The molecular structure of thio-ether fatty acids influences PPAR-dependent regulation of lipid metabolism

Article abstract of DOI:10.1016/j.bmc.2016.01.045

Thio-ether fatty acids (THEFAs), including the parent 2-(tetradecylthio)acetic acid (TTA), are modified fatty acids (FAs) that have profound effects on lipid metabolism given that they are blocked for β-oxidation, and able to act as peroxisome proliferator-activated receptor (PPAR) agonists. Therefore, TTA in particular has been tested clinically for its therapeutic potential against metabolic syndrome related disorders. Here, we describe the preparation of THEFAs based on the TTA scaffold with either a double or a triple bond. These are tested in cultured human skeletal muscle cells (myotubes), either as free acid or following esterification as phospholipids, lysophospholipids or monoacylglycerols. Metabolic effects are assessed in terms of cellular bioavailabilities in myotubes, by FA substrate uptake and oxidation studies, and gene regulation studies with selected PPAR-regulated genes. We note that the inclusion of a triple bond promotes THEFA-mediated FA oxidation. Furthermore, esterification of THEFAs as lysophospholipids also promotes FA oxidation effects. Given that the apparent clinical benefits of TTA administration were offset by dose limitation and poor bioavailability, we discuss the possibility that a selection of our latest THEFAs and THEFA-containing lipids might be able to fulfill the therapeutic potential of the parent TTA while minimizing required doses for efficacy, side-effects and adverse reactions.

Full text of DOI:10.1016/j.bmc.2016.01.045

Bioorganic & Medicinal Chemistry 24 (2016) 1191–1203
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
The molecular structure of thio-ether fatty acids influences PPAR-
dependent regulation of lipid metabolism
Jenny Lund a,y, Camilla Stensrud a,y, Rajender b,c, Pavol Bohov d, G. Hege Thoresen a,e, Rolf K. Berge d,f
,
Michael Wright b,g, Ahmed Kamal c, Arild C. Rustan a, Andrew D. Miller c,g,h, Jon Skorve d,
⇑
a Department of Pharmaceutical Biosciences, School of Pharmacy, University of Oslo, Norway
b Imperial College Genetic Therapies Centre, Department of Chemistry, Flowers Building, Imperial College London, UK
c Division of Organic Chemistry, Indian Institute of Chemical Technology, Hyderabad, India
d Department of Clinical Science, University of Bergen, N-5021 Bergen, Norway
e Department of Pharmacology, Institute of Clinical Medicine, Faculty of Medicine, University of Oslo and Oslo University Hospital, Norway
f Department of Heart Disease, Haukeland University Hospital, Norway
g Institute of Pharmaceutical Science, Franklin-Wilkins Building, King’s College London, UK
h GlobalAcorn Ltd, London, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
Thio-ether fatty acids (THEFAs), including the parent 2-(tetradecylthio)acetic acid (TTA), are modified
fatty acids (FAs) that have profound effects on lipid metabolism given that they are blocked for b-oxida-
tion, and able to act as peroxisome proliferator-activated receptor (PPAR) agonists. Therefore, TTA in
particular has been tested clinically for its therapeutic potential against metabolic syndrome related dis-
orders. Here, we describe the preparation of THEFAs based on the TTA scaffold with either a double or a
triple bond. These are tested in cultured human skeletal muscle cells (myotubes), either as free acid or
following esterification as phospholipids, lysophospholipids or monoacylglycerols. Metabolic effects
are assessed in terms of cellular bioavailabilities in myotubes, by FA substrate uptake and oxidation
studies, and gene regulation studies with selected PPAR-regulated genes. We note that the inclusion of
Received 20 November 2015
Revised 19 January 2016
Accepted 21 January 2016
Available online 23 January 2016
Keywords:
Metabolic syndrome
b-Oxidation
Diabetes type 2
PPAR receptors
Glycerolipids
TTA
Thio-ether fatty acids
THEFAs
a
triple bond promotes THEFA-mediated FA oxidation. Furthermore, esterification of THEFAs as
lysophospholipids also promotes FA oxidation effects. Given that the apparent clinical benefits of TTA
administration were offset by dose limitation and poor bioavailability, we discuss the possibility that a
selection of our latest THEFAs and THEFA-containing lipids might be able to fulfill the therapeutic
potential of the parent TTA while minimizing required doses for efficacy, side-effects and adverse
reactions.
Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction
In the past decade, it has been found that structurally modified
Abbreviations: THEFA, thio-ether fatty acid; FA, fatty acid; MUFA, monounsat-
urated fatty acid; SFA, saturated fatty acid; TTA, 2-(tetradecylthio)acetic acid; TTA-PC,
1,2-di-[2-(tetradecylthio)acetyl]-sn-glycero-3-phosphocholine; TTA-MAG, 1-[2-(tetrade
cylthio)acetyl]-sn-glycerol; dTTA, 2-(tetradec-[7-enyl]thio)acetic acid; dTTA-PC,
sulfur-containing thio-ether fatty acids (THEFAs) are more potent
than natural fatty acids (FAs) in modulating critical steps in the
regulation of lipid metabolism.1 In particular, the THEFA known
as 2-(tetradecylthio)acetic acid (TTA) 1 was found especially
potent as a regulator of lipid metabolism (Fig. 1).2,3 TTA 1 can be
described as palmitic acid with a sulfur atom inserted between car-
bon atoms 2 and 3, or as septadecanoic acid with a sulfur atom in
position 3 of the FA main chain, introduced as a means to limit b-
oxidation. THEFAs studied with their sulfur atom positioned at
higher odd number positions in the FA main chain were found
more prone to b-oxidation than TTA 1. Given the clear resistance
1,2-di-[2-(tetradec-[7-enyl]thio)acetyl]-sn-glycero-3-phosphocholine;
lyso-dTTA-PC,
1-[2-(tetradec-[7-enyl]thio)acetyl]-sn-glycero-3-phosphocholine; dTTA-MAG, 1-[2-(tet-
radec-[7-enyl]thio)acetyl]-sn-glycerol; tTTA, 2-(tetradec-[7-ynyl]thio)acetic acid; tTTA-
PC, 1,2-di-[2-(tetradec-[7-ynyl]thio)acetyl]-sn-glycero-3-phosphocholine; lyso-tTTA-PC,
1-[2-(tetradec-[7-ynyl]thio)acetyl]-sn-glycero-3-phosphocholine; tTTA-MAG, 1-[2-(tet-
radec-[7-ynyl]thio)acetyl]-sn-glycerol; OA, oleic acid; ANGPTL4, angiopoietin-like 4;
FABP3, fatty acid binding protein 3; CD36, fatty acid translocase; CPT1A, carnitine
palmitoyltransferase 1A; NR, nuclear receptor; CA, cell-associated radioactivity.
⇑
Corresponding author. Tel.: +47 91731452; fax: +47 5597 23115.
E-mail address: jon.skorve@k2.uib.no (J. Skorve).
These authors contributed equally.
y
http://dx.doi.org/10.1016/j.bmc.2016.01.045
0968-0896/Ó 2016 Elsevier Ltd. All rights reserved.

1192
J. Lund et al. / Bioorg. Med. Chem. 24 (2016) 1191–1203
O
metabolic diseases because of their critical role in the regulation
S
of lipid metabolism and fat cell differentiation. Natural ligands of
the PPARs include a wide variety of saturated and unsaturated
FAs plus eicosanoid derivatives. As a PPAR ligand, TTA 1 is known
to modulate the expression of lipid-metabolizing enzymes
involved in catabolic pathways, all in a PPAR dependent man-
ner.4–7 This array of biological responses has suggested that TTA
1 could be a potent therapeutic agent matched to a variety of indi-
cations.1 Accordingly, clinical trials have been carried out that indi-
cate that TTA 1 could beneficially affect plasma lipid levels and
inflammatory parameters in humans.8–10 However, although TTA
OH
O
1
O
O
S
P
N
O
O
O
O
S
2
O
O
O
O
S
P
N
O
O
O
3
OH
O
S
O
OH
1 operates by clear mechanisms involving PPAR
a agonism (and
4
OH
O
to some extent d and
c
agonism too) with modest acute toxicity,11
the clinical development of this compound has been significantly
impaired by low water solubility and poor bioavailability due to
poor absorption, distribution, and excretion properties.12 Hence,
the ultimate clinical indication for TTA 1 use and the most appro-
priate form and means of administration remain to be established.
Following this, we reported fairly recently on a comparison of
biological effects following peroral administration of free TTA 1,
TTA-containing phosphatidylcholine (1,2-di-[2-(tetradecylthio)
acetyl]-sn-glycero-3-phosphocholine, TTA-PC) 2 and TTA-contain-
ing triacylglycerol (1,2,3-tri-[2-(tetradecylthio)acetyl]-sn-glycerol,
TTA-TAG) to an obese, male Wistar rat model (Fig. 1).13 Intrigu-
ingly, the TTA-PC 2 demonstrated metabolic properties (i.e., lower-
ing triacylglycerol and cholesterol levels in plasma plus hepatic
triacylglycerol levels; increasing rates of oxidation of palmitoyl-
CoA; enhancing carnitine palmitoyltransferase II, 3-hydroxy-3-
methylglutaryl-CoA synthase, and fatty acyl-CoA oxidase activities)
that were much enhanced relative to the situation observed with
2 mol equiv of TTA 1. On the other hand, effects of TTA-TAG were
essentially indistinguishable from effects mediated by 3 mol equiv
of TTA 1. Therefore, administration of TTA-PC 2 appeared to be a
very useful chemical formulation for TTA 1 in terms of improving
S
OH
5
6
O
O
O
S
P
N
O
O
O
O
S
O
O
O
O
S
P
N
O
O
O
7
8
9
OH
O
S
O
OH
OH
O
S
OH
O
O
O
O
S
P
N
O
O
O
S
10
O
O
O
O
S
P
N
O
O
O
TTA
1 bioavailability and TTA-mediated biological effects in
11
12
OH
O
particular FA oxidation and lipid lowering capacity. Subsequently,
we were able to demonstrate that TTA-containing lysophos-
phatidylcholine (1-[2-(tetradecylthio)acetyl]-sn-glycero-3-phos-
phocholine, lyso-TTA-PC) 3 (Fig. 1) was even more potent as a
PPARa agonist than TTA 1 itself when administered to MCF-7 cells
in vitro, suggesting that lyso-TTA-PC 3 might be a key active form
S
O
OH
OH
Figure 1. Chemical structures of THEFAs plus their corresponding glycerolipids
synthesized and used here in the following biological studies. The first is TTA 1 with
corresponding glycerolipids TTA-PC 2, lyso-TTA-PC
3 and 1-[2-(tetradecylthio)
of TTA 1 in vivo either in terms of PPAR interactions or in terms of
acetyl]-sn-glycerol (TTA-MAG) 4; plus 2-(tetradec-[7-enyl]thio)acetic acid (dTTA) 5
and corresponding glycerolipids 1,2-di-[2-(tetradec-[7-enyl]thio)acetyl]-sn-glyc-
ero-3-phosphocholine (dTTA-PC) 6, 1-[2-(tetradec-[7-enyl]thio)acetyl]-sn-glycero-
3-phosphocholine (lyso-dTTA-PC) 7, 1-[2-(tetradec-[7-enyl]thio)acetyl]-sn-glycerol
(dTTA-MAG) 8; plus 2-(tetradec-[7-ynyl]thio)acetic acid (tTTA) 9 and corresponding
glycerolipids 1,2-di-[2-(tetradec-[7-ynyl]thio)acetyl]-sn-glycero-3-phosphocholine
(tTTA-PC) 10, 1-[2-(tetradec-[7-ynyl]thio)acetyl]-sn-glycero-3-phosphocholine
(lyso-tTTA-PC) 11, 1-[2-(tetradec-[7-ynyl]thio)acetyl]-sn-glycerol (tTTA-MAG) 12.
promoting TTA
interactions.14
1
bioavailability for intracellular PPAR
One reason for these data could be that TTA 1 functions as any
other FA and is readily incorporated in vivo into glycerolipids, with
a preference for phospholipids over acylglycerols. In fact TTA 1 is
frequently found esterified with cholesterol in lipoprotein parti-
cles.15 Also, during intestinal digestion, lipids are typically metab-
olized in the gut into monoacylglycerols, lysophospholipids, and
free FAs prior to absorption into enterocytes. Here we report on
the next stage of investigation comparing biological effects of
TTA 1 with effects mediated by other THEFAs and corresponding
THEFA-containing lipids. These new THEFAs are prepared with
double or triple bond unsaturations in their hydrocarbon chains.
The reason for such inclusions is as follows. TTA 1 with a double
bond is a metabolite formed in vivo when feeding animals with
this analog.15 Therefore, we speculated that inclusion. of such a
double bond unsaturation might be beneficial to biological effects
by providing an additional structural chemical block to b-oxida-
tion, and potentially enhance PPAR binding in order to promote
PPAR agonism. Furthermore, we speculated that inclusion of a tri-
ple bond might further potentiate PPAR agonism by providing the
chemical structural means to attenuate b-oxidation even more
effectively and enhance PPAR interactions. Here we describe syn-
thetic routes to these THEFAs and for the preparation of
of TTA 1 to b-oxidation, this THEFA has been demonstrated to have
many useful metabolic properties over the past 15 years.1
In mechanistic terms, TTA 1 has been found to act on mitochon-
dria in cells. Broadband pleiotropic biological responses include
increased hepatic and muscle mitochondrial FA oxidation in mam-
mals, reduced body fat (as seen in rats), decreased plasma lipid
levels, modification of plasma and tissue FA profiles, improved
insulin sensitivity, potent antioxidant effects, and anti-inflamma-
tory actions. Underlying many of these effects may be the fact that
TTA 1 is a ligand for peroxisome proliferator-activated receptors
(PPARs). Indeed, post reporter cell transfection, TTA 1 was found
to induce reporter gene activities at concentrations comparable
to those of the known PPARa ligand WY 14.643, but at concentra-
tions markedly lower than required with eicosapentaenoic acid
(EPA) and docosahexaenoic acid (DHA).1 PPARs have for the last
decades been targeted in drug discovery for the treatment of

J. Lund et al. / Bioorg. Med. Chem. 24 (2016) 1191–1203
1193
corresponding THEFA-containing phospholipids and monoacyl-
glycerols (see Fig. 1). Thereafter, we describe the testing of these
in dry tetrahydrofuran (THF), in the presence of potassium t-butox-
ide (KOBut), in order to obtain cis and trans compounds 20 in 50%
and 10% yield, respectively. Product 20 was then combined with
triphenyl phosphine dibromide (PPh3Br2) in dry CH2Cl2 to give
bromo compound 21 that was then coupled with thioglycolic acid
14 to give the desired compound dTTA 5 as white solid in 65% yield
(Scheme 2).
compounds in vitro in cultured human myotubes,
a well-
established cell system for characterization of FA metabolism such
as FA uptake and oxidation.16 Data suggest that lyso-TTA-PC 3 and
triple bond THEFA structures could be useful as next generation
TTA 1 replacements for use in clinical scenarios in due course.
THEFA-containing lipids dTTA-PC 6 and lyso-dTTA-PC 7 were
synthesized by similar methods to 2 and 3. Firstly, dTTA 5 was acti-
vated with DCC and DMAP as coupling reagents in dry CH2Cl2, then
2. Results
coupled to L-a-GPC to give crude product that was purified by flash
2.1. Synthetic chemistry
column chromatography using two different solvent mixtures:
firstly, CH2Cl2/MeOH/H2O (345:90:10 v/v/v); and secondly, CH2Cl2/
MeOH/H2O in the ratio 65:25:4 (v/v/v). The product dTTA-PC 6 was
obtained in 50% yield and lyso-dTTA-PC 7 in 35% yield, both as
white solids (Scheme 2). Finally, dTTA-MAG 8 was prepared as
for 4 by coupling of Solketal to dTTA 5, using HBTU and DMAP as
coupling reagents. The resulting diol-protected ester intermediate
22 that was then subject to TFA deprotection leading to the prepa-
ration of 8 as white solid in 50% yield (Scheme 2).
In the final set of synthetic activities, tTTA 9 was synthesized
using 6-bromohexyl-1-ol 23 as starting material. Initial THP
protection was followed by halogen exchange using NaI in acetone
followed by alkynylation with 1-octyne, in a THF-hexamethylphos-
phoramide (HMPA) solvent mixture, resulting in THP-protected
alkyne intermediate 26. Afterwards 26 was combined with PPh3Br2
to give bromo compound 27 in excellent yield, followed by
coupling with thioglycolic acid 14 to yield tTTA 9 (Scheme 3).
Thereafter, tTTA-PC 10 and lyso-tTTA-PC 11 were synthesized by
similar methods to 2 and 3. In brief, tTTA 9 was activated with
New THEFAs and THEFA-containing glycerolipids (2–12) were
prepared for comparison with TTA 1 as shown (Schemes 1–3).
Following previous precedent, synthesis of TTA 1 followed a
simple SN2 mechanism with tetradecyl bromide 13 as the elec-
trophile and thioglycolic acid 14 as the nucleophile, and was car-
ried out as previously described.17 The synthesis of TTA-PC 2 and
lyso-TTA-PC 3 has also been reported earlier, by activation of TTA
1 using 1,10-carbonyldiimidazole (CDI) prior to combining with
L-
a
-glycerolphosphocholine (
L-a
-GPC).13 This procedure was how-
ever replaced by a one-pot reaction using ester-coupling condi-
tions with dicyclohexyldiimide (DCC) and dimethylamin
opyridine (DMAP) as coupling reagents in dry dimethylformamide
(DMF) to obtain TTA-PC 2 in 50% yield and lyso-TTA-PC 3 in 20%
yield as white solid compounds (Scheme 1). TTA-MAG 4 was
synthesized by means of ester-coupling between TTA 1 and
D
-1,2-isopropylideneglycerol (Solketal), making use of 2-(1H-Ben-
zotriazole-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate
(HBTU) with DMAP as coupling reagents to yield a diol-protected
ester intermediate 15. This was smoothly converted using trifluo-
roacetic acid (TFA) to TTA-MAG 4 recovered as a white solid in
55% yield (Scheme 1).
Synthesis of dTTA 5 was carried out by employing commercially
available 1,7-heptanediol 16 as a starting material, which was
combined with 2,3-dihydro-2H-pyran (DHP) in the presence of
p-toluene sulfonic acid (PTSA) to give a mono pyran-protected 7-
(2-tetrahydropyranyloxy)heptane-1-ol 17. Iodination of 17 by
reaction with iodine led to compound 18, which were treated with
triphenyl phosphine (TPP, PPh3) in the presence of calcium carbon-
ate to give phosphonium salt 19. A highly stereo-selective Wittig
condensation between 19 and heptaldehyde was then performed
DCC and DMAP and coupled to L-a-GPC in dry DMF resulting in
tTTA-PC 10 and lyso-tTTA-PC 11 in 60% and 20% yield, respectively
(Scheme 3). tTTA-MAG 12 was prepared as 4, through coupling
of Solketal to afford 28 using HBTU and DMAP as coupling
reagents, followed by TFA deprotection in methanol, resulting in
tTTA-MAG 12 as a white solid in a yield of 60% (Scheme 3).
2.2. Cellular uptake of THEFAs, lipids and fatty acid composition
in myotubes
The FA composition in cultured myotubes was determined after
incubating the cells for 96 h with a selection of seven THEFAs and
O
HS
+
( )11
13
Br
OH
14
i
O
S
OH
1
ii
iii
O
O
O
S
O
O
O
O
S
S
S
( )11
P
N
( )11
O
O
( )11
O
15
4
O
O
iv
2
O
+
O
O
O
3
O
O
S
P
N
( )11
O
OH
( )11
O
OH
OH
Scheme 1. Reagents and conditions: (i) NaOH/MeOH, rt, 72 h, 75%; (ii)
19 h, 80%; (iv) TFA, MeOH, at 0 °C–rt, 30 min, 55%.
L-a-GPC, DCC, DMAP, dry DMF, N2, rt, 16 h, 2 50%, 3 20%; (iii) HBTU, DMAP, Solketal, dry CH2Cl2, rt,

1194
J. Lund et al. / Bioorg. Med. Chem. 24 (2016) 1191–1203
ii
i
( )5
16
( )5
iii
HO
OH
I
OTHP
18
HO
( )5
17
OTHP
iv
I Ph3P
( )5
19
OTHP
OTHP
Cis 50 % + Trans 10%
20
v
O
Br
+
HS
OH
OH
21
14
vi
O
S
5
vii
viii
O
O
O
O
O
O
S
P
P
N
N
S
ix
( )4
( )3
O
( )4
( )3
O
O
O
O
22
( )3
S
( )4
O
6
+
O
O
O
O
O
O
S
S
( )4
( )3
O
OH
( )4
( )3
O
OH
7
OH
8
Scheme 2. Reagents and conditions: (i) DHP, PTSA, dry CH2Cl2, rt, 24 h, 80%; (ii) PPh3, imidazole, dry THF, iodine, –10 °C–rt, 30 min, 75%; (iii) PPh3, CaCO3, CH3CN, reflux 85 °C,
22 h, 93%; (iv) KOBut, 1-heptaldehyde, dry THF, ꢀ5 °C, 3 h, rt, 5 h, cis 50%, trans 10%; (v) PPh3Br2, PPh3, dry CH2Cl2, 0 °C, 15 min, 90%; (vi) 25% NaOH/MeOH, rt, 72 h, 65%; (vii)
L-
a
-GPC, DCC, DMAP, dry CH2Cl2, N2, rt, 16 h, 6: 50%, 7: 35%; (viii) HBTU, DMAP, Solketal, dry CH2Cl2, rt, 19 h, 83%; (ix) TFA, MeOH, at 0 °C–rt, 30 min, 50%.
THEFA-containing lipids (1, 2, 3, 4, 6, 9, 11, syntheses of which are
described above). After harvesting and lipid extraction of the cells,
levels of recovered FAs were determined as described in the exper-
imental section. Two samples with dimethyl sulphoxide (DMSO)
(0.1%) but no THEFA added were also analyzed as controls. The
levels of free TTA 1, dTTA 5 or tTTA 9 were used as a measure for
the cellular uptake of THEFAs or corresponding THEFA-containing
lipids from the medium (Table 1). The uptake of THEFAs or
THEFA-containing lipids by the myotubes was found in each case
to be mainly concentration-dependent, thus 3-fold increases in
extracellular THEFA or THEFA-containing lipid concentrations,
TTA-MAG 4 resulted in similar SFA decreases although levels of
incorporation were generally lower and peaked at a 38% decrease.
In contrast, treatment with THEFAs and THEFA-containing lipids
with double and triple bonds resulted in minimal effects on rela-
tive SFA levels, but dTTA-PC 6, tTTA 9 and lyso-tTTA-PC 11 treat-
ment did impact on the relative MUFA levels in a concentration-
dependent manner (up to 21%). Finally, treatment with TTA 1,
lyso-TTA-PC 3 and TTA-MAG 4 also resulted in comparable reduc-
tions in levels of MUFAs. Cell protein levels and total fatty acid
levels per mg of protein were constant between controls, 10
lM
and 30 M of analogs, indicating no toxic effect on the skeletal
l
from 10 to 30
l
M, often resulted in approximately 3-fold increases
muscle cells.
in the measured intracellular levels of corresponding THEFAs in
myotubes. In this, TTA 1 was the most bioavailable of all the THE-
FAs or THEFA-containing lipids administered, while uptake of the
TTA-containing lipids lyso-TTA-PC 3 and TTA-MAG 4 resulted in
intracellular levels of TTA 1 30–40% lower than those same levels
observed following treatment with equimolar amounts of TTA 1
(Table 1). Similarly, uptake of tTTA 9 also resulted in lower
intracellular levels of tTTA 9 than those levels observed following
treatment with equimolar amounts of TTA 1 (Table 1). In contrast,
a 3-fold increase in extracellular lyso-tTTA-PC 11 concentration
only resulted in a doubling of the intracellular levels of the corre-
sponding tTTA 9, while a tripling of the extracellular concentration
of TTA-PC 2 resulted in no significant increase in the corresponding
intracellular levels of TTA 1.
2.3. Fatty acid oxidation
Myotube mediated uptake of [14C]oleic acid (OA) and oxida-
tive capacity of cells for FAs in the myotubes were assessed dur-
ing 4 h incubation after pre-incubation of cells for 96 h with 10
and 30 lM THEFAs and THEFA-containing lipids or 10 nM of the
PPARd agonist GW501516 as positive control. For negative con-
trol purposes, myotubes were treated with DMSO (0.1%). Initially,
we noted that the extent of intracellular [14C]OA uptake was
unaffected by extracellular concentration rises of 10–30 lM in
extracellular THEFAs and THEFA-containing lipid concentrations
(Fig. 2A). On the other hand, [14C]OA oxidation was increased in
myotubes in
a dose-dependent manner following treatment
The FA composition in the myotubes was further analyzed and
compared with control myotube cultures in which no THEFAs or
THEFA-containing lipids were added (Table 1). The main impact
of the treatment with THEFAs or THEFA-containing lipids was a
decrease in the relative content of saturated fatty acids (SFAs)
and monounsaturated fatty acids (MUFAs), mainly oleic acid. The
largest decrease in SFAs (up to 50%) was observed in the myotubes
with TTA 1, lyso-TTA-PC 3, TTA-MAG 4 and lyso-tTTA-PC 11
(Fig. 2B).
In particular, we observed that TTA 1 treatment (30 lM)
resulted in a nearly 3-fold increase in [14C]OA oxidation above
control, in keeping with previous observations.18 Furthermore,
treatment with lyso-TTA-PC 3 and TTA-MAG 4 resulted in [14C]
OA oxidation levels approximately 2-fold higher than those
incubated with 30
lM TTA 1. Incubation with lyso-TTA-PC 3 and
levels observed following treatment with an equimolar (30 lM)

J. Lund et al. / Bioorg. Med. Chem. 24 (2016) 1191–1203
1195
i
( )
Br
OH
( )
Br
OTHP
OTHP
4
4
23
24
ii
( )
I
4
25
iii
OTHP
26
iv
Br
O
27
v
S
OH
9
vii
vi
O
O
O
O
S
S
P
N
( )4
( )3
O
O
O
( )4
( )3
O
O
O
10
O
28
12
O
O
( )4
( )3
S
viii
O
+
O
O
O
O
S
S
P
N
( )4
( )3
OH
( )4
( )3
O
O
OH
OH
11
Scheme 3. Reagents and conditions: (i) DHP, PTSA, dry CH2Cl2, rt, 16 h, 95%; (ii) NaI, acetone, reflux, 15 min, 89%; (iii) 1-octyne, nBuLi, dry THF, HMPA, N2, 0 °C–rt, 50 h, 72%;
(iv) PPh3Br2, PPh3, dry CH2Cl2, N2, 0 °C, 15 min, 90%; (v) thioglycolic acid, NaOH/MeOH, rt, 72 h, 85%; (vi)
HBTU, DMAP, Solketal, dry CH2Cl2, rt, 19 h, 81%; (viii) TFA, MeOH, 0 °C–rt, 15 min, 60%.
L-a-GPC, DCC, DMAP, dry DMF, N2, rt, 16 h, 10: 60%, 11: 20%; (vii)
amount of TTA 1. This outcome is also in keeping with previous
data showing lyso-TTA-PC 3 to be more efficient as a PPAR ago-
nist relative to TTA 1.14 This efficacy difference is made more
striking considering that the measured intracellular levels of
TTA 1 in myotubes following administration of TTA 1 alone were
30–40% higher than the intracellular levels of TTA 1 observed
following administration of either lyso-TTA-PC 3 or TTA-MAG 4
(Table 1). Furthermore, administration of lyso-tTTA-PC 11 was
also able to increase [14C]OA oxidation to levels approximately
2-fold higher than those levels observed following treatment
2.4. Increased expression of PPAR regulated genes
The impact of THEFAs and THEFA-containing lipids on regulated
gene expression was investigated when myotubes were incubated
for 96 h with THEFAs and THEFA-containing lipids (30 lM) fol-
lowed by qPCR of selected genes. Since TTA 1 is a known PPAR ago-
nist, genes that are PPAR regulated were primarily selected for
study.16 The genes that were examined were angiopoietin-like 4
(ANGPTL4), fatty acid binding protein 3 (FABP3), fatty acid translo-
case (CD36) and carnitine palmitoyltransferase 1A (CPT1A). The
encoded protein corresponding with ANGPTL4 functions as a hor-
mone that regulates glucose homeostasis and lipid metabolism.19
The FABP3 gene encodes the fatty acid-binding protein found in
liver. The encoded protein corresponding with CD36 is a FA trans-
porter involved with long-chain FA uptake as well as signaling.20
The CPT1A gene encodes a mitochondrial enzyme responsible for
the formation of acyl carnitines.21 Within experimental error,
administration of THEFAs and THEFA-containing lipids all induced
broadly similar levels of gene activation with the exception of TTA-
PC 2 (Fig. 3A, B and D). On closer examination, ANGPTL4 was
induced the strongest, up to 35-fold over the control by tTTA 9
(Fig. 3A), followed by up to 25-fold over control by lyso-tTTA-LPC
11 and up to 20-fold over control by positive control TTA 1. Simi-
larly the CPT1A gene was induced the strongest up to 10-fold over
the control by tTTA 9 (Fig. 3D), followed by up to 7-fold over con-
trol by lyso-tTTA-PC 11 and up to approximately 5-fold over
with an equimolar (30 lM) amount of TTA 1. This 2-fold
enhancement was also all the more striking given that the intra-
cellular levels of tTTA 9 post administration of lyso-tTTA-PC 11
were approximately 20% or less than the corresponding intracel-
lular levels of TTA 1 in myotubes following administration of
TTA 1 alone (Table 1). One surprising result here was that
TTA-PC 2 was found to be less effective at stimulating lipid oxi-
dation than the other glycerolipids in spite of the significantly
greater efficacy of TTA-PC 2 in comparison to TTA 1 when used
in vivo.13 This lack of efficacy may be related to the fact that
extent of cellular uptake of TTA-PC 2 was unexpectedly low.
Indeed the intracellular levels of TTA 1 following administration
of TTA-PC 2 were only a fraction of the corresponding intracellu-
lar levels of TTA 1 observed in myotubes following administra-
tion of TTA 1 alone at the same extracellular concentration
(Table 1).

1196
J. Lund et al. / Bioorg. Med. Chem. 24 (2016) 1191–1203
control by the positive control TTA 1. On the other hand, adminis-
tration of THEFAs and THEFA-containing lipids with double and
triple bonds did appear to modestly but significantly stimulate
CD36 gene expression compared with TTA 1 and other TTA-con-
taining lipids (Fig. 3C). Curiously, FABP3 was induced best by
TTA-MAG 4 followed by lyso-tTTA-PC 11 (Fig. 3B).
These stimulatory effects of tTTA 9 and lyso-tTTA-LPC 11 are
particularly interesting given previous observations that measured
intracellular levels of tTTA 9 post administration of either tTTA 9 or
lyso-tTTA-PC 11 were less than 20% of the corresponding level of
TTA 1 in myotubes following administration of TTA 1 alone at
the same extracellular concentration (Table 1). Overall, the data
shown (Figs. 2 and 3) do provide some support for the notion that
tTTA 9 and other tTTA containing lipids could surpass TTA 1 in
their capacities to stimulate FA oxidation and to some measure-
able extent expression of the PPAR-associated genes ANGPTL4,
CPT1A and CD36.
3. Discussion
The aim of this study was to prepare new and novel THEFAs and
THEFA-containing lipids whose biological properties could be
compared against TTA 1, the parent THEFA. The clinical potential
and limitations of TTA 1 are outlined in Section 1, therefore our
hope was to identify alternative THEFAs and/or THEFA-containing
lipids that might be useful as next generation TTA 1 replacements
for use in future potential clinical scenarios.
Of the THEFA and THEFA-containing lipids prepared and stud-
ied here, lyso-TTA-PC 3, TTA-MAG 4, tTTA 9, and lyso-tTTA-PC 11
were all able to increase OA oxidation levels the highest, approxi-
mately 4–5-fold above the positive PPARd control GW501516, and
nearly 2-fold over TTA 1. The efficacies of these compounds were
all the more remarkable given that the measured intracellular
levels of TTA 1 or tTTA 9 as appropriate in cells post administration
were consistently at least 30–40% lower than levels seen post
administration of equimolar amounts of TTA 1 alone. These obser-
vations are consistent with previous conclusions that THEFA bio-
logical effects should be better realized through administration
of THEFA-containing lipids in place of free-FA THEFAs (Table 1
and Fig. 2).13,14 Furthermore, these data also suggest that the intro-
duction of a triple bond into the TTA 1 structure may promote
PPAR agonist effects compared with TTA 1 alone. On the other
hand, the introduction of a double bond into the TTA 1 structure
suggests the opposite given the weaker performance of dTTA-con-
taining lipids under investigation (Table 1 and Fig. 2). Finally the
beneficial effects for PPAR agonism of introducing a triple bond
were further supported by the apparent increases in mRNA
expression of ANGPTL4, CPT1A and CD36 induced by administration
of either tTTA 9 or lyso-tTTA-PC 11 in comparison to effects
observed post administration of other THEFA and THEFA-contain-
ing lipids (Fig. 3).
Intriguingly, administration of THEFA and THEFA-containing
lipids was found to remodel FA distributions in myotubes (Table 1).
The competition by THEFAs like TTA 1 was particularly strong
towards the SFAs, which were decreased by up to 50% after incu-
bation at the highest concentration. The replacement of MUFA
was less dependent on the cellular TTA content. The unsaturated
FAs dTTA 6 and tTTA 9 did not compete with the endogenous SFAs,
but mainly replaced MUFA, decreasing the MUFA content by up to
20% at the highest concentrations used. Compared to the effect of
TTA 1 this seems reasonable since the intracellular levels of the
unsaturated THEFAs were markedly lower than those of TTA 1.
Finally, why might the inclusion of THEFAs into glycerolipid
structures result in more pronounced metabolic effects than those
effects elicited by treatment with free THEFAs? Certainly, what is

J. Lund et al. / Bioorg. Med. Chem. 24 (2016) 1191–1203
1197
Figure 2. Oleic acid uptake and oxidation. (A) Cellular uptake (CO2 + CA) and (B) oxidation (CO2) from [14C]oleic acid (OA) after 4 h incubation of human myotubes treated for
96 h with THEFAs, THEFA-containing lipids or 10 nM GW501516 as indicated. Concentrations of THEFAs and intracellular concentrations of THEFAs are as given in Table 1. For
negative control purposes, myotubes were treated with DMSO (0.1%). Values are given as percentage relative to basal uptake or oxidation of OA (100%) in DMSO-treated
myotubes. Number of experiments (n) = 8–15. Range control values: OA uptake: 16.4–165.8 nmol/mg protein; OA oxidation: 3.4–46.0 nmol/mg protein.
Figure 3. Gene expression. Gene expression of (A) ANGPTL4, (B) FABP3, (C) CD36 and (D) CPT1A relative to DMSO (0.1%) as control in human myotubes incubated for 96 h with
THEFAs and THEFA-containing lipids (30 lM). Number of experiments (n) = 3.
known is that TTA 1 is found endogenously esterified into glyc-
erolipids, triacylglycerols and cholesteryl esters. Therefore, it
would be logical to impute that primary PPAR ligands might be
glycerolipids in which a THEFA is located in at least one of the ester
positions. It is reasonable to assume that when incubating the
myotubes with esterified THEFAs, a higher cellular level of the
actual PPAR ligand will be achieved compared to incubating the
cells with free THEFAs. Consistent with this statement, data have

1198
J. Lund et al. / Bioorg. Med. Chem. 24 (2016) 1191–1203
been produced to show that lyso-glycerolipids can be endogenous
activators of nuclear receptors (NRs).22,23 Indeed, four NRs have so
far been identified as PL binding proteins: the steroidogenic factor-
like NRs liver receptor homolog 1 (LRH-1) and steroidogenic factor
1 (SF-1), PPARa, and ultraspiracle (insect homolog of retinoid X
receptor). Additionally, a family of PL transporters that stimulate
dC = 77.00 ppm; CD3OD, dC = 49.05 ppm). Data are reported as fol-
lows: C, CH, CH2, CH3; d = doublet, q = quadruplet; coupling con-
stant(s) J in Hz to the nearest 0.5 Hz; assignment. 13C NMR data
were only fully assigned by correlation between 1H and 13C NMR
spectral data. Mass spectra were recorded using VG Platform II,
VG-070B, Joel SX- 102 or Bruker Esquire 3000 ESI instruments.
Mass accuracy is indicated to the nearest 0.1 ppm.
NR transactivation has been identified and shown to up-regulate
the transcriptional activity of both PPAR
a and hepatocyte nuclear
factor 4
a
.
24,25 Additionally, several groups have also demonstrated
4.1.1. 2-(Tetradecylthio)acetic acid (TTA) 1
that the mammalian NRs SF-1 and LRH-1 interact with phos-
phatidylcholines, phosphatidylethanolamines, phosphatidylglyc-
erols, and perhaps even phosphatidylinositol-phosphates. Indeed,
in one study mass spectrometry experiments identified phos-
Tetradecylbromide 13 (5.36 mL, 18.0 mmol, 1.00 equiv) was
added to a solution of thioglycolic acid 14 (1.25 mL, 18.0 mmol,
1.00 equiv) in 25% NaOH in MeOH (6.03 g in 24.0 mL) and the mix-
ture was stirred vigorously at room temperature (rt) for 72 h. After
dilution with water, the mixture was acidified to pH 1 with con-
centrated HCl, and the aqueous phase was extracted with diethyl
ether. The ether phase was dried over MgSO4, filtered and concen-
trated in vacuo to give a crude white solid that was further purified
by silica gel (60–120) flash column chromatography, eluting with
CH2Cl2. Desired compound 1 was obtained as an amorphous solid
(3.86 g, 75%), TLC: Rf 0.50 (hexane/EtOAc 3:1 v/v). 1H NMR
(400 MHz, CDCl3): d 3.27 (s, 2H, Ha), 2.68–2.65 (t, 2H, J = 7.5 Hz,
Hb), 1.66–1.58 (m, 2H, Hc), 1.41–1.36 (m, 2H, Hd), 1.27 (s, 20H,
10ꢁCH2), 0.908–0.874 (t, 3H, J = 7.0 Hz, CH3); (EI+) m/z 289 [M
+H]+.
phatidylcholine 16:0/18:1 as one of several lipids bound to PPAR
isolated from murine liver tissue.26 Binding of this PC species was
selective for PPAR over PPARd and PPAR , and could be inhibited
by treatment with a standard PPAR agonist (WY14.643). In a
comparable way, studies in Chinese Hamster Ovary cells have been
a
a
c
a
used to demonstrate that PPARc can be activated by lysophospha-
tidic acid that is generated intracellularly.27
The data obtained in this study regarding THEFAs are thus con-
sistent with the concept of glycerolipids and lyso-glycerolipids and
similar molecules as endogenous PPAR ligands. Hence, although
our current data supports a potential importance for lyso-TTA-PC
3 and lyso-tTTA-PC 11 as significant lipid mediators of TTA 1 and
tTTA 9 effects respectively, THEFA-containing lipids such as these
are likely to be transferred and become integrated into other glyc-
erolipid species too, owing to transesterification reactions. Owing
to the small number of genes studied other PPAR regulated genes
will have to be studied to confirm this conclusion. Further detailed
lipidomics type studies would be very helpful to confirm not only
the functional importance of THEFA-containing lysophospholipids
to PPAR agonism, but also to identify other potentially important
PPAR lipid agonists deriving from THEFAs by esterification fol-
lowed by transesterification, giving rise to next generations of
TTA 1 replacements for potential use in clinical scenarios.
4.1.2. 1,2-Di-[2-(tetradecylthio)acetyl]-sn-glycero-3-
phosphocholine (TTA-PC) 2
To a solution of TTA 1 (261 mg, 0.905 mmol, 1.00 equiv) in dry
DMF (5 mL), and DCC (190 mg, 1.17 mmol, 1.30 equiv), DMAP
(135
lL, 0.905 mmol, 1.00 equiv) and
L-a-GPC (133 mg,
0.301 mmol, 0.33 equiv; 5 mL) were added. The mixture was stir-
red vigorously at rt under N2 for 16 h. A clear yellow solution
was obtained. After completion of the reaction the DMF was
removed under vacuo, the residue was purified by silica gel (60–
120) flash column chromatography, eluting with solvent mix A fol-
lowed by solvent mix B, to yield TTA-PC 2 (120 mg, 50%) and lyso-
TTA-PC 3 (32 mg, 20%) as white amorphous solids, TLC: Rf 0.40 [sol-
vent B]. 1H NMR (400 MHz, CDCl3): d 5.25 (br, 1H, H2), 4.45–4.41
(dd, 2H, J = 12.0 Hz, J = 3.0 Hz, H1), 4.28–4.23 (m, 2H, 2ꢁH3),
4.00–3.97 (t, 2H, J = 6.0 Hz, H4), 3.85–3.74 (m, 2H, H5), 3.31–3.28
(br, 13H, 3ꢁNCH3, Ha), 2.63–2.58 (m, 4H, Hb), 1.61–1.54 (m, 4H,
Hc), 1.38–1.35 (m, 2H, Hd), 1.26 (s, 40H, 20ꢁCH2), 0.85–0.82 (t,
6H, J = 7.0 Hz, 2ꢁCH3), 13C (125 MHz; CDCl3) 170.36, 170.12 (C,
2ꢁC@O), 71.47–71.41 (CH, C2), 66.30–66.25 (CH2, C5), 63.49
(CH2, C1), 63.33–63.29 (CH2, C3), 59.45–59.41 (CH2, C4), 54.37
(3ꢁNCH3), 33.60, 33.45 (CH2, 2ꢁCa), 32.74, 32.66 (CH2, 2ꢁCb),
31.90 (CH2, 2ꢁCc), 29.66, 29.62, 29.60, 29.35, 29.32, 29.02, 29.00,
28.85, 28.82 (18ꢁCH2), 22.66 (2ꢁCH2), 14.09 (2ꢁCH3); LC–MS,
m/z 798 [M+H]+.
4. Experimental section
4.1. Synthetic chemistry
All reactions were carried out under an atmosphere of nitrogen
(N2) or argon, in oven-dried glassware, unless otherwise stated.
CH2Cl2 was distilled over P2O5, and other solvents were bought
and pre-dried as required. All chemicals were purchased from
Sigma–Aldrich. Flash column chromatography was performed on
silica gel 60 (Merck Kieselgel 60 F254 230–240 mesh) according
to the method of Still.28 Thin layer chromatography (TLC) was per-
formed on pre-coated Merck silica gel (0.2 mm, 60 F254) alu-
minum-backed plates, and visualized with a UV lamp (254 nm)
and/or stained with acidic ammonium molybdate, basic potassium
manganate (KMnO4), iodine and phosphomolybdic acid. Special
chromatography solvent mixtures are: solvent mix A: CH2Cl2/
MeOH/H2O (345:90:10 v/v/v); solvent mix B: CH2Cl2/MeOH/H2O
(65:25:4 v/v/). 1H NMR and 13C NMR spectra were recorded on
Bruker Avance 400. 1H NMR was recorded at 400 MHz and chem-
ical shifts, dH are quoted in parts per million (ppm), using residual
isotopic solvent as internal reference (CDCl3, dH = 7.27 ppm;
CD3OD, dH = 3.30 ppm). Data are reported as follows: integration;
br = broad; s = singlet, d = doublet, t = triplet, q = quartet, m = mul-
tiplet; coupling constant(s) J in Hz to the nearest 0.5 Hz; assign-
ment. Peaks split by the presence of a phosphorus atom are
indicated with a superscript p. 13C NMR spectra were recorded at
100 MHz and chemical shifts, dC are quoted in parts per million
(ppm), using residual isotopic solvent as internal reference (CDCl3,
4.1.3. 1-[2-(Tetradecylthio)acetyl]-sn-glycero-3-phosphocholine
(lyso-TTA-PC) 3
1H NMR (400 MHz, CDCl3/MeOD 2:1): d 4.28–4.11 (m, 3H,
2ꢁH1, H2, 1ꢁH3), 3.95–3.86 (m, 3H, 2ꢁH4, 1ꢁH3), 3.59 (br, 2H,
2ꢁH5), 3.23–3.12 (m, 11H, 2ꢁHa, 3ꢁNCH3), 2.60–2.56 (t, 2H, J
7.5, 2ꢁHb), 1.58–1.51 (m, 2H, 2ꢁHc), 1.34–1.32 (m, 2H, 2ꢁHd),
1.21 (s, 20H, 10ꢁCH2), 0.85–0.81 (t, 3H, J 7.0, CH3), 13C NMR
(125 MHz, CDCl3) d 170.60 (C, C@O), 68.55 (CH, C2), 67.00 (CH2,
C5), 66.08 (CH2, C3), 59.45 (CH2, C1), 54.52 (3ꢁNCH3), 40.06,
(CH2), 33.62 (CH2, C3), 32.69 (CH2, 2ꢁCb), 31.89 (CH2, 2ꢁCc),
29.68, 29.64, 29.58, 29.46, 29.33, 29.29, 29.14, 29.03, 28.81,
(10ꢁCH2), 22.65 (2ꢁCH2), 16.30 (CH3); LC–MS, m/z 528 [M+H]+.
4.1.4. 2,3-Acetonide-1-[2-(tetradecylthio)acetyl]-sn-glycerol 15
HBTU (1.16 g, 3.05 mmol, 3.28 equiv) and DMAP (1.09 g,
8.95 mmol, 9.64 equiv) were added to
a solution of TTA 1

J. Lund et al. / Bioorg. Med. Chem. 24 (2016) 1191–1203
1199
(834.0 mg, 2.89 mmol, 3.11 equiv) and Solketal (85.5 mg,
0.928 mmol, 1.00 equiv) in dry CH2Cl2 under argon. The mixture
was stirred at rt for 19 h, and then the reaction was quenched with
15 mL of citric acid solution (7%). The aqueous phase was separated
from the organic phase and extracted with CH2Cl2 (4 ꢁ 25 mL). The
organic phases were then combined, dried over MgSO4, filtered and
concentrated in vacuo to give a white solid crude (2.49 g). The
crude solid was further purified by silica gel (60–120) flash column
chromatography, eluting with hexane/EtOAc (9:1, 8:2, 6:4 v/v), to
yield 15 as a white amorphous powder (297 mg, 80%), TLC: Rf
0.40 (hexane/EtOAc 9:1 v/v). 1H NMR (400 MHz, CDCl3): d 4.42–
4.36 (m, 1H), 4.35–4.18 (m, 2H), 4.17–4.09 (m, 1H), 3.80–3.76
(m, 1H), 3.27 (s, 2H), 2.67–2.63 (t, 2H), 1.64–1.57 (m, 2H), 1.46
(s, 3H), 1.39 (s, 3H), 1.27 (s, 20H), 0.91–0.88 (t, 3H); LC–MS, m/z
423 [M+Na]+.
100 mL of acetonitrile was heated at 85 °C for 22 h. Then the reac-
tion mixture was filtered and concentrated under reduced pressure
to obtain a gum-like residue that was dissolved in a minimum
amount of CH2Cl2. Addition of diethyl ether resulted in a precipi-
tate that was collected by filtration then dried under vacuum
(0.05 mmHg) at rt to give phosphonium salt 19 as a gum (19.9 g,
92.6%).
4.1.9. 2-Tetradec-(7Z)-enyloxy-tetrahydropyran 20
To a solution of 19 (12.91 g, 31.97 mmol) in dry THF (60 mL),
KOBut (2.98 g, 26.62 mmol) was added at ꢀ5 °C and stirred for
3 h at the same temperature. A solution of heptaldehyde (4.7 g,
10.65 mmol) in THF (15 mL) was added dropwise and stirred at
rt for 5 h. Saturated aqueous NH4Cl solution (30 mL) was added
and extracted with EtOAc (3 ꢁ 40 mL). The organic layer was
washed with water (3 ꢁ 50 mL), brine (3 ꢁ 50 mL), dried over
MgSO4, filtered and evaporated under reduced pressure to give a
liquid (crude weight 19 g). This liquid was purified further by silica
gel (60–120) column chromatography, eluting with EtOAc/hexane
(1:9 v/v) to yield 20 as an oil (4.74, 50%), TLC: Rf 0.65 (EtOAc/hex-
ane 1:9 v/v). 1H NMR (400 MHz, CDCl3): d 5.32–5.38 (m, 2H), 4.58–
4.60 (m, 2H), 3.86–3.89 (m, 1H), 3.50–3.53 (m, 2H), 3.40–3.49 (m,
2H), 2.01 (m, 4H), 1.71–1.83 (m, 6H), 1.44–1.59 (t, 8H), 1.21–1.31
(m, 6H), 0.73–0.85 (m, 3H); ESI, m/z 297 [M+H]+.
4.1.5. 1-[2-(Tetradecylthio)acetyl]-sn-glycerol (TTA-MAG) 4
To a solution of 15 (2.59 g, 5.39 mmol) in MeOH (15 mL) at 0 °C,
TFA (1 mL) was added. The reaction was stirred at rt for 30 min
before being neutralized with concentrated ammonium hydroxide.
The solvent was removed under reduced pressure to obtain a gum-
like residue. This residue was purified by silica gel (60–120) col-
umn chromatography, eluting with EtOAc/hexane (6:4 v/v), to
yield 4 was obtained as a white solid (1.1 g, 55%). 1H NMR
(400 MHz, CDCl3): d 4.42–4.36 (m, 1H), 4.35–4.18 (m, 2H), 4.17–
4.09 (m, 1H), 3.80–3.76 (m, 1H), 3.27 (s, 2H), 2.67–2.63 (t, 2H),
1.64–1.57 (m, 2H), 1.46 (s, 3H), 1.39 (s, 3H), 1.27 (s, 20H), 0.91–
0.88 (t, 3H); LC–MS, m/z 384 [M+Na]+.
4.1.10. 1-Bromo-tetradec-7-ene 21
To a stirred solution of dibromo triphenylphosphine (PPh3Br2;
5.87 g, 13.9 mmol) and PPh3 (1.41 g, 5.36 mmol) in anhydrous CH2
Cl2 (150 mL) at 0 °C under N2, 20 (3.61 g, 10.7 mmol) was added.
After 15 min, excess PPh3Br2 was quenched with 10% K2CO3 solu-
tion (100 mL). The organic phase was separated, and the aqueous
phase was re-extracted with CH2Cl2 (3 ꢁ 150 mL). The combined
organic layers were washed with water (150 mL), brine (150 mL),
dried over MgSO4, filtered, concentrated and obtained as light yel-
low oily liquid (crude weight 6.5 g). This oily liquid was purified by
silica gel (60–120) column chromatography, eluting with EtOAc/
hexane (1:9 v/v), to afford 21 as a pale yellow oil (2.64 g, 90%),
TLC: Rf 0.68 (hexane/EtOAc 9:2 v/v). 1H NMR (400 MHz, CDCl3): d
5.34–5.42 (m, 2H), 3.35–3.45 (t, 2H), 2.06–2.29 (m, 4H), 1.76–
1.90 (m, 4H), 1.37–1.50 (m, 4H), 1.20–1.33 (m, 8H), 0.88–1.01
(m, 3H); ESI, m/z 275 [M+H]+.
4.1.6. 7-(2-Tetrahydropyranyloxy)heptane-1-ol 17
To 4.00 g of 1,7-heptanediol 16 in 30 mL of CH2Cl2, DHP
(35.85 mmol) and catalytic amount of PTSA were added dropwise.
The reaction was stirred for 24 h at rt. After completion of the reac-
tion the organic layer was washed with water (2 ꢁ 50 mL), NaHCO3
(3 ꢁ 50 mL), dried over MgSO4, filtered, and evaporated in vacuo.
The resulting crude oil (4.5 g) was purified by silica gel (60–120) col-
umn chromatography, eluting with EtOAc/hexane (7:3 v/v) to yield
17 as an off-white liquid (5.18 g, 80%). 1H NMR (400 MHz, CDCl3): d
4.42–4.36 (m, 2H), 4.58–4.59 (m, 2H), 3.80–3.91 (m, 3H), 3.723–3.36
(m, 1H), 3.37–3.40 (m, 2H), 3.18–3.20 (t, 2H), 178–1.85 (m, 2H),
1.50–1.61 (m, 4H), 1.39–1.41 (m, 6H); ESI, m/z 217 [M+H]+.
4.1.7. 1-Iodo-7-(2-tetrahydropyranyloxy)heptane 18
4.1.11. 2-(Tetradec-[7-enyl]thio)acetic acid (dTTA) 5
To a solution containing triphenylphosphine (PPh3; 20.1 g,
76.3 mmol), imidazole (10.34 g, 152 mmol) and 17 (10.67,
50.8 mmol) in 140 mL of THF, iodine (19.36 g, 76.3 mmol) was
added at -10 °C. The solution was allowed to warm up to rt and
then stirred for 30 min. The reaction mixture was diluted with
diethyl ether, washed successively with a saturated solution of
sodium thiosulfate (100 mL) and a saturated solution of NaHCO3
(100 mL), dried over MgSO4, filtered, and evaporated to dryness.
The residue was stirred with petroleum ether to precipitate triph-
enylphosphine oxide, then filtered and concentrated under
Compound 21 (5.36 mL, 18.0 mmol, 1.00 equiv) was added to a
solution of thioglycolic acid (1.25 mL, 18.0 mmol, 1.00 equiv) in
25% NaOH in MeOH (6.03 g in 24 mL) and the mixture was stirred
vigorously at rt for 72 h. After dilution with water, the mixture was
acidified to pH 1 with concentrated HCl, and the aqueous phase
was extracted with diethyl ether. The ether phase was dried over
MgSO4, filtered and concentrated in vacuo to obtain a white solid,
that was purified by silica gel (60–120) flash column chromatogra-
phy, eluting with CH2Cl2, to yield 5 as an amorphous solid com-
pound (3.36 g, 65%), TLC: Rf 0.50 (EtOAc/hexane 1:3 v/v). 1H NMR
(400 MHz, CDCl3): d 5.35–5.40 (m, 2H), 3.38 (s, 2H), 2.67 (t, 2H),
2.03–2.10 (m, 4H), 1.60–1.64 (m, 8H), 1.17–1.38 (m, 8H), 0.73–
0.89 (t, 3H); 13C (100 MHz, CDCl3), 176.87 (C, C@O), 130.08,
129.56 (HC@CH), 77.00, 33.43 (CH2, Ca), 32.73 (CH2, Cb), 31.75
(CH2, Cc), 29.69, 29.54, 28.95, 28.81, 28.77, 28.59, 27.19, 27.06,
22.63 (9ꢁCH2), 14.08 (CH3); LC–MS, m/z 285 [M+H]+.
reduced pressure to give
a crude oily liquid (crude weight
28.5 g). This oily liquid that was further purified by silica gel
(60–120) column chromatography, eluting with EtOAc/hexane
(3:97 v/v), to yield 18 as a clear oil (12.4 g, 75%), TLC: Rf 0.43 (hex-
ane/EtOAc 9:1 v/v). 1H NMR (CDCl3, 400 MHz): d 4.6 (m, 2H), (m,
2H), 3.85–3.90 (m, 2H), 3.49–3.57 (m, 1H), 37–3.41 (m, 2H), 3.32
(t, 2H), 1.53–1.60 (t, 2H), 1.39–1.41 (m, 4H), 1.36–1.38 (m, 6H);
ESI, m/z 327 [M+H]+.
4.1.12. 1,2-Di-[2-(tetradec-[7-enyl]thio)acetyl]-sn-glycero-3-
phosphocholine (dTTA-PC) 6
4.1.8. (7-[2-Tetrahydropyranyloxy]heptyl)triphenyl phosphonium
iodide 19
To a solution of 5 (261 mg, 0.905 mmol, 1.00 equiv) in CH2Cl2
(5 mL), DCC (190 mg, 1.17 mmol, 1.30 equiv), DMAP (135
0.905 mmol, 1.00 equiv) and -GPC (133 mg, 0.301 mmol,
0.33 equiv; 5 mL) were added. The mixture was stirred vigorously
lL,
A
solution containing triphenylphosphine (PPh3; 19.3 g,
L-a
73.6 mmol), 18 (12 g, 36.7 mmol) and calcium carbonate (2 g) in

1200
J. Lund et al. / Bioorg. Med. Chem. 24 (2016) 1191–1203
at rt under N2 for 16 h. A clear yellow solution was obtained. After
completion of the reaction the DMF was removed under vacuo and
the residue purified by silica gel (60–120) flash column chromatog-
raphy, eluting with solvent mix A followed by solvent mix B, to
yield 6 (119 mg, 50%) and lyso-dTTA-PC 7 (55 mg, 35%) as white
amorphous solids, Rf 0.40 (solvent B). 1H NMR (400 MHz, CDCl3):
d 5.32–5.40 (m, 4H), 4.39–4.45 (m, 4H), 4.05 (m, 2H), 3.89 (m,
2H), 3.42 (s, 4H), 3.25 (t, 8H), 2.61–2.65 (m, 10H), 2.03–2.04 (m,
12H), 1.57 (m, 4H), 1.30–1.36 (m, 20H), 0.90 (t, 6H), 13C
(100 MHz, CDCl3) d 170.13, 169.88 (c, C@O), 130.37, 129.89,
129.38 (2ꢁHC@CH), 71.38–71.31 (CH, C2), 66.18–66.13 (CH2, C5),
63.36 (CH2, C1), 63.08 (CH2, C3), 59.24 (CH2, C4), 54.27 (N
3ꢁCH3), 33.42, 33.26 (CH2, 2ꢁCa), 32.53, 32.47 (CH2, 2ꢁCb),
32.35 (CH2, 2ꢁCc), 31.59, 29.53, 29.47, 28.79, 28.76, 28.54, 28.52,
27.56, (18ꢁCH2), 22.65 (2ꢁCH2), 16.10 (2ꢁCH3); LC–MS, m/z 794
[M+H]+.
N2 atmosphere, by which time the solution had turned slightly
brown in color. After completion, the reaction mixture was neu-
tralized with aqueous NaHCO3 (50 mL), the organic phase was col-
lected and the aqueous phase re-extracted with CH2Cl2
(2 ꢁ 30 mL). Organic fractions were combined, washed with water
(30 mL), brine (30 mL) over MgSO4, filtered and concentrated to a
crude residue. This crude residue was purified by silica gel (60–
120) flash column chromatography, eluting with hexane/EtOAc
(9:0.5 v/v), to furnish 24 as a bright yellow oil (2.5 g, 95%): Rf
0.60 (hexane/EtOAc, 9:1 v/v). 1H NMR (400 MHz, CDCl3): d 1.31–
1.62 (m, 10H 5ꢁCH2), 1.66–1.71 (m, 1H CH2), 1.88–2.0 (m, 3H),
3.37–3.48 (m, 3H), 3.50–3.54 (m, 1H), 3.72–3.75 (m, 1H, CHO),
3.85–3.87 (m, 1H, CHO), 4.57–4.59 (t, 1H, CHO); ESI+, m/z 264 [M
+H]+.
4.1.17. 2-(6-Iodohexyl-1-oxy)tetrahydropyran 25
NaI (8.59 g, 57.3 mmol) was dissolved in a solution of 24
(8.40 g, 28.6 mmol) in acetone (160 mL). The stirred, slightly
cloudy solution was then placed under reflux under N2 for
15 min, during which period a slide precipitated out of the solu-
tion. The reaction mixture was filtered and concentrated and the
residue was dissolved in water (150 mL) and CH2Cl2 (200 mL).
Then, the organic layer was separated, washed with water
(100 mL), brine over MgSO4, filtered, and concentrated to afford
crude product 25 as bright non-viscous oil (12.5 g). This residue
was purified by silica gel (60–120) column chromatography, elut-
ing with hexane/EtOAc (13:1 v/v), to obtain 25 as a pale yellow,
non-viscous oil (7.68 g, 89%), TLC: Rf 0.62 [hexane/EtOAc, 1:1 v/v].
1H NMR (400 MHz, CDCl3): d 1.36–1.39 (m, 4H, 2ꢁCH2), 1.40–
1.60 (m, 6H, 3ꢁCH2), 1.65–1.78 (m, 1H), 1.79–189 (m, 3H),
3.19–3.22 (t, 2H, CH2I), 3.37–3.41 (m, 1H, CH2O), 3.49–3.53 (m,
1H, CHO), 3.72–3.76 (m, 1H, CH2O), 4.58–4.59 (m, 1H, OCHO);
ESI+, m/z 302 [M]+.
4.1.13. 1-[2-(Tetradec-[7-enyl]thio)acetyl]-sn-glycero-3-
phosphocholine (lyso-dTTA-PC) 7
1H NMR (400 MHz, CDCl3): d 5.27–5.35 (m, 2H), 4.35 (m, 2H),
4.15 (m, 1H), 3.85–3.98 (m, 2H), 3.37 (s, 2H), 3.12–3.23 (m, 6H),
2.60 (t, 2H), 2.02–2.11 (m, 8H), 1.55 (m, 2H), 1.29–1.37 (m, 20H),
0.09 (3H, t); LC–MS, m/z 526 [M+H]+
4.1.14. 2,3-Acetonide-1-[2-(tetradec-[7-enyl]thio)acetyl]-
sn-glycerol 22
HBTU (1.16 g, 3.05 mmol, 3.28 equiv) and DMAP (1.09 g,
8.95 mmol, 9.64 equiv) were added to a solution of dTTA 5
(834.0 mg, 2.89 mmol, 3.11 equiv) and Solketal (85.5 mg,
0.928 mmol, 1.00 equiv) in dry CH2Cl2 under argon. The mixture
was stirred at rt for 19 h before reaction was quenched with
15 mL of citric acid solution (7%). The aqueous phase was separated
from the organic phase and extracted with CH2Cl2 (4 ꢁ 25 mL). The
organic phases were combined, dried over MgSO4, filtered and con-
centrated in vacuo to obtain as a crude white solid (2.49 g). The
solid was purified by silica gel (60–120) flash column chromatog-
raphy, eluted with hexane/EtOAc (9:1, 8:2, 6:4 v/v, EtOAc), to yield
22 as a white amorphous powder (308 mg, 83%), TLC: Rf 0.40 (hex-
ane/EtOAc 9:3 v/v). 1H NMR (400 MHz, CDCl3): d 5.36–5.40 (m,
2H), 4.26 (m, 2H), 4.09 (m, 1H), 4.30 (m, 2H), 3.32 (s, 2H), 2.64–
2.68 (t, 2H), 2.01–2.04 (m, 4H), 1.55–1.60 (m, 4H), 1.30–1.46 (m,
18H), 0.91–0.92 (t, 3H); ESI-MS, m/z 400 [M]+.
4.1.18. 2-(Hexadec-7-ynyl-1-oxy)tetrahydropyran 26
To a stirred solution of 1-octyne (2.4 mL, 13.5 mmol) in anhy-
drous THF (23 mL) at 0 °C under N2 atmosphere, n-BuLi (2.17 M
in hexane; 8.71 mL, 18.9 mmol) was added dropwise. After
15 min the solution became orange-red in color. HMPA (23 mL)
was added to the solution, turning the reaction mixture blood
red in color. After 15 min 25 (5.50 g, 16.2 mmol) was added with
anhydrous THF (11 mL), turning the solution yellow-orange.
Within 1.5 h the solution turned darker in color. The reaction mix-
ture was allowed to stir at rt for a further 48 h and then poured
onto an ice/water mixture (150 mL). Crude 26 was extracted with
diethyl ether (3 ꢁ 150 mL). The combined organic layer was
washed with water (75 mL), brine (75 mL), dried over MgSO4, fil-
tered, and concentrated. The crude residue (5.6 g) was purified
by silica gel (60–120) flash column chromatography, eluting with
hexane/EtOAc (2:1 v/v), to yield 26 as a pale-yellow oil (2.98 g,
72%), TLC: Rf 0.58 (hexane/EtOAc, 2:1 v/v). 1H NMR (CDCl3,
400 MHz): d 0.89–0.93 (t, 3H, CH3), 1.27–1.82 (m, 20H, 10ꢁCH2),
1.84–1.86 (m, 1H), 1.87–1.90 (m, 1H), 2.13–2.18 (m, 4H, H2CC„
CCH2), 3.38–3.55 (m, 2H, CHO), 3.86–3.88 (m, 1H, CHO),
3.89–3.92 (m, 1H, CHO), 4.59–4.61 (t, 1H, OCHO); ESI+, m/z 295
[M+H]+.
4.1.15. 1-[2-(Tetradec-[7-enyl]thio)acetyl]-sn-glycerol (dTTA-
MAG) 8
To a solution of compound 22 (2.59 g, 5.39 mmol) in MeOH
(15 mL) at 0 °C, TFA (1 mL) was added. The reaction was stirred
at rt for 30 min. After completion of the reaction it was neutralized
with concentrated ammonium hydroxide and the solvent removed
under reduced pressure. The resulting crude solid was purified by
silica gel (60–120) column chromatography, using EtOAc/hexane
(4:6 v/v), to obtain 8 as a white solid (0.97 g, 50%). 1H NMR
(400 MHz, CDCl3): d 5.34–5.39 (m, 2H), 4.24–4.27 (m, 2H), 3.98
(m, 1H), 3.73 (m, 2H), 3.64 (m, 2H), 3.32 (s, 2H), 2.64 (t, 2H),
2.23 (m, 2H), 2.03 (m, 4H), 1.62–1.64 (m, 2H), 1.30–1.48 (m,
12H), 0.89–0.92 (t, 3H), 13C (100 MHz, CDCl3), 170.88 (C@O),
130.08, 129.54 (HC@CH), 69.98, 66.00, 63.22, 62.00, 33.46, 32.71,
32.56, 32.44, 31.73, 29.68, 29.54, 29.41, 28.94, 28.87, 28.78,
28.60, 27.19, 27.06, 22.61(10ꢁCH2), 16.10 (CH3); ESI+, m/z 360 [M
+H]+.
4.1.19. 1-Bromohexadec-7-yne 27
To a stirred solution of PPh3Br2 (5.87 g, 13.9 mmol) and PPh3
(1.41 g, 5.36 mmol) in anhydrous CH2Cl2 (150 mL) at 0 °C under
N2, 26 was added (3.61 g, 10.7 mmol). After 15 min, excess PPh3-
Br2 was quenched with 10% K2CO3 solution. The organic phase
was separated, and the aqueous phase was re-extracted with
CH2Cl2 (3 ꢁ 150 mL). The combined organic layers were washed
with water (150 mL), brine (150 mL), dried (MgSO4), filtered,
and concentrated. Crude (4.65 g) 27 was purified by silica gel
4.1.16. 2-(6-Bromohexyl-1-oxy)tetrahydropyran 24
To a stirred solution of 6-bromohexyl-1-ol 23 (2.0 g, 9.61 mmol)
in anhydrous CH2Cl2 (40 mL) at 0 °C, under N2, DHP (1.051 mL,
11.53 mmol) and PTSA (603 mg, 2.40 mmol) were added. The reac-
tion mixture was allowed to stir for 16 h in the water bath, under

J. Lund et al. / Bioorg. Med. Chem. 24 (2016) 1191–1203
1201
(60–120) column chromatography, eluting with hexane/EtOAc
4.1.23. 2,3-Acetonide-1-[2-(tetradec-[7-ynyl]thio)acetyl]-
sn-glycerol 28
9:1 (v/v), to furnish the title compound 27 as a pure pale-yellow
oil (2.62 g, 90%), TLC: Rf 0.61 [hexane/EtOAc, 9:2 v/v]. 1H
NMR (400 MHz, CDCl3): d 0.90–0.93 (t, 3H, CH3), 1.25–1.46 (m,
14 H, 7ꢁCH2), 1.86–1.93 (quart, 2H, CH2CH2Br), 2.14–2.20 (m,
4H, H2CC„CCH2) 3.42–3.45 (t, 2H, J 6.9, CH2Br), (ESI+) m/z 272
[M]+.
HBTU (1.16 g, 3.05 mmol, 3.28 equiv) and DMAP (1.09 g,
8.95 mmol, 9.64 equiv) were added to a solution of 9 (834.0 mg,
2.89 mmol, 3.11 equiv) and Solketal (85.5 mg, 0.928 mmol,
1.00 equiv) in dry CH2Cl2 under argon. The mixture was stirred at
rt for 19 h before the reaction was quenched with 15 mL of citric
acid solution (7%). The aqueous phase was separated from the
organic phase and extracted with CH2Cl2 (4 ꢁ 25 mL). The organic
phases were combined, dried over MgSO4, filtered, and concen-
trated in vacuo to give a white solid (2.49 g). The solid was purified
by silica gel (60–120) flash column chromatography, eluting with
hexane/EtOAc (9:1, 8:2, 6:4 v/v, EtOAc), to yield 28 as a white
amorphous powder (298 mg, 81%), TLC: Rf 0.40 (hexane/EtOAc
9:1 v/v), 1H NMR (400 MHz, CDCl3): d 0.89–0.93 (t, 3H, CH3),
1.25–1.43 (m, 20H, 2ꢁCH3, 7ꢁCH2), 1.59–1.64 (m, 2H, CH2),
2.13–2.20 (m, 4H, H2CC„CCH2), 2.64–2.68 (t, 2H, CH2AS), 3.27
(s, 2H, SCH2ACOOH), 3.76–3.80 (m, 1H, CHO), 4.09–4.20 (m, 3H,
CH2O), 4.35–4.38 (quint, 1H, OCHO), 4.27 (s, 20H,), 0.91–0.88 (t,
3H,), m/z 398 [M]+.
4.1.20. 2-(Tetradec-[7-ynyl]thio)acetic acid (tTTA) 9
Compound 27 (5.36 mL, 18.0 mmol, 1.00 equiv) was added to
a solution of thioglycolic acid (1.25 mL, 18.0 mmol, 1.00 equiv) in
25% NaOH in MeOH (6.03 g in 24 mL) and the mixture was stir-
red vigorously at rt for 72 h. After dilution with water the mix-
ture was acidified to pH 1 with concentrated HCl and the
aqueous phase was extracted with diethyl ether. The ether phase
was dried over MgSO4, filtered, and concentrated in vacuo to
obtain a white solid, that was purified by silica gel (60–120)
flash column chromatography, eluting with CH2Cl2, to yield tTTA
9 as an amorphous solid (4.36 g, 85%), TLC: Rf 0.50 (hexane/
EtOAc 3:1 v/v). 1H NMR (400 MHz, CDCl3): d 0.89–0.96 (t, 3H,
CH3), 1.25–1.42 (m, 14H, 7ꢁCH2), 1.61–1.68 (m, 2H, CH2),
2.14–2.19 (m, 4H, H2CC„CCH2) 2.67–2.70 (t, 2H, CH2-S), 3.28
(s, 2H, SACH2COOH), 13C (500 MHz, CDCl3), 176.53 (C@O),
80.42, 79.91 (HC„CH), 33.44, 32.72, 31.34, 29.09, 28.91, 28.77,
28.51, 28.31, 28.22, 22.54, 18.72, 18.64 (9ꢁCH2) 16.05 (CH3);
ESI-MS, m/z 283 [MꢀH]+.
4.1.24. 1-[2-(Tetradec-[7-ynyl]thio)acetyl]-sn-glycerol (tTTA-
MAG) 12
To a solution of compound 28 (2.59 g, 5.39 mmol) in MeOH
(15 mL) at 0 °C, TFA (1 mL) was added. The reaction was stirred
at rt for 15 min before neutralized with concentrated NH4OH and
the solvent was removed on reduced pressure. The crude product
was purified by silica gel (60–120) column chromatography, elut-
ing with EtOAc/hexane (6:4 v/v), to obtain 12 as a white solid
(1.16 g, 60%). 1H NMR (400 MHz, CDCl3): d 0.89–0.93 (t, 3H, CH3),
1.25–1.43 (m, 14H, 7ꢁCH2), 1.50–1.65 (m, 2H), 1.99 (br, 2H),
2.14–2.16 (m, 4H, H2CC„CCH2), 2.64–2.68 (t, 2H, CH2AS), 3.25–
3.28 (s, 2H, SACH2ACOOH) 3.64–3.82 (m, 2H, CHO), 3.97–4.0 (m,
1H, CHO), 4.21–4.32 (m, 2H, CH2AOH), 13C (125 MHz, CDCl3),
170.85 (C@O), 80.42, 79.91 (HC„CH), 69.99 (CH, C2), 66.01 (CH2,
C5), 63.23 (CH2, C1), 63.05 (CH2, C3), 33.42, 32.68, 31.32, 29.08,
28.91, 28.82, 28.50, 28.46, 28.31, 28.25, 28.22, (10ꢁCH2), 22.53,
18.71 (CH2), 16.10, (CH3); ESI-MS, m/z 359 [M+H]+.
4.1.21. 1,2-Di-[2-(tetradec-[7-ynyl]thio)acetyl]-sn-glycero-3-
phosphocholine (tTTA-PC) 10
To a solution of 9 (261 mg, 0.905 mmol, 1.00 equiv) in DMF
(5 mL) and DCC (190 mg, 1.17 mmol, 1.30 equiv), DMAP (135
lL,
0.905 mmol, 1.0 equiv) and -GPC (133 mg, 0.301 mmol,
L-a
0.33 equiv; 5 mL) were added. The mixture was stirred vigorously
at rt under N2 atmosphere for 16 h. A clear yellow solution was
obtained. After completion, DMF was removed in vacuo and the
residue purified by silica gel (60–120) flash column chromatogra-
phy, eluting with solvent mix A followed by solvent mix B, to yield
10 (143 mg, 60%) and lyso-tTTA-PC 11 (32 mg, 20%) as white amor-
phous solids, TLC: Rf 0.40 (Solvent A, Solvent B), 1H NMR (400 MHz,
CDCl3): d 0.73–0.90 (t, 6H, 2ꢁCH3), 1.24–1.61 (m, 32H, 16ꢁCH2),
2.01–2.15 (t, 8H, 2ꢁH2CC„CCH2), 2.60–2.64 (m, 4H, 2ꢁCH2AS),
3.20–3.30 (s, 4H, 2ꢁSACH2ACOOH), 3.76 (t, 2H, CH2O), 3.84 (t,
2H, CH2AN), 4.24–4.46 (m, 4H, 2ꢁCH2O), 5.26–5.32 (m, 1H,
CHO), 13C (125 MHz, CDCl3), 170.27, 170.02 (C@O), 80.34, 79.88
(HC„CH), 71.48–71.42, (CH, C2), 66.40–66.35 (CH2, C5), 63.47
(CH2, C1), 63.24 (CH2, C3), 59.42 (CH2, C4), 54.50 (N 3ꢁCH3),
33.57, 33.42 (CH2, 2ꢁCa), 32.65, 32.58 (CH2, 2ꢁCb), 31.32 (CH2,
2ꢁCc), 29.63, 29.08, 29.05, 29.00, 28.99, 28.94, 28.84, 28.50,
28.42, 28.32, 28.26 (18ꢁCH2), 22.52, 18.71 (2ꢁCH2), 14.05, 14.01
(2ꢁCH3); EI+, m/z 790 [M]+.
4.2. Human skeletal muscle cell cultures
Human skeletal muscle cells (myotubes) grown from satellite
cells were isolated as previously described16 from musculus obli-
quus internus abdominis from healthy donors, aged 48 3 years
and body mass index 24.5 0.8 kg/m2. The biopsies were obtained
with informed consent and approved by the National Committee
for Research Ethics (Oslo, Norway). The satellite cells were cultured
in DMEM-GlutamaxTM (5.5 mM glucose, Gibco, Life Technologies,
Paisley, UK) with supplements during proliferation and differenti-
ation into myotubes as previously described.29 Experiments were
performed after 7 days of differentiation, and the cells were
exposed to DMSO (0.1%, Sigma Aldrich), different concentrations
of THEFAs, THEFA-containing lipids or GW501516 for 96 h before
initiation of experiments.
4.1.22. 1-[2-(Tetradec-[7-ynyl]thio)acetyl]-sn-glycero-3-
phosphocholine (lyso-tTTA-PC) 11
1H NMR (400 MHz, CDCl3): d 0.8–0.93 (t, 3H, CH3), 1.2–1.6 (m,
16H, 8ꢁCH2), 2.13–2.15 (t, 4H, H2CC„CCH2), 2.61–2.64 (m, 2H,
ASACH2), 3.25 (s, 2H, SACH2ACOOH), 3.30 (s, 9 H, (CH3)3AN),
3.66–3.76 (m, 2 H, CH2AN), 3.82 (m, 2H, CH2O), 3.98 (m, 2H,
CH2O), 4.16 (br, 2H, CH2O), 4.99 (br, 1H, ACHAOH), 13C (CDCl3,
125 MHz), 170.51 (C@O), 80.28, 79.85 (HC„CH), 68.49, 68.45
(CH, C2), 66.87–66.06 (CH2, C5), 59.38 (CH2, C1), 57.89 (CH2, C3),
54.24 (N, 3ꢁCH3), 39. 82, (CH2, C4), 33.55, 32.55 (CH2, Ca), 32.49
(CH2, Cb), 31.49 (CH2, Cc), 31.27, 29.58, 29.04, 28.96, 28.96,
28.84, 28.51, 28.45, 28.38, 28.24, 28.26 (9ꢁCH2), 22.50, 18.90
(CH2), 16.05, (CH3); ESI-MS, m/z 524 [M+H]+.
4.3. Fatty acid analysis
Human myotubes were cultured in 25 cm2 NuncTM Cell Culture
Treated EasYFlasksTM (Thermo Scientific, Roskilde, Denmark) and
the cells were treated with DMSO (0.1%), THEFAs or THEFA-con-
taining lipids (10 lM or 30 lM) in culture media for 96 h. The cells
were then washed with DPBS (with Ca2+ and Mg2+), harvested in
cold DPBS (0.5 mL) and then stored at ꢀ80 °C. Protein content
was measured, lipids from washed myotubes were extracted and

1202
J. Lund et al. / Bioorg. Med. Chem. 24 (2016) 1191–1203
methyl esters were obtained by heating of lipids with methanol at
90 °C for one hour as previously described.30,31 Sulfuric acid was
used as a catalyst. After extraction into an organic solvent, FA
methyl esters were analyzed by gas–liquid chromatography. Gas
chromatograph GC 8000 TOP (Finnigan, USA) was equipped with
a programmed temperature vaporization injector, flame-ionization
detector, AS 800 autosampler, and a fused silica capillary column
coated with dimethylpolysiloxane stationary phase, DB1-ms
(J&W Scientific, USA). Hydrogen was used as carrier gas. Column
temperature was programmed from 110 to 310 °C with a gradient
of 2.5 °C/min. GC signal was acquired and evaluated with Chrome-
leon software (Dionex, USA). Peaks were identified by means of
known FA standards and by means of mass spectra, obtained by
GC/MS analysis (GCQ, Finnigan, USA) on the same column. Internal
standard C21:0 was used for quantitation after calibration with
known mixtures of FA standards.
Authors contribution
J.L.
Conducted biological assays,
interpretation of data, wrote the
manuscript
C.S.
Conducted biological assays,
interpretation of data
P.B.
H.T.
R.
M.W.
A.K.
A.C.R.
Conducted fatty acid analysis
Supervised assays, interpretation of data
Conducted synthetic work
Conducted synthetic work
Conducted synthetic work
Supervised biological assays,
interpretation of data
R.K.B.
A.D.M.
Designed and supervised the study
Designed and supervised chemical
synthesis, wrote the manuscript
Designed and supervised the study,
interpretation of data, wrote the
manuscript
4.4. Substrate oxidation assay
J.S.
Human myotubes were cultured on 96-well CellBINDÒ micro-
plates (Corning Life-Sciences, Schiphol-Rijk, The Netherlands),
and the cells were treated for 96 h with DMSO (0.1%), THEFAs or
THEFA-containing lipids (10 lM or 30 lM), or the PPARd agonist
GW501516 (10 nM, gift from Trond Vidar Hansen, School of Phar-
macy, University of Oslo) in culture media. [1-14C]oleic acid (OA,
Acknowledgements
0.5
Phosphate Buffered Saline (DPBS, with Mg2+ and Ca2+, Gibco, Life
Technologies, Paisley, UK) with HEPES (10 mM) and -carnitine
lCi/mL, 100 lM, PerkinElmer, Boston, MA, US) in Dulbecco’s
This work was supported by grants from UKIERI, University of
Bergen and the University of Oslo. Rajender acknowledges CSIR-
UGC, New Delhi for the award of a senior research fellowship.
L
(1 mM) was given during 4 h incubation for CO2 oxidation assess-
ment. A 96-well UniFilter-96Ò GF/B micro plate (PerkinElmer, Shel-
ton, CT, US) was mounted on top of the CellBINDÒ plate as
previously described32 and the cells were incubated at 37 °C for
4 h. The [14C]OA trapped in the filter was counted by liquid scintil-
lation, and the result reflects CO2 production. The remaining cell-
associated radioactivity (CA) was also assessed by liquid scintillation,
and the sum of CO2 and CA was considered as total substrate
uptake. After completion of the substrate oxidation assay, protein
content in each well was measured according to Bradford.33
The authors would like to thank Kari Williams, Liv-Kristine Øys
and Kari Helland Mortensen for excellent technical assistance.
æd
References and notes
1. Berge, R. K.; Skorve, J.; Tronstad, K. J.; Berge, K.; Gudbrandsen, O. A.; Grav, H.
Curr. Opin. Lipidol. 2002, 13, 295.
2. Dyroy, E.; Wergedahl, H.; Skorve, J.; Gudbrandsen, O. A.; Songstad, J.; Berge, R.
K. Lipids 2006, 41, 169.
3. Gudbrandsen, O. A.; Dyrøy, E.; Bohov, P.; Skorve, J.; Berge, R. K. Chem. Biol.
Interact. 2005, 155, 71.
4. Berger, J.; Moller, D. E. Annu. Rev. Med. 2002, 53, 409.
5. Fuentes, E.; Guzmán-Jofre, L.; Moore-Carrasco, R.; Palomo, I. Mol. Med. Report
2013, 8, 1611.
4.5. RNA-isolation and qPCR
6. Hihi, A.; Michalik, L.; Wahli, W. Cell. Mol. Life Sci. 2002, 59, 790.
7. Monsalve, F. A.; Pyarasani, R. D.; Delgado-Lopez, F.; Moore-Carrasco, R.
Mediators Inflamm. 2013, 2013.
8. Fredriksen, J.; Ueland, T.; Dyrøy, E.; Halvorsen, B.; Melby, K.; Melbye, L.;
Skalhegg, B.; Bohov, P.; Skorve, J.; Berge, R. Eur. J. Clin. Invest. 2004, 34, 709.
9. Løvås, K.; Røst, T.; Skorve, J.; Ulvik, R.; Gudbrandsen, O.; Bohov, P.; Wensaas, A.;
Rustan, A.; Berge, R.; Husebye, E. Diabetes Obes. Metab. 2009, 11, 304.
10. Morken, T.; Bohov, P.; Skorve, J.; Ulvik, R.; Aukrust, P.; Berge, R. K.; Livden, J. K.
Scand. J. Clin. Lab. Invest. 2011, 71, 269.
11. Pettersen, R. J.; Salem, M.; Skorve, J.; Ulvik, R. J.; Berge, R. K.; Nordrehaug, J. E. J.
Cardiovasc. Pharmacol. 2008, 51, 410.
12. Skrede, S.; Sørensen, H. N.; Larsen, L. N.; Steineger, H. H.; Høvik, K.; Spydevold,
Ø. S.; Horn, R.; Bremer, J. Metabolism 1997, 1344, 115.
Human myotubes were cultured in NuncTM Cell Culture Treated
EasYFlasksTM (25 cm2, Thermo Scientific) and the cells were treated
with DMSO (0.1%), THEFAs or THEFA-containing lipids (10
lM or
30 M) in culture media for 96 h. The myotubes were washed with
l
cold DPBS and centrifuged to a pellet before total RNA was isolated.
Total RNA was purified using 1 mL TRIzolÒ reagent (AmbionTM, Life
Technologies, Carlsbad, CA, USA) combined with the RNeasy Mini
Kit from QIAGEN (Hilden, Germany). The quantity of RNA was
assessed spectrophotometrically using NanoDrop ND-1000 (Nano-
Drop Products, Wilmington, DE, USA). The purity of RNA was mea-
sured using Agilent RNA 6000 Nano Kit on an Agilent 2100
Bioanalyzer (Agilent Technologies Inc., Santa Clara, CA, USA). The
quality limit for further use of RNA was set to a RNA Integrity
Number P7 (out of 10). cDNA was obtained using the High Capac-
ity Reverse Transcription Kit with RNase inhibitor (Applied Biosys-
tems, Foster City, USA). Real-time PCR was performed on an ABI
prism 7900 H sequence detection system (Applied Biosystems)
using 384-well multiply PCR plates (Sarstedt Inc., Newton, USA),
with gene specific primers from Applied Biosystems. Three differ-
ent reference genes were included: 18s (Kit-FAM-TAMRA) from
Eurogentec, Belgium, glyceraldehyde-3-phosphate dehydrogenase
(Taqman Rodent GAPDH) and ribosomal protein, large, P0 (RPLP0)
from Applied Biosystems. The NormFinder software was used to
evaluate the reference genes and data normalized against all three
reference genes are presented.34
13. Jorgensen, M. R.; Bhurruth-Alcor, Y.; Røst, T.; Bohov, P.; Müller, M.; Guisado, C.;
Kostarelos, K.; Dyrøy, E.; Berge, R. K.; Miller, A. D. J. Med. Chem. 2009, 52, 1172.
14. Bhurruth-Alcor, Y.; Rost, T. H.; Jorgensen, M. R.; Müller, M.; Skorve, J.; Berge, R.
K.; Miller, A. D. Bioorg. Med. Chem. Lett. 2010, 20, 1252.
15. Madsen, L.; Frøyland, L.; Grav, H. J.; Berge, R. K. J. Lipid Res. 1997, 38, 554.
´
16. Feng, Y. Z.; Nikolic, N.; Bakke, S. S.; Boekschoten, M. V.; Kersten, S.; Kase, E. T.;
Rustan, A. C.; Thoresen, G. H. Arch. Physiol. Biochem. 2014, 120, 12.
17. Kersten, S.; Desvergne, B.; Wahli, W. Nature 2000, 405, 421.
18. Wensaas, A. J.; Rustan, A. C.; Just, M.; Berge, R. K.; Drevon, C. A.; Gaster, M.
Diabetes 2009, 58, 527.
19. Zhu, P.; Goh, Y. Y.; Chin, H. F.; Kersten, S.; Tan, N. S. Biosci. Rep. 2012, 32, 211.
20. Su, X.; Abumrad, N. A. Trends Endocrinol. Metab. 2009, 20, 72.
21. Bonnefont, J.-P.; Djouadi, F.; Prip-Buus, C.; Gobin, S.; Munnich, A.; Bastin, J. Mol.
Aspects Med. 2004, 25, 495.
22. Forman, B. M. Cell Metab. 2005, 1, 153.
23. Musille, P. M.; Kohn, J. A.; Ortlund, E. A. FEBS Lett. 2013, 587, 1238.
24. Krylova, I. N.; Sablin, E. P.; Moore, J.; Xu, R. X.; Waitt, G. M.; MacKay, J. A.;
Juzumiene, D.; Bynum, J. M.; Madauss, K.; Montana, V. Cell 2005, 120, 343.
25. Li, Y.; Choi, M.; Cavey, G.; Daugherty, J.; Suino, K.; Kovach, A.; Bingham, N. C.;
Kliewer, S. A.; Xu, H. E. Mol. Cell 2005, 17, 491.

J. Lund et al. / Bioorg. Med. Chem. 24 (2016) 1191–1203
1203
26. Chakravarthy, M. V.; Lodhi, I. J.; Yin, L.; Malapaka, R. R.; Xu, H. E.; Turk, J.;
Semenkovich, C. F. Cell 2009, 138, 476.
30. Bligh, E. G.; Dyer, W. J. Can. J. Biochem. Physiol. 1959, 37, 911.
31. Christie, W. W. The Preparation of Methyl and Other Esters of Fatty Acids. In
Gas Chromatography and Lipids: A Practical Guide; Barnes, P. J., Associates, Eds.;
The Oily Press: Bridgewater, Somerst, Scotland, 1989; pp 37–44.
32. Wensaas, A.; Rustan, A.; Lövstedt, K.; Kull, B.; Wikström, S.; Drevon, C.; Hallen,
S. J. Lipid Res. 2007, 48, 961.
27. Stapleton, C. M.; Mashek, D. G.; Wang, S.; Nagle, C. A.; Cline, G. W.; Thuillier, P.;
Leesnitzer, L. M.; Li, L. O.; Stimmel, J. B.; Shulman, G. I. PLoS One 2011, 6.
28. Still, W. C.; Kahn, M.; Mitra, A. J. Org. Chem. 1978, 43, 2923.
29. Hessvik, N. P.; Bakke, S. S.; Fredriksson, K.; Boekschoten, M. V.; Fjørkenstad, A.;
Koster, G.; Hesselink, M. K.; Kersten, S.; Kase, E. T.; Rustan, A. C. J. Lipid Res.
2010, 51, 2090.
33. Bradford, M. M. Anal. Biochem. 1976, 72, 248.
34. Andersen, C. L.; Jensen, J. L.; Ørntoft, T. F. Cancer Res. 2004, 64, 5245.