critical factors involved in intestinal chylomicron … · 1.4. factors affecting postprandial...
TRANSCRIPT
CRITICAL FACTORS INVOLVED IN
INTESTINAL CHYLOMICRON ASSEMBLY
by
Jennifer Paula Webb
A thesis submitted in conformity with the requirements
for the degree of Master of Science
Graduate Department of Biochemistry
University of Toronto
©Copyright by Jennifer Paula Webb 2010
ii
CRITICAL FACTORS INVOLVED IN CHYLOMICRON ASSEMBLY
Master of Science 2010
Jennifer Paula Webb
Graduate Department of Biochemistry
University of Toronto
ABSTRACT
Assembly of intestinal chylomicron particles (lipid-protein complexes) is the fundamental
mechanism by which we absorb dietary fat. Two intestinal lipid transporters, Cluster of
Differentiation 36 (CD36) and fatty acid-binding protein 1 (FABP1), have been shown to play a
role in lipid absorption, however, it remains unclear how knockdown of these proteins leads to
aberrant intestinal chylomicron secretion. In an enterocyte-like cell culture model, Caco-2 cells,
we hypothesized that knockdown of CD36 or FABP1 using short-hairpin RNA interference
techniques would impair triacylglycerol (TG) and apolipoprotein B (apoB) secretion.
Surprisingly, knockdown of these lipid transporters lead to an increase in TG and apoB
secretion that was associated with an increase in fatty acid synthase and fatty acid transport
protein 4 (FATP4) protein levels. De novo fatty acid synthesis was slightly increased in CD36-,
but not FABP1-knockdown Caco-2 cells. This study highlights the importance of fatty acid
targeting in regulating chylomicron production.
©Copyright by Jennifer Paula Webb 2010
iii
TABLE OF CONTENTS
ABSTRACT……………………………………………………………………………………...ii
TABLE OF CONTENTS………………………………………………………………………..iii
LIST OF TABLES.......................................................................................................................vii
LIST OF FIGURES......................................................................................................................vii
LIST OF ABBREVIATIONS.....................................................................................................viii
CHAPTER 1……………………………………………………………………………………...1
1. INTRODUCTION………………………………………………………………………..1
1.1. The metabolic syndrome……………………………..…………………………...1
1.2. Metabolic dyslipidemia: A common component of the metabolic syndrome..…..2
1.3. Postprandial hyperlipidemia in insulin resistant states…………………..……….2
1.4. Factors affecting postprandial metabolism ………………………………….…...3
1.5. Intestinal lipoprotein overproduction in insulin resistance……………………….3
1.6. Dietary fatty acid uptake………………………………………………………….4
1.6.1. The role of CD36 in fatty acid uptake……………………………...…….5
1.6.2. The role of fatty acid binding proteins in fatty acid uptake……………....6
1.6.3. The role of fatty acid transport proteins in fatty acid uptake……………..9
1.7. Intestinal cholesterol absorption………………………………………………...10
1.8. Chylomicron assembly: A characteristic property of enterocytes………………11
1.9. Triacyglycerol synthesis………………………………………………………...13
1.10. The process of chylomicron assembly……………………………………..……15
1.11. Specific aims of the study…………………………………………………...…..16
1.11.1. Experimental models……………………………………………………16
iv
1.11.2. Rationale………………………………………………………………...17
1.11.3. Hypothesis……………………………………………………………....18
1.11.4. Objectives……………………………………………………………….19
CHAPTER 2…………………………………………………………………………………….20
2. MATERIALS & METHODS................………………………..……………………….20
2.1. Animals………………………………………………………………………….20
2.2. Chemicals and reagents………………………………………………………....20
2.3. Laboratory supplies and apparata…………………………………………..…...21
2.4. Animal surgery………………………………………………………………….22
2.4. Primary enterocyte isolation…………………………………………….………22
2.5. Primary enterocyte cell culture……………………………………………...…..23
2.6. Caco-2 cell culture……………………………………………………………....23
2.7. LB broth and agar plate preparation………………………………………….…23
2.8. Transformation and plasmid DNA amplification……………………………….24
2.9. shRNA transfection in Caco-2 cells………………………………………..…...24
2.11. Selection of stably transfected Caco-2 cell clones………………………………25
2.12. Total RNA extraction………………………………………………………..….25
2.13. Quantitative real-time polymerase chain reaction................................................26
2.14. SDS-PAGE and western blotting………………………………………..……....27
2.15. Preparation of lipid micelles……………………………………………..……...28
2.16. Lipid micelle stimulation of Caco-2 cells or primary enterocytes………...….…29
2.17. Lipid extraction and thin layer chromatography………………….……………..30
2.18. Metabolic pulse labeling and immunoprecipitation of ApoB……………….…..30
2.19. Chylomicron density ultracentrifugation……………………………………......31
2.20. Triglyceride, cholesterol, and liver enzyme assays….…………………….……32
v
2.21. Cerulenin treatment…………………………………………………………......32
2.22. Statistical analysis……………………………………………………………….32
CHAPTER 3………………………………………………………………………………….....33
3. RESULTS……………………………………………………………………………….33
3.1. shRNA-mediated knockdown of CD36 and FABP1 in Caco-2 cells……….…..33
3.2. Depletion of CD36 or FABP1 does not impair FA upake in Caco-2 cells….…..34
3.3. CD36- or FABP1-deficiency is associated with increased TG secretion in Caco-2
cells……………………………………………………………………………..……….35
3.4. TG content of lipoproteins secreted by CD36- or FABP1-deficient Caco-2 cells
…………………………………………………………………………………………..36
3.5. CD36- or FABP1-deficiency is associated with increased ApoB secretion in
Caco-2 cells……………………………………………………………………………..37
3.6. Expression of genes involved in fatty acid synthesis………………………...…40
3.7. Cerulenin decreases fatty acid synthesis in Caco-2 cells……………………….43
3.8. De novo lipogenesis may contribute to the increase in TG or ApoB secretion
observed in CD36-, but not FABP1-deficient Caco-2 cells………………...……..……45
3.9. CD36-/- have impaired cholesterol synthesis and TG secretion......……..............47
3.10. Cellular levels of proteins involved in TG synthesis in CD36- and FABP1-
deficient Caco-2 cells.......................................................................................................49
3.11. Factors involved in chylomicron assembly in CD36- and FABP1-deficient
Caco-2 cells......................................................................................................................51
CHAPTER 4………………………………………………………………………………….....55
4. DISCUSSION……………………………………………………………………...........55
4.1. shRNA-mediated knockdown of CD36 and FABP1 in Caco-2 cells……….......55
4.2. CD36- or FABP1-deficiency in Caco-2 cells did not impair fatty acid uptake…56
vi
4.3. CD36- or FABP1-deficiency in Caco-2 cells is associated with an increase in TG
and ApoB secretion……………………………………………………………………...57
4.4. Depletion of CD36 in Caco-2 cells results in a shift in secreted lipoprotein
profile…………………………………………………………………………………....59
4.5. Evidence for an increase in de novo lipogenesis in CD36-deficient Caco-2
cells……………………………..…………………………………………………….....61
4.6. TG synthesis in CD36- or FABP1-deficient Caco-2 cells…...................……….64
4.7. Factors involved in chylomicron assembly and secretion....................................66
4.8. Fatty acid synthesis in enterocytes isolated from CD36-/- mice...........................67
4.9. Concluding Remarks…………………………………………………………....69
CHAPTER 5………………………………………………………………………………….....71
5. FUTURE DIRECTIONS………………………………………………………………..71
REFERENCES………………………………………………………………………………….72
vii
LIST OF TABLES
Table 1. RT-PCR Primer Sequences……………………………………………...………27
LIST OF FIGURES
Figure 1. Absorption of dietary lipid, and the assembly and secretion of chylomicrons….12
Figure 2. TG synthesis pathways.........................................................................................14
Figure 3. shRNA-mediated knockdown of CD36 or FABP1 in Caco-2 cells.....................34
Figure 4. Lipid uptake in CD36- or FABP1-deficient Caco-2 cells....................................35
Figure 5. Lipid assimilation and secretion in CD36 & FABP1 knockdown Caco-2 cells...36
Figure 6. TG content of lipoproteins secreted from normal and CD36-deficient Caco-2
cells.......................................................................................................................37
Figure 7. Apolipoprotein B secretion from CD36 or FABP1 knockdown Caco-2 cells......39
Figure 8. FAS protein levels in CD36 and FABP1 knockdown Caco-2 cells treated with or
without 1.6 mM oleate-containing lipid micelles for 12 hours............................41
Figure 9. Real-time PCR measurement of ACC1 mRNA in control and CD36 or FABP1
knockdown Caco-2 cells.......................................................................................42
Figure 10. Dose response curve for cerulenin, a FA synthase inhibitor, in Caco-2 cells......43
Figure 11. Cellular [3H]-lipids extracted from CD36 or FABP1 knockdown Caco-2 cells...46
Figure 12. Secreted [3H]-lipids extracted from the medium CD36 or FABP1 knockdown
Caco-2 cells..........................................................................................................47
Figure 13. Lipogenesis in enterocytes isolated from WT and CD36-/- mice..........................48
Figure 14. FATP4 protein expression levels in CD36 and FABP1 knockdown cells...........50
Figure 15. Real-time PCR measurement of DGAT2 mRNA in control and CD36 or FABP1
knockdown Caco-2 cells.......................................................................................51
Figure 16. Microsomal triglyceride transfer protein levels in CD36 and FABP1 knockdown
cells.......................................................................................................................52
viii
Figure 17. Sar1 GTPase protein expression levels in CD36 and FABP1 knockdown cells..53
Figure 18. Real-time PCR measurement of ApoAIV mRNA in control and CD36 or FABP1
knockdown Caco-2 cells.......................................................................................54
LIST OF ABBREVIATIONS
ApoB Apolipoprotein B APS Ammonium persulfate BSA Bovine serum albumin ß-Me ß-Mercaptoethanol DTT Dithiothreitol DGAT Diacylglycerol acyltransferase ECL Enhanced chemiluminescence EDTA Ethylenediaminetetraacetic acid ER Endoplasmic reticulum FABP Fatty acid binding protein FFA Free fatty acid HDL High density lipoprotein HRP Horseradish peroxidase KBr Potassium bromide mRNA messenger RNA MGAT Monoacylglyerol acyltransferase MTP Microsomal triglyceride transfer protein PBS Phosphate-buffered saline PI Protease inhibitor cocktail PPAR Peroxisome proliferators activated receptor PVDF Polyvinylidene fluoride SDS Sodium dodecyl sulfate SDS-PAGE Sodium dodecyl sulfate polyacrylamide gel electrophoresis TG Triacylglycerol TEMED N,N,N’,N’-tetra-methyl-ethylenediamine VLDL Very low density lipoprotein
1
CHAPTER 1
1. INTRODUCTION
1.1. The Metabolic Syndrome
Within the past century, a global trend towards a more sedentary lifestyle and an
increase in caloric intake has increased the prevalence of chronic disorders such as obesity and
type 2 diabetes. The rise in obesity and type 2 diabetes is so overwhelming that it is now
deemed a worldwide “epidemic” (Zimmet et al., 2001). The metabolic syndrome, formely
refered to as insulin resistance syndrome, is the culmination of the complications of obesity and
insulin resistant states. Furthermore, it is characterized by pathologies such as glucose
intolerance, insulin resistance, obesity, dyslipidemia, and hypertension (Avramoglu et al.,
2006). It has been predicted that, unless preventative measures are taken, the total number of
people with diabetes will rise to 366 million in less than 30 years (Meetoo et al., 2007). The
development of complications associated with diabetes such as retinopathy, nephropathy,
neuropathy, cardiovascular diseases, peripheral vascular diseases and stroke, will undoubtedly
lead to major socioeconomic costs (Meetoo et al., 2007).
The metabolic syndrome is a common pathological state that include a number of
complications such as obesity, insulin resistance, glucose intolerance, dyslipidemia, and
hypetension (Avramoglu et al., 2003). This cluster of risk factors significantly increases the risk
of cardiovascular disease and type 2 diabetes (Klein et al., 2002; Malik et al., 2004). Insulin
resistance, a decreased response to circulating plasma insulin levels, usually develops as the first
indicator of type 2 diabetes. At this stage, the pancreas becomes stressed as it tries to keep up
with the increasing demand for insulin due to tissue insulin resistance in order to maintain blood
glucose levels. Type 2 diabetes manifests when the pancreas fails and can no longer produce
enough insulin (Avramoglu et al., 2003).
2
1.2. Metabolic Dyslipidemia: A Common Component of the Metabolic
Syndrome
The dyslipidemia that is observed in insulin resistance and type 2 diabetes is the most
common and important risk factor associated with the development of cardiovascular disease
and atherosclerosis (reviewed in (Avramoglu et al., 2006)). Metabolic dyslipidemia is
characterized by an abnormal lipoprotein profile with elevated plasma TG, elevated small dense
LDL particles, and a decrease in HDL particles (Krauss, 2004; Taskinen, 2005). This abnormal
fasting lipemia is indicative of the magnitude of the postprandial dyslipidemia (Annuzzi et al.,
1989).
1.3. Postprandial Hyperlipidemia in Insulin Resistant States
Patients with type 2 diabetes present with abnormal plasma lipoprotein profiles in the
postprandial state (Mero et al., 1998). A study by Annuzzi, et al. investigated the role of insulin
resistance on the development of postprandial dyslipidemia in type 2 diabetic patients (Annuzzi
et al., 2004). Their sample consisted of type 2 diabetic patients with good control over blood
glucose, normal fasting lipoprotein profiles, yet they displayed an exaggerated postprandial
response (Rivellese et al., 2004). A hyperinsulinemic glycemic clamp ensured that the glucose
and insulin remained at similar levels between groups, such that the only difference was the
level of insulin sensitivity (Annuzzi et al., 2004). Independent of circulating glucose and insulin
levels, insulin resistant patients displayed abnormal postprandial lipoprotein profiles, pointing to
a direct role for insulin resistance in the development of the postprandial hyperlipemia observed
in the metabolic syndrome (Annuzzi et al., 2004).
3
1.4. Factors Affecting Postprandial Metabolism
The mechanism responsible for aberrant postprandial metabolism in insulin resistant
conditions is unclear. There is evidence of a role for lipoprotein lipase (Lpl), the enzyme that
catalyzes the hydrolosis of TG-rich lipoproteins for uptake into peripheral tissues (reviewed in
(Goldberg et al., 2009)). In a study by Annuzzi, et al. 2008, adipose tissue LpL activity was
significantly lower in diabetic subjects compared to control subjects, both in the fasted and
postprandial states (Annuzzi et al., 2008). These results suggest that LpL may play a role in the
exaggerated postprandial response observed in diabetic subjects (Annuzzi et al., 2008).
The free fatty acids (FFAs) that are liberated from lipoproteins through Lpl mediated
hydrolysis enter peripheral cells via passive diffusion or protein facilitated process. CD36,
which will be discussed in greater detail in section 1.6.1, plays a role in fatty acid (FA) uptake in
the heart, skeletal muscle and adipose tissue among other peripheral tissues (Coburn et al.,
2000). Moreover, CD36 deficiency significantly impairs FFA uptake in these tissues (Coburn et
al., 2000).
Lipoprotein TG is progressively hydrolyzed, forming what is referred to as a remnant
lipoprotein. There is substantial evidence that intestinal lipoprotein remnants pose an
atherosclerotic risk when they are chronically elevated (Havel, 2000; Karpe et al., 1994;
Weintraub et al., 1996). A myriad of proteins and receptors have been linked to the clearance of
chylomicron remnants, such as LDL receptor, LRP, SR-B1, hepatic lipase, and ABCA1
(reviewed in (Cianflone et al., 2008)).
1.5. Intestinal Lipoprotein Overproduction in Insulin Resistance
Although intestinal lipoprotein assembly is required for the absorption of dietary fat and
fat-soluble vitamins (Hussain et al., 2005), it has long been regarded as a passive process. There
is growing evidence that intestinal lipoprotein overproduction is a major contributor to the both
4
fasting and postprandial lipemia observed in these pathological conditions. The intestine
contributes to the pathologies in a number of ways. Intestinally derived lipoproteins,
particularly chylomicrons (CMs), can accelerate the development of obesity by delivering
energy-rich TG to adipose depots (Williams, 2008). Due to an increased activity of lipoprotein
lipase (LpL) in adipose and a decreased activity in muscle, there is a diversion of CM
triacylglycerol (TG) from combustion in muscle, into adipose tissue for storage in sedentary
individuals (Mead et al., 2002; Hamilton et al., 2007). As the CM delivers TG to peripheral
tissues, it becomes a more dense, cholesterol (CH)-rich particle. These small dense particles
enter the arterial wall and promote the formation of atherogenic plaques. Finally, CMs interfere
with the catabolism of atherogenic hepatic lipoproteins, which also act as key mediators for the
formation of atherosclerotic plaques (Karpe and Hultin, 1995).
1.6. Dietary FA Uptake
In the lumen of the small intestine, dietary fat is hydrolyzed into its components,
monoacylglycerol (MG) and FFAs that are then dispersed in bile acids (Fig. 1.). In close
proximity to the enterocyte surface, the pH is lower, causing protonation of the FAs. Free FAs
then dissociate from the bile salt micelles and either passively diffuse or are transported across
the brush border membrane by protein mediated transport. Several proteins have been
implicated in the facilitated uptake of FAs by the enterocyte including cluster determinant 36
(CD36), fatty acid binding protein (FABPpm), and fatty acid transport protein family members
(FATPs) (Abumrad et al., 1999). Both CD36 and FABPpm are thought to reside in specialized
microdomains called lipid rafts (Ehehalt et al., 2006). Lipid rafts have a higher content of
highly saturated hydrocarbon chain shingolipids, which pack more tightly than other
phospholipids species, and are therefore more ordered microdomains of the plasma membrane
(Ehehalt et al., 2006).
5
1.6.1. The Role of CD36 in FA Uptake
CD36 is a scavenger receptor that is expressed in many cell types such as
megakaryocytes, erythroid precursors, platelets, monocytes, dendritic cells, adipocytes,
myocytes, retinal and mammary epithelial cells, and endothelial cells of the microvasculature
and the small intestine. CD36 deficiency is common in subpopulations that are at higher risk of
developing type 2 diabetes (Abumrad et al., 1999). However, studies in humans and mice do
not consistently present an association of CD36 deficiency and insulin responsiveness. Thus,
the effect of CD36 deficiency on the development of pathological conditions may strongly
depend on dietary conditions.
CD36 was initially recognized as a modulator of FA uptake in adipocytes (Abumrad et
al., 1981). Subsequent labeling of the protein with membrane impermeable FA analogs lead to
the identification of a membrane protein that was shown to have 85% identity to CD36
(Abumrad et al., 1993). Nada Abumrad’s group studied FA uptake and utilization in CD36-/-
mouse tissues using [3H]-palmitic acid and a non-oxidizible FA analog, β-methyl 15-(p-
iodophenyl) pentadecanoic acid (BMIPP) (Coburn et al., 2000). They showed that FA uptake
was decreased in the fed state in the heart, skeletal muscle, and adipose tissues of CD36-/- mice
(Coburn et al., 2000), however, FA uptake in the intestine was not significantly different
between CD36-/- and wild type mice. There was a 2-fold reduction in BMIPP and [3H]-palmitic
acid incorporation into TG in muscle and adipose tissue, and an 2-fold increase in labeled
diacylglycerols in isolated adipocytes. However, there was no change in the activity of the
enzyme responsible for the conversion of diacylglycerols to TG, diacylglycerolacyltransferase
(DGAT). There was no change in the long chain acyl-CoA synthetase enzymes in muscle and
adipose tissues. The authors postulate that a decreased rate of TG production in these tissues
may be due to the lower affinity of DGAT for fatty acyl-CoAs (Coburn et al., 2000).
6
In the intestine, CD36 deficiency was not associated with decreased uptake of either
[3H]-palmitic acid or BMIPP (Coburn et al., 2000), despite its similar localization to other
proteins implicated in FA uptake (Poirier et al., 1996). However, Drover, et al. demonstrated
that CD36 is involved in the uptake of FA through its involvement in chylomicron formation in
CD36-/- mice enterocyte (Drover et al., 2005). In this study, CD36-/- or wild type mice were
given a bolus of olive oil of FAs by intragastric gavage and the secretion of lipids into the lymph
was measured. There was an increase in BMIPP accumulation in the proximal section of the
small intestine in CD36-/- mice compared to wild type mice, which was not due to an impairment
of FA uptake (Drover et al., 2005). There was a decrease in microsomal TG in CD36-/-
enterocytes compared to wild type animals, but unlike adipocytes and muscle, DAG did not
accumulate in the microsome fraction of isolated enterocytes from CD36-/- mouse intestine
(Drover et al., 2005). These data suggest that CD36 is responsible for directing FAs towards the
microsomal secretion TG pool (Drover et al., 2005). How CD36 accomplishes the intracellular
trafficking of FAs is completely unknown, but Drover, et al. hypothesized that CD36-rich
membrane vesicles shuttle FA with the help of trafficking proteins that cycle between the ER
and plasma membrane (Drover et al., 2005).
Examination of lipid output from the intestine of these animals revealed that TG
secretion into the lymph was significantly decreased in CD36-/- mice compared to wild type
mice (Drover et al., 2005). Again, these data emphasize a role for CD36 in intestinal lipid
absorption and CM secretion.
1.6.2. The Role of Fatty Acid Binding Proteins in FA Uptake
Once the FAs are transported across the plasma membrane, they encounter a family of
lipid chaperones called fatty acid binding proteins (FABP). In the intestine, there are a two
main FABPs, intestinal-FABP (FABP2) and liver-FABP (FABP1) (Iseki et al., 1990). L-FABP
7
is a small (14 kDa) cytoplasmic protein that can bind long chain FAs, long chain fatty-acyl CoA
and an array of other molecules including carcinogens, lysophospholipids, acyl-coenzyme A,
eicosanoids and heme (Coe and Bernlohr, 1998). The promoter of L-FABP contains a
peroxisome proliferators response element, but its mRNA is also regulated by FAs, dicarboxylic
acid and retinoic acid (Rolf et al., 1995). It has been suggested that L-FABP can act as a co-
activator of PPARα-mediated gene activation, since L-FABP and PPARα physically interact
(Furuhashi and Hotamisligil, 2008). Unlike other members of the FABP family, L-FABP can
bind two ligands simultaneously (Furuhashi and Hotamisligil, 2008). This property of L-FABP
is suggested to serve as a feature enabling ligand delivery through interactions with target
receptors (Furuhashi and Hotamisligil, 2008).
L-FABP deficient mice have no change in appearance, gross morphology or viability.
They were of normal weight and had normal serum TG and FA levels (Cianflone et al., 2008).
However, the metabolic parameters in these mice upon exposure to high-fat/cholesterol diet
differed between studies (Cianflone et al., 2008). Some evidence suggests that L-FABP is
necessary for the formation of chylomicrons (also discussed in section 1.10. The Process of
Chylomicron Assembly and Secretion). Charles Mansbach’s group demonstrated the
involvement of L-FABP in the formation of chylomicrons. L-FABP was shown to select cargo
and its presence in the COPII protein-deficient cytosol was sufficient to bud prechylomicron
transport vesicles (PCTV) from the ER in an in vitro budding assay (Neeli et al., 2007).
I –FABP is only expressed in the small intestine and thought to be involved in the
transport of long-chain FAs (Kim et al., 2001). Studies in Caco-2 cells suggest that I-FABP
may actually be involved in the reduction of FA uptake (Darimont et al., 2000). Through an
unknown mechanism, overexpression of I-FABP inhibited FA incorporation into triglyceride in
differentiated Caco-2 cells supplemented with FA (Darimont et al., 2000). Characterization of
the I-FABP-null mouse has lead to a new hypothesis for the role of I-FABP; its deletion did not
8
affect development or dietary fat absorption. The mice were hyperinsulinemic, but showed
genetic differences in body weight and plasma parameters (Darimont et al., 2000). Male I-
FABP-null mice had elevated plasma TG and weighed more than wild type animals on both low
and high fat diets (Darimont et al., 2000), while female I-FABP-null mice gained less weight
than wild-type mice. The authors suggest that I-FABP functions as a lipid-sensing factor in the
maintenance of energy homeostasis and may not be directly involved in the uptake of FAs
(Darimont et al., 2000).
In a recent study, protein levels of both intestinal I-FABP and L-FABP were measured in
two hereditary lipid malabsorption syndromes, Abetalipoproteinemia (ABL) and Anderson's
disease (AD) (Guilmeau et al., 2007). Both of these disorders are characterized by absent or
abnormal lipid and lipoprotein metabolism. ABL results from a mutation in microsomal
triglyceride transfer protein (MTP), a chaperone necessary for the lipidation of apoB (Wetterau
et al., 1992). In patients with this disorder, apoB100 and apoB48 are absent from the plasma
and accumulation of lipid droplets in the cytoplasm of enterocytes (Wetterau et al., 1992;
Shoulders et al., 1993). A mutation in the SARA 2 gene encoding the Sar1 GTPase causes
Anderson’s disease (Jones et al., 2003). This disorder is characterized by a retention of
chylomicrons in the enterocyte, presumably because of a lack of the appropriate intracellular
trafficking machinery (Jones et al., 2003). Protein levels, but not mRNA levels, of I-FABP and
L-FABP were decreased in patients with either ABL or AD compared to normal subjects
(Guilmeau et al., 2007). The authors suggest that a reduction of I-FABP or L-FABP may be
beneficial in these pathological conditions associated with intracellular lipid accumulation
(Guilmeau et al., 2007). Their postulation is that a decrease in I-FABP or L-FABP would
prevent conversion of FA to TG, preventing additional intestinal injury (Guilmeau et al., 2007).
9
1.6.3. The Role of Fatty Acid Transport Proteins in Fatty Acid Uptake
CD36 is thought to be involved in the uptake of FAs by binding them at the plasma
membrane (Ibrahimi and Abumrad, 2002; Abumrad et al., 1999), and FABPs are chaperones
that are involved in lipid signaling, trafficking, esterification, and chylomicron formation.
Another family of lipid binding proteins, FATPs, play a role in the translocation of FAs across
the plasma membrane (Stahl et al., 2001). There are six mammalian FA transport protein
members that are expressed in fat utilizing tissues (Ehehalt et al., 2006). Of the six, only
FATP4 is expressed in the intestine (Ehehalt et al., 2006; Stahl et al., 1999). FATP4 deletion
causes embryonic lethality or perinatal lethality with a phenotype that similar to lethal restrictive
dermopathy (Herrmann et al., 2003). Herrmann, et al. showed that FATP4 is implicated in
ceramide biosynthesis, perhaps by acylating very long chain FAs. There was an increase in
ceramide and cholesterol content in the skin of these animals, however, very long chain FA
substitutes were significantly reduced in the ceramide fraction (Herrmann et al., 2003). In
another study, FATP4 null mice died shortly after birth and it appeared that FATP4 plays an
essential role in the formation of the epidermal barrier. FATP4 heterozygosity in the same study
lead to reduced LCFA uptake by the enterocyte, but did not affect lipid absorption or body
composition in vivo (Hall et al., 2005).
Recently, a group rescued the defective skin phenotype in FATP4 null mice by using an
FATP4 transgene driven by a keratinocyte-specific promoter (Fatp4–/–;Ivl-Fatp4tg/+) in order to
study intestinal lipid absorption (Shim et al., 2009). In short, there were no differences in food
consumption, weight gain, plasma lipid parameters, or intestinal lipid absorption in WT vs.
Fatp4–/–;Ivl-Fatp4tg/+ on normal chow diets, although Western diet fed Fatp4–/–;Ivl-Fatp4tg/+
mice showed a significant increase in enterocyte TG and FA content. There was no
compensation observed by any other FATP family member, or by any FA or cholesterol
10
transporter in Fatp4–/–;Ivl-Fatp4tg/+ mice (Shim et al., 2009). The current proposed mechanism
by which FATP4 facilitates uptake of FAs is by providing acyl CoA synthetase activity
(Herrmann et al., 2003). Moreover, FA uptake is indirectly driven by FATP4 activity;
activation of FAs to fatty acyl-CoA moieties shifts the equilibrium towards uptake of FAs at the
plasma membrane (Herrmann et al., 2003). The products of FATP4 activity, fatty acyl-CoAs,
are substrates for TG synthesis and are directed towards TG synthesis enzymes that are located
on the ER membrane.
1.7. Intestinal Cholesterol Absorption
In the lumen of the intestine, cholesterol is taken up by enterocytes and secreted into the
intestinal lymph mostly as cholesteryl ester. Cholesterol is esterified through the action of acyl-
CoA cholesterol acyl transferase (ACAT) (Gallo et al., 1984). Our understanding of the process
cholesterol absorption increased dramatically from the discovery of ezetimibe, an inhibitor of
cholesterol absorption (Clader, 2004). The target of ezetimibe is now known to be the
Niemman-Pick C1-like protein 1 (NPC1L1) (Garcia-Calvo et al., 2005). Scavenger Receptor
Class B Type I (SR-BI) is another key player in the absorption of cholesterol. SR-BI binds
lipoprotein particles on the surface of cells to facilitate cholesterol uptake (Connelly and
Williams, 2004). CD36 has also been shown to mediate cholesterol uptake in enterocytes (Nauli
et al., 2006). Important to this study, is the regulation of intestinal chylomicron assembly and
secretion by cholesterol. Pal, et al. examined the regulatory role of sterols, cholesterols, and
cholesteryl esters on the secretion apoB48 containing lipoproteins in Caco-2 cells (Pal et al.,
2002). Their results indicate that cholesteryl ester formation regulates apoB48 secretion in these
cells (Pal et al., 2002).
11
1.8. Chylomicron Assembly: A Characteristic Property of Enterocytes
CMs are very large buoyant lipoprotein particles produced by the intestine upon
ingestion of fat. Their physiological role is to transport dietary fat throughout the circulation,
thereby delivering energy to peripheral tissues. Chylomicrons consist of a neutral lipid core of
unordered TG and cholesteryl ester (CE) and are surrounded by a phospholipid monolayer
embedded with free cholesterol (Hussain, 2000). The key structural compenent of chylomicrons
is a large monomeric protein apolipoprotein B48 (apoB48). The liver synthesizes apoB100 for
the assembly of VLDL particles. In the intestine, apoB48 is translated from the full length
message, or apoB100, but is posttranscriptionally modified by deamination of a cytosine to
uracil, ending protein synthesis at codon 2153 (reviewed in (Hussain, 2000)). Since the liver
does not have the capacity to assembly CM or synthesize apoB48, it was thought that CM
synthesis dependent on apoB48 synthesis. This hypothesis was investigated by Dr. Hussain’s
group, who created Caco-2 cells that stably express apoB48 independent of mRNA editing
activity (Luchoomun et al., 1997). Their observation that undifferentiated Caco-2 cells secrete
apoB-containing lipoprotein particles of different sizes, but not CM particles, suggested that
expression of apoB48 could not drive the assembly of larger CM particles (Luchoomun et al.,
1997). Since apoB48 expression was not sufficient for CM assembly, this group postulated that
formation of CM particles may be a unique property of differentiated enterocytes. To this end,
Caco-2 cells were allowed to differentiate and challenged with fatty acid (FA) (Luchoomun and
Hussain, 1999). However, both apoB100 and apoB48 had the capacity to assemble CM
particles (Luchoomun and Hussain, 1999). Studies in mice that produce only apoB100 from the
liver and intestine showed that apoB100 could replace apoB48 in the intestine (Farese, Jr. et al.,
1996). Numerous studies in rodent hepatocytes that can synthesize apoB48 in addition to
12
apoB100 showed that this cell type cannot be induced to assemble CM particles (Boren et al.,
1994; Hussain et al., 1989; Hussain et al., 1995). Therefore, CM assembly is a characteristic
property of differentiated enteroctes (reviewed in (Hussain, 2000)).
ApoB48
VAMP7VAMP7Sar1Sar1
MTPMTP
2. TG2. TG--rich lipid rich lipid droplet formationdroplet formation
Phospholipid
Golgi
FABPs
Endoplasmic reticulum
FABP1FABP1
FFA
Dietary TriacylglycerolDietary Triacylglycerol MG + Free Fatty Acids (FFA)MG + Free Fatty Acids (FFA)Intestinal LumenIntestinal Lumen
CD36CD36
PCTVPCTV
MTPMTP
ChylomicronsChylomicrons
dede novo novo lipogenesislipogenesis
FFA
MGATMGATDGATDGAT FATP4FATP4
Lymphatic CirculationLymphatic Circulation
ApoAIVApoAIV
1. Primordial particle
MTPMTP 3. Nascent chylomicron/PCTV
budding
ApoB48
VAMP7VAMP7Sar1Sar1
MTPMTP
2. TG2. TG--rich lipid rich lipid droplet formationdroplet formation
Phospholipid
Golgi
FABPs
Endoplasmic reticulum
FABP1FABP1
FFA
Dietary TriacylglycerolDietary Triacylglycerol MG + Free Fatty Acids (FFA)MG + Free Fatty Acids (FFA)Intestinal LumenIntestinal Lumen
CD36CD36
PCTVPCTV
MTPMTP
ChylomicronsChylomicrons
dede novo novo lipogenesislipogenesis
FFA
MGATMGATDGATDGAT FATP4FATP4
Lymphatic CirculationLymphatic Circulation
ApoAIVApoAIV
1. Primordial particle
MTPMTP 3. Nascent chylomicron/PCTV
budding
Figure 1. Absorption of dietary lipid, and the assembly and secretion of chylomicrons. CD36 facilitates FA uptake. FAs are bound by cytosolic FA binding proteins that deliver substrates to TG synthesis enzymes (MGAT and DGAT) on the ER membrane. Chylomicron assembly is thought to proceed in three distinct steps: the co-translational lipidation of apoB with phospholipids and some TG by MTP, to form a primordial particle (1); formation of TG-rich lipid droplets (2); and ‘core expansion’, or fusion of the primordial particle and TG-rich lipid droplets (3). The budding of the prechylomicron transport vesicle may be the site of ‘core expansion’ and involves the selection of cargo by FABP1. PCTVs fuse with the Golgi in a COPII dependent manner. Mature chylomicrons are secreted via the basolateral membrane into the lymphatic circulation.
13
1.9. TG Synthesis
The intestinal enterocyte can assimilate TG via two separate pathways: glycerol
phosphate (G3P) or Kennedy pathway (CLARK and HUBSCHER, 1961) and the
monoacylglycerol (MG) pathway (HUBSCHER and CLARK, 1960) (Fig. 2.). The enzymes
involved in the two pathways of TG synthesis are found in different subcellular regions, keeping
the products of acylation compartmentalized (Higgins and Barrnett, 1971). Higgins, et al.
showed that acyltransferases from the MG pathway were localized to the smooth ER while those
from the G3P pathway were localized to the rough ER (Higgins and Barrnett, 1971). The first
step in the MG pathway is catalyzed by monoacylglycerol acyltransferase (MGAT), which
acylates monoacylglycerol with fatty acyl-CoA (FA-CoA) to form diacyglycerol (DG)
(reviewed in (Mansbach and Gorelick, 2007)). Subsequent acylation of DG by DGAT yields
TG (Mansbach and Gorelick, 2007).
Two mammalian DGAT enzymes have been identified. DGAT1 was first identified in
1998 as having a similar sequence to acyl-CoA:cholesterol transferase (ACAT) and possessing
diacylglycerol transferase activity (Cases et al., 1998). DGAT1 is a member of a family of
membrane-bound O-acyltransferases (MBOAT) that catalyze the addition of acyl-CoA moieties
to hydroxyl or thiol groups of lipids or protein molecules (reviewed in (Yen et al., 2008)).
Other MBOAT family members include acyl-CoA: cholesterol acyltransferase (ACAT) which
covalently joins cholesterol and fatty acyl-CoA to form cholesteryl ester. DGAT2 was sought
after following the development and characterization of the DGAT1-/- mouse, since lack of
DGAT1 did not cause a defect in TG assimilation (reviewed in (Yen et al., 2008)).
Mammalian DGAT2 was cloned in 2001, followed by identification of other members of
DGAT2 family, including MGAT1, MGAT2 and MGAT3 (reviewed in (Yen et al., 2008)).
Studies in DGAT2-/- mice suggest that DGAT2 is the major DGAT involved in TG synthesis
14
and that it may be involved in FA biosynthesis (Stone et al., 2004). Evidence for this comes
from studies by Man, et al. who have shown that DGAT2 and stearoyl-CoA desaturase1 (SCD1)
colocalize and interact in HeLa cells (Man et al., 2006). This suggests that DGAT2 may be
responsible for incorporating the monounsaturated FA, oleoyl-CoA, into TG (Man et al., 2006).
TG can also be synthesized by the G3P pathway, which begins with the consecutive
acylation of glycerol-3-phosphate by GPAT and AGPAT to form lysophosphatidate and
phosphatidate, respectively. Phosphatidate is then dephosphorylated by PAP to form DG, which
undergoes acylation to yield TG. The DG intermediate in the G3P pathway may also go on to
form phospholipids, however, DG formed by the MG pathway can only form TG (Mansbach
and Gorelick, 2007).
The MG pathway is used in tissues that re-esterify TG from its dietary lipolytic products,
such as the liver and the intestine. The MG pathway contributes to approximately 80% of CM
TG, while the G3P pathway contributes only 20% of CM TG (Breckenridge and Kuksis, 1975).
Figure 2. TG synthesis pathways. The predominant pathway for synthesis of TG in normal primary enterocytes is the monoglyceride (MG) pathway, while the predominant pathway used by Caco-2 cell is the glycerol phosphate (G3P) pathway, involving the enzymes glycerol-3-phosphate acyltransferase (GPAT), 1-acylglycerol-3-P-acyltransferase (AGPAT), phosphatidic acid phosphatase (PAP), and diacylglycerol acyltransferase (DGAT). DAG produced via the G3P pathway can be used for the biosynthesis of phospholipids, while DAG produced by the MG pathway cannot.
GPAT (acylCoA:glycerol-3-phosphate acyltransferase)
Monoacylglycerol (MG) Pathway
Monoacylglycerol (MG)
Glycerol Phosphate (G3P) Pathway
Glycerol-3-Phosphate (G3P)
Lysophosphatidic acid (LPA)
Phosphatidic acid (PA)
MGAT (monoacylglcyerol
acyltransferase)
PAP (phosphatidic
acid phosphotase)
AGPAT (1-acylglycerol-3-P-acyltransferase)
FA-CoA FA-CoA
FA-CoA
Diacylglycerol (DAG) Diacylglycerol (DAG)
Triacylglycerol (TG) Phospholipids (PL)
DGAT (diacylglcyerol acyltransferase)
GPAT (acylCoA:glycerol-3-phosphate acyltransferase)
Monoacylglycerol (MG) Pathway
Monoacylglycerol (MG)
Glycerol Phosphate (G3P) Pathway
Glycerol-3-Phosphate (G3P)
Lysophosphatidic acid (LPA)
Phosphatidic acid (PA)
MGAT (monoacylglcyerol
acyltransferase)
PAP (phosphatidic
acid phosphotase)
AGPAT (1-acylglycerol-3-P-acyltransferase)
FA-CoA FA-CoA
FA-CoA
Diacylglycerol (DAG) Diacylglycerol (DAG)
Triacylglycerol (TG) Phospholipids (PL)
DGAT (diacylglcyerol acyltransferase)
15
1.10. The Process of Chylomicron Assembly and Secretion
Overall, the assembly of chylomicrons is thought to proceed in a similar manner to that
of VLDL assembly in the liver, since lack of MTP has a similar detrimental effect on the
assembly of CM and VLDL particles. The co-translational lipidation of apoB by MTP is
necessary for the release of apoB from the ER and for the formation of primordial lipoprotein
particles (reviewed in (Hussain, 2000)). However, the subsequent steps involved in lipidation of
the primordial particle to form either VLDL or CM are still unclear. Tso, et al. proposed an
independent model of chylomicron assembly, suggesting that VLDL and CM assembly proceed
by different mechanisms (Tso et al., 1984). Moreover, that VLDL assembly is constitutive,
while CM assembly occurs in the postprandial state. A key piece of evidence supporting this
model is that Pluronic L81 (PL81) affects the secretion of CM but not VLDL particles (Tso and
Gollamudi, 1984). PL81 is a hydrophobic surfactant that has been shown to increase cytosolic
TG (Fatma et al., 2006). Mammoud Hussain’s group proposes that PL81 disrupts the transport
of cytosolic TG to the ER for secretion (Fatma et al., 2006), which may explain why PL81
inhibits assembly of more buoyant CM particles but not of smaller particles.
On the other hand, Dr. Hussain proposed a sequential model of assembly, involving the
same initial lipidation of apoB by MTP (Hussain et al., 2005), followed by ‘core expansion’ of
the primordial particle by the addition of TG to form a nascent chylomicron particle (Hussain,
2000). This model involves three distinct steps, the first being the co-translational lipidation of
apoB, where synthesis of apoB is associated with interaction with phospholipids and requires
the activity of MTP. Second, is the formation of lipid droplets in the lumen of the smooth ER.
This stage is independent of apoB synthesis, but would differ in the postprandial state as more
substrate is available for TG synthesis. Presumably, the increase in substrate would lead to the
formation of larger TG droplets and hence, yield larger lipoprotein particles during the third step
16
of this model, ‘core expansion’(Hussain, 2000). This step involves the fusion of the primordial
particles with the preformed lipid droplets residing in the smooth ER.
Charles Mansbach’s group proposed that a vesicle transporting the developing
chylomicron from the ER to the Golgi, is the rate limiting step in TG transport across the
enterocyte (Kumar and Mansbach, 1997; Kumar and Mansbach, 1999). Evidence to support this
came from studies that showed the rapid esterification of oleic acid into TG, despite a long (4
hour) lag time to obtain steady state TG secretion (Mansbach and Nevin, 1998). They identified
a vesicular subcellular fraction, named the prechylomicron transport vesicle (PCTV) that
functions to transport the nascent chylomicron particles from the ER to the Golgi (Kumar and
Mansbach, 1999). The formation of this vesicle is hypothesized to be the site of the ‘core
expansion’ step (Hussain, 2000). The developing CM particle traverses the Golgi, undergoing
further maturation, and is secreted via the basolateral membrane into the lymphatic circulation.
1.11. Specific Aims of the Study
1.11.1. Experimental Models
Caco-2 cells are an intestinal cell line derived from a human colorectal carcinoma and
have been extensively used to study intestinal lipid absorption. Caco-2 cells are the only
enterocyte cell culture model available for the study of lipid and lipoprotein processing
(reviewed in (Levy et al., 1995)). The differentiation of Caco-2 cells proceeds through three
states, the first being a homogenous undifferentiated state. Upon plating on polycarbonate
microporous membrane plates (Transwell® filter plates), the cells begin to heterogeneously
differentiate, displaying differences in morphology and brush border organization. Finally, after
approximately 21-30 days of culture, Caco-2 cells become homogenously polarized and
differentiated (reviewed in (Levy et al., 1995)). The potential for genetic manipulation and their
17
enterocyte-like secretory profile makes Caco-2 cells a suitable model for the study of intestinal
chylomicron assembly and secretion.
Differentiated Caco-2 cells secrete mainly LDL-sized lipoprotein particles when they
are not stimulated with lipid (Hughes et al., 1987a). Several investigators have explored the
effect of different fatty acids on the formation of larger lipoproteins (Dashti et al., 1990; Hughes
et al., 1987b; Luchoomun et al., 1997; Luchoomun and Hussain, 1999; Traber et al., 1987).
Specifically, Luchoomum, et al. performed a detailed analysis of chylomicron secretion from
Caco-2 cells using different concentrations of oleate and the bile salt derivative, taurocholate
(Luchoomun and Hussain, 1999). The optimal concentration of oleate and taurocholate for the
induction of CM assembly and secretion was found to be 1.6 mM and 1.0 mM, respectively
(Luchoomun and Hussain, 1999).
Importantly, there is a difference in the TG synthesis pathway that predominates in
Caco-2 cells and normal absorptive enterocytes. As mentioned above, the MG pathway is
characteristic of absorptive enterocytes, and is especially important in the postprandial state. In
mature Caco-2 cells, however, MGAT activity was less than 1/10th of that in rat jejunal cells
(Levin et al., 1992). The G3P pathway, therefore, is the main pathway for TG synthesis in
Caco-2 cells under both basal and lipid stimulated conditions.
In order to compare the in vitro results obtained in this study to an in vivo model, we
made use of the CD36-/- mouse model. To study the effect of CD36 knockdown on intestine
lipid metabolism, and more specifically, FA and TG synthesis, we followed a protocol that has
been optimized in our lab for the isolation of primary enterocytes (Haidari et al., 2002).
1.11.2. Rationale
The process of lipid absorption, defined as the uptake, assembly into lipoproteins, and
secretion of lipid into the lymphatic circulation, may be subject to regulation at several steps.
18
Under pathological conditions, such as obesity and insulin resistance, dysregulation of this
process leads to dyslipidemia. Changes in levels of FA transporters such as CD36 and FABPs
may lead to aberrant uptake or mistargetting of FAs (Coburn et al., 2000; Drover et al., 2005;
Kim et al., 2001; Masuda et al., 2008; Nassir et al., 2007; Nauli et al., 2006; Neeli et al., 2007)
and changes in proteins involved in the assembly and trafficking of lipoprotein particles causes
lipid malabsorption disorders (Avramoglu et al., 2003; Jones et al., 2003; Lewis et al., 2004;
Masuda et al., 2008; Wetterau et al., 1992). It remains unclear however, how deficiency of
these intestinal lipid transporters contributes to aberrant intestinal lipoprotein production.
Furthermore, whether the expression of these lipid transporters reflects the rate of lipid
absorption, or if it determines the incorporation of lipid into lipoprotein particles, is also of
significant interest. Elucidating the role of proteins potentially involved in the assembly of
chylomicrons in an enterocyte cell culture model may provide insight into the mechanisms
responsible for the dyslipidemia observed under pathological conditions.
1.11.3. Hypothesis
The hypothesis of the proposed research is that depletion of either intestinal FA
transporter, CD36 or FABP1, will impair intestinal lipoprotein assembly and secretion. Both
proteins are highly expressed in the proximal intestine and their involvement in lipid absorption
has been well documented. However, in studies of whole body CD36 or FABP1 knock out
animals, their influence on lipid and lipoprotein metabolism may be masked by peripheral
factors. To delineate a role for these proteins in intestinal chylomicron assembly and secretion,
we will make use of an enterocyte cell culture model. In this isolated model, external factors
such as hormonal changes and peripheral clearance of lipoproteins should not impede the
interpretation of the role of these proteins in the intestine.
19
1.11.4. Objectives
GENERAL OBJECTIVE: The overall objective of this project was to examine potential
molecular factors involved in intestinal chylomicron assembly.
SPECIFIC AIM 1: To knock down putative factors involved in chylomicron
assembly, CD36 and FABP1, in Caco-2 cells, an
enterocyte cell culture model.
SPECIFIC AIM 2: To examine assembly of CM particles, apoB synthesis and
secretion, and TG synthesis and secretion under
knockdown conditions.
20
CHAPTER 2
2. MATERIALS and METHODS
2.1. Animals
Male CD36-/- mice bred on the C57BL/6 background, and wild type C57BL/6 mice from
12 to 15 weeks of age were obtained from Dr. Maria Febbraio, Cleveland Clinic, via Dr. Kevin
Kain, University of Toronto. Following a one week acclimatization period, animals underwent
surgery as described below. Anaesthetics used for surgery (isofluorane, nitrous oxide, and
oxygen) were from Baxter (Toronto, ON).
2.2. Chemicals and Reagents
Cell Recovery Solution was from BD Biosciences (Bedford, MA). Complete-Mini,
EDTA-free protease inhibitor cocktail (PI) and Colourimetric TG assay was purchased from
Roche (Mississauga, ON). [3H]-oleic acid and [3H]-acetate were obtained from PerkinElmer
(Boston, MA). [35S]-methionine was from PerkinELmer Life Science, Inc. (Woodbridge, ON).
Bovine Serum Albumin (BSA), β-mercaptoethanol, glycerol, polyoxyethylenesorbitam
monolaurate (Tween 20), calcium chloride, cholesterol, cholesteryl oleate, L-α-
phosphatidylcholine, and oleic acid etc etc etc, were purchased from Sigma Aldrich (St. Louis,
MO). Acetic acid (glacial) and petroleum ether were from Fisher Scientific (Nepean, ON).
Methanol, ethanol, and 2-propanol were from Caledon Laboratories Ltd. (Georgetown, ON).
SureSilencingTM shRNA plasmids and CD36 and FABP1 RT-PCR primers were
purchased from SA Biosciences Corp. (Frederick, MD). LipofectamineTM 2000 was purchased
from Invitrogen (Carlsbad, CA). Power SYBR Green master mix was purchased from Applied
Biosystems (CA, USA). Antibiotics G418 and Puromycin were purchased from Sigma Aldrich
(St. Louis, MO). Eagle's Minimal Essential Medium (EMEM), Dulbecco’s Minimal Essential
21
Medium (DMEM), methionine, cysteine, and glutamine-free DMEM, L-glutamine, fetal bovine
serum, and Trypsin-EDTA, were purchased from Wisent, Inc. (Montreal, QC).
Agarose was purchased from Invitrogen Life Technologies (Grand Island, NY).
Molecular weight marker was from Fermentas (Burlington, ON). All Real-time polymerase
chain reaction primers were purchased from Sigma Aldrich (St. Louis, MO). All SDS-PAGE
including 40% acrylamide/bis solution (29:1 with 3.3%C), tris(hydroxymethyl)-aminomenthane
(TRIS), glycine, sodium dodecyl sulfate (SDS), ammonium persulfate (APS), and N,N,N,N’,N’-
tetramethyl-ethylenediamine (TEMED), blocking-grade non-fat dry milk and the Bradford-
based protein assay kit were purchased from Bio-Rad. Horse radish peroxidase (HRP)-
conjugated secondary antibodies and Western Lightning chemiluminescence detection reagents
(ECL) were purchased from Amersham Biosciences. PVDF membranes were purchased from
PerkinElmer (Boston, MA).
The anti-MTP, anti-FATP4, and anti-β-actin antibodies were purchased from Sigma
Aldrich (St. Louis, MO). The anti-Sar1 antibody was from Stressgen Bioreagents (Ann Arbor,
MI). The anti-FAS antibody was from Novus Biologicals (Littleton, CO). The anti-CD36
antibody was purchased from Abcam Inc. (Cambridge, MA).
2.3. Laboratory Supplies and Apparata
Micropipette tips (10 μl- 5 ml), screw-cap tubes (15 ml & 50 ml) and microfuge tubes
(1.5 ml & 2 ml) were from Sarstedt Inc. (Montreal, QC). Syringes (1 ml, 3 ml, 5 ml, and 10 ml)
and needles (18- 25 gauge) were from Becton Dickinson & Co. (Franklin Lakes, NJ). Petri
dishes were from Fisher Scientific (Nepean, ON). Beckman Optima LE-80 ultracentrifuge,
rotor (SW55Ti), ultracentrifuge tubes (5 ml), and table top refrigerated centrifuge were from
Beckman (Palao Alto, CA). Transwell filter plates (1.12cm2 insert diameter) and 6-well cell
culture plates were purchased from Corning Inc. (Corning, NY).
22
Mini-gel electrophoresis/transfer systems and power supplies were from Bio-Rad. The
orbital shaker was from VWR (Mississauga, ON). The recording film was from Kodak (Cedex,
France). The gel dryer was from Savant Instruments (Holbrook, N.Y.).
2.4. Animal Surgery
The animal surgery was performed essentially as described by Haidari et al. 2002
(Haidari et al., 2002). Briefly, male CD36-/- or wild type mice were housed separately with free
access to food and water. The mice were fasted for 8 hours prior to surgery. The animals were
anesthetized with a continuous flow mixture of isofluorane, nitrous oxide, and oxygen. An
incision was made in the animal’s abdomen and the jejunum (10 cm of the proximal end of the
small intestine) was excised and placed in a Petri dish of ice cold 1xPBS to be used for primary
enterocyte collection. The intestine was rinsed several times with ice cold 1xPBS using an 18-
gauge needle and 5 ml syringe to clear intestinal contents. The intestine was cut longitudinally
and subsequently into 2 cm fragments.
2.5. Primary Enterocyte Isolation
The intestine fragments were immersed in 5 ml Matrisperse Cell Recovery Solution for 1
h at 4°C. The fragments were then washed with 1xPBS with agitation on an orbital shaker for 5
min at 4°C. On the shaker, 5 ml of 1xPBS was added to the fragments, which were then gently
taped with a sterile dissection instrument. At this time, the villi begin dissociating from the
intestinal fragments. The PBS with suspended villi was collected and the villi collection was
repeated 5 more times or until no more villi dissociated from the fragments. The villi were
pelleted with a 3 min centrifugation at 1000 RPM at 4°C. The supernatant was removed and the
pelleted villi were weighed to estimate the number of cells. The pelleted villi were then washed
with 1x PBS followed by another centrifugation at 1000 RPM at 4°C.
23
2.6. Primary Enterocyte Cell Culture
Primary enterocytes isolated from CD36-/- or wild type mice were resuspended in a
volume of media in order to obtain a culture with 1 x 106 cells/ ml. To achieve this cell density,
2.5 ml of media was added per 0.1g of cells. One ml of cells plated per well of a 6-well culture
dish. Primary cells underwent lipid micelle stimulation and lipid labeling as described below.
2.7. Caco-2 Cell Culture
Caco-2 cells (American Type Culture Collection, Manassas, VA) were grown in 75 cm2
tissue culture flasks at 37oC in air and 5% CO2 in Eagle’s Minimal Essential Medium (EMEM,
GIBCO, USA), supplemented with 20% fetal bovine serum (FBS, GIBCO, USA), 100IU/ml
penicillin. The culture medium was changed every other day. For the subculture, the medium
was removed and the cells were detached from the culture dish with 0.25% trypsin diluted in
phosphate-buffer saline (PBS) containing 0.2g/L EDTA. Culture medium with FBS was added
to stop trypsinization. For all experiments, Caco-2 cells were plated at a density of 5x104 cells
in 400μl onto the apical compartment of Transwell filter plates. The basolateral compartment
was supplemented with 1 ml of culture media. The media was changed every two days and
transepithelial resistance (TER) was monitored using an Ohmmeter (homemade by the Grinstein
lab). After 21 days, the Caco-2 cells had formed a confluent monolayer (TER was >600Ω) and
were fully differentiated.
2.8. LB broth and agar plate preparation
LB broth was made by dissolving 25g of LB broth MILLER in 1L Millipore-purified water and
autoclaving. LB agar plates were made by dissolving 37g of LB agar MILLER in 1L of purified
water and autoclaved. When the agar had cooled down to approximately 50 ºC, ampicillin was
added to a final concentration of 50µg/ml, and poured into Petri dishes. Once the agar had
polymerized, the plates were stored at 4 ºC.
24
2.9. Transformation and plasmid DNA amplification
shRNA plasmids obtained were amplified by transforming into DH5α competent E. coli
cells by heat- shock method. Approximately 2 ng of plasmid DNA was combined with 50µL of
freshly thawed DH5α competent cells. The mixture was incubated on ice for 1 hr, placed in a
37ºC water bath for 1 min, and returned on ice for another 5 min. The transformed culture was
plated onto LB-ampicillin agar plates (50µg/ml ampicillin) and incubated overnight at 37ºC.
Several colonies were picked off the plates using sterile pipettes and were used to inoculate
separate tubes of LB broth containing 50µg/ml ampicillin. The tubes were incubated overnight
at 37ºC with shaking at250rpm.
Plasmid DNA was isolated using a QIAprep Miniprep kit according to the
manufacturer’s protocol. Plasmids were eluted using purfied, autoclaved water. Their
concentrations were determined by Nanodrop. The purity of the plasmid DNA was assessed by
the ratio of optical density units at A260/A280, which was between 1.8 and 2.
2.10. shRNA Transfection in Caco-2 Cells
Transfection of CD36- or FABP1-specific shRNA plasmid, or their respective
transfection control shRNA plasmid was performed essentially as described by the
manufacturer. Briefly, one day before transfection, Caco-2 cells were plated at a density of
2x104 cells in 12-well culture plates. For each well, 4.0 μl LipofectamineTM 2000 was diluted in
100 μl EMEM without serum and incubated at room temperature for 5 min. Two μg of either
control shRNA plasmid or CD36- or FABP1-targetting shRNA plasmid DNA was diluted in
100 μl EMEM without serum and mixed gently. The diluted LipofectamineTM 2000 and DNA
mixtures were mixed together and incubated for 45 min at room temperature. The complexed
DNA and LipofectamineTM 2000 solutions were added to each well and mixed by rocking the
plate back and forth. The cells were returned to the 37oC incubator and after 6 hours, the media
25
was replaced with EMEM supplemented with 20% FBS. Three replicate transfections for each
of the four gene-specific and the scrambled control shRNA plasmids were performed.
2.11. Selection of Stably Transfected Caco-2 Cell Clones
CD36-specific shRNA plasmid and its scrambled control shRNA plasmid conferred
puromycin resistance, while the FABP1-specific shRNA plasmid and its scrambled control
shRNA plasmid conferred G418 resistance. Dose response curve for G418 and puromycin
selection were generated and concentrations of 800 μg/ml G418 and 4 μg/ml puromycin were
found to be the effective concentrations for selection in Caco-2 cells. Forty-eight hours post-
transfection, Caco-2 cells were subjected to selection using the appropriate antibiotic. The
antibiotic containing media was changed every two days for up to 21 days. When a stably-
transfected colony was observed, it was aspirated from the plate and re-plated into a 24-well
plate to generate separate populations of stably-transfected cells, or a monoclonal population.
These clones were grown for one week and then re-plated onto 6 well plates and grown until
enough cells were available for generating a frozen stock and for isolating total RNA.
2.12. Total RNA Extraction
Total RNA was extracted from Caco-2 cells. Caco-2 cells grown on Transwell filters
were washed with 1xPBS and the cells were disrupted directly in the Transwell filter plates by
adding Buffer RLT. RNA was extracted using RNeasy Plus reagent (Qiagen, ON, Canada). All
subsequent steps were performed as recommended by the manufacturer. RNA was quantified
using a NanoDrop Spectrophotometer ND-1000 (Nanodrop Technologies, Wilmington, USA).
RNA integrity was examined by running a 2% agarose gel and by measuring the ratio of the
optical density at 260 nm to that at 280 nm. RNA was considered to be of good quality if the
28S:18S RNA was 2:1, and if the ratio of optical density at 260 nm to that at 280 nm exceeded
1.8.
26
2.13. Quantitative Real-Time Polymerase Chain Reaction
Quantitative real-time polymerase chain reaction (qRT-PCR) analysis was used to
quantify the expression of specific genes (Table 1). cDNA was synthesized using Taq Man
Reverse Transcription Reagent cDNA Synthesis Kit (Applied Biosystems, CA, USA). Reverse
transcription was performed using 200 ng of total RNA and random hexamer primers according
to the manufacturer’s instructions. The qRT-PCR conditions were 95°C for 10 min, 40 cycles at
95°C for 15s, 60-62oC (primer specific, optimal annealing temperature) for 1 min. Melting
curve analysis was performed by increasing the temperature (1oC/s) from 65oC to 95oC, with
continuous fluorescence acquisition. Primers for the target genes, ACC1, DGAT2, ApoAIV,
SCD1, and FABP2 were designed with Primer 3.0 (Rozen and Skaletsky, 2000), using publicly
available sequences. PCR conditions for all primers were optimized and specificities of all
reactions were verified by melting curves and electrophoresis on 1% agarose gels. Primers for
target genes and 3 reference genes are shown in Table 1. Threshold cycle (Ct) values were
obtained in triplicate for each sample using the Power SYBR Green master mix according to the
manufacturer’s protocol. The mean of three experiments in triplicate was used for the statistical
analyses. PCR efficiencies for each set of primers were calculated for each sample using serial
dilution curve method. The differences in expression between control and treated group means
were assessed for statistical significance by pair-wise fixed reallocation randomization test,
using the relative expression software tool (REST)(Pfaffl et al., 2002). An index (part of the
REST program) was generated with three reference genes selected: actin and 18S and GAPDH.
These genes were not differentially expressed between samples (assessed by qRT-PCR) and the
index generated with their values was used to normalize the data. Probability less than 0.05 was
considered statistically significant.
27
Table 1. RT-PCR primer sequences
Gene Forward Primer Reverse primer Amplicon
size (bp)
GAPDH 5’- AGGTCGGTGTGAACGGATTTG -3’ 5’- TGTAGACCATGTAGTTGAGGTCA -3’ 123
18S 5’-TAAGTCCCTGCCCTTTGTACACA -3’ 5’-GATCCGAGGGCCTCACTAAAC-3’ 71
Actin 5’-CCTTTTCCAGCCTTCCTTC -3’ 5’-TACTCCTGCTTGCTGATCC -3’ 300
Acetyl CoA Carboxylase 1 (ACC1)
5’- GAATCCTCATTGGCTATAATGAG-3’ 5’- CAATGGCCCGGCACGT-3’ 61
Apolipoprotein AIV (ApoAIV)
5’- CAGCAAGCAGCTCAGGATGTT-3’ 5’- TCCACGGCCTCCTTGGCATT-3’ 87
FA binding protein 2 (I-FABP)
5’- CTGATGCAGCCTGTCCACTA-3’ 5’- TTTTGAGCCCTGGAAAATTG-3’ 143
Diacylglycerol acyltransferase 2 (DGAT2)
5’-CAGCTACAGGTCATCTCAGTGC-3’ 5’-CAAACACCAGCCAAGTGAAGTA-3’ 139
SCD1 5’- ATGAATCCATTGGAGAAGCG-3’ 5’-AAAAATCCCACCCAATCACA-3’ 117
CD36 Primers obtained with shRNA plasmids from
SABiosciencesTM
Sequences were not disclosed, Cat #
PPH01356A 121
FABP1 Primers obtained with shRNA plasmids from
SABiosciencesTM
Sequences were not disclosed, Cat #
PPH01718A 88
2.14. SDS-PAGE and Western Blotting
For western blotting analysis of Caco-2 cells lysate, Transwell filters were washed with
1xPBS and cut out of the plastic insert. The filter was immersed in 500 μl solubilizing buffer
(150 mM NaCl, 1.0 mM Tris, 1.0 mM EDTA, 1.0 mM EGTA, 0.01% Triton X-100, 0.01% NP-
40, 100 mM NaF, 10 mM Na4P2O7, 2 mM sodium orthovanadate, 2 mM PMSF, pH 7.4) and
vortexed for 2 min. Lysates were centrifuged at 13,000 g for 10 min to remove debris and the
protein concentrations in the resulting supernatants were determined as described by Bradford
(Bradford, 1976). Aliquots of lysate containing equal amounts of protein were used for SDS-
PAGE. 6-12% SDS-PAGE resolving gels were cast and allowed to polymerize for 30 min. The
resolving gels were overlaid with 3-5% stacking gels and either 10 or 15-well combs were
inserted. The cell lysates from Caco-2 cells were prepared for PAGE by diluting in 4X sample
buffer (8% SDS, 0.25 M Tris, 40% glycerol) containing β-mercaptoethanol. The sample
mixture was boiled for 5 min at 100°C to fully denature the proteins in the sample. The
28
denatured samples were spun at 13,000 RPM in a tabletop centrifuge to pellet any remaining
cell membranes and loaded into the wells cast in the stacking gel, along with a protein molecular
weight marker. The Bio-Rad mini-gel electrophoresis system was filled with 1x running buffer
(25 mM Tris, 192 mM glycine, 0.1% (w/v) SDS, pH 8.5) and the gels were subjected to
electrophoresis at 100V for a period of approximately 2 hours. For apoB and FAS SDS-PAGE,
gels were subjected to electrophoresis overnight at 35V. The gels were then transferred to
methanol activated PVDF membranes in the Bio-Rad mini-gel protein transfer apparatus filled
with 1x transfer buffer (pH 7.5) and run at 225 mA for 2 hours at 4°C. When the transfer was
complete, the PVDF membranes were soaked in methanol for 15 seconds and left to air-dry.
The dried membranes were soaked in 5% blocking-grade non-fat milk in 1x PBST (pH 7.5) for
at least 1 hour on an orbital shaker at RT. The membranes were then incubated in a primary
antibody dilution (1:1000 in 1% BSA, 1xPBST) overnight at 4°C. The membranes were then
washed with 1xPBS 3 times for 5 minutes each, followed by incubation with a secondary
antibody dilution (for anti-mouse and ant-rabbit secondary antibodies the dilution was 1:10,000
in 1% BSA, 1xPBST, and for anti-goat secondary antibody the dilution was 1:80,000 in 1%
BSA, 1xPBST) for 1 hour at RT. The membranes were then washed with 1xPBS 3 times for 10
minutes each, followed by incubation in ECL reagent containing HRP substrate to yield
chemiluminescence. The membranes were exposed to recording film and then developed.
ImageJ software (available at rsb.info.nih.gov/ij/) was used for quantification of intensity of
each band. The graphs show the densitometry (mean ± SE). In the case of western blotting, a
typical blot was shown. Data comparisons were analyzed for statistical significance using a
student’s t-test. Differences with probability values lower than 0.05 were considered
significant.
29
2.15. Preparation of lipid micelles
Stock solutions of cholesterol (0.5 mg/ml) and oleic acid (100 mM) were prepared in
chloroform and ethanol, respectively. L-α-phosphatidylcholine, oleic acid, and cholesterol were
added to a large Pyrex tube to obtain the final concentrations of 0.5 mg/ml, 1.6 mM, and 0.2
mg/ml, respectively. With constant rotation, the tube was held nearly horizontal and lipids were
evaporated with a mild stream of Argon gas until a film was deposited on the Pyrex tube. The
Pyrex tube was returned upright and maintained under a constant stream of Argon gas for 5 min
to continue drying the lipids. During this time, 5.37g taurocholate was dissolved in 0.5 ml
media to obtain a final concentration of 1.0 mM. The dried lipids were resuspended in 0.5 ml
taurocholate-containing media and left to hydrate for 1 min. The solution was dissolved by
sonication for 1 min and dissolved in 9.5 ml of EMEM supplemented with 20% FBS to obtain a
total of 10 ml of 1.6 mM oleate-containing lipid micellar media. The micellar solution was
filtered through a 0.45 μm cellulose acetate filter before adding to cells. For lipid labeling
experiments, 125 μCi [3H]-oleate (5 μCi of [3H]-oleate per well) was added to the Pyrex tube
with the other lipids, followed by drying and resuspension as described above.
2.16. Lipid Micelle Stimulation of Caco-2 Cells
Differentiated Caco-2 cells were grown for 21 days and the basolateral media was
replaced with serum-free EMEM immediately before lipid micelle stimulation. The apical
compartment of the cells was washed with sterile 1xPBS that had been warmed to 37°C in a
water bath. The apical compartment was supplemented with 400 μl of 1.6 mM oleate-
containing lipid micellar media or with 400 μl EMEM supplemented with 20% FBS. The cells
were incubated for 12 hours at 37oC in air and 5% CO2.
30
2.17. Lipid Extraction and Thin Layer Chromatography
Following the 12 hour lipid micelle stimulation, the basolateral media was transferred to
2 ml tubes. A volume of 600 μl of organic solvent hexane:isopropanol (3:2, v/v) was added to 1
ml of basolateral media and vortexed for 1 min. The media sat on ice for 30min to allow
separation of organic and aqueous phases. During this time, the apical and basolateral
membrane of cells were washed with cold 1X PBS and 600 μl of hexane:isopropanol was added
directly to the apical membrane of washed cells. After 30 min, the upper layer of organic
solvent was carefully removed from the basolateral media by Pasteur pipette and transferred to a
1.5 ml tube and the lipid extraction was repeated. The organic solvent was removed from the
apical membrane of the cells and transferred to a 1.5 ml tube and the lipid extraction was
repeated. The extracted lipids were dried in a dessicator with vacuum connection (NALGENE®
Labware, Thermo Fischer Scientific Inc., NY) for 12 hours and resuspended by adding 30 μl ice
cold hexane and vortexing for 1 min. Solubilized lipids were spotted onto Silica plates in 2 μl
volumes for a total of 10 μl for each sample. Each sample spot was overlayed with 2 μl of
standards cholesteryl oleate, cholesterol (50 mg/ml), TG (1:20 dilution of olive oil), and oleate
(10mM). Development was with petroleum ether: diethyl ether: glacial acetic acid (80:20:1) in
a glass chamber for 90 minutes. The plates were air-dried at room temperature in and the lipid
standards were visualized by iodine vapour. The lipid spots were cut out from the plates and
placed into scintiliation vials with 4 ml Ready Safe® liquid scintillation cocktail (Beckman
Coulter, USA). Scintillation vials were vortexed for 1 min and allowed to sit overnight before
counting in scintillation spectrometer.
2.18. Metabolic Pulse Labeling and Immunoprecipitation of ApoB
To assess newly synthesized apoB, differentiated Caco-2 cells were incubated in 20%-
FBS containing methionine and cysteine-free DMEM for 1 hr (pre-pulse) before they were pulse
31
labeled with 50 μCi [35S]-methionine. The pulse media was removed from the apical
compartment and replaced with chase media (20% FBS containing EMEM). Basolateral media
was collected after a 3 hour chase period and subjected to immunoprecipitation. Anti-apoB-
antibody (5 μl/ per ml of media) was added to the samples and incubated for 18 hours at 4°C on
an orbital shaker. The following day, 50 μl of zysorbin was added to each sample and incubated
at room temperature for 1 hour on an orbital shaker. The samples were subjected to
centrifugation at 13,000 RPM for 10 min to pellet antibody-bound proteins. The supernatant
was removed and saved for a subsequent immunoprecipitation of a control secretory protein
used for normalization. The pellet was washed 3x10 min each with 1xPBS supplemented with
PI, and finally resuspended in 50 μl 1x sample buffer containing β-mercaptoethanol. The
sample was then centrifuged to pellet zysorbin. The supernatant containing labeled apoB was
resolved by SDS-PAGE as described above, and the SDS-PA gels were dried onto filter paper
using a gel dryer for approximately 2.5 hours. Protein bands were visualized by exposing the
dried gels to a phosphoimager screen and quantified using ImageJ, as described above.
2.19. Chylomicron Density Ultracentrifugation
Isolation of secreted lipoproteins of different sizes was accomplished by using a method
to isolate large chylomicrons, small chylomicrons, and VLDL from cell culture media by
sequential ultracentrifugation (Haidari et al., 2002). The basolateral media from experiments
where Caco-2 cells were stimulated with lipid micelles was adjusted to a density of 1.1 g/ml
using KBr. The adjusted media was then overlaid with 3 ml of 1.006 g/ml KBr solution
containing PI in a 4 ml ultracentrifuge tube. To obtain large CM (Sf > 400), samples were
ultracentrifuged for 33 min at 40,000 rpm 10°C, and the top 333 μl was collected. The samples
in the ultracentrifuge were then overlaid with 333 μl of 1.006 g/ml KBr solution. The gradient
was spun again at 40,000 rpm at 10°C for 2 hours, and the top 333 μl containing small CM (Sf =
32
60–400) was collected. The gradient was then spun at 40,000 RPM for 17 hours at 10°C, and
then top 333 μl, containing VLDL (Sf = 20–60) was collected.
2.20. Triglyceride, cholesterol, and liver enzyme assays
To measure triglyceride mass in different lipoprotein fractions, 10 μl of each fraction
was used in a colourimetric assay from Roche according to the manufacturer’s protocol. For
mice studies, plasma TG, cholesterol, and liver ezymes, aspartate aminotransferase (AST) and
alanine aminotransferase (ALT) were measured by colourimetric assay using the VITROS
250/350/950 and 5,1 FS Chemistry Systems and the VITROS 5600 Integrated System in the
Clinical Biochemistry lab at The Hospital for Sick Children.
2.21. Cerulenin treament
Cerulenin, a fatty acid analog, was used to inhibit fatty acid synthesis in Caco-2 cells. A
dose response curves was generated to determine the optimal concentration of cerulenin to
inhibit fatty acid synthesis. Cerulenin was added to the apical compartment of differentiated
Caco-2 cells to obtain a final concentration of 0, 5, 10, or 15 µg/mL. The apical media was
EMEM supplemented with 20% FBS, while the basolateral compartment media was replaced
with serum-free EMEM. Five µCi of 3H-acetate was added to label newly synthesized fatty
acids. After 12 hours, cellular lipids were resolved by TLC and counted, as described above.
The optimal concentration was found to be 10 µg/mL and further experiments were conducted
as above using 10 µg/mL cerulenin.
2.22. Statistical Analysis
Data were entered into, and all graphs were created using GraphPad Prism Software
(Version 4, April 2003). Data were analyzed by two-way ANOVA or by student’s T-test where
appropriate. A p-value les than 0.05 was considered significant.
33
CHAPTER 3
3. RESULTS
3.1 . shRNA Mediated Knockdown of CD36 and FABP1 in Caco-2 Cells
After selection of Caco-2 cell clones that stably express control, or CD36- or FABP1-
specific shRNA plasmids, the knockdown of CD36 or FABP1 gene products were verified by
real-time quantitative polymerase chain reaction. Messenger RNA expression in each clone was
normalized to 3 housekeeping (internal control) genes and expressed as a percentage of CD36
mRNA in the control puromycin transfected clones. Clones CD36-1 and CD36-3 had 3.66 fold
(p=0.001) and 3.75 fold (p=0.04) level of CD36 knockdown, respectively (Fig. 3A). To assess
protein knockdown in these clones, cell lysates were resolved by SDS-PAGE and
immunoblotting for CD36 was performed (Fig. 3B). The trend in RNA knockdown correlated
with the level of protein knockdown (except for shRNA-CD36-4) (Fig. 3C). Clones CD36-1
and CD36-3 express CD36 protein at 40% (p=0.02) and 60% (p=0.03), respectively, of control
cell protein levels (Fig. 3C). One clone expressing FABP1-specific shRNA was quantified (Fig.
3D). This clone had an 18.8 fold decrease in expression of FABP1 at the mRNA level (Fig.
3D). Immunoblotting for FABP1 protein knockdown was not performed because we could not
obtain an antibody for FABP1 that could detect FABP1 by immunoblot.
34
3.2. Depletion of CD36 or FABP1 does not impair FA upake in Caco-2 cells
To examine whether CD36-or FABP1-deficiency results in impaired lipid uptake, Caco-
2 cells were treated with 3H-oleate containing micelles, and uptake was measured after 1, 2,
4,and 12 hours (Fig. 4). There were no differences in 3H cellular counts between CD36- or
FABP1- deficient Caco-2 cells and their respective controls at any time point (Fig. 4).
Therefore, differences in fatty acid uptake do not contribute to the increase in TG secretion from
CD36- or FABP1- deficient Caco-2 cells (Fig. 4).
Figure 3. shRNA mediated knockdown of CD36 or FABP1 in Caco-2 cells. Cells were transfected with SureSilencingTM shRNA plasmids containing short-hairpin sequences with scrambled shRNA or shRNA sequences targeting CD36 or FABP1 (SuperArray, Bioscience Corporation) and selected using Puromycin or Neomycin, respectively. A. Relative CD36 mRNA expression of clones CD36-1, CD36-2, and CD36-4, compared to the control clone. B. CD36 Protein expression in clones CD36-1, CD36-2, and CD36-4. C. Quantification of Immunoblot in Panel B. D. Relative FABP1 mRNA levels in FABP1 knockdown clone, FABP1-4C, compared to its scrambled control. *p<0.05, **p<0.01, n=3, **p,0.01.
Control CD36-1 CD36-3 CD36-40
50
100
150
** **
CD36 mRNA Expression
CD
36 m
RN
A E
xpre
ssio
n(%
of C
ontr
ol)
Control CD36-1 CD36-3 CD36-40
50
100
150
** **
CD36 Protein Expression
Nor
mal
ized
to B
eta-
Act
in(%
of C
ontr
ol)
Control FABP1-40
25
50
75
100
125
**
FABP1 mRNA Expression
FAB
P1 m
RN
A E
xpre
ssio
n (%
of C
ontr
ol)
A. B.
C.D.
CD36 Immunoblot
Puromycin CD36-1 CD36-3 CD36-4(control)
CD36 ~60kDa
Beta-actin43kDa
Puromycin CD36-1 CD36-3 CD36-4(control)
CD36 ~60kDa
Beta-actin43kDa
CD36 mRNA Expression
CD36 Protein Expression
CD36 Immunoblot
FABP1 mRNA Expression
Control CD36-1 CD36-3 CD36-40
50
100
150
** **
CD36 mRNA Expression
CD
36 m
RN
A E
xpre
ssio
n(%
of C
ontr
ol)
Control CD36-1 CD36-3 CD36-40
50
100
150
** **
CD36 Protein Expression
Nor
mal
ized
to B
eta-
Act
in(%
of C
ontr
ol)
Control FABP1-40
25
50
75
100
125
**
FABP1 mRNA Expression
FAB
P1 m
RN
A E
xpre
ssio
n (%
of C
ontr
ol)
A. B.
C.D.
CD36 Immunoblot
Puromycin CD36-1 CD36-3 CD36-4(control)
CD36 ~60kDa
Beta-actin43kDa
Puromycin CD36-1 CD36-3 CD36-4(control)
CD36 ~60kDa
Beta-actin43kDa
Control CD36-1 CD36-3 CD36-40
50
100
150
** **
CD36 mRNA Expression
CD
36 m
RN
A E
xpre
ssio
n(%
of C
ontr
ol)
Control CD36-1 CD36-3 CD36-40
50
100
150
** **
CD36 Protein Expression
Nor
mal
ized
to B
eta-
Act
in(%
of C
ontr
ol)
Control FABP1-40
25
50
75
100
125
**
FABP1 mRNA Expression
FAB
P1 m
RN
A E
xpre
ssio
n (%
of C
ontr
ol)
A. B.
C.D.
CD36 Immunoblot
Puromycin CD36-1 CD36-3 CD36-4(control)
CD36 ~60kDa
Beta-actin43kDa
Puromycin CD36-1 CD36-3 CD36-4(control)
CD36 ~60kDa
Beta-actin43kDa
CD36 mRNA Expression
CD36 Protein Expression
CD36 Immunoblot
FABP1 mRNA Expression
35
3.3. CD36 or FABP1-Deficiency is Associated with Increased TG Secretion in
Caco-2 Cells
CD36- or FABP1-deficient Caco-2 cells were stimulated with [3H]-oleate-containing
lipid micelles to determine whether CD36- or FABP1-deficiency affects TG assimilation or
secretion in Caco-2 cells. Assimilation of neutral lipids, TG and CE, were not affected by
knockdown of CD36 or FABP1 in Caco-2 cells (Fig. 5A & B). Levels of FA and cholesterol
were not different between CD36- or FABP1-deficient Caco-2 cells and their respective controls
(Fig. 5A & B). Although intracellular [3H]-TG content was not different between control and
knockdown cell lines, TG secretion from CD36 knockdown Caco-2 cells was significantly
increased (Fig. 5C). A trend towards increased TG secretion was also observed in FABP1
knockdown cells compared to the control cells, although this was not significant (Fig. 5D).
Secretion of CE, FA, or cholesterol was not significantly altered by knockdown of CD36 (Fig.
5C). Although secretion of CE and FA were not altered, secretion of cholesterol from FABP1
Total Cellular Lipid
1 2 4 120
100000
200000
300000ControlCD36 KD
Hours
DPM
per
mg
prot
ein
Total Cellular Lipid
1 2 4 120
100000
200000
300000ControlFABP1 KD
Hours
DPM
per
mg
prot
ein
A. B.
Figure 4. Lipid Uptake in CD36- or FABP1-Deficient Caco-2 Cells. Cells were treated for 1, 2, 4, or 12 hours with 1.6 mM [3H]-oleate containing lipid micelles. Lipids were extracted from the CD36-deficient Caco-2 cells (A) or FABP1-deficient Caco02 cells (B) using hexane:isopropanol (3:2), dried, and resuspended in hexane. Lipids were resolved by thin layer chromatography using petroleum ether:diethyl ether:acetic acid (80:20:2) as solvent. The 3H-oleate incorporation into each lipid was measured by scintillation counting. n=3.
36
knockdown Caco-2 cells was significantly increased compared to control cells (Fig. 5D).
Cellular 3H Lipids
Control CD36 KD0
1000
2000
CETGFACH
50007500
10000100000300000500000
DPM
Nor
mal
ized
toPr
otei
n Co
ncen
tratio
nCellular 3H Lipids
Control FABP1 KD0
1000
2000
CETGFACH
4000
8000
200000400000600000
DPM
Nor
mal
ized
toPr
otei
n Co
ncen
tratio
n
Secreted 3H Lipids
Control CD36 KD0
40
80
TGFA
CE
CH
300600900
1000
2000 *
DPM
Nor
mal
ized
toPr
otei
n Co
ncen
tratio
n
Secreted 3H Lipids
Control FABP1 KD0
200400600800
1000
5000
9000CETGFACH
*DPM
Nor
mal
ized
toPr
otei
n Co
ncen
tratio
n
A. B.
C. D.
3.4. TG Content of Lipoproteins Secreted by CD36- or FABP1-Deficient
Caco-2 Cells
To examine which lipoprotein fraction contained the observed increase in TG and apoB,
density gradient ultracentrifugation of the basolateral media was performed. Without lipid
stimulation, the amount of TG present in the large CM and small CM lipoproteins secreted from
control and CD36-deficient cells was not significantly different (Fig. 6). Interestingly, the
amount of TG present in VLDL secreted from CD36 knockdown Caco-2 cells was significantly
Figure 5. Lipid assimilation and secretion in CD36 & FABP1 knockdown Caco-2 cells. Cellular 3H-lipids in CD36- (A) and FABP1- (B) deficient Caco-2 cells. 3H-lipids secreted from CD36- (C) and FABP1- (D) deficient Caco-2 cells. The apical compartment of control or CD36, or FABP1 knockdown Caco-2 cells were treated with 1.6 mM oleate-containing lipid micelles also containing 5 μCi 3H-oleate for 12 hours. Lipids were extracted from the cells (A & B) or the basolateral media (C & D) using hexane:isopropanol (3:2), dried, and resuspended in hexane. Lipids were resolved by thin layer chromatography using petroleum ether:diethyl ether:acetic acid (80:20:2) as solvent. The 3H-oleate incorporation into each lipid was measured by scintillation counting. n=9, *p=<0.05.
37
greater than in control cells (Fig. 6). When control cells were treated with lipid, most of the TG
was present in the CM fraction (Fig. 6). The same trend was observed in CD36 knockdown
Caco-2 cells treated with lipid micelles, however, the small CM fraction contained more TG
than in control cells (Fig. 6).
Lipoprotein Triacylglycerol
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5Lg Chylomicron
Sm Chylomicron
VLDL
Lipid - - + +
**
**
Control CD36 KD
mm
ol/L
Tria
cylg
lyce
rol
3.5. CD36 or FABP1-Deficiency in Caco-2 cells is Associated With an
Increase in ApoB Secretion
The basolateral media from experiments in which Caco-2 cells were treated with oleate-
containing lipid micelles was used for immunoblot analysis of apoB. As expected, lipid
stimulation increased apoB secretion in control cells and in CD36 knockdown Caco-2 cells (Fig.
7A & B). An increase in apoB secretion upon lipid stimulation was not observed in FABP1-
Figure 6. TG content of lipoproteins secreted from normal and CD36-deficient Caco-2 cells. The basolateral media from cells treated with (+) or without (-) 1.6 mM oleate-containing lipid micelles was subjected to density gradient utracentrifugation to isolate large chylomicron (LG CM), small chylomicron (SM CM), and VLDL. To determine the mas of TG in each fraction, TG was detected in 10 µL from each fraction by colourimetric assay. n=3. *p<0.05 compared to the control VLDL fraction without lipid treatment.
38
deficient Caco-2 cells (Fig. 7B). Under basal conditions, depletion of CD36 was associated with
an increase in apoB100 secretion (Fig. 7A & C).
Metabolic labeling of proteins was used to examine apoB secretion under basal
conditions. Caco-2 cells were pulse-labeled with [35S]-methionine for one hour, followed by a 3
hour chase period (Fig. 7E & F). Radiolabeled apoB100 and apoB48 could be detected in these
samples by immunoprecipitation of apoB using an anti-apoB antibody followed by resolution of
bound proteins by SDS-PAGE. Under basal conditions, CD36-deficient Caco-2 cells secreted
significantly more apoB100 and apoB48 compared to control cells (Fig. 7G). Compared to
control cells, FABP1-knockdown Caco-2 cells secreted significantly more apoB100, while
levels of secreted apoB48 remained the same (Fig. 7H).
39
Secreted ApoB100
0
50
100
150
*
Control CD36 KDLipid Micelles - + - +
Rel
ativ
e A
poB
100
Secr
etio
n (A
U)
Secreted ApoB100
0
50
100
Lipid Micelles - + - + Control FABP1 KD
Rel
ativ
e A
poB
100
Secr
etio
n (A
U)
Secreted ApoB
Control CD36 KD0.0
2.5
5.0
7.5ApoB 100ApoB 48
**
**
Labe
led
Apo
B N
orm
aliz
edto
TC
A C
ount
s (A
U)
Secreted ApoB
Control FABP1 KD012345678
ApoB100ApoB48
**
Labe
led
Apo
B N
orm
aliz
edto
TC
A C
ount
s (A
U)
C. D.
G. H.
A. ApoB Immunoblot B. ApoB Immunoblot
E. Labeled ApoB F. Labeled ApoB
Figure 7. Apolipoprotein B secretion from CD36 or FABP1 knockdown Caco-2 cells. Immunoblot of apoB100 secreted from CD36- (A) or FABP1- (B) deficient Caco-2 cells. Cells were treated with 1.6 mM oleate-containing lipid micelles for 12 hours, the basolateral media was resolved by SDS-PAGE. Quantification of immunoblots in A (C) & B (D). Metabolic labeling of apoB100 and apoB48 in CD36- (E) and FABP1- (F) deficient Caco-2 cells. Cells were labeled for 1 hr with [35S]-methionine and chased for 3 hours. ApoB was immunoprecipitated from the basolateral media and was resolved by SDS-PAGE. Quantification of phosphoimages in E (G) & F (H). n=6, *p<0.05, **p<0.01
Control FABP1 KD
Lipid Micelles - + - +Control CD36 KD
Lipid Micelles - + - +
Control FABP1 KD
ApoB 100
ApoB 48
Media
Control CD36 KD
ApoB 100
ApoB 48
Media
ApoB 100 ApoB100
40
3.6. Expression of Genes Involved in FA Synthesis
We hypothesized that depletion of either CD36 or FABP1 would impair lipoprotein
assembly and secretion in Caco-2 cells. Upregulation of FA synthesis could lead to increased
formation of neutral lipid and, in turn, may provide more substrate for apoB lipidation. We thus
sought to determine the mechanisms responsible these paradoxical results. The expression of
genes involved in FA synthesis, fatty acid synthase (FAS) and acetyl coA carboxylase1 (ACC1),
was examined.
Cells were treated for 12 hours with 1.6 mM oleate-containing lipid micelles, lysates
were resolved, and immunoblotting for FAS was performed. FAS levels were normalized to the
amount of E-Cadherin, a loading control protein (Beaslas et al., 2009). In CD36 control cells
(Fig. 8A), FAS expression increased with lipid micelle stimulation, while in FABP1 control
cells (Fig. 8B), levels of FAS remained the same with or without lipid stimulation. The
differences in response to lipid stimulation in the control cell lines and this may be due to
variation in mRNA levels in different control cell preparations (Fig. 8C and 8D). Under basal
conditions, an increase in FAS expression in both CD36 and FABP1 knockdown cells was
observed, compared to their respective controls (Fig. 8A & B). Levels of FAS did not change
significantly in either CD36 or FABP1 knockdown cells when treated with lipid micelles (Fig.
8A & B).
41
FAS
0.00
0.25
0.50
0.75
1.00
1.25
Control FABP1 KDLipid Micelle - + - +
**
**
**
FAS
Prot
ein
Expr
essi
on (A
U)
FAS
0.00
0.25
0.50
0.75
1.00
1.25
Lipid Micelle - + - + Control CD36 KD
*
*
FAS
Prot
ein
Nor
mal
ized
toE-
Cad
herin
(AU
)
C. D.
A. B.
Levels of ACC1 mRNA were measured in control and knockdown cells by real-time
PCR. Control cells that had been treated with lipid for 12 hours had a significantly reduced
level of ACC1 mRNA, compared to control cells that were not treated with lipid (Fig. 9A & B).
Surprisingly though, significantly less ACC1 mRNA was detected in both CD36 and FABP1
knockdown cells compared to their respective controls under basal conditions (Fig. 9A & B). A
trend of decreased ACC1 mRNA was observed in both knockdown cell lines compared to their
controls when treated with lipid, but this was statistically significant only in CD36 knockdown
cells (Fig. 9A & B).
Figure 8. FAS protein levels in CD36 and FABP1 knockdown Caco-2 cells treated with or without 1.6 mM oleate-containing lipid micelles for 12 hours. A. Representative immunoblot of FAS and E-cadherin in control and CD36 knockdown cells. B. Representative immunoblot of FAS and E-cadherin in FABP1 knockdown cells. C & D. Quantification of immunoblots in A & B. n=9, *p<0.05, **p<0.01.
Control CD36 KD
Lipid Micelles - + - +FAS 272 kDa
E-Cadherin
Control FABP1 KD
Lipid Micelles - + - +FAS 272 kDa
E-Cadherin
42
ACC1
Control CD36 KD CD36 KD -4
-3
-2
-1
0
**
**
**
+ lipid + lipid
Rel
ativ
e G
ene
Expr
essi
on(F
old
chan
ge c
ompa
red
to C
ontr
ol) ACC1
Control FABP1 KD FABP1 KD -3
-2
-1
0
** **
+ lipid + lipid
Rel
ativ
e G
ene
Expr
essi
on(F
old
chan
ge c
ompa
red
to C
ontr
ol)
A. B.
Since FAS levels were increased under basal conditions in both CD36- and FABP1-
deficient Caco-2 cells, we decided to investigate whether de novo lipogenesis (endogenous FA
synthesis) was increased, and if cerulenin (a FAS inhibitor) would inhibit FA synthesis in these
cells.
3.7. Cerulenin Decreases FA Synthesis in Caco-2 Cells
Differentiated Caco-2 cells were treated overnight with acetate and varying
concentrations of cerulenin. In both control and knockdown Caco-2 cells, doses of 5 or 10
μg/ml cerulenin reduced [3H]-acetate incorporation into FAs (Fig. 10). In CD36 control cells
(Fig. 10A), inhibition of [3H]-acetate incorporation into FAs was greatest at 5 μg/ml cerulenin,
and cerulenin decreased the [3H]-acetate incorporation into TG. Levels of CE and cholesterol
decreased as the concentration of cerulenin was increased (Fig. 10A). FA synthesis was
decreased in a dose-dependent manner in CD36 knockdown cells (Fig. 10B), however, it was
only decreased by approximately 30% at 5 μg/ml cerulenin as compared to a 60% decrease in
FA synthesis observed in control cells (Fig 10A). Similar to control cells, [3H]-acetate
incorporation into CE, cholesterol, and TG were decreased with cerulenin treatment in CD36
Figure 9. Real-time PCR measurement of ACC1 mRNA in control and CD36 or FABP1 knockdown Caco-2 cells. Differentiated Caco-2 cells were treated with or without 1.6 mM oleate-containing lipid micelles for 12 hours. Ct values for ACC1 were normalized to an internal control gene, 18S, and expressed as fold-change compared to their respective controls that were not treated with lipid. n=9, **p<0.01, compared the control without lipid treatment.
43
knockdown cells (Fig 10B). We observed a similar level of inhibition of FA synthesis in control
and CD36 knockdown Caco-2 cells, therefore a concentration of 10 μg/ml was used in further
experiments. Also, our laboratory had previously used a concentration of 10 μg/ml cerulenin in
primary enterocytes to successfully inhibit FA synthesis (Haidari et al., 2002).
Control
0 2 4 6 8 10 12 14 160
20
40
60
80
100CETGFACH
Cerulenin (μg/ml)
Labe
led
Lipi
d (%
of c
ontr
ol)
CD36 KD
0 2 4 6 8 10 12 14 160
25
50
75
100
125CETGFACH
Cerulenin (μg/ml)
Labe
led
Lipi
d (%
of c
ontr
ol)
Control
0 2 4 6 8 10 12 14 160
50
100
150CETGFACH
Cerulenin (μg/ml)
Labe
led
Lipi
d (%
of c
ontr
ol)
FABP1 KD
0 2 4 6 8 10 12 14 160
25
50
75
100
125CETGFACH
Cerulenin (μg/ml)
Labe
led
Lipi
d (%
of c
ontr
ol)
A. B.
C. D.
Cerulenin treatment also inhibited FA synthesis in FABP1 control cells, as measured by
[3H]-acetate incorporation into FA (Fig. 10C). However, the dose-dependent decrease in [3H]-
acetate incorporation into CE, TG or FA, was not observed in FABP1 control cells (Fig. 10C).
FA synthesis was also decreased by approximately (80%) in FABP1 knockdown Caco-2 cells
by 5 μg/ml cerulenin treatment (Fig. 10D). At 10 μg/ml cerulenin treatment, [3H]-acetate
Figure 10. Dose-dependent effect of different concentrations (0, 5, 10, 15 mg/ml) of cerulenin, a FA synthase inhibitor, on incorporation of [3H]-acetate into CE, TG, FA, and CH in control Caco-2 cells (A & C) and CD36 knockdown (B) or FABP1 knockdown (C) Caco-2 cells. Lipids were extracted from the cells using hexane: isopropanol (3:2), dried, and resuspended in hexane. Lipids were resolved by thin layer chromatography using petroleum ether: diethyl ether: acetic acid (80:20:1) as solvent. The [3H]-acetate incorporation into each lipid species was assessed by scintillation counting. [3H]-acetate incorporation into lipid species at different concentrations of cerulenin are expressed as a percent of the untreated condition. n=3.
44
incorporation into FA increased to approximately 60% of untreated cell levels (Fig. 10D). In
FABP1 knockdown Caco-2 cells, [3H]-acetate incorporation into cholesterol, TG and CE were
decreased in a dose-dependent manner (Fig. 10D). We next determined if an increase in de
novo lipogenesis may be responsible for the increase in TG and apoB secretion observed in
CD36 or FABP1 knockdown Caco-2 cells.
3.8. De novo Lipogenesis May Contribute to the Increase in TG or ApoB
Secretion Observed in CD36, but not FABP1-Deficient Caco-2 Cells
The incorporation of [3H]-acetate into CE, TG, and FA was examined in differentiated
control and knockdown Caco-2 cells after a 12 hour treatment with or without 1.6 mM oleate-
containing lipid micelles and with or without 10 μg/ml cerulenin. The incorportation of [3H]-
acetate into CE was increased following treatment with oleate-containing lipid micelles in
CD36- or FABP1-deficient Caco-2 cells and in their respective control cell lines (Fig. 11A &
B). This is expected, as the availability of FA substrate for esterification of cholesterol would
increase with oleate supplementation. Overall, cerulenin had no effect on the incorporation of
[3H]-acetate into CE except in FABP1 knockdown cells, where cerulenin treatment decreased
the incorporation of [3H]-acetate into CE when the cells were also treated with lipid micelles
(Fig. 11B).
The incorporation of [3H]-acetate into TG followed a similar trend in CD36 and FABP1
knockdown, and the FABP1 control cell lines. That is, with lipid micelle supplementation, [3H]-
acetate incorporation into TG increased significantly, although this was not seen in the CD36
knockdown control cells in the absence of cerulenin treatment (Fig. 11C & D). Cerulenin
treatment had no major effect on [3H]-acetate incorporation into TG in CD36 knockdown cells,
however, in their respective control cell line, cerulenin treatment decreased [3H]-acetate
incorporation into TG in the absence of lipid stimulation (Fig. 11C).
45
The objective of this experiment was to determine whether CD36 or FABP1 knockdown
cells have an increase in de novo lipogenesis compared to their respective controls. With no
lipid stimulation or cerulenin treatment, the level of de novo lipogenesis was not different
between control and CD36 knockdown cell lines (Fig. 11E, first and fifth bars). When the cells
were treated with lipid, there was a trend towards an increased level of FA synthesis in CD36
knockdown cells compared to control cells (Fig. 11E, second and sixth bars). When CD36
knockdown cells were treated with cerulenin, FA synthesis was significantly greater than in
control cells (Fig. 11E).
Surprisingly, the level of de novo lipogenesis was decreased in FABP1 knockdown cells
compared to their control (Fig. 11D, first and fifth bars). Cerulenin treatment decreased the
incorporation of [3H]-acetate into FA, an indication that FAS was inhibited (Fig. 10E & F), the
exception being in FABP1 knockdown cells, where cerulenin had no effect on the already
blunted incorporation of [3H]-acetate into FA (Fig. 11F).
There were no significant differences between the level of secreted [3H]-CE from CD36
or FABP1 knockdown cells and their respective control cell lines (Fig. 12A & B). Without
cerulenin treatment, secreted [3H]-TG levels from CD36 knockdown cells were increased
compared to controls, while cerulenin treatment reduced the level of secreted [3H]-TG to control
levels (Fig. 12C). There were no significant differences between secreted [3H]-TG from FABP1
knockdown cells compared to the control with or without cerulenin treatment (Fig. 12D). There
were no significant differences between secreted [3H]-FA from CD36 knockdown cells
compared to the control with or without cerulenin treatment (Fig. 12E). FABP1 knockdown
cells secreted significantly less [3H]-FA with cerulenin treatment (Fig. 12F).
46
Cellular 3H-CE
0
1000
2000
3000
4000
5000
6000
7000
Control CD36 KD
Cerulenin(10ug/ml) - - + + - - + +Lipid Micelles - + - + - + - +
***
*
DPM
Nor
mal
ized
toPr
otei
n Co
ncen
tratio
n
Cellular 3H-CE
0
5000
10000
15000
Cerulenin(10ug/ml) - - + + - - + +Lipid Micelles - + - + - + - +
Control FABP1 KD
DPM
Nor
mal
ized
toPr
otei
n Co
ncen
tratio
n
Cellular 3H-TG
0
1000
2000
3000
4000
5000
6000
7000
Cerulenin(10ug/ml) - - + + - - + +Lipid Micelles - + - + - + - +
Control CD36 KD
**
*
**
*
DPM
Nor
mal
ized
toPr
otei
n Co
ncen
tratio
n
Cellular 3H-TG
0
2500
5000
7500
10000
12500
Cerulenin(10ug/ml) - - + + - - + +Lipid Micelles - + - + - + - +
Control FABP1 KD
**
DPM
Nor
mal
ized
toPr
otei
n Co
ncen
tratio
n Cellular 3H-FA
0
5000
10000
15000
Cerulenin(10ug/ml) - - + + - - + +Lipid Micelles - + - + - + - +
Control CD36 KD
***
DPM
Nor
mal
ized
toPr
otei
n Co
ncen
tratio
n
Cellular 3H-FA
0
5000
10000
15000
20000
25000
Cerulenin(10ug/ml) - - + + - - + +Lipid Micelles - + - + - + - +
Control FABP1 KD
*
DPM
Nor
mal
ized
toPr
otei
n Co
ncen
tratio
n
A. B.
C. D.
E. F.
Overall, these results suggest that de novo lipogenesis may be contributing to the
increase in TG or apoB secretion in CD36 knockdown Caco-2 cells (Fig 10E & F). Cerulenin
treatment showed a trend towards decreased cellular FA in CD36 knockdown cells treated with
lipid (Fig. 11, bars six and eight). Finally, cerulenin did not have an effect on the secretion of
[3H]-acetate labeled lipids from Caco-2 cells (Fig. 12).
Figure 11. Cellular [3H]-lipids extracted from CD36 or FABP1 knockdown Caco-2 cells. A & B. Cellular [3H]-esterified CH. C & D. Cellular [3H]-TG. E & F. Cellular [3H]-FA. The apical compartment of control of CD36 or FABP1 knockdown Caco-2 cells were treated with (+) or without (-) cerulenin (10μg/ml), and with (+) or without (-) 1.6 mM oleate-containing lipid micelles and 10μCi [3H]-acetate for 12 hours. Lipids were extracted from the cells using hexane: isopropanol (3:2), dried, and resuspended in hexane. Lipids were resolved by thin layer chromatography using petroleum ether: diethyl ether: acetic acid (80:20:2) as solvent. The [3H]-acetate incorporation into each lipid species was measured by scintillation counting. Data is plotted as mean DPM per mg protein + SEM. Comparisons are indicated by lines and statistical significance was determined by two-way ANOVA. n=3, *p<0.05, **p=<0.001.
47
Secreted 3H-CE
05
1015202530354045
Cerulenin(10ug/ml) - - + + - - + +Lipid Micelles - + - + - + - +
Control CD36 KD
DPM
/ m
l med
ia
Secreted 3H-CE
0
10
20
30
40
Cerulenin(10ug/ml) - - + + - - + +Lipid Micelles - + - + - + - +
Control FABP1 KD
DPM
/ m
l med
ia
Secreted 3H-TG
0
10
20
30
40
Cerulenin(10ug/ml) - - + + - - + +Lipid Micelles - + - + - + - +
Control CD36 KD
**
DPM
/ m
l med
ia
Secreted 3H-TG
0
10
20
30
40
50
Cerulenin(10ug/ml) - - + + - - + +Lipid Micelles - + - + - + - +
Control FABP1 KDD
PM /
ml m
edia
Secreted 3H-FA
0
25
50
75
100
125
Cerulenin(10ug/ml) - - + + - - + +Lipid Micelles - + - + - + - +
Control CD36 KD
DPM
/ m
l med
ia
Secreted 3H-FA
0
50
100
150
200
250
Cerulenin(10ug/ml) - - + + - - + +Lipid Micelles - + - + - + - +
Control FABP1 KD
*
DPM
/ m
l med
ia
A. B.
C. D.
E. F.
3.9. CD36-/- Mice Have Impaired cholesterol and TG Production
We observed that de novo lipogensis may be contributing to the increase in TG and apoB
secretion in CD36-deficient Caco-2 cells so we wanted to test the hypothesis that an impairment
Figure 12. Secreted [3H]-lipids extracted from the medium CD36 or FABP1 knockdown Caco-2 cells. A & B. Cellular [3H]-esterified CH. C & D. Cellular [3H]-TG. E & F. Cellular [3H]-FA. The apical compartment of control of CD36 or FABP1 knockdown Caco-2 cells were treated with (+) or without (-) Cerulenin (10μg/ml) to inhibit FA synthesis, and with (+) or without (-) 1.0mM oleate-containing lipid micelles and 10μCi [3H]-acetate for 12 hours. Lipids were extracted from the cells using hexane: isopropanol (3:2), dried, and resuspended in hexane. Lipids were resolved by thin layer chromatography using petroleum ether: diethyl ether: acetic acid (80:20:2) as solvent. The [3H]-acetate incorporation into
48
in de novo lipogenesis contributes to the dyslipidemia observed in the CD36-/- mouse. CD36-/-
mice were anesthetized, the intestine was removed and primary enterocytes were isolated using
an established method (Haidari et al., 2002). The cells were radiolabeled with either [3H]-
acetate, or [3H]-oleate. The cellular and secreted lipids were then analyzed. Overall, we
observed that cholesterol synthesis (Fig. 13A) and TG synthesis (Fig. 13B) were significantly
decreased in CD36-/- mouse enterocytes, compared to their WT littermates.
Lipogenesis in CD36 -/- Mouse Enterocytes
CE TG FA CH0
25
50
75
100
125WTCD36 -/-
***
Lipids
3 H- A
ceta
te L
abel
ed L
ipid
(% o
f con
trol
)
Lipogenesis in CD36 -/- Mouse Enterocytes
CE TG FA CH0
25
50
75
100
125WTCD36 -/
**
Lipids
3 H- O
leat
e La
bele
d Li
pid
(% o
f con
trol
)
CE TG FA CH0
100
200WTCD36 -/-
***
Secreted Lipids
3 H- O
leat
e La
bele
d Li
pid
(% o
f con
trol
) WT CD36 -/-0
60
120
180
AST
(U/L
)
WT CD36 -/-0
10
20
30
40
ALT
(U/L
)
WT CD36 -/-0.0
1.5
3.0
4.5
Cho
lest
erol
(mm
ol/L
)
WT CD36 -/-0.0
0.5
1.0
1.5
Trig
lyce
ride
(mm
ol/L
)
A. B.
C. D. E.
F. G.
There were no significant differences in
Figure 13. Lipogenesis in enterocytes isolated from WT and CD36-/- mice. Enterocytes were isolated from WT and CD36 -/- animal and resuspended in serum-free DMEM. Cells were pulse labeled with A. 5μCi [3H]-acetate or B. 1.0mM oleate-BSA + 5μCi [3H]-oleate. After 3 hour labeling period, lipids were extracted from the cells (A & B) or the medium (C) using hexane: isopropanol (3:2), dried, and resuspended in hexane. Lipids were resolved by thin layer chromatography using petroleum ether: diethyl ether: acetic acid (80:20:2) as solvent. The 3H-acetate or 3H-oleate incorporation into each lipid was assessed by measuring radioactivity by scintillation counting. Data is expressed as a percent of WT 3H-acetate or 3H-oleate incorporation. (D) Plasma ALT (U/l) (E) Plasma ALT (U/l) (F) Plasma CH (mmol/l) (G) Plasma triglyceride (mmol/l). n=3, *p<0.05, **p<0.01, ***p<0.001 are statistically significant compared to WT based on a student’s t-test.
49
plasma levels of liver enzymes AST and ALT There were no significant differences in plasma
levels of liver enzymes AST and ALT (Fig. 13D & E) or in levels of TG or cholesterol (Fig. 13F
& G) between CD36-/- and WT mice. We also wanted to confirm our findings from Caco-2
cells that CD36 deficiency is associated with an increase in TG secretion. In contrast to the
increase secretion observed from CD36-deficient Caco-2 cells, we found that TG secretion from
CD36-/- enterocytes was decreased compared to WT enterocytes (Fig. 13C).
It is possible that in CD36-deficient Caco-2 cells, an increase in FA substrate would feed
into TG synthesis pathways. In FABP1-deficient Caco-2 cells, there was no observed increase
in FA synthesis, as measured by [3H]-acetate incorporation into FA, although FAS levels were
increased. Thus, protein levels of enzyme involved in TG synthesis were measured next.
3.10. Cellular levels of proteins involved in TG synthesis in CD36- and FABP1-
deficient Caco-2 cells
To determine if there was an upregulation of proteins involved in TG synthesis, we
measured levels of fatty acid transport protein 4 (FATP4) and diacylglycerol acyltransferase
(DGAT). FATP4 is a membrane protein found in the plasma membranes and ER membranes of
the small intestine and has both FA transport and acyl-coA synthetase activities (Hall et al.,
2005), allowing it to convert long-chain FAs to fatty acyl-CoA. Lipid treatment increased levels
of FATP4 in control cells and in FABP1 knockdown cells (Fig. 14A & B). Under basal
conditions, FATP4 was increased in both CD36 and FABP1 knockdown cells compared to
control cells (Fig. 14A & B). Lipids did not have a stimulatory effect on FATP4 protein levels
in CD36-deficient Caco-2 cells (Fig. 14A). The increase in FATP4 under basal conditions
might provide more activated substrate (fatty acyl-CoA) for the enzymes in either the MG or
G3P pathways of TG synthesis (Fig. 2).
50
The synthesis of TG is mediated by two diacylglycerol transferases, DGAT1 and
DGAT2. DGAT1 knockout mice can synthesize TG and have normal serum TG levels, even
when fed a high fat diet (Buhman et al., 2002). DGAT2, on the other hand, is important for TG
synthesis and is essential for survival (Stone et al., 2004). In control cells, DGAT2 expression
was decreased by lipid supplementation (Fig. 15), and in both knockdown cell lines, DGAT2
mRNA levels were decreased under basal and lipid stimulated conditions compared to their
respective controls (Fig. 15).
FATP4
0.0
0.5
1.0
1.5
Lipid Micelle - + - + Control CD36 KD
*
FATP
4 Pr
otei
n N
orm
aliz
edto
E-C
adhe
rin (A
U)
FATP4
0.0
0.5
1.0
1.5
Lipid Micelle - + - + Control FABP1 KD
*
*
*
FATP
4 Pr
otei
n N
orm
aliz
edto
E-C
adhe
rin (A
U)
A. B.
C. D.
Control CD36 KD Lipid Micelles - + - +
FATP4 70 kDa
E-Cadherin
Lipid Micelles - + - +
Control FABP1KD
FATP4 70 kDa
E-Cadherin
Figure 14. FATP4 protein expression levels in CD36 and FABP1 knockdown cells. Cells were treated with or without 1.6 mM oleate containing micelles and harvested after 12 hours. Representative immunoblot of cell lysates from CD36- (A) or FABP1- (B) deficient Caco-2 cells using an anti-FATP4 antibody. C & D. FATP4 protein expression was normalized to the loading control, E-Cadherin. n=9, *p<0.05.
51
DGAT2
Control CD36 KD CD36 KD
-1.5
-1.0
-0.5
0.0
+ lipid + lipid
Rel
ativ
e G
ene
Expr
essi
on(F
old
chan
ge c
ompa
red
to C
ontr
ol) DGAT2
Control FABP1 KD FABP1 KD
-1.5
-1.0
-0.5
0.0
*
+ lipid + lipid
Rel
ativ
e G
ene
Expr
essi
on(F
old
chan
ge c
ompa
red
to C
ontr
ol)
A. B.
To further investigate mechanisms responsible for the increase in TG and apoB secretion
observed from CD36 and FABP1 knockdown Caco-2 cells, we examined the expression of
genes or proteins involved in chylomicron assembly.
3.11. Factors involved in chylomicron assembly in CD36- and FABP1-
deficient Caco-2 cells
Differentiated CD36 and FABP1 knockdown Caco-2 cells were treated with or without
1.6 mM oleate-containing micelles for 12 hours, after which protein or RNA were isolated. The
mRNA expression or protein level of proteins involved in chylomicron assembly, such as
microsomal triglyceride transfer protein (MTP), Sar1 GTPase and ApoAIV was examined.
MTP possesses lipid transfer activity and plays a role not only in the initial lipidation of
newly translated apoB with ER-membrane bound TG (Black, 2007), but also in the core
expansion of the chylomicron particle. CD36 and FABP1 knockdown Caco-2 cells were probed
for MTP, however, MTP protein levels were variable among the samples, and the expected
Figure 15. Real-time PCR measurement of DGAT2 mRNA in control and CD36 or FABP1 knockdown Caco-2 cells. Differentiated Caco-2 cells were treated with or without 1.0mM oleate-containing lipid micelles for 12 hours. Ct values for DGAT2 were normalized to an internal control gene, 18S, and expressed as fold-change compared to their respective untreated controls. n=9, *p<0.05.
52
increase in MTP after lipid stimulation was not observed, except in CD36 knockdown cells (Fig.
16A).
Nascent chylomicrons exit the ER in a specialized transport vesicle, the prechylomicron
transport vesicle or PCTV. Work by Charles Mansbach’s group has shown that PCTVs can bud
from the ER in a COPII independent manner, but that the trafficking and fusion with the Golgi
are dependent on Sar1, Sec23/24, and its unique SNARE complex (Siddiqi et al., 2006). Sar1
expression was examined in CD36 and FABP1 knockdown cells, and in all cells, Sar1 levels
MTP
0
1
2
Lipid Micelles - + - +Control CD36 KD
** **
MTP
Pro
tein
Exp
ress
ion
Nor
mal
ized
to B
eta-
Act
in (
AU
)
MTP
0.0
0.5
1.0
1.5
2.0
2.5
Lipid Micelles - + - +Control FABP1 KD
*
MTP
Pro
tein
Exp
ress
ion
Nor
mal
ized
to B
eta-
Act
in( A
U)
A. B.
C. D.
Control FABP1 KD Lipid Micelles - + - +
MTP 95 kDa
Beta Actin
Control CD36 KD Lipid Micelles - + - +
MTP 95 kDa
Beta Actin
Figure 16. Microsomal triglyceride transfer protein levels in CD36 and FABP1 knockdown cells. Cells were treated with or without 1.6 mM oleate-containing micelles and harvested after 12 hours. Representative immunoblot of cell lysates from CD36- (A) or FABP1- (B) deficient Caco-2 cells using an anti-MTP antibody. C & D. MTP protein expression was normalized to the loading control, beta actin. n=6, *p<0.05, **p<0.01.
53
increased slightly after lipid micelle stimulation (Fig. 17). However, there were no differences
in levels of Sar1 between knockdown cells and their respective controls (Fig. 17).
ApoAIV is a 46 kDa glycoprotein that associates with nascent chylomicrons in the ER
and is secreted on the surface of chylomicrons at the basolateral membrane (Black, 2007). It has
been proposed to play a role in the enlargement and stabilization of the maturing chylomicron
(Black, 2007). The control cells for each knockdown cell lines responded differently to lipid
stimulation, that is, apoAIV mRNA levels decreased in the CD36 KD control cell line (Fig.
18A) and increased significantly in the FABP1 KD control cell line (Fig. 18B). There were no
statistically significant differences in the amount of apoAIV mRNA between the knockdown
Sar1
0.00
0.25
0.50
0.75
1.00 *
Control CD36 KDLipid Micelle - + - +
Sar1
Pro
tein
Expr
essi
on (A
U)
Sar1
0.00
0.25
0.50
0.75
1.00
1.25
Lipid Micelles - + - + Control FABP1 KD
Sar1
Pro
tein
Exp
ress
ion
Nor
mal
ized
to E
-Cad
herin
(AU
)
C. D.
A. B. Control FABP1KD
Sar1 21 kDa
E-Cadherin
Control CD36 KD
Lipid Micelles - + - +
Sar1 21 kDa
E-Cadherin
Lipid Micelles - + - +
Figure 17. Sar1 GTPase protein expression levels in CD36 and FABP1 knockdown cells. Cells were treated with or without 1.6 mM oleate-containing micelles and harvested after 12 hours. Representative immunoblot of cell lysates from CD36- (A) or FABP1- (B) deficient Caco-2 cells using an anti-Sar1 antibody. C & D. Sar1 protein expression was normalized to the loading control, E-Cadherin. n=9, *p<0.05, **p<0.01.
54
cells and their respective control, untreated cells (Fig. 18). Changes in the expression of
proteins involved in chylomicron assembly may not explain the increase in TG and apoB
secretion that we observe in the CD36 and FABP1 knockdown cell lines.
ApoAIV
Control CD36 KD CD36 KD-2
-1
0
1
2
+ lipid + lipid
Rel
ativ
e G
ene
Expr
essi
on(F
old
chan
ge c
ompa
red
to C
ontro
l)
ApoAIV
Control FABP1 KD FABP1 KD-2
-1
0
1
2
+ lipid + lipid
*
Rel
ativ
e G
ene
Expr
essi
on(F
old
chan
ge c
ompa
red
to C
ontro
l)
A. B.
Figure 18. Real-time PCR measurement of ApoAIV mRNA in control and CD36 or FABP1 knockdown Caco-2 cells. Differentiated Caco-2 cells were treated with or without 1.0mM oleate-containing lipid micelles for 12 hours. Ct values for ApoAIV were normalized to an internal control gene, 18S, and expressed as fold-change compared to their respective controls that were not treated with lipid. n=9, **p<0.01.
55
CHAPTER 4
4. DISCUSSION
4.1. shRNA Mediated Knockdown of CD36 and FABP1 in Caco-2 Cells
The discovery of RNA interference (RNAi) in 1998 (Fire et al., 1998) brought about
exciting opportunities for elucidating the role of your gene of interest. RNAi is a conserved
process of double-stranded RNA post-transcriptional gene expression modulation. Elaborate.
Since its discovery, RNAi technology has been harnessed for uses in cell biology and
therapeutics. With the improved algorithms for the design of short interfering and short hairpin
RNA sequences, investigators can attribute the observed phenotype to the loss of gene of
interest.
We employed the use of shRNA plasmids, containing sequences complementary to
CD36 and FABP1 to study the effect of knockdown of these proteins on intestinal lipid
absorption and lipoprotein assembly. CD36 and FABP1 mRNA levels were reduced by more
than 70% in Caco-2 cells stably expressing the shRNA plasmids compared to control cells
stably expressing the scrambled shRNA plasmid (Fig. 3A &D). CD36 protein was found to be
decreased by less than 60% of control cell levels (Fig 3C). As mentioned in Section 4.1., there
is no commercially available anti-FABP1 antibody that is suitable for detection of FABP1 by
immunoblot.
Researchers have struggled to delineate the role of CD36 in chylomicron production
because its intestinal function is complicated by its functions in peripheral tissues. The
intestinal specific contribution of FABP1 to lipoprotein metabolism is not clear in studies of
FABP1-/- mouse, however, Mansbach and colleagues, showed that FABP1 is important for
chylomicron secretion, since FABP1-depleted enterocyte cytosol had a decreased capacity to
bud nascent chylomicrons from the ER (Neeli et al., 2007). Therefore, the development of these
56
cell lines should aid in the resolution of the role of CD36 and FABP1 in intestinal chylomicron
assembly.
4.2. CD36- or FABP1-deficiency in Caco-2 cells did not impair fatty acid
uptake
Fatty acid uptake was measured by treating cells with [3H]-oleate-containing lipid
micelles and measuring the amount of radioactivity in the cells after 1, 2, 4, or 12 hours (Fig. 4).
At every time point, we did not observe a difference in oleic acid uptake by CD36-deficient
Caco-2 cells compared to their control cells (Fig. 4A). Several studies have implicated CD36 in
the absorption of very-long chain fatty acids (Drover et al., 2008d; Nassir et al., 2007). Drover,
et al. observed that FA absorption in CD36 expressing COS-7 cells was directly related to the
length of the acyl chain when the FA backbone was greater than 18 carbon atoms in length
(Drover et al., 2008c). However, there were no differences in absorption properties of oleic acid
(18:0), and palmitic acid (16:0) . We did not measure the uptake of very long chain fatty acids
in the current study. In contrast, a study by Nassir and colleagues suggests that CD36 is
important for oleic acid uptake. Isolated enterocytes from WT and CD36-/- mice were
incubated with different ratios of FA/BSA for 0 to 30 min to examine FA uptake (Nassir, et al.,
2007). CD36-deficient enterocytes had significantly reduced levels of oleic acid uptake
between 5 and 30 minutes, compared to WT enterocytes (Nassir et al., 2007). We conclude that
CD36-deficiency does not impair oleic acid uptake in Caco-2 cells, and these results are in line
with previous studies in cell culture (Drover et al., 2008b).
Fatty acid uptake was not different, at any measured time point, between FABP1-
deficient Caco-2 cells compared to its control cell line (Fig. 4B). Previous studies have shown
that FABP1 enhances fatty acid uptake (McArthur et al., 1999; Prows et al., 1995; Wolfrum et
57
al., 1999). Deficiency of FABP1 has not been shown to affect the absorption of oleic acid. Our
studies indicate that in Caco-2 cells, FABP1-deficiency does not impair FA uptake.
4.3. CD36- or FABP1-deficiency in Caco-2 cells is associated with an increase
in TG and ApoB secretion
The intracellular lipid content, as measured by [3H]-oleate labelling experiments, was
not significantly different in either CD36- or FABP1- deficient Caco-2 cells (Fig. 5A & B).
However, there was a significant increase in TG secretion from CD36- deficient cells with lipid
stimulation (Fig. 5A). A number of studies have documented a postprandial increase in plasma
TG in both CD36-/- mice and humans (Masuda et al., 2008; Miyaoka et al., 2001), however,
some investigators attribute this to impaired clearance of lipoprotein TG (Masuda et al., 2008).
Masuda, et al. investigated the mechanisms of postprandial hypertriglyceridemia
observed in CD36-states (Masuda et al., 2008). Patients with CD36-deficiency displayed
increased FA, TG and apoB48 levels in both fasting and postprandial states (Miyaoka et al.,
2001). After an oral fat load, there was a delayed peak in plasma TG and apoB48 in CD36-
deficient patients (Masuda et al., 2008; Miyaoka et al., 2001), suggesting that there is either an
overproduction of CM or impaired clearance of CM remnants in CD36-deficient states (Masuda
et al., 2008). These results support those observed in a study by Drover, et al. who showed that
TG secretion into the lymph was delayed in CD36-/- mice compared to wild type (Drover et al.,
2005). It seemed that there may be a defect in lipoprotein lipase activity, since the size of CM
particles (measured by electron microscopy) were much larger (from 110nm to 200-700nm in
diameter) in the plasma of CD36-/- mice (Drover et al., 2005). However, Masuda, et al.
measured lipase activity in the plasma of CD36-/- and wild type animals and found no difference
in either lipoprotein lipase or hepatic lipase activity (Masuda et al., 2008). This supports the
58
premise that synthesis of lipoproteins is increased in the intestines of CD36-deficient animals
and humans.
In our study, FABP1-deficient Caco-2 cells treated with lipid showed a trend towards
increased TG secretion (Fig. 5B). These are consistent with another study examining intestinal
lipid output in FABP1-/- mice (Newberry et al., 2003). When TG secretion rates were examined
using tyloxapol, there were no apparent differences in intestinal TG output between FABP1-/-
and WT mice (reported as unpublished observations in (Newberry et al., 2003)).
Using two different approaches; quantifying apoB mass by immunoblotting or
radiolabelling newly synthesized apoB, we have shown that under basal conditions, knockdown
of FABP1 results in increased secreted apoB100, and that knockdown of CD36 results in
increased apoB100 and apoB48 secretion in Caco-2 cells. It is well established that apoB
secretion is enhanced by lipid stimulation (reviewed in (Ginsberg and Fisher, 2009)). TG and
phospholipids act to stabilize the nascent apoB polypeptide, thereby rescuing it from
degredation pathways (reviewed in (Fisher and Ginsberg, 2002)). Although Liao and Chan have
shown that apoB does not undergo intracellular degradation in Caco-2 cells (Liao and Chan,
2000), our results are consistent with the former observations. Scrambled shRNA control Caco-
2 cells secrete apoB under basal conditions, and the secretion of apoB is enhanced upon lipid
micelle stimulation (Fig. 6A and B). Interestingly, CD36-deficient Caco-2 cells secrete
significantly more apoB than control cells under basal conditions (Fig. 6A). When stimulated
with lipid, there is an increase in the amount of secreted apoB, however, this is comparable to
lipid-stimulated control cell amounts (Fig. 6A).
A potentially novel finding from these studies is the absence of lipid-induced
enhancement of apoB secretion from FABP1-deficient Caco-2 cells (Fig. 6B). Newberry, et al.,
reported no differences in plasma apoB100 or apoB48 from FABP1-/- mice compared to WT
mice (Newberry et al., 2003). The link between FABP1 and apoB secretion has been
59
investigated by Mansbach and colleagues, who found that FABP1 was necessary for the exit of
nascent chylomicrons (or prechylomicrons) from the ER (Neeli et al., 2007). In these studies,
WT mouse enterocytes were pre-incubated with [3H]-oleate for 30 minutes, followed by
isolation of the ER. Nascent chylomicron particles were budded from the ER by incubation
with WT or FABP1-/- cytosol. FABP-/- cytosol provided only 60% of the budding activity of
WT cytosol, indicating that FABP1 plays a role in the budding of the prechylomicron particle
(Neeli et al., 2007). The decrease in TG output from the ER treated with FABP1-/- cytosol, and
the authors are currently investigating whether this is a result of a decrease in number of
particles, or a decrease in the size of the budded prechylomicron particles (Neeli et al., 2007).
In our study, depletion of FABP1 resulted in a greater amount of secreted TG, but the
same amount of apoB. A possible explanation for these observations is that FABP1 restricts the
size of nascent chylomicron particles exiting the ER. This scenario would allow for the
formation of bigger, but not more, prechylomicron particles, and explain the absence of the
lipid-induced enhancement in apoB secretion.
4.4. Depletion of CD36 in Caco-2 cells results in a shift in secreted lipoprotein
profile
The TG in large and small CM, and in VLDL particles secreted from Caco-2 cells was
measured (Fig. 6). Under basal conditions, Caco-2 cells secrete small, dense lipoproteins of
HDL size (Levy et al., 1995). We also made this observation in our control Caco-2 cell line.
The three fractions of lipoproteins analyzed contained very little TG in control cells that were
not stimulated with lipid (Fig. 6). However, Fatma, et al. showed that when stimulated with an
appropriate concentration of oleate and taurocholate-containing lipid micelles, Caco-2 cells
secrete large CM particles (Fatma et al., 2006). The TG content in the CM fraction of control
cells that were with stimulated with lipid micelles was significantly greater than in control cells
60
in the basal condition. This indicates that the control Caco-2 cells can assemble TG-rich large
CM-sized particles (Fig. 6).
In CD36-deficient Caco-2 cells, there is a significant increase in the TG content of the
VLDL fraction compared to control cells under basal conditions (Fig. 6). This was an
unexpected observation, and would require a significant upregulation of de novo TG synthesis.
The culture conditions used expose Caco-2 cells to a high concentration of FA (20% FBS),
therefore, there may have been an increase in esterification of FA to neutral lipids.
In mice lacking CD36, there is a shift in particle size from LDL to VLDL in the fasting
state (Febbraio et al., 1999). The authors attributed this increase in particle size to a defect in
peripheral fatty acid utilization. As mentioned above, a deficiency in lipoprotein lipase activity
was suggested to account for the defect in uptake by peripheral tissues, although no change in
activity was noted (Masuda et al., 2008). In a more recent study, fasting and postprandial
lipoprotein profiles were examined in CD36-/- mice (Masuda et al., 2008). The fasting TG
content of the VLDL-sized fraction was significantly higher in CD36-/- mice (Masuda et al.,
2008). In fact, when lipoprotein lipase activity was blocked by an inhibitor of Lpl, triton WR-
1339, fasting plasma TG levels were not different between wild type and CD36-/- animals
(Masuda et al., 2008). This supports the notion that hepatic lipoprotein synthesis is not different
between wild type and CD36-/- animals, hence, the difference in fasting levels of VLDL-
associated TG may be attributed to intestinal output.
In our CD36-deficient Caco-2 cell model, we observed an increase in VLDL-associated
TG secreted under basal conditions, and an increase in small CM-associated TG secreted under
lipid stimulated conditions. In the postprandial state, significant increases in CM, VLDL, and
LDL TG was observed in CD36-/- mice (Masuda et al., 2008). Furthermore, inhibition of Lpl
activity after an oral fat load resulted in a significant accumulation of TG in the plasma of
CD36-/- mice compared to wild type (Masuda et al., 2008). Perhaps, upon lipid stimulation,
61
there is an even further induction of the assembly of buoyant particles (>VLDL in size) in
CD36-deficient states. We examined this possibility by measuring levels of de novo lipogenesis
and proteins involved in the assembly of chylomicrons.
4.5. Evidence for an increase in de novo lipogenesis in CD36-deficient Caco-2
cells
FAS protein levels were elevated under basal conditions in both CD36- and FABP1-
deficient Caco-2 cells compared to control Caco-2 cell levels (Fig. 8). When stimulated with
lipid, FAS levels increased in CD36 knockdown control cells and in FABP1-deficient cells (Fig.
8C & D). No change was observed in CD36-deficient cells or in FABP1 knockdown control
cells (Fig. 8C &D). This suggests that, under basal conditions, de novo lipogenesis is increased
in CD36- and FABP1-deficient Caco-2 cells. In animals, FAS is transcriptionally regulated by
feeding and insulin. FAS expression is stimulated by insulin in the liver (Paulauskis and Sul,
1989) and by leptin in adipose tissue, through the action of and sterol regulatory element
binding protein-1c (SREBP-1c) and Upstream Stimulatory Factor (USF) (Latasa et al., 2003).
FAS expression is also indirectly regulated by fatty acids. Studies in HepG2 cells showed that
SREBP-1-activated gene transcription was decreased by supplementation with oleic acid and
other polyunsaturdated fatty acids, while saturated fatty acids did not regulate SREBP-1-
activated genes (Worgall et al., 1998). Thus, we would expect control Caco-2 cells treated with
oleate-containing lipid micelles to have decreased levels of FAS compared to basal conditions.
The presence of taurocholate in the culture media may also provide an explanation for this
observation, since transcription of FAS is also regulated by bile acids through an FXR
dependent mechanism (Matsukuma et al., 2006). It is unclear why FAS expression in the
control cell lines did not respond to lipid micelle stimulation in the same way.
62
The reaction catalyzed by ACC is the first committed step in FA synthesis, converting
malonyl CoA to acetyl CoA. ACC1 expression is controlled by at least three promoters, PI-PIII,
and key transcription factors include sterol-regulatory-element-binding protein 1C (SREBP1C)
as well as liver X receptor, retinoid X receptor, PPARs, FOXO, and PPARγ co-activator
(Brownsey et al., 2006). Levels of ACC mRNA were significantly reduced in control and
CD36- and FABP1-deficent Caco-2 cells treated with oleate-containing lipid micelles (Fig. 9A
and B). Like FAS, ACC expression is controlled by SREBP1C (Brownsey et al., 2006).
Therefore, the above results are consistent with the regulation of ACC1 by SREBP1C. On the
other hand, we observed a decreased in ACC1 expression in CD36 and FABP1 knockdown
Caco-2 cells under basal conditions (Fig. 9A and B). Reduced activity of ACC would limit the
supply of malonyl-CoA, the essential substrate for fatty acid synthesis.
Our observations that an increase in FAS protein levels under basal conditions led us to
hypothesize that there may be an increase in basal levels of FA synthesis. However, the
expression of ACC does not support this hypothesis. To directly measure the level of FA
synthesis in Caco-2 cells, we labeled cells with [3H]-acetate and measured its incorporation into
FA, CE, and TG (Fig. 11). Oleate treatment significantly reduced FA synthesis in CD36 control
cells. Interestingly, FA synthesis in CD36-deficient Caco-2 cells seemed resistant to the
inhibitory effects of either FA or cerulenin, however, FA and cerulenin together modestly
reduced FA synthesis (Fig. 11E). The observation that CD36-deficient cells are more resistant
to cerulenin treatment than control Caco-2 cells points to a role for CD36 in the targeting of FA
to the appropriate subcellular locations. Moreover, CD36 may act in FA targeting or signalling,
and its ablation could result in a loss of FAS feedback inhibition.
As mentioned above, evidence from our lab suggests that upon induction of an insulin
resistant state, there is an increase in FA synthesis in the intestine as measured by [3H]-oleate
incorporation into FA (Haidari et al., 2002). The increase in FA synthesis was associated with
63
the observed fasting and postprandial increase in TG and apoB48 secretion observed in insulin
resistant hamsters (Haidari et al., 2002). Interestingly, Masuda, et al. showed that mRNA levels
of FAS were increased in the intestines of CD36-/- mice in both the fasted and postprandial
conditions, although they did not measure protein levels or FA synthesis directly (Masuda et al.,
2008).
These studies both support the idea that FA biosynthesis is susceptible to both dietary
and genetic influences and may perturb normal FA metabolism. How CD36 and FABP1 are
involved in the regulation of FA metabolism in the intestine has yet to be investigated. It is
tempting to speculate that these lipid binding proteins are involved in the targeting of fatty acids
to their appropriate metabolic fate. CD36 and FABP1 may interact with other proteins to
accomplish their role in targeting of FA. For example, FABP1 has been shown to directly
interact with PPARα, which induces a number of genes involved in lipid metabolism (Hostetler
et al., 2009). Furthermore, the observed interaction between FABP1 and PPARα suggest that
FABP1-bound ligands may be transferred to PPARα (Hostetler et al., 2009). PPARα induces a
number of genes involved in the oxidation of fatty acids. If FABP1 is responsible for the
delivery of FA to induce PPARα-mediated activation of genes, and ultimately the oxidation of
FA, its absence could reduce the level of FA oxidation in the cell.
We propose that CD36 is acting as a lipid sensor in Caco-2 cells, such that its absence
perturbs endogenous FA metabolism. SREBP1 induces transcription of genes involved in de
novo lipogenesis, including FAS, ACC1 and SCD1. FA and cholesterol are suppressors of de
novo FA synthesis pathway through inhibition of cleavage of SREBP1. Moreover, we suggest
that CD36 normally functions to sense FA and cholesterol. In support of this, another member
of scavenger receptors, SR-B1, plays a role in lipid sensing in the intestine (Beaslas et al.,
2009). Perhaps CD36 and FABP1 function together, and the absence of one of these key
regulatory molecules would perturb normal FA metabolism.
64
4.6. Triacylglycerol synthesis in CD36- or FABP1-deficient Caco-2 cells
Intracellular [3H]-oleate incorporation into TG was not different, while secreted [3H]-TG
was increased from CD36- or FABP1-deficient Caco-2 cells compared to their respective
control cell lines (Fig. 4). In this study, differentiated Caco-2 cells were incubated with [3H]-
oleate-containing lipid micelles for 12 hours, according to a previously published protocol
(Luchoomun and Hussain, 1999). Drover, et al. showed that FA transport to microsomes was
defective in CD36-/- enterocytes treated with [3H]-oleate-containing micelles (Drover et al.,
2005). The [3H] incorporation into CE, FA, and DAG, was not significantly different in
microsomes isolated from CD36-/- enterocytes compared to wild type (Drover et al., 2005),
although there was a significant increase in [3H] incorporation into TG. In their study, the
primary enterocytes were incubated with lipid micelles for 1 hour (Drover et al., 2005).
Therefore, the defect in FA delivery to microsomes can explain the initial lag in TG output to
the lymphatic circulation observed in this model (Drover et al., 2005). In the present study,
there may be an initial defect in targeting of FA in CD36- or FABP1-deficient Caco-2 cells,
however, at the end of the 12 hour incubation with lipid micelles, a mistargetting of FA or lag in
TG secretion was not observed. As mentioned above, CD36 was proposed to play a role in the
targeting of exogenous fatty acids to the microsomal TG secretion pool (Drover et al., 2005).
On the contrary, we conclude that CD36 is not necessary for the formation and secretion of TG,
as its deficiency was actually associated with an increase in TG secretion.
We have previously shown that DGAT activity in the intestine is associated with an
increase in TG accumulation and secretion in a dietary model of insulin resistance, the fructose-
fed Syrian golden hamster (Casaschi et al., 2005). DGAT activity was increased 2.5- and 5-fold
in the proximal intestine upon feeding of fructose and fat-enriched diets, respectively (Casaschi
et al., 2005), but was not associated with an increase in mRNA levels of either DGAT1 or
65
DGAT2. In Caco-2 cells deficient in CD36, DGAT2 mRNA levels were not changed between
control and knockdown cell lines, despite an increase in TG secretion (Fig. 15A). In addition,
fatty acid supplementation did not increase DGAT2 mRNA levels in either control or CD36-
deficient Caco-2 cells (Fig. 15A). Under basal conditions, DGAT2 expression was not changed
in FABP1 knockdown Caco-2 cells compared to their control, however, there was a significant
decrease in DGAT2 mRNA levels when these cells were supplemented with lipid (Fig. 15B).
The regulation of DGAT enzymes under some condition is at the level of mRNA, however,
transcription factors responsible for its regulation are not well characterized (reviewed in (Yen
et al., 2008)).
Recently, the contribution of DGAT1 activity to the synthesis of TG in Caco-2 cells was
investigated (Cheng et al., 2008). They showed that, in vitro, approximately 76% TG synthesis
from MG is mediated by DGAT1 in Caco-2 cell (Cheng et al., 2008). Recall that the majority
of TG synthesis in Caco-2 cells is formed via the G3P pathway (reviewed in (Levy et al.,
1995)), therefore, the relative amount of TG produced by DGAT1 may not account for 76% of
total TG in normal Caco-2 cells. DGAT2 may catalyze the formation of TG from intermediates
produced by the G3P pathway. It is also possible that DGAT1 catalyzes the assimilation of TG
from G3P pathway-derived DAG, however, this was not tested.
Incorporation of oleate into TG was reduced in adipocytes from CD36-/- mice, while
incorporation into DAG was increased (Febbraio et al., 1999). It followed then, that there was
an impairment of DAG conversion to TG in CD36-deficient conditions (Febbraio et al., 1999).
In the intestine, however, Drover, et al. showed that DAG species do not accumulate. They also
did not detect changes in either DGAT1 or DGAT2 mRNA levels (Drover et al., 2005).
Fatty acids that are taken up by cells either diffuse or are transported to the appropriate
subcellular location, for example, the mitochondria, peroxisomes, or the ER and are activated by
their conversion to CoA thioesters. FATP4 has been implicated in fatty acid uptake via this
66
process, which is termed vectorial acylation. In the current study, we observed an increase in
FATP4 protein levels under basal conditions in both CD36- and FABP1-deficient Caco-2 cells.
Since FATP4 is present on the ER membrane (Ehehalt et al., 2006; Stahl et al., 1999), it is
possible that FATP4 is curbing normal cellular metabolism towards secretion.
4.7. Factors involved in chylomicron assembly and secretion
We have previously shown that MTP gene expression is increased by oleate in an
SREBP1-independent manner in Hep-G2 cells (Qiu et al., 2005), while cholesterol increases
MTP expression in a SREBP1-dependent manner (reference- chapter 14). MTP expression is
highly regulated in the liver, but little is known about the regulation of MTP in the intestine. If
MTP expression is regulated in a similar manner in liver and enterocytes, we may expect to see
an increase in MTP levels in our Caco-2 cells treated with oleate-containing lipid micelles.
However, we do not see an increase in MTP upon lipid stimulation except for in CD36-deficient
Caco-2 cells (Fig. 16). Although the lack of FABP1 in FABP-deficient Caco-2 cells could
perturb normal FA signaling mechanism, this is probably not the case in this experiment, since
control Caco-2 cells did not respond to oleate either.
The intestine is the first organ to process dietary lipid, and the absorption of lipid is
clearly very complex and highly regulated, although it is not well understood. Many of the
findings from hepatocyte lipoprotein assembly are assumed to be true for enterocyte lipoprotein
assembly as well. However, there is evidence that regulation of protein involved in chylomicron
assembly do not respond in a similar fashion in liver and the intestine. For example, MTP
expression in the liver of Sprague-Dawley rats is downregulated in response to a 6- and even
more so in a 12-hour fast, while its expression in the intestine does not change (Lally et al.,
2007).
67
There may also be differences in the regulation of proteins involved in chylomicron
assembly during different developmental stages. Lu, et al. have shown that in enterocytes of
newborn swine, both MTP and ApoAIV mRNAs are upregulated by duodenal lipid infusion
(Leng et al., 2007b; Lu et al., 2002). HNF4α regulates transcription of ApoAIV in enterocytes
from newborn swine (Leng et al., 2007a). The MTP gene also contains an HNF4α binding
elements in its promoter region (Hussain et al., 2003). In our hands, the expression of ApoAIV
did not change in response to oleate-containing lipid micelle stimulation (Fig. 18).
The involvement of Sar1 to the secretion of lipoprotein is unique to the intestine, as
mutations in the SARA2 gene lead to chylomicron retention disease (Dannoura et al., 1999;
Jones et al., 2003). Patients with this rare-lipid malabsorption disorder secrete VLDL particles
from the liver but retain chylomicrons in enterocytes. Previous work in our lab has shown that
the protein level of Sar1 are increased 2.13-fold in chow fed hamsters after receiving a bolus of
olive oil (Wong, et al. Manuscript under revision). Furthermore, we have shown that Sar1
protein levels are increased 2.33-fold in fructose-fed hamsters, a well-established model of a
mild-insulin resistant state (Wong, et al. Manuscript under revision; Avramoglu et al., 2003;
Avramoglu et al., 2006; Taghibiglou et al., 2000). On this dietary background, Sar1 protein
expression is not induced further by a lipid bolus (Wong, et al. Manuscript under revision). In
this study, Sar1 levels were not induced by oleate-containing micelles stimulation in Caco-2
cells, except when CD36 was deficient.
4.8. Fatty acid synthesis in enterocytes isolated from CD36-/- mice
In order to determine whether our observations in Caco-2 cells accurately represent what
is happening in vivo, we examined secretion of lipids from isolated entercoytes from CD36-/-
mice. We observed a significant reduction in secreted TG from CD36-/- enterocytes and these
results are consistent with several studies examining intestinal lipid absorption and secretion in
68
CD36-/- mice (Drover et al., 2005; Drover et al., 2008a). We next wanted to examine FA
synthesis in CD36-/- enterocytes, since a more recent study has suggested that FA synthesis may
be upregulated in the intestine of CD36-/- mice (Masuda et al., 2008), and we have similar
evidence of this in CD36-deficient Caco-2 cells. To assess FA synthesis in isolated enterocytes,
we measured the incorporation of [3H]-acetate into CE, TG, FA, and cholesterol (Fig. 13A).
There were no siginificant differences in the incorporation of [3H]-acetate into CE, TG, or FA,
but we observed a significant reduction in the incorporation of [3H]-acetate into cholesterol.
CD36 has been implicated in intestinal cholesterol metabolism (Nassir et al., 2007; Nauli et al.,
2006); our observations also support a relationship between CD36 and cholesterol metabolism.
To assess enterocyte de novo lipogenesis, we measured the incorporation of [3H]-oleate
into CE, TG, FA, and cholesterol (Fig. 13B). There were no significant differences in the
incorporation of [3H]-oleate into CE, FA, or cholesterol, but we observed a significant reduction
in the incorporation of [3H]-acetate into TG. The reduction in TG synthesis in CD36-/-
enterocytes is consistent with the decrease in microsomal TG observed in CD36-/- enterocytes
(Drover et al., 2005).
Fatty acid and triacylglycerol metabolism in Caco-2 cells is different from mature villus
enterocytes (discussed above). The impairment of TG synthesis associated with CD36-
deficiency implicates CD36 in the MAG (dominant) pathway of TG synthesis. Differences in
TG synthesis pathway predominance in Caco-2 cells and normal enterocytes may explain the
differences in TG synthesis and secretion observed in these two models. It is also possible that
the level of CD36 expression is important to its function, and that a model in which CD36 is
completely absent would react differently than a model in which CD36 is expressed at 50% of
control cell levels.
69
4.9. Concluding remarks
The wealth of information concerning CD36 and FABP1 and their role in FA
metabolism is complicated by their tissue-specific roles and the differences in phenotypes under
different dietary conditions. It is clear that both CD36 and FABP1 are important players in FA
and lipoprotein metabolism in the intestine, as polymorphisms in both genes are associated with
the metabolic syndrome (Ma et al., 2004). We have investigated the role of these two proteins
in a cell culture model of mature villus enterocytes, Caco-2 cells. We have used a Caco-2 cell
culture model in order to avoid interference of various circulating factors that can metabolize
intestinally secreted lipoproteins. From these studies, we report a number of observations: (i)
shRNA-mediated knockdown of CD36 or FABP1 in Caco-2 cells was associated with increased
TG secretion under lipid stimulated conditions, (ii) shRNA-mediated knockdown of CD36 or
FABP1 in Caco-2 cells was associated with increased apoB secretion under basal conditions,
(iii) deficiency of CD36 results in a shift in lipoprotein associated TG, from large CM only, to
both large and small CM, (iv) deficiency of CD36 or FABP1 results in increased FAS
expression under basal conditions, (v) cerulenin reduces FA synthesis in Caco-2 cells, but that
CD36-deficient cells are more resistant to cerulenin-mediated inhibition of FA synthesis, (vi)
FATP4 levels are increase under basal conditions in CD36- or FABP1-deficient Caco-2 cells,
(vii) levels of MTP, Sar1, and ApoAIV were not dramatically different in the knockdown cells
versus control cells, under basal or lipid-stimulated conditions.
We have also performed several experiments in CD36-/- mice in order to compare our in
vitro findings to an in vivo model. From these studies, we report that CD36-/- mice (i) have
impaired cholesterol and TG synthesis and (ii) a significant reduction in secreted TG compared
to WT animals.
70
Patients with CD36-deficiency and the CD36-/- mouse display an increase in fasting and
postprandial apoB48 and TG (Miyaoka et al., 2001), which is not associated with a defect in Lpl
activity (Masuda et al., 2008). This suggests that overproduction of intestinal lipoproteins is
responsible for this postprandial increase in apoB48 and TG (Masuda et al., 2008). In our study,
CD36 or FABP1-deficiency is associated with an increase in FAS expression. Although FAS
protein levels did not correlate with fatty acid synthesis in these cells, there may be an increase
in FA synthesis in CD36 knockdown Caco-2 cells.
As mentioned above, FABP1 has been shown to directly interact with PPARα, a
transcription factor responsible for the induction of genes involved in FA oxidation (Hostetler et
al., 2009). Rivabene, et al. have shown that the pre-existing redox state of Caco-2 cells affects
cellular lipid metabolism (Rivabene et al., 2001a). Modulation of the redox status in Caco-2
cells using antioxidant N-acetylcysteine (NAC) or the pro-oxidizing agent CuSO4, altered the
metabolism of oleate supplemented to Caco-2 cells (Rivabene et al., 2001b). The synthesis of
TG was significantly lowered under more reducing conditions using NAC. In the absence of
FABP1, the normal oxidative and reductive potential of Caco-2 cells may be altered.
Upregulation of a number of genes would be necessary to accommodate diminished FA
oxidation in order to relieve the cells of the potential lipotoxic effects of FFA. For example,
upregulation of FATP4 could be an adaptive response, activating FA and shuttling them towards
secretion or storage as neutral TG.
We postulate that CD36 and FABP1 are intimately involved in lipid sensing and
downstream regulation of lipid metabolism in the cell, and our results indicate that modulation
of their expression leads to aberrant intestinal lipid and lipoprotein production.
71
CHAPTER 5
5. FUTURE DIRECTIONS
The absence of CD36 may result in the overcompensation by a molecule that was not
measured in our studies, SR-B1. SR-B1 is a scavenger receptor that is present on the apical
membrane of Caco-2 cells and acts as a lipid sensor (Beaslas et al., 2009). Supplementation of
Caco-2 cells with postprandial lipid micelles results in downstream signaling events, dependent
on SR-B1, and a re-localization of apoB to apical compartments (Beaslas et al., 2009). The
relationship between SR-B1 and CD36 in Caco-2 cells should be examined; SR-B1 levels in
Caco-2 cells deficient in CD36 could be measured by immunoblotting. Co-
immunoprecipitation studies should be performed to determine whether CD36 and SR-B1
interact. Confocal immunofluoresce based assays could be used to determine the subcellular
localization of CD36 in relation to SR-B1 and in response different stimuli (basal and
postprandial conditions).
Another experimental approach could have yielded some important information with
respect to the importance of CD36 or FABP1 to fatty acid uptake in Caco-2 cells. A study
should be designed in which CD36- or FABP1-knockdown Caco-2 cells are challenged with
various concentrations of fatty acid. This experiment could answer the question of whether
CD36 or FABP1 are critical for fatty acid uptake when fatty acid concentrations are extremely
low.
A novel observation in the current study was the lack of lipid-induced enhancement of
apoB secretion from FABP1-deficient Caco-2 cells. As discussed above, this observation is
supported by the findings of Neeli, et al., which have shown that FABP1 participates in the
selection of cargo for, and budding of, the prechylomicron transport vesicle (Neeli et al., 2007).
When FABP1 is deficient, we propose that there is less regulation of cargo budding, which may
72
allow for the formation of larger particles. Future experiments could focus on characterizing
pre-chylomicron particles formed from FABP1-deficient Caco-2 cells or from isolated
enterocytes from FABP1-/- mice. Our lab has performed previous experiments on the proteome
of prechylomicron transport vesicles isolated from enterocytes from the Syrian golden hamster
(Wong et al., 2009).
73
REFERENCES
Abumrad N, Coburn C, Ibrahimi A. Membrane proteins implicated in long-chain fatty acid uptake by mammalian cells: CD36, FATP and FABPm. Biochim. Biophys. Acta 1441: 4-13, 1999.
Abumrad NA, el Maghrabi MR, Amri EZ, Lopez E, Grimaldi PA. Cloning of a rat adipocyte membrane protein implicated in binding or transport of long-chain fatty acids that is induced during preadipocyte differentiation. Homology with human CD36. J. Biol. Chem. 268: 17665-17668, 1993.
Abumrad NA, Perkins RC, Park JH, Park CR. Mechanism of long chain fatty acid permeation in the isolated adipocyte. J. Biol. Chem. 256: 9183-9191, 1981.
Annuzzi G, De Natale C, Iovine C, Patti L, Di Marino L, Coppola S, Del Prato S, Riccardi G, Rivellese AA. Insulin resistance is independently associated with postprandial alterations of triglyceride-rich lipoproteins in type 2 diabetes mellitus. Arterioscler. Thromb. Vasc. Biol. 24: 2397-2402, 2004.
Annuzzi G, Giacco R, Patti L, Di Marino L, De Natale C, Costabile G, Marra M, Santangelo C, Masella R, Rivellese AA. Postprandial chylomicrons and adipose tissue lipoprotein lipase are altered in type 2 diabetes independently of obesity and whole-body insulin resistance. Nutr. Metab Cardiovasc. Dis. 18: 531-538, 2008.
Annuzzi G, Holmquist L, Carlson LA. Concentrations of apolipoproteins B, C-I, C-II, C-III, E and lipids in serum and serum lipoproteins of normal subjects during alimentary lipaemia. Scand. J. Clin. Lab Invest 49: 73-81, 1989.
Avramoglu RK, Basciano H, Adeli K. Lipid and lipoprotein dysregulation in insulin resistant states. Clin. Chim. Acta 368: 1-19, 2006.
Avramoglu RK, Qiu W, Adeli K. Mechanisms of metabolic dyslipidemia in insulin resistant states: deregulation of hepatic and intestinal lipoprotein secretion. Front Biosci. 8: d464-d476, 2003.
Beaslas O, Cueille C, Delers F, Chateau D, Chambaz J, Rousset M, Carriere V. Sensing of dietary lipids by enterocytes: a new role for SR-BI/CLA-1. PLoS. One. 4: e4278, 2009.
Black DD. Development and physiological regulation of intestinal lipid absorption. I. Development of intestinal lipid absorption: cellular events in chylomicron assembly and secretion. Am. J. Physiol Gastrointest. Liver Physiol 293: G519-G524, 2007.
Boren J, Rustaeus S, Olofsson SO. Studies on the assembly of apolipoprotein B-100- and B-48-containing very low density lipoproteins in McA-RH7777 cells. J. Biol. Chem. 269: 25879-25888, 1994.
Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72: 248-254, 1976.
74
Breckenridge WC, Kuksis A. Triacylglycerol biosynthesis in everted sacs of rat intestinal mucosa. Can. J. Biochem. 53: 1184-1195, 1975.
Brownsey RW, Boone AN, Elliott JE, Kulpa JE, Lee WM. Regulation of acetyl-CoA carboxylase. Biochem. Soc. Trans. 34: 223-227, 2006.
Buhman KK, Smith SJ, Stone SJ, Repa JJ, Wong JS, Knapp FF, Jr., Burri BJ, Hamilton RL, Abumrad NA, Farese RV, Jr. DGAT1 is not essential for intestinal triacylglycerol absorption or chylomicron synthesis. J. Biol. Chem. 277: 25474-25479, 2002.
Casaschi A, Maiyoh GK, Adeli K, Theriault AG. Increased diacylglycerol acyltransferase activity is associated with triglyceride accumulation in tissues of diet-induced insulin-resistant hyperlipidemic hamsters. Metabolism 54: 403-409, 2005.
Cases S, Smith SJ, Zheng YW, Myers HM, Lear SR, Sande E, Novak S, Collins C, Welch CB, Lusis AJ, Erickson SK, Farese RV, Jr. Identification of a gene encoding an acyl CoA:diacylglycerol acyltransferase, a key enzyme in triacylglycerol synthesis. Proc. Natl. Acad. Sci. U. S. A 95: 13018-13023, 1998.
Cheng D, Iqbal J, Devenny J, Chu CH, Chen L, Dong J, Seethala R, Keim WJ, Azzara AV, Lawrence RM, Pelleymounter MA, Hussain MM. Acylation of acylglycerols by acyl coenzyme A:diacylglycerol acyltransferase 1 (DGAT1). Functional importance of DGAT1 in the intestinal fat absorption. J. Biol. Chem. 283: 29802-29811, 2008.
Cianflone K, Paglialunga S, Roy C. Intestinally derived lipids: metabolic regulation and consequences--an overview. Atheroscler. Suppl 9: 63-68, 2008.
Clader JW. The discovery of ezetimibe: a view from outside the receptor. J. Med. Chem. 47: 1-9, 2004.
CLARK B, HUBSCHER G. Biosynthesis of glycerides in subcellular fractions of intestinal mucosa. Biochim. Biophys. Acta 46: 479-494, 1961.
Coburn CT, Knapp FF, Jr., Febbraio M, Beets AL, Silverstein RL, Abumrad NA. Defective uptake and utilization of long chain fatty acids in muscle and adipose tissues of CD36 knockout mice. J. Biol. Chem. 275: 32523-32529, 2000.
Coe NR, Bernlohr DA. Physiological properties and functions of intracellular fatty acid-binding proteins. Biochim. Biophys. Acta 1391: 287-306, 1998.
Connelly MA, Williams DL. SR-BI and HDL cholesteryl ester metabolism. Endocr. Res. 30: 697-703, 2004.
Dannoura AH, Berriot-Varoqueaux N, Amati P, Abadie V, Verthier N, Schmitz J, Wetterau JR, Samson-Bouma ME, Aggerbeck LP. Anderson's disease: exclusion of apolipoprotein and intracellular lipid transport genes. Arterioscler. Thromb. Vasc. Biol. 19: 2494-2508, 1999.
Darimont C, Gradoux N, Persohn E, Cumin F, De Pover A. Effects of intestinal fatty acid-binding protein overexpression on fatty acid metabolism in Caco-2 cells. J. Lipid Res. 41: 84-92, 2000.
75
Dashti N, Smith EA, Alaupovic P. Increased production of apolipoprotein B and its lipoproteins by oleic acid in Caco-2 cells. J. Lipid Res. 31: 113-123, 1990.
Drover VA, Ajmal M, Nassir F, Davidson NO, Nauli AM, Sahoo D, Tso P, Abumrad NA. CD36 deficiency impairs intestinal lipid secretion and clearance of chylomicrons from the blood. J. Clin. Invest 115: 1290-1297, 2005.
Drover VA, Nguyen DV, Bastie CC, Darlington YF, Abumrad NA, Pessin JE, London E, Sahoo D, Phillips MC. CD36 mediates both cellular uptake of very long chain fatty acids and their intestinal absorption in mice. J. Biol. Chem. 283: 13108-13115, 2008d.
Ehehalt R, Fullekrug J, Pohl J, Ring A, Herrmann T, Stremmel W. Translocation of long chain fatty acids across the plasma membrane--lipid rafts and fatty acid transport proteins. Mol. Cell Biochem. 284: 135-140, 2006.
Farese RV, Jr., Veniant MM, Cham CM, Flynn LM, Pierotti V, Loring JF, Traber M, Ruland S, Stokowski RS, Huszar D, Young SG. Phenotypic analysis of mice expressing exclusively apolipoprotein B48 or apolipoprotein B100. Proc. Natl. Acad. Sci. U. S. A 93: 6393-6398, 1996.
Fatma S, Yakubov R, Anwar K, Hussain MM. Pluronic L81 enhances triacylglycerol accumulation in the cytosol and inhibits chylomicron secretion. J. Lipid Res. 47: 2422-2432, 2006.
Febbraio M, Abumrad NA, Hajjar DP, Sharma K, Cheng W, Pearce SF, Silverstein RL. A null mutation in murine CD36 reveals an important role in fatty acid and lipoprotein metabolism. J. Biol. Chem. 274: 19055-19062, 1999.
Fire A, Xu S, Montgomery MK, Kostas SA, Driver SE, Mello CC. Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature 391: 806-811, 1998.
Fisher EA, Ginsberg HN. Complexity in the secretory pathway: the assembly and secretion of apolipoprotein B-containing lipoproteins. J. Biol. Chem. 277: 17377-17380, 2002.
Furuhashi M, Hotamisligil GS. Fatty acid-binding proteins: role in metabolic diseases and potential as drug targets. Nat. Rev. Drug Discov. 7: 489-503, 2008.
Gallo LL, Clark SB, Myers S, Vahouny GV. Cholesterol absorption in rat intestine: role of cholesterol esterase and acyl coenzyme A:cholesterol acyltransferase. J. Lipid Res. 25: 604-612, 1984.
Garcia-Calvo M, Lisnock J, Bull HG, Hawes BE, Burnett DA, Braun MP, Crona JH, Davis HR, Jr., Dean DC, Detmers PA, Graziano MP, Hughes M, Macintyre DE, Ogawa A, O'neill KA, Iyer SP, Shevell DE, Smith MM, Tang YS, Makarewicz AM, Ujjainwalla F, Altmann SW, Chapman KT, Thornberry NA. The target of ezetimibe is Niemann-Pick C1-Like 1 (NPC1L1). Proc. Natl. Acad. Sci. U. S. A 102: 8132-8137, 2005.
Ginsberg HN, Fisher EA. The ever-expanding role of degradation in the regulation of apolipoprotein B metabolism. J. Lipid Res. 50 Suppl: S162-S166, 2009.
76
Goldberg IJ, Eckel RH, Abumrad NA. Regulation of fatty acid uptake into tissues: lipoprotein lipase- and CD36-mediated pathways. J. Lipid Res. 50 Suppl: S86-S90, 2009.
Guilmeau S, Niot I, Laigneau JP, Devaud H, Petit V, Brousse N, Bouvier R, Ferkdadji L, Besmond C, Aggerbeck LP, Bado A, Samson-Bouma ME. Decreased expression of Intestinal I- and L-FABP levels in rare human genetic lipid malabsorption syndromes. Histochem. Cell Biol. 128: 115-123, 2007.
Haidari M, Leung N, Mahbub F, Uffelman KD, Kohen-Avramoglu R, Lewis GF, Adeli K. Fasting and postprandial overproduction of intestinally derived lipoproteins in an animal model of insulin resistance. Evidence that chronic fructose feeding in the hamster is accompanied by enhanced intestinal de novo lipogenesis and ApoB48-containing lipoprotein overproduction. J. Biol. Chem. 277: 31646-31655, 2002.
Hall AM, Wiczer BM, Herrmann T, Stremmel W, Bernlohr DA. Enzymatic properties of purified murine fatty acid transport protein 4 and analysis of acyl-CoA synthetase activities in tissues from FATP4 null mice. J. Biol. Chem. 280: 11948-11954, 2005.
Hamilton MT, Hamilton DG, Zderic TW. Role of low energy expenditure and sitting in obesity, metabolic syndrome, type 2 diabetes, and cardiovascular disease. Diabetes 56: 2655-2667, 2007.
Havel RJ. Remnant lipoproteins as therapeutic targets. Curr. Opin. Lipidol. 11: 615-620, 2000.
Herrmann T, van der HF, Grone HJ, Stewart AF, Langbein L, Kaiser I, Liebisch G, Gosch I, Buchkremer F, Drobnik W, Schmitz G, Stremmel W. Mice with targeted disruption of the fatty acid transport protein 4 (Fatp 4, Slc27a4) gene show features of lethal restrictive dermopathy. J. Cell Biol. 161: 1105-1115, 2003.
Higgins JA, Barrnett RJ. Fine structural localization of acyltransferases. The monoglyceride and -glycerophosphate pathways in intestinal absorptive cells. J. Cell Biol. 50: 102-120, 1971.
Hostetler HA, McIntosh AL, Atshaves BP, Storey SM, Payne HR, Kier AB, Schroeder F. L-FABP directly interacts with PPARalpha in cultured primary hepatocytes. J. Lipid Res. 50: 1663-1675, 2009.
HUBSCHER G, CLARK B. Metabolism of phospholipids. II. Isolation and properties of phosphatidic acid from mammalian liver. Biochim. Biophys. Acta 41: 45-54, 1960.
Hughes TE, Sasak WV, Ordovas JM, Forte TM, Lamon-Fava S, Schaefer EJ. A novel cell line (Caco-2) for the study of intestinal lipoprotein synthesis. J. Biol. Chem. 262: 3762-3767, 1987b.
Hughes TE, Sasak WV, Ordovas JM, Forte TM, Lamon-Fava S, Schaefer EJ. A novel cell line (Caco-2) for the study of intestinal lipoprotein synthesis. J. Biol. Chem. 262: 3762-3767, 1987a.
Hussain MM. A proposed model for the assembly of chylomicrons. Atherosclerosis 148: 1-15, 2000.
Hussain MM, Fatma S, Pan X, Iqbal J. Intestinal lipoprotein assembly. Curr. Opin. Lipidol. 16: 281-285, 2005.
77
Hussain MM, Shi J, Dreizen P. Microsomal triglyceride transfer protein and its role in apoB-lipoprotein assembly. J. Lipid Res. 44: 22-32, 2003.
Hussain MM, Zanni EE, Kelly M, Zannis VI. Synthesis, modification, and flotation properties of rat hepatocyte apolipoproteins. Biochim. Biophys. Acta 1001: 90-101, 1989.
Hussain MM, Zhao Y, Kancha RK, Blackhart BD, Yao Z. Characterization of recombinant human apoB-48-containing lipoproteins in rat hepatoma McA-RH7777 cells transfected with apoB-48 cDNA. Overexpression of apoB-48 decreases synthesis of endogenous apoB-100. Arterioscler. Thromb. Vasc. Biol. 15: 485-494, 1995.
Ibrahimi A, Abumrad NA. Role of CD36 in membrane transport of long-chain fatty acids. Curr. Opin. Clin. Nutr. Metab Care 5: 139-145, 2002.
Iseki S, Kondo H, Hitomi M, Ono T. Localization of liver fatty acid-binding protein and its mRNA in the liver and jejunum of rats: an immunohistochemical and in situ hybridization study. Mol. Cell Biochem. 98: 27-33, 1990.
Jones B, Jones EL, Bonney SA, Patel HN, Mensenkamp AR, Eichenbaum-Voline S, Rudling M, Myrdal U, Annesi G, Naik S, Meadows N, Quattrone A, Islam SA, Naoumova RP, Angelin B, Infante R, Levy E, Roy CC, Freemont PS, Scott J, Shoulders CC. Mutations in a Sar1 GTPase of COPII vesicles are associated with lipid absorption disorders. Nat. Genet. 34: 29-31, 2003.
Karpe F, Hultin M. Endogenous triglyceride-rich lipoproteins accumulate in rat plasma when competing with a chylomicron-like triglyceride emulsion for a common lipolytic pathway. J. Lipid Res. 36: 1557-1566, 1995.
Karpe F, Steiner G, Uffelman K, Olivecrona T, Hamsten A. Postprandial lipoproteins and progression of coronary atherosclerosis. Atherosclerosis 106: 83-97, 1994.
Kim CH, Yun SK, Byun DW, Yoo MH, Lee KU, Suh KI. Codon 54 polymorphism of the fatty acid binding protein 2 gene is associated with increased fat oxidation and hyperinsulinemia, but not with intestinal fatty acid absorption in Korean men. Metabolism 50: 473-476, 2001.
Klein BE, Klein R, Lee KE. Components of the metabolic syndrome and risk of cardiovascular disease and diabetes in Beaver Dam. Diabetes Care 25: 1790-1794, 2002.
Krauss RM. Lipids and lipoproteins in patients with type 2 diabetes. Diabetes Care 27: 1496-1504, 2004.
Kumar NS, Mansbach CM. Determinants of triacylglycerol transport from the endoplasmic reticulum to the Golgi in intestine. Am. J. Physiol 273: G18-G30, 1997.
Kumar NS, Mansbach CM. Prechylomicron transport vesicle: isolation and partial characterization. Am. J. Physiol 276: G378-G386, 1999.
Lally S, Owens D, Tomkin GH. Genes that affect cholesterol synthesis, cholesterol absorption, and chylomicron assembly: the relationship between the liver and intestine in control and streptozotosin diabetic rats. Metabolism 56: 430-438, 2007.
78
Latasa MJ, Griffin MJ, Moon YS, Kang C, Sul HS. Occupancy and function of the -150 sterol regulatory element and -65 E-box in nutritional regulation of the fatty acid synthase gene in living animals. Mol. Cell Biol. 23: 5896-5907, 2003.
Leng S, Lu S, Yao Y, Kan Z, Morris GS, Stair BR, Cherny MA, Black DD. Hepatocyte nuclear factor-4 mediates apolipoprotein A-IV transcriptional regulation by fatty acid in newborn swine enterocytes. Am. J. Physiol Gastrointest. Liver Physiol 293: G475-G483, 2007a.
Levin MS, Talkad VD, Gordon JI, Stenson WF. Trafficking of exogenous fatty acids within Caco-2 cells. J. Lipid Res. 33: 9-19, 1992.
Levy E, Mehran M, Seidman E. Caco-2 cells as a model for intestinal lipoprotein synthesis and secretion. FASEB J. 9: 626-635, 1995.
Lewis GF, Naples M, Uffelman K, Leung N, Szeto L, Adeli K. Intestinal lipoprotein production is stimulated by an acute elevation of plasma free fatty acids in the fasting state: studies in insulin-resistant and insulin-sensitized Syrian golden hamsters. Endocrinology 145: 5006-5012, 2004.
Liao W, Chan L. Apolipoprotein B, a paradigm for proteins regulated by intracellular degradation, does not undergo intracellular degradation in CaCo2 cells. J. Biol. Chem. 275: 3950-3956, 2000.
Lu S, Huffman M, Yao Y, Mansbach CM, Cheng X, Meng S, Black DD. Regulation of MTP expression in developing swine. J. Lipid Res. 43: 1303-1311, 2002.
Luchoomun J, Hussain MM. Assembly and secretion of chylomicrons by differentiated Caco-2 cells. Nascent triglycerides and preformed phospholipids are preferentially used for lipoprotein assembly. J. Biol. Chem. 274: 19565-19572, 1999.
Luchoomun J, Zhou Z, Bakillah A, Jamil H, Hussain MM. Assembly and secretion of VLDL in nondifferentiated Caco-2 cells stably transfected with human recombinant ApoB48 cDNA. Arterioscler. Thromb. Vasc. Biol. 17: 2955-2963, 1997.
Ma X, Bacci S, Mlynarski W, Gottardo L, Soccio T, Menzaghi C, Iori E, Lager RA, Shroff AR, Gervino EV, Nesto RW, Johnstone MT, Abumrad NA, Avogaro A, Trischitta V, Doria A. A common haplotype at the CD36 locus is associated with high free fatty acid levels and increased cardiovascular risk in Caucasians. Hum. Mol. Genet. 13: 2197-2205, 2004.
Malik S, Wong ND, Franklin SS, Kamath TV, L'Italien GJ, Pio JR, Williams GR. Impact of the metabolic syndrome on mortality from coronary heart disease, cardiovascular disease, and all causes in United States adults. Circulation 110: 1245-1250, 2004.
Man WC, Miyazaki M, Chu K, Ntambi J. Colocalization of SCD1 and DGAT2: implying preference for endogenous monounsaturated fatty acids in triglyceride synthesis. J. Lipid Res. 47: 1928-1939, 2006.
Mansbach CM, Gorelick F. Development and physiological regulation of intestinal lipid absorption. II. Dietary lipid absorption, complex lipid synthesis, and the intracellular packaging
79
and secretion of chylomicrons. Am. J. Physiol Gastrointest. Liver Physiol 293: G645-G650, 2007.
Mansbach CM, Nevin P. Intracellular movement of triacylglycerols in the intestine. J. Lipid Res. 39: 963-968, 1998.
Masuda D, Hirano KI, Oku H, Sandoval JC, Kawase R, Yuasa-Kawase M, Yamashita Y, Takada M, Tsubakio-Yamamoto K, Tochino Y, Koseki M, Matsuura F, Nishida M, Kawamoto T, Ishigami M, Hori M, Shimomura I, Yamashita S. Chylomicron remnants are increased in the postprandial state in CD36 deficiency. J. Lipid Res. 2008.
Matsukuma KE, Bennett MK, Huang J, Wang L, Gil G, Osborne TF. Coordinated control of bile acids and lipogenesis through FXR-dependent regulation of fatty acid synthase. J. Lipid Res. 47: 2754-2761, 2006.
McArthur MJ, Atshaves BP, Frolov A, Foxworth WD, Kier AB, Schroeder F. Cellular uptake and intracellular trafficking of long chain fatty acids. J. Lipid Res. 40: 1371-1383, 1999.
Mead JR, Irvine SA, Ramji DP. Lipoprotein lipase: structure, function, regulation, and role in disease. J. Mol. Med. 80: 753-769, 2002.
Meetoo D, McGovern P, Safadi R. An epidemiological overview of diabetes across the world. Br. J. Nurs. 16: 1002-1007, 2007.
Mero N, Syvanne M, Taskinen MR. Postprandial lipid metabolism in diabetes. Atherosclerosis 141 Suppl 1: S53-S55, 1998.
Miyaoka K, Kuwasako T, Hirano K, Nozaki S, Yamashita S, Matsuzawa Y. CD36 deficiency associated with insulin resistance. Lancet 357: 686-687, 2001.
Nassir F, Wilson B, Han X, Gross RW, Abumrad NA. CD36 is important for fatty acid and cholesterol uptake by the proximal but not distal intestine. J. Biol. Chem. 282: 19493-19501, 2007.
Nauli AM, Nassir F, Zheng S, Yang Q, Lo CM, Vonlehmden SB, Lee D, Jandacek RJ, Abumrad NA, Tso P. CD36 is important for chylomicron formation and secretion and may mediate cholesterol uptake in the proximal intestine. Gastroenterology 131: 1197-1207, 2006.
Neeli I, Siddiqi SA, Siddiqi S, Mahan J, Lagakos WS, Binas B, Gheyi T, Storch J, Mansbach CM. Liver fatty acid-binding protein initiates budding of pre-chylomicron transport vesicles from intestinal endoplasmic reticulum. J. Biol. Chem. 282: 17974-17984, 2007.
Newberry EP, Xie Y, Kennedy S, Han X, Buhman KK, Luo J, Gross RW, Davidson NO. Decreased hepatic triglyceride accumulation and altered fatty acid uptake in mice with deletion of the liver fatty acid-binding protein gene. J. Biol. Chem. 278: 51664-51672, 2003.
Pal S, Allister E, Thomson A, Mamo JC. Cholesterol esters regulate apoB48 secretion in CaCo2 cells. Atherosclerosis 161: 55-63, 2002.
Paulauskis JD, Sul HS. Hormonal regulation of mouse fatty acid synthase gene transcription in liver. J. Biol. Chem. 264: 574-577, 1989.
80
Pfaffl MW, Horgan GW, Dempfle L. Relative expression software tool (REST) for group-wise comparison and statistical analysis of relative expression results in real-time PCR. Nucleic Acids Res. 30: e36, 2002.
Poirier H, Degrace P, Niot I, Bernard A, Besnard P. Localization and regulation of the putative membrane fatty-acid transporter (FAT) in the small intestine. Comparison with fatty acid-binding proteins (FABP). Eur. J. Biochem. 238: 368-373, 1996.
Prows DR, Murphy EJ, Schroeder F. Intestinal and liver fatty acid binding proteins differentially affect fatty acid uptake and esterification in L-cells. Lipids 30: 907-910, 1995.
Qiu W, Taghibiglou C, Avramoglu RK, Van Iderstine SC, Naples M, Ashrafpour H, Mhapsekar S, Sato R, Adeli K. Oleate-mediated stimulation of microsomal triglyceride transfer protein (MTP) gene promoter: implications for hepatic MTP overexpression in insulin resistance. Biochemistry 44: 3041-3049, 2005.
Rivabene R, Napolitano M, Cantafora A, Bravo E. Redox-dependent modulation of lipid synthesis induced by oleic acid in the human intestinal epithelial cell line Caco-2. Exp. Biol. Med. (Maywood. ) 226: 191-198, 2001a.
Rivellese AA, De Natale C, Di Marino L, Patti L, Iovine C, Coppola S, Del Prato S, Riccardi G, Annuzzi G. Exogenous and endogenous postprandial lipid abnormalities in type 2 diabetic patients with optimal blood glucose control and optimal fasting triglyceride levels. J. Clin. Endocrinol. Metab 89: 2153-2159, 2004.
Rolf B, Oudenampsen-Kruger E, Borchers T, Faergeman NJ, Knudsen J, Lezius A, Spener F. Analysis of the ligand binding properties of recombinant bovine liver-type fatty acid binding protein. Biochim. Biophys. Acta 1259: 245-253, 1995.
Rozen S, Skaletsky H. Primer3 on the WWW for general users and for biologist programmers. Methods Mol. Biol. 132: 365-386, 2000.
Shim J, Moulson CL, Newberry EP, Lin MH, Xie Y, Kennedy SM, Miner JH, Davidson NO. Fatty acid transport protein 4 is dispensable for intestinal lipid absorption in mice. J. Lipid Res. 50: 491-500, 2009.
Shoulders CC, Brett DJ, Bayliss JD, Narcisi TM, Jarmuz A, Grantham TT, Leoni PR, Bhattacharya S, Pease RJ, Cullen PM, . Abetalipoproteinemia is caused by defects of the gene encoding the 97 kDa subunit of a microsomal triglyceride transfer protein. Hum. Mol. Genet. 2: 2109-2116, 1993.
Siddiqi SA, Siddiqi S, Mahan J, Peggs K, Gorelick FS, Mansbach CM. The identification of a novel endoplasmic reticulum to Golgi SNARE complex used by the prechylomicron transport vesicle. J. Biol. Chem. 281: 20974-20982, 2006.
Stahl A, Gimeno RE, Tartaglia LA, Lodish HF. Fatty acid transport proteins: a current view of a growing family. Trends Endocrinol. Metab 12: 266-273, 2001.
81
Stahl A, Hirsch DJ, Gimeno RE, Punreddy S, Ge P, Watson N, Patel S, Kotler M, Raimondi A, Tartaglia LA, Lodish HF. Identification of the major intestinal fatty acid transport protein. Mol. Cell 4: 299-308, 1999.
Stone SJ, Myers HM, Watkins SM, Brown BE, Feingold KR, Elias PM, Farese RV, Jr. Lipopenia and skin barrier abnormalities in DGAT2-deficient mice. J. Biol. Chem. 279: 11767-11776, 2004.
Taghibiglou C, Carpentier A, Van Iderstine SC, Chen B, Rudy D, Aiton A, Lewis GF, Adeli K. Mechanisms of hepatic very low density lipoprotein overproduction in insulin resistance. Evidence for enhanced lipoprotein assembly, reduced intracellular ApoB degradation, and increased microsomal triglyceride transfer protein in a fructose-fed hamster model. J. Biol. Chem. 275: 8416-8425, 2000.
Taskinen MR. Type 2 diabetes as a lipid disorder. Curr. Mol. Med. 5: 297-308, 2005.
Traber MG, Kayden HJ, Rindler MJ. Polarized secretion of newly synthesized lipoproteins by the Caco-2 human intestinal cell line. J. Lipid Res. 28: 1350-1363, 1987.
Tso P, Drake DS, Black DD, Sabesin SM. Evidence for separate pathways of chylomicron and very low-density lipoprotein assembly and transport by rat small intestine. Am. J. Physiol 247: G599-G610, 1984.
Tso P, Gollamudi SR. Pluronic L-81: a potent inhibitor of the transport of intestinal chylomicrons. Am. J. Physiol 247: G32-G36, 1984.
Weintraub MS, Grosskopf I, Rassin T, Miller H, Charach G, Rotmensch HH, Liron M, Rubinstein A, Iaina A. Clearance of chylomicron remnants in normolipidaemic patients with coronary artery disease: case control study over three years. BMJ 312: 935-939, 1996.
Wetterau JR, Aggerbeck LP, Bouma ME, Eisenberg C, Munck A, Hermier M, Schmitz J, Gay G, Rader DJ, Gregg RE. Absence of microsomal triglyceride transfer protein in individuals with abetalipoproteinemia. Science 258: 999-1001, 1992.
Williams KJ. Molecular processes that handle -- and mishandle -- dietary lipids. J. Clin. Invest 118: 3247-3259, 2008.
Wolfrum C, Buhlmann C, Rolf B, Borchers T, Spener F. Variation of liver-type fatty acid binding protein content in the human hepatoma cell line HepG2 by peroxisome proliferators and antisense RNA affects the rate of fatty acid uptake. Biochim. Biophys. Acta 1437: 194-201, 1999.
Wong DM, Webb JP, Malinowski PM, Macri J, Adeli K. Proteomic profiling of the prechylomicron transport vesicle involved in the assembly and secretion of apoB-48-containing chylomicrons in the intestinal enterocytes. Proteomics. 9: 3698-3711, 2009.
Worgall TS, Sturley SL, Seo T, Osborne TF, Deckelbaum RJ. Polyunsaturated fatty acids decrease expression of promoters with sterol regulatory elements by decreasing levels of mature sterol regulatory element-binding protein. J. Biol. Chem. 273: 25537-25540, 1998.
82
Yen CL, Stone SJ, Koliwad S, Harris C, Farese RV, Jr. Thematic review series: glycerolipids. DGAT enzymes and triacylglycerol biosynthesis. J. Lipid Res. 49: 2283-2301, 2008.
Zimmet P, Alberti KG, Shaw J. Global and societal implications of the diabetes epidemic. Nature 414: 782-787, 2001.