Lipid in the livers of adolescents with nonalcoholic steatohepatitis: combined effects of pathways on steatosis.

Abstract

Fatty liver is a prerequisite for the development of nonalcoholic steatohepatitis (NASH). The homeostasis of hepatic lipid is determined by the dynamic balance of multiple pathways introducing lipids into or removing lipids from hepatocytes. We aim to study the different contributions of major lipid pathways to fat deposition in NASH livers. Expression of the lipid metabolism-related genes was analyzed by microarray and quantitative real-time polymerase chain reaction analysis. The expression levels of genes responsible for the rate-limiting steps of fatty acid uptake (CD36, FABPpm, SLC27A2, and SLC27A5), de novo synthesis (ACACB), oxidation (CPT-1), and very low-density lipoprotein (VLDL) secretion (ApoB) were used to evaluate the relative activity of each pathway. The expression levels for CD36 and CPT-1 were confirmed by Western blot analysis. Fatty acid uptake pathways were up-regulated to a higher degree than other pathways. The de novo synthesis pathway was also up-regulated more than both VLDL secretion and fatty acid oxidation pathways. In contrast to other NASH livers, one NASH liver exhibited lower ApoB and CPT-1 expression levels than normal controls. The increased fatty acid uptake and de novo synthesis were the most common causes for steatosis in NASH patients. In a rare case, impaired VLDL secretion and fatty acid oxidation contributed to the development of steatosis. Our study promises a simple method for the determination of why hepatic steatosis occurs in individual patients. This method may allow specific targeting of therapeutic treatments in individual patients.

Full text

Page 1

Lipid in the livers of adolescents with nonalcoholic
steatohepatitis: combined effects of pathways on steatosis
Lixin Zhua,⁎, Susan S. Bakera, Wensheng Liua, Meng-Hua Taob, Raza Patela,
Norma J. Nowakc,d, Robert D. Bakera,⁎
aDepartment of Pediatrics, Digestive Diseases and Nutrition Center, The State University of New York, Buffalo, NY 14214, USA
bDepartment of Social and Preventive Medicine, The State University of New York, Buffalo, NY 14214, USA
cDepartment of Biochemistry and the New York State Center of Excellence in Bioinformatics and Life Sciences,
The State University of New York, Buffalo, NY 14214, USA
dRoswell Park Cancer Institute, Microarray and Genomics Facility, Buffalo, NY 14263, USA
A R T I C L E I N F OA B S T R A C T
Article history:
Received 22 July 2010
Accepted 10 October 2010
Fatty liver is a prerequisite for the development of nonalcoholic steatohepatitis (NASH). The
homeostasis of hepatic lipid is determined by the dynamic balance of multiple pathways
introducing lipids into or removing lipids from hepatocytes. We aim to study the different
contributions of major lipid pathways to fat deposition in NASH livers. Expression of the
lipid metabolism–related genes was analyzed by microarray and quantitative real-time
polymerase chain reaction analysis. The expression levels of genes responsible for the rate-
limiting steps of fatty acid uptake (CD36, FABPpm, SLC27A2, and SLC27A5), de novo synthesis
(ACACB), oxidation (CPT-1), and very low-density lipoprotein (VLDL) secretion (ApoB) were
used to evaluate the relative activity of each pathway. The expression levels for CD36 and
CPT-1 were confirmed by Western blot analysis. Fatty acid uptake pathways were up-
regulated to a higher degree than other pathways. The de novo synthesis pathway was also
up-regulated more than both VLDL secretion and fatty acid oxidation pathways. In contrast
to other NASH livers, one NASH liver exhibited lower ApoB and CPT-1 expression levels than
normal controls. The increased fatty acid uptake and de novo synthesis were the most
common causes for steatosis in NASH patients. In a rare case, impaired VLDL secretion and
fatty acid oxidation contributed to the development of steatosis. Our study promises a
simple method for the determination of why hepatic steatosis occurs in individual patients.
This method may allow specific targeting of therapeutic treatments in individual patients.
© 2011 Elsevier Inc. All rights reserved.
M E T A B O L I S M C L I N I C A L A N D E X P E R I M E N T A L 6 0 ( 2 0 1 1 ) 1 0 0 1 – 1 0 1 1
Author contributions: L Zhu: conceived and designed the experiments, performed the experiments, analyzed the data, and wrote the
first draft. S Baker: conceived and designed the experiments, performed the experiments, analyzed the data, and revised the draft. W Liu:
performed the experiments, analyzed the data, and revised the draft. M-H Tao: analyzed the data and revised the draft. R Patel: performed
the experiments and revised the draft. N Nowak: performed the experiments and revised the draft. R Baker: conceived and designed the
experiments, performed the experiments, analyzed the data, and revised the draft.
⁎ Corresponding authors. Tel.: +1 716 829 2191; fax: +1 716 829 3585.
E-mail addresses: lixinzhu@buffalo.edu (L. Zhu), rbaker@upa.chob.edu (R.D. Baker).
0026-0495/$ – see front matter © 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.metabol.2010.10.003
available at www.sciencedirect.com
www.metabolismjournal.com

Page 2

1.Introduction
Nonalcoholic fatty liver disease (NAFLD) is the most common
cause of abnormal liver enzymes in the US population [1]. The
advanced form of NAFLD with inflammation or fibrosis is
termed nonalcoholic steatohepatitis (NASH). According to the
current “2-hit” hypothesis [2], steatosis is a prerequisite for
NASH patients to develop inflammation and fibrosis. Regard-
ing the etiology for steatosis, studies focusing on 1 or 2
pathways of hepatic lipid metabolism have led to conflicting
conclusions [3-9].
Homeostasis of hepatic lipids depends on the dynamic
balance of several pathways including fatty acid uptake, de
novo synthesis, oxidation, and very low-density lipoprotein
(VLDL) secretion (reviewed in Goldberg and Ginsberg [10],
Fabbrini et al [11], and Lavoie et al [12]). Therefore, to
understand the mechanism for excess lipid accumulation in
liver, all of the major pathways for lipid metabolism should be
studied in parallel. Because a convenient quantitative proteo-
mic method is not available, our first choice was a well-
characterized NASH microarray data set [13], complemented
by quantitative real-time polymerase chain reaction (qRT-
PCR). Here, the expression levels of the lipid metabolism
related genes were examined to test our hypothesis that fatty
livers are of varied etiology. We found that several abnormal-
ities in different lipid metabolism pathways collaboratively
contributed to the development of steatosis.
2.Methods
2.1.Patients
This study was approved by the Institutional Review Board of
the State University of New York at Buffalo. Only children and
adolescents were included in this study to ensure that our
patients were not sustained alcohol users. All our adolescent
patients claimed that they were not regular drinkers of
alcoholic beverages. We were assured by their parents that
these adolescent patients had no access to alcoholic bev-
erages. Even if our patients did ingest some alcohol, it could
not be consistent long-term alcohol use simply because of
their age. We considered other possible sources of alcohol
(aspartame, fruit), but a 3-day food record before any biopsy
failed to identify a significant source of alcohol. All NASH
patients included in this study had a body mass index (BMI)
greater than the 95th percentile for age and exhibited
significant insulin resistance (IR). Insulin resistance was
calculated based on the homeostasis model assessment
(HOMA) method [14]. Liver biopsies were obtained, with prior
written consent, from parents of patients suspected of having
NASH as part of regular medical care. Patients signed an
assent to the research. Diagnosis of NASH was based on
hepatic fat infiltration, inflammation, and fibrosis as revealed
by liver biopsy, following the criteria of Kleiner et al [15]. The
clinical information for the patients and normal controls (NCs)
subjected to qRT-PCR is listed in Table 1. Information for
patients and NCs subjected to microarray analysis was
described previously [13].
2.2. RNA extraction and microarray hybridization
Liver biopsies were stored in RNAlater before total RNA was
extracted with RNeasy and treated with RNase-free DNase I
set (Qiagen, Valencia, CA). RNA samples obtained from Admet
were also treated with RNase free DNase before downstream
experiments. Quality of the RNA samples was ensured with
Bioanalyzer (Agilent Technologies, Santa Clara, CA) before
downstream biotin labeling and hybridization to the CodeLink
Human Whole Genome Bioarray (GE Health Care–Amersham
Biosciences, Piscataway, NJ) following the manufacturer's
manual. The original microarray data have been uploaded to
the Gene Expression Omnibus Web site: http://www.ncbi.nlm.
nih.gov/geo/index.cgi. The accession numbers are GSM435821
to GSM435827 for NASH liver data sets and GSM435828 and
GSM435833 to GSM435835 for NC data sets.
2.3.Quantitative RT-PCR
Selected genes (Fig. 1A) were analyzed by qRT-PCR. Primers
(Table 2) were designed with the assistance of Primer 3 [16]
and BLAST [17] search. Complementary DNA was synthesized
with the iScript complementary DNA (cDNA) synthesis kit
(Bio-Rad Laboratories, Hercules, CA). Quantitative RT-PCR was
performed on an iCycler iQ real-time detection system (Bio-
Rad Laboratories) using Sybergreen (iQ SYBR Green Supermix,
Bio-Rad Laboratories) for real-time monitoring. For normali-
zation purposes, GAPD RNA levels were analyzed in parallel
with the genes of interest. The presence of a single specific
PCR product was verified by melting curve analysis (data not
shown) and confirmed on agarose gels (Fig. 1B).
The concentration of messenger RNA ([mRNA]) is repre-
sented by the following equation: [mRNA] = M/ECt, where
constant M is an arbitrary threshold, E is the efficiency of
PCR, and Ct is the threshold cycle. All PCR reactions had
efficiencies around 1.9, as determined experimentally with
Table 1 – Characteristics of NASH patients and NCs
RT-PCRWestern blot
NASH NCa
NASHNon-NASH
controle
Sex
(female-male)
Age (y)
BMI
Fasting insulin
(mU/mL)
Fasting glucose
(mmol/L)
IR (HOMA)d
12:152:44:53:1
9-18
37.4 ± 1.6b
23.3 ± 2.6
1-19
18.4 ± 1.3c
NA
11-18
32.7 ± 1.6b
20.7 ± 5.1
4-12
20.6 ± 2.9c
<2
5.4 ± 0.2 NA5.7 ± 0.44.7 ± 0.2
5.6 ± 0.7 NA5.7 ± 1.7 <1
NA indicates not available.
aNormal healthy liver intended for transplantation; no liver
disease reported.
bPatients are all greater than the 95th percentile of the population.
cNormal controls are all less than the 80th percentile of the
population.
dIR (HOMA) for healthy subjects is around 1.
eLiver biopsies from patients with hepatitis C, autoimmune
hepatitis, gall stone, and cystic fibrosis, respectively. They were free
from steatosis.
1002
M E T A B O L I S M C L I N I C A L A N D E X P E R I M E N T A L 6 0 ( 2 0 1 1 ) 1 0 0 1 – 1 0 1 1

Page 3

4-times serial diluted samples. The relative mRNA concen-
tration of each target gene was calculated as the mRNA
concentration of target gene normalized against that of
GAPD, as represented by the following equation: [mRNA]
target/[mRNA]GAPD= EGAPDCt.GAPD/EtargetCt.target.
2.4.Western blot analysis
Liver biopsies stored in RNAlater were homogenized in
phosphate-buffered saline and then boiled for 5 minute in
sodium dodecyl sulfate polyacrylamide gel electrophoresis
loading buffer. Samples (15 μg total protein each) were
separated on 10% sodium dodecyl sulfate polyacrylamide gel
electrophoresis gels. After blotting onto nitrocellulose mem-
branes, CD36 [18,19] (cat#sc-9154, Santa Cruz Biotechnology,
Santa Cruz, CA), carnitine palmitoyltransferase I (CPT-1) [20]
(cat#sc-98834, Santa Cruz Biotechnology), and β-actin [13]
(Clone C4; MP Biomedicals, Cleveland, OH) were probed. The
results were visualized using the SuperSignal West Dura
Extended Duration Substrate (Invitrogen, Carlsbad, CA) and
recorded with an image reader (LAS-1000; Fujifilm, Edison, NJ).
2.5. Statistical analysis
Characteristics and expression levels of genes in liver of NASH
cases and controls were compared using the nonparametric
test (Mann-Whitney U test) and the Student t test. All
statistical tests were based on 2-sided probability and a
significant level of P ≤ .05. Statistical analyses were conducting
using SAS, Version 9.2 (SAS Institute, Cary, NC).
Fig. 1 – Genes targeted in qRT-PCR. A, Schematic representation of the major pathways introducing lipid into or removing lipid
from liver. Four fatty acid transporters—CD36, FABPpm, SLC27A2, and SLC27A5—are known to mediate fatty acid uptake into
liver; ACACB catalyzes the rate-limiting step in fatty acid de novo synthesis; microsomal triglyceride transfer protein (MTP) is a
key protein in the secretion of VLDL, the secretion rate of which is determined by the availability of ApoB. The rate limiting step
for fatty acid oxidation in mitochondria is the translocation of fatty acid into mitochondria by CPT-1, which delivers fatty acids
to a myriad of enzymes including enoyl-CoA:hydratase 3-hydroxyacyl-CoA dehydrogenase (EHHADH). B, Agarose gel
electrophoresis of the qRT-PCR products. The end products of qRT-PCR were analyzed on a 3% agarose gel. The expected sizes
are: (1) GAPD, 110 base pairs (bp); (2) CD36, 133 bp; (3) FABPpm, 142 bp; (4) SLC27A2, 116 bp; (5) CPT-1, 132 bp; (6) EHHADH, 77 bp;
(7) MTP, 148 bp; (8) ApoB, 107 bp; (9) SLC27A5, 82 bp; and (10) ACACB, 107 bp. M is the GeneRuler 100-bp DNA Ladder, 100 to 1000
bp (Fermentas Life Sciences, Glen Burnie, MD).
1003
M E T A B O L I S M C L I N I C A L A N D E X P E R I M E N T A L 6 0 ( 2 0 1 1 ) 1 0 0 1 – 1 0 1 1

Page 4

3.Results
3.1.Gene expression for hepatic fatty acid uptake
Fatty acid uptake by hepatocytes depends on both the
concentration of plasma fatty acid and the capacity of the
uptake transporter systems [21]. The very elevated plasma
fatty acid concentration [22] found in NASH patients suggests
that the uptake transporter systems are saturated; therefore,
the capacity (quantity) of the fatty acid transporters is the
rate-limiting factor in fatty acid uptake.
Four different fatty acid transporters were reported on the
plasma membrane of hepatocytes. They are CD36 (also known
as fatty acid translocase) [23], plasma membrane–associated
fatty acid binding protein (FABPpm) [24], and solute carrier
family 27 member 2 (SLC27A2) [25] and member 5 (SLC27A5)
[26]. From the transcriptome data generated on the Codelink
whole genome microarray, increased signals were observed in
NASH livers for CD36 (NASH/NC=5.76, P = .004), FABPpm
(NASH/NC = 4.27, P = .005), SLC27A2 (NASH/NC = 2.27, P = .015),
and SLC27A5 (NASH/NC = 1.72, P = .08), although the latter did
not achieve statistical significance (Table 3). This increase was
observed in the context of similar gene transcription levels for
most of the 45 000 targeted sequences as represented by
several housekeeping genes (Table 3).
To validate these results, cDNA was prepared from a
different but overlapping set of NASH and normal livers for
qRT-PCR analysis (Fig. 2A-D). Increased expression of CD36,
FABPpm, SLC27A2,and SLC27A5inNASHliverswas confirmed.
Thedifferencesarestatisticallysignificantforallofthesegenes.
3.2.
metabolism pathways
Elevated expression of genes for other lipid
De novo synthesis is another important source of lipids in
liver, with the rate-limiting reaction being the carboxylation of
acetyl–coenzyme A (CoA), catalyzed by acetyl-CoA carboxyl-
ase β (ACACB). Because microarray analysis did not give good
signals forthisgene,ACACBexpressionlevelwasexamined by
qRT-PCR (Fig. 2E). A significant difference was observed
between NASH livers and NCs (NASH/NC = 3.8, P = .002). This
result was confirmed with a different primer pair for ACACB in
a similarly conducted qRT-PCR analysis (data not shown).
Many other genes involved in lipid synthesis (Table 3),
including fatty acid synthase gene, stearoyl-CoA desaturase
1 gene, and several acyltransferase genes, also exhibited
significantly elevated expression in NASH livers.
Similarly, elevated expression was observed for genes
involvedinfattyacidoxidation(Table3andFig.2F,G),although
statistical significance was not achieved for CPT-1, the rate-
limiting factor for fatty acid oxidation in mitochondria.
The VLDL secretion pathway was also elevated in NASH
livers at the gene expression level (Table 3 and Fig. 2H, I).
Statistical significance was achieved for apolipoprotein B
(ApoB) in qRT-PCR analysis, but not in microarray analysis.
The finding that all 4 hepatic lipid metabolism pathways
were more or less up-regulated was intriguing, but there were
2 immediate major concerns. First, were the gene expression
patterns consistent among NASH patients? Second, a com-
parisonof the extent to which each pathway was up-regulated
relative to the other lipid metabolism pathways had to be
made to understand the molecular basis of the fatty livers.
3.3.Lipid metabolism in NASH patient P112
In examining the gene expression patterns of individual NASH
livers and NCs, NASH liver P112 stood out because the gene
expression pattern was different from all other NASH livers.
Whereas other NASH livers exhibited high expression levels in
all of the 4 hepatic fatty acid transporters, NASH liver P112 only
had a high level in CD36. FABPpm, SLC27A2, and SLC27A5
exhibited similar expression levels as NCs (Fig. 3A-D). Similarly,
theACACBexpressionlevelinP112wasnothigherthanthatin
Table 2 – Primer pairs for qRT-PCR analysis
Symbol DescriptionSequence
GAPD Glyceraldehyde-3-phosphate dehydrogenaseAGCCTCAAGATCATCAGCAATG
ATGGACTGTGGTCATGAGTCCTT
GCCAAGGAAAATGTAACCCAGG
CCACAGCCAGATTGAGAA
ATCCCACGGGAGTGGACCCG
CGCACAGCCCAGGCATCCTT
GTATTGTGGCTGGTGCTACTC
CCGAAGCAGTTCACCGATA
AGAAGGCAACATGGGCTTAG
GGACAGCATTCGGAGGAG
TTGTGATGGTGACCCCCGAGGACCTTAAG
CGGGGATTCTCTTGGCAATGTCCACAATC
CTGGACTTCATTCCTGGAAAAAGAAG
CGATCTTGGCGTACATCGTTGTCATC
CCACGCAGAGGCTCAAGTT
GGAGAAGCTGGGTTCCTCTT
CTCTGCTTCATTTCCTCATATTCAGCTTC
CGGTAGCCCACGCTGTCTTGCAGTTTTCC
CAACCCTGAGGGCAAAGCCTTGCTG
CCTGCTTCCCTTCTGGAATGGCC
(Forward)
(Reverse)
(Forward)
(Reverse
(Forward)
(Reverse
(Forward)
(Reverse
(Forward)
(Reverse
(Forward)
(Reverse)
(Forward)
(Reverse)
(Forward)
(Reverse)
(Forward)
(Reverse)
(Forward)
(Reverse)
CD36 CD36 molecule (thrombospondin receptor)
FABPpm Plasma membrane fatty acid binding protein
(aspartate aminotransferase 2)
Solute carrier family 27 (fatty acid transporter), member 2 SLC27A2
SLC27A5 Solute carrier family 27 (fatty acid transporter), member 5
ACACBAcetyl-CoA carboxylase β
CPT-1 Carnitine palmitoyltransferase I
EHHADHEnoyl-CoA, hydratase/3-hydroxyacyl CoA dehydrogenase
MTP Microsomal triglyceride transfer protein
ApoB Apolipoprotein B
1004
M E T A B O L I S M C L I N I C A L A N D E X P E R I M E N T A L 6 0 ( 2 0 1 1 ) 1 0 0 1 – 1 0 1 1

Page 5

NCs (Fig. 3E). The more striking observation was that the
expression of CPT-1 (Fig. 3F) and ApoB (Fig. 3G) in P112 was
negligible compared with NCs, unlike all other NASH livers.
3.4.
liver P112
Lipid metabolism in all NASH livers except NASH
Because all major lipid metabolism pathways were elevated in
NASH livers compared with NCs, the relative increase among
all these pathways was compared to reveal their relative
contributions to steatosis. For this purpose, genes whose
products govern the rate-limiting step in each pathway were
compared among NASH livers and NCs.
The ratio of the gene expressions for CD36 vs CPT-1 (CD36/
CPT-1) was used to evaluate the relative up-regulation of the
fatty acid uptake pathway (CD36) vs the fatty acid oxidation
pathway (CD36 uptake/oxidation) in each individual subject.
Calculated from the qRT-PCR results, the CD36 uptake to
oxidation ratio was much greater in NASH patients than that
in NCs (Fig. 4A, NASH/NC = 13.0, P < .001). Similar results were
obtained for CD36/ApoB (Fig. 4B, NASH/NC = 15.9, P < .001),
indicating that the up-regulation of CD36-mediated fatty acid
uptake is far greater than that of fatty acid oxidation and
VLDL secretion.
The ratio of the gene expression for FABPpm vs CPT-1 was
greater in NASH livers than in NCs (Fig. 4C). However this
increase was small and did not achieve statistical significance
(NASH/NC = 1.8, P = .13). The ratio of the gene expression for
FABPpm vs ApoB was also modestly increased in NASH livers
withoutstatisticalsignificance(Fig.4D,NASH/NC= 1.7,P= .12).
Table 3 – Microarray analysis of gene expression between NASH livers and NCsa
GenBank
accession no.
Gene descriptionNASH liver
(n = 7)
Normal liver
(n = 4)
NASH/normalb
P
valuec
HousekeepingNM_000194.1 Hypoxanthine phosphoribosyltransferase
(HPRT)
Glucose-6-phosphate dehydrogenase (G6PD)
TATA box binding protein (TBP)
β-2-microglobulin (B2M)
Ribosomal protein L13a (RPL13A)
Actin, β (ACTB)
Tubulin, β1 (TUBB1)
2.99 ± 0.62 3.21 ± 0.62 0.93.804
NM_000402.2
NM_003194.2
NM_004048.2
NM_012423.2
NM_001101.2
NM_030773.1
1.58 ± 0.10
7.24 ± 0.44
167.17 ± 32.92
63.13 ± 19.19
18.19 ± 2.91
0.30 ± 0.08
1.53 ± 0.27
8.70 ± 1.33
194.66 ± 16.32
80.55 ± 15.34
16.07 ± 4.88
0.31 ± 0.07
1.03
0.83
0.86
0.78
1.13
0.96
.875
.364
.475
.497
.723
.915
Fatty acid
uptake
NM_003645.2 Solute carrier family 27 (fatty acid transporter),
member 2
Solute carrier family 27 (fatty acid transporter),
member 5
FABPpm; mitochondrial aspartate
aminotransferase
CD36 antigen
22.18 ± 3.529.79 ± 2.202.27 .015
NM_012254.1 19.12 ± 2.1211.13 ± 3.111.72 .080
NM_002080.1 42.05 ± 7.579.85 ± 3.05 4.27.005
NM_000072.116.55 ± 3.102.87 ± 1.635.76 .004
Lipid
synthesis
NM_004104.3
NM_005063.3
NM_003578.2
NM_178176.2
Fatty acid synthase (FASN)
Stearoyl-CoA desaturase 1 (SCD1)
Acyl-CoA:cholesterol acyltransferase 2
Acyl-CoA:monoacylglycerol acyltransferase
3 (MGAT3)
Acyl-CoA:monoacylglycerol acyltransferase
2 (MGAT2)
10.50 ± 2.34
34.59 ± 10.41
1.43 ± 0.26
4.61 ± 0.45
1.80 ± 0.65
3.80 ± 1.34
0.87 ± 0.21
2.00 ± 0.17
5.83
9.11
1.65
2.31
.009
.025
.130
.001
NM_025098.22.19 ± 0.340.55 ± 0.13 3.96 .002
Fatty acid
oxidation
NM_000098.1
NM_000387.2
NM_001608.1
NM_001966.1
Carnitine palmitoyltransferase II (CPT2)
Carnitine/acylcarnitine translocase
Long chain acyl-CoA dehydrogenase (ACADL)
Enoyl-CoA:hydratase 3-hydroxyacyl-CoA
dehydrogenase (EHHADH)
Mitochondrial 3-oxoacyl-CoA thiolase
Peroxisomal D3,D2-enoyl-CoA isomerase (PECI)
Branched chain acyl-CoA oxidase
Catalase (CAT), mRNA
12.12 ± 1.23
15.74 ± 2.62
8.61 ± 1.48
37.93 ± 5.45
3.49 ± 2.52
3.65 ± 0.99
2.50 ± 0.08
5.47 ± 1.43
3.47
4.32
3.45
6.93
.034
.003
.006
.001
NM_006111.1
NM_006117.1
NM_003500.1
NM_001752.1
49.23 ± 10.43
95.83 ± 13.01
140.15 ± 18.30
12.27 ± 2.16
21.29 ± 9.95
20.28 ± 3.02
73.41 ± 20.30
0.97 ± 0.39
2.31
4.73
1.91
12.70
.088
.001
.043
.002
VLDL
secretion
NM_000253.1
NM_000384.1
NM_001645.2
NM_000483.3
NM_000040.1
NM_001646.1
NM_000041.1
Microsomal triglyceride transfer protein (MTP)
Apolipoprotein B-100
Apolipoprotein C-I (APOC1)
Apolipoprotein C-II (APOC2)
Apolipoprotein C-III (APOC3)
Apolipoprotein C-IV (APOC4)
Apolipoprotein E (APOE)
46.19 ± 7.43
13.86 ± 3.19
614.75 ± 82.47
501.93 ± 30.57
625.93 ± 126.08
57.96 ± 6.55
353.83 ± 49.92
16.86 ± 4.24
7.62 ± 2.55
301.14 ± 61.67
288.10 ± 49.95
312.83 ± 67.97
15.61 ± 5.89
258.38 ± 53.30
2.74
1.82
2.04
1.74
2.00
3.71
1.37
.008
.161
.014
.013
.058
.001
.229
aThe gene expression levels (sample mean ± standard error) shown were median normalized.
bFold difference of gene expression levels (sample mean) between NASH liver tissues and normal liver controls.
cTwo-tailed student t test.
1005
M E T A B O L I S M C L I N I C A L A N D E X P E R I M E N T A L 6 0 ( 2 0 1 1 ) 1 0 0 1 – 1 0 1 1

Page 6

Similar results were obtained when examining fatty
uptake pathways mediated by SLC27A2 (Fig. 4E, F) and
SLC27A5 (Fig. 4G, H).
To evaluate the role of fatty acid de novo synthesis in
steatosis, the relative expression level of ACACB vs CPT-1 was
examined and found to be significantly higher in NASH livers
(Fig. 4I, NASH/NC = 2.8, P = .03), as was the relative expression
level of ACACB vs ApoB (Fig. 4J, NASH/NC = 2.4, P = .005).
However, the up-regulation of the ACACB expression level in
NASH livers was much lower than that of CD36 (compare with
Fig. 4A and B).
3.5.
NASH livers
Elevated protein expression of CD36 and CPT-1 in
To examine the expression of CD36 and CPT-1 at the protein
level, Western blot analyses were performed with the lysates
made from NASH liver biopsies (NASH, n = 9; control, n = 4). As
normal healthy liver tissue from adolescent is not available,
non-NASH liver biopsies free from steatosis were used as
controls. Blots were probed with antibodies specific for CD36
and CPT-1, respectively. A separate blot was also probed for β-
actin as a loading controls. Whereas all samples had similar
signals for actin, NASH patients exhibited higher expression of
CD36 and CPT-1 (Fig. 5A). Quantitation of the results with
National Institutes of Health Image software indicated that
there were significant increases in CD36 and CPT-1 proteins in
NASH livers: for CD36, NASH/NC = 6.7, P = .001; for CPT-1,
NASH/NC = 3.3, P = .001 (Fig. 5B).
4.Discussion
4.1.
common in NASH livers
Elevated fatty acid uptake and de novo synthesis are
Most of the lipid metabolism activity in liver can be grouped
into 4 major pathways: (1) fatty acid uptake, (2) de novo fatty
0.00
0.70
1.40
2.10
2.80
Normal NASH NormalNASHNormal NASH
NormalNASH Normal NASHNormalNASH
Normal NASHNormalNASH Normal NASH
0.00
0.15
0.30
0.45
0.60
0.00
0.60
1.20
1.80
2.40
0.00
0.60
1.20
1.80
2.40
0.00
2.00
4.00
6.00
8.00
0.00
0.04
0.08
0.12
0.16
0.00
0.15
0.30
0.45
0.60
0.00
0.90
1.80
2.70
3.60
0.00
0.50
1.00
1.50
2.00
A CD36
B FABPpm
C SLC27A2
D SLC27A5
E ACACB
F CPT-1
G EHHADH
H ApoB
I MTP
Gene expression levels with GAPD as reference
Fig. 2 – Quantitative RT-PCR analysis of the expression of genes involved in hepatic lipid metabolism. Gene expression levels of
(A) CD36, (B) FABPpm, (C) SLC27A2, (D) SLC27A5, (E) ACACB, (F) CPT-1, (G) EHHADH, (H) ApoB, and (I) MTP were analyzed by qRT-
PCR as described in “Methods.” Specific primer pairs are specified in Table 2. The cDNAs prepared from NASH patient samples
(n = 27) and NC samples (n = 6) were analyzed in duplicate. The gene expression level of each sample was normalized with that
of GAPD. Sample means of the gene expression levels were plotted with error bars indicating the standard errors. The fold
differences of gene expression between NASH livers and NCs (NASH/normal) and the P values of Student t tests are indicated.
Statistically significant increases in gene expression were observed in NASH livers for all genes tested except for (F) CPT-1.
1006
M E T A B O L I S M C L I N I C A L A N D E X P E R I M E N T A L 6 0 ( 2 0 1 1 ) 1 0 0 1 – 1 0 1 1

Page 7

acid synthesis, (3) oxidation of fatty acids, and (4) secretion of
VLDL. Abnormalities in any of these pathways may impair the
dynamic balance of the homeostasis of lipids in hepatocytes,
causing abnormal lipid depletion or accumulation. To study
major lipid metabolism pathways in NASH livers, we exam-
inedthese pathways at the sametime, taking advantageof the
-2
0
2
4
6
8
10
-0.1
0.0
0.1
0.2
0.3
0.4
-5
0
5
10
15
20
25
-1
0
1
2
3
4
5
-1
0
1
2
3
4
5
6
A CD36
B FABPpm
-0.5
C SLC27A2
0.0
0.5
1.0
1.5
-1
0
1
2
3
4
5
D SLC27A5
E ACACB
F CPT-1
G ApoB
Normal NASH P112
Normal NASH P112
Gene expression levels with GAPD as reference
Normal: n = 6
NASH: n = 26, all NASH patients
except P112.
P112: the NASH patient exhibiting
outstanding gene expression
pattern.
Fig. 3 – Two different expression patterns for lipid metabolism–related genes in NASH livers. Most of the NASH livers (26 of 27)
share a similar gene expression pattern with increased gene expression for all genes examined; the other liver (P112) exhibited
a very different pattern with several genes down-regulated. The gene expression levels of (A) CD36, (B) FABPpm, (C) SLC27A2,
(D) SLC27A5, (E) ACACB, (F) CPT-1, and (G) ApoB were compared among NCs and 2 groups of NASH livers. P112 was similar to
other NASH livers in that CD36 expression was increased, but the expression of all other genes examined differed from other
NASH livers. Note that the CPT-1 and ApoB gene expression levels in P112 are negligible comparing to the means of the NCs.
These data suggested that the down-regulated pathways for fatty acid oxidation (represented by the rate-limiting CPT-1) and
VLDL secretion (represented by the rate-limiting ApoB), together with the up-regulated fatty acid uptake (through CD36), caused
the development of steatosis in P112.
1007
M E T A B O L I S M C L I N I C A L A N D E X P E R I M E N T A L 6 0 ( 2 0 1 1 ) 1 0 0 1 – 1 0 1 1

Page 8

high throughput microarray technique complemented by
qRT-PCR assays.
In general, NASH livers exhibited elevated activity in all
lipid metabolism pathways. In the liver, 4 independent fatty
acid transporters were reported: CD36, FABPpm, SLC27A2, and
SLC27A5. Increased expression of all these transporters was
observed, with CD36 exhibiting the highest increase. However,
elevated geneexpressionwasalsoobservedforgenesinvolved
0.00
0.40
0.80
1.20
1.60
NormalNASH
0.00
0.25
0.50
0.75
1.00
NormalNASH
0.00
0.06
0.12
0.18
0.24
NormalNASH
0.00
0.25
0.50
0.75
1.00
NormalNASH
0.00
0.90
1.80
2.70
3.60
NormalNASH
B Uptake (CD36)/VLDL (ApoB)
D Uptake (FABPpm)/VLDL (ApoB)
F Uptake (SLC27A2)/VLDL (ApoB)
H Uptake (SLC27A5)/VLDL (ApoB)
J Synthesis (ACACB)/VLDL (ApoB)
0.00
8.00
16.00
24.00
32.00
NormalNASH
0.00
5.00
10.00
15.00
20.00
NormalNASH
0.00
1.20
2.40
3.60
4.80
NormalNASH
0.00
6.00
12.00
18.00
24.00
NormalNASH
0.00
25.00
50.00
75.00
100.00
Normal NASH
A Uptake (CD36)/Oxidation (CPT-1)
C Uptake (FABPpm)/Oxidation (CPT-1)
E Uptake (SLC27A2)/Oxidation (CPT-1)
G Uptake (SLC27A5)/Oxidation (CPT-1)
I Synthesis (ACACB)/Oxidation (CPT-1)
Relative Gene expression levels
1008
M E T A B O L I S M C L I N I C A L A N D E X P E R I M E N T A L 6 0 ( 2 0 1 1 ) 1 0 0 1 – 1 0 1 1

Page 9

in fatty acid de novo synthesis, fatty acid oxidation, and VLDL
secretion. By comparing the gene expression levels of the
genes governing the rate-limiting step in each pathway, it
became clear that the up-regulation of fatty acid uptake
(through CD36) is more pronounced than other lipid metab-
olism pathways, suggesting a major role for CD36 in the
development of steatosis. Our data also showed that fatty acid
uptake mediated by SLC27A5, SLC27A2, and FABPpm, respec-
tively, was up-regulated to a greater degree than the pathways
of fatty acid oxidation or VLDL secretion, suggesting that these
fatty acid uptake transporters also contribute to the develop-
ment of fatty livers, albeit with a smaller contribution
compared with CD36. The elevated expression for CD36 and
CPT-1 at the protein level was also observed in NASH livers
(NASH, n = 9; control, n = 4), consistent with the elevated
mRNAlevelsfor thesegenes. Itis noteworthy that the increase
in the protein level of CD36 in NASH livers was higher than
that of CPT-1, consistent with the mRNA results.
The elevated gene transcriptions of these fatty acid
transporters (Table 3 and Fig. 2) were consistent with the
previous studies where individual transporters were exam-
ined [25,26]. Our new finding is that increases in the
expression of fatty acid transporters were greater than those
of the fatty acid oxidation and VLDL secretion pathways.
Our data also revealed that fatty acid de novo synthesis
was up-regulated. Our results are consistent with those
previously reported by several groups that the expression of
genes involved in de novo fatty acid synthesis is increased in
NASH livers [6,7]. Furthermore, we demonstrated that this
increase in genes involved in de novo synthesis is greater
than the increase in genes involved in fatty acid oxidation
and VLDL secretion. The elevated de novo synthesis could be
explained by the elevated insulin levels in our NASH patients
[10]. Our previous work suggests that alcohol metabolism
also contributes to the elevated lipid synthesis in NASH
livers [13].
Lipid can also be introduced into liver through lipoprotein
receptors including low-density lipoprotein (LDL) receptor-
related proteins and LDL receptor. A recent study with
nonobese patients carrying apolipoprotein C3 gene variants
(C482T, T455C, or both) [27] implicated elevated activity of
the LDL receptor in steatosis. These mutations are
Fig. 4 – Comparison of the gene expression levels among lipid metabolism pathways. To compare different lipid metabolism
pathways, genes whose products govern the rate-limiting step in each lipid metabolism pathway were considered. Four fatty
acid transporters (CD36, FABPpm, SLC27A2, and SLC27A5) mediate the rate-limiting steps of fatty acid uptake. ACACB, CPT-1,
and ApoB were the protein factors mediating the rate-limiting step for de novo fatty acid synthesis, fatty acid oxidation, and
VLDL secretion, respectively. A, Comparison between fatty acid uptake (CD36) and fatty acid oxidation (CPT-1) pathways. The
ratio of the gene expression levels for CD36 and CPT-1 was calculatedfor each NASH liver (n = 26, excluding P112) and NC (n = 6).
This ratio is significantly greater in NASH liver than in NC liver, indicating that the up-regulation of fatty acid uptake is greater
than the up-regulation of fatty acid oxidation. B, Comparison between fatty acid uptake (CD36) and VLDL secretion (ApoB)
pathways. The ratio of the gene expression levels for CD36 and ApoB was calculated for each NASH liver and NC. This ratio is
significantly greater in NASH livers than in NCs, indicating that the up-regulation of fatty acid uptake is greater than the up-
regulation of VLDL secretion. C, Comparison between fatty acid uptake (FABPpm) and fatty acid oxidation (CPT-1) pathways.
D, Comparison between fatty acid uptake (FABPpm) and VLDL secretion (ApoB) pathways. E, Comparison between fatty acid
uptake (SLC27A2) and fatty acid oxidation (CPT-1) pathways. F, Comparison between fatty acid uptake (SLC27A2) and VLDL
secretion (ApoB) pathways. G, Comparison between fatty acid uptake (SLC27A5) and fatty acid oxidation (CPT-1) pathways.
H, Comparison between fatty acid uptake (SLC27A5) and VLDL secretion (ApoB) pathways. I, Comparison between de novo fatty
acid synthesis (ACACB) and fatty acid oxidation (CPT-1) pathways. J, Comparison between de novo fatty acid synthesis (ACACB)
and VLDL secretion (ApoB) pathways. Note that the expression levels were always greater in fatty acid uptake pathways and de
novo synthesis pathway than in fatty acid oxidation pathway and VLDL secretion pathway, although statistical significance
was not achieved in panels C and F.
Fig. 5 – Elevated expression of CD36 and CPT-1 proteins in
NASH livers. A, Western blot analyses were performed with
lysates prepared from liver biopsies of NASH and non-NASH
patients. The non-NASH controls were of normal BMI and
free from steatosis. The blots were probed for CD36, CPT-1,
and β-actin as these proteins. Whereas similar signals for
actin weredetected for all samples, the NASHlivers exhibited
stronger signals for both CD36 and CPT-1. B, The Western
blot results were quantitated with National Institutes of
Health Image software. The normalized quantities of the
CD36 and CPT-1 signals were the densities of the CD36 and
CPT-1 bands divided by those of the β-actin bands, respec-
tively. The mean values for NASH and non-NASH patients
were plotted with error bars representing the standard error
of the means.
1009
M E T A B O L I S M C L I N I C A L A N D E X P E R I M E N T A L 6 0 ( 2 0 1 1 ) 1 0 0 1 – 1 0 1 1

Page 10

associated with higher levels of fasting plasma apolipopro-
tein C3 concentration and NAFLD. It is known that excess
apolipoprotein C3 could inhibit lipoprotein lipase leading to
hypertriglyceridemia, which leads to steatosis through
increased activity of the LDL receptor. However, examination
of our microarray data indicated that none of these receptors
were significantly changed in NASH livers at the mRNA level
(data not shown).
Therefore, we conclude that elevated fatty acid uptake and
de novo synthesis are the common pathways leading to
steatosis in NASH patients and that the CD36 mediated fatty
acid uptake is the dominant cause.
Many of the genes studied here are regulated by the
transcriptional factors peroxisome proliferator-activated
receptors and sterol regulatory element binding proteins.
However, examination of the microarray data indicated that
none of these transcriptional factors were differentially
expressed in NASH livers compared with NCs (data not
shown). Although this finding needs to be confirmed, a
different mechanism may cause the elevated gene expression
in the NASH livers.
4.2.
oxidation and VLDL secretion
NASH livers commonly exhibit elevated fatty acid
Nonalcoholic steatohepatitis livers also featured elevated
fatty acid oxidation and VLDL secretion, which consume/
remove lipid from liver. The enhanced activities of fatty acid
oxidation and VLDL secretion pathways indicated that the
NASH livers tried but failed to reduce the lipid load toward a
normal level.
It is argued that impaired fatty acid oxidation could lead
to steatosis [8,9]. Blocking fatty acid oxidation by inhibiting
CPT-1 induces steatosis in a mouse model [28]. Yet our data
suggested an opposite scenario for NASH patients. Our
microarray data suggested a significant increase of the
CPT-1 expression in NASH livers. Moderate increase of
CPT-1 was also observed by qRT-PCR, although it was not
statistically significant. Nevertheless, Western blot analysis
suggested that CPT-1 protein level is higher in NASH livers.
Overall, fatty acid oxidation seemed to be increased in
mitochondria and peroxisomes in our NASH patients, albeit
to a level insufficient to compensate for the import/
synthesis of fatty acids. Our data are consistent with the
report that obese ob/ob mice exhibit elevated fatty acid
oxidation [29]. However, it is noteworthy that all our NASH
patients, being children and adolescents, have relatively
mild fibrosis compared with adult NASH patients. Therefore,
our conclusion may not apply to adult NASH patients, as
more severe fibrosis is likely associated with more extensive
damage in mitochondria and decreased activity of mito-
chondria oxidation.
Charlton et al [5] reported that all 7 NASH patients in
their study groups had a consistent lower level of expression
of ApoB than normal and obese (non-NASH) subjects. Their
results suggested that impaired VLDL secretion is a universal
phenomenon in NASH. However, a recent study by Fujita
et al [30] led to the opposite conclusion that the VLDL
secretion by NASH liver is significantly higher than that of
NCs. As ApoB is the critical protein component of VLDL and
the availability of ApoB is rate limiting in the assembly and
secretion of VLDL, our result that NASH livers exhibited
elevated ApoB expression is in harmony with the report of
Fujita et al.
4.3. Atypical lipid metabolism in NASH patient P112
In examining the gene expression patterns of individual
subjects, one NASH liver (P112) did not fit into the general
picture described above. P112 had elevated expression of CD36.
However, this subject did not exhibit elevated gene expression
for other fatty acid transporters. Neither did this subject exhibit
elevated expression in genes responsible for fatty acid de novo
synthesis, fatty acid oxidation, or VLDL secretion. Instead, a
much lower expression (lower than both NCs and NASH livers)
was observed for ApoB gene and CPT-1, indicating impaired
VLDL secretion and fatty acid oxidation. Apparently, the
suppressed VLDL secretion and fatty acid oxidation could lead
to lipid accumulation in hepatocytes and worsen the situation
causedbyelevatedactivityofCD36.Althoughrare(1of27NASH
subjects), this particular patient is very important because the
patient represents a different gene expression pattern and
suggests a distinct etiology for steatosis.
TheinterestingmRNAexpressionlevelofP112promptedus
to revisit the pathologic data for this patient and other NASH
patients. However, the BMI, serum markers for lipid, glucose,
and steatosis and fibrosis stages of this patient are similar to
other NASH patients. High levels of VLDL were found in many
oftheNASHpatientsincludingP112(datanotshown).Wewere
surprised that the serum VLDL level of P112 was not lower, as
the liver VLDL pathway was down-regulated, at least at the
mRNAlevel.Possibleexplanationsforthisinconsistencyareas
follows: (1) VLDL is also produced in the intestine, and so VLDL
production by the intestine might compensate for a decrease
in production by the liver; and (2) the serum level may also be
affected by the consumption of VLDL, which could also be
down-regulated. Further study with a larger study group is
needed to determine the prevalence of NASH livers with
suppressed fatty acid oxidation and VLDL secretion.
One concern about the strategy of our study is whether the
“rate-limiting” steps at normal physiologic condition were
still rate limiting in NASH livers. The fact that many genes
were activated in one pathway suggested that many steps are
rate limiting in NASH livers. Nevertheless, the “rate-limiting”
enzymes showed a profound increase among all the enzymes
in the same pathways. Likely, these enzymes are those of the
rate-limiting steps and are at least good representations of
the pathways.
In summary, comparison of the major pathways for lipid
metabolism in NASH livers suggested that up-regulated fatty
acid uptake and de novo synthesis are the accompanying
causes for lipid accumulation in most NASH livers. Evidence is
also provided to show that impaired VLDL secretion together
with suppressed fatty acid oxidation could contribute to the
development of steatosis. This is the first evidence at the
molecular level that more than one pathway leading to
steatosis is dysregulated in NASH patients. Our study pro-
mises a simple method for the accurate identification of the
cause for steatosis in individual patients. This method may
1010
M E T A B O L I S M C L I N I C A L A N D E X P E R I M E N T A L 6 0 ( 2 0 1 1 ) 1 0 0 1 – 1 0 1 1

Page 11

allow us to specifically target therapeutic treatments for
individual patients.
Acknowledgment
This work was supported by an unrestricted grant from the
Peter and Tommy Fund, Buffalo, NY. The funder had no role in
study design, data collection and analysis, decision to publish,
or preparation of the manuscript.
R E F E R E N C E S
[1] Clark JM, Brancati FL, Diehl AM. The prevalence and etiology
of elevated aminotransferase levels in the United States. Am J
Gastroenterol 2003;98:960-7.
[2] Day CP, James OF. Steatohepatitis: a tale of two “hits”?
Gastroenterology 1998;114:842-5.
[3] Villanueva CJ, Monetti M, Shih M, et al. Specific role for acyl
CoA:diacylglycerol acyltransferase 1 (Dgat1) in hepatic stea-
tosis due to exogenous fatty acids. Hepatology (Baltimore, Md
2009;50:434-42.
[4] Diraison F, Moulin P, Beylot M. Contribution of hepatic de
novo lipogenesis and reesterification of plasma non esterified
fatty acids to plasma triglyceride synthesis during non-
alcoholic fatty liver disease. Diabetes Metab 2003;29:478-85.
[5] Charlton M, Sreekumar R, Rasmussen D, et al. Apolipoprotein
synthesis in nonalcoholic steatohepatitis. Hepatology (Balti-
more, MD) 2002;35:898-904.
[6] Kohjima M, Enjoji M, Higuchi N, et al. Re-evaluation of fatty
acid metabolism-related gene expression in nonalcoholic
fatty liver disease. Int J Mol Med 2007;20:351-8.
[7] Mitsuyoshi H, Yasui K, Harano Y, et al. Analysis of hepatic
genes involved in the metabolism of fatty acids and iron in
nonalcoholic fatty liver disease. Hepatol Res 2009;39:366-73.
[8] Pessayre D, Fromenty B. NASH: a mitochondrial disease.
J Hepatol 2005;42:928-40.
[9] Rao MS, Reddy JK. Peroxisomal beta-oxidation and steatohe-
patitis. Semin Liver Dis 2001;21:43-55.
[10] Goldberg IJ, Ginsberg HN. Ins and outs modulating hepatic
triglyceride and development of nonalcoholic fatty liver
disease. Gastroenterology 2006;130:1343-6.
[11] Fabbrini E, Sullivan S, Klein S. Obesity and nonalcoholic fatty
liver disease: biochemical, metabolic, and clinical implica-
tions. Hepatology (Baltimore, MD) 2010;51:679-89.
[12] Lavoie JM, Gauthier MS. Regulation of fat metabolism in the
liver: link to non-alcoholic hepatic steatosis and impact of
physical exercise. Cell Mol Life Sci 2006;63:1393-409.
[13] Baker SS, Baker RD, Liu W, et al. Role of alcohol metabolism in
non-alcoholic steatohepatitis. PLoS ONE 2010;5:e9570.
[14] Matthews DR, Hosker JP, Rudenski AS, et al. Homeostasis
model assessment: insulin resistance and beta-cell function
from fasting plasma glucose and insulin concentrations in
man. Diabetologia 1985;28:412-9.
[15] KleinerDE,BruntEM,VanNattaM,etal.Designandvalidation
of a histological scoring system for nonalcoholic fatty liver
disease. Hepatology (Baltimore, MD) 2005;41:1313-21.
[16] Rozen S, Skaletsky H. Primer3 on the WWW for general users
and for biologist programmers. Methods Mol Biol 2000;132:
365-86.
[17] Altschul SF, Gish W, Miller W, et al. Basic local alignment
search tool. J Mol Biol 1990;215:403-10.
[18] Liang CP, Han S, Okamoto H, et al. Increased CD36 protein as a
response to defective insulin signaling in macrophages. J Clin
Invest 2004;113:764-73.
[19] Yabe U, Sato C, Matsuda T, et al. Polysialic acid in human
milk. CD36 is a new member of mammalian polysialic acid-
containing glycoprotein. J Biol Chem 2003;278:13875-80.
[20] Richmond SR, Carper MJ, Lei X, et al. HIV-protease inhibitors
suppress skeletal muscle fatty acid oxidation by reducing
CD36 and CPT1 fatty acid transporters. Biochim Biophys Acta
2010;1801:559-66.
[21] Bradbury MW. Lipid metabolism and liver inflammation. I.
Hepatic fatty acid uptake: possible role in steatosis. Am J
Physiol Gastrointest Liver Physiol 2006;290:G194-G198.
[22] de Almeida IT, Cortez-Pinto H, Fidalgo G, et al. Plasma total
and free fatty acids composition in human non-alcoholic
steatohepatitis. Clin Nutr 2002;21:219-23.
[23] Stahlberg N, Rico-Bautista E, Fisher RM, et al. Female-
predominant expression of fatty acid translocase/CD36 in rat
and human liver. Endocrinology 2004;145:1972-9.
[24] Martin GG, Danneberg H, Kumar LS, et al. Decreased liver fatty
acid binding capacity and altered liver lipid distribution in
mice lacking the liver fatty acid-binding protein gene. J Biol
Chem 2003;278:21429-38.
[25] DiRusso CC, Li H, Darwis D, et al. Comparative biochemical
studies of the murine fatty acid transport proteins (FATP)
expressed in yeast. J Biol Chem 2005;280:16829-37.
[26] Doege H, Baillie RA, Ortegon AM, et al. Targeted deletion of
FATP5 reveals multiple functions in liver metabolism:
alterations in hepatic lipid homeostasis. Gastroenterology
2006;130:1245-58.
[27] Petersen KF, Dufour S, Hariri A, et al. Apolipoprotein C3 gene
variants in nonalcoholic fatty liver disease. N Engl J Med 2010;
362:1082-9.
[28] Grefhorst A, Hoekstra J, Derks TG, et al. Acute hepatic
steatosis in mice by blocking beta-oxidation does not
reduce insulin sensitivity of very-low-density lipoprotein
production. Am J Physiol Gastrointest Liver Physiol 2005;289:
G592-G598.
[29] Brady LJ, Brady PS, Romsos DR, et al. Elevated hepatic
mitochondrial and peroxisomal oxidative capacities in fed
and starved adult obese (ob/ob) mice. Biochem J 1985;231:
439-44.
[30] Fujita K, Nozaki Y, Wada K, et al. Dysfunctional very-low-
density lipoprotein synthesis and release is a key factor in
nonalcoholic steatohepatitis pathogenesis. Hepatology 2009;
50:772-80.
1011
M E T A B O L I S M C L I N I C A L A N D E X P E R I M E N T A L 6 0 ( 2 0 1 1 ) 1 0 0 1 – 1 0 1 1