β-Lactones: A Novel Class of Ca2+-Independent Phospholipase A2 (Group VIA iPLA2) Inhibitors with the Ability To Inhibit β-Cell Apoptosis

Article abstract of DOI:10.1021/acs.jmedchem.8b01216

Ca²⁺-independent phospholipase A₂ (GVIA iPLA₂) has gained increasing interest recently as it has been recognized as a participant in biological processes underlying diabetes development and autoimmune-based neurological disorders. The development of potent GVIA iPLA₂ inhibitors is of great importance because only a few have been reported so far. We present a novel class of GVIA iPLA₂ inhibitors based on the β-lactone ring. This functionality in combination with a four-carbon chain carrying a phenyl group at position-3 and a linear propyl group at position-4 of the lactone ring confers excellent potency. trans-3-(4-Phenylbutyl)-4-propyloxetan-2-one (GK563) was identified as being the most potent GVIA iPLA₂ inhibitor ever reported (X_I(50) 0.0000021, IC₅₀ 1 nM) and also one that is 22 000 times more active against GVIA iPLA₂ than GIVA cPLA₂. It was found to reduce β-cell apoptosis induced by proinflammatory cytokines, raising the possibility that it can be beneficial in countering autoimmune diseases, such as type 1 diabetes.

Full text of DOI:10.1021/acs.jmedchem.8b01216

Subscriber access provided by Macquarie University
Article
#-Lactones: A Novel Class of Ca2+-Independent Phospholipase A2
(Group VIA iPLA2) Inhibitors with Ability to Inhibit #-Cell Apoptosis
Christina Dedaki, Maroula G. Kokotou, Varnavas D Mouchlis, Dimitris Limnios, Xiaoyong Lei,
Carol T Mu, Sasanka Ramanadham, Victoria Magrioti, Edward A. Dennis, and George Kokotos
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.8b01216 • Publication Date (Web): 23 Feb 2019
Downloaded from http://pubs.acs.org on February 27, 2019
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 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 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.
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 49
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
β-Lactones: A Novel Class of Ca2+-
Independent Phospholipase A2 (Group VIA
iPLA2) Inhibitors with Ability to Inhibit β-
Cell Apoptosis
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
Christina Dedaki, † Maroula G. Kokotou,†,§ Varnavas D. Mouchlis,§ Dimitris
Limnios,†,§ Xiaoyong Lei,#, ‡ Carol T. Mu, § Sasanka Ramanadham, #,‡ Victoria
Magrioti,*,† Edward A. Dennis,*, § George Kokotos*,†
†Laboratory of Organic Chemistry, Department of Chemistry, National and
Kapodistrian University of Athens, Panepistimiopolis, Athens 15771, Greece
§Department of Chemistry and Biochemistry and Department of Pharmacology,
School of Medicine, University of California, San Diego, La Jolla, California 92093-
0601, USA
#Department of Cell, Developmental and Integrative Biology, University of Alabama
at Birmingham, Birmingham, AL 35294, USA
‡Comprehensive Diabetes Center, University of Alabama at Birmingham,
Birmingham, AL 35294, USA
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 49
1
2
3
4
KEYWORDS. β-cell apoptosis, β-lactones, Ca2+-independent phospholipase A2,
5
6
cytosolic phospholipase A2, inhibitors.
7
8
9
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
ABSTRACT: Interest in Ca2+-independent phospholipase A2 (GVIA iPLA2) has
accelerated recently as it is being recognized as a participant in biological processes
underlying diabetes development and autoimmune-based neurological disorders. The
development of potent GVIA iPLA2 inhibitors is of great importance, because only a
few have been reported so far. We present a novel class of GVIA iPLA2 inhibitors
based on the β-lactone ring. This functionality in combination with a four-carbon
chain carrying a phenyl group at position-3, and a linear propyl group at position-4 of
the lactone ring confers excellent potency. trans-3-(4-Phenylbutyl)-4-propyloxetan-2-
one (GK563) was identified as being the most potent GVIA iPLA2 inhibitor ever
reported (XI(50) 0.0000021, IC50 1 nM) and also one that is 22,000 times more active
against GVIA iPLA2 than GIVA cPLA2. It was found to reduce β-cell apoptosis
induced by pro-inflammatory cytokines, raising the possibility that it can be beneficial
in countering autoimmune diseases, such as type 1 diabetes.
INTRODUCTION
The phospholipase A2 (PLA2) superfamily consists of diverse enzymes, which are
currently categorized into sixteen groups and many subgroups and all are able to
hydrolyze the ester bond at the sn-2 position of glycerophospholipids.1 A number of
these enzymes do not require Ca2+ ions either for their activity or for the translocation
to membranes and are classified as Ca2+-independent PLA2s.1-5 The initial reports of
ACS Paragon Plus Environment

Page 3 of 49
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Ca2+-independent PLA2 activity referred to a 40-kDa enzyme described as iPLA2.6,7
Subsequently, an 85-kDa iPLA2 was purified and characterized from macrophages8
and cloned from hamster, mouse, and rat9-11 and is now designated as Group VIA
(GVIA) iPLA2 (also iPLA2β). To date, the group VI Ca2+-independent PLA2 includes
six different subgroups: GVIA (iPLA2β), GVIB (iPLA2γ), GVIC (iPLA2δ), GVID
(iPLA2ε), GVIE (iPLA2ζ), and GVIF (iPLA2η).1 Among them, GVIA is the most well
studied and recognized iPLA2. The human GVIA iPLA2 (806 amino acids) contains
seven ankyrin repeats (residues 152-382), a linker region (residues 383-474) with the
eighth repeat disrupted by a 54-amino-acid insert, and a catalytic domain (residues
475-806). The active site serine of GVIA iPLA2 lies within a lipase consensus
sequence (Gly486-X-Ser519-X-Gly487) in the catalytic domain. Although GVIA
iPLA2 and the other major intracellular PLA2, the calcium-dependent GIVA cPLA2,
both use a serine/aspartate catalytic dyad for their catalytic mechanism, GVIA iPLA2
does not show arachidonic acid-selectivity while GIVA cPLA2 does.12
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
PLA2s have been implicated in a number of physiological and pathophysiological
processes. Thus, a variety of synthetic inhibitors have been generated and studied in
vitro, as well as in vivo.1,13-15 The majority of these inhibitors have been developed to
target cytosolic GIVA cPLA216 and secreted sPLA2.17 Inhibitors of GVIA iPLA2 had
attracted less interest, because the enzyme’s role was less well understood. The first
introduced inhibitor for iPLA2 was a bromoenol lactone compound (1, BEL,18 Figure
1) which is an irreversible, covalent inhibitor of GVIA iPLA2.18 It has been used
widely to delineate the specific role of GVIA iPLA2 in variety of systems and
biological processes.1,13 Although BEL is selective against GVIA iPLA2 versus other
PLA2s, it also inhibits other serine enzymes (i.e. magnesium-dependent phosphatidate
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 49
1
2
3
4
phosphohydrolase)19 and therefore the data obtained from ex vivo and in vivo studies
5
6
of its inhibitory activity must be carefully considered.
7
8
9
The GVIA iPLA2 is involved in lipid signaling and pathological conditions including
diabetes,4,5,20 Barth syndrome21 and progesterone-induced acrosome exocytosis.22 The
recent emergence of GVIA iPLA2 as a contributor to pathophysiology prompted us to
explore the possibility that more potent and selective GVIA iPLA2 inhibitors can be
developed.
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
In our first series of chemical synthesis, we generated polyfluoroketone-based
compounds that proved to be more potent and selective than BEL, with the added
feature of manifesting reversible inhibition of GVIA iPLA2. These compounds
contained an aromatic ring and a small aliphatic chain as a spacer between the two
functional groups.23-25 One of these first generation fluoroketones, FKGK1123 (2a,
Figure 1, XI(50) 0.001424), was used in vivo to demonstrate a role for GVIA iPLA2 in
both the onset and progression of experimental autoimmune encephalomyelitis, an
animal model of multiple sclerosis.26 Further, the combination of BEL or FKGK11
with anticancer drug paclitaxel was highly effective in blocking ovarian cancer
development.27
Subsequent structure-activity relationship studies with the polyfluoroketones led to
the generation of FKGK1824 (3, Figure 1, XI(50) 0.000224), which contained a
naphthyl ring and a trifluoromethyl group instead of a phenyl ring and a
pentafluoroethyl group, respectively. FKGK18 was found to be 195 and >455 times
more potent for GVIA iPLA2 than for GIVA cPLA2 and GV sPLA2, respectively. In
view of this, FKGK18 was deemed a valuable tool to explore the role of GVIA iPLA2
in cells and in vivo models. Those studies revealed that FKGK18 was able to inhibit
ACS Paragon Plus Environment

Page 5 of 49
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
β-cell apoptosis28 and that its administration to spontaneous diabetes-prone non obese
diabetic (NOD) mice significantly reduced diabetes incidence in association with
reduced insulitis, improved glucose homeostasis, higher circulating insulin and β-cell
preservation.20 Subsequently, GK18725 a more potent and selective GVIA iPLA2
inhibitor (2b, Figure 1, XI(50) 0.000125) was identified. To gain a greater
understanding of the enzyme-inhibitor interactions, a robust homology model was
developed based on hydrogen/deuterium exchange mass specrtometry experimental
data and molecular dynamics simulations.29-31 Combining this with computational
chemistry, organic synthesis and in vitro assays led to identification of new thioether
fluoroketone inhibitors as well as a novel thioether keto-1,2,4-oxadiazole inhibitor (4,
Figure 1, XI(50) 0.0057) of GVIA iPLA2.32
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
Br
O
C2F5
O
O
R
2a
2b
, R=H, FKGK11
, R=CH3O, GK187
1
, BEL
O
O
S
N
CF3
O
N
O
3
4
, FKGK18
Figure 1. Known inhibitors of GVIA iPLA2.
The need for more potent and selective inhibitors of GVIA iPLA2 and the potential
pharmacological limitations of fluoroketones as human therapeutics, led us to the
quest of additional functional series of inhibitors. In this work, we developed a novel
class of GVIA iPLA2 inhibitors based on a β-lactone ring. The design and the
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 49
1
2
3
4
5
6
7
8
9
synthesis of a variety of these inhibitors, as well as assessment of their selectivity
towards the three main human PLA2s are reported. Furthermore, the ability to reduce
β-cell apoptosis induced by pro-inflammatory cytokines is demonstrated.
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
RESULTS AND DISCUSSION
Design and synthesis of inhibitors. Lipstatin (5a, Figure 2), a natural product
isolated from Streptomyces toxytricini, is a potent inhibitor of pancreatic lipase33 and
the semisynthetic derivative tetrahydrolipstatin (Orlistat, 5b, Figure 2) is an approved
drug for the treatment of obesity inhibiting lipase and thus preventing the absorption
of fats from the human diet. Studies on the mode of action of tetrahydrolipstatin
revealed that an enzyme-inhibitor complex of an acyl-enzyme type is formed, which
slowly decomposes, and that the β-lactone ring is the functional group of
tetrahydrolipstatin reacting with the active site of the enzyme.34 Several other natural
products containing a β-lactone ring are enzyme inhibitors and present attractive
pharmacological properties, for example marizomib (6, Figure 2) is a proteasome
inhibitor recently approved as an orphan drug by FDA for the treatment of multiple
myeloma.35 Structural modifications of naturally occurring β-lactones have been
proposed as an effective strategy for generating new drugs for treating bacterial
infections, cancer, obesity and hyperlipidemia.36
In general, the strained β-lactone ring is expected to be attacked by the hydroxyl
group of the active site serine of a serine-hydrolase. GVIA iPLA2 utilizes a serine
residue in its catalytic mechanism, thus, in principle it may interact with a β-lactone
ring. Our previous studies on GVIA iPLA2 inhibitors have shown that a potential
inhibitor has to be a small non-polar molecule.23-25 In addition, an aromatic ring
ACS Paragon Plus Environment

Page 7 of 49
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
attached to a four-carbon chain seems to fit very well into the binding site of GVIA
iPLA2. Thus, we designed a β-lactone and chose one of the substituents to be a
medium carbon chain carrying an aromatic ring in varying distances (Ar-Cn) from the
lactone ring. The second substituent is a small aliphatic chain (-R) containing one to
six carbon atoms. The general structure of the lactones we designed is depicted in
Figure 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
NHCHO
H
OH
H
O
O
O
N
O
O
O
O
( )4
R
Cl
5a
5b
, Lipstatin R=
6,
Marizomib
, Tetrahydrolipstatin R=
n
O
( )
Ar
O
R
Figure 2. Structures of lipstatin, tetrahydrolipstatin and marizomib and general
structure of β-lactones designed in this study.
The general route for the synthesis of the designed β-lactones is depicted in Scheme 1.
A variety of carboxylic acids containing an aromatic group at the end of the chain
were chosen as starting materials. Carboxylic acids 7a-f were deprotonated by
treatment with LDA and after reaction with commercially available aliphatic
aldehydes RCHO,37 β-hydroxy acids 8a-k were obtained. Finally, cyclization of the
intermediate α,β-substituted β-hydroxy acids 8a-k upon treatment with p-
toluenesulfonyl chloride38 led to α,β-substituted β-lactones 9a-k (Scheme 1).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 49
1
2
3
4
5
6
7
8
9
O
O
( )n
R
O
a
( )n
b
( )n
Ar
Ar
OH
Ar
OH
O
R
OH
8a-k
7a-f
9a-k
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
n
3
3
3
3
3
3
3
3
3
4
4
R
7
a
Ar
n
8, 9
Ar
3
a
b
c
d
e
f
nC6H13
nC6H13
nC6H13
nC3H7
nC3H7
nC3H7
iC3H7
C2H5
3
3
3
4
4
b
c
d
e
f
O
O
g
h
i
CH3
nC3H7
nC3H7
j
k
Scheme 1. Reagents and conditions. (a) (i) LDA (produced in situ by (i-Pr)2NH and
sol. n-BuLi 1.6 M/hexane), dry THF, 0 oC, 1 h, (ii) solution of RCHO in dry THF, 0
oC, 1 h  r.t, o.n, (b) p-TsCl in dry pyridine, 0 oC, 1 h  4 oC, 3 days.
ACS Paragon Plus Environment

Page 9 of 49
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Both β-hydroxy acids 8a-k and β-lactones 9a-k were obtained as mixtures of
diastereomers, whose ratio was estimated by 1H NMR spectroscopy. For β-hydroxy
acids 8a-k, the ratio of the peak integrations corresponding to methinic CHOH signals
was used to estimate the ratio of anti:syn diastereomers and varied from 7:3 to 6:4.
For β-lactones, the ratio of the peak integrations corresponding to methinic proton of
either C-3 or C-4 indicated a ratio of trans:cis diastereomers varying from 9:1 to 7:3.
trans β-Lactones were obtained in excess being thermodynamically more stable than
their counterparts cis products, and their geometry was determined by 1H NMR based
on the chemical shifts reported in literature for similar compounds. In accordance to
literature data, characteristic peaks for 3-CH and 4-CH are reported at 3.2 and 4.2
ppm, respectively, for trans β-lactones,39 while the corresponding chemical shifts for
cis β-lactones are reported at 3.6 and 4.5 ppm, respectively.40 The trans and cis
diastereomers of 9d, e, f, j and k were separated by column chromatography. In
particular for 9k, the coupling constants between the C-3 and C-4 protons were
measured to be 4.0 Hz and 6.7 Hz for the trans and the cis diastereomer (see,
Supporting Information), respectively, values which are in accordance with those
reported in the literature.40 Further, in 13C NMR spectra, the chemical shifts
corresponding to C-3 and C-4 are at 56.0 ppm and 77.9 ppm for the trans
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
diastereomer, while at 52.6 ppm and 75.4 ppm for the cis diastereomer.
In vitro inhibition of GVIA iPLA2, GIVA cPLA2 and GV sPLA2. All synthesized
β-lactones were tested for their in vitro inhibitory activity on recombinant human
GVIA iPLA2 using mixed micelle assays. In addition, their selectivity over human
GIVA cPLA2 and GV sPLA2 was also studied using similar group-specific mixed
micelle assays. The initial screening assays for the in vitro inhibition of human GVIA
iPLA2, GIVA cPLA2 and GV sPLA2 for the racemic lactones and their comparison
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 49
1
2
3
4
5
6
7
8
9
with previously reported fluroketones and oxadiazoles inhibitors were carried out
using our previously described radioactivity-based mixed micelle assay.41-43 For the
most potent lactones, the trans and cis diastereomers were prepared and our
previously described lipidomics-based mixed micelle assay was employed to
determine their activities.44,45 The inhibition results presented in Table 1 are either as
percent inhibition or as XI(50) values. At first, the percent of inhibition for each PLA2
enzyme at 0.091 mole fraction of each inhibitor was determined. Then, the XI(50)
values were measured for compounds that displayed greater than 95% inhibition of
GVIA iPLA2. The XI(50) is the mole fraction of the inhibitor in the total substrate
interface required to inhibit the enzyme activity by 50%. Data for inhibitors 2a,24
2b,25 324 and 432 (tested under the same radioactivity-based assay conditions) are
included in Table 1 for comparison purposes.
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
The curves for the concentration dependence of the inhibition of GVIA iPLA2 by β-
lactones were fit to sigmoidal curves and those of trans-9k (GK563) and cis-9k
(GK564) are presented as examples in Figure 3.
ACS Paragon Plus Environment

Page 11 of 49
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Figure 3. Dose-response inhibition curves for GVIA iPLA2 inhibitors trans-9k and
cis-9k. The curves were generated using GraphPad Prism with a nonlinear regression
targeted at symmetrical sigmoidal curves based on plots of % inhibition versus
log(inhibitor concentration). The reported XI(50) values were calculated from the
resultant plots.
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
At first, β-lactones carrying an aromatic group at the end of a three-carbon atom chain
and an n-hexyl chain as substituents (compounds 9a, 9b and 9c) were tested.
Irrespective of the nature of the aromatic group (phenyl, naphthyl, biphenyl), none of
these presented significant inhibition (< 90%) of GVIA iPLA2 (entries 1-3).
However, it seemed that a simple phenyl (9a), instead of a naphthyl (9c) or a biphenyl
(9b), group led to greater inhibition. Then, the hexyl chain was reduced to a shorter
chain of three-carbon atoms. Interestingly, all three compounds (9d, 9e and 9f)
combining an aromatic group at the end of a three-carbon atom chain and a n-propyl
chain (entries 4, 7 and 10) presented significant inhibition (96-98%) of GVIA iPLA2.
The mixtures of these compounds were separated by column chromatography and the
potencies of the corresponding trans and cis diastereomers were evaluated. Both
trans-9d and cis-9d presented significant inhibition for both GVIA iPLA2 and GIVA
cPLA2 (entries 5 and 6). However, trans-9d seemed to be more potent inhibitor of
GVIA iPLA2 with an XI(50) value of 0.00019 (IC50 95 nM, entry 5), while cis-9d
more potent for GIVA cPLA2 with an XI(50) value of 0.0019 (IC50 0.95 µM, entry 6).
Among the naphthyl derivatives trans-9e and cis-9e (entries 8 and 9), trans-9e was
found more potent inhibiting GVIA iPLA2 with an XI(50) value of 0.00030 (IC50 0.15
µM, entry 8). For the para-methoxyphenyl derivatives trans-9f and cis-9f (entries 11
and 12), an interesting selectivity seems to take shape. trans-9f inhibited GVIA iPLA2
with an XI(50) value equal to that estimated for trans-9d (0.00019, IC50 95 nM, entry
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 49
1
2
3
4
5
6
7
8
9
11). However, cis-9f proved to be a potent inhibitor of GIVA cPLA2 (XI(50) 0.00004,
IC50 20 nM, entry 12) being almost 1000 times more potent for GIVA cPLA2 than for
GVIA iPLA2 (XI(50) 0.038, IC50 20 µM, entry 12).
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
When the linear n-propyl group of 9d was replaced by a branched isopropyl group
(compound 9g, entry 13), the inhibitory potency over GVIA iPLA2 was reduced.
Replacement of the n-propyl group of 9e by an ethyl or a methyl group (compounds
9h, 9i, entries 14 and 15) resulted in a reduction of the inhibitory potency. Clearly, a
small linear chain of three-carbon atoms led to superior inhibitory results over GVIA
iPLA2 in comparison to a medium chain of six-carbon atoms or a short chain of one or
two carbon atoms.
Thus, by keeping a linear three-carbon chain at position-4, the distance between the
aromatic group and the lactone ring at position-3 was increased by one carbon atom
resulting in compounds 9j and 9k, which both presented high inhibition of GVIA
iPLA2 (97-100%, entries 16 and 19). The diastereomers were separated by column
chromatography and the potencies of both trans and cis diastereomers of 9j and 9k
were estimated. Both the naphthyl derivatives trans-9j and cis-9j were found to
inhibit GVIA iPLA2 with XI(50) values of 0.00009 (IC50 45 nM, entry17) and 0.0021
(IC50 1 µM, entry 18), respectively, but did not present significant inhibition of GIVA
cPLA2 (59% and 72% at a high concentration of 0.091 mole fraction, respectively,
entries 17 and 18). Gratifyingly, the combination of a four-carbon chain carrying a
phenyl group at position-3 and a linear propyl group at position-4 of the lactone ring
led to the best results. The trans diastereomer of 9k [trans-(±)-3-(4-phenylbutyl)-4-
propyloxetan-2-one, GK563] was found to be a highly potent inhibitor of GVIA
iPLA2 with a XI(50) value of 0.0000021 (IC50 1 nM, entry 20), while the cis
diastereomer of 9k [cis-(±)-3-(4-phenylbutyl)-4-propyloxetan-2-one, GK564] was a
ACS Paragon Plus Environment

Page 13 of 49
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
dramatically weaker inhibitor, presenting a XI(50) value of 0.007 (IC50 3.5 µM, entry
21). Both trans-9k and cis-9k were found to be weaker inhibitors of GIVA cPLA2. In
particular for inhibitor trans-9k, its XI(50) value for GIVA cPLA2 was measured to be
0.042 (IC50 22 µM) indicating 22,000 times selectivity.
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
None of the β-lactones presented any appreciable inhibition of GV sPLA2. The
percentage inhibition of GV sPLA2 did not exceed 47% (entry 1) at a high
concentration of 0.091 mole fraction.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 49
1
2
3
4
5
Table 1. In vitro potency and selectivity of β-lactones.
6
7
8
9
GVIA iPLA2
GIVA cPLA2
% Inhibitiona
GV sPLA2
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
No
% Inhibitiona
XI(50)
XI(50)
% Inhibitiona
Entry
Structure
O
1
9a
89 ± 2
91 ± 0.5
47 ± 8
O
O
O
2
3
9b
9c
59 ± 7
78 ± 4
54 ± 1
58 ± 3
40 ± 8
40 ± 6
O
O
O
9d
96 ± 1
99 ± 0
81 ± 1
90 ± 1
4
5
O
O
O
0.00019 ±
0.00004
trans-9d
25 ± 14
ACS Paragon Plus Environment

Page 15 of 49
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
O
0.0019 ±
0.0004
cis-9d
N.D.b
92 ± 1
98 ± 2
97 ± 0
77 ± 3
O
6
7
O
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
9e
O
O
O
0.00030 ±
98 ± 0
trans-9e
cis-9e
73 ± 3
89 ± 1
30 ± 3
O
8
9
0.00004
81 ± 1
96 ± 1
27 ± 10
O
O
O
9f
76 ± 4
10
O
O
O
0.00019 ±
99 ± 1
trans-9f
cis-9f
85 ± 2
95 ± 1
26 ± 3
29 ± 1
11
12
O
0.00004
O
0.00004 ±
0.00001
0.038 ±
85 ± 1
0.004
O
O
O
13
9g
93 ± 0
85 ± 0.5
29 ± 2
O
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 49
1
2
3
4
5
O
6
7
8
9
14
15
16
O
9h
9i
93 ± 0.0
89 ± 0.0
97 ± 3
85 ± 0.5
29 ± 2
26 ± 9
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
O
O
85 ± 1
71 ± 4
O
9j
O
O
O
0.00009 ±
99 ± 0
trans-9j
cis-9j
59 ± 2
72 ± 6
29 ± 5
N.D.
17
18
0.00001
O
O
0.0021 ±
95 ± 1
0.0007
ACS Paragon Plus Environment

Page 17 of 49
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
O
19
20
21
9k
100 ± 0
84 ± 0
88 ± 3
94 ± 1
O
9
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
0.0000021
O
±
0.042 ± 0.004
trans-9k
cis-9k
100 ± 0
25 ± 8
34 ± 6
O
0.0000004
O
0.007 ±
99 ± 0
0.001
O
O
0.0014 ±
99 ± 024
2a
2b
3
N.D.24
N.D.25
28 ± 124
C2F5
O
22
23
0.000124
0.0001 ±
10025
3325
C2F5
0.000025
O
O
0.0002 ±
100 ± 024
81 ± 124
37 ± 824
CF3
24
0.000024
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 49
1
2
3
4
5
O
6
7
8
9
S
N
25
O
0.0057 ±
98 ± 032
4
70 ± 232
38 ± 432
N
0.001232
O
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
a% Inhibition at 0.091 mole fraction of each inhibitor.
bN.D. signifies compounds with less than 25% inhibition (or no detectable inhibition).
ACS Paragon Plus Environment

Page 19 of 49
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
The results of in vitro inhibition clearly confirm our assumption that β-lactones inhibit
the serine-based GVIA iPLA2. However, a careful selection of the heterocyclic ring
substituents is critical for potent inhibition. β-Lactone trans-9k stands out as the most
potent inhibitor of GVIA iPLA2 ever reported in literature, outperforming the potent
fluoroketone FKGK18 (XI(50) value of 0.000224, IC50 100 nM), which has been used
successfully for in vivo studies.20
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
Binding mode and interactions of 9k diastereomers. Lactones constitute a novel
class of compounds identified as potent GVIA iPLA2 inhibitors. The binding mode of
the most active compound in the active site of the enzyme was determined in our
effort to understand its interactions with critical residues of the active site. For the
docking calculations, the previously published docked structures of GIVA cPLA2 and
GVIA iPLA2 based on our molecular dynamics simulations with two different
fluoroketone compounds in the active site were used.31,32,44 An average theoretical
score of 6.0 kcal/mol was calculated for all four diastereomers of the most potent
lactone GVIA iPLA2 inhibitor 9k (Table S3, Supporting Information). The binding
mode of trans-(S,S)-9k in the resulting optimized docked structure showed close
proximity of the carbonyl group to the oxyanion hole (Gly486/Gly487) of GVIA
iPLA2, while the aromatic chain was placed in the hydrophobic area of the active site,
interacting with residues such as Tyr541, Met544, Val548, Phe549, Tyr643, and
Leu770. The small aliphatic tail of the inhibitor was located close to Ala640 and
Pro641 (Figure 4B). The lactone inhibitor trans-(S,S)-9k exhibited lower inhibition
towards GIVA cPLA2. The binding mode in the active site of GVIA cPLA2 also
showed close proximity of the carbonyl group to the oxyanion hole (Gly197/Gly198)
(Figure 4A). The aromatic chain was also located in the hydrophobic area of GVIA
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 49
1
2
3
4
5
6
cPLA2, but its small size does not complement the suitable aromatic interactions with
the active site of GIVA cPLA2.
7
8
9
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
Figure 4. Binding mode of trans-(S,S)-9k in the active site of (A) GIVA cPLA2 (PDB
ID: 1CJY) and (B) GVIA iPLA2 (HM based on PDB ID: 1OXW).
Suppression of cytokine-induced β-cell apoptosis. Type 1 diabetes (T1D) is a
consequence of autoimmune destruction of islet β-cells. It is recognized that
eicosanoids play important roles in promoting inflammatory responses in several
diseased states, including diabetes.46 We reported that inhibition of the GVIA iPLA2
(iPLA2β) mitigates β-cell death47-52 raising the possibility that inhibitors of GVIA
iPLA2 may be beneficial in reducing β-cell death that leads to T1D incidence. To
date, several inhibitors that can inhibit GVIA iPLA2 are available, but they have
limitations.1,13-15 As we noted in introduction, recent efforts to generate more
selective and potent GVIA iPLA2 inhibitors identified reversible fluoroketone
compounds as being selective towards GVIA.23-25 One such inhibitor, designated
FKGK18, was recently described to be more selective towards iPLA2β than iPLA2γ
(GVIB iPLA2).28 Under in vitro conditions, FKGK18 inhibited insulin secretion and
β-cell apoptosis.28 Under in vivo conditions, it was devoid of cytotoxicity and
ACS Paragon Plus Environment

Page 21 of 49
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
effective in reducing T1D incidence.20 Thus, novel GVIA iPLA2 inhibitors are very
attractive as candidates for preventing β-cell apoptosis and as potential new agents for
preventing T1D development.
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
Here, we assessed the ability of trans-9k in reducing β-cell apoptosis by treating INS-
1 cells with pro-inflammatory cytokines (IL-1β + IFNγ) in the absence and presence
of trans-9k. As expected, cytokine exposure resulted in a dramatic increase in β-cell
apoptosis (Fig. 5). At 0.10 µM and 1.0 µM, trans-9k alone had no effect, but it
promoted a slight but modest rise in cell death. Co-treatment of the cells with
cytokines and trans-9k produced a concentration-dependent inhibition of β-cell
apoptosis; with 0.10 µM showing minimal and non-significant effect, but significant
decreases evident with 1.0 µM (28%) and 10.0 µM (41%). In comparison, these
results are similar to those seen with S-BEL,47 a selective inhibitor of iPLA2β.
However, in contrast to S-BEL, continuous exposure of trans-9k to cells was not
cytotoxic at 0.10 or 1.0 µM, and induced only a modest rise in percent cell death
(DMSO, 7.06 ± 0.36 vs. trans-9k, 9.51 ± 0.78, p = 0.012) at 10 µM. These findings
suggest that trans-9k is another candidate inhibitor of GVIA iPLA2 suitable for
further studies and raise the possibility that its use in vivo may be beneficial in
reducing β-cell death leading to T1D.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 49
1
2
3
4
5
6
7
8
9
35
30
25
20
15
10
5
9%
28%
b
41%
b
(3)
(12)
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
b,c,d
(6)
e,f,g,h
(3)
a
(3)
(3)
(6)
(12)
0
D
0.1
1
10
D
0.1
1
10
- CTK
+ CTK
[GK563] (µM)
aSig diff from other - CTK groups, p = 0.0065
bSig diff from DMSO, p < 10-10
cSig diff from + CTK, p = 0.00045
dSig diff from + CTK + 0.1 GK563, p = 0.0065
eSig diff from DMSO, p < 10-6
fSig diff from + CTK, p = 0.00056
gSig diff from + CTK + 1.0 GK563, p = 0.0087
hSig diff from + CTK + 0.1 GK563, p = 0.0093
Figure 5. INS-1 cell apoptosis ± trans-9k. INS-1 cells were treated with vehicle
(DMSO) or cytokines (CTK, 100 U/mL IL-1 +  U/mL IFN) for 16 h in the
absence or presence of trans-9k (0.10-10 µM). The cells were then processed for
TUNEL analyses and the means ± SEM (n=3-12) of percent apoptotic cells relative to
total number of cells are presented.
ACS Paragon Plus Environment

Page 23 of 49
Journal of Medicinal Chemistry
1
2
3
4
CONCLUSION
5
6
7
8
9
Herein, we describe a novel class of GVIA iPLA2 inhibitors based on the β-lactone
ring. This reactive functionality in combination with a four-carbon chain carrying a
phenyl group at position-3, and a linear propyl group at position-4 of the lactone ring
produced the best candidate inhibitor of GVIA iPLA2. Inhibitor trans-9k with a XI(50)
value of 0.0000021 (IC50 1 nM) is the most potent inhibitor of GVIA iPLA2 ever
reported in the literature, being a hundred times more potent than the fluoroketone
inhibitor FKGK18. In addition, it is selective for GVIA iPLA2, because it is 22,000
more potent for GVIA iPLA2 than for GIVA cPLA2. It reduces β-cell apoptosis
induced by pro-inflammatory cytokines (IL-1β + IFNγ) in a concentration-dependent
manner, suggesting that its use in vivo may be beneficial in reducing β-cell death
leading to type 1 diabetes. This novel, highly potent and selective GVIA iPLA2
inhibitor may be an excellent tool for the study of the role of the enzyme in cells and
in animals and might help in developing novel medicinal agents.
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
EXPERIMENTAL SECTION
General. Chromatographic purification of products was accomplished using Merck
Silica Gel 60 (70-230 or 230-400 mesh). Thin-layer chromatography (TLC) was
performed on Silica Gel 60 F254 aluminum plates. TLC spots were visualized with
UV light and/or phosphomolybdic acid in EtOH. Melting points were determined
using a Büchi 530 apparatus and were uncorrected. 1H and 13C NMR spectra were
recorded on a Varian Mercury (200 MHz and 50 MHz, respectively) and a Bruker
Avance III (600 MHz and 150 MHz, respectively) in CDCl3. Chemical shifts are
given in ppm, and coupling constants (J) in Hz. Peak multiplicities are described as
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 49
1
2
3
4
5
6
7
8
9
follows: s, singlet, d, doublet, t, triplet and m, multiplet. Low resolution electron spray
ionization (ESI) mass spectra were recorded on a Finnigan, Surveyor MSQ Plus
spectrometer, while HRMS spectra were recorded on a Bruker Maxis Impact QTOF
Spectrometer. Dichloromethane was dried by standard procedures and stored over
molecular sieves. All other solvents and chemicals were reagent grade and used
without further purification. The purity of all compounds subjected to biological tests
was determined by analytical HPLC, and was found to be ≥95%. HPLC analyses were
carried out on a Shimadzu LC-2010AHT system and an ODS Hypersil (250 x 4.6
mm, 5 μm) analytical column, using H2O/acetonitrile 20/80 v/v, at a flow rate of 1.0
mL/min. HPLC analyses of trans-9k and cis-9k were carried out on a Agilent 1100
system and a Daicel Chiralcel OD-H (250  4.6 mm, 5 m) using hexane/i-PrOH 95/5
v/v, at a flow rate of 1.0 mL/min.
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
Carboxylic acids 5a and 5f were commercially available. Carboxylic acids 5b,53 5c,25
5d25 and 5e53 have been described elsewhere and their analytical data are in
accordance with literature. β-Hydroxy acids 8a-k and β-lactones 9a-k were obtained
as mixtures of diastereomers. The diastereomeric ratio (d.r.) of the mixtures was
determined by 1H NMR spectroscopy.
General method for the synthesis of β-hydroxy acids (8a-k). To a stirred solution
ο
of diisopropylamine (3 mmol) in anhydrous THF (2 mL), under argon at 0 C, a
solution of n-BuLi 1.6 Μ in hexane (3 mmol, 1.9 mL) was slowly added via syringe
ο
and the solution of LDA was stirred at 0 C for 10 minutes. The carboxylic acid 7a-f
(1 mmol) in anhydrous THF (6 mL) was then added and the solution was stirred at 0
οC for 1 h. Then, aldehyde (1.3 mmol) in anhydrous THF (2 mL) was added and the
ο
solution was stirred at 0 C for 1 h and at room temperature overnight. The solvent
ACS Paragon Plus Environment

Page 25 of 49
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
was removed under reduced pressure. The reaction mixture was acidified with HCl
1Ν to pH 2 and extracted with Et2O (3x 20 mL). The organic layers were combined,
washed with brine and dried. The solvent was removed and the product was purified
by column chromatography eluting with a gradient of CHCl3/MeOH 95/5 to 9/1.
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
3-Hydroxy-2-(3-phenylpropyl)nonanoic acid (8a). Mixture of diastereomers (d.r.
1
6:4 anti:syn). Yield 51%; Oil; H NMR (200 MHz, CDCl3): δ 7.37-7.09 (m, 5H),
3.92-3.77 (m, 0.4H), 3.76-3.61 (m, 0.6H), 2.73-2.55 (m, 2H), 2.54-2.37 (m, 1H), 1.87-
1.56 (m, 4H), 1.55-1.13 (m, 10H), 0.90 (t, J = 6.6 Hz, 3H); 13C NMR (50 MHz,
CDCl3): δ 180.6, 141.8, 128.3, 128.2, 125.8, 72.2, 72.1, 50.9, 50.6, 35.7, 35.6, 35.2,
33.9, 31.7, 29.1, 29.0, 28.9, 28.7, 25.6, 22.6, 14.1; MS (ESI) m/z (%): 291.3 [(M-H)-,
100]; HRMS: 315.1947 (M+Na)+, (315.1931).
3-Hydroxy-2-(3-(naphthalen-2-yl)propyl)nonanoic acid (8b). Mixture of
1
diastereomers (d.r. 7:3 anti:syn). Yield 35%; Oil; H NMR (200 MHz, CDCl3): δ
7.86-7.65 (m, 3H), 7.55 (s, 1H), 7.48-7.21 (m, 3H), 3.93-3.75 (m, 0.3H), 3.74-3.57
(m, 0.7H), 2.85-2.65 (m, 2H), 2.56-2.29 (m, 1H), 1.91-1.56 (m, 4H), 1.56-1.03 (m,
13
10H), 0.84 (t, J = 6.4 Hz, 3H); C NMR (50 MHz, CDCl3): δ 180.7, 139.3, 133.5,
131.9, 127.8, 127.5, 127.4, 127.2, 126.3, 125.8, 125.1, 72.3, 72.2, 51.1, 50.9, 36.3,
35.8, 35.4, 34.0, 31.8, 29.2, 29.0, 28.9, 28.7, 25.6, 22.6, 14.1; MS (ESI) m/z (%):
341.3 [(M-H)-, 100]; HRMS: 365.2090 (M+Na)+, (365.2087).
2-(3-([1,1'-Biphenyl]-4-yl)propyl)-3-hydroxynonanoic acid (8c). Mixture of
diastereomers (d.r. 7:3 anti:syn). Yield 52%; Solid; mp 40-42 οC; 1H NMR (200 MHz,
CDCl3): δ 7.60-7.21 (m, 9H), 3.95-3.83 (m, 0.3H), 3.79-3.63 (m, 0.7H), 2.75-2.60 (m,
13
3H), 1.88-1.60 (m, 4H), 1.53-1.13 (m, 10H), 0.88 (t, J = 6.0 Hz, 3H); C NMR (50
MHz, CDCl3): δ 180.6, 141.0, 138.8, 128.8, 128.7, 127.0, 126.9, 72.3, 72.2, 51.1,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 49
1
2
3
4
5
6
50.9, 35.5, 35.3, 33.9, 33.9, 31.8, 29.5, 29.1, 29.0, 28.7, 25.6, 22.6, 14.1; MS (ESI)
m/z (%): 367.2 [(M-H)-, 100].
7
8
9
3-Hydroxy-2-(3-phenylpropyl)hexanoic acid (8d). Mixture of diastereomers (d.r.
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
7:3 anti:syn). Yield 28%; Oil; H NMR (200 MHz, CDCl3): δ 7.37-7.07 (m, 5H),
3.94-3.80 (m, 0.3H), 3.79-3.67 (m, 0.7H), 2.72-2.57 (m, 2H), 2.57-2.42 (m, 1H), 1.89-
1.56 (m, 4H), 1.56-1.30 (m, 4H), 0.93 (t, J = 6.2 Hz, 3H); 13C NMR (50 MHz,
CDCl3): δ 180.6, 141.9, 128.3, 128.2, 125.8, 71.9, 71.8, 50.8, 50.7, 37.4, 36.1, 35.8,
35.6, 29.5, 29.0, 28.9, 26.3, 19.1, 18.8, 13.9; MS (ESI) m/z (%): 249.3 [(M-H)-, 100];
HRMS: 273.1475 (M+Na)+, (273.1461).
3-Hydroxy-2-(3-(naphthalen-2-yl)propyl)hexanoic acid (8e). Mixture of
diastereomers (d.r. 7:3 anti:syn). Yield 20%; Solid; mp 38-40 οC; 1H NMR (200 MHz,
CDCl3): δ 7.90-7.67 (m, 3H), 7.60 (s, 1H), 7.50-7.19 (m, 3H), 3.94-3.80 (m, 0.3H),
3.80-3.62 (m, 0.7H), 2.78 (t, J = 6.0 Hz, 2H), 2.58-2.38 (m, 1H), 1.94-1.57 (m, 4H),
13
1.57-1.15 (m, 4H), 0.89 (t, J = 6.2 Hz, 3H); C NMR (50 MHz, CDCl3): δ 180.4,
139.3, 133.5, 131.9, 127.9, 127.5, 127.4, 127.1, 126.4, 125.9, 125.1, 71.9, 71.8, 51.0,
50.7, 37.4, 36.0, 35.9, 35.7, 29.3, 29.0, 28.9, 26.2, 19.1, 18.8, 13.9; MS (ESI) m/z
(%): 299.3 [(M-H)-, 100].
3-Hydroxy-2-(3-(4-methoxyphenyl)propyl)hexanoic acid (8f). Mixture of
diastereomers (d.r. 6:4 anti:syn). Yield 46%; Oil; 1H NMR (200 MHz, CDCl3): δ 7.06
(d, J = 8.0 Hz, 2Η), 6.79 (d, J = 8.0 Hz, 2Η), 3.95-3.80 (m, 0.4H), 3.79 (s, 3H), 3.78-
3.59 (m, 0.6H), 2.56 (t, J = 6.8 Hz, 2H), 2.52-2.43 (m, 1H), 1.86-1.53 (m, 4H), 1.53-
1.19 (m, 4H), 1.02-0.80 (m, 3H); 13C NMR (50 MHz, CDCl3): δ 180.3, 157.6, 134.0,
129.2, 113.7, 71.9. 71.8, 55.2, 50.9, 50.7, 37.4, 36.0, 34.8, 34.7, 29.7, 29.3, 28.9, 26.2,
19.1, 18.8, 13.9; MS (ESI) m/z (%): 279.2 [(M-H)-, 100].
ACS Paragon Plus Environment

Page 27 of 49
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
3-Hydroxy-4-methyl-2-(3-phenylpropyl)pentanoic acid (8g). Mixture of
diastereomers (d.r. 7:3 anti:syn). Yield 28%; Solid; mp 85-90 οC; 1H NMR (200 MHz,
CDCl3): δ 7.46-7.03 (m, 5H), 3.64-3.45 (m, 0.3H), 3.45-3.25 (m, 0.7H), 2.72-2.44 (m,
13
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
2H and 0.7Η), 2.43-2.27 (m, 0.3Η), 1.89-1.40 (m, 5H), 1.05-0.72 (m, 6H); C NMR
(50 MHz, CDCl3): δ 180.9, 180.7, 141.8, 128.3, 128.2, 125.8, 77.2, 48.3, 48.0, 35.8,
35.6, 31.8, 30.9, 29.5, 29.3, 29.0, 26.1, 19.6, 19.5, 17.5, 17.4; MS (ESI) m/z (%):
+
268.2 [(M+NH4 ), 100].
3-Hydroxy-2-(3-(naphthalen-2-yl)propyl)pentanoic acid (8h). Mixture of
ο
1
diastereomers (d.r. 7:3 anti:syn). Yield 30%; Solid; mp 115-120 C; H NMR (200
MHz, CDCl3): δ 7.83-7.70 (m, 3H), 7.60 (s, 1H), 7.48-7.28 (m, 3H), 3.78-3.68 (m,
0.3H), 3.60-3.53 (m, 0.7H), 2.75 (t, J = 6.0 Hz, 2H), 2.57-2.37 (m, 1H), 1.88-1.37 (m,
6H), 0.93 (t, J = 7.0 Hz, 3H); 13C NMR (50 MHz, CDCl3): δ 179.1, 139.5, 133.4,
131.8, 127.7, 127.4, 127.2, 127.1, 126.2, 125.7, 124.9, 73.7, 73.6, 51.1, 50.6, 35.8,
+
35.6, 29.3, 28.9, 28.8, 27.8, 10.3, 9.8; MS (ESI) m/z (%): 304.2 [(M+ΝH4 ), 100].
2-(1-Hydroxyethyl)-5-(naphthalen-2-yl)pentanoic
acid
(8i).
Mixture
of
ο
1
diastereomers (d.r. 6:4 anti:syn). Yield 33%; Solid; mp 125-130 C; H NMR (200
MHz, CDCl3): δ 7.86-7.70 (m, 3H), 7.60 (s, 1H), 7.50-7.25 (m, 3H), 4.12-3.99 (m,
0.4H), 3.99-3.86 (m, 0.6H), 2.79 (t, J = 6.6 Hz, 2H), 2.58-2.36 (m, 1H), 1.93-1.54 (m,
13
4H), 1.29-1.16 (m, 3H); C NMR (50 MHz, CDCl3): δ 180.0, 179.8, 139.2, 133.5,
132.0, 127.9, 127.6, 127.4, 127.1, 126.3, 125.9, 125.1, 68.3, 68.1, 52.6, 51.7, 35.9,
+
35.7, 29.3, 28.8, 26.7, 21.4, 20.0; MS (ESI) m/z (%): 290.4 [(M+ΝH4 ), 100].
3-Hydroxy-2-(4-(naphthalen-2-yl)butyl)hexanoic
acid
(8j).
Mixture
of
diastereomers (d.r. 6:4 anti:syn). Yield 20%; Solid; mp 36-38 οC; 1H NMR (200 MHz,
CDCl3): δ 7.83-7.70 (m, 3H), 7.60 (s, 1H), 7.48-7.28 (m, 3H), 3.93-3.79 (m, 0.4H),
3.79-3.65 (m, 0.6H), 2.77 (t, J = 7.2 Hz, 2H), 2.53-2.39 (m, 1H), 1.85-1.59 (m, 4H),
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 49
1
2
3
4
13
1.58-1.31 (m, 6H), 0.91 (t, J = 6.6 Hz, 3H); C NMR (50 MHz, CDCl3): δ 179.8,
5
6
7
8
139.8, 133.5, 131.9, 127.8, 127.6, 127.4, 127.3, 126.3, 125.8, 125.0, 71.8, 71.7, 50.7,
50.6, 37.6, 36.1, 35.8, 31.3, 31.1, 29.3, 27.4, 27.0, 26.5, 19.1, 18.9, 13.9; MS (ESI)
9
+
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
m/z (%): 332.2 [(M+ΝH4 ), 100].
3-Hydroxy-2-(4-phenylbutyl)hexanoic acid (8k). Mixture of diastereomers (d.r. 7:3
anti:syn). Yield 68%; Solid; mp 38-40 οC; H NMR (200 MHz, CDCl3): δ 7.37-7.07
1
(m, 5H), 3.93-3.80 (m, 0.3H), 3.80-3.64 (m, 0.7H), 2.62 (t, J = 7.6 Hz, 2H), 2.53-2.37
(m, 1H), 1.85-1.55 (m, 4H), 1.55-1.27 (m, 6H), 0.94 (t, J = 7.0 Hz, 3H); 13C NMR (50
MHz, CDCl3): δ 180.5, 142.3, 128.3, 128.2, 125.7, 71.9, 71.8, 50.9, 50.7, 37.5, 36.1,
35.6, 31.4, 31.3, 29.2, 27.4, 26.9, 26.4, 19.1, 18.9, 13.9; MS (ESI) m/z (%): 282.2
+
[(M+NH4 ), 100].
General method for the cyclization of β-hydroxy acids to β-lactones (9a-k). To a
stirred solution of β-hydroxy acid 8a-k (1 mmol) in anhydrous pyridine (2 mL), under
argon at 0 οC, p-toluenesulfonyl chloride (2 mmol) in anhydrous pyridine (1 mL) was
added slowly via syringe. The solution was stirred at 0 οC for 1 h and kept at 4 οC for
3 days. Then, Et2O was added and the organic layer was washed with 10% Na2CO3,
HCl 1Ν to pH 2 and brine. The organic layer was dried and the solvent was removed
in vacuo. The product was purified by column chromatography eluting with a gradient
of petroleum ether (bp 40-60 °C) /AcOEt 95/5 to 9/1.
4-Hexyl-3-(3-phenylpropyl)oxetan-2-one (9a). Mixture of diastereomers (d.r. 8:2
1
trans:cis).Yield 75%; Oil; H NMR (200 MHz, CDCl3): δ 7.35-7.11 (m, 5H), 4.58-
4.44 (m, 0.2H), 4.27-4.13 (m, 0.8H), 3.68-3.53 (m, 0.2H), 3.26-3.11 (m, 0.8H), 2.65
(t, J = 6.8 Hz, 2H), 1.93-1.66 (m, 6H), 1.40-1.22 (m, 8H), 0.89 (t, J = 6.4 Hz, 3H); 13C
NMR (50 MHz, CDCl3): δ 172.1, 171.4, 141.3, 128.4, 128.3, 126.0, 78.0, 75.6, 55.9,
ACS Paragon Plus Environment

Page 29 of 49
Journal of Medicinal Chemistry
1
2
3
4
52.4, 35.4, 34.4, 31.5, 30.1, 29.1, 28.8, 28.6, 27.3, 25.4, 24.9, 23.3, 22.5, 14.0; MS
5
6
+
(ESI) m/z (%): 292.3 [(M+ΝH4 ), 100]; HRMS: 297.1842 (M+Na)+, (297.1825).
7
8
9
4-Hexyl-3-(3-(naphthalen-2-yl)propyl)oxetan-2-one (9b). Mixture of diastereomers
(d.r. 9:1 trans:cis).Yield 44%; Oil; 1H NMR (200 MHz, CDCl3): δ 7.87-7.72 (m, 3H),
7.60 (s, 1H), 7.52-7.23 (m, 3H), 4.60-4.41 (m, 0.1H), 4.27-4.10 (m, 0.9H), 3.69-3.55
(m, 0.1H), 3.27-3.10 (m, 0.9H), 2.82 (t, J = 7.0 Hz, 2H), 1.97-1.63 (m, 6H), 1.45-1.16
(m, 8H), 0.87 (t, J = 6.0 Hz, 3H); 13C NMR (50 MHz, CDCl3): δ 171.4, 138.8, 133.5,
132.0, 128.1, 127.6, 127.4, 127.0, 126.5, 126.0, 125.3, 78.0, 55.9, 35.6, 34.4, 31.6,
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
+
28.9, 28.5, 27.4, 25.0, 22.5, 14.0; MS (ESI) m/z (%): 342.2 [(M+ΝH4 ), 100]; HRMS:
347.1986 (M+Na)+, (347.1982).
3-(3-([1,1'-Biphenyl]-4-yl)propyl)-4-hexyloxetan-2-one
(9c).
Mixture
of
diastereomers (d.r. 7:3 trans:cis).Yield 57%; Solid; mp 35-38 οC; 1H NMR (200 MHz,
CDCl3): δ 7.64-7.16 (m, 9H), 4.61-4.44 (m, 0.3H), 4.27-4.15 (m, 0.7H), 3.71-3.55 (m,
0.3H), 3.28-3.12 (m, 0.7H), 2.70 (t, J = 6.8 Hz, 2H) 2.02-1.51 (m, 6H), 1.51-1.14 (m,
8H), 0.88 (t, J = 6.8 Hz, 3H); 13C NMR (50 MHz, CDCl3): δ 172.1, 171.4, 140.9,
140.4, 139.0, 128.8, 128.7, 127.2, 127.1, 127.0, 78.0, 75.6, 55.9, 52.5, 35.1, 34.4,
31.6, 30.1, 29.1, 28.9, 28.6, 27.4, 25.5, 25.0, 23.6, 22.5, 14.0; MS (ESI) m/z (%):
+
368.3 [(M+ΝH4 ), 100]; HRMS: 373.2141 (M+Na)+, (373.2138).
3-(3-Phenylpropyl)-4-propyloxetan-2-one (9d). Mixture of diastereomers (d.r. 9:1
1
trans:cis). Yield 68%; Oil; H NMR (200 MHz, CDCl3): δ 7.40-7.10 (m, 5H), 4.60-
4.45 (m, 0.1H), 4.28-4.13 (m, 0.9H), 3.68-3.54 (m, 0.1H), 3.26-3.10 (m, 0.9H), 2.66
(t, J = 7.0 Hz, 2H), 1.96-1.56 (m, 6H), 1.56-1.30 (m, 2H), 0.97 (t, J = 7.2 Hz, 3H); 13C
NMR (50 MHz, CDCl3) δ 172.1, 171.4, 141.3, 128.4, 128.3, 126.0, 77.9, 75.4, 56.0,
ACS Paragon Plus Environment