Evaluation of Dietary Effects on Hepatic Lipids in High
Fat and Placebo Diet Fed Rats by In Vivo MRS and LC-MS
Jadegoud Yaligar1, Venkatesh Gopalan1, Ong Wee Kiat1, Shigeki Sugii1, Guanghou Shui2, Buu
Duyen Lam3, Christiani Jeyakumar Henry4, Markus R. Wenk3, E. Shyong Tai5, S. Sendhil Velan1,4*
1 Laboratory of Molecular Imaging, Singapore Bioimaging Consortium, A*STAR, Singapore, Singapore, 2 Life Sciences Institute, National University of Singapore,
Singapore, Singapore, 3 Department of Biochemistry and Department of Biological Sciences, National University of Singapore, Singapore, Singapore, 4 Singapore Institute
for Clinical Sciences, A*STAR, Singapore, Singapore, 5 Department of Medicine, National University of Singapore, Singapore, Singapore
Background & Aims: Dietary saturated fatty acids contribute to the development of fatty liver and have pathogenic link to
systemic inflammation. We investigated the effects of dietary fat towards the pathogenesis of non-alcoholic fatty liver
disease by longitudinal in vivo magnetic resonance spectroscopy (MRS) and in vitro liquid chromatography coupled with
mass spectrometry (LC-MS).
Methods: All measurements were performed on rats fed with high fat diet (HFD) and chow diet for twenty four weeks.
Longitudinal MRS measurements were performed at the 12th, 18th and 24th weeks. Liver tissues were analyzed by LC-MS,
histology and gene transcription studies after terminal in vivo experiments.
Results: Liver fat content of HFD rats for all ages was significantly (P,0.05) higher compared to their respective chow diet
fed rats. Unsaturation indices estimated from MRS and LC-MS data of chow diet fed rats were significantly higher (P,0.05)
than HFD fed rats. The concentration of triglycerides 48:1, 48:2, 50:1, 50:2, 50:3, 52:1, 52:2, 52:3, 54:3 and 54:2 was
significantly higher (P,0.05) in HFD rats. The concentration for some polyunsaturated triglycerides 54:7, 56:8, 56:7, 58:11,
58:10, 58:9, 58:8 and 60:10 was significantly higher (P,0.05) in chow diet fed rats compared to HFD rats. Lysophospholipid
concentrations including LPC and LPE were higher in HFD rats at 24 weeks indicating the increased risk of diabetes. The
expression of CD36, PPARa, SCD1, SREBF1 and UCP2 were significantly upregulated in HFD rats.
Conclusions: We demonstrated the early changes in saturated and unsaturated lipid composition in fatty liver by in vivo
MRS and ex vivo LC-MS. The higher LPC concentration in HFD rats indicated a higher risk of developing diabetes. Early
metabolic perturbations causing changes in lipid composition can be evaluated by the unsaturation index and correlated to
the non alcoholic fatty liver disease.
Citation: Yaligar J, Gopalan V, Kiat OW, Sugii S, Shui G, et al. (2014) Evaluation of Dietary Effects on Hepatic Lipids in High Fat and Placebo Diet Fed Rats by In Vivo
MRS and LC-MS Techniques. PLoS ONE 9(3): e91436. doi:10.1371/journal.pone.0091436
Editor: Herve´ Guillou, INRA, France
Received October 1, 2013; Accepted February 12, 2014; Published March 17, 2014
Copyright: ß 2014 Yaligar et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This research was supported by the intramural funding from Singapore Bioimaging Consortium, A*STAR, Singapore & National University of Singapore,
Singapore. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: Sendhil_Velan@sbic.a-star.edu.sg
Introduction insulin-resistant type-2 diabetes mellitus and atherosclerotic
cardiovascular disease are associated with NAFLD [1,2,3,4,5].
Non alcoholic fatty liver disease (NAFLD) results from an
imbalance between lipid availability (from circulating lipid uptake The estimation of saturated and unsaturated lipids during the
or de novo lipogenesis) and lipid disposal (via fatty acid oxidation transition of normal liver to NAFLD provides information on early
or triglyceride-rich lipoprotein secretion). The accumulated lipids biochemical changes and may be utilized to evaluate response
induce oxidative stress, resulting in production of cytokines and during interventions including exercise and drugs. Indeed, the
reactive oxygen species which in turn activate apoptosis thereby composition of the fatty acids (FA, the building blocks for
initiating a sequence of disease events from steatosis to nonalco- triglyceride) in the triglycerides (TG) is suspected to contribute
holic steatohepatitis (NASH), which progress into fibrosis and to the pathogenesis of NAFLD. In particular, saturated fatty acids
cirrhosis [1,2,3,4,5]. Hepatic steatosis causes insulin resistance play a prominent role in the progression of obesity and diabetes
which may act as pathogenic link between obesity and its [8,9,10,11]. Diets with high saturated fatty acids are associated
metabolic complications [6,7]. Many patients suffering from with insulin resistance and NAFLD . Saturated fatty acids are
metabolic syndrome involving obesity, dyslipidemia, hypertension, poorly oxidized compared to unsaturated fat and hence are more
likely to accumulate in insulin resistant tissues . Saturated
PLOS ONE | www.plosone.org 1 March 2014 | Volume 9 | Issue 3 | e91436
Hepatic Lipid Composition by In Vivo MRS and LC-MS
lipids are also the most inhibitory lipids on insulin sensitivity positioned prone on the surface coil with the liver on top of the
[14,15]. The degree of unsaturation of the FAs modulate the coil. The respiratory rate and body temperature were monitored
metabolic signaling and energy metabolism . As such, the using physiological monitoring system (ML880 16/30 power lab
assessment of both the ratio of saturated and unsaturated fat in the system, AD Instruments, Spechbach, Germany). The temperature
liver, as well as the quantity of fat in the liver represent an probe was placed in the rectum of the rat and body temperature
important aspect in the pathogenesis of chronic liver disease and was maintained at 37uC by circulating hot water through a rat
metabolic disease. cradle on which the animal was resting. In vivo measurements were
performed on animals in both the chow diet group and HFD
Magnetic resonance spectroscopy (MRS) is a non-invasive group at 12, 18 and 24 weeks of age. Volume localized point
method for investigating metabolism allowing longitudinal studies resolved spectroscopy sequence (PRESS) with water suppression
on humans and rodents. It estimates both the total fat content and experiments were performed on a 46464 mm3 voxel within the
unsaturation index. However, its main limitation deals with the liver using TR = 4000 ms, TE = 13 ms, 128 averages, and 2048
impossibility of estimating the individual lipid composition. On complex points with a spectral width of 3500 Hz. A respiratory
another hand, liquid chromatography with mass spectrometry triggered gating module was incorporated into the PRESS
(LC-MS) permits analysis of individual saturated and unsaturated sequence with a trigger delay of 20 ms and an animal breathing
lipids. stabilized at 60–65 cycles per minute. Water unsuppressed spectra
were acquired under identical conditions with similar parameters
In this study, we used in vivo MRS to evaluate the longitudinal but only 4 averages. Lipid estimates were corrected for T2
changes in liver fat content and unsaturation, on rodents fed with a relaxation. Fat content was estimated by the ratio of n-methylene
high-fat diet (HFD). HFD is a significant source of fatty acids taken signal (1.30 ppm) to the sum of n-methylene and water signals. An
up by the liver . We identified the specific lipid species that are unsaturation index was estimated using the ratio of olefinic signal
altered in liver of HFD fed animals due to insulin resistance and (5.30 ppm) to the sum of olefinic, methylenes (1.30 ppm and
fatty liver conditions predisposed to diabetes. 2.06 ppm) and methyl signals (0.90 ppm) .
Materials and Methods Liquid Chromatography Coupled with Mass
Animals and Animal Diet Spectrometry (LC-MS)
All animal experiments were approved by the institutional Animals were sacrificed at the 24th week just after the final in vivo
animal care and use committee of the biological resource center, MR experiments. The liver tissues were snap frozen in liquid
A*STAR, Singapore. Male F344 rats (CLEA Japan, Inc. Tokyo, nitrogen. Lipids were extracted and quantified using methodology
Japan) were received at 4 weeks of age, placed in individual cages, as described in our earlier work [20,21,22,23]. About 20–30 mg of
and randomly assigned to chow diet (n = 8) and HFD groups liver tissue was homogenized in 900 mL of chloroform/methanol,
(n = 8). The animals were fed with their respective diets from the 1/2, v/v (Merck Pte. Ltd., Singapore) and incubated on a vacuum
5th week until the completion of the study (24th week). The FA chamber in a dark room for 1 h with agitation. After incubation,
composition of the HFD (Research Diets, New Brunswick, New 0.3 mL of chloroform was added to the homogenate, followed by
Jersey, USA) was 62.4% saturated fat, 30.7% monounsaturated 0.4 mL of ice-cold water. The homogenate was then vortexed for
fat, 6.9% polyunsaturated fat. 30 s followed by centrifugation for 2 min at 9000 rpm. The
bottom organic phase was carefully transferred to an empty tube
Biochemical Measurements and 0.5 mL of ice-cold chloroform was added and re-extracted to
We measured plasma triglyceride, glucose, cholesterol and collect the residual lipids fractions. The two organic extracts were
then combined and dried under nitrogen. Individual classes of
insulin levels at 12, 18 and 24 weeks of age. All the animals were polar lipids were separated using an Agilent 1200 HPLC system
fasted for 13 hours before blood collection. Bleeding was before introduction into a 3200 Q-Trap mass spectrometer
performed from the lateral tail vein using a rodent restrainer. (Applied Biosystems) with HPLC conditions: Luna 3-mm silica
Plasma glucose, cholesterol and triglyceride were evaluated by column (i.d. 15062.0 mm), mobile phase A (chloroform:methano-
enzymatic colorimetric method using hexokinase, cholesterol l:ammonium hydroxide, 89.5:10:0.5), mobile phase B (chloroform:-
esterase, cholesterol oxidase and fossati steps respectively (Quest methanol:ammonium hydroxide:water, 55:39:0.5:5.5); flow rate
Lab Pvt. Ltd Singapore). Plasma insulin was measured by using an 300 mL min21; 5% B for 3 min, then linear increase of B up to 30%
ultra-sensitive ELISA kit (Crystal Chem Inc., Illinois, USA). An in 24 min, follow by 5 min under these conditions, and then linear
oral glucose tolerance test (OGTT) was performed at week 18 after change to 70% B in 5 min. Mass spectrometry was recorded under
an overnight fasting. Glucose (2 g kg21) was administered to rats both positive and negative electron-spray ionization (ESI) modes
by oral gavage injection and blood samples were collected from depending on the fraction  (conditions: Turbo Spray source
the tail vein at 0, 10, 30, 60 and 120 min. The total area under the voltage, 5000 and - 4500 V for positive and negative modes,
glucose curve was determined from time 0 to 120 min (AUC 0– respectively) with EMS scan type (conditions: source temperature,
120 min) after glucose administration as described in our earlier 300uC; GS1: 40.00, GS2: 40.00, curtain gas: 25.). Various lipid
work . fractions including phosphatidylcholine (PC), sphingomyelin (SM),
ceramide (Cer) and Glucocyl-ceramide (GluCer) were acquired in
In vivo MR Imaging and Spectroscopy the positive ESI mode while phosphatidylethanolamine (PE),
All animals were subjected to magnetic resonance imaging phosphatidylinositol (PI), phosphatidylserine (PS), phosphatidylgly-
cerol (PG), phosphatidic acids (PA) and gangliosides mannoside 3
(MRI) and localized MRS. Prior to in vivo experiments, animals (GM3) were measured in the negative ESI mode . Individual
were initially anesthetized with 3% isofluorane in a dedicated lipid species were quantified by comparison with spiked internal
chamber. During the course of MRS experiments, isofluorane standards PC-14:0/14:0, PE-14:0-14:0, PS- 14:0/14:0, PA-17:0/
levels were reduced to 1.5–2.0% in combination with medical air 17:0, PG-14:0/14:0, d31-PI18:1/16:0, C17-LPC, C17-LPA, C17-
and medical oxygen. In vivo imaging and spectroscopy were LPS, C17-Cer, C8-GluCer, C12-SM and C17-ganglioside GM3
performed using a 7 T Bruker ClinScan (Siemens VB15) MRI/
MRS scanner equipped with a 72 mm volume resonator for RF
transmit and 20 mm receive only surface coil. Rats were
PLOS ONE | www.plosone.org 2 March 2014 | Volume 9 | Issue 3 | e91436
Hepatic Lipid Composition by In Vivo MRS and LC-MS
obtained from Avanti polar lipids (Alabaster, AL, USA). The molar Data Analysis and Statistics
fractions of individual lipid species and each lipid class were Spectroscopic data were processed and analyzed using the
normalized by summation of all polar lipid species. TGs were
separated from polar lipids on an Agilent Zorbax Eclipse XDB-C18 LCmodel software . Lipid concentrations for both chow diet
column (i.d. 15064.66 mm), with chloroform:methanol:0.1 M and HFD groups were estimated by fitting the resonances of
ammonium acetate (100:100:4) as mobile phase at a flow rate of methyl, n-methylene, allylic methylene, and the unsuppressed
0.25 mL min21. TGs were analyzed by using a modified version of water signal. The results are expressed as the mean 6 s.e.m.
reversed phase HPLC/ESI/MS with d5-TG 48:0 (CDN isotopes) (standard error of the mean). Statistical analysis was performed by
as internal standard . Cholesterol esters were analyzed with MedCalc, with significant differences between means identified
corresponding d6-C18 cholesterol ester (CDN isotopes) as internal using paired two-tailed t-test. Differences were considered
standards . The unsaturation index from the LC-MS data were significant at P,0.05.
computed by the ratio of the absolute concentration of all
unsaturated lipids to the concentration of saturated lipids. Unsupervised multivariate factor analysis was performed using
Triglycerides with no double bonds were defined as saturated, with principal component analysis (PCA) (Unscrambler 10.2 software)
1 to 3 unsaturations as mono-unsaturated (considering up to one on all quantitative data measured through LC-MS techniques.
double bond for each fatty acyl chain) and more than 3 double The PCA results are represented in terms of scores and correlation
bonds as poly-unsaturated. Two unsaturation indices were derived loadings plots. These scores and correlation loading vectors
considering either all unsaturated fatty acids (mono- and poly- provide a concise and simplified description of the variance
unsaturated, n$1) or only the poly-unsaturated ones (n.3). hidden in the dataset [26,27]. In the current PCA model the term
‘explained validation variance’ (EVV), expressed in percentage, is
mRNA Analysis defined as the proportion of the variance in the data explained by
Total RNA was extracted from the liver samples using trizol the model.
reagent (Invitrogen) and treated with DNase I prior to cDNA Results
conversion using the revertAid H minus first strand cDNA
synthesis kit (Fermentas, USA) with oligo d(T) 18 primer according Bodyweight and Biochemical Measurements
to manufacturer’s instructions. For real time qPCR, cDNA HFD fed animals gained significant body weight during the 9th
samples were analyzed in triplicates using the SYBR Green
PCR Master Mix reagent kit (Applied Biosystems) on a and 10th weeks. The average body weight of HFD group at the
StepOnePlus Real-Time PCR System (Applied Biosystems). 12th, 18th and 24th week was significantly (P,0.001) higher than
Relative mRNA levels were calculated and normalized to chow diet group (Table 1). The relative increase in percentage
glyceraldehyde 3-phosphate dehydrogenase GAPDH (CAAGGT- body weight of HFD rats at 12, 18 and 24 weeks were 20%, 35%
CATCCATGACAACTTTG) and (GGCCATCCACAGTCTT- and 42% higher than the chow diet rats respectively. Blood plasma
CTGA), used as an endogenous control gene. The primer glucose, insulin, total cholesterol and TG measured in HFD and
sequences used were as follows: peroxisome-proliferator-activated chow diet groups of rats are listed in Table 1. The total TGs
receptor a; PPARa (TGTCATCACAGACACCCTCTCTC) and (P#0.002) and cholesterol content (P#0.001) of HFD animals
(TCATCTGTACTGGTGGGGACA), sterol regulatory element were significantly higher than the chow diet fed animals for all
binding factor; SREBF1 (CTGCTTTGGAACCTCGTCCG) and measurements. The plasma glucose and insulin concentrations
(GCCTCCTGTGTACTTGCCCAT), stearoyl-CoA desaturase after OGTT performed at 18 weeks are shown in Figure 1A and
1; SCD1 (CCTACGACAAGAACATTCAATCTC) and (TT- B, respectively. Plasma glucose and insulin concentrations
GATGTGCCAGCGGTACTCACTG). Fatty acid translocase; calculated by the area under the curve from 0–120 mins were
CD36 (Rn02115479_g1), mitochondrial uncoupling protein 2; significantly (P,0.001) higher in HFD fed rats compared to chow
UCP2 (Rn01754856_m1) (Taqmann, Life Technologies, CA, diet fed rats. Plasma glucose and insulin concentration from 0–
USA). 120 mins for HFD rats was 29108 mg/dL and 876 mg/dL
compared to 22607 mg/dL and 296 mg/dL in chow diet fed rats
Histology respectively. The plasma glucose and insulin concentrations of
Rat liver sections were stained for Oil Red O and hematoxylin HFD rats at the 12th, 18th and 24th weeks were significantly higher
(P,0.001) than chow diet fed rats (Table 1).
& eosin (H & E) to assess the fat accumulation and hepatocyte
inflammation, respectively. After the terminal study the livers were Magnetic Resonance Spectroscopy
excised and fixed in 10% formalin for 24 h and embedded in The liver fat was measured longitudinally in chow diet and
paraffin wax after dehydration. Formalin fixed rat livers were
sectioned at 8 mm and slides were rinsed with PBS (pH 7.4). After HFD fed animals at the 12th, 18th and 24th weeks of age. Most of
passing dry air, the slides were placed in 100% propylene glycol the lipid resonances are well resolved at 7 T compared to lower
for 2 min, and stained in 0.5% Oil Red O solution in propylene field magnets. The representative in vivo spectra obtained from
glycol for 30 min. The slides were transferred to a 85% propylene both HFD and chow diet fed rats are shown in Figure S1A and B.
glycol solution for 1 min, rinsed in distilled water for 2 times and The signals from methyl (0.9 ppm), n-methylene (1.30 ppm),
processed for hematoxylin counter staining. The steatosis/ allylic methylene (2.06 ppm), a-methylene (2.20 ppm) and olefinic
steatohepatitis was evaluated using a semi-quantitative scoring (5.30 ppm) groups are assigned in the spectra. Figure 2A shows the
system . The scoring was based on a chow diet group of liver fat content of chow diet and HFD fed rats for different age
animals for which if liver acini did not show lipid vacuoles, a score groups. Total liver fat contents of the HFD and chow diet fed
of zero was given (baseline). Acini having lipid vacuoles up to 33% animals at 12 weeks were 15.1861.47% and 2.1160.12%,
(mainly macrovesicular type) were considered as score 1. Acini respectively. At 18 weeks the fat content in HFD animals
with 34–66% of lipid vacuoles were scored 2 while acini having increased to 18.1460.91% compared to the chow diet group
over 66% of lipid vacuoles were ranked 3. 2.5560.23%. At 24 weeks the liver fat content increased to
20.9761.71% and 3.2960.13% for the HFD and chow diet
animals, respectively. The liver fat content was significantly
(P,0.001) higher in HFD animals compared to the chow diet
PLOS ONE | www.plosone.org 3 March 2014 | Volume 9 | Issue 3 | e91436
Hepatic Lipid Composition by In Vivo MRS and LC-MS
Table 1. Biochemical and body weight measurements of chow diet and HFD fed rats.
Biochemical measurements Chow diet animals High fat diet animals*
Cholesterol (mg/dL) 12th week 18th week 24th week 12th week 18th week 24th week
Glucose (mg/dL) 46.24 49.06 53.64 119.50 133.37 151.63
Insulin (ng/mL) 78.54 82.65 85.97 401.74 423.61 469.94
Body weight (gm) 102 115 122 136 148 164
1.14 1.31 1.40 4.26 4.71 5.21
231 274 342 278 371 487
The blood plasma cholesterol, triglycerides, glucose, insulin and body weight were significantly (P,0.05).
* higher in HFD fed rats.
group for all time points in the study. The liver fat of HFD fed rest of the TG concentrations were in the lower range of 0.8 to
animals increased significantly with age from 12 to 18 to 24 weeks. 33.01 nmol/mL but were still higher in HFD rats compared to
Figure 2B shows the unsaturation indices (UI) of chow diet and chow diet fed rats (P,0.05). In chow diet group of animals the
HFD animals at different ages. The UI of chow diet fed animals concentrations of all the saturated TGs were in the lower range of
was always higher than the HFD fed group at all ages (P,0.005). 0.1 to ,50 nmol/ml. The unsaturated TGs including 50:4, 52:5,
52:4, 54:5, 54:4, and 57:4 with concentration in the range 5 to
In vitro LC-MS Studies 102 nmol/ml in HFD rats were significantly (P,0.05) higher than
The liver was harvested after the terminal in vivo study at 24 chow diet fed rats. However, the poly-unsaturated triglycerides
54:7, 56:8, 56:7, 58:11, 58:10, 58:9, 58:8 and 60:10 were
weeks and tissue samples were subjected to LC-MS analysis. The significantly higher (P,0.05) in chow diet fed rats than in HFD
mean TG content in HFD animals (Figure 3A) was rats.
19836356 nmol/mL while it was significantly lower
(5116221 nmol/mL, P,0.001) for the chow diet group. In addition to the TGs, we also estimated the concentration of
Figure 3B and C show the two unsaturation indices of HFD and phospholipids [PC, PE, PI, PG, PS, PA] and sphingolipids [SM,
chow diet groups determined by considering as unsaturated FAs Cer, GluCer, GM3]. Both chow diet and HFD fed groups exhibit
either from n$1 or from n.3, respectively (both definitions given a similar concentration for each of the molecule studied (Figure
in the materials and methods section). The unsaturation index for S2A and B), except for lysophospholipids (LPC and LPE, Table 2).
the chow diet fed animals was significantly higher (P,0.002) when Total LPC levels, as well as the LPCs 18:1, 18:0 and 20:0 and
considering the poly-unsaturated fatty acids for its calculation. LPEs 16:0, 18:1 and 18:0 were significantly (P,0.05) higher in
HFD rats (Table 2). Cholesterol ester (CE) content evaluated using
Figure 4 shows the saturated and unsaturated TGs in HFD and LC-MS in HFD animals was higher 226650 nmol/mL than chow
chow diet groups. Individual concentrations are provided in Table diet group 1562 nmol/ml mL (P,0.001).
S1. The triglycerides 50:2, 50:1, 52:3, 52:2 and 52:1 were found to
be the most abundant (120.99 to 273.60 nmol/mL) and signifi- The PCA was performed on LC-MS data. The model optimally
cantly higher (P,0.05) in HFD rats than in the chow diet fed fitted the data with 96% and 3% of the total variance explained on
group. Concentrations of the triglycerides 48:2, 48:1, 50:3, 54:2, PC1 and PC2, respectively. The PCA score plot (figure not shown)
54:3 were in the range 56.73 to 85.44 nmol/mL in HFD rats. The of the quantitative LC-MS data showed well delineated clusters of
HFD and chow diet fed rats in the multidimensional space. The
Figure 1. OGTT measurements. Time course of oral glucose tolerance test measurements from (A) plasma glucose and (B) plasma insulin for chow
diet and HFD fed rats.
PLOS ONE | www.plosone.org 4 March 2014 | Volume 9 | Issue 3 | e91436
Hepatic Lipid Composition by In Vivo MRS and LC-MS
Figure 2. Liver fat and unsaturation indices estimated by in vivo MRS. Estimation of (A) liver fat (%) and (B) unsaturation indices by in vivo
MRS from chow diet and high fat diet fed animals at the 12th, 18th and 24th weeks. Liver fat content was significantly (P,0.001) higher in HFD fed rats
and unsaturation indices were significantly (P,0.005) higher in chow diet fed rats for all age groups.
loading plot (Figure S3) highlighted those key metabolites which displacing the nucleus or the presence of smaller well-defined
were predominantly accounted for variability along PC1 and PC2 intracytoplasmic droplets. Based on the scoring method , the
vectors. The PCA isolated the unsaturated TGs 52:2, 52:3, 52:1, HFD fed animals showed histopatholgical features of hepatic
54:2, 50:2, 50:1, 48:2, 48:1, 50:3, 54:3 on PC1 axis and were steatosis with a score of 3 indicating over 68% of acini were
present in higher concentrations. occupied by lipid vacuoles compared to chow diet fed animals.
Accumulation of lipid vacuoles was of both macro- and micro-
mRNA Analysis vesicular patterns. There were clear, well delineated and the
The real-time PCR mRNA expression analysis of various genes cytoplasmic vacuoles in the hepatocytes were of variable sizes with
mainly midzonal and paracentral distributions. There were
for both chow diet and HFD rats are shown in Figure 5. We found 6861.08% lipid vacuoles in the liver of HFD rats compared to
an increased expression (P,0.05) of mRNA levels of fatty acid 4.5961.54% only for the chow diet group. HFD animals displayed
translocase (CD36: 9 fold), peroxisome-proliferator-activated histopatholgical features of hepatic steatosis.
receptor a (PPARa: 2.35 fold), sterol regulatory element binding
factor (SREBF1: 2.40 fold), stearoyl-CoA desaturase 1 (SCD1: Discussion
2.20 fold), mitochondrial uncoupling protein 2 (UCP2: 4.17 fold)
in the HFD animal livers. Dietary intake of high amounts of saturated FAs and low
amounts of polyunsaturated fatty acids can cause NAFLD . In
Histopathologic Assessment of Hepatic Steatosis this study, we investigated the longitudinal changes of lipid
Representative H & E and Oil Red O stained sections of 24 composition in a high saturated fat diet fed rodent model using in
vivo MRS and LC-MS techniques.
weeks old rat liver from HFD and chow diet animals are showed in
Figure 6A and B, respectively. The fat deposition (lipid droplets) in Our HFD model showed significant weight gain starting at the
stained sections of the HFD fed animals is of mixed type including 9th week. The blood plasma analysis of the HFD animals showed
macrovesicular type where in either single large fat droplet hyperinsulinemia and hypertriglyceridemia confirming insulin-
Figure 3. Triglycerides and unsaturation indices estimated from LC-MS. Estimation of (A) total triglycerides (B) unsaturation indices (by
considering unsaturated FAs with n$1 and (C) unsaturation indices (by considering unsaturated FAs with n.3) at 24 weeks in high fat diet and chow
diet fed rats.
PLOS ONE | www.plosone.org 5 March 2014 | Volume 9 | Issue 3 | e91436
Hepatic Lipid Composition by In Vivo MRS and LC-MS
Figure 4. Saturated and unsaturated triglycerides by LC-MS. Concentrations of saturated and unsaturated TGs in HFD and chow diet fed rats
estimated by LC-MS at 24 weeks. Triglycerides C50:2, C50:1, C52:3, C52:2 and C52:1 were found to be most abundant (.100 nmol/ml) and
significantly higher in HFD fed rats compared to chow diet fed rats. Concentrations of other TGs were less than 100 nmol/ml but significantly higher
in HFD rats.
resistant conditions [29,30]. High saturated fat diet intervention HFD rat livers suggested enhanced fatty acid transport in the fatty
resulted in an increase of body weight, increase in triglycerides and liver resulting in an increased demand for the oxidation of fatty
cholesterol which lead to the metabolic consequence of insulin acids. Hepatic PPARa associated with the attenuation of insulin
resistance from the 12th week. The liver fat of HFD groups as signaling and hepatic steatosis was upregulated by 2.35 fold in
assessed by MRS, was higher by 13–17% compared to the HFD rats. The SREBF1 was elevated 2.4 fold in HFD fed rats
respective chow diet groups. This was confirmed by measuring the implicating the cause of insulin resistance and hepatosteatosis as
liver TG content by LC-MS and also on histology. Histology confirmed by histology. Hepatic mitochondrial oxidant production
results from HFD fed rats showed excessive accumulation of is one of the primary mechanisms that promotes oxidative stress
triglycerides with 68% of lipid vacuoles occupying the hepatocytes during NAFLD, NASH and type 2 diabetes conditions [31,32].
at 24 weeks. The 9-fold increase in CD36 mRNA expression in UCP2 is mitochondrial inner membrane protein present in variety
PLOS ONE | www.plosone.org 6 March 2014 | Volume 9 | Issue 3 | e91436
Hepatic Lipid Composition by In Vivo MRS and LC-MS
Table 2. Concentrations of LPCs and LPEs in liver tissues of of tissues including liver (hepatocytes) and its main function
chow diet and HFD fed rats at 24 weeks. involves the control of mitochondria-derived oxidant production
[33,34,35]. Studies on different tissues of animal models confirmed
LPC and LPE Chow diet (mmol/ml) HFD (mmol/ml) P value that the baseline expression of UCP2 gene in hepatocytes is
undetectable but it shows significant up-regulation in the liver
LPC18:0 0.00636561024 0.0086661024 0.005 during pathologic conditions associated with steatosis . The 4-
LPC18:1 0.00126661025 0.00236261024 ,0.005 fold increased expression of UCP2 levels in HFD group supported
LPC20:0 0.00276261024 0.0046561024 ,0.04 the involvement of UCP2 gene in the pathogenesis of NAFLD.
LPE16:0 0.01476161023 0.01846761024 ,0.005
LPE18:1 0.00236961025 0.00486561024 ,0.001 In addition, we showed that the FA composition of the liver is
LPE18:0 0.0106661024 0.0176961024 0.001 altered by HFD feeding. The unsaturation indices as determined
by in vivo MRS at 12, 18 and 24 weeks and by LC-MS at 24 weeks
Concentrations of lysophosphatidylcholines (LPC) and were significantly lower in HFD rats. The UIs estimated by MRS
lysophosphatidylethanolamines (LPE) were significantly higher in liver tissues of using resonances of olefinic, methyl, methylene and allylic
HFD fed rats. methylene at 24 weeks showed a 2.6 fold increase in chow diet
doi:10.1371/journal.pone.0091436.t002 fed rats. The UIs estimated by LC MS were 1.12 and 2.9 fold
higher in chow diet fed rats when considering both mono- and
poly-unsaturated FA and only polyunsaturated fatty acids,
respectively. This 2.9 fold higher UI obtained by calculating only
Figure 5. mRNA expression analysis. mRNA expression of (A) CD36, (B) SREBF1, (C) PPARa, (D) SCD1, and (E) UCP2. The expression of all these
genes were significantly higher for HFD rats than for chow diet fed animals.
PLOS ONE | www.plosone.org 7 March 2014 | Volume 9 | Issue 3 | e91436
Hepatic Lipid Composition by In Vivo MRS and LC-MS
Figure 6. Liver histology. (A) Oil red O staining and (B) Hematoxylin- degree of unsaturation in skeletal muscle of normal, overweight
eosin staining of HFD and chow diet fed rat liver tissues at 24 weeks. and obese human subjects  and bone marrow composition in
The HFD fed rats showed histopathological features of hepatic steatosis osteoporosis subjects . Relatively high levels of saturated fatty
with a score of 3 indicating over 68% of acini occupied by lipid vacuoles acids and low levels of polyunsaturated fatty acids are found in
compared to chow diet fed rats. individuals with insulin resistance and metabolic syndrome .
doi:10.1371/journal.pone.0091436.g006 Our study suggested that the same phenomenon was observed in
the liver of rats fed with a HFD and may also be relevant to the
polyunsaturated fatty acids is comparable with the MRS result (2.6 pathogenesis of NAFLD and insulin resistance associated with
fold). The in vivo MRS and in vitro LC-MS techniques provided high fat feeding.
complimentary information and had a good agreement for the
estimation of unsaturation in chow diet and HFD fed groups. Both We also noticed a decreasing trend of FA unsaturation in chow
methods demonstrated decreased unsaturation in HFD rats diet fed rats over the age which might be due to the redistribution
compared to chow diet rats. The accuracy of estimating the of the fatty acid composition. It was shown that ageing can alter
unsaturation by 1D MR methods can be improved by including liver mitochondrial membranes influencing the free radical
bi-allylic methylene (2.8 ppm) signal which is not well resolved. production and reduce unsaturation . During ageing there is
Localized 2D L-COSY technique  provides improved spectral a redistribution between types of unsaturated fatty acids resulting
resolution where J coupled multiplet resonances are dispersed over in transition from highly unsaturated fatty acids to less unsaturated
two spectral dimensions. The L-COSY techniques has been fatty acids . In addition, it was found that impairment in
utilized to estimate unsaturation using the cross peaks generated activity of delta-6-desaturase (D6D) due to ageing process results in
by the scalar couplings between olefinic and allylic, diallylic decreased synthesis of n-6 and n-3 polyunsaturated fatty acids
methylene protons  in skeletal muscle. This technology can be which is also a contributing factor in reducing the unsaturation
further developed for liver applications with appropriate motion index over time .
compensation and optimization of acquisition time within a
clinical setting. The multivariate analysis showed the triglycerides (52:2, 52:3,
52:1, 54:2, 50:2, 50:1, 48:2, 48:0, 48:1, 50:3, 54:3) accounted for
The decrease in unsaturation in the HFD group might be due to increased fat in the liver. Rhee et al.  showed that triglycerides
alteration in fatty acid composition because of changes in 48:0, 48:1 and 52:1 in human plasma of diabetic individuals are
desaturase expression or activities. Donelly et al. have shown that associated with increased risk of diabetes mellitus. Our current
dietary fatty acids contribute to the fatty acid pool in the liver . results suggests that the accumulation of these fatty acids in liver
As such, the high proportion of saturated fat and decreased tissues of HFD fed rats may also be important in increasing the risk
availability of polyunsaturated lipid components in the HFD of diabetes mellitus. Cholesterol ester hydrolase (CEH) plays a vital
contributed to the low unsaturation index. Degree of unsaturation role in hepatic cholesterol homeostasis and its activity is
can be influenced by the activity of desaturases. SCD1 catalyzes proportional to production of cholesterol ester. Three fold higher
the conversion of the saturated fatty acyl-CoAs, to their respective CEH activity was reported in patients with acute hepatitis
monounsaturated fatty acyl CoAs. The SCD1 was up regulated in compared to normal livers [46,47]. In the present study the
the livers of HFD mice by 2.2 fold. SCD1 up regulation excessive accumulation of cholesterol ester in HFD rats confirmed
contributed to increase in monounsaturated fatty acids (Table the steatosis condition of the liver indicating the risk of hepatitis.
S1). In spite of increased mono-unsaturation, the proportion of the Increased incorporation of saturated fat and cholesterol into the
available saturated fatty acids was still abundant in the HFD group cell membranes increase the membrane rigidity thereby reducing
resulting in lower unsaturation index. The upregulation of SCD1 the number of insulin receptors and their affinity to insulin,
in HFD rats confirmed its crucial role in the pathogenesis of diet- causing an insulin resistance in the local tissue. This increased
induced hepatic insulin resistance . It was established that rate- availability of hepatic saturated fatty acids in HFD fed animals
limiting nature of desaturation enzymes contribute to the reduces hepatic fatty acid oxidation and triglyceride export. This
development of the diabetic condition . A reduced degree of in turn would increase hepatic fatty acid and triglyceride synthesis
unsaturation was reported in the skeletal muscle of overweight and which would augment the triglyceride accumulation in the liver of
obese subjects . Earlier MRS studies showed the changes in HFD fed animals. LPC is an important signaling molecule with
diverse biological functions and involved in regulating cellular
inflammation [48,49]. Plasma, liver and skeletal muscle LPC levels
are increased in the obese diabetic db/db mouse and these
findings support that LPC may be involved in mediating insulin
resistance in obesity . Increased abundance of LPCs was
shown to induce hepatocellular death caused by mitochondrial
membrane depolarization [51,52].
The HFD fed rat livers showed increased concentration of
lysophospholipids (LPC and LPE). In particular, excessive
accumulation of LPC 18:1, 18:0, 20:0 and LPE 16:0, 18:0, 18:1
in HFD fed rats might be due to liver inflammation indicating the
increased risk of diabetes . Production of these lysopho-
spholipids in vivo is usually mediated by the release and/or
activation of the enzyme phospholipase A2 (PLA2) under
inflammatory conditions. PLA2 is present in neutrophilic granu-
locytes and its activation under inflammatory condition is also
associated with generation of reactive oxygen species. In a recent
clinical study, the LPE and LPCs were significantly higher in
inflammatory livers compared to healthy liver as confirmed by 31P
PLOS ONE | www.plosone.org 8 March 2014 | Volume 9 | Issue 3 | e91436
Hepatic Lipid Composition by In Vivo MRS and LC-MS
In conclusion, HFD increased the total fat fraction, and altered line (PC), phosphatidylethanolamine (PE), phosphatidylinositol
the fatty acid composition which might be due to the composition (PI), phosphatidylserine (PS). B. Glucocyl-ceramide (GluCer),
of the diet, or alterations in desaturase enzymes. The change in phosphatidic acids (PA), phosphatidylglycerol (PG), gangliosides
unsaturation is relevant to the pathogenesis of NAFLD and insulin mannoside 3(GM3) in HFD and chow diet fed rats.
resistance. Perturbations in lysophospholipids (LPC and LPE) (TIF)
might provide early information on oxidative stress/inflammation
in NAFLD before the incidence of diabetes. These markers may Figure S3 Multivariate analysis of lipid components.
be useful to study the link between a HFD and medical conditions Multivariate analysis of lipid components in HFD and chow diet
in humans. Further studies are required to explore the possibilities fed rats. Loading plot highlighted the key TGs contributing to the
of using unsaturation index in scaling the severity of the NAFLD maximum variance between the two groups.
in clinical practice. These studies could be extrapolated to evaluate (TIF)
the metabolic response during interventions including exercise and
drugs. Table S1 Concentrations of saturated and unsaturated
triglycerides. Concentrations of saturated and unsaturated
Supporting Information triglycerides in liver of chow diet and HFD fed rats at 24 weeks
Figure S1 In vivo liver spectra from HFD and chow diet
fed rats. Representative in-vivo liver spectra from (A) HFD and Acknowledgments
(B) chow diet fed rats. The signals from methyl (0.9 ppm), n-
methylene (1.30 ppm), allylic methylene (2.06 ppm), a-methylene We thank Drs. Guilhem Pages and Karthikeyan Narayanan for
(2.20 ppm) and olefinic (5.3 ppm) groups are assigned in the discussions.
(TIF) Author Contributions
Figure S2 Concentration of phospholipids and sphin- Conceived and designed the experiments: SSV MRW EST CJH SS.
gholipids from HFD and chow diet fed rats. A. Concen- Performed the experiments: JY VG OWK GS BDL. Analyzed the data: JY
trations of sphingomyelin (SM), ceramide (Cer), phosphatidylcho- VG OWK GS BDL. Contributed reagents/materials/analysis tools: SSV
MRW. Wrote the paper: JY EST MRW SSV.
16. Vessby B, Gustafsson IB, Tengblad S, Boberg M, Andersson A (2002)
1. Seppala-Lindroos A, Vehkavaara S, Hakkinen AM, Goto T, Westerbacka J, Desaturation and elongation of Fatty acids and insulin action. Ann N Y Acad
et al. (2002) Fat accumulation in the liver is associated with defects in insulin Sci 967: 183–195.
suppression of glucose production and serum free fatty acids independent of
obesity in normal men. J Clin Endocrinol Metab 87: 3023–3028. 17. Donnelly KL, Smith CI, Schwarzenberg SJ, Jessurun J, Boldt MD, et al. (2005)
Sources of fatty acids stored in liver and secreted via lipoproteins in patients with
2. Lakka HM, Laaksonen DE, Lakka TA, Niskanen LK, Kumpusalo E, et al. nonalcoholic fatty liver disease. J Clin Invest 115: 1343–1351.
(2002) The metabolic syndrome and total and cardiovascular disease mortality in
middle-aged men. JAMA 288: 2709–2716. 18. Nagarajan V, Gopalan V, Kaneko M, Angeli V, Gluckman P, et al. (2013)
Cardiac function and lipid distribution in rats fed a high-fat diet: in vivo
3. Marchesini G, Brizi M, Morselli-Labate AM, Bianchi G, Bugianesi E, et al. magnetic resonance imaging and spectroscopy. Am J Physiol Heart Circ Physiol
(1999) Association of nonalcoholic fatty liver disease with insulin resistance. 304: H1495–1504.
Am J Med 107: 450–455.
19. Yeung DK, Griffith JF, Antonio GE, Lee FK, Woo J, et al. (2005) Osteoporosis
4. Ikai E, Ishizaki M, Suzuki Y, Ishida M, Noborizaka Y, et al. (1995) Association is associated with increased marrow fat content and decreased marrow fat
between hepatic steatosis, insulin resistance and hyperinsulinaemia as related to unsaturation: a proton MR spectroscopy study. J Magn Reson Imaging 22: 279–
hypertension in alcohol consumers and obese people. J Hum Hypertens 9: 101– 285.
20. Shui G, Stebbins JW, Lam BD, Cheong WF, Lam SM, et al. (2011)
5. Zavaroni I, Mazza S, Dall’Aglio E, Gasparini P, Passeri M, et al. (1992) Comparative plasma lipidome between human and cynomolgus monkey: are
Prevalence of hyperinsulinaemia in patients with high blood pressure. J Intern plasma polar lipids good biomarkers for diabetic monkeys? PLoS ONE 6:
Med 231: 235–240. e19731.
6. Dunn W, Xu R, Wingard DL, Rogers C, Angulo P, et al. (2008) Suspected 21. Shui G, Lam SM, Stebbins J, Kusunoki J, Duan X, et al. (2013) Polar lipid
nonalcoholic fatty liver disease and mortality risk in a population-based cohort derangements in type 2 diabetes mellitus: potential pathological relevance of
study. Am J Gastroenterol 103: 2263–2271. fatty acyl heterogeneity in sphingolipids. Metabolomics 9: 786–799.
7. Fabbrini E, deHaseth D, Deivanayagam S, Mohammed BS, Vitola BE, et al. 22. Shui G, Guan XL, Low CP, Chua GH, Goh JS, et al. (2010) Toward one step
(2009) Alterations in fatty acid kinetics in obese adolescents with increased analysis of cellular lipidomes using liquid chromatography coupled with mass
intrahepatic triglyceride content. Obesity (Silver Spring) 17: 25–29. spectrometry: application to Saccharomyces cerevisiae and Schizosaccharo-
myces pombe lipidomics. Mol Biosyst 6: 1008–1017.
8. Manco M, Mingrone G, Greco AV, Capristo E, Gniuli D, et al. (2000) Insulin
resistance directly correlates with increased saturated fatty acids in skeletal 23. Shui G, Cheong WF, Jappar IA, Hoi A, Xue Y, et al. (2011) Derivatization-
muscle triglycerides. Metabolism 49: 220–224. independent cholesterol analysis in crude lipid extracts by liquid chromatogra-
phy/mass spectrometry: applications to a rabbit model for atherosclerosis.
9. Goodpaster BH, Wolf D (2004) Skeletal muscle lipid accumulation in obesity, J Chromatogr A 1218: 4357–4365.
insulin resistance, and type 2 diabetes. Pediatr Diabetes 5: 219–226.
24. Wei Y, Clark SE, Morris EM, Thyfault JP, Uptergrove GM, et al. (2008)
10. Shoelson SE, Herrero L, Naaz A (2007) Obesity, inflammation, and insulin Angiotensin II-induced non-alcoholic fatty liver disease is mediated by oxidative
resistance. Gastroenterology 132: 2169–2180. stress in transgenic TG(mRen2)27(Ren2) rats. J Hepatol 49: 417–428.
11. Dorfman SE, Laurent D, Gounarides JS, Li X, Mullarkey TL, et al. (2009) 25. Provencher SW (1993) Estimation of metabolite concentrations from localized in
Metabolic implications of dietary trans-fatty acids. Obesity (Silver Spring) 17: vivo proton NMR spectra. Magn Reson Med 30: 672–679.
26. Pearson K (1901) On Lines and Planes of Closest Fit to Systems of Points in
12. Zivkovic AM, German JB, Sanyal AJ (2007) Comparative review of diets for the Space. Philosophical Magazine 2.
metabolic syndrome: implications for nonalcoholic fatty liver disease. Am J Clin
Nutr 86: 285–300. 27. Abdi H, W LJ (2010) Principal component analysis. Wiley Interdisciplinary
Reviews: Computational Statistics 2.
13. Gaster M, Rustan AC, Beck-Nielsen H (2005) Differential utilization of saturated
palmitate and unsaturated oleate: evidence from cultured myotubes. Diabetes 28. Musso G, Gambino R, De Michieli F, Cassader M, Rizzetto M, et al. (2003)
54: 648–656. Dietary habits and their relations to insulin resistance and postprandial lipemia
in nonalcoholic steatohepatitis. Hepatology 37: 909–916.
14. Chavez JA, Knotts TA, Wang LP, Li G, Dobrowsky RT, et al. (2003) A role for
ceramide, but not diacylglycerol, in the antagonism of insulin signal transduction 29. Pratchayasakul W, Kerdphoo S, Petsophonsakul P, Pongchaidecha A,
by saturated fatty acids. J Biol Chem 278: 10297–10303. Chattipakorn N, et al. (2011) Effects of high-fat diet on insulin receptor function
in rat hippocampus and the level of neuronal corticosterone. Life Sci 88: 619–
15. Chavez JA, Summers SA (2003) Characterizing the effects of saturated fatty 627.
acids on insulin signaling and ceramide and diacylglycerol accumulation in 3T3-
L1 adipocytes and C2C12 myotubes. Arch Biochem Biophys 419: 101–109.
PLOS ONE | www.plosone.org 9 March 2014 | Volume 9 | Issue 3 | e91436
Hepatic Lipid Composition by In Vivo MRS and LC-MS
30. Pipatpiboon N, Pratchayasakul W, Chattipakorn N, Chattipakorn SC (2012) from long-lived species: the pigeon and human case. Mech Ageing Dev 86: 53–
PPARgamma agonist improves neuronal insulin receptor function in hippo- 66.
campus and brain mitochondria function in rats with insulin resistance induced 43. Pamplona R, Portero-Otin M, Riba D, Requena JR, Thorpe SR, et al. (2000)
by long term high-fat diets. Endocrinology 153: 329–338. Low fatty acid unsaturation: a mechanism for lowered lipoperoxidative
modification of tissue proteins in mammalian species with long life spans.
31. Pessayre D (2007) Role of mitochondria in non-alcoholic fatty liver disease. J Gerontol A Biol Sci Med Sci 55: B286–291.
J Gastroenterol Hepatol 22 Suppl 1: S20–27. 44. Bordoni A, Hrelia S, Lorenzini A, Bergami R, Cabrini L, et al. (1998) Dual
influence of aging and vitamin B6 deficiency on delta-6-desaturation of essential
32. Mailloux RJ, Harper ME (2011) Uncoupling proteins and the control of fatty acids in rat liver microsomes. Prostaglandins Leukot Essent Fatty Acids 58:
mitochondrial reactive oxygen species production. Free Radic Biol Med 51: 417–420.
1106–1115. 45. Rhee EP, Cheng S, Larson MG, Walford GA, Lewis GD, et al. (2011) Lipid
profiling identifies a triacylglycerol signature of insulin resistance and improves
33. Fleury C, Neverova M, Collins S, Raimbault S, Champigny O, et al. (1997) diabetes prediction in humans. J Clin Invest 121: 1402–1411.
Uncoupling protein-2: a novel gene linked to obesity and hyperinsulinemia. Nat 46. Simon JB, Poon RW (1978) Hepatic cholesterol ester hydrolase in human liver
Genet 15: 269–272. disease. Gastroenterology 75: 470–473.
47. Simon JB, Poon RW (1978) Studies on human hepatic cholesterol ester
34. Chavin KD, Yang S, Lin HZ, Chatham J, Chacko VP, et al. (1999) Obesity hydrolase in liver disease. Scand J Clin Lab Invest Suppl 150: 218–222.
induces expression of uncoupling protein-2 in hepatocytes and promotes liver 48. Xu Y (2002) Sphingosylphosphorylcholine and lysophosphatidylcholine: G
ATP depletion. J Biol Chem 274: 5692–5700. protein-coupled receptors and receptor-mediated signal transduction. Biochim
Biophys Acta 1582: 81–88.
35. Brand MD, Esteves TC (2005) Physiological functions of the mitochondrial 49. Xu Y, Fang XJ, Casey G, Mills GB (1995) Lysophospholipids activate ovarian
uncoupling proteins UCP2 and UCP3. Cell Metab 2: 85–93. and breast cancer cells. Biochem J 309 (Pt 3): 933–940.
50. Han MS, Lim YM, Quan W, Kim JR, Chung KW, et al. (2011)
36. Thomas MA, Yue K, Binesh N, Davanzo P, Kumar A, et al. (2001) Localized Lysophosphatidylcholine as an effector of fatty acid-induced insulin resistance.
two-dimensional shift correlated MR spectroscopy of human brain. Magn Reson J Lipid Res 52: 1234–1246.
Med 46: 58–67. 51. Kalous M, Rauchova H, Drahota Z (1992) The effect of lysophosphatidylcholine
on the activity of various mitochondrial enzymes. Biochim Biophys Acta 1098:
37. Velan SS, Durst C, Lemieux SK, Raylman RR, Sridhar R, et al. (2007) 167–171.
Investigation of muscle lipid metabolism by localized one- and two-dimensional 52. Basanez G, Sharpe JC, Galanis J, Brandt TB, Hardwick JM, et al. (2002) Bax-
MRS techniques using a clinical 3T MRI/MRS scanner. J Magn Reson type apoptotic proteins porate pure lipid bilayers through a mechanism sensitive
Imaging 25: 192–199. to intrinsic monolayer curvature. J Biol Chem 277: 49360–49365.
53. Puri P, Baillie RA, Wiest MM, Mirshahi F, Choudhury J, et al. (2007) A
38. Gutierrez-Juarez R, Pocai A, Mulas C, Ono H, Bhanot S, et al. (2006) Critical lipidomic analysis of nonalcoholic fatty liver disease. Hepatology 46: 1081–1090.
role of stearoyl-CoA desaturase-1 (SCD1) in the onset of diet-induced hepatic 54. Schober C, Schiller J, Pinker F, Hengstler JG, Fuchs B (2009) Lysophosphatidyl
insulin resistance. J Clin Invest 116: 1686–1695. ethanolamine is - in contrast to - choline - generated under in vivo conditions
exclusively by phospholipase A2 but not by hypochlorous acid. Bioorg Chem 37:
39. Horrobin DF (1993) Fatty acid metabolism in health and disease: the role of 202–210.
delta-6-desaturase. Am J Clin Nutr 57: 732S–736S; discussion 736S–737S.
40. Reznick AZ PL, Sen CK, Holloszy JO, Jackson MJ. (1998) Oxidative Stress in
Skeletal Muscle: Basel: Birkhauser Verlag.
41. Velan SS, Said N, Durst C, Frisbee S, Frisbee J, et al. (2008) Distinct patterns of
fat metabolism in skeletal muscle of normal-weight, overweight, and obese
humans. Am J Physiol Regul Integr Comp Physiol 295: R1060–1065.
42. Pamplona R, Prat J, Cadenas S, Rojas C, Perez-Campo R, et al. (1996) Low
fatty acid unsaturation protects against lipid peroxidation in liver mitochondria
PLOS ONE | www.plosone.org 10 March 2014 | Volume 9 | Issue 3 | e91436