Lipoic acid analogs with enhanced pharmacological activity

Article abstract of DOI:10.1016/j.bmc.2013.10.057

Lipoic acid (1,2-dithiolane-3-pentanoic acid) is a pharmacophore with unique antioxidant and cytoprotective properties. We synthesized a library based upon the condensation of natural and unnatural amino acids with the carboxylic acid moiety of lipoic acid. SAR studies were conducted using a cardiac ischemia-reperfusion animal model. Cytoprotective efficacy was associated with the R-enantiomer of the dithiolane. Potency of library compounds was dictated by the acidic strength of the adduct. α-N-[(R)-1,2-dithiolane-3-pentanoyl]- l-glutamyl-l-alanine, designated CMX-2043, was chosen for further pharmacologic evaluation.

Full text of DOI:10.1016/j.bmc.2013.10.057

Bioorganic & Medicinal Chemistry 22 (2014) 505–512
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Lipoic acid analogs with enhanced pharmacological activity q
⇑
Steven A. Kates , Ralph A. Casale, Alexander Baguisi, Reinier Beeuwkes III
Ischemix, LLC, 63 Great Road, Maynard, MA 01759, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Lipoic acid (1,2-dithiolane-3-pentanoic acid) is a pharmacophore with unique antioxidant and cytopro-
tective properties. We synthesized a library based upon the condensation of natural and unnatural amino
acids with the carboxylic acid moiety of lipoic acid. SAR studies were conducted using a cardiac ischemia-
reperfusion animal model. Cytoprotective efficacy was associated with the R-enantiomer of the
Received 19 September 2013
Revised 23 October 2013
Accepted 31 October 2013
Available online 14 November 2013
dithiolane. Potency of library compounds was dictated by the acidic strength of the adduct.
a
-N-[(R)-
1,2-dithiolane-3-pentanoyl]-
pharmacologic evaluation.
L-glutamyl-L-alanine, designated CMX-2043, was chosen for further
Keywords:
Lipoic acid
Ischemia-reperfusion
SAR studies
Ó 2013 The Authors. Published by Elsevier Ltd. All rights reserved.
Polymerization
Amino acids
1. Introduction
liver,14–17 kidney,18–21 heart,6,22–26 and brain.27–29 It is available
in Germany in oral and parenteral formulations.30,31
Pharmacological protection against cellular damage and apop-
totic death is a serious clinical need.1 Although a large number of
candidate molecules have been evaluated for this property in ani-
mal studies and human trials,2,3 none has yet attained significant
The purpose of the present study was to prepare and screen a
selection of LA analogs for cytoprotective activity. LA is an attrac-
tive candidate for modification based on both its chemical and
biological properties. Structure activity relationships were investi-
gated by assessing pharmacological efficacy of the analogs in a
standard ischemia-reperfusion injury animal model.
clinical usefulness. One such molecule is a-lipoic acid (LA), known
primarily as a naturally occurring cellular antioxidant. Its activity
in both aqueous and lipophilic, and intracellular or extracellular
environments, both in oxidized and reduced forms, suggests
potential opportunities for broad pharmacological action.4
Lipoic acid is naturally present in both prokaryotic and eukary-
2. Results and discussion
The molecular design adopted for these studies was an out-
growth of earlier work showing pharmacological activity of the
fatty acid DHA (docosahexanoic acid) linked to a dodecapeptide.32
This earlier molecule included an amide bond to an initial Asp-Gly
sequence. In the present work the fatty acid was lipoic acid and the
peptide sequence was shortened.33,34 Cardiac ischemia-reperfusion
injury was chosen to define structure–activity relationships.
otic cells. It is recognized as a complement of a-keto acid dehydro-
genase complexes of mitochondria and thus fundamental
mammalian metabolism.5 Due to its water and lipid soluble prop-
erties, exogenously administered LA is distributed in cellular mem-
branes, cytosol and extracellular spaces.6 Although considered as
an antioxidant,7–10 LA has regulatory action on signal transduction
processes involved in tissue damage and protection11 including
activation of AKT phosphorylation.6 Following an oral dose it is ex-
creted largely by the kidneys.12 It is metabolized extensively in the
liver, including a high first-pass effect.13 LA has been shown to be
efficacious for prevention or treatment of ischemic injury to
2.1. Screening assay
In a multi-factorial condition such as ischemia-reperfusion in-
jury (IRI) with complex contributory mechanisms, approaches
based upon traditional receptor-based drug design have not suc-
ceeded clinically.35 It has been recently recommended that the pre-
clinical testing of potential cardioprotective agents should employ
clinical trial-like approaches.36 Since the goal of our development
program was to develop an agent effective for IRI, an unconven-
tional screening approach was selected based upon efficacy in an
animal model of cardiac injury. Although labor intensive, this
q
This is an open-access article distributed under the terms of the Creative
Commons Attribution-NonCommercial-No Derivative Works License, which per-
mits non-commercial use, distribution, and reproduction in any medium, provided
the original author and source are credited.
⇑
Corresponding author. Tel.: +1 978 897 5139.
E-mail address: skates@ischemix.com (S.A. Kates).
0968-0896/$ - see front matter Ó 2013 The Authors. Published by Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmc.2013.10.057

506
S. A. Kates et al. / Bioorg. Med. Chem. 22 (2014) 505–512
approach simultaneously addressed issues related to ADME (‘dru-
gability’) and toxicity within the single efficacy measure. Efficacy
was measured in a widely used rat model of myocardial ische-
mia-reperfusion injury (IRI).37,38 Male Sprague Dawley rats (6–
12/group dependent on the experiment) were used in accordance
with the Guide for the Care and Use of Laboratory Animals (National
Research Council, The National Academies Press Eighth Edition,
2011). Animal studies were approved by the Ischemix Animal Care
and Use Committee. In this model, following anesthesia, thoracot-
omy, and vertical pericardotomy, the left circumflex (LCX) coro-
nary artery was ligated for 30 min to induce ischemia. During the
ischemic period, fluorescent microspheres were injected through
the apex of the heart into the left ventricular cavity to delineate
the area at risk (AR) by the absence of fluorescence. Following
the ligation period, the heart was allowed to reperfuse and the ani-
mals to recover. After 24 h, the animals were sacrificed and the
hearts excised and sliced transversely into sections 2 mm thick.
The slices were stained to distinguish live tissue from the un-
stained myocardial infarct (MI) area. Test animals were adminis-
tered a single injection of test agent into the left ventricular
cavity prior to LCX ligation. The AR and MI areas were quantitated
to obtain the MI/AR ratio. More efficacious compounds have smal-
ler MI/AR ratios.
The N-terminal amino function of the peptide H2N-LGlu-LAla-OH
(5) was attached to the carboxylic acid of either the R- or S-stereo-
isomers, as well as the R/S racemic mixture, to provide 4a, 4b, and
4c, respectively. The optical activity of 4a, 4b, and 4c is +36.1,
À69.5, and À24.2, respectively (Table 2). A single IC bolus of 4a
at doses of 1 and 2 mg/kg provided an MI/AR ratio of 0.33 and
0.29, respectively, correlating to a MI/AR reduction of ꢀ21%. A sin-
gle IC bolus of 4b at doses of 1 and 2 mg/kg provided an MI/AR ra-
tio of 0.40 correlating to a MI/AR reduction of ꢀ10%. A single IC
bolus of 4c at a dose of 2 mg/kg provided an MI/AR ratio of 0.33
correlating to a MI/AR reduction of ꢀ21%. Finally, a dose of a mix-
ture 4a (1 mg/kg) and 4b (1 mg/kg) provided an MI/AR ratio of 0.33
correlating to an MI/AR reduction of 21%. These results clearly
demonstrate that the R-enantiomer of LA (4a) provides more effi-
cacy than its corresponding S-isomer, and the former was chosen
as the lead candidate for further investigation.
O
S
S
H
N
OH
N
H
O
O
CO2H
4
2.3. Chemistry
2.2. Peptide and stereoisomer selection
Since it is known that lipoic acid can polymerize, this possibility
had to be taken into consideration in synthesis of the planned ana-
logs. Photo instability of lipoic acid was described by Barltrop et al.
in 195443 and Wagner et al. in 1956.44 Its degradation in the pres-
ence of light was characterized by a physical change in the com-
pound and a shift in the ultraviolet spectrum. The instability was
proposed to be due to the polymerization of lipoic acid via opening
of the strained dithiolane ring followed by formation of intermo-
lecular disulfide linkages.45
The effects of solvent on lipoic acid photolysis and polymeriza-
tion have also been evaluated.46 Brown and Edwards found that
solvents with readily extractable hydrogen inhibited light induced
polymerization of lipoic acid and reported that this tendency was
minimized in the presence of 2-propanol. This phenomenon is
consistent of a mechanism of lipoic acid polymerization via a pho-
tolytic opening of the dithiolane ring structure resulting in a di-
radical, followed by propagation through intermolecular disulfide
bond formation.
In initial studies, the efficacy of peptide sequences capped with
an acetyl group was evaluated. The Asp-Gly sequence in the
dodecapeptide was replaced with the more chemically stable
dipeptide Glu-Ala (EA) to avoid known aspartimide formation.39
These studies showed that the di- and tetrapeptide Glu-Ala motifs
had greater efficacy than that of a heptapeptide (Table 1). Based on
this finding, the shorter dipeptide sequence Glu-Ala was selected
for attachment to lipoic acid to undergo further evaluation.
LA occurs naturally as the R-enantiomer and acts as a coenzyme
in many reactions. It constitutes a growth factor for a number of
bacteria and protozoa. In studies with synthetic lipoic acid the R-
isomer has been described as the more active form as well as the
enantiomer that is more readily absorbed following oral adminis-
tration (38% for the R-form vs 28% for the S-form in humans).40,41
The R-enantiomer has primarily an anti-inflammatory activity
while the S-enantiomer has an antinociceptive (analgesic) activity.
In addition to these properties, both R- and S-enantiomers possess
cytoprotective activity. Wolz and Krieglstein describe that the S-
enantiomer of lipoic acid had a stronger neuroprotective effect
than the R-enantiomer in a mouse model of cerebral ischemia.42
In view of the demonstrated variability in effects of lipoic acid ste-
reospecificity, the two stereoisomers of Lip-EA-OH were tested to
determine their relative efficacies in the rat IRI model (Table 2).
Efforts to produce lipoic acid free from contaminating polymer
have focused on crystallization solvents47,48 and salt formation.49–
51
Beisswenger et al. crystallized lipoic acid at low temperature
from mixtures of solvents selected from pentane, cyclohexane,
ethyl acetate and ethers.48 Klatt et al. crystallized pharmaceutical
grade lipoic acid at low temperature from organic solvents with
a dielectric constant between 1.95 and 2.4, such as hexane or hep-
tane.47 Hettche, Rischer, and Sarlikiotis prepared sodium and trom-
ethane (tris) salts and drug product formulations containing mixed
and single isomers of lipoic acid.49 Ames stabilized lipoic acid via
the production of a nicotinamide salt.50
Majeed and Nagabhushanam prepared an adduct of lipoic acid
consisting of the N’,N’-(dimethylamino)ethylamino amide.51 This
compound was crystallized as a stable nonhydroscopic fumarate
or maleate salt, although propensity toward polymerization was
not directly addressed. The potential of other lipoic acid derivatives
to polymerize has not been generally addressed in the synthetic
literature.33
Table 1
MI/AR reduction of lead sequence
Entry
Peptide sequence
MI/AR reduction (%)
1
2
3
Ac-EAEAEGA-OH
Ac-EAEA-OH
Ac-EA-OH
17 (p < 0.01)
23 (p < 0.05)
23 (p < 0.05)
Table 2
Stereoisomers of Lip-Glu-Ala-OH
Entry
Compound name
(R)-Lip- Glu- Ala-OH
(S)-Lip- Glu- Ala-OH
(R/S)-Lip- Glu- Ala-OH
Optical rotation (EtOH, 25 °C, 10 mg/mL)
To prepare 4, a-lipoic acid was added to a solution of N,N-diiso-
4a
4b
4c
L
L
+36.1
À69.5
À24.2
propylethylamine (DIPEA) in acetone and treated with N,N’-disucc-
inimidyl carbonate (DSC) (Scheme 1). After CO2 evolution, the
resulting succinimidyl ester solution 5a was cooled and treated
L
L
L
L

S. A. Kates et al. / Bioorg. Med. Chem. 22 (2014) 505–512
507
O
O
O
ii.
OH
N
i.
O
S
S
O
5
S
S
5a
CO2H
O
H
N
CO2H
NH
O
S
S
4
Scheme 1. Reagents and conditions: (i) DSC, DIPEA/acetone. (ii) Glutamyl-alanine, NaOH/acetone/water.
rapidly with an aqueous NaOH solution of glutamyl-alanine. After
the reaction was complete, workup involved removal of acetone
under vacuum, addition of water and ethyl acetate, and acidificat-
ion with aqueous HCl. For small scale reactions (<1 g), purification
of 4 was accomplished via preparative HPLC. For initial large-scale
preparations (>100 g), the final step in the synthetic process was
crystallization from ethyl acetate and n-butanol.
lyzed by HPLC. Analysis of the HPLC chromatogram indicated the
disappearance of the broad, late eluting impurity peak and the
peak corresponding to 4 was substantially diminished. Analysis
of the HPLC chromatogram also indicated the appearance of a peak
corresponding to the disulfhydryl analog of 4 containing the dihy-
drolipoyl moiety. DTT treatment reduced both the dithiolane in 4
and the disulfide bridges of the polymeric material to a single
product corresponding to the dithiol analog of 4. Therefore, the
impurity observed in 4 was determined to be the intermoleclular
disulfide polymer of the lipoyl function in 4.
2.4. Polymerization
During the initial large-scale preparations of 4c, and subse-
quently 4a, analysis of the isolated product via HPLC indicated a
peak corresponding to 4 and a broad later eluting peak (Fig. 1).
The late eluting impurity was observed upon isolation of the solid
product. The mother liquor from the crystallization was analyzed
by HPLC and did not contain the impurity. Therefore, the impurity
was formed during crystallization or precipitation of 4. The impu-
rity was detected and characterized as intermolecular linked disul-
fide polymers.
To avoid the formation of the intermoleclular disulfide polymer
during large scale preparation, a secondary alcohol such as 2-pro-
panol has been reported to inhibit the polymerization of lipoic
acid.46 Crystallization conditions were changed to isopropyl ace-
tate–2-propanol from the original solvent mixture of ethyl ace-
tate–n-butanol. In addition, steps were taken to avoid reducing
the solvent mixture to a low volume and inducing crystallization.
2.5. Chiral purity
A size exclusion chromatographic (SEC) HPLC method was
developed to detect potential polymer contamination in 4 using
a Tosoh Bioscience TSK guard column PWXL (6 mm  4 cm) with
a Tosoh Bioscience TSK-Gel PW2500XL (7.8 mm  30 cm) column.
This method was able to separate analytes based on molecular
weight and it was determined that the later eluting impurity in
Figure 1 was a higher molecular weight moiety. Furthermore, an
approximation of the molecular weight impurity size was deter-
mined using molecular weight cutoff spin cartridges (Micron filter
device Ultracel YM-3 and YM-10). The material retained on the
YM-10 filter (regenerated cellulose 10,000 molecular weight cut-
off) was predominantly the late eluting impurity.
Efforts to resolve the diastereomers of 4c were undertaken. We
observed that 4a and 4b do not resolve via reversed-phase HPLC
and chiral chromatography. Alternatively, 4d in which
placed with Ala is readily resolved from 4a and 4b by reversed-
phase HPLC on a C18 column.
LAla is re-
D
A survey of the literature provided a procedure for resolution of
the enantiomers of lipoic acid.52 Treatment of lipoic acid with a
reducing agent followed by reaction with o-phthalaldehyde and L-
phenylalanine provides lipoic acid derivatives that are resolved by
HPLC. An investigation to determine if this procedure could be ap-
plied to the resolution of 4c was performed (Boston Analytical).
Compounds 4a, 4b, and 4c were treated with a reducing agent
2followed by reaction with o-phthalaldehyde and L-phenylalanine.
The HPLC analysis method for the derivatives of 4a, 4b, and 4c was
To determine if the impurity in the crystallized product was a
disulfide linked polymer of 4, the mixture was treated with a
reducing agent. The product containing both 4 and the impurity
was treated with dithiothreitol (DTT) and the mixture was ana-
not specific due to the co-elution of side products from the
Figure 1. HPLC analysis of preparation of 4.

508
S. A. Kates et al. / Bioorg. Med. Chem. 22 (2014) 505–512
derivatization procedure and thus the chiral purity of 4a, 4b, and 4c
could not be established. Due to the distance between the chiral cen-
ters of the lipoyl and glutamyl moieties, chiral resolution of the dia-
stereomers of 4c and corresponding assay for purity of the single
stereoisomerof4aisaseriouschallengeandwillbeexploredfurther.
Based upon the enhanced biological efficacy of 4a compared to
solution followed by the condensation of the activated lipoic acid
with a commercially available amino-containing compound in a
semi-aqueous solution. Polymer-free compounds were isolated
from the crude reaction mixture either by preparative HPLC or
after extractive removal of reactants by crystallization from a solu-
tion of isopropyl or ethyl acetate containing 2-propanol and water.
Structure confirmation and purity was obtained by HPLC, mass
spectrometry and 1H NMR spectroscopy.
4b, the development program focused on the preparation of
a-li-
poic acid adducts incorporating the R-enantiomer. A small library
of compounds was prepared to explore further the role of the
acidic strength. Both alkyl and aryl carboxylic, sulfonic and phos-
ponic derivatives were synthesized using the procedure outlined
2.6. Structure activity relationships
above.
a
-Lipoic acid adducts were produced by first in situ activa-
The results of the cytoprotection assay for the various LA ana-
logs are presented in Table 3. Test compounds were dissolved in
isotonic saline at a dose volume of 1 mL/kg and were administered
tion of the carboxylic moiety of lipoic acid with disuccinimidyl car-
bonate (DSC) in the presence of a tertiary amine in an acetone
Table 3
MI/AR reduction and pKa values of lipoic acid adducts
O
O
% MI/AR reduction
% MI/AR reduction
*
*
S
S
Compound number
pKa
Compound number
pKa
(p value)**
(p value)**
S
S
HO3S
OH
5
19 (p > 0.05)
4.52
3.50
15
30 (p < 0.05)
À1.28
N
H
CO2H
CO2H
CO2H
CO2H
N
H
N
4a
31 (p < 0.01)
16
21 (p > 0.05)
3.31
CO2H
CO2H
N
H
O
CONH2
H
N
CO2H
H
N
6
7
17 (p > 0.05)
28 (p < 0.05)
3.73
3.38
17
18
39 (p < 0.01)
4.45
N
H
O
HO2C
H
N
H
N
SO3H
CO2H
45–52 (p < 0.001)
À1.01
N
H
O
HO
H
N
OSO3H
8
31 (p < 0.05)
3.73
19
30 (p < 0.05)
À1.74
H
N
CO2H
N
H
O
CO2H
H
N
PO3H2
H
9
33 (p < 0.01)
24 (p < 0.05)
26 (p < 0.05)
29 (p < 0.05)
À1.22
3.52
3.39
3.71
20
21
22
53 (p < 0.001)
47 (p < 0.01)
22 (p > 0.05)
1.79
1.54
4.16
N
N
H
SO3H
O
O
H
N
H
N
CO2H
OPO3H2
N
H
10
11
12
HO2C
HO2C
HO
F
H
N
CO2H
N
CO2H
H
SO3H
H
N
23
24
33 (p < 0.05)
44 (p < 0.01)
À2.05
À2.16
N
H
CO2H
H
N
+/-
SO3H
13
14
10 (p > 0.05)
3.89
3.52
N
H
CO2H
CO2H
33 (p < 0.01)
N
H
CO2H
*
Theoretical calculation using MarvinSketch Version 5.5.0.1 with pKa plugin.
p Values paired against placebo.
**

S. A. Kates et al. / Bioorg. Med. Chem. 22 (2014) 505–512
509
as a single intravenous bolus 15 min prior to the 30 min ligation at
60
50
40
30
20
10
0
doses between 1–10 mg/kg. Specifically, the free acid of 4a was
added to a solution that included sufficient 0.5 N NaOH for its
neutralization, NaCl, and water in quantities calculated to produce
285 mOsm, that is, physiological osmolarity. Prior to efficacy test-
ing, formulated solutions were confirmed for both 4a concentra-
tion and absence of polymer by analysis of the HPLC
chromatogram. The MI/AR ratio was calculated and the reduction
of test compound MI/AR to that for the vehicle was calculated.
The parent
model.
a-lipoic acid yielded only a 19% MI/AR reduction in this
To explore the nature of the dipeptide adduct of 4a, 7 was pre-
pared in which one methylene group was removed in each side-
chain (LAsp for LGlu; Gly for LAla). Similar efficacy to 4a (28% MI/
AR reduction compared to placebo) was achieved indicating that
the methylene tether is not critical. In addition, inverting the
-3
-2
-1
0
1
2
3
4
5
pKa
dipeptide sequence to LAla-LGlu in 10 also provided good efficacy
(24% MI/AR reduction compared to placebo). However, removal
of the acidic side-chain in 4a and replacement with a carboxamide
(6) eliminated efficacy. This result corroborates in vitro data dem-
onstrating the requirement of a carboxylic acid function, although
the location of the acidic residue is not critical. The phenolic func-
tion of the side-chain of a tyrosine residue in 8 provides enough
acidity to provide efficacy (31% MI/AR reduction compared to pla-
cebo). Lastly, to explore the role of the C-terminal acidic function,
Figure 2. Scatter plot analysis of lipoic acid analogs efficacy as a function of pKa.
only one acidic function. 14 (
carboxylic acid was less efficacious (33%) than 17 (
(39%); 9 ( Lip- Glu-Tau) containing a carboxylic acid side-chain
and terminal sulfonic acid and 15 ( Lip- Cya-OH) containing a
side-chain sulfonic acid were less efficacious (33% and 30%, respec-
tively) than both 17 and 18 Lip-Tau-OH) (45–52%). The theoretical
R
Lip-
L
Glu-OH) containing a side-chain
RLip-bAla-OH)
R
L
R
L
LAla was replaced with taurine in 9 to incorporate the stronger sul-
R
fonic acid function. The sulfonic acid residue provided slightly bet-
ter efficacy than its corresponding carboxylic acid (33% vs 31%).
pKa values for lipoic acid analogs 5–24 were calculated using Mar-
vinSketch Version 5.5.0.1 with pKa plugin (Table 3). A scatter plot
correlating the efficacy determined by reduction in MI/AR ratio
compared to placebo as a function of theoretical pKa values for li-
poic acid analogs 5–24 is displayed in Figure 2. Analysis of the scat-
To determine the role of the alkyl side chain of
pared in which the residue was removed. Slightly improved effi-
cacy was achieved compared to 4a which incorporates Ala (33%
vs 31%). Replacement of the carboxylic acid function of Glu with
a sulfonic acid (cysteic acid [ Cya]) 15, phenol ( Tyr) 12, and aryl
carboxylic acid (4-carboxy- Phe) 11 provided compounds with
similar efficacy (30%, 29%, and 26%, respectively). Minimal efficacy
was achieved when 3-fluoro- -alanine (10% MI/AR reduction
LAla, 14 was pre-
L
ter plot indicates that modifications to
a-lipoic acid with adducts
L
L
that yield a final product with pKa values between 3 and 4 do
not provide compounds that were as efficacious based upon reduc-
tion of MI/AR ratio to placebo than analogs with pKa values below
2. Interestingly, the optimal results were obtained with com-
pounds that have a pKa value between 1 and 2. Although optimiza-
tion for dose administration for each compound was not
performed, the data clearly indicates a trend that the highest level
L
D/L
compared to placebo). These results demonstrate the significance
of the side-chain function. Potency is achieved when one side-
chain residue contains an acidic function such as a carboxylic,
sulfonic acid or phenol. Compounds 4a, 7, 8, 9, 10, 11, 12, 14, 15
contain 2 acidic functions derived from proteogenic and nonprote-
of efficacy should be achieved with a-lipoic acid analogs possess-
ing a calculated pKa values below 2.
ogenic amino acids.
a-Lipoic acid was treated with iminodiacetic
acid (Idaa) to form 16 containing a tertiary amide with a bis-car-
boxylic acid. Surprisingly, 16 provided a low level of efficacy
(21% MI/AR reduction compared to placebo). Presumably the lower
efficacy is attributed to the incorporation of a tertiary amide as op-
posed to other analogs that contain a secondary amide.
3. Conclusion
A discrete library based upon condensation of natural and
unnatural amino acids with the carboxylic acid moiety of LA was
prepared and evaluated for cytoprotective activity. SAR studies in
a cardiac IRI model indicated that potency was dictated by the
acidic strength of the modification. Linear alkyl acidic (carboxylic,
sulfonic, phosphoric) adducts to LA yielded the most efficacious
compounds. Linear alkyl adducts were more effective than their
corresponding aryl adducts. Analogs containing two acidic func-
tions (carboxylic or sulfonic acid) were less efficacious than com-
pounds modified with only one acidic function. Modifications
that did not contain an acidic function were the least efficacious.
From these studies, sulfonic acid/sulfuric acid > phosphonic acid/
phosphoric acid > carboxylic acid in reducing of MI/AR ratio effects
in a rat model of ischemia-reperfusion injury.
Linear alkyl adducts as opposed to
-lipoic acid and their efficacy was measured. bAla, taurine,
aminoethyl hydrogen sulfate, phosphorylethanolamine, and
aminoethylphosphonic acid were condensed with activated -li-
a-amino acids were attached
to
a
a
poic acid to provide 17, 18, 19, 20, and 21, respectively. The MI/
AR reduction compared to placebo for 17, 18, 19, and 20 was
39%, 45–52%, 30%, 53%, and 47%, respectively. The linear alkyl ad-
ducts provided the highest level of efficacy. Presumably this effect
was due to both the incorporation of a linear alkyl chain and more
importantly acid functions with stronger pKa values.
The corresponding aryl carboxylic and sulfonic acids 22, 23, and
24 were prepared and their efficacy was determined. The MI/AR
reduction compared to placebo for 22, 23, and 24 was 22%, 33%,
and 44%, respectively. These results clearly demonstrate that an al-
kyl acidic function is more favored than an aryl function (14 vs 11,
17 vs 22 and 18 vs 23, 24). Aryl substitution is favored in the 4-po-
sition (24) as opposed to the 3-position 23.
These studies support the potential utility of lipoic acid-peptide
analogs in treating cellular ischemia-reperfusion injury in clinical
settings. The choice of a clinical candidate involves more than
biological efficacy in an animal model. This work also provided in-
sights to the ease of synthesis and compound stability. The analog
Interestingly, adducts to the lipoyl moiety containing 2 acidic
functions generally were less efficacious than a similar adduct with
4a (N-[(R)-1,2-dithiolane-3-pentanoyl]-
selected based upon these attributes as well as a favorable ADMET
L-glutamyl-L-alanine) was

510
S. A. Kates et al. / Bioorg. Med. Chem. 22 (2014) 505–512
and receptor screen profile for further preclinical and clinical inves-
tigation.53 In a human trial, 4a, designated CMX-2043, showed sta-
tistically significant protection against myocardial damage
associated with percutaneous coronary intervention (PCI).54
4.2.2. N-[(S)-1,2-dithiolane-3-pentanoyl]-L-glutamyl-L-alanine
(4b)
Prepared from S-lipoic acid. HPLC: >99% (rt: 12.3 min); À69.5
(EtOH, 25 °C, 10 mg/mL).
4. Material and methods
4.2.3. N-[(R/S)-1,2-dithiolane-3-pentanoyl]-L-glutamyl-L-alanine
(4c)
4.1. General experimental procedures
Prepared from R/S-lipoic acid. HPLC: >99% (rt: 12.3 min); À24.2
(EtOH, 25 °C, 10 mg/mL).
All reagents were purchased from commercial sources and used
without further purification. Racemic lipoic acid was obtained
from Sigma–Aldrich (US), R-lipoic acid from Labochim (Italy), and
S-lipoic acid from Shanghai Freeman Lifescience (China). Natural
and unnatural amino acids were obtained from either Sigma–Al-
drich (US) or Bachem (Switzerland); dipeptides from Bachem,
chemical reagents from Sigma–Aldirch, and solvents from VWR
(US). Acetylated peptides 1, 2, and 3 were obtained either from
CS Bio (San Carlos, CA) or MidWest Biotech (Fishers, IN). Optical
rotation was performed with a Rudolph Polarimeter model Digi-
Pol-781 SDV. RP-HPLC chromatography was performed with a
Waters 1525 binary pump, 717 plus autosampler coupled to a dual
4.2.4. N-[(R)-1,2-dithiolane-3-pentanoyl]-L-glutaminyl-glycine
(6)
HPLC: 100% (rt: 10.0 min); 1H NMR (400 MHz, d6-DMSO) d 8.2
(d, 1, J = 7.15, NH), 7.9 (d, 1, J = 7.93, NH), 4.85 (m, 1, CH), 3.90 (d,
2H, CH), 3.10 (m, 1), 3.60 (m, 1, CH-S), 3.19-3.05 (m, 2, CH2-S);
ESI-MS: m/z calcd for C15H25N3O5S2 = 392 (MÀ1).
4.2.5. N-[(R)-1,2-dithiolane-3-pentanoyl]-L-aspartyl-glycine (7)
HPLC: >99% (rt: 9.19 min); 1H NMR (400 MHz, d6-DMSO) d 8.2
(d, 1, J = 7.15, NH), 7.9 (d, 1, J = 7.93, NH), 4.30 (m, 1, CH), 4.17
(m, 1, CH), 3.60 (m, 1, CH-S), 3.19-3.05 (m, 2, CH2-S); ESI-MS: m/
z calcd for C14H22N2O6S2 = 379 (M+1).
band detector 2487. The analyses were conducted using
a
YMC-PackPro C18 column (100 Â 4.6 mm I.D), a mobile phase
linear gradient, and monitoring by UV absorption at 220 nm. Mass
spectra were provided by HT Laboratories or Creagen. Identity of
all final compounds was assessed by mass spectrometry.
4.2.6. N-[(R)-1,2-dithiolane-3-pentanoyl]-L-tyrosinyl-L-alanine
(8)
HPLC: >98% (rt: 13.1 min); 1H NMR (400 MHz, d6-DMSO) d 8.1
(d, 1, J = 7.15, NH), 8.0 (d, 1, J = 7.93, NH), 4.63 (m, 1, CH), 4.40
(m, 1, CH), 3.60 (m, 1, CH-S), 3.19-3.05 (m, 2, CH2-S); ESI-MS: m/
z calcd for C20H28N2O5S2 = 439 (MÀ1).
4.2. Library synthesis
General procedure for the synthesis of
a-lipoic acid analogs:
4.2.7. N-[(R)-1,2-dithiolane-3-pentanoyl]-L-glutamyl-L-taurine
(R)Lipoic acid ( Lip-OH, 10.0 g, 48.5 mmol) was dissolved in acetone
R
(9)
HPLC: 100% (rt: 9.05 min); 1H NMR (400 MHz, d6-DMSO) d 8.3
(d, 1, J = 7.15, NH), 7.9 (d, 1, J = 7.93, NH), 4.20 (m, 1, CH), 3.36
(m, 1, 2H, CH2), 3.60 (m, 1, CH-S), 3.19–3.05 (m, 2, CH2-S); ESI-
MS: m/z calcd for C15H26N2O7S3 = 441 (MÀ1).
(100 mL, 10 mL/g). The solution was protected from direct light by
covering the reaction flask with foil. N,N-Disuccinimidylcarbonate
(15.5 g, 1.25 equiv, 60.6 mmol) and N,N-diisopropylethylamine
(DIEA, 10.5 mL, 1.25 equiv, 60.6 mmol) were added sequentially
and the reaction was stirred vented for 2 h at room temperature
to form Lip-OSu in situ. The corresponding amines (1.15 equiv) were
added to the solution of Lip-OSu in acetone, followed by the addition
of water (50 mL) and DIEA (19.4 mL, 2.3 equiv, 112 mmol). The
combined solution was stirred overnight. Approximately one third
of the reaction mixture was then reduced to approximately half vol-
ume on a rotary evaporator. The remaining reaction mixture was in-
jected multiple times directly onto a semi-preparative high-
performance liquid chromatography (HPLC) system and the product
isolated on a YMC Pack Pro C18 reverse phase column using a gradi-
ent of increasing acetonitrile (0.5% acetic acid) in water (0.5% acetic
acid). Product-containing fractions were identified by analytical
HPLC, frozen, and lyophilized.
4.2.8. N-[(R)-1,2-dithiolane-3-pentanoyl]-L-alaninyl-L-glutamic
acid (10)
HPLC: 98% (rt: 11.25 min); 1H NMR (400 MHz, d6-DMSO) d 8.3
(d, 1, J = 7.15, NH), 7.9 (d, 1, J = 7.93, NH), 4.20 (m, 1, CH), 3.36
(m, 1, 2H, CH2), 3.60 (m, 1, CH-S), 3.19-3.05 (m, 2, CH2-S); ESI-
MS: m/z calcd for C16H26N2O6S2 = 405 (MÀ1).
4.2.9. N-[(R)-1,2-dithiolane-3-pentanoyl]-L-(4-carboxy)-
phenylalanine (11)
HPLC: 99% (rt: 12.69 min); 1H NMR (400 MHz, d6-DMSO) d 7.9
(d, 1, J = 8.0, NH), 4.20 (m, 1, CH), 3.36 (m, 1, 2H, CH2), 3.60 (m,
1, CH-S), 3.19–3.05 (m, 2, CH2-S); ESI-MS: m/z calcd for C18H23NO5-
S2 = 398 (M+1).
4.2.1. N-[(R)-1,2-dithiolane-3-pentanoyl]-L-glutamyl-L-alanine
(4a)
4.2.10. N-[(R)-1,2-dithiolane-3-pentanoyl]-L-tyrosine (12)
HPLC: 95% (rt: 12.98 min); 1H NMR (400 MHz, d6-DMSO) d 7.9
(d, 1, J = 8.0, NH), 4.20 (m, 1, CH), 3.60 (m, 1, CH-S), 3.19–3.05
(m, 2, CH2-S); ESI-MS: m/z calcd for C17H23NO4S2 = 370 (M+1).
6.9 g (35%); HPLC: >99% (rt: 12.3 min); 1H NMR (400 MHz, d6-
DMSO) d 12.29 (br s, 2, OH), 8.19 (d, 1, J = 7.15, NH), 7.93 (d, 1,
J = 7.93, NH), 4.30 (m, 1, CH), 4.17 (m, 1, CH), 3.60 (m, 1, CH-S),
3.19–3.05 (m, 2, CH2-S), 2.45–2.35 (m, 1), 2.28–2.34 (m, 2), 2.18-
2.08 (m, 2), 1.91–1.84 (m, 2), 1.75–1.62 (m, 2), 1.56–1.48 (m, 3),
1.36–1.32 (m, 2), 1.27 (d, 3, J = 7.25, CH3); 13C NMR (d6-DMSO,
100 MHz) d 174.0 (CO2H, CO2H), 172.0 (CONH), 171.1 (CONH),
56.1 (NHCHCO), 51.3 (NHCHCO), 47.4, (SCH(CH2)2) 39.5 (SCH2),
38.1 (CH2), 34.9 (CH2), 34.1(CH2), 30.0(CH2), 28.2(CH2), 27.6(CH2),
25.0 (CH2), 17.0 (CH3); mp 128–130 °C; IR (KBr pellet) 3281,
3300–2500, 1722, 1638, 1546, 1453, 1425, 1273 cmÀ1; ESI-MS:
m/z = 405.33 (MÀ1); Anal. Calcd for C16H26N2O6S2: C, 47.3; H,
6.4; N, 6.9; S, 15.8. Found C, 46.8 H, 6.4; N, 6.9; S, 15.4; +36.1 (EtOH,
25 °C, 10 mg/mL).
4.2.11. N-[(R)-1,2-dithiolane-3-pentanoyl]-D/L-b-flouro-alanine
(13)
HPLC: 100% (rt: 12.49 min); 1H NMR (400 MHz, d6-DMSO) d 8.1
(d, 1, J = 8.0, NH), 4.13 (m, 1, CH), 3.60 (m, 1, CH-S), 3.19-3.05 (m, 2,
CH2-S); ESI-MS: m/z calcd for C11H18FNO3S2 = 296 (M+1).
4.2.12. N-[(R)-1,2-dithiolane-3-pentanoyl]-L-glutamic acid (14)
HPLC: 95% (rt: 12.98 min); 1H NMR (400 MHz, d6-DMSO) d 8.0
(d, 1, J = 8.0, NH), 4.37 (m, 1, CH), 3.60 (m, 1, CH-S), 3.19–3.05
(m, 2, CH2-S); ESI-MS: m/z calcd for C13H21NO5S2 = 334 (MÀ1).

S. A. Kates et al. / Bioorg. Med. Chem. 22 (2014) 505–512
511
4.2.13. N-[(R)-1,2-dithiolane-3-pentanoyl]-
L
-cysteic acid (15)
potential of lipoic acid analogs.12,13 This model is analogous to
the I/R injury observed in patients following coronary occlusions
and cardiac surgery procedures, such as percutaneous coronary
intervention (PCI) and coronary artery bypass grafting (CABG).
Male Sprague Dawley rats (6–12/group dependent on the experi-
ment) (Charles River) between 300 and 350 g were anesthetized
with 3–4% isoflurane in an induction chamber. Anesthesia was
maintained at a surgical plane with 1.5–2.0% isoflurane, adminis-
tered by a rodent ventilator (Harvard Apparatus Inc.) through a
16-gauge endotracheal tube. The ventilator was set at 2.5 cc at a
rate of 60–65 breaths per minute to maintain ventilation during
surgery. The core temperature of the animal was monitored with
a rectal probe and maintained at 37 °C with a heating lamp at-
tached to a temperature controller (Physitemp Instruments Inc).
A left anterior thoracotomy was performed and the heart was
exposed using a vertical pericardotomy. Ischemia in the left ventri-
cle was induced by ligating the left coronary artery approximately
4 mm from the base of the aorta using a cardiovascular 7.0 mono-
filament suture on an 11 mm needle (Ethicon Inc.). Fluorescent
HPLC: 97% (rt: 7.61 min); 1H NMR (400 MHz, d6-DMSO) d 8.0 (d,
1, J = 8.0, NH), 4.55 (m, 1, CH), 3.75 (m, 1, CH-S), 3.19–3.05 (m, 2,
CH2-S); ESI-MS: m/z calcd for C11H19NO6S3 = 360 (M+1).
4.2.14. N-[(R)-1,2-dithiolane-3-pentanoyl]-iminodiacetic acid
(16)
HPLC: 90% (rt: 11.40 min); 1H NMR (400 MHz, d6-DMSO) d 8.2
(d, 1, J = 8.0, NH), 4.20 (d, 4H, CH2), 3.75 (m, 1, CH-S), 3.19–3.05
(m, 2, CH2-S); ESI-MS: m/z calcd for C12H19NO5S2 = 320 (MÀ1).
4.2.15. N-[(R)-1,2-dithiolane-3-pentanoyl]-b-alanine (17)
HPLC: 100% (rt: 10.95 min); 1H NMR (400 MHz, d6-DMSO) d 8.2
(d, 1, J = 8.0, NH), 4.20 (d, 4H, CH2), 3.75 (m, 1, CH-S), 3.19–3.05 (m,
2, CH2-S); ESI-MS: m/z calcd for C11H19NO3S2 = 278 (M+1).
4.2.16. N-[(R)-1,2-dithiolane-3-pentanoyl]-taurine (18)
HPLC: 100% (rt: 8.29 min); 1H NMR (400 MHz, d6-DMSO) d 8.1
(d, 1, J = 8.0, NH), 4.16 (t, 2H, J = 7.10 CH2), 3.75 (m, 1, CH-S),
3.19–3.05 (m, 2, CH2-S), 2.98 (t, 2H, J = 7.05); ESI-MS: m/z calcd
for C10H19NO4S3 = 312 (MÀ1).
microspheres (Invitrogen, FluoSpheres 10 lm, 300 lL) were in-
jected into the left ventricular cavity 15 min after the ligation to
delineate the ischemic area. The suture was removed 30 min after
ligation and the ischemic area was checked to insure reperfusion.
The chest was then closed using 5–0 silk sutures for the muscle
layers, and wound clips for the cutaneous layer. The animals were
allowed to recover from anesthesia under temperature-controlled
conditions before they were returned to the colony.
4.2.17. N-[(R)-1,2-dithiolane-3-pentanoyl]-2-aminoethyl hydro-
gen sulfate (19)
HPLC: 100% (rt: 9.05 min); 1H NMR (400 MHz, d6-DMSO) d 8.0
(d, 1, J = 8.0, NH), 3.65 (m, 2H, CH2), 3.75 (m, 1, CH-S), 3.19–3.05
(m, 2, CH2-S), 2.98 (t, 2H, J = 7.05); ESI-MS: m/z calcd for C10H19-
NO5S3 = 328 (MÀ1).
Twenty-four hours after reperfusion, anesthesia was induced
using ketamine hydrochloride and the chest was opened. The ani-
mals were sacrificed by injecting a 15% potassium chloride solution
(w/v) into the left ventricular cavity to arrest the heart in diastole.
The heart was excised distal to the aortic valve and washed with
saline to remove residual blood. Sagittal slices of the heart were
prepared between the base of the ventricle and the apex. Five slices
of heart tissue were obtained, each 2 mm thick. The slices were
stained by immersion in a 1% solution of 2,3,5-triphenyl-2H-tetra-
zolium chloride (TTC) (Sigma) in isotonic saline. Images of the
slices were obtained under bright field microscopy to document
tissue TTC staining; and fluorescence microscopic images were
generated to observe the deposition of microspheres. The area at
risk was delineated by the absence of microspheres and the infarct
area was determined by the absence of TTC staining. The AR and MI
areas were quantitated to obtain the MI/AR ratio. Statistical analy-
sis (Student’s t-test) was determined (data not shown) and the
reduction of MI/AR ratio compared to placebo reported.
4.2.18. N-[(R)-1,2-dithiolane-3-pentanoyl]-2-aminoethyl phos-
phonic acid (20)
HPLC: 98% (rt: 8.63 min); 1H NMR (400 MHz, d6-DMSO) d 7.9 (d,
1, J = 8.0, NH), 3.75 (m, 1, CH-S), 3.19–3.05 (m, 2, CH2-S), 2.05 (m,
2H, CH2); ESI-MS: m/z calcd for C10H20NO4PS2 = 312 (MÀ1).
4.2.19. N-[(R)-1,2-dithiolane-3-pentanoyl]-O-phosphoryl-etha-
nolamine (21)
HPLC: 89% (rt: 8.32 min); 1H NMR (400 MHz, d6-DMSO) d 8.2 (d,
1, J = 8.0, NH), 3.84 (q, 2H, J = 7.0, CH2) 3.75 (m, 1, CH-S), 3.19–3.05
(m, 2, CH2-S),; ESI-MS: m/z calcd for C10H20NO5PS2 = 328 (MÀ1).
4.2.20. N-[(R)-1,2-dithiolane-3-pentanoyl]-4-aminobenzoic acid
(22)
HPLC: 99% (rt: 15.25 min); 1H NMR (400 MHz, d6-DMSO) d 8.1
(d, 1, J = 8.0, NH), 3.75 (m, 1, CH-S), 3.19–3.05 (m, 2, CH2-S),; ESI-
MS: m/z calcd for C15H19NO3S2 = 326 (M+1).
Acknowledgment
4.2.21. N-[(R)-1,2-dithiolane-3-pentanoyl]-3-aminobenzene
sulfonic acid (23)
We acknowledge Cedarburg Hauser Pharmaceuticals for their
valuable advice and expertise.
HPLC: 95% (rt: 12.50 min); 1H NMR (400 MHz, d6-DMSO) d 8.1
(d, 1, J = 8.0, NH), 3.75 (m, 1, CH-S), 3.19–3.05 (m, 2, CH2-S),; ESI-
MS: m/z calcd for C14H19NO4S3 = 360 (MÀ1).
References and notes
4.2.22. N-[(R)-1,2-dithiolane-3-pentanoyl]-3-aminobenzene
sulfonic acid (24)
1. Piper, H. M.; Abdallah, Y.; Schafer, C. Cardiovasc. Res. 2004, 61, 365.
2. Yellon, D. M.; Hausenloy, D. J. N. Eng. J. Med. 2007, 357, 1121.
3. Otani, H. Antioxid. Redox Signaling 2008, 10, 207.
4. Whiteman, M.; Tritschler, H.; Halliwell, B. FEBS Lett. 1996, 379, 74.
5. Packer, L. Drug Metab. Rev. 1998, 30, 245.
HPLC: 92% (rt: 12.37 min); 1H NMR (400 MHz, d6-DMSO) d 8.0
(d, 1, J = 8.0, NH), 3.75 (m, 1, CH-S), 3.19-3.05 (m, 2, CH2-S); ESI-
MS: m/z calcd for C14H19NO4S3 = 360 (MÀ1).
6. Deng, C.; Sun, Z.; Tong, G.; Yi, W.; Ma, L.; Zhao, B.; Cheng, L.; Zhang, J.; Cao, F.;
Yi, D. PLoS One 2013, 8, e58371.
7. Zweier, J. L. J. Biol. Chem. 1988, 263, 1353.
4.3. Biological study
8. Werns, S. W.; Lucchesi, B. R. Cardiovasc. Drugs Ther. 1989, 2, 761.
9. Ganguly, N. K.; Nalini, K.; Wahi, S.; Dhawan, V.; Meenakshi, S.; Chakravarti, R.
N. Mol. Cell. Biochem. 1992, 111, 71.
10. Giordano, F. J. J. Clin. Investig. 2005, 115, 500.
11. Pugazhenthi, S.; Akhov, L.; Selvaraj, G.; Wang, M.; Alam, J. Am. J. Physiol.
Endocrinol. Metab. 2007, 293, E645.
Rat model of I-R injury. Male Sprague Dawley rats were used
in accordance with the Guide for the Care and Use of Laboratory
Animals. Animal studies were approved by the Ischemix Animal
Care and Use Committee. A standard rat model of myocardial
ischemia-reperfusion injury was used to evaluate the protective
12. Teichert, J.; Kern, J.; Tritschler, H. J.; Ulrich, H.; Preiss, R. Int. J. Clin. Pharmacol.
Ther. 1998, 36, 625.

512
S. A. Kates et al. / Bioorg. Med. Chem. 22 (2014) 505–512
13. Schupke, H.; Hempel, R.; Peter, G.; Hermann, R.; Wessel, K.; Engel, J.; Kronbach,
T. Drug Metab. Dispos. 2001, 29, 855.
32. Shashoua, V. E.; Adams, D. S.; Boyer-Boiteau, A.; Cornell-Bell, A.; Li, F.; Fisher,
M. Brain Res. 2003, 963, 214.
14. Muller, C.; Dunschede, F.; Koch, E.; Vollmar, A. M.; Kiemer, A. K. Am. J. Physiol.
Gastrointest. Liver Physiol. 2003, 285, G769.
15. Glantzounis, G. K.; Yang, W.; Koti, R. S.; Mikhailidis, D. P.; Seifalian, A. M.;
Davidson, B. R. Curr. Pharm. Des. 2006, 12, 2891.
33. Guillonneau, C.; Charton, Y.; Ginot, Y. M.; Fouquier-d’Herouel, M. V.; Bertrand,
M.; Lockhart, B.; Lestage, P.; Goldstein, S. Eur. J. Med. Chem. 2003, 38, 1.
34. Rosini, M.; Andrisano, V.; Bartolini, M.; Bolognesi, M. L.; Hrelia, P.; Minarini, A.;
Tarozzi, A.; Melchiorre, C. J. Med. Chem. 2005, 48, 360.
16. Dulundu, E.; Ozel, Y.; Topaloglu, U.; Sehirli, O.; Ercan, F.; Gedik, N.; Sener, G.
Pharmacology 2007, 79, 163.
17. Duenschede, F.; Erbes, K.; Riegler, N.; Ewald, P.; Kircher, A.; Westermann, S.;
Schad, A.; Miesmer, I.; Albrecht-Schock, S.; Gockel, I.; Kiemer, A. K.; Junginger,
T. World J. Gastroenterol. 2007, 13, 3692.
18. Takaoka, M.; Ohkita, M.; Kobayashi, Y.; Yuba, M.; Matsumura, Y. Clin. Exp.
Pharmacol. Physiol. 2002, 29, 189.
19. Sehirli, O.; Sener, E.; Cetinel, S.; Yuksel, M.; Gedik, N.; Sener, G. Clin. Exp.
Pharmacol. Physiol. 2008, 35, 249.
20. Bae, E. H.; Lee, K. S.; Lee, J.; Ma, S. K.; Kim, N. H.; Choi, K. C.; Frokiaer, J.; Nielsen,
S.; Kim, S. Y.; Kim, S. Z.; Kim, S. H.; Kim, S. W. Am. J. Physiol. Renal Physiol. 2008,
294, F272.
21. Koga, H.; Hagiwara, S.; Kusaka, J.; Goto, K.; Uchino, T.; Shingu, C.; Kai, S.;
Noguchi, T. J. Surg. Res. 2012, 174, 352.
35. Hausenloy, D. J.; Yellon, D. M. Cardiovasc. Res. 2004, 61, 448.
36. Bolli, R.; Becker, L.; Gross, G.; Mentzer, R., Jr.; Balshaw, D.; Lathrop, D. A. Circ.
Res. 2004, 95, 125.
37. Kobara, M.; Tatsumi, T.; Takeda, M.; Mano, A.; Yamanaka, S.; Shiraishi, J.; Keira,
N.; Matoba, S.; Asayama, J.; Nakagawa, M. Basic Res. Cardiol. 2003, 98, 319.
38. Matsui, T.; Tao, J.; del Monte, F.; Lee, K. H.; Li, L.; Picard, M.; Force, T. L.; Franke,
T. F.; Hajjar, R. J.; Rosenzweig, A. Circulation 2001, 104, 330.
39. Tam, J. P.; Riemen, M. W.; Merrifield, R. B. Pept. Res. 1988, 1, 6.
40. Packer, L.; Kraemer, K.; Rimbach, G. Nutrition 2001, 17, 888.
41. Biewenga, G. P.; Haenen, G. R.; Bast, A. Gen. Pharmacol. 1997, 29, 315.
42. Wolz, P.; Krieglstein, J. Neuropharmacology 1996, 35, 369.
43. Barltrop, J. A.; Hayes, P. M.; Calvin, M. J. Am. Chem. Soc. 1954, 76, 4348.
44. Wagner, A. F.; Walton, E.; Boxer, G. E.; Pruss, M. P.; Holly, F. W.; Folkers, K. J.
Am. Chem. Soc. 1956, 78, 5079.
22. Freisleben, H. J. Toxicology 2000, 148, 159.
45. Thomas, R. C.; Reed, L. J. J. Am. Chem. Soc. 1956, 78, 6148.
46. Brown, P. R.; Edwards, J. O. J. Org. Chem. 1969, 34, 3131.
47. Klatt, M. J.; Niebel, M.; Paust, J.; Rieger, J. US Patent 6,441,024, 2002.
48. Beisswenger, T.; Bethge, H.; Goede, J.; Hubner, F.; Huthmacher, K.; Klenk, H.;
Moller, R. US Patent 5,455,264, 1992.
23. Ko, K. M.; Yiu, H. Y. Mol. Cell. Biochem. 2001, 220, 141.
24. Oh, S. K.; Yun, K. H.; Yoo, N. J.; Kim, N. H.; Kim, M. S.; Park, B. R.; Jeong, J. W.
Korean Circ. J. 2009, 39, 359.
25. Wang, X.; Yu, Y.; Ji, L.; Liang, X.; Zhang, T.; Hai, C. X. Food Chem. Toxicol. 2011,
49, 2750.
26. He, L.; Liu, B.; Dai, Z.; Zhang, H. F.; Zhang, Y. S.; Luo, X. J.; Ma, Q. L.; Peng, J. Eur. J.
Pharmacol. 2012, 678, 32.
27. Connell, B. J.; Saleh, M.; Khan, B. V.; Saleh, T. M. Brain Res. 2011, 1375, 128.
28. Yan, G. T.; Si, Y. I.; Zhang, J. Y.; Deng, Z. H.; Xue, H. Zhongguo Wei Zhong Bing Ji
Jiu Yi Xue 2011, 23, 345.
29. Richard, M. J.; Connell, B. J.; Khan, B. V.; Saleh, T. M. Neurosci. Res. 2011, 69, 299.
30. Bitar, M. S.; Ayed, A. K.; Abdel-Halim, S. M.; Isenovic, E. R.; Al-Mulla, F. Life Sci.
2010, 86, 844.
31. Ziegler, D.; Nowak, H.; Kempler, P.; Vargha, P.; Low, P. A. Diabet. Med. 2004, 21,
114.
49. Hettche, H.; Rischer, M.; Sarlikiotis, W. US Patent 6,348,490, 2002.
50. Ames, B. N. US Patent 7,030,154 2006.
51. Majeed, M.; Nagabhushanam, K. US Patent 7,202,270, 2007.
52. Niebch, G.; Buchele, B.; Blome, J.; Grieb, S.; Brandt, G.; Kampa, P.; Raffel, H. H.;
Locher, M.; Borbe, H. O.; Nubert, I.; Fleischhauer, I. Chirality 1997, 9, 32.
53. Baguisi, A.; Beeuwkes, R.; Casale, R.; Kates, S.A.; Lader, A. manuscripts in
preparation.
54. Krucoff, M.W., Ponde, C.K., Hiremath, J., Ajit, M., Ramsaran, E., Patel, M.R., Lader,
A.S., Beeuwkes, R., Gibson, C.M., Tcheng, J.E. Presented at the 24th Annual
Transcatheter Cardiovascular Therapeutics (TCT) Scientific Symposium, Miami,
FL; October 2012, Featured Clinical Research I Session.