dry toluene (10 mL). The solvent was eliminated under
reduced pressure. Dispersion in dry toluene and solvent elimination was repeated twice
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
7
according to the previously described protocol. The crude acid chloride was dispersed in
anhydrous dioxane (10 mL) and supplemented under stirring with the appropriate phenol (5.0
mmol) and anhydrous pyridine (10.0 mmol). After 12ꢀ48 h stirring at room temperature, the
solvent was eliminated by distillation under reduced pressure. The residue was partitioned
between chloroform (60 mL) and 0.1 N HCl (30 mL), and the organic layer was decanted,
4
dried over MgSO , filtered, and evaporated to dryness under reduced pressure. The crude final
compound was then recrystallized in the appropriate solvent.
According to this method, the following compounds were obtained:
2
-Carboxyphenyl 6-(chloromethyl)-2-oxo-2H-1-benzopyran-3-carboxylate (18): the
compound was recrystallized in methanol/water and then in dioxane/water; yields: 45%; m.p.:
ꢀ
1
1
152ꢀ155 °C; IR (KBr) 3081, 1775, 1673, 1624, 1575, 1487, 1374, 1246 cm ; H NMR
DMSOꢀd ) δ (ppm): 4.88 (s, 2H, CH Cl), 7.36 (d, J = 2.1 Hz, 1H, 6'ꢀH), 7.47 (td, J = 7.6
Hz/1.0 Hz, 1H, 4'ꢀH), 7.53 (d, J = 8.6 Hz, 1H, 8ꢀH), 7.73 (td, J = 7.9 Hz/1.7 Hz, 1H, 5'ꢀH),
.87 (dd, J = 8.6 Hz/2.2 Hz, 1H, 7ꢀH), 8.01 (dd, J = 7.8 Hz/1.6 Hz, 1H, 3'ꢀH), 8.09 (d, J = 2.1
(
6
2
7
1
3
Hz, 1H, 5ꢀH), 9.09 (s, 1H, 4ꢀH), 13.19 (bs, 1H, COOH). C NMR (DMSOꢀd
6
) δ (ppm): 45.0;
16.7; 116.8; 117.7; 123.9; 124.0; 126.6; 130.6; 131.6; 134.1; 134.6; 135.6; 149.8; 150.2;
54.5; 155.6; 160.6; 165.4. Anal. (C H ClO ) theoretical (%): C: 60.27; H: 3.09. Found (%):
1
1
18
11
6
C: 59.94; H: 3.07.
2
-Methoxycarbonylphenyl 6-(chloromethyl)-2-oxo-2H-1-benzopyran-3-carboxylate (19):
the compound was recrystallized in acetone; yields: 46%; m.p.: 152ꢀ155 °C ; IR (KBr) 3069,
ꢀ1 1
1
769, 1716, 1622, 1575, 1485, 1445, 1376, 1217, 1202 cm ; H NMR (DMSOꢀd ) δ 3.76 (s,
6
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 52
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3H, OCH ), 4.89 (s, 2H, CH Cl), 7.41 (dd, J = 8.1 Hz/0.9 Hz, 1H, 6'ꢀH), 7.50 (td, J = 7.7
3
2
Hz/1.1 Hz, 1H, 4'ꢀH), 7.53 (d, J = 8.6 Hz, 1H, 8ꢀH), 7.77 (td, J = 7.9 Hz/1.7 Hz, 1H, 5'ꢀH),
.87 (dd, J = 8.6 Hz/2.2 Hz, 1H, 7ꢀH), 8.02 (dd, J = 7.8 Hz/1.7 Hz, 1H, 3'ꢀH), 8.10 (d, J = 2.1
7
13
Hz, 1H, 5ꢀH), 9.11 (s, 1H, 4ꢀH). C NMR (DMSOꢀd ) δ 45.0; 52.4; 116.6; 116.8; 117.8;
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
122.9; 124.1; 126.8; 130.6; 131.4; 134.5; 134.6; 135.6; 149.6; 150.3; 154.6; 155.6; 160.6;
+
1
64.4. HRMS for C H ClO + H calcd, 373.0473; found, 373.0509. Anal. (C H ClO )
19
13
6
19 13
6
theoretical (%): C: 61.22; H: 3.51. Found (%): C: 61.17; H: 3.57.
-Carboxyphenyl 6-(chloromethyl)-2-oxo-2H-1-benzopyran-3-carboxylate (28): the
compound was recrystallized in methanol/water; yields: 56%; m.p.: 231ꢀ235 °C ; IR (KBr)
3
ꢀ
1 1
3
7
5
078, 1770, 1685, 1622, 1577, 1490, 1374 cm ; H NMR (DMSOꢀd
6
2
) δ 4.89 (s, 2H, CH Cl),
.53 (d, J = 8.6 Hz, 1H, 8ꢀH), 7.57 (dd, J = 8.1 Hz/1.2 Hz, 1H, 6'ꢀH), 7.64 (t, J = 7.9 Hz, 1H,
'ꢀH), 7.83 (s, 1H, 2'ꢀH), 7.86 (dd, J = 8.6 Hz/2.0 Hz, 1H, 7ꢀH), 7.91 (d, J = 7.6 Hz, 1H, 4'ꢀH),
13
8.06 (d, J = 1.8 Hz, 1H, 5ꢀH), 9.12 (s, 1H, 4ꢀH), 13.25 (bs, 1H, COOH). C NMR (DMSOꢀ
d
1
3
6
) δ 45.0; 116.8; 116.8; 117.8; 122.6; 126.4; 127.0; 130.1; 130.5; 132.4; 134.6; 135.5; 150.2;
+
50.2; 154.5; 155.7; 160.8; 166.4. HRMS for C H ClO + H calcd, 359.0317; found,
18
11
6
6
59.0350. Anal. (C18H11ClO ) theoretical (%): C: 60.27; H: 3.09. Found (%): C: 60.62; H:
3.12.
-Methoxycarbonylphenyl 6-(chloromethyl)-2-oxo-2H-1-benzopyran-3-carboxylate (29):
the compound was recrystallized in acetone; yields: 32%; m.p.: 128ꢀ133 °C ; IR (KBr) 3076,
3
ꢀ1 1
1
761, 1724, 1622, 1574, 1486, 1432, 1377, 1220, 1204 cm ; H NMR (DMSOꢀd
6
) δ 3.89 (s,
3H, OCH ), 4.89 (s, 2H, CH Cl), 7.53 (d, J=8.6 Hz, 1H, 8ꢀH), 7.61 (ddd, J = 8.1 Hz/2.3
3
2
Hz/1.1 Hz, 1H, 6'ꢀH), 7.67 (t, J = 7.9 Hz, 1H, 5'ꢀH), 7.86 (m, 2H, 7ꢀH/2'ꢀH), 7.93 (dt, J = 7.7
13
Hz/1.3 Hz, 1H, 4'ꢀH), 8.06 (d, J = 2.1 Hz, 1H, 5ꢀH), 9.12 (s, 1H, 4ꢀH). C NMR (DMSOꢀd
6
)
δ 45.0; 52.4; 116.7; 116.8; 117.8; 122.5; 126.8; 126.9; 130.3; 130.5; 131.1; 134.6; 135.6;
+
150.2; 150.3; 154.5; 155.7; 160.7; 165.4. HRMS for C19
H13ClO
6
+ H calcd, 373.0473; found,
24
ACS Paragon Plus Environment

Page 25 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
373.0513. Anal. (C H ClO ) theoretical (%): C: 61.22; H: 3.51. Found (%): C: 60.89; H:
1
9
13
6
3.47.
3-Carbamoylphenyl 6-(chloromethyl)-2-oxo-2H-1-benzopyran-3-carboxylate (30): the
compound was recrystallized in chloroform/petroleum ether (40ꢀ60 °C) and then in
dioxane/water; yields: 82%; m.p.: 190ꢀ192 °C ; IR (KBr) 3367, 3187, 3074, 1772, 1754,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ꢀ
1 1
1652, 1622, 1575, 1491, 1380, 1241, 1210 cm ; H NMR (DMSOꢀd
.46 (ddd, J = 8.1 Hz/2.3 Hz/0.8 Hz, 1H, 6'ꢀH), 7.51 (bs, 1H, NH), 7.53 (d, J = 8.6 Hz, 1H, 8ꢀ
H), 7.58 (t, J = 7.9 Hz, 1H, 5'ꢀH), 7.78 (m, 1H, 2'ꢀH), 7.86 (m, 2H, 7ꢀH/4'ꢀH), 8.06 (d, J = 2.1
6
) δ 4.89 (s, 2H, CH
2
Cl),
7
1
3
Hz, 1H, 5ꢀH), 8.09 (bs, 1H, NH), 9.11 (s, 1H, 4ꢀH). C NMR (DMSOꢀd
6
) δ 45.0; 116.8;
1
1
16.8; 117.8; 121.0; 124.7; 125.2; 129.6; 130.6; 134.6; 135.6; 135.9; 150.1; 150.2; 154.5;
+
55.7; 160.9; 166.7. HRMS for C H ClNO + H calcd, 358.0477; found, 358.0511. Anal.
18
12
5
18 5
(C H12ClNO ) theoretical (%): C: 60.43 ; H: 3.38 ; N: 3.91. Found (%): C: 59.92; H: 3.72; N:
3.73.
4-Methoxycarbonylphenyl 6-(chloromethyl)-2-oxo-2H-1-benzopyran-3-carboxylate (33):
the compound was recrystallized in chloroform/petroleum ether (40ꢀ60 °C) and then in
acetone/petroleum ether (40ꢀ60°C); yields: 35%; m.p.: 212ꢀ215 °C ; IR (KBr) 3069, 1777,
ꢀ
1 1
1
718, 1623, 1576, 1491, 1433, 1377, 1224, 1208 cm ; H NMR (DMSOꢀd
OCH ), 4.89 (s, 2H, CH Cl), 7.46 (d, J = 8.6 Hz, 2H, 2'ꢀH/6'ꢀH), 7.53 (d, J = 8.6 Hz, 1H, 8ꢀ
H), 7.86 (dd, J = 8.6 Hz/1.8 Hz, 1H, 7ꢀH), 8.06 (d, J = 1.4 Hz, 1H, 5ꢀH), 8.09 (d, J = 8.6 Hz,
6
) δ 3.88 (s, 3H,
3
2
13
2
6
H, 3'ꢀH/5'ꢀH), 9.12 (s, 1H, 4ꢀH). C NMR (DMSOꢀd ) δ 44.6; 52.3; 117.5; 117.6; 117.7;
121.6; 128.2; 129.5; 131.3; 134.7; 135.3; 149.8; 153.9; 155.2; 156.0; 160.9; 166.2. HRMS for
+
C
19
H
13ClO
6
+ H calcd, 373.0473; found, 373.0510. Anal. (C19
H13ClO
6
) theoretical (%): C:
6
1.22; H: 3.51. Found (%): C: 60.97; H: 3.50.
Enzyme and inhibition assays. Kallikreins activities were determined by monitoring
the hydrolysis of the appropriate fluorogenic substrate (λexc = 355 nm, λem = 460 nm for AMC
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 52
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
substrates) at 37 °C in the presence of untreated kallikreins (control) or kallikreins that had
been incubated with a test compound. Substrates and compounds were previously dissolved in
DMSO, with the final solvent concentration kept constant at 2 % (v/v). The composition of
the activity buffers was (pH 8.0): 50 mM TrisꢀHCl, Tween 20 0.01 % (v/v), and 150 mM
NaCl for KLK5, 1M NaCl for KLK7 and KLK14. The final concentrations were 0.6 nM
(KLK5), 100 µM (BocꢀValꢀProꢀArgꢀAMC); 7.6 nM (KLK7), 40 µM (SucꢀLeuꢀLeuꢀValꢀTyrꢀ
AMC); 0.2 nM (KLK14), 10 µM (BocꢀValꢀProꢀArgꢀAMC); 0.1 nM (matriptase), 10 µM
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(BocꢀGlnꢀAlaꢀArgꢀAMC). Compounds (0.1ꢀ100 µM) were tested in triplicate for each
inhibitor concentration to detect their inhibitory potential. The enzyme and the inhibitors were
incubated for 15 min before the determination of the enzyme activity. Initial rates determined
in control experiments (V ) were considered to be 100 % of the proteinase activity; initial
0
rates (V ) that were below 100 % in the presence of a tested compound were considered to be
i
inhibitions. The inhibitory activity of compounds was expressed as IC50 (inhibitor
concentrations giving 50% inhibition). The values of IC50 were calculated by fitting the
experimental data either to equation 1: % Inhibition = 100 x (1ꢀV
i
/V
0 0 0
) = 100 [I] /(IC50 + [I] ),
nH
nH
nH
or equation 2, % Inhibition = 100 [I]0 /(IC50 + [I]0 ) with n = Hill number.
H
Mechanistic studies. Reversibility was evaluated either by dilution or by treatment
with hydroxylamine. The reaction mixtures were usually diluted 40ꢀfold after 15 and 60 min
preincubation of the enzyme with inhibitor. Aliquots of reaction mixtures (2.5 µL) were
added to 97.5 µL of buffer containing the fluorogenic substrate (experimental conditions
identical to the routine protocol used for a given enzyme) or eventually immediately treated
by 0.5 M of hydroxylamine at pH 8.0 during 60 min before determination of remaining
activity. For nonꢀcovalent inhibitors, the mechanisms of inhibition were determined by
varying substrates and inhibitors concentrations and using classical representations
(LineweaverꢀBurk). For suicide inhibitors, inactivation can be represented by the minimum
26
ACS Paragon Plus Environment

Page 27 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
2
6
kinetic scheme (eq. 3) , where E and I are the free forms of enzyme and inhibitor, E*I a
kinetic chimera of the Michaelis complex and the acylꢀenzyme, E
and P the products.
i
the inactivated enzyme,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(
eq. 3)
were determined for KLK7 using the progress curve method.
The ratio k /K was obtained by fitting the experimental data to the equations (F.U.,
i I
The kinetic constants k and K
2
6
i
I
fluorescence unit):
(eq. 4)
with
and
Linear and nonꢀlinear regression fits of the experimental data to the equations were performed
with Kaleidagraph Software. The experimental conditions were: [KLK7] = 7,6 nM , [S] = 40
µM, [I] = 0ꢀ200 nM.
0
0
0
Cell culture assay. Human skin was obtained from healthy donors undergoing
abdominal plastic surgery. Samples were obtained after informed consent and used according
to the guidelines of the Declaration of Helsinki Principles. Healthy keratinocytes are spread at
a rate of 4,000 cells per well of 96ꢀwell plates in 125 µL of medium/well consisting of: 50%
of complete culture medium Green containing 60% of DMEM (Dulbecco's Modified Eagle
Media), 30% of Ham's F12, 10% of FCS, 180 mM of adenine, 5 mg/mL of insulin, 0.4
mg/mL of hydrocortisone, 10 nM of cholera toxin, 2 nM of triiodothyronine, 10 ng/mL of
human recombinant EGF, 100 U /mL of penicillin G / streptomycin, and 50% of basal
medium (Epilife ꢀ Cascade Biologics) for the culture of human keratinocytes without feeder
layer. 48 h after seeding the cells, 125 µL of fresh medium were mixed with the inhibitor in
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 52
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
DMSO to give a concentration of 2 or 20 µM, and then this solution was added to the wells to
obtain a final concentration of 1 or 10 µM. For each condition, at least three wells are
realized. After 48 h incubation, cells were washed with PBS. Then 250 µL of NR medium
(
culture medium containing neutral red 33 µg/mL) were added, and cells were incubated at
7°C and 5% of CO for 3 h. Then, the medium containing neutral red was removed; the cells
were washed with PBS. Then, 100 µL of NR desorbed solution (1% acetic acid, 50% ethanol,
9% H O) were added, the plate was shaked for 20ꢀ40 min and the absorbance at 540 nm
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
2
4
2
read. Relative cell viability was expressed as a percentage of the viability of cells treated with
DMSO alone.
In situ zymography. In situ zymography was carried out on skin sections of transgenic
46
mice overexpressing human KLK5. Cryosections of non lesional skin (thickness of 5 µm)
were washed with a PBS solution containing 2% of Tween 20 during 2 min, then 5 min with
PBS. The skin sections were then incubated (or not) with the inhibitors at 37 ° C overnight
with 100 uL of BODIPYFL casein (10 µg /mL) using the Ultra EnzChek Protease Assay kit
(Invitrogen) in a buffer containing TrisꢀHCl 100 mM, pH 8. Sections were rinsed with PBS.
The inhibition of the release of fluorochrome was then analyzed, visualized and quantified
using a Leica TCS SP5 AOBS confocal microscope.
ASSOCIATED CONTENT
Supporting information. Three figures showing the postulated mechanisms for the inhibition
of serine proteases by coumarinꢀ3ꢀcarboxylic acid derivatives, IC50 curves and Lineweaverꢀ
Burk plot; molecular docking specification; references of supporting information. This
material is available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
28
ACS Paragon Plus Environment

Page 29 of 52
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Corresponding authors
*
Prof. Dr. Michèle ReboudꢀRavaux, Sorbonne Universités, UPMC University Paris 06, UMR
8256, B2A, Biological Adaptation and Ageing, Integrated Cellular Ageing and Inflammation,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Molecular & Functional Enzymology, Institut de Biologie Paris Seine ,7 Quai St Bernard, Fꢀ
75005 Paris, France. Phone: (33) 1 44 27 50 78. Eꢀmail: michele.reboud@upmc.fr
.
*
Dr. Chahrazade El Amri, Sorbonne Universités, UPMC University Paris 06, UMR 8256,
B2A, Biological Adaptation and Ageing, Integrated Cellular Ageing and Inflammation,
Molecular & Functional Enzymology, Institut de Biologie Paris Seine, 7 Quai St Bernard, Fꢀ
75005 Paris, France. Phone: (33) 1 44 27 59 62. Eꢀmail: chahrazade.el_amri@upmc.fr
.
*Prof. Dr. Alain Hovnanian, University Paris Descartes ꢀ Sorbonne Paris Cité, INSERM
UMR 1163, Laboratory of Genetic Skin Diseases, Imagine Institute, 24 boulevard du
Montparnasse, Fꢀ75015 Paris, France. Phone: (33) 1 71 19 63 95. Eꢀmail:
alain.hovnanian@inserm.fr.
Author contributions
The manuscript was written through contributions of all authors. All authors have given
approval to the final version of the manuscript.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
This work was supported by the Institut National pour la Recherche Médicale (INSERM), the
Université Pierre et Marie Curie (UPMC), the Centre National de la Recherche Scientifique
(CNRS), and the Agence Nationale de Recherche (ANRꢀGENOPAT). The authors are
29
ACS Paragon Plus Environment