Hepatic Cholesterol Homeostasis
Is the Low-Density Lipoprotein Pathway a Regulatory or a
Shunt Pathway?
Allan D. Sniderman, Yanqin Qi, Cheng-I J. Ma, Rui Hao Leo Wang, Mark Naples, Chris Baker,
Jing Zhang, Khosrow Adeli, Robert S. Kiss
Downloaded from http://atvb.ahajournals.org/ by guest on June 16, 2017
Objective—The hypothesis that cholesterol that enters the cell within low-density lipoprotein (LDL) particles rapidly
equilibrates with the regulatory pool of intracellular cholesterol and maintains cholesterol homeostasis by reducing
cholesterol and LDL receptor synthesis was validated in the fibroblast but not in the hepatocyte. Accordingly, the present
studies were designed to compare the effects of cholesterol that enters the hepatocyte within an LDL particle with those
of cholesterol that enters via other lipoprotein particles.
Approach and Results—We measured cholesterol synthesis and esterification in hamster hepatocytes treated with LDL
and other lipoprotein particles, including chylomicron remnants and VLDL. Endogenous cholesterol synthesis was not
significantly reduced by uptake of LDL, but cholesterol esterification (280%) and acyl CoA:cholesterol acyltransferase 2
expression (870%) were increased. In contrast, cholesterol synthesis was significantly reduced (70% decrease) with other
lipoprotein particles. Furthermore, more cholesterol that entered the hepatocyte within LDL particles was secreted within
VLDL particles (480%) compared with cholesterol from other sources.
Conclusions—Much of the cholesterol that enters the hepatocyte within LDL particles is shunted through the cell and
resecreted within VLDL particles without reaching equilibrium with the regulatory pool. (Arterioscler Thromb Vasc
Biol. 2013;33:2481-2490.)
Key Words: cholesterol ◼ 3-hydroxy-3-methylglutaryl-coenzyme A ◼ high-density lipoproteins
◼ very–low-density lipoproteins
T
he low-density lipoprotein (LDL) receptor pathway plays
a major role in the regulation of the plasma level of LDL
in humans. When clearance of LDL particles through the LDL
receptor pathway is markedly impaired or abolished, as in
familial hypercholesterolemia,1 profound elevations in plasma
LDL result. At the other extreme, more effective clearance
of LDL particles by the LDL receptor pathway is the major
mechanism by which statins lower plasma LDL. However,
these changes in the activity of the LDL receptor pathway are
produced by disease or pharmacological intervention and are
not evidence of its physiological role.
The LDL receptor pathway paradigm stipulates that intracellular cholesterol homeostasis is based on a reciprocal
relationship between the rate at which cholesterol within an
LDL particle enters the cell through the LDL receptor pathway and the rate at which cholesterol and LDL receptors are
synthesized within the cell.2 In brief, cholesterol that enters
the cell within an LDL particle is released from the particle
in the lysosome and equilibrates with the cholesterol in the
endoplasmic reticulum (ER) membrane. The cholesterol mass
within the ER membrane determines the activity of the sterol regulatory element–binding protein 2 (SREBP2) pathway, which regulates the synthesis of cholesterol and the
LDL receptor.3–5 Additional, more rapid, adaptation mechanisms exist, including esterification of cholesterol by acyl
CoA:cholesterol acyltransferase (ACAT) and acute inhibition of 3-hydroxy-3-methyl-glutaryl-coenzyme A reductase
(HMGCoA reductase, the rate-limiting step of endogenous
cholesterol biosynthesis) by hydroxysterols.6,7
The regulatory paradigm of the LDL receptor pathway is
a simple, closed-loop, seesaw homeostatic model: increased
uptake is followed by decreased synthesis; decreased synthesis is followed by increased uptake. However, this paradigm
was based on studies in cultured fibroblasts, a cell that plays
no important role in total body cholesterol homeostasis and
a cell with little capacity to secrete cholesterol. In contrast,
Received on: August 25, 2011; final version accepted on: August 13, 2013.
From the Department of Medicine, Cardiovascular Research Laboratories, Royal Victoria Hospital, McGill University, Montreal, Quebec, Canada (A.D.S.,
Y.Q., C.J.M., R.H.L.W., R.S.K.); Molecular Structure and Function, Research Institute, The Hospital for Sick Children, Toronto, Ontario, Canada (M.N.,
C.B., J.Z., K.A.); and Department of Biochemistry and Laboratory Medicine, University of Toronto, Toronto, Ontario, Canada (M.N., C.B., J.Z., K.A.).
The online-only Data Supplement is available with this article at http://atvb.ahajournals.org/lookup/suppl/doi:10.1161/ATVBAHA.113.301517/-/DC1.
Correspondence to Robert S. Kiss, 687 Pine Ave West, Royal Victoria Hospital, Room H7.02, Montreal, Quebec H3A 1A1, Canada. E-mail
[email protected]
© 2013 American Heart Association, Inc.
Arterioscler Thromb Vasc Biol is available at http://atvb.ahajournals.org
2481
DOI: 10.1161/ATVBAHA.113.301517
2482 Arterioscler Thromb Vasc Biol November 2013
Nonstandard Abbreviations and Acronyms
apoB
apolipoprotein B
CE
cholesteryl ester
CR
chylomicron remnants
ER
endoplasmic reticulum
HDL
high-density lipoprotein
HMGCoA reductase3-hydroxy-3-methyl-glutaryl-coenzyme
A reductase
LPDS
lipoprotein-deficient serum
LDL
low-density lipoprotein
PCSK9
proprotein convertase subtilisin/kexin type 9
SREBP2
sterol regulatory element–binding protein 2
VLDL
very–low-density lipoprotein
Downloaded from http://atvb.ahajournals.org/ by guest on June 16, 2017
the hepatocyte is at the center of all the major cholesterol
fluxes within the organism, with multiple lipoprotein receptor–mediated endocytic pathways that contribute cholesterol
to the total hepatocyte cholesterol pool. There are also multiple outputs of cholesterol from the hepatocyte: cholesterol
can be metabolized to bile acids, as well as dissolved in bile;
cholesterol and cholesterol ester can be secreted within apolipoprotein B (apoB)–containing very low-density lipoprotein
(VLDL) particles; and finally cholesterol can be transferred
from the hepatocyte plasma membranes via ATP-binding cassette transporter A1 and ATP-binding cassette transporter G1
to apolipoprotein A-I and particles of high-density lipoprotein
(HDL) as well as secreted directly with HDL.8,9
Although the LDL receptor pathway plays an important
role in determining the concentration of plasma LDL, it does
not follow that the LDL receptor pathway paradigm accurately describes cholesterol homeostasis in the liver. There is
evidence that it does not, in that LDL receptor activity and
cholesterol synthesis have been documented to persist in the
face of increased delivery of LDL particles to the liver.10–19
Based on this evidence, we propose that LDL-derived cholesterol is handled differently by the hepatocyte, invoking separate regulatory and homeostatic mechanisms. We now show
that much of the cholesterol that enters the hepatocyte within
an LDL particle is resecreted, either within a VLDL particle
or via HDL, without ever entering the regulatory pool of cholesterol. This allows cholesterol and LDL receptor synthesis
to continue in the face of the entry of large amounts of cholesterol into the hepatocyte via LDL particles, and it means
that the LDL receptor pathway is principally functioning as a
shunt rather than a regulatory pathway.20
Materials and Methods
Materials and Methods are available in the online-only Supplement.
Results
CR Versus LDL Uptake
Preliminary experiments demonstrated, as anticipated, relatively similar binding but significantly greater uptake of chylomicron remnants (CR) compared with LDL particles (1 hour:
6.86±0.48 versus 12.24±3.74; 4 hours: 20.30±2.80 versus
51.37±3.34; 8 hours: 31.37±2.02 versus 74.47±7.59 cpm/ng
cell protein; all significant differences [P<0.05] by Student t
test; Figure I in the online-only Data Supplement). Accordingly,
the relative concentrations of CR and LDL in all subsequent
experiments were adjusted to ensure that the total cholesterol
(in LDL or CR) taken up by the cells was equivalent.
Regulatory Mechanisms
Cholesteryl ester (CE) synthesis was measured after uptake
of 3H-cholesterol–labeled LDL particles versus uptake of
3
H-cholesterol–labeled CR particles in cholesterol-depleted
and cholesterol-loaded hamster hepatocytes. Under both conditions, formation of CE was substantially less when cells are
incubated with CR than when they are incubated with LDL.
As would be expected, formation of CE from CR cholesterol
was substantially greater in cholesterol-loaded cells than in
cholesterol-depleted hepatocytes. Under both conditions, formation of CE from exogenous cholesterol was substantially
greater from LDL than CR (3.44±0.69 versus 2.11±0.21%
of total radioactivity in cholesterol-loaded cells, Figure 1A;
3.45±0.43 versus 0.42±0.11% of total radioactivity in cholesterol-depleted cells, Figure 1B). These results indicate that
more of the cholesterol within LDL is esterified than the cholesterol within CR. These findings were confirmed by measuring CE formation based on incorporation of [14C]-oleate
into CE. Formation of CE in cholesterol-loaded cells was significantly greater with LDL than with CR (14.4±5.6 versus
1.27±0.14 cpm/ng; Figure 1C).
Effects of LDL- and CR-Derived Cholesterol
on Synthesis of Cholesterol and CE
One mechanism to regulate total cellular cholesterol levels is
the downregulation of endogenous cholesterol synthesis by
uptake of exogenous cholesterol. The de novo synthesis of cholesterol was measured by incorporation of [3H]-mevalonate.
Equivalent amounts of exogenous cholesterol were taken up
by the hepatocytes, and LDL-treated cells synthesized 3-fold
more cholesterol than CR-treated cells (2.85±0.30 versus
0.86±0.15 cpm/ng; Figure 1D). In addition, more newly synthesized cholesterol was esterified when cells are incubated
with LDL compared with CR particles (6.53±1.11 versus
1.96±0.44%; Figure 1E). These results demonstrate that LDLderived cholesterol does not downregulate de novo synthesis
as CR-derived cholesterol does. Furthermore, more newly
synthesized cholesterol is esterified in LDL-treated cells compared with CR-treated cells, suggesting higher ACAT activity
in LDL-treated cells.
All these observations are consistent with the hypothesis
that more of the CR-derived cholesterol comes into equilibrium with the regulatory pool than the cholesterol that is liberated from LDL particles. These results are consistent with the
observation that the intracellular route followed by CR particles differs from that followed by LDL particles.
Fluorescence Microscopy of LDL and CR
Fluorescent hydrophobic dyes (3,3-dioctadecyloxacarbocyanine perchlorate [DiO] and 1,1ʹ-dioctadecyl-3,3,3ʹ,3ʹtetramethylindodicarbocyanine,
4-chlorobenzenesulfonate
[DiD]) are irreversibly incorporated into separate lipoproteins
Sniderman et al LDL-Derived Cholesterol Is Shunted Into VLDL 2483
Downloaded from http://atvb.ahajournals.org/ by guest on June 16, 2017
Figure 1. Low-density lipoprotein (LDL)
cholesterol is preferentially esterified in
hepatocytes. [3H]-cholesterol-labeled
LDL (black bar) and chylomicron remnants (CR; white bar) were added to
hepatocytes grown in fetal bovine serum–
containing media (A) or lipoproteindeficient serum–containing media (B) for
24 h, and then radioactive cholesteryl
ester (CE) was determined. Results are
expressed as the average percent CE
of the total radioactive cholesterol±SD.
*P<0.05, **P<0.001, compared with the
LDL condition. C, [14C]-oleate was incubated with hepatocytes treated with LDL
(black bar) or CR (white bar) for 24 h to
measure incorporation into CE. Results
are expressed as the average cpm/ng cell
protein±SD. **P<0.001, compared with
the LDL condition. D, [3H]-mevalonate
was incubated with hepatocytes treated
with LDL (black bar) or CR (white bar) for
24 h, and then radioactive cholesterol
was determined. Results are expressed
as cpm/ng cell protein±SD. **P<0.001,
compared with the LDL condition.
E, [3H]-mevalonate was incubated with
hepatocytes treated with LDL (black
bar) or CR (white bar) for 24 h, and then
radioactive CE was determined. Results
are expressed as average percent CE
of the total radioactive cholesterol±SD.
*P<0.05, compared with the LDL condition. ACAT indicates acyl CoA:cholesterol
acyltransferase.
and then incubated with primary hamster hepatocytes for
increasing periods of time. During that time, these hydrophobic dyes traffic through the hepatocyte within the lipoprotein
particle into which they are incorporated. Unlike the natural
lipid components of the lipoprotein particles, the hydrophobic
dyes are reliable guides until they aggregate within the lysosomes (≤1 hour). Within the first 15 minutes, DiO-CR (green)
and DiD-LDL (red) are in separate and distinct endosomes
(Figure 2). By 1 hour, there was partial overlap of DiO-CR
and DiD-LDL in lysosome compartments (as detected by
colocalization with lysotracker blue). These results indicate
that during the initial phases of internalization, the vast majority of LDL and CR particles take separate endocytic routes.
LDL-Specific Effect on Hepatic
Cholesterol Homeostasis
The initial series of experiments suggested that metabolic
consequences for hepatic cholesterol homeostasis differed
after uptake of LDL particles versus CR particles. To test
this hypothesis further, we extended the initial observations
using a variety of other lipoproteins, including human chylomicrons, CR, HDL, VLDL, and mouse β-VLDL. The effects
of lipoprotein-deficient serum (LPDS), LDL and non-lipoprotein–derived free cholesterol were also studied.
Increasing concentrations of LDL, VLDL, chylomicrons, CR, β-VLDL, and HDL were labeled with 2 μCi of
[3H]-cholesteryl oleate and added to hepatocytes for 24 hours.
Incorporation of radioactivity into the hepatocytes increased
over time (Figure IIA in the online-only Data Supplement).
The concentration of each lipoprotein where the equivalent
amount of radioactivity was incorporated into the hepatocytes after 24 hours was considered to be the equivalent concentration, and mass experiments were conducted to ensure
that equivalent cholesterol mass loading was achieved. After
loading hepatocytes with the lipoprotein-derived and nonlipoprotein–derived cholesterol, we measured the total mass
of cholesterol within the hepatocyte by 3 different methods: (1) total cholesterol and free cholesterol kits (Roche
Diagnostics); (2) high-performance liquid chromatography
separation and quantification of free and cholesteryl ester;
and (3) gas chromatographic separation and quantification of
free and cholesteryl ester. All 3 methods confirmed that the
incubation of hepatocytes with lipoprotein and non-lipoprotein–derived cholesterol resulted in equivalent cholesterol
loading (Figure 3A; Figure IIB and IIC in the online-only
Data Supplement). As a negative control, LPDS-treated hepatocytes contained the lowest mass of cholesterol (Figure 3A).
Incubation with LDL, chylomicrons, CR, VLDL, β-VLDL,
HDL, and non-lipoprotein–derived cholesterol resulted in a
significant and almost identical increase in cholesterol mass
compared with the LPDS control. These results establish that
uptake of exogenous cholesterol from multiple lipoprotein and
non-lipoprotein sources increases the total mass of cholesterol
within the cell to a similar extent. For this reason, we can
presume that, under the loading conditions used here for this
and subsequent experiments, the hepatocytes were cholesterol
2484 Arterioscler Thromb Vasc Biol November 2013
Downloaded from http://atvb.ahajournals.org/ by guest on June 16, 2017
Figure 2. Low-density lipoprotein (LDL) and chylomicron
remnants (CR) particles have separate endocytic itineraries.
A, 1,1ʹ-dioctadecyl-3,3,3ʹ,3ʹ-tetramethylindodicarbocyanine,
4-chlorobenzenesulfonate (DiD)-labeled LDL (red) and DiOlabeled CR (green) were incubated with the same hepatocytes
for 15 min and visualized by fluorescence microscopy. B, After
1 h, all lipoproteins are found in acidic lysosomal compartments
(blue); LDL (red, yellow arrow) and CR (green, white arrow) particles are found with minor overlap (purple arrow).
loaded to an equivalent degree, and any differences noted are
not because of differences in the degree of cholesterol loading.
Importantly, uptake of non-lipoprotein–derived cholesterol,
presumably to the plasma membrane, resulted in the same
amount of cholesterol loading, thereby serving as a suitable
positive control.
Under the conditions described above, each of these exogenous lipoprotein-derived and non-lipoprotein–derived cholesterol sources was labeled with an equivalent amount of
[3H]-cholesterol, and a cholesterol esterification assay was
performed. LDL-derived cholesterol promoted a significant
increase in CE formed compared with other sources of cholesterol (Figure 3B). Interestingly, VLDL-derived cholesterol
also led to increased CE synthesis that was significantly greater
than human chylomicrons, human CR, and mouse β-VLDL,
but not as high as LDL, perhaps caused by contaminant LDL
because the VLDL sample was generated by sequential density gradient ultracentrifugation or perhaps because VLDL
itself may bind to the LDL receptor. In any case, the appearance of exogenous cholesterol in CE from LDL was considerably greater than for any of the other sources of cholesterol.
Endogenous cholesterol synthesis was measured by incorporation of [3H]-mevalonate into free and esterified cholesterol. As a positive control, hepatocytes treated with LPDS,
which does not contain significant exogenous cholesterol,
had the highest radioactive cholesterol formed (Figure 4).
Endogenous cholesterol synthesis was virtually as high when
LDL-derived cholesterol was added, and this amount was significantly greater than the amount of endogenous cholesterol
synthesis when the other sources of lipoprotein were added
to the medium. Interestingly, VLDL also resulted in a small
Figure 3. Total cellular cholesterol mass increases with lipoprotein loading. A, Hepatocytes were cholesterol loaded with
the different lipoproteins (low-density lipoprotein [LDL], human
chylomicrons [Chylo], chylomicron remnants [CR], β-very–lowdensity lipoprotein [β-VLDL], very–low-density lipoprotein [VLDL],
high-density lipoprotein [HDL], and non-lipoprotein–associated
free cholesterol), and then the total cholesterol mass was
measured using the high-performance liquid chromatography
method. Results are presented as the mean±SD of quadruplicate
measurements. Cholesterol loading increased total cholesterol
mass, free cholesterol (gray bar), and cholesteryl ester (CE; black
bar), and hepatocytes were almost identically cholesterol loaded
with the different lipoproteins. *P<0.05, compared with the
lipoprotein-deficient serum (LPDS) condition. B, Total CE formed
is increased with LDL cholesterol loading. Hepatocytes grown
in LPDS were treated with the different lipoproteins (containing
[3H]-cholesterol) for 24 h. Then, cellular lipids were extracted and
separated by thin layer chromatography. Radioactivity associated with the CE band was measured by scintillation counting,
and the results are presented as the mean±SD of quadruplicate
measures of 3 independent experiments. *P<0.05, **P<0.001,
compared with the LPDS condition.
increase in de novo–synthesized cholesterol, suggesting that
VLDL-derived cholesterol partially mimics the effect of
LDL-derived cholesterol. These results are consistent with the
hypothesis that cholesterol released from LDL does not enter
the regulatory pool as does the cholesterol from the other lipoprotein particles, and therefore, LDL-derived cholesterol does
not downregulate endogenous cholesterol synthesis.
The amount of newly synthesized cholesterol that was
incorporated into CE was also determined for each condition.
CE synthesis mirrors cholesterol synthesis (Figure 4). For the
positive control, treatment with LPDS resulted in the highest percent of CE formed from de novo–synthesized cholesterol. That cholesterol ester synthesis is upregulated in LPDS
medium, in which there is no source of exogenous cholesterol
and, therefore, endogenous cholesterol synthesis is increased,
is anticipated. LDL treatment also resulted in significantly
Sniderman et al LDL-Derived Cholesterol Is Shunted Into VLDL 2485
ACAT Activity and ACAT1 and ACAT2 Expression
Downloaded from http://atvb.ahajournals.org/ by guest on June 16, 2017
Figure 4. Total de novo–synthesized cholesterol was increased
with low-density lipoprotein (LDL) cholesterol loading. Hepatocytes grown in lipoprotein-deficient serum (LPDS) were treated
with the different lipoproteins and an equivalent amount of [3H]mevalonate for 24 h. Then, cellular lipids were extracted and
separated by thin layer chromatography. Radioactivity associated with the free cholesterol band (gray bar) and the cholesteryl
ester band (black bar) was measured by scintillation counting,
and the results are presented as the mean±SD of quadruplicate
measures of 3 independent experiments. *P<0.05, **P<0.001,
compared with LPDS condition. β-VLDL indicates β-very–lowdensity lipoprotein; Chylo, human chylomicrons; CR, chylomicron
remnants; HDL, high-density lipoprotein; and VLDL, very–lowdensity lipoprotein.
increased CE formation. However, addition of chylomicrons,
CR, VLDL, β-VLDL, HDL, and non-lipoprotein–derived
cholesterol resulted in a significantly lower percentage of CE
formed. These results are consistent with the hypothesis that
exogenous cholesterol released from LDL particles does not
downregulate endogenous cholesterol synthesis as effectively
as other lipoprotein and non-lipoprotein sources of cholesterol, and incorporation of both exogenous and endogenous
cholesterol into cholesterol ester under conditions of LDLderived cholesterol loading is substantially greater than with
other lipoprotein particles.
The enzyme responsible for esterification of cholesterol intracellularly is ACAT (also known as SOAT). In hepatocytes,
there are 2 independent enzymes (ACAT1 and ACAT2) with
putative separate physiological functions and location. These
experimental results to date are consistent with a model that
posits that a substantial portion of the cholesterol in LDL that
is internalized, trafficked to, and released from the lysosome
tends to be rapidly exposed to ACAT, whereas cholesterol
from other lipoprotein and non-lipoprotein sources is not.
Accordingly, to further explicate the role of ACAT activity
in these different responses, under the same conditions as in
Figure 1A, CE synthesis was measured after treatment with
[3H]-cholesterol–labeled LDL or CR without and with a total
ACAT inhibitor (58035). As observed previously, CE synthesis was substantially greater on addition of LDL compared
with CR (Figure IV in the online-only Data Supplement).
Importantly, addition of the ACAT inhibitor resulted in a
significant decrease in CE formed to the same background
level for LDL and CR, demonstrating that the increase in CE
synthesis on LDL treatment is entirely dependent on ACATmediated newly esterified cholesterol.
These findings suggest that LDL might induce the activity of
either or both ACAT1 and ACAT2. To determine whether this
was the case, Western blots of cell lysates from LDL-, CR-,
and other lipoprotein-treated hepatocytes were performed.
There was no difference in ACAT1 expression between cells
treated with different lipoproteins (Figure 5; Figure V in the
online-only Data Supplement). However, ACAT2 expression
was significantly upregulated in LDL-treated cells (8.7-fold)
compared with LPDS-treated cells. ACAT2 expression was
also significantly upregulated in VLDL-treated cells (4.4fold) although not to the same degree as LDL. These results
suggest that LDL treatment particularly upregulates ACAT2
mRNA Expression of HMGCoA
Reductase and LDL Receptor
SREBP2 responds to intracellular cholesterol levels in the
regulatory pool in the ER membrane by controlling the mRNA
expression of HMGCoA reductase and the LDL receptor. To
confirm the measured activities of endogenous cholesterol
synthesis, we measured the mRNA expression by quantitative polymerase chain reaction. Incubation with LDL resulted
in an mRNA expression level of HMGCoA reductase similar
to treatment with LPDS (Figure III in the online-only Data
Supplement). Incubation with other lipoproteins or with nonlipoprotein–derived cholesterol resulted in a significantly lower
mRNA expression level, suggesting that the cholesterol derived
from these treatments did enter the regulatory pool. LDL receptor mRNA was similar to HMGCoA reductase mRNA with
regard to treatments, except to a lesser extent. Therefore, these
results indicate that LDL-derived cholesterol does not enter the
regulatory pool to diminish the SREBP2-mediated expression
of HMGCoA reductase and LDL receptor mRNA.
Figure 5. Acyl CoA:cholesterol acyltransferase 2 (ACAT2) expression is upregulated by low-density lipoprotein (LDL) treatment.
Hepatocytes were treated with lipoprotein-deficient serum
(LPDS), very–low-density lipoprotein (VLDL), chylomicrons,
LDL, chylomicron remnants (CR), β-very–low-density lipoprotein
(β-VLDL), high-density lipoprotein (HDL), or non-lipoprotein–
derived cholesterol for 24 h, and then cell lysates were prepared
and Western blots performed for acyl CoA:cholesterol acyltransferase 1 (ACAT1), ACAT2, LDL receptor, proprotein convertase
subtilisin/kexin type 9 (PCSK9), and inducible degrader of LDL
receptor (IDOL). GAPDH was used as a loading control. Western
blots from 3 separate experiments were performed, and a representative Western blot is shown here. Quantification is shown in
Figure V in the online-only Data Supplement.
2486 Arterioscler Thromb Vasc Biol November 2013
Downloaded from http://atvb.ahajournals.org/ by guest on June 16, 2017
expression, despite the fact that an equivalent amount of cholesterol was delivered by other treatments.
Important regulators of LDL receptor activity and expression on the cell surface of the hepatocyte are proprotein convertase subtilisin/kexin type 9 (PCSK9) and inducible degrader
of LDL receptor (IDOL). Both proteins, although they are
regulated differently and act mechanistically differently, contribute to the downregulation, internalization, and degradation
of LDL receptors. We performed Western blots of the LDL
receptor, PCSK9, and IDOL of cell lysates from hepatocytes
treated with different lipoproteins (Figure 5; Figure V in the
online-only Data Supplement). LDL receptor expression was
relatively increased in VLDL- and LDL-treated hepatocytes
(2.8- and 6.3-fold, respectively) compared with LPDS, consistent with the lack of an effect of LDL- or VLDL-derived
cholesterol on the cholesterol regulatory pool to downregulate
LDL receptor expression. However, chylomicrons, HDL, and
non-lipoprotein–derived cholesterol had lower LDL receptor expression, as expected (1.1-, 0.7-, and 0.9-fold, respectively, compared with LPDS control). Interestingly, CR- and
β-VLDL–treated hepatocytes had increased LDL receptor
expression compared with the LPDS control (5.6- and 5.8-fold,
respectively). We postulated that expression levels of PCSK9
and IDOL may determine the LDL receptor expression levels (Figure 5; Figure V in the online-only Data Supplement).
Interestingly, PCSK9 levels increased moderately in VLDLand LDL-treated cells compared with LPDS (1.1- and 3.7fold, respectively), despite increased LDL receptor levels.
However, PCSK9 protein levels were inversely correlated with
LDL receptor levels in chylomicron-, CR-, β-VLDL-, HDLand non-lipoprotein–derived cholesterol-treated hepatocytes,
explaining LDL receptor expression levels. IDOL was significantly upregulated in all lipoprotein-treated cells in comparison with LPDS (Figure 5; Figure V in the online-only Data
Supplement), demonstrating that IDOL is not a factor explaining the increased LDL receptor expression levels. However,
expression of PCSK9 does explain the expression level of the
LDL receptor in most lipoprotein-treated hepatocytes but not
in LDL- and VLDL-treated hepatocytes. Therefore, LDLinduced changes in hepatocytes represent a unique homeostatic mechanism not explained by PCSK9 or IDOL.
Double Labeling
Because experiments with LDL and CR were performed
in separate primary cell cultures and then compared, it was
important to extend these results by performing LDL and
CR treatment experiments in the same cells. For this reason, we used double labeling of the same hepatocytes with
[3H]-cholesterol and [14C]-cholesterol labels. Assuring similar labeling efficiency, the results were that 14.98±2.92%
of [3H]-cholesterol derived from LDL was incorporated
into CE, whereas 4.69±0.99% of [14C]-cholesterol derived
from CR was incorporated into CE (Figure 6A). In a second experiment, reverse labeling of [3H]-cholesterol-CR and
[14C]-cholesterol-LDL was performed. Once again assuring
similar labeling efficiency, results showed that 24.54±1.73%
of [14C]-cholesterol derived from LDL was incorporated into
CE, whereas 8.35±1.95% of [3H]-cholesterol derived from CR
was incorporated into CE (Figure 6A). These results confirm
that LDL-derived cholesterol is preferentially esterified compared with CR-derived cholesterol.
ApoB Secretion
To determine the fate of delivered cholesterol, VLDL secretion
was measured. In this experiment, [3H]-cholesterol–labeled
LDL and [3H]-cholesterol–labeled CR were incubated with
hepatocytes for 24 hours in the presence of a lecithin:cholesterol
acyltransferase inhibitor. The cell medium was collected, and
the CE lipids were collected by thin layer chromatography and
submitted for scintillation counting. More than 3-fold LDLderived cholesterol was secreted as CE into the medium than
CR-derived cholesterol (152.3±21.1 versus 46.6±16.3 cpm/μg
cell protein; Figure 6B). This experiment was repeated using
anti-apoB antibodies to immunoprecipitate secreted apoB and
its cholesterol-associated radioactivity. Immunoprecipitation
of media from cells grown in fetal bovine serum with
[3H]-cholesterol or LPDS with [3H]-cholesterol (representing the positive and negative controls, respectively) showed
23927±3551 and 4159±420 cpm/mg cell protein, indicating
that apoB secretion increases in cholesterol-loaded compared
with cholesterol-depleted cells. By comparison, LDL-treated
cells had 27548±3319 cpm/mg cell protein, and CR had
5710±1714 cpm/mg cell protein (Figure 6C). These results
demonstrate that LDL-derived cholesterol is preferentially
esterified and secreted as part of an apoB-containing particle
(VLDL) compared with CR-derived cholesterol.
Efflux Results
Hepatocytes were dually labeled by adding separately labeled
LDL and CR for uptake, and then an efflux assay was performed. To be complete and to confirm the effect observed,
we labeled the LDL and CR in 3 different ways: (1) [3H]
cholesterol-LDL and [14C]cholesterol-CR; (2) [3H]cholesteryl
ester-LDL and [14C]cholesterol-CR; and (3) [3H]cholesteryl
ester-CR and [14C]cholesterol-LDL. In each case, the cholesterol-associated LDL label was preferentially transported to
HDL by cholesterol efflux (Figure 6D). The results demonstrate that LDL-derived cholesterol, no matter how labeled
and delivered, is preferentially transported to an efflux-accessible site for mediated cholesterol efflux.
Discussion
The conventional model states that cholesterol, which enters
the cell within LDL particles, rapidly equilibrates with the
cholesterol within the regulatory pool and, by doing so,
induces a coordinated, physiologically purposeful series of
responses, which restore cholesterol balance by reducing
its synthesis and uptake. Increased input leads to decreased
input. Alterations in output are not part of the equation. But
the conventional model, which was based on studies in fibroblasts, does not take into account that cholesterol can be
secreted from the hepatocyte either as bile acids or dissolved
in bile acids or within VLDL or HDL particles and that cholesterol esterification within the hepatocyte can be driven by
either ACAT1 or ACAT2, whereas only ACAT1 is present in
other cells.
Sniderman et al LDL-Derived Cholesterol Is Shunted Into VLDL 2487
Downloaded from http://atvb.ahajournals.org/ by guest on June 16, 2017
Figure 6. Low-density lipoprotein (LDL)–derived cholesterol is preferentially esterified, secreted in apolipoprotein B (apoB) particles, and
effluxed to high-density cholesterol (HDL). A, Hepatocytes, dually labeled with [3H]-free cholesterol LDL ([3H]-FC-LDL; black bar) and
[14C]-FC-chylomicron remnants (CR; gray bar; left) or [14C]-FC-LDL (black bar) and [3H]-FC-CR (gray bar; right), were grown in lipoproteindeficient serum (LPDS) media for 24 h, and then cellular radioactive cholesteryl ester (CE) was determined. Results are calculated as the
percent of radioactive CE of total radioactivity for each isotope and presented as the mean±SD of quadruplicates of 3 identical experiments. **P<0.01, compared with the LDL condition. B, Hepatocytes were labeled with [3H]-cholesterol–labeled LDL (black bar) and CR
(white bar) for 24 h, the cells were washed extensively, and then the cells were incubated with plain media for another 8 h. Lipids were
extracted from the media, and radioactive CE was determined. Results are expressed as the average cpm/μg cell protein±SD of quadruplicates of 3 identical experiments. ***P<0.001, compared with the LDL condition. C, Experiments were performed similar to B, except
the apoB was immunoprecipitated from the media. As controls, media from fetal bovine serum (FBS) and LPDS-treated cells were also
included. Results are calculated as the percent of radioactive CE of total radioactivity and presented as the mean±SD of quadruplicates
of 3 identical experiments. ***P<0.001, compared with the LDL condition. D, Hepatocytes were labeled with the following: (1) [3H]-FC-LDL
and [14C]FC-CR; (2) [3H]-CE-LDL and [14C]-FC-CR; (3) [3H]-CE-CR and [14C]-FC-LDL for 24 h, the cells were then washed extensively, and
then the cells were incubated with HDL (10 μg) for 4 h. Media were collected, centrifuged to remove intact cells, and radioactivity was
determined by scintillation counting. Results are expressed as the average percent radioactive cholesterol (medium radioactivity/total
radioactivity [cellular and media counts])±SD of quadruplicates of 3 identical experiments. *P<0.05, compared with the LDL condition.
Lange et al6,7,21 demonstrated that the pools of cholesterol
in the plasma membrane, the mitochondria, and the ER are
in rapid reversible equilibrium. Cholesterol delivered to the
plasma membrane moves first to the ER and mitochondria
where both immediate and delayed adaptive responses are
stimulated. Only after equilibration in these compartments
does cholesterol move to the site of ACAT activity so that
cholesterol esterification, which renders cholesterol bioinactive, can occur. Our results suggest that more of the cholesterol released from the CR lysosome is delivered to the
plasma membrane-ER-mitochondria compartment than from
the LDL lysosome. In contrast, compared with CR-containing
lysosomes, a greater portion of the cholesterol released from
LDL-containing lysosomes is delivered to ACAT2 and rapidly
esterified and incorporated into nascent apoB lipoproteins.
Our data suggest that much of the cholesterol taken up within
LDL particles is so rapidly esterified that it does not gain
access to the regulatory pool but rather is almost immediately
resecreted within apoB or nascent HDL particles. However,
this pattern seems to be characteristic of cholesterol from LDL
particles but not of cholesterol from CR, VLDL, and β-VLDL
particles or of cholesterol from non-lipoprotein particle
sources. Cholesterol from all these sources seems to equilibrate rapidly with the cholesterol regulatory pool rather than
primarily and initially interacting with ACAT. In fibroblasts,
by contrast, ACAT activity is less, cholesterol ester cannot
exit the cell, and, therefore, cholesterol derived from LDL will
eventually equilibrate with the regulatory pool and deactivate
the SREBP2 pathway.22 These differences may also be related
to differences in endosomal pathways for lipoprotein particles
because our fluorescent microscopic studies demonstrated
that LDL and CR seem to enter the cell by separate pathways.
These observations are supported by other reports of a delay
before lysosomes containing CR particles reach the perinuclear region of the cell, whereas lysosomes containing LDL
particles seemed to move rapidly to the perinuclear region.23,24
Given that the mass of CE seems to be a major determinant
of the rate of secretion of apoB particles from hepatocytes,8,9
linkage of the LDL endocytosis pathway to the apoB secretion pathway via esterification of cholesterol creates a shunt
pathway for cholesterol, which enters and leaves the hepatocyte without interacting biologically with the cholesterol
within the regulatory pools. Interestingly, entry of LDL cholesterol upregulates ACAT2 expression in this model system,
2488 Arterioscler Thromb Vasc Biol November 2013
Downloaded from http://atvb.ahajournals.org/ by guest on June 16, 2017
and future studies are required to determine how LDL-derived
cholesterol upregulates ACAT2 expression and how LDLderived cholesterol may be specifically targeted to ACAT2mediated esterification.
Involvement of PCSK9 and IDOL in the hepatic processing
of LDL-derived cholesterol is still uncertain. PCSK9, synthesized as a 72-kDa zymogen, which autocatalytically cleaves
itself into a mature 60-kDa protein, is secreted and binds to the
LDL receptors promoting their uptake and lysosomal degradation.25 PCSK9 is regulated by SREBP2, in that low intracellular
cholesterol levels stimulate PCSK9 expression and high intracellular cholesterol levels reduce PCSK9 levels (confirmed in
Figure 5). The role of PCSK9 is paradoxical, in that under low
intracellular cholesterol conditions, HMGCoA reductase, the
rate-limiting enzyme of endogenous cholesterol biosynthesis,
and LDL receptor expression are also stimulated (to increase
intracellular cholesterol levels), whereas PCSK9 downregulates the LDL receptor achieving the opposite. Western blotting showed that the LDL receptor and PCSK9 are inversely
correlated, providing evidence that SREBP2-mediated upregulation of PCSK9 is the strongest determinant of LDL receptor
expression levels.26 However, this is not the case in LDL- and
VLDL-treated hepatocytes, as these cells have high PCSK9
(presumably because of the shunting effect on LDL-derived
cholesterol away from the regulatory pool and, therefore,
increased SREBP2) and increased LDL receptor. This result
suggests there is a complexity in the regulation of the LDL
receptor by LDL and VLDL treatment that has not been previously appreciated. This can also be demonstrated by the
inverse relationships between PCSK9 and the LDL receptor
depending on the lipoprotein treatment: chylomicron, HDL,
and non-lipoprotein–derived cholesterol treatment increased
PCSK9 expression and decreased LDL receptor expression,
whereas β−VLDL and CR decreased PCSK9 expression and
increased LDL receptor expression. Thus, it seems that the
precise intracellular location of delivery of cholesterol is an
important determinant of the homeostatic response.
It also seems that all cholesterol pools are not equivalent
and that cholesterol that enters hepatocytes does not automatically enter the regulatory pool (by which we mean the
actual mass of cholesterol that regulates the SREBP2 pathway). Interestingly, the immature 72-kDa form of PCSK9 is
observed in the LDL-treated cells but not in the CR-treated
cells, suggesting a level of regulation of PCSK9 that controls
the processing and secretion of apoB lipoprotein particles.27,28
Newly synthesized intracellular PCSK9 might interact with
apoB in the lumen of the ER, preventing intracellular VLDL
degradation and enhancing VLDL secretion,29 and this may
be responsible for the lack of downregulation of the LDL
receptor, despite SREBP2-mediated stimulation of PCSK9
expression. In any case, these results imply more complicated
regulatory pathways for PCSK9 in hepatocytes than are usually understood to exist.30–32
IDOL is a liver X receptor nuclear receptor–regulated target, which mediates ubiquitination and targeted degradation
of the LDL receptor.33 IDOL is stimulated under conditions
of high intracellular cholesterol but operates independently of
SREBP2 regulation. Therefore, the observation that all cholesterol treatment resulted in increased IDOL expression is not
inconsistent with the understanding of regulation of IDOL.34,35
In this case, we do not see any difference between LDL- and
CR-treated cells36 and do not predict that IDOL plays a role in
the observed experimental results.
These results in hepatocytes contrast with those in fibroblasts, in which uptake of an LDL particle produces acute
downregulation of cholesterol and LDL receptor synthesis,
observations that constitute core evidence for the LDL receptor paradigm. In the fibroblast, however, hydrolysis of lysosomal CE is more rapid than esterification.8,37 Because ACAT
activity is relatively low in the fibroblast compared with the
hepatocyte38 and because cholesterol cannot be secreted from
the fibroblast, cholesterol released from the lysosome does
enter the regulatory pool in the fibroblast, whereas it does not
in the hepatocyte.
There are important limitations to our study. Although there
are many similarities with regard to cholesterol homeostasis
and apoB metabolism, both in vivo and in vitro, we cannot
assume that these results in hamster hepatocytes apply without reservation to human hepatocytes. Furthermore, our study
of cholesterol homeostasis even in this system is incomplete
because we did not estimate the impact on bile acid synthesis
and secretion of cholesterol within bile. It also seems that published results in macrophages may be contrary to our results
in hepatocytes. In the macrophage, the cholesterol within CR
is rapidly reesterified notwithstanding that there seems to be
a similar pause of CR particles just after uptake.23,24 This difference may relate to differences in mechanisms of uptake,
with syndecan-1 playing a more prominent role in uptake in
the hepatocyte membrane compared with the macrophage.39
Alternatively, the intracellular distribution of ACAT may differ
in macrophages versus hepatocytes.40,41 The high expression
of ACAT2 in hepatocytes is the most likely cause of the physiological differences in handling LDL-derived cholesterol.
In summary, we demonstrate that LDL- and VLDL-derived
cholesterol is handled differently than cholesterol derived
from other lipoproteins by the hepatocyte. LDL-derived cholesterol upregulates ACAT2 expression, leading to increased
cholesterol esterification, enhanced secretion of cholesterol
within VLDL particles, and increased efflux of cholesterol
to HDL particles. Consequently, entry of cholesterol into the
regulatory pool is decreased, and endogenous cholesterol synthesis is not downregulated. These results are consistent with
in vivo apoB kinetic data obtained in humans, demonstrating
that the LDL receptor and endogenous cholesterol synthesis
remain active in the face of large amounts of LDL cholesterol
delivered to the liver and that increased apoB secretion is associated with increased delivery of cholesterol to the liver.8,9,20
These findings do not invalidate the importance of the LDL
pathway, but they do point to a more multifaceted and complex model of cholesterol homeostasis in the liver than the
fibroblast. Furthermore, they indicate that intracellular cholesterol homeostasis is dependent, at least in part, on site-specific
metabolic linkages that create micrometabolic domains within
the cell. Thus, cholesterol released from a remnant particle at
one site within the cell will tend to enter the regulatory compartment more easily than cholesterol released from an LDL
particle at another site. In contrast, if cholesterol is released
in close proximity to active ACAT2, the result will be a shunt
Sniderman et al LDL-Derived Cholesterol Is Shunted Into VLDL 2489
pathway of cholesterol through the cell without that cholesterol ever entering the regulatory compartment.
Acknowledgments
We thank Cyril Martin for his expert technical assistance and Audric
Moses of the Women and Children’s Health Research Institute and
the University of Alberta’s Faculty of Medicine and Dentistry for
cholesterol analysis by high-performance liquid chromatography.
Sources of Funding
We acknowledge funding from Canadian Institutes of Health Research
(grant MOP-89972 to R.S. Kiss), the Heart and Stroke Foundation of
Quebec (to R.S. Kiss), the Heart and Stroke Foundation of Ontario (to
K. Adeli), and the Mike Rosenbloom Fund for Cardiovascular Research
(to A.D. Sniderman) for support.
Disclosures
None.
Downloaded from http://atvb.ahajournals.org/ by guest on June 16, 2017
References
1. Goldstein JL, Kita T, Brown MS. Defective lipoprotein receptors and atherosclerosis. Lessons from an animal counterpart of familial hypercholesterolemia. N Engl J Med. 1983;309:288–296.
2. Brown MS, Goldstein JL. A receptor-mediated pathway for cholesterol
homeostasis. Science. 1986;232:34–47.
3. Radhakrishnan A, Goldstein JL, McDonald JG, Brown MS. Switch-like
control of SREBP-2 transport triggered by small changes in ER cholesterol: a delicate balance. Cell Metab. 2008;8:512–521.
4. Brown MS, Goldstein JL. Cholesterol feedback: from Schoenheimer’s
bottle to Scap’s MELADL. J Lipid Res. 2009;50(Suppl):S15–S27.
5. Goldstein JL, Brown MS. The LDL receptor. Arterioscler Thromb Vasc
Biol. 2009;29:431–438.
6.Lange Y, Ory DS, Ye J, Lanier MH, Hsu FF, Steck TL. Effectors
of rapid homeostatic responses of endoplasmic reticulum cholesterol and 3-hydroxy-3-methylglutaryl-CoA reductase. J Biol Chem.
2008;283:1445–1455.
7. Lange Y, Steck TL, Ye J, Lanier MH, Molugu V, Ory D. Regulation of
fibroblast mitochondrial 27-hydroxycholesterol production by active
plasma membrane cholesterol. J Lipid Res. 2009;50:1881–1888.
8. Sniderman AD, Zhang Z, Genest J, Cianflone K. Effects on apoB-100
secretion and bile acid synthesis by redirecting cholesterol efflux from
HepG2 cells. J Lipid Res. 2003;44:527–532.
9. Zhang Z, Cianflone K, Sniderman AD. Role of cholesterol ester mass in
regulation of secretion of ApoB100 lipoprotein particles by hamster hepatocytes and effects of statins on that relationship. Arterioscler Thromb
Vasc Biol. 1999;19:743–752.
10.Pangburn SH, Newton RS, Chang CM, Weinstein DB, Steinberg D.
Receptor-mediated catabolism of homologous low density lipoproteins in
cultured pig hepatocytes. J Biol Chem. 1981;256:3340–3347.
11. Havekes LM, Verboom H, de Wit E, Yap SH, Princen HM. Regulation
of low density lipoprotein receptor activity in primary cultures of human
hepatocytes by serum lipoproteins. Hepatology. 1986;6:1356–1360.
12.Cohen LH, Princen HM, Kwekkeboom J, Havekes LM, Kempen HJ.
Regulation of cholesterol metabolism in the liver in vivo and in vitro.
Biochem Soc Trans. 1987;15:339–340.
13. Havekes LM, de Wit EC, Princen HM. Cellular free cholesterol in Hep
G2 cells is only partially available for down-regulation of low-densitylipoprotein receptor activity. Biochem J. 1987;247:739–746.
14.Wu GY, Wu CH, Rifici VA, Stockert RJ. Activity and regulation of
low density lipoprotein receptors in a human hepatoblastoma cell line.
Hepatology. 1984;4:1190–1194.
15.Harders-Spengel K, Wood CB, Thompson GR, Myant NB, Soutar
AK. Difference in saturable binding of low density lipoprotein to liver
membranes from normocholesterolemic subjects and patients with heterozygous familial hypercholesterolemia. Proc Natl Acad Sci USA.
1982;79:6355–6359.
16.Dueland S, Trawick JD, Nenseter MS, MacPhee AA, Davis RA.
Expression of 7 alpha-hydroxylase in non-hepatic cells results in liver
phenotypic resistance of the low density lipoprotein receptor to cholesterol
repression. J Biol Chem. 1992;267:22695–22698.
17. Ye ZJ, Go GW, Singh R, Liu W, Keramati AR, Mani A. LRP6 protein
regulates low density lipoprotein (LDL) receptor-mediated LDL uptake. J
Biol Chem. 2012;287:1335–1344.
18. Truong TQ, Auger A, Denizeau F, Brissette L. Analysis of low-density
lipoprotein catabolism by primary cultures of hepatic cells from normal
and low-density lipoprotein receptor knockout mice. Biochim Biophys
Acta. 2000;1484:307–315.
19.Strong A, Ding Q, Edmondson AC, Millar JS, Sachs KV, Li X, et al.
Hepatic sortilin regulates both apolipoprotein B secretion and LDL catabolism. J Clin Invest. 2012;122:2807–2816.
20. Sniderman AD, Kiss RS. The strengths and limitations of the apoB/apoAI ratio to predict the risk of vascular disease: a Hegelian analysis. Curr
Atheroscler Rep. 2007;9:261–265.
21. Lange Y, Steck TL. Cholesterol-rich intracellular membranes: a precursor
to the plasma membrane. J Biol Chem. 1985;260:15592–15597.
22.Brown MS, Dana SE, Goldstein JL. Receptor-dependent hydrolysis of
cholesteryl esters contained in plasma low density lipoprotein. Proc Natl
Acad Sci USA. 1975;72:2925–2929.
23. Myers JN, Tabas I, Jones NL, Maxfield FR. Beta-very low density lipoprotein is sequestered in surface-connected tubules in mouse peritoneal
macrophages. J Cell Biol. 1993;123(6 Pt 1):1389–1402.
24. Tabas I, Myers JN, Innerarity TL, Xu XX, Arnold K, Boyles J, et al. The
influence of particle size and multiple apoprotein E-receptor interactions
on the endocytic targeting of beta-VLDL in mouse peritoneal macrophages. J Cell Biol. 1991;115:1547–1560.
25.Lambert G, Sjouke B, Choque B, Kastelein JJ, Hovingh GK. The
PCSK9 decade: thematic review series: new lipid and lipoprotein
targets for the treatment of cardiometabolic diseases. J Lipid Res.
2012;53:2515–2524.
26. Tavori H, Fan D, Blakemore JL, Yancey PG, Ding L, Linton MF, et al.
Serum proprotein convertase subtilisin/kexin type 9 and cell surface
low-density lipoprotein receptor: evidence for a reciprocal regulation.
Circulation. 2013;127:2403–2413.
27. Xu W, Liu L, Hornby D. c-IAP1 binds and processes PCSK9 protein: linking the c-IAP1 in a TNF-α pathway to PCSK9-mediated LDLR degradation pathway. Molecules. 2012;17:12086–12101.
28.Benjannet S, Saavedra YG, Hamelin J, Asselin MC, Essalmani R,
Pasquato A, et al. Effects of the prosegment and pH on the activity of PCSK9: evidence for additional processing events. J Biol Chem.
2010;285:40965–40978.
29. Sun H, Samarghandi A, Zhang N, Yao Z, Xiong M, Teng BB. Proprotein
convertase subtilisin/kexin type 9 interacts with apolipoprotein B and prevents its intracellular degradation, irrespective of the low-density lipoprotein receptor. Arterioscler Thromb Vasc Biol. 2012;32:1585–1595.
30. Wu M, Dong B, Cao A, Li H, Liu J. Delineation of molecular pathways
that regulate hepatic PCSK9 and LDL receptor expression during fasting
in normolipidemic hamsters. Atherosclerosis. 2012;224:401–410.
31.Li H, Liu J. The novel function of HINFP as a co-activator in ste
rol-regulated transcription of PCSK9 in HepG2 cells. Biochem J.
2012;443:757–768.
32. Cao A, Wu M, Li H, Liu J. Janus kinase activation by cytokine oncostatin M
decreases PCSK9 expression in liver cells. J Lipid Res. 2011;52:518–530.
33.Sorrentino V, Zelcer N. Post-transcriptional regulation of lipoprotein
receptors by the E3-ubiquitin ligase inducible degrader of the low-density
lipoprotein receptor. Curr Opin Lipidol. 2012;23:213–219.
34.Zhang L, Reue K, Fong LG, Young SG, Tontonoz P. Feedback regulation of cholesterol uptake by the LXR−IDOL−LDLR axis. Arterioscler
Thromb Vasc Biol. 2012;32:2541–2546.
35.Ishibashi M, Masson D, Westerterp M, Wang N, Sayers S, Li R, et al.
Reduced VLDL clearance in Apoe(−/−)Npc1(−/−) mice is associated with
increased Pcsk9 and Idol expression and decreased hepatic LDL-receptor
levels. J Lipid Res. 2010;51:2655–2663.
36. Dong B, Wu M, Cao A, Li H, Liu J. Suppression of Idol expression is an
additional mechanism underlying statin-induced up-regulation of hepatic
LDL receptor expression. Int J Mol Med. 2011;27:103–110.
37. Brown MS, Goldstein JL. Analysis of a mutant strain of human fibroblasts
with a defect in the internalization of receptor-bound low density lipoprotein. Cell. 1976;9(4 Pt 2):663–674.
38. Lange Y, Steck TL. Quantitation of the pool of cholesterol associated with
acyl-CoA:cholesterol acyltransferase in human fibroblasts. J Biol Chem.
1997;272:13103–13108.
39. Stanford KI, Bishop JR, Foley EM, Gonzales JC, Niesman IR, Witztum
JL, et al. Syndecan-1 is the primary heparan sulfate proteoglycan mediating hepatic clearance of triglyceride-rich lipoproteins in mice. J Clin
Invest. 2009;119:3236–3245.
2490 Arterioscler Thromb Vasc Biol November 2013
40. Kiss RS, You Z, Genest J Jr, Behm DJ, Giaid A. Urotensin II differentially regulates macrophage and hepatic cholesterol homeostasis. Peptides.
2011;32:956–963.
41.Khelef N, Buton X, Beatini N, Wang H, Meiner V, Chang TY, et al.
Immunolocalization of acyl-coenzyme A:cholesterol O-acyltransferase in
macrophages. J Biol Chem. 1998;273:11218–11224.
Significance
The low-density lipoprotein receptor paradigm demonstrates the exquisite homeostasis of intracellular cholesterol regulation. The liver is
central to total body cholesterol regulation and, therefore, represents the most important cell type to understand cholesterol homeostasis.
Here, we demonstrate that this paradigm is incomplete when describing cholesterol homeostasis in hepatocytes. We also demonstrate that
the cholesterol derived from low-density lipoprotein does not enter the normal intracellular regulatory pool but rather is shunted out as part of
a very–low-density lipoprotein particle. These findings have enormous clinical relevance not only on the accumulation of cholesterol in hepatocytes as part of fatty liver disease but also on the plasma concentration of low-density lipoprotein as a contributing factor to atherosclerosis.
Downloaded from http://atvb.ahajournals.org/ by guest on June 16, 2017
Downloaded from http://atvb.ahajournals.org/ by guest on June 16, 2017
Hepatic Cholesterol Homeostasis: Is the Low-Density Lipoprotein Pathway a Regulatory
or a Shunt Pathway?
Allan D. Sniderman, Yanqin Qi, Cheng-I J. Ma, Rui Hao Leo Wang, Mark Naples, Chris Baker,
Jing Zhang, Khosrow Adeli and Robert S. Kiss
Arterioscler Thromb Vasc Biol. 2013;33:2481-2490; originally published online August 29,
2013;
doi: 10.1161/ATVBAHA.113.301517
Arteriosclerosis, Thrombosis, and Vascular Biology is published by the American Heart Association, 7272
Greenville Avenue, Dallas, TX 75231
Copyright © 2013 American Heart Association, Inc. All rights reserved.
Print ISSN: 1079-5642. Online ISSN: 1524-4636
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://atvb.ahajournals.org/content/33/11/2481
Data Supplement (unedited) at:
http://atvb.ahajournals.org/content/suppl/2013/08/29/ATVBAHA.113.301517.DC1
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Arteriosclerosis, Thrombosis, and Vascular Biology can be obtained via RightsLink, a service of the
Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for
which permission is being requested is located, click Request Permissions in the middle column of the Web
page under Services. Further information about this process is available in the Permissions and Rights
Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Arteriosclerosis, Thrombosis, and Vascular Biology is online
at:
http://atvb.ahajournals.org//subscriptions/
B
125I-bound
lipoprotein
(cpm/μg)
A
150
100
Binding
100
50
0
LDL
Lipoprotein 3H-cholesteryloleate uptake
(cpm/ng)
Supplemental Material
CR
**
Uptake
80
**
60
40
20
0
0
2
4
6
8
Time (hours)
Supp. Figure I: Binding and internalization studies. A) 125I-labeled LDL (100 ng; black bar)
and CR (100 ng; white bar) were added to hepatocytes at 4°C to measure cell surface receptor
binding. Results are expressed as cpm/μg cell protein ±SD. B) 3H-cholesteryl oleate (2μCi),
incorporated into 50 μg of LDL (black bar) and CR (white bar), were added for increasing
times to hepatocytes at 37°C to measure internalization rates. Results are expressed as the
average cpm/ng cell protein ±SD. **p<0.001 compared to the LDL condition.
1
Radioactive CE incorporated into
cells (cpm/μg total cell protein)
320
240
160
80
0
0
20
40
60
80
100
120
Lipoprotein (μg)
Supp. Figure IIA: Uptake and loading studies. 3H-cholesteryl oleate (2 μCi) was incorporated
into the different lipoproteins: VLDL (closed circles); Chylomicrons (open circles); LDL
(closed inverted triangles); CR (open inverted triangles); βVLDL (closed squares); HDL (open
squares); and non-lipoprotein derived cholesterol (closed diamond). Lipoproteins were added
to hepatocytes for 24 hours at increasing concentrations of lipoprotein with the same amount
of radioactivity. After 24 hours, lipids were extracted and cellular radioactivity was
determined by scintillation counting. Results are expressed as cpm/μg cell protein as a mean
of quadriplicates±SD. We chose the concentration of lipoproteins that would be utilizable for
all loading conditions and was at a point of non-saturation (at 80 cpm/μg).
2
*
*
*
CR
25
LD
L
*
LP
DS
VL
DL
Ch
ylo
Total cholesterol mass
(μg/mg total cell protein)
30
*
*
*
20
15
10
5
βV
LD
L
ch HD
ole
L
ste
ro
l
0
Supp. Figure IIB: Total cellular cholesterol mass increases with lipoprotein loading.
Hepatocytes were cholesterol loaded with the different lipoproteins (LDL, human
chylomicrons (Chylo), CR, β-VLDL, VLDL, HDL and non-lipoprotein associated free
cholesterol), and then the total cholesterol mass was measured using the enzymatic kit assay
method. Results are presented as the mean ±SD of quadruplicate measurements. Cholesterol
loading increased total cholesterol mass, free cholesterol (grey bar) and cholesteryl ester
(black bar), and hepatocytes were almost identically cholesterol loaded with the different
lipoproteins. *p<0.05, compared to the LPDS condition.
3
Total cholesterol mass
(μg/mg total cell protein)
30
*
25
*
*
*
*
*
*
20
15
10
5
βV
LD
L
ch HD
ole L
ste
ro
l
LD
L
CR
LP
DS
VL
DL
Ch
ylo
0
Supp. Figure IIC: Total cellular cholesterol mass increases with lipoprotein loading.
Hepatocytes were cholesterol loaded with the different lipoproteins (LDL, human
chylomicrons (Chylo), CR, β-VLDL, VLDL, HDL and non-lipoprotein associated free
cholesterol), and then the total cholesterol mass was measured using the GC method. Results
are presented as the mean ±SD of quadruplicate measurements. Cholesterol loading increased
total cholesterol mass, free cholesterol (grey bar) and cholesteryl ester (black bar), and
hepatocytes were almost identically cholesterol loaded with the different lipoproteins.
*p<0.05, compared to the LPDS condition.
4
1.25
HMGCoA red
mRNA fold change
1.00
0.75
*
*
0.50
*
**
**
0.25
0.00
LDL receptor
1.00
*
*
LD
L
CR
0.75
DS
VL
DL
Ch
yl o
mRNA fold change
1.25
0.50
*
*
0.25
βV
LD
L
ch H D
ole L
st e
ro
l
LP
0.00
Supp. Figure III: Hepatocytes were loaded with lipoprotein derived and non-lipoprotein
derived cholesterol for 24 hours. mRNA was extracted using the RNeasy mini RNA
extraction kit (Qiagen). Analysis of mRNA expression of HMGCoA reductase, LDL receptor
and β-tubulin by real-time quantitative PCR from 200 ng total RNA was carried out as
described in Methods. Values shown are means ± SD of triplicate experiments. The expression
of each gene was normalized to β-tubulin expression, and mRNA fold changes relative to
LPDS control were determined. *p < 0.05, **p < 0.01 by Student's t-test.
5
incorporated into CE
(% of total cellular radioactivity)
3H-cholesterol
†
3
2
1
*
**
0
-
+
-
LDL
+
ACATi
CR
Supp. Figure IV: ACAT inhibitor drastically reduces CE formation in LDL-treated
hepatocytes. To determine the relative contribution of ACAT activity to the accumulation of
CE in LDL and CR-treated hepatocytes, the ACAT inhibitor (ACATi) 58035 was added at the
same time as lipoprotein. Lipoproteins were labeled with [3H]-cholesterol and treated as
before and the radioactive CE was presented as the mean ±SD of quadruplicates of three
identical experiments. *p<0.05, compared to the CR (no ACATi) condition. **p<0.001,
compared to the LDL (no ACATi) condition. † p<0.001, comparing the LDL (no ACATi)
condition to the CR (no ACATi) condition.
6
7
Supp. Figure V: Quantification of western blots. Cell lysates of hepatocytes treated with
different lipoproteins from three independent experiments were prepared and western blots
were performed for ACAT1, ACAT2, LDL receptor, PCSK9 and IDOL. GAPDH was used as
a loading control. Bands were scanned with the AlphaImager scanning program, normalized
to the LPDS control and average ±SD was computed and graphed for each protein. *p<0.05
compared to the LPDS condition.**p<0.01 compared to the LPDS condition.
8
Materials and Methods
Materials—Plastic cell culture dishes were obtained from Corning (Primaria for primary
cell cultures) and culture media from Invitrogen. Fetal bovine serum, lipoprotein
deficient serum, and bovine serum albumin were purchased from Sigma. Na125I was from
Amersham Biosciences, and [3H]cholesterol, [14C]cholesterol, [3H]mevalonate,
[14C]oleate, and [3H]cholesteryl oleate were from PerkinElmer Life Sciences.
[3H]cholesteryl oleate and 3, 3-dioctyloxacarbocyanine perchlorate (DiO) and 1,1’dioctadecyl-3,3,3’,3’-tetramethylindodicarbocyanine perchlorate (DiD) labeling dyes
(Invitrogen) were incorporated into lipoproteins by microinjection according to Vassiliou
et al. 1, 2
Isolation of Lipoproteins – Chylomicrons, VLDL, LDL, and HDL were isolated from a
human peripheral blood sample by sequential density gradient ultracentrifugation.
Chylomicrons were treated with hepatic lipase to produce lipolyzed chylomicron
remnants (CR; 3) and then reisolated by density gradient ultracentrifugation. Mouse
chylomicrons were isolated from the plasma of apoE deficient mice, then re-enriched by
incubation with bacterially expressed apoE (β-VLDL)) as previously described 4, and
then reisolated by density gradient ultracentrifugation. The concentration of lipoproteins
was calculated using the Markwell Lowry protein assay.
Isolation and assay of hepatocytes - All experiments performed were in accordance with
protocols approved by the McGill University Health Centre Animal Care Committee.
Male Syrian Golden hamsters were housed (3 animals/cage) under controlled temperature
(72°F) and lighting (12 h light/dark cycle). Animals had free access to autoclaved water
and food. Primary hepatocytes from Syrian Golden hamsters were isolated as described
previously 5 and seeded onto Primaria plates in DMEM containing 10% FBS or 10%
lipoprotein deficient serum (LPDS). After allowing 4 h for the cells to adhere, cells were
incubated with 10% LPDS for 16 h prior to the uptake and binding assays. For the
binding assay, lipoproteins (LDL and CR) were radiolabeled with 125I using iodobeads
(Pierce; according to manufacturer’s instructions). The binding assay was performed at
4°C for 1 h before extensive washing and determination of bound radioactivity
(normalized to non-specific binding and total cell protein). [3H]cholesteryl oleate
radiolabeled lipoproteins (25 μg) were added to the cells for 1 h, 4 h, or 8h and after
extensive washing, the amount of radiolabel taken up by the cells was normalized for
non-specific uptake and total cell protein.
Due to the differences in the size and composition of the lipoprotein particles, we
compared the uptake of cholesterol from each with increasing concentrations of total
lipoprotein cholesterol. These experiments were first performed with 3H-cholesteryloleate labeled lipoproteins (as above except that a range of concentrations was used at a
24 h time point). Next, the same experiment was performed except that a 3H-cholesterol
label was used. Importantly, 3H-cholesteryl oleate and 3H-cholesterol labeled the cells to
an equivalent degree over the concentration curve, demonstrating that both labels are
reliable indicators of cholesterol uptake. 3H-cholesteryl oleate labeling of lipoproteins
results in a 95-100% incorporation into internal compartments (non-methyl βcyclodextrin accessible) while
3
H-cholesterol labeling of lipoproteins 65-70%
incorporation into internal compartments (non-methyl β-cyclodextrin accessible) with the
remaining radioactivity in the plasma membrane (as previously shown by Zheng et al. 6).
Experiments with both labels were included in order to find a concentration of each
lipoprotein that gave a similar degree of radioactive labeling. The concentration of each
lipoprotein (and non-lipoprotein derived cholesterol) served as a guide for the mass
assays (LDL 50 μg/mL; CR 12 μg/mL; β-VLDL 18 μg/mL; Chylomicrons 8 μg/mL;
VLDL 25 μg/mL; HDL 100 μg/mL; non-lipoprotein-derived cholesterol 100 μg/mL). We
then performed the same experiment with unlabeled lipoproteins at these specific
concentrations for 24 h and performed mass measurements. Total cellular cholesterol
mass was measured by three different methods: 1) total cholesterol and free cholesterol
kits (Roche Diagnostics; according to manufacturer’s protocol); 2) high performance
liquid chromatography (HPLC) separation and quantitation of free and cholesteryl ester;
3) gas chromatographic separation and quantitation of free and cholesteryl ester.
Chromatography of lipids – Gas chromatography experiments were performed as
previously described 7. HPLC was performed as follows. Cellular lipids were extracted
by a one hour incubation with hexane. The effluent was evaporated under a stream of
nitrogen and re-dissolved in 100 μL of chloroform: isooctane (1:1). The solution was
transferred to Agilent HPLC vials and stored at -200C until the chromatography. Samples
were analyzed using an Agilent 1100 series equipped with a quaternary pump and an
Alltech ELSD 2000 evaporative light-scattering detector. The column used was an Onyx
monolithic Si (Phenomenex) and the solvent system was based on the method of Graeve
and Janssen 8. Solvent A consisted of isooctane:ethylacetate (99.8:0.2); solvent B was
acetone:ethylacetate (2:1) with 0.02% acetic acid; solvent C was isopropanol:water
(85:15) with acetic acid and ethanolamine each at 0.05%. Gas flow was 3.0 L/min and
drift tube temperature was set at 60 degrees C. Peaks were analyzed using Agilent
Chemstation software and quantified using calibration curves prepared with commercial
lipid standards (Sigma-Aldrich, Avanti Polar Lipids). All three methods confirmed that
the incubation of hepatocytes with lipoprotein and non-lipoprotein-derived cholesterol
resulted in equivalent cholesterol loading. Therefore, for all subsequent experiments, we
achieved the equivalent loading and labeling of hepatocytes for each lipoprotein
(including non-lipoprotein-derived cholesterol).
For the ACAT assay, hepatocytes were incubated with [3H]-cholesterol (5 μCi)
incorporated into the predetermined concentrations of lipoproteins, FBS or LPDS for 24
h to uniformly label the cells. Then the radioactive lipids were extracted with hexane.
Lipids were separated by thin layer chromatography using hexane/diethyl ether/acetic
acid (105:45:1.5, v/v/v) as a solvent system. Cholesterol and cholesteryl ester (CE) spots
were scraped from the plate and radioactivity determined by liquid scintillation. Results
are expressed as percent CE of the total radioactive cholesterol (free and esterified) ±SD.
Alternatively, cells incubated with unlabeled lipoproteins for 24 h were incubated with
[14C]oleate (1 μCi) for the last 12 h and then the radioactive lipids quantified as above.
Results are expressed as cpm of radioactive CE/ng total cell protein ±SD. For de novo
synthesis, cells were incubated with unlabeled lipoproteins for 24 h in the presence of
[3H] mevalonate (5μCi) and then lipids were quantified as above. Results are expressed
as cpm of radioactive cholesterol or cholesteryl ester/ng total cell protein ±SD. For de
novo synthesized CE, the amount of CE was taken as a percent of total radioactive
cholesterol ±SD, to account for the significantly increased amount of newly synthesized
cholesterol in LDL-treated cells. Typical specific activities for [3H] cholesterol labeling
are (in cpm/ng lipoprotein protein): chylomicrons 15803 cpm/ng; CR 10280 cpm/ng; βVLDL 6905 cpm/ng; VLDL 5086 cpm/ng; HDL 1122 cpm/ng; LDL 2543 cpm/ng.
mRNA Quantitation - Total RNA was prepared from primary hamster hepatocytes cells
using the RNeasy mini RNA extraction kit (Qiagen), according to the manufacturer's
instructions. Total RNA (200 ng) was reverse-transcribed using the QuantiTect Reverse
Transcription kit (Qiagen). Real-time quantitative PCR was carried out using the
Quantitect SYBR Green PCR kit and QuantiTect Primer assays (Qiagen) using the
primers: β-actin 5′-GCACCAAGGTGTGATGGTG-3′ and antisense primer 5′CGGTTGGCCTTCAGGGTTC-3′; HMGCoA reductase 5′CGAGGAAAGACTGTGGTTTG-3′ and antisense primer 5′CACGTTCCTTGAAGATCTTG-3′; LDL receptor 5′-GTG TGA AGA TAT TGA CGA
GTG-3′, and antisense primer 5′-AGT AGA TTC TAT TGT TGG TCA-3′. All reactions
were performed on an ABI PRISM 7300 Sequence Detection System (Applied
Biosystems). Amplifications were carried out in a 96-well plate with 50 μl reaction
volumes and 40 amplification cycles (94°C, 15s; 55°C, 30s; 72°C, 34s). Experiments
were carried out in triplicate, and the mRNA expression was taken as the mean of three
separate experiments. The expression of each gene was normalized to β-actin expression.
Fold changes relative to controls were determined using the ΔΔCt method.
Double
Labeling
-
We
labeled
hepatocytes
with
[3H]cholesterol-LDL
and
[14C]cholesterol-CR. To assure completeness, we converted and normalized cpm values
to dpm values based on the efficiency of detection in a scintillation counter (according to
Beckman protocols). Total radioactivity incorporated into the same hepatocyte wells of
[3H]cholesterol-LDL and [14C]cholesterol-CR were 277860±62550 and 260540±39110
dpm/mg cell protein, respectively. Therefore, essentially the same radioactivity of each
isotope was incorporated. The converse labeling of [3H]cholesterol-CR and
[14C]cholesterol-LDL was also performed.
Total radioactivity incorporated was
253690±41820 and 267390±43980 dpm/mg cell protein, respectively.
Results are
expressed as percent radioactive CE of the total radioactivity ±SD, for each isotope.
Fluorescence Microscopy— Images were collected on a Zeiss LSM-510 Meta laser
scanning microscope with either 63x or 40x oil immersion lens. Cells were placed on
coverslip culture dishes (MatTek Corp., Ashland, MA) at ∼70% confluence. For live cell
imaging, cells were maintained at 37 °C in a heated chamber and incubated with Hepesbuffered complete media, pH 7.4.
ACAT1, ACAT2, LDL receptor, PCSK9, IDOL and GAPDH Expression – Western
blotting of hepatocyte samples treated with different lipoproteins were probed with
antibodies to ACAT1, ACAT2 and LDL receptor (Santa Cruz), PCSK9 (kind gift from
Dr. Jingwen Liu, VA Palo Alto Health Care System), IDOL and GAPDH (Santa Cruz).
Cells treated with FBS or LPDS alone were used as positive and negative controls,
respectively.
ApoB Secretion - We labeled hepatocytes with [3H]cholesterol-LDL (50 μg with 5μCi) or
[3H]cholesterol-CR (12 μg with 5μCi) for 24 h in the presence of an LCAT inhibitor,
washed cells extensively, then added media with no serum. The media was collected after
8 h, concentrated by a spin column, lipids were extracted and CE quantified by TLC and
liquid scintillation. To assure specificity, alternatively, we immunoprecipitated apoB
(rabbit anti apoB (Sigma)) from the media samples and then performed the lipid isolation
and quantification (validated and confirmed according to
9-11
). Immunoprecipitation of
media from cells grown in FBS with [3H]-cholesterol or LPDS with [3H]-cholesterol
(representing the positive and negative controls, respectively) showed 23927±3551 and
4159±420 cpm/mg cell protein.
HDL-mediated cholesterol efflux - We performed an efflux assay according to Kiss et al.
12-14
Briefly, we labeled hepatocytes with: a) [3H]cholesterol-LDL and [14C]cholesterol-
CR; b) [3H]cholesteryl ester-LDL and [14C]cholesterol-CR; c) [3H]cholesteryl ester-CR
and [14C]cholesterol-LDL for 24 h. We washed the cells extensively and then we added
HDL for 4 h and measured the amount of radioactivity that associated with the HDL.
References
(1) Vassiliou G, Benoist F, Lau P, Kavaslar GN, McPherson R. The low density
lipoprotein receptor-related protein contributes to selective uptake of high density
lipoprotein cholesteryl esters by SW872 liposarcoma cells and primary human
adipocytes. J Biol Chem 2001;276:48823-48830.
(2) Vassiliou G, McPherson R. A novel efflux-recapture process underlies the
mechanism of high-density lipoprotein cholesteryl ester-selective uptake mediated
by the low-density lipoprotein receptor-related protein. Arterioscler Thromb Vasc
Biol 2004;24:1669-1675.
(3) Crawford SE, Borensztajn J. Plasma clearance and liver uptake of chylomicron
remnants generated by hepatic lipase lipolysis: evidence for a lactoferrin-sensitive
and apolipoprotein E-independent pathway. J Lipid Res 1999;40:797-805.
(4) Bousette N, D'Orleans-Juste P, Kiss RS, You Z, Genest J, Al-Ramli W, Qureshi
ST, Gramolini A, Behm D, Ohlstein EH, Harrison SM, Douglas SA, Giaid A.
Urotensin II receptor knockout mice on an ApoE knockout background fed a
high-fat diet exhibit an enhanced hyperlipidemic and atherosclerotic phenotype.
Circ Res 2009;105:686-695.
(5) Taghibiglou C, Rudy D, Van Iderstine SC, Aiton A, Cavallo D, Cheung R, Adeli
K. Intracellular mechanisms regulating apoB-containing lipoprotein assembly and
secretion in primary hamster hepatocytes. J Lipid Res 2000;41:499-513.
(6) Zheng H, Kiss RS, Franklin V, Wang MD, Haidar B, Marcel YL. ApoA-I
lipidation in primary mouse hepatocytes. Separate controls for phospholipid and
cholesterol transfers. J Biol Chem 2005;280:21612-21621.
(7) Sahoo D, Trischuk TC, Chan T, Drover VA, Ho S, Chimini G, Agellon LB,
Agnihotri R, Francis GA, Lehner R. ABCA1-dependent lipid efflux to
apolipoprotein A-I mediates HDL particle formation and decreases VLDL
secretion from murine hepatocytes. J Lipid Res 2004;45:1122-1131.
(8) Graeve M, Janssen D. Improved separation and quantification of neutral and polar
lipid classes by HPLC-ELSD using a monolithic silica phase: application to
exceptional marine lipids. J Chromatogr B Analyt Technol Biomed Life Sci
2009;877:1815-1819.
(9) Sun H, Samarghandi A, Zhang N, Yao Z, Xiong M, Teng BB. Proprotein
convertase subtilisin/kexin type 9 interacts with apolipoprotein B and prevents its
intracellular degradation, irrespective of the low-density lipoprotein receptor.
Arterioscler Thromb Vasc Biol 2012;32:1585-1595.
(10) Bamji-Mirza M, Sundaram M, Zhong S, Yao EF, Parks RJ, Yao Z. Secretion of
triacylglycerol-poor VLDL particles from McA-RH7777 cells expressing human
hepatic lipase. J Lipid Res 2011;52:540-548.
(11) Sundaram M, Zhong S, Bou KM, Links PH, Zhao Y, Iqbal J, Hussain MM, Parks
RJ, Wang Y, Yao Z. Expression of apolipoprotein C-III in McA-RH7777 cells
enhances VLDL assembly and secretion under lipid-rich conditions. J Lipid Res
2010;51:150-161.
(12) Kiss RS, McManus DC, Franklin V, Tan WL, McKenzie A, Chimini G, Marcel
YL. The lipidation by hepatocytes of human apolipoprotein A-I occurs by both
ABCA1-dependent and -independent pathways. J Biol Chem 2003;278:1011910127.
(13) Kiss RS, Maric J, Marcel YL. Lipid efflux in human and mouse macrophagic
cells: evidence for differential regulation of phospholipid and cholesterol efflux. J
Lipid Res 2005;46:1877-1887.
(14) Kiss RS, You Z, Genest J, Jr., Behm DJ, Giaid A. Urotensin II differentially
regulates macrophage and hepatic cholesterol homeostasis. Peptides 2011;32:956963.