UNIVERSITY OF CINCINNATI
Date:______
I, ______, hereby submit this work as part of the requirements for the degree of: in:
It is entitled:
This work and its defense approved by:
Chair: ______
THE STRUCTURE AND FUNCTION OF
APOLIPOPROTEIN A-IV
By Kevin Joseph Pearson
B.S., University of Pittsburgh, 1999
May 2005
A dissertation presented to the faculty of the University of Cincinnati in partial fulfillment of the requirements for the degree of Doctor of Philosophy in
Pathobiology and Molecular Medicine
Committee Members
W. Sean Davidson-Chair
Patrick Tso
Ronald J. Jandacek
Min Liu
Simon L. Newman
Randall R. Sakai
THESIS ABSTRACT
Apolipoprotein (apo) A-IV is a protein synthesized by the small intestine in
response to lipid absorption. It has been proposed to play a role in cholesterol efflux,
lipoprotein metabolism and food intake. Unfortunately, little information on its
structure/function relationship is known. Therefore, it was important to establish a
recombinant expression system for apoA-IV so deletion mutagenesis could be performed
to achieve the specific aims.
The following aims address the hypothesis that specific regions of apoA-IV are
involved in its ability to interact with lipid and inhibit food intake.
Aim 1: Determine the region of apoA-IV that is responsible for its ability to bind lipid as well as to identify the generalities of its structure. Initially, it was found that removing the C-terminal 44 amino acids from human apoA-IV caused a significant increase in lipid binding ability as compared to WT. Eventually, a smaller region from amino acid 333-343 was established as the ‘inhibitory’ region of apoA-IV at the C- terminus. Using the ∆333-343 as a template, the N-terminus was also mutated and it was
found that the N-terminal amino acids from 11-20 were required for the mutants to be
fast lipid binders. Therefore, we propose there is an interaction between the ‘inhibitory’
region from 333-343 and the N-terminus of the protein that does not allow the proper
conformation for accelerated lipid binding.
Aim 2: Determine the region of apoA-IV that is responsible for its role in the
inhibition of food intake. Originally, several C-terminal deletion mutants were made in
rat apoA-IV. The mutants were injected into the 3rd ventricle of rat brains and their food intake was determined. It was found that the food intake region was located in the N- terminal 116 amino acids. Next a 3rd mutant was made that removed the N-terminal 61 amino acids. This mutant did not inhibit food intake as compared to WT suggesting the region involved in food intake is located in the first 61 amino acids. In separate studies, a
14-residue peptide from amino acid 17-30 also inhibited food intake in rats suggesting this is the active region.
ACKNOWLEDGEMENTS
This work was supported by a pre-doctoral fellowship to K. Pearson from the American
Heart Association, Southern and Ohio Valley Affiliate. I would like to thank Patrick Tso for funding several years of this project.
I would also like to thank my advisor, W. Sean Davidson, for his constant support
throughout this long and arduous journey. He has taught me how to become a successful
scientist in many ways. He also made a pretty good target for me to fire at (or mostly
complain to) when I was stressed or frustrated. When I leave UC, I will leave having had
a great thesis advisor, but also a close friend.
As for the rest of the lab…thanks - only one personal shout-out and he probably won’t
ever read this…Nick, you’re the dude and thanks for helping out a scared kid from
Pennsylvania when he didn’t know anything! Outside the lab…Thanks mom and dad for helping out so I could be a student until I was 27 (the car/insurance)…Skot, Jake, Wael,
Kara and Kimberly – Those first years could have been so much harder…1214 McKeone
- enough said…and finally three things I couldn’t have made it without (I know you’re thinking my girlfriend here but that’s too easy) 1. Cherry, my 1992 Dodge Spirit. 2.
Pretty blue bullet, my 1997 Chevrolet Cavalier. 3. Cincinnati style chili and currently
Subway.
TABLE OF CONTENTS
LIST OF TABLES 7
LIST OF FIGURES 8
ABBREVIATIONS 12
CHAPTER 1: GENERAL INTRODUCTION 13 OVERVIEW OF APOA-IV METABOLISM 13
GENETIC ORGANIZATION OF APOA-IV 15
PRIMARY SEQUENCE OF APOA-IV 16
SECONDARY STRUCTURE OF APOA-IV 17
TERTIARY STRUCTURE OF APOA-IV 18
LIPID ASSOCIATED FORMS OF APOA-IV 19
ROLE IN LIPOPROTEIN METABOLISM 20
ROLE OF APOA-IV IN INFLAMMATION 23
ROLE OF APOA-IV IN FOOD INTAKE 24
GENETIC MANIPULATIONS 25
THE AIMS OF THIS THESIS 27
THE ORGANIZATION OF THIS THESIS 28
REFERENCES 30
FIGURES 39
CHAPTER 2: BACTERIAL EXPRESSION AND CHARACTERIZ- ATION OF RAT APOLIPOPROTEIN A-IV AND APOLIPOPROTEIN E 41 ABSTRACT/SUMMARY 41
INTRODUCTION 42
EXPERIMENTAL PROCEDURES 44
Construction of the expression vector for apoA-IV 45
Cloning and mutagenesis of rat apoE 46
Protein expression and purification 47
SDS-PAGE and immunoblot 49
Electrospray mass spectrometry 50
N-terminal amino acid sequencing 50
Lipoprotein particle reconstitution 50
Biological activity of recombinant apoA-IV 51
Conditioned taste aversion 52
Native protein purification 53
Western blot analysis for apoE 53
Circular dichroism spectroscopy 54
DMPC liposome solubilization 54
Cholesterol efflux assay 54
RESULTS: RAT APOA-IV 55
DISCUSSION: RAT APOA-IV 57
2 RESULTS: RAT APOE 59
DISCUSSION: RAT APOE 62
REFERENCES 63
TABLES 67
FIGURES 68
CHAPTER 3: THE STRUCTURE OF HUMAN APOLIPOPROTEIN A-IV: A DISTINCT DOMAIN ARCHITECTURE AMONG EXCHANGEABLE APOLIPOPROTEINS WITH POTENTIAL FUNCTIONAL IMPLICATIONS 81 ABSTRACT 81
INTRODUCTION 82
EXPERIMENTAL PROCEDURES 86
Cloning and mutagenesis of human apoA-IV 86
Protein expression and purification 87
Circular dichroism spectroscopy 88
Fluorescence measurements 88
Self-association determination 89
DMPC liposome solubilization 89
Cholesterol efflux assay 90
RESULTS 90
DISCUSSION 97
3 REFERENCES 103
TABLES 106
FIGURES 108
CHAPTER 4: A POTENTIAL INTERACTION BETWEEN THE N- AND C-TERMINI OF APOLIPOPROTEIN A-IV AND ITS ROLE IN LIPID ASSOCIATION 117 ABSTRACT 117
INTRODUCTION 117
EXPERIMENTAL PROCEDURES 119
Mutagenesis of human apoA-IV 120
Protein expression and purification 120
DMPC liposome solubilization 121
Interfacial Behavior at the Oil/water Interface 122
Thermal denaturation and fluor. spectroscopy 122
RESULTS 123
DISCUSSION 127
REFERENCES 132
TABLES 135
FIGURES 137
4 CHAPTER 5: IDENTIFICATION OF AN APOLIPOPROTEIN A-IV SEQUENCE RESPONSIBLE FOR CENTRAL INHIBITION OF FOOD INTAKE 144 ABSTRACT 144
INTRODUCTION 145
EXPERIMENTAL PROCEDURES 147
Cloning and purification of rat apoA-IV mutants 148
Removal of LPS 149
Purification of native apoA-IV 150
Peptide synthesis 150
Food intake studies 151
BBB studies 152
RESULTS 153
DISCUSSION 157
REFERENCES 161
FIGURES 164
CHAPTER 6: GENERAL DISCUSSION 174 SUMMARY: AIM 1 174
DISCUSSION: AIM 1 175
FUTURE STUDIES: AIM 1 176
SUMMARY: AIM 2 177
5 DISCUSSION: AIM 2 178
FUTURE STUDIES: AIM 2 179
FINAL THOUGHTS 179
REFERENCES 181
6 TABLES:
CHAPTER 2: BACTERIAL EXPRESSION AND CHARACTERIZ- ATION OF RAT APOLIPOPROTEIN A-IV AND APOLIPOPROTEIN E
Table 2-1: Purification of recombinant rat apolipoprotein E. 67
CHAPTER 3: THE STRUCTURE OF HUMAN APOLIPOPROTEIN A-IV: A DISTINCT DOMAIN ARCHITECTURE AMONG EXCHANGEABLE APOLIPOPROTEINS WITH POTENTIAL FUNCTIONAL IMPLICATIONS
Table 3-1: Conformation and stability properties of WT apoA-IV and the various deletion mutants. 106
Table 3-2: Kinetic parameters for cholesterol efflux from RAW264 macrophages to lipid-free apoA-IV and its deletion mutants. 107
CHAPTER 4: A POTENTIAL INTERACTION BETWEEN THE N- AND C-TERMINI OF APOLIPOPROTEIN A-IV AND ITS ROLE IN LIPID ASSOCIATION
Table 4-1. Rate constants for WT and mutant apoA-IV in the DMPC clearance assay. 135
Table 4-2. Fluorescence and thermal denaturation parameters of WT and mutant apoA-IV 136
7 FIGURES:
CHAPTER 1: GENERAL INTRODUCTION
Figure 1-1: The primary sequence of human apoA-IV. 39
Figure 1-2: A general schematic of lipid metabolism and the potential roles of apoA-IV. 40
CHAPTER 2: BACTERIAL EXPRESSION AND CHARACTERIZ- ATION OF RAT APOLIPOPROTEIN A-IV AND APOLIPOPROTEIN E
Figure 2-1: Schematic representation of the pET 30 vector. 68
Figure 2-2: SDS polyacrylamide gel of recombinant apoA-IV produced by E.Coli at different steps of the purification process. 69
Figure 2-3: Immunoblot analysis of recombinant rat apoA-IV expressed in E. coli. 70
Figure 2-4: Electrospray mass spectrum of recombinant apoA-IV. 71
Figure 2-5: Native gradient polyacrylamide gel of reconstituted particles made with recombinant apoA-IV. 72
Figure 2-6: Mean food intake following i3vt recombinant rat apoA-IV and natural rat apoA-IV. 73
Figure 2-7: Conditioned taste aversion. 74
Figure 2-8: Forward oligonucleotide primer sequence. 75
Figure 2-9: SDS-PAGE analysis of recombinant rat apoE at different stages of purification and western blot analysis following purification. 76
Figure 2-10: SDS-PAGE analysis of rat apoE under different storage conditions. 77
8 Figure 2-11: Far UV circular dichroism spectra of apoE. 78
Figure 2-12. Dimyristoylphosphatidylcholine (DMPC) liposome solubilization by apoE. 79
Figure 2-13: ABCA1-mediated cholesterol efflux from RAW264 macrophages. 80
CHAPTER 3: THE STRUCTURE OF HUMAN APOLIPOPROTEIN A-IV: A DISTINCT DOMAIN ARCHITECTURE AMONG EXCHANGEABLE APOLIPOPROTEINS WITH POTENTIAL FUNCTIONAL IMPLICATIONS
Figure 3-1: Linear diagrams of WT apoA-IV and truncation mutants. 108
Figure 3-2: SDS-PAGE analysis of WT apoA-IV and representative mutant forms. 109
Figure 3-3: Thermal unfolding of N- and C-terminal mutants of apoA-IV. 110
Figure 3-4: Thermal unfolding of combined N- and C-terminal mutants. 111
Figure 3-5: Effects of apoA-I, WT apoA-IV and mutant forms of apoA-IV on ANS fluorescence. 112
Figure 3-6: SDS-PAGE analysis of the self-association of WT apoA-IV and representative mutant forms. 113
Figure 3-7: Dimyristoylphosphatidylcholine (DMPC) liposome solubilization by apoA-I, apoA-IV and deletion mutants of apoA-IV. 114
Figure 3-8: ABCA1-mediated cholesterol efflux from RAW264 macrophages. 115
Figure 3-9: A generalized model for the effects of deletions on the structure and function of lipid-free apoA-IV. 116
9
CHAPTER 4: A POTENTIAL INTERACTION BETWEEN THE N- AND C-TERMINI OF APOLIPOPROTEIN A-IV AND ITS ROLE IN LIPID ASSOCIATION
Figure 4-1. Schematic map of human apoA-IV deletion mutants used in this study. 137
Figure 4-2. SDS-PAGE and native-PAGE analysis of WT apoA-IV and selected truncation mutants. 138
Figure 4-3. Lipid association of WT apoA-IV and apoA-IV ∆333-376. 139
Figure 4-4. Lipid association of WT apoA-IV and apoA-IV ∆344-376 and ∆354-367. 140
Figure 4-5. Sequence alignment of the putative “inhibitory” region from amino acid 333 to 343 and the lipid binding performance of resulting mutants. 141
Figure 4-6. DMPC liposome solubilization by WT apoA-IV and N-terminal mutants also lacking the inhibitory region from 333-343. 142
Figure 4-7. Steady-state Stern-Volmer plot of WT apoA-IV and deletion mutants. 143
CHAPTER 5: IDENTIFICATION OF AN APOLIPOPROTEIN A-IV SEQUENCE RESPONSIBLE FOR CENTRAL INHIBITION OF FOOD INTAKE
Figure 5-1: Linear diagram of WT rat apoA-IV and recombinant deletion mutants. 164
Figure 5-2: Food intake measurements in i3vt cannulated rats. 165
Figure 5-3: The effect of residual LPS on suppression of rat food
10 intake by recombinant apoA-IV. 166
Figure 5-4: SDS-PAGE analysis of WT rat apoA-IV and recombinant deletion mutants. 167
Figure 5-5: Effect of C-terminal truncation mutants of apoA-IV on rat food intake when injected centrally. 168
Figure 5-6: Effect of an N-terminal deletion of apoA-IV on rat food intake when injected centrally. 169
Figure 5-7: Multi-species sequence alignment of the N-terminal 61 amino acids of apoA-IV. 170
Figure 5-8: Effect of N-terminal and control synthetic peptides of rat apoA-IV on food intake. 171
Figure 5-9: Leptin has the ability to cross the BBB. 172
Figure 5-10: ApoA-IV does not appear to cross the BBB. 173
11 ABBREVIATIONS
A600, absorbance at 600 nm; ABCA1, ATP binding cassette transporter A1; ANS, 8- anilino-1-naphthalenesulfonic acid; apo, apolipoprotein; BSA, bovine serum albumin;
BS3, Bis[sulfosuccinimidyl] suberate; CETP, cholesteryl ester transfer protein; CD,
circular dichroism; CTA, conditioned taste aversion; DMEM, Dulbecco’s modified Eagle
medium; DMPC, dimyristoyl phosphotidylcholine; DMSO, dimethylsulfoxide; DSS,
dextran sulfate sodium; HDL, high density lipoprotein; i3vt, 3rd intra-cerebroventricle;
IPTG, isopropyl-β-D-thiogalactoside; LCAT, lecithin: cholesteryl acyl transferase; LPL, lipoprotein lipase; LPS, lipopolysaccharide; LDL, low density lipoprotein; LB, Luria-
Bertani; STB, standard Tris buffer; TG, triglyceride; λ-max, wavelength of maximum fluorescence; WT, wild type; VLDL, very low density lipoprotein
12 CHAPTER 1:
GENERAL INTRODUCTION
Apolipoprotein A-IV was discovered in 1974 as a constituent of rat plasma high density lipoprotein (HDL) (1) and was found to be a member of the family of related exchangeable apolipoproteins that include apoA-I and apoE (2). It is synthesized by the small intestine in response to lipid absorption and the formation of chylomicrons (3;4).
Despite its responsiveness to lipid intake, an unambiguous function for apoA-IV has not yet been widely recognized. It has been proposed to play many functions in vivo including: food intake regulation (5;6), gastrointestinal motility (7), structural constituent of lipoproteins (8), protection against lipid oxidation and atherosclerosis (9), and can also mimic many of the roles of apoA-I in terms of lipid binding and cholesterol efflux (10-
13).
Overview of apoA-IV metabolism: ApoA-IV is synthesized by enterocytes in the small intestine of humans (4). Its production by the small intestine, especially the jejunum, is increased in response to lipid absorption and the formation of chylomicrons.
Also, the ileum produces peptide tyrosine-tyrosine (PYY) in response to lipid exposure and this peptide can act as a signal for the jejunum to increase apoA-IV synthesis (14).
Unlike in humans, rodent hepatocytes are also capable of synthesizing apoA-IV, but the liver only contains about one-tenth of the mRNA as compared to the small intestine (15).
Thyroid hormone is one of the factors responsible for stimulating rat hepatocytes to
13 produce apoA-IV (16;17). However, it is not known whether there is a difference in function between apoA-IV produced by the liver or the small intestine.
After production by the enterocytes, apoA-IV is packaged at the surface of chylomicrons and secreted into the lymph (4). The lymph empties into the bloodstream through the thoracic duct into the left subclavian vein (18). Evidence has shown that apoA-IV dissociates from chylomicrons upon entering the blood and this is probably due
to displacement by the apoC proteins (19). However, there is conflicting data on what
proportion of apoA-IV is present in the lipid-free or lipoprotein fraction. Earlier studies
showed that human apoA-IV was mostly found in the lipid-free fraction.
Ultracentrifugation techniques have shown as much as 98% can be found in the lipid free
fraction (4;20). However, more recent data suggests that more than 70-90% of apoA-IV
can be found associated with lipid using affinity chromatography or immunoprecipitation
(21-23). The lipid-bound fraction is thought to be made up of two major forms, apoA-IV
that is associated with apoA-I in high density lipoproteins and apoA-IV that is not
associated with apoA-I (22-25). The discrepancy in the lipid association of apoA-IV in
plasma is probably due to different separation techniques. The majority of lipid-free
apoA-IV could be an artifact of harsh separation conditions. Indeed, apoA-IV is thought
to be the most hydrophilic of the apolipoproteins (26;27); therefore, any extreme
conditions could cause its disassociation from lipids. In contrast to human apoA-IV, it
has been consistently shown that at least half of rat apoA-IV is found associated with
various apolipoproteins (20;28;29). Interestingly, Dallinga-Thie et al. have shown that
lipid associated apoA-IV levels do not change when compared between fed and fasted
animals. However, the lipid-free apoA-IV level is significantly decreased in the fasted
14 state suggesting there is a potential role for lipid-free apoA-IV as well (20). Weinberg
has suggested that rat apoA-IV is more hydrophobic than human apoA-IV and may cause rat apoA-IV to remain associated with lipid more frequently following separation techniques (30).
ApoA-IV has a rapid catabolic rate as compared to other apolipoproteins as its half-life is between 6-18 h (31-33). Ghiselli et al. have also shown that radiolabeled apoA-IV associated with HDL has a longer retention time than radiolabeled lipid-free apoA-IV (34). In either case, the liver is the major catabolic site, but the kidney has some function as well (32;34;35). Indeed, those patients with chronic renal failure have been shown to have up to three times the level of apoA-IV as compared to individuals with normal kidney function (35).
Genetic organization of apoA-IV: The apoA-IV gene is located in the same 15- kilobase gene complex as apoA-I and apoC-III on the long arm of chromosome 11 in humans and on chromosome 9 in rats (36-40). The genes for apoA-IV and apoA-I are transcribed in the same direction but the apoC-III gene is transcribed in the opposite direction (41). There are several control elements for the apoA-IV gene although they have not been studied thoroughly. There is a proximal promoter region sensitive to intestinal fatty acids located in the 5’ nontranscribed region of the gene (15). A distal promoter also exists and acts as regulatory region (42). In addition, an apoC-III/A-IV mutual enhancer is located upstream of both genes that contains a hormone-response element (43;44). Several mutations in the apoA-I/apoC-III/apoA-IV gene cluster have been reported and can lead to premature atherosclerosis (45-47).
15 The apoA-IV gene consists of three exons separated by two introns rather than
four exons and three introns like the rest of the apolipoprotein family (37). Because of
this, it is thought that intraexonic duplication of the apoA-I gene about 300 million years
ago probably led to the formation of the apoA-IV gene. Further evidence is that the
apoA-IV gene is similar to apoA-I and apoE whereas each consists of multiple 33 and 66
nucleotide repeats (37;39;40).
The human apoA-IV coding region consists of 1188 nucleotides that encode a 396
amino acid pre-protein (15). The rat's coding region is made up of 1173 nucleotides that
translate a 391 amino acid pre-protein (48). ApoA-IV is produced in both rats and
humans with a 20 amino acid pre-sequence that is thought to aid in its packaging in the
endoplasmic reticulum. This signal sequence is proteolytically cleaved during co-
translational translocation across the ER membrane. After this cleavage, the amino
terminal sequence is identical to that of apoA-IV found in plasma and lymph (49;50).
Exon 1 of apoA-IV encodes most of the signal sequence. Meanwhile, exon 2 encodes the
rest of the signal sequence and the first 39 amino acids that are predicted to form a class
G amphipathic helix. Finally, exon 3 encodes the 11 and 22 amino acid repeats that form class A and class Y amphipathic helices as will be discussed in more detail below (15).
Primary sequence of apoA-IV: The mature, secreted forms of human and rat apoA-IV are 376 and 371 amino acids in length, with corresponding molecular weights of
46 and 43 kDa, respectively (48;51). One main difference in molecular weight is that
human apoA-IV is glycosylated and 6% of its weight is carbohydrate (mannose,
galactose, N-acetyl glucosamine and sialic acid), but rat apoA-IV is not thought to be
glycosylated (52). Unfortunately, there has not been any research on the purpose of its
16 glycosylation but, as in other proteins; it probably has some signaling potential. Rat and
human apoA-IV primary sequence is generally well conserved at greater than 60%
identity and the only discrepancy in length is seen near the C-terminus where 5
consecutive amino acids have been deleted in the rat (2;15). Figure 1-1 shows the
primary sequence of human apoA-IV. In both rat and human apoA-IV, there is a region
from 354-367 that has a series of Glu-Gln-(Gln/Val)-Gln repeats that have not been
found in any other apolipoprotein (53).
ApoA-IV secondary structure: The primary and secondary structure of apoA-IV shares several features with other exchangeable apolipoproteins such as apoA-I and apoE.
All three have multiple sets of an 11 or 22-amino acid repeat throughout their primary sequence and this repeat is predicted to form amphipathic alpha helices (2;48).
Amphipathic helices have both a hydrophobic and hydrophilic region around a central axis (or a polar/nonpolar interface) and are thought to give apolipoproteins the ability to bind to lipids. As stated above, apoA-IV contains a class G helix at its N-terminus and many class A and class Y helices throughout the remainder of the protein. Class G helices are typically found in globular proteins and they have positive and negative charges scattered on the polar interface. A class A helix has positively charged residues clustered near the polar-nonpolar interface and negatively charged amino acids in the center of the polar face. Finally, a class Y helix is similar to a class A helix, but it also has a positive charge inside of the negative charges creating a Y shape among the positive charges (54-56). Amphipathic helices are a common feature of apolipoproteins and give them the ability to interact with lipids. However, they all do not interact with lipids with the same affinity. ApoA-IV has the most class Y helices of the
17 apolipoproteins and this is one reason that apoA-IV is considered the most hydrophilic of all the apolipoproteins (30;57).
The C-terminal region of apoA-IV from amino acids 333-376 in the human and
333-371 of the rat are not predicted to form these amphipathic helices. Secondary structure was predicted to be 81% helical and 17% random coil using the Garnier program on the World Wide Web on www.expasy.org (58). Using Chou-Fasman analysis, Weinberg has shown there are multiple helical regions separated by short random coil and beta-turns segments. This analysis has also shown there are short beta- sheet segments from amino acid 6-15 and 331-336 (30).
The secondary structure of apoA-IV is likely dominated by alpha helices because of its putative 22 amino acid amphipathic repeats. Its helical content has been found to be 35-55% by circular dichroism (57;59). However, the protein is easily denatured in the presence of minimal (less than 0.5 M) guanidine hydrochloride. ApoA-IV has a much lower free energy of denaturation (0.2 Kcal/mol) as compared to apoA-I (2.4-3.7
Kcal/mol) or apoE (12 Kcal/mol) (60-62). This suggests that apoA-IV is more sensitive to chemical denaturation and therefore, is less stable. It is thought that apoA-IV may form a ‘molten’ globule which has its secondary structure loosely organized but can also move fluidly with its amphipathic helices facing the inside of the molecule (57;63;64).
Tertiary structure of apoA-IV: Despite many years of study, there is not a complete 3-dimensional structure of the human exchangeable apolipoproteins. Based on an X-ray crystal structure of an N-terminal fragment of apoE, it has been suggested that the N-terminal 22 kDa of apoE forms a helical bundle that is relatively stable in solution
(65). The remaining C-terminal domain exhibits a less defined structure and is thought to
18 be responsible for lipid binding (66). Without the benefit of a monomeric, lipid-free crystal structure of apoA-I, several studies have suggested a similar domain organization for apoA-I (60;67;68). This suggests that apoA-IV may have a similar overall structure.
In addition to what is known about other apolipoproteins, Weinberg et al. have suggested that apoA-IV is a relatively unstable ‘molten’ globule that is found as both monomer and dimer in vivo (57;69). It is thought that the globule formation may be made up of helical bundles (57;63;64) and the C-terminus is important in the tertiary structure of the protein.
Removing the C-terminus caused the Trp at position 12 to be more exposed than in the
WT protein (70). There is very little information on the structure of human apoA-IV, but the studies in Chapters 2 and 3 will directly address this deficiency.
Lipid associated forms of apoA-IV: As stated previously, amphipathic helical repeats give apoA-IV the ability to interact with lipid. It has been shown that its interaction with lipid increases its helicity to greater than 80% (26;57;71). This is closer to the predicted helicity for apoA-IV by computer analysis. Interestingly, apoA-IV is more resistant to chemical denaturation when associated with lipid. It can remain stable in the presence of guanidine HCl up to 2 M as opposed to less than 0.5 M guanidine HCl in the lipid-free form (26;57;71). Also, its free energy of denaturation is also increased from 0.2 to 6.3 Kcal/mol in the presence of phospholipid (26). This suggests that apoA-
IV undergoes a conformation change when bound to lipid that is more helical but also more stable. Reconstituted particles have also been made with apoA-IV and synthetic phospholipids in vitro and were found to be discoidal. These particles are similar to nascent high density lipoproteins (HDL) found in vivo before they are acted on by the
19 lecithin: cholesteryl acyl transferase (LCAT) enzyme as discussed in more detail below
(71). Jonas et al. also made reconstituted lipoprotein particles and found that 2
apolipoproteins are found at the surface of each particle. The particles also form two
distinct sizes with a diameter of either 116 or 145 D. These particles are larger than those
particles made with apoA-I or apoE. The difference is not due to a difference in mass,
but they predict it is due to the number of amphipathic helices in each (72). There is little
information as to how apoA-IV binds to lipids and what makes it stay in contact with
lipids. This problem is addressed in Chapters 2 and 3.
Role of apoA-IV in lipoprotein metabolism: Figure 1-2 is a general schematic of
lipid metabolism. After eating a lipid-filled meal, the small intestine takes up lipid and
forms chylomicrons that enter into the lymphatics and eventually the plasma. ApoA-IV
is found at the surface of the chylomicron particles and upon reaching the plasma is replaced by the apoC proteins. It is unclear, but it is thought that these proteins may work together to activate lipoprotein lipase (LPL) (19;73). LPL hydrolyzes triglycerides and releases fatty acids into adipose tissue and the resulting chylomicron remnants are taken up by the liver. The liver then packages the endogenous lipids into very low density lipoproteins (VLDL) and LPL hydrolyzes the triglycerides and releases fatty acids to adipose and muscle tissues (74). Low density lipoproteins (LDL) result from this hydrolysis and have extremely high cholesterol content. The role of LDL is to deliver cholesterol to peripheral tissues. However, if the LDL is not taken up quickly, it can become oxidized and then scavenged by macrophages. Macrophages can form foam cells if they take up excess oxidized LDL and this can be a precursor to atherosclerosis.
20 ApoA-IV is thought to act as an antioxidant which can decrease the amount of oxidized
LDL in the periphery (9;75).
Macrophages can decrease their cholesterol content by several mechanisms including ATP-binding cassette transporter (ABCA1)-mediated cholesterol efflux, by simple diffusion or by decreasing production of cholesterol. Nascent HDL or lipid poor apolipoproteins such as apoA-IV are loaded with cholesterol and phospholipid by the
ABCA1 transporter. The cholesterol is then esterified by the LCAT enzyme and can then be packaged into the center of the mature HDL particle (76-80). Mature HDL particles are more spherical than the nascent HDL because cholesteryl ester has been packaged into the core of the particle. ApoA-IV has been shown to activate LCAT in vitro (81;82).
Eventually, HDL can return cholesterol to the liver for catabolism or it can be transferred to LDL or VLDL by cholesteryl ester transfer protein (CETP). ApoA-IV has also been shown to activate CETP (83;84).
It is known that apoA-IV is found at the surface of chylomicrons and is important in their formation. Lu et al. have shown overexpression of apoA-IV in newborn swine intestinal epithelial cells causes increased secretion of lipoproteins including chylomicrons (85). Also, Weinberg et al. have suggested the presence of apoA-IV at the surface of chylomicrons increases their size (27). Quarfordt et al. have shown large chylomicrons were cleared more rapidly than small chylomicrons if equal masses of lipid were injected into rats (86). In contrast, Martins et al. showed that particle number was more important than particle size for clearance rate (87). Therefore, the effect on chylomicron size is important because if the chylomicrons are smaller, a greater number
21 of particles will be produced in order to carry the same amount of lipid and this could potentially be more atherogenic (88).
One of the prerequisites for activating the LCAT enzyme is the presence of amphipathic helices (89). As discussed previously, apoA-IV is made up of 12 putative amphipathic helices. It has been shown using deletion mutagenesis that the region from
117-160 is required for LCAT activation. Removal of this region decreases LCAT activity by at least 75% (90). Using the same set of mutants, it was also shown that no particular region of apoA-IV is required for activation of CETP, but instead, the water/phospholipid interfacial exclusion pressure of the mutants directly correlated to the rate constant of cholesterol ester transfer (91). Therefore, the higher the exclusion pressure, the more cholesterol ester is transferred.
ApoA-IV has been shown to promote cholesterol efflux via the ABCA1-mediated pathway. ApoA-IV, together with phosphotidylcholine in a liposome or in serum lipoproteins, is an efficient cholesterol acceptor from fibroblasts or from mouse adipose cells (12). In addition, an increase in the presence of apoA-IV in transgenic animals has shown similar results. Serum, collected from mice with human apoA-IV overexpression, induces cAMP-stimulated cholesterol efflux from mouse macrophages and has increased cholesterol efflux as compared to WT serum (92). The presence of cAMP increases
ABCA1 in the macrophage membrane and results in an increase of apolipoprotein- mediated cholesterol efflux (93;94).
Qin et al. first proposed apoA-IV might act as an antioxidant when they showed that apoA-IV protected LDL from copper-induced oxidation (75). Unlike humans, mice do not readily get atherosclerosis; therefore, apoE knockout mice are generally used
22 instead because they do get atherosclerotic lesions. ApoE knockout mice overexpressing
human apoA-IV had decreased lesion size as compared to the apoE knockout alone.
However, there was no change in the amount of HDL in these animals even though a
correlation had been shown such that increased HDL leads to decreased atherosclerosis in
humans and other animal models (9). Therefore, another possibility for this
atherosclerotic protection could be the role of apoA-IV as an antioxidant. Indeed, Ostos
et al. showed that the human apoA-IV transgenic on an apoE knockout background had
reduced oxidation parameters as compared to the apoE knockout alone. They showed
both anti-oxidized LDL antibodies and the aggregation state of LDL were decreased in
the apoA-IV transgenic animals (9).
Role of apoA-IV in inflammation: More recently, apoA-IV has been suggested to play
an anti-inflammatory role in vivo. In the same model discussed above, apoE knockout mice overexpressing human apoA-IV had decreased lesion size as compared to the apoE knockout alone even though their HDL profile was similar. As discussed, there was a difference in the antioxidant capabilities between the two groups of mice. Another difference was that when the knockout animals were treated with lipopolysaccharide
(LPS), their lesion size increased. However, the apoA-IV transgenic, apoE knockout had a protective role in the presence of LPS. One reason for this was that there was a decrease in the proinflammatory cytokines including IL-4, INF-gamma and TNF-α.
Studies were also performed in vitro showing that recombinant apoA-IV also led to a decrease in cytokine production by monocytes suggesting at least one possible mechanism (95).
23 Another study looking at its anti-inflammatory properties was recently published by Vowinkel et al (96). In this study, they used a mouse model giving the mice 3% dextran sulfate sodium (DSS) in their diet. This induces colitis which is a type of inflammation with similar characteristics to human irritable bowel disease such as weight loss, diarrhea, mucosal damage, cytokine production and leukocyte infiltration. These symptoms were more pronounced in the apoA-IV knockout as compared to the WT.
Also, giving recombinant apoA-IV intraperitoneally to both the KO and WT mice
(treated with DSS), significantly decreased their inflammatory response and the overall morphology of the intestine was similar to that of normal WT mice (not treated with
DSS) (96). Taken as a whole, these studies suggest that apoA-IV may indeed act as an anti-inflammatory agent in vivo.
Role of apoA-IV in food intake: Another hypothesis about apoA-IV function is that it inhibits food intake and may act as a satiety factor. This function is unlike all of the other proposed functions for apoA-IV as it is not shared by additional apolipoproteins; instead, this one is unique to apoA-IV. As stated above, apoA-IV synthesis is stimulated by fat absorption in both the human and rat small intestine and is secreted into the lymph at the surface of chylomicrons. Intravenous injection of rat lymph from fed rats into unfed animals significantly inhibited their food intake (97). This finding, along with the fact that lipid absorption causes a two to three-fold increase of apoA-IV but not other apolipoproteins (3), led to the proposal that apoA-IV controls food intake. Indeed, apoA-
IV antibodies were able to decrease this effect on food intake. Also, intravenous
24 injection of purified apoA-IV but not apoA-I significantly inhibited food intake (97).
These peripheral injections all had an effect on food intake.
The hypothalamus is a likely target for this inhibitory action by apoA-IV because
it has been shown to be involved in the regulation of food intake (98-100). In addition,
much smaller amounts of native rat apoA-IV infused into the 3rd cerebral ventricle
inhibited food intake in a dose dependent manner (6). For this reason, the role of apoA-
IV in food intake is thought to be mediated centrally. Liu et al. have shown that apoA-IV
mRNA is present in the hypothalamus and is increased in response to lipid feeding
(101;102). Therefore, it appears that some sort of signal is transferred between the gut
(sensing lipid and producing apoA-IV) and the brain producing apoA-IV as well and may
act as a satiety signal. In addition, apoA-IV has also been shown to inhibit gastric
emptying and gastric acid secretion centrally (7;103;104). This could be a second
potential mechanism for decreased food intake. At this point, it is unknown whether
apoA-IV crosses the blood-brain barrier or if it is acting through the nervous system for its behavior as a satiety signal. There is some evidence that it may be through the nervous system as it has been shown recently that apoA-IV stimulates vagal afferent activity in order to inhibit gastric motility (105).
Genetic manipulations. Whether intentional or unintentional: Data exists using both transgenic and apoA-IV KO mice in order to further define its role in vivo. Their role has already been discussed for cholesterol efflux, antioxidation and anti-inflammation. In addition, animals overexpressing mouse apoA-IV were protected from atherosclerosis when the mice were on an atherogenic diet. This protection was thought to be due to
25 increased levels of HDL (106). Similarly, mice overexpressing human apoA-IV only in the liver also had increased HDL levels and were protected against atherosclerosis (107).
Supporting the role of apoA-IV in lipid transport, its overexpression in a newborn swine enterocyte cell line led to increased transport of triglyceride, cholesteryl ester and phospholipid as lipoproteins (85). Also, the KO animal exhibited decreased HDL levels as compared to the WT (108), presumably because apoA-IV is a component of HDL particles and can act as a lipid poor cholesterol acceptor from the ABCA1 transporter.
However, all of the suggested roles for apoA-IV are not supported by the transgenic and KO animals. One example is that intestinal overexpression of human apoA-IV in mice had no influence on dietary lipid or feeding behavior (109). In addition, it has also been shown that the apoA-IV KO animals show normal growth, lipid absorption and feeding behavior (108). One potential problem with these studies is that the knockout mice were generated on a heterogeneous genetic background but will be studied more extensively by the Tso laboratory. Another potential problem in the case of transgenic and KO animals, is that other proteins levels can be increased or decreased in order to make up for these changes in apoA-IV.
There have also been several mutations found occurring in nature. Normal human apoA-IV-1 has a Gln encoded by CAG at position 360. However, apoA-IV-2 (Q360H), has a His encoded by a single point mutation CAT at that position. ApoA-IV-1 is much more common with an allele frequency of 0.91, while apoA-IV-2 is found at 0.07-0.09
(110-112). Another polymorphism is seen when an ACT has the point mutation to TCT resulting in a Thr to Ser substitution, T347S. This mutant occurs with an allele frequency
26 from 0.12-0.22 (110). Several other polymorphisms exist, but to a much lesser extent and
little is known about their effects.
Q360H is thought to bind lipid to a greater extent as compared to apoA-IV-1;
however, the T347S variant is thought to bind lipid to a lesser extent (113-115). This has
a range of effects in vivo in terms of lipid metabolism. For example, humans
heterozygous for the Q360H allele have been shown to be slower at clearing postprandial
triglycerides while those heterozygous for the T347S allele have the opposite effects
(116). Also, as discussed previously, apoA-IV has been proposed to play a role in food
intake. Humans heterozygous for the Q360H allele have been shown to have a decreased
body mass index (BMI) but those that are heterozygous for the T347S allele have an
increased BMI and body fat percentage (117-119). Taken together, this suggests that the
T347S mutation would lead to an increased risk for cardiovascular disease whereas the
Q360H mutation may have a more protective effect. Wong et al. have recently provided data supporting the notion that the T347S mutation leads to increased coronary heart disease (120). In the last 10 years, many papers have been published involving the polymorphisms and their effects in humans. However, like the genetically engineered mice, there are many factors that can change in response to the changes in apoA-IV which can lead to conflicting findings.
The aims of this Thesis: The overall goal of this research was to derive a better understanding of apolipoprotein A-IV in terms of its structure and function. The general hypothesis that was tested was:
27 HYPOTHESIS: Specific regions of apoA-IV are responsible for its ability to interact with lipid and inhibit food intake.
The broad strategy consisted of generating mutant forms of the protein and then testing them in various structural and functional assays. The hypothesis was broken into two aims.
Aim 1: Determine the region of apoA-IV that is responsible for its ability to bind lipid as well as to identify the generalities of its structure.
Aim 2: Determine the region of apoA-IV that is responsible for its role in the central inhibition of food intake.
The organization of this Thesis:
This thesis is organized into six chapters. Chapter 1 is the background and summary of the literature of apoA-IV. The remaining chapters are set up as separate manuscripts with an individual abstract, introduction, materials and methods, results and discussion and references. Chapter 2 discusses the importance and generation of a recombinant expression system. Specifically, rat apoA-IV and rat apoE were produced in a bacterial expression system and are discussed in detail in this chapter. This chapter is important because it focuses on the purification and comparison of recombinant proteins with native proteins purified from rats. The proteins compared favorably in all assays tested.
28 Chapters 3-4 focus on the knowledge gained on Aim 1. These chapters used human
apoA-IV in the same expression system as discussed in Chapter 2. Mutant forms of the
protein were made in order to determine whether a specific region of the protein was
involved in lipid binding. Interestingly, two important regions were discovered in apoA-
IV when tested for lipid binding ability. An ‘inhibitory’ region was found from amino
acids 333-343. When this region is removed, the mutant is better at binding lipid than the
WT. However, the region from amino acid 11-20 was also important because if it was
removed in addition to amino acids 333-343, the mutant was slower than the ∆333-343 mutant alone. This suggests there is a complex interaction between the C-terminus that keeps the N-terminus from putting the protein in a favorable conformation for lipid binding.
Chapter 5 uses the expression system from Chapter 2 and focuses on Aim 2. Mutant forms of the protein were made in order to determine which region was responsible for inhibiting food intake when injected centrally into rats. Using a series of mutants, the region was located to the N-terminal 61 amino acids of apoA-IV. This region of the protein is very well conserved between species. Studies were also performed that showed apoA-IV does not appear to have the ability to cross the blood-brain barrier when injected into the periphery. This is important because it suggests that nerve induction may be the mechanism that apoA-IV uses to regulate food intake in vivo.
Chapter 6 is a general discussion of the entire project and includes future goals and the
importance of the findings of this thesis.
29 REFERENCES
1. Swaney,J.B., Reese,H., and EDER,H.A. (1974) Polypeptide composition of rat high density lipoprotein: characterization by SDS-gel electrophoresis, Biochem. Biophys. Res. Commun. 59, 513-519.
2. Boguski,M.S., Elshourbagy,N., Taylor,J.M., and Gordon,J.I. (1985) Comparative analysis of repeated sequences in rat apolipoproteins A-I, A-IV, and E, Proc. Natl. Acad. Sci. USA 82, 992- 996.
3. Hayashi,H., Nutting,D.F., Fujimoto,K., Cardelli,J.A., Black,D., and Tso,P. (1990) Transport of lipid and apolipoproteins A-I and A-IV in intestinal lymph of the rat, J. Lipid Res. 31, 1613-1625.
4. Green,P.H., Glickman,R.M., Riley,J.W., and Quinet,E. (1980) Human apolipoprotein A-IV. Intestinal origin and distribution in plasma, J. Clin. Invest 65, 911-919.
5. Fujimoto,K., Machidori,H., Iwakiri,R., Yamamoto,K., Fujisaki,J., Sakata,T., and Tso,P. (1993) Effect of intravenous administration of apolipoprotein A-IV on patterns of feeding, drinking and ambulatory activity of rats, Brain Res. 608, 233-237.
6. Fujimoto,K., Fukagawa,K., Sakata,T., and Tso,P. (1993) Suppression of food intake by apolipoprotein A-IV is mediated through the central nervous system in rats, J. Clin. Invest 91, 1830-1833.
7. Okumura,T., Fukagawa,K., Tso,P., Taylor,I.L., and Pappas,T.N. (1996) Apolipoprotein A-IV acts in the brain to inhibit gastric emptying in the rat, Am. J. Physiol 270, G49-G53.
8. Lefevre,M. and Roheim,P.S. (1984) Metabolism of apolipoprotein A-IV. [Review] [46 refs], J. Lipid Res. 25, 1603-1610.
9. Ostos,M.A., Conconi,M., Vergnes,L., Baroukh,N., Ribalta,J., Girona,J., Caillaud,J.M., Ochoa,A., and Zakin,M.M. (2001) Antioxidative and antiatherosclerotic effects of human apolipoprotein A- IV in apolipoprotein E-deficient mice, Arterioscler. Thromb. Vasc. Biol. 21, 1023-1028.
10. Remaley,A.T., Stonik,J.A., Demosky,S.J., Neufeld,E.B., Bocharov,A.V., Vishnyakova,T.G., Eggerman,T.L., Patterson,A.P., Duverger,N.J., Santamarina-Fojo,S., and Brewer,H.B., Jr. (2001) Apolipoprotein specificity for lipid efflux by the human ABCAI transporter, Biochem. Biophys. Res. Commun. 280, 818-823.
11. Steinmetz,A., Barbaras,R., Ghalim,N., Clavey,V., Fruchart,J.C., and Ailhaud,G. (1990) Human apolipoprotein A-IV binds to apolipoprotein A-I/A-II receptor sites and promotes cholesterol efflux from adipose cells, J. Biol. Chem. 265, 7859-7863.
12. Stein,O., Stein,Y., Lefevre,M., and Roheim,P.S. (1986) The role of apolipoprotein A-IV in reverse cholesterol transport studied with cultured cells and liposomes derived from an ether analog of phosphatidylcholine, Biochim. Biophys. Acta 878, 7-13.
13. Oram,J.F. and Yokoyama,S. (1996) Apolipoprotein-mediated removal of cellular cholesterol and phospholipids, J. Lipid Res. 37, 2473-2491.
14. Kalogeris,T.J., Fukagawa,K., Tsuchiya,T., Qin,X., and Tso,P. (1999) Intestinal synthesis and lymphatic secretion of apolipoprotein A-IV after cessation of duodenal fat infusion: mediation by bile, Biochim. Biophys. Acta 1436, 451-466.
30 15. Elshourbagy,N.A., Walker,D.W., Paik,Y.K., Boguski,M.S., Freeman,M., Gordon,J.I., and Taylor,J.M. (1987) Structure and expression of the human apolipoprotein A-IV gene, J. Biol. Chem. 262, 7973-7981.
16. Lin-Lee,Y.C., Strobl,W., Soyal,S., Radosavljevic,M., Song,M., Gotto,A.M., Jr., and Patsch,W. (1993) Role of thyroid hormone in the expression of apolipoprotein A-IV and C-III genes in rat liver, J. Lipid Res. 34, 249-259.
17. Davidson,N.O., Carlos,R.C., Drewek,M.J., and Parmer,T.G. (1988) Apolipoprotein gene expression in the rat is regulated in a tissue-specific manner by thyroid hormone, J. Lipid Res. 29, 1511-1522.
18. Redgrave,T.G. (1970) Formation of cholesteryl ester-rich particulate lipid during metabolism of chylomicrons, J. Clin. Invest 49, 465-471.
19. Weinberg,R.B. and Spector,M.S. (1985) Human apolipoprotein A-IV: displacement from the surface of triglyceride-rich particles by HDL2-associated C-apoproteins, J. Lipid Res. 26, 26-37.
20. Dallinga-Thie,G.M., Groot,P.H., and van Tol,A. (1985) Distribution of apolipoprotein A-IV among lipoprotein subclasses in rat serum, J. Lipid Res. 26, 970-976.
21. Malmendier,C.L., Lontie,J.F., Lagrost,L., Delcroix,C., Dubois,D.Y., and Gambert,P. (1991) In vivo metabolism of apolipoproteins A-IV and A-I associated with high density lipoprotein in normolipidemic subjects, J. Lipid Res. 32, 801-808.
22. Ezeh,B., Haiman,M., Alber,H.F., Kunz,B., Paulweber,B., Lingenhel,A., Kraft,H.G., Weidinger,F., Pachinger,O., Dieplinger,H., and Kronenberg,F. (2003) Plasma distribution of apoA-IV in patients with coronary artery disease and healthy controls, J. Lipid Res. 44, 1523-1529.
23. Lagrost,L., Gambert,P., Boquillon,M., and Lallemant,C. (1989) Evidence for high density lipoproteins as the major apolipoprotein A-IV-containing fraction in normal human serum, J. Lipid Res. 30, 1525-1534.
24. Duverger,N., Ghalim,N., Ailhaud,G., Steinmetz,A., Fruchart,J.C., and Castro,G. (1993) Characterization of apoA-IV-containing lipoprotein particles isolated from human plasma and interstitial fluid, Arteriosclerosis & Thrombosis 13, 126-132.
25. Bottcher,A., Schlosser,J., Kronenberg,F., Dieplinger,H., Knipping,G., Lackner,K.J., and Schmitz,G. (2000) Preparative free-solution isotachophoresis for separation of human plasma lipoproteins: apolipoprotein and lipid composition of HDL subfractions, J. Lipid Res. 41, 905-915.
26. Weinberg,R.B. and Jordan,M.K. (1990) Effects of phospholipid on the structure of human apolipoprotein A-IV, J. Biol. Chem. 265, 8081-8086.
27. Weinberg,R.B., Ibdah,J.A., and Phillips,M.C. (1992) Adsorption of apolipoprotein A-IV to phospholipid monolayers spread at the air/water interface. A model for its labile binding to high density lipoproteins, J. Biol. Chem. 267, 8977-8983.
28. DeLamatre,J.G., Hoffmeier,C.A., Lacko,A.G., and Roheim,P.S. (1983) Distribution of apolipoprotein A-IV between the lipoprotein and the lipoprotein-free fractions of rat plasma: possible role of lecithin:cholesterol acyltransferase, J. Lipid Res. 24, 1578-1585.
29. DeLamatre,J.G. and Roheim,P.S. (1983) The response of apolipoprotein A-IV to cholesterol feeding in rats, Biochim. Biophys. Acta 751, 210-217.
31 30. Weinberg,R.B. (1987) Differences in the hydrophobic properties of discrete alpha-helical domains of rat and human apolipoprotein A-IV, Biochim. Biophys. Acta 918, 299-303.
31. Fidge,N.H. (1980) The redistribution and metabolism of iodinated apolipoprotein A-IV in rats, Biochim. Biophys. Acta 619, 129-141.
32. Ghiselli,G., Krishnan,S., Beigel,Y., and Gotto,A.M., Jr. (1986) Plasma metabolism of apolipoprotein A-IV in humans, J. Lipid Res. 27, 813-827.
33. Dallinga-Thie,G.M., van't Hooft,F.M., and van Tol,A. (1986) Comparison of the metabolic behavior of rat apolipoproteins A-I and A-IV, isolated from both lymph chylomicrons and serum high density lipoproteins, International Journal of Biochemistry 18, 383-388.
34. Ghiselli,G., Crump,W.L., Musanti,R., Sherrill,B.C., and Gotto,A.M.J. (1989) Metabolism of apolipoprotein A-IV in rat, Biochim. Biophys. Acta 1006, 26-34.
35. Dieplinger,H., Lobentanz,E.M., Konig,P., Graf,H., Sandholzer,C., Matthys,E., Rosseneu,M., and Utermann,G. (1992) Plasma apolipoprotein A-IV metabolism in patients with chronic renal disease, Eur. J. Clin. Invest 22, 166-174.
36. Karathanasis,S.K. (1985) Apolipoprotein multigene family: tandem organization of human apolipoprotein AI, CIII, and AIV genes, Proc. Natl. Acad. Sci. U. S. A 82, 6374-6378.
37. Karathanasis,S.K., Oettgen,P., Haddad,I.A., and Antonarakis,S.E. (1986) Structure, evolution, and polymorphisms of the human apolipoprotein A4 gene (APOA4), Proc. Natl. Acad. Sci. U. S. A 83, 8457-8461.
38. Elshourbagy,N.A., Walker,D.W., Boguski,M.S., Gordon,J.I., and Taylor,J.M. (1986) The nucleotide and derived amino acid sequence of human apolipoprotein A-IV mRNA and the close linkage of its gene to the genes of apolipoproteins A-I and C-III, J. Biol. Chem. 261, 1998-2002.
39. Boguski,M.S., Birkenmeier,E.H., Elshourbagy,N.A., Taylor,J.M., and Gordon,J.I. (1986) Evolution of the apolipoproteins. Structure of the rat apo-A-IV gene and its relationship to the human genes for apo-A-I, C-III, and E, J. Biol. Chem. 261, 6398-6407.
40. Omori,K., Vergnes,L., Zakin,M.M., and Ochoa,A. (1995) The apolipoprotein AICIII-AIV gene cluster: sequence of the ApoCIII-ApoAIV intergenic region, Gene 159, 231-234.
41. Bisaha,J.G., Simon,T.C., Gordon,J.I., and Breslow,J.L. (1995) Characterization of an enhancer element in the human apolipoprotein C-III gene that regulates human apolipoprotein A-I gene expression in the intestinal epithelium, J. Biol. Chem. 270, 19979-19988.
42. Sauvaget,D., Chauffeton,V., Citadelle,D., Chatelet,F.P., Cywiner-Golenzer,C., Chambaz,J., Pincon-Raymond,M., Cardot,P., Le Beyec,J., and Ribeiro,A. (2002) Restriction of apolipoprotein A-IV gene expression to the intestine villus depends on a hormone-responsive element and parallels differential expression of the hepatic nuclear factor 4alpha and gamma isoforms, J. Biol. Chem. 277, 34540-34548.
43. Vergnes,L., Taniguchi,T., Omori,K., Zakin,M.M., and Ochoa,A. (1997) The apolipoprotein A-I/C- III/A-IV gene cluster: ApoC-III and ApoA-IV expression is regulated by two common enhancers, Biochim. Biophys. Acta 1348, 299-310.
44. Le Beyec,J., Chauffeton,V., Kan,H.Y., Janvier,P.L., Cywiner-Golenzer,C., Chatelet,F.P., Kalopissis,A.D., Zannis,V., Chambaz,J., Pincon-Raymond,M., and Cardot,P. (1999) The -700/- 310 fragment of the apolipoprotein A-IV gene combined with the -890/-500 apolipoprotein C-III
32 enhancer is sufficient to direct a pattern of gene expression similar to that for the endogenous apolipoprotein A-IV gene, J. Biol. Chem. 274, 4954-4961.
45. Wojciechowski,A.P., Farrall,M., Cullen,P., Wilson,T.M., Bayliss,J.D., Farren,B., Griffin,B.A., Caslake,M.J., Packard,C.J., Shepherd,J., and . (1991) Familial combined hyperlipidaemia linked to the apolipoprotein AI-CII-AIV gene cluster on chromosome 11q23-q24, Nature 349, 161-164.
46. Ordovas,J.M., Cassidy,D.K., Civeira,F., Bisgaier,C.L., and Schaefer,E.J. (1989) Familial apolipoprotein A-I, C-III, and A-IV deficiency and premature atherosclerosis due to deletion of a gene complex on chromosome 11, J. Biol. Chem. 264, 16339-16342.
47. Karathanasis,S.K., Ferris,E., and Haddad,I.A. (1987) DNA inversion within the apolipoproteins AI/CIII/AIV-encoding gene cluster of certain patients with premature atherosclerosis, Proc. Natl. Acad. Sci. U. S. A 84, 7198-7202.
48. Boguski,M.S., Elshourbagy,N., Taylor,J.M., and Gordon,J.I. (1984) Rat apolipoprotein A-IV contains 13 tandem repetitions of a 22-amino acid segment with amphipathic helical potential, Proc. Natl. Acad. Sci. U. S. A 81, 5021-5025.
49. Gordon,J.I., Bisgaier,C.L., Sims,H.F., Sachdev,O.P., Glickman,R.M., and Strauss,A.W. (1984) Biosynthesis of human preapolipoprotein A-IV, J. Biol. Chem. 259, 468-474.
50. Gordon,J.I., Smith,D.P., Alpers,D.H., and Strauss,A.W. (1982) Proteolytic processing of the primary translation product of rat intestinal apolipoprotein A-IV mRNA. Comparison with preproapolipoprotein A-I processing, J. Biol. Chem. 257, 8418-8423.
51. Beisiegel,U. and Utermann,G. (1979) An apolipoprotein homolog of rat apolipoprotein A-IV in human plasma. Isolation and partial characterisation, Eur. J. Biochem. 93, 601-608.
52. Weinberg,R.B. and Scanu,A.M. (1983) Isolation and characterization of human apolipoprotein A- IV from lipoprotein-depleted serum, J. Lipid Res. 24, 52-59.
53. Weinberg,R.B. (1994) Identification of functional domains in the plasma apolipoproteins by analysis of inter-species sequence variability, J. Lipid Res. 35, 2212-2222.
54. Brouillette,C.G. and Anantharamaiah,G.M. (1995) Structural models of human apolipoprotein A- I. [Review], Biochim. Biophys. Acta. 1256, 103-129.
55. Segrest,J.P., De Loof,H., Dohlman,J.G., Brouillette,C.G., and Anantharamaiah,G.M. (1990) Amphipathic helix motif: classes and properties, Proteins 8, 103-117.
56. Segrest,J.P., Jones,M.K., De Loof,H., Brouillette,C.G., Venkatachalapathi,Y.V., and Anantharamaiah,G.M. (1992) The amphipathic helix in the exchangeable apolipoproteins: a review of secondary structure and function. [Review], J. Lipid Res. 33, 141-166.
57. Weinberg,R.B. and Spector,M.S. (1985) Structural properties and lipid binding of human apolipoprotein A-IV, J. Biol. Chem. 260, 4914-4921.
58. Garnier,J., Gibrat,J.F., and Robson,B. (1996) GOR method for predicting protein secondary structure from amino acid sequence, Methods Enzymol. 266, 540-553.
59. Dvorin,E., Mantulin,W.W., Rohde,M.F., Gotto,A.M., Jr., Pownall,H.J., and Sherrill,B.C. (1985) Conformational properties of human and rat apolipoprotein A-IV, Journal. of. Lipid Research. 26, 38-46.
33 60. Davidson,W.S., Hazlett,T., Mantulin,W.W., and Jonas,A. (1996) The role of apolipoprotein AI domains in lipid binding, Proc. Natl. Acad. Sci. USA 93, 13605-13610.
61. Edelstein,C. and Scanu,A.M. (1980) Effect of guanidine hydrochloride on the hydrodynamic and thermodynamic properties of human apolipoprotein A-I in solution, J. Biol. Chem. 255, 5747- 5754.
62. Wetterau,J.R., Aggerbeck,L.P., Rall,S.C., Jr., and Weisgraber,K.H. (1988) Human apolipoprotein E3 in aqueous solution. I. Evidence for two structural domains, Journal. of. Biological. Chemistry. 263, 6240-6248.
63. Weinberg,R.B. (1988) Exposure and electronic interaction of tyrosine and tryptophan residues in human apolipoprotein A-IV, Biochemistry 27, 1515-1521.
64. Weinberg,R.B., Anderson,R.A., Cook,V.R., Emmanuel,F., Denefle,P., Tall,A.R., and Steinmetz,A. (2002) Interfacial exclusion pressure determines the ability of apolipoprotein A-IV truncation mutants to activate cholesterol ester transfer protein, J. Biol. Chem. 277, 21549-21553.
65. Wilson,C., Wardell,M.R., Weisgraber,K.H., Mahley,R.W., and Agard,D.A. (1991) Three- dimensional structure of the LDL receptor-binding domain of human apolipoprotein E, Science 252, 1817-1822.
66. Weisgraber,K.H. (1994) Apolipoprotein E: structure-function relationships. [Review] [201 refs], Advances. in Protein Chemistry. 45, 249-302.
67. Oda,M.N., Forte,T.M., Ryan,R.O., and Voss,J.C. (2003) The C-terminal domain of apolipoprotein A-I contains a lipid-sensitive conformational trigger, Nat. Struct. Biol 10, 455-460.
68. Saito,H., Dhanasekaran,P., Nguyen,D., Holvoet,P., Lund-Katz,S., and Phillips,M.C. (2003) Domain structure and lipid interaction in human apolipoproteins A-I and E, a general model, J. Biol. Chem. 278, 23227-23232.
69. Weinberg,R.B. and Spector,M.S. (1985) The self-association of human apolipoprotein A-IV. Evidence for an in vivo circulating dimeric form, J. Biol. Chem. 260, 14279-14286.
70. Weinberg,R.B., Anderson,R.A., Cook,V.R., Emmanuel,F., Denefle,P., Hermann,M., and Steinmetz,A. (2000) Structure and interfacial properties of chicken apolipoprotein A-IV, J. Lipid Res. 41, 1410-1418.
71. Lins,L., Brasseur,R., Rosseneu,M., Vanloo,B., and Ruysschaert,J.M. (1993) Structure of the apolipoprotein A-IV/lipid discoidal complexes: an attenuated total reflection polarized Fourier transform infrared spectroscopy study, Biochim. Biophys. Acta 1149, 267-277.
72. Jonas,A., Steinmetz,A., and Churgay,L. (1993) The number of amphipathic alpha-helical segments of apolipoproteins A-I, E, and A-IV determines the size and functional properties of their reconstituted lipoprotein particles, J. Biol. Chem. 268, 1596-1602.
73. Sato,K., Takahashi,Y., Takahashi,T., Katoh,N., and Akiba,Y. (2002) Identification of factors regulating lipoprotein lipase catalyzed hydrolysis in rats with the aid of monoacid-rich lipoprotein preparations(1), J. Nutr. Biochem. 13, 528.
74. Nilsson-Ehle,P., Garfinkel,A.S., and Schotz,M.C. (1980) Lipolytic enzymes and plasma lipoprotein metabolism, Annu. Rev. Biochem. 49, 667-693.
34 75. Qin,X., Swertfeger,D.K., Zheng,S., Hui,D.Y., and Tso,P. (1998) Apolipoprotein AIV: a potent endogenous inhibitor of lipid oxidation, American Journal of Physiology 274, H1836-H1840.
76. Phillips,M.C., Johnson,W.J., and Rothblat,G.H. (1987) Mechanisms and consequences of cellular cholesterol exchange and transfer. [Review], Biochim. Biophys. Acta. 906, 223-276.
77. Fielding,C.J. and Fielding,P.E. (1995) Molecular physiology of reverse cholesterol transport. [Review], J. Lipid Res. 36, 211-228.
78. Brown,M.S. and Goldstein,J.L. (1986) A receptor-mediated pathway for cholesterol homeostasis. [Review], Science 232, 34-47.
79. Lin,G. (2002) Insights of high-density lipoprotein apolipoprotein-mediated lipid efflux from cells, Biochem. Biophys. Res. Commun. 291, 727-731.
80. Santamarina-Fojo,S., Remaley,A.T., Neufeld,E.B., and Brewer,H.B., Jr. (2001) Regulation and intracellular trafficking of the ABCA1 transporter, J. Lipid Res. 42, 1339-1345.
81. Chen,C.H. and Albers,J.J. (1985) Activation of lecithin: cholesterol acyltransferase by apolipoproteins E-2, E-3, and A-IV isolated from human plasma, Biochim. Biophys. Acta 836, 279-285.
82. Steinmetz,A. and Utermann,G. (1985) Activation of lecithin: cholesterol acyltransferase by human apolipoprotein A-IV, J. Biol. Chem. 260, 2258-2264.
83. Lagrost,L., Gambert,P., Dangremont,V., Athias,A., and Lallemant,C. (1990) Role of cholesteryl ester transfer protein (CETP) in the HDL conversion process as evidenced by using anti-CETP monoclonal antibodies, J. Lipid Res. 31, 1569-1575.
84. Barter,P.J., Rajaram,O.V., Chang,L.B., Rye,K.A., Gambert,P., Lagrost,L., Ehnholm,C., and Fidge,N.H. (1988) Isolation of a high-density-lipoprotein conversion factor from human plasma. A possible role of apolipoprotein A-IV as its activator, Biochem. J. 254, 179-184.
85. Lu,S., Yao,Y., Meng,S., Cheng,X., and Black,D.D. (2002) Overexpression of apolipoprotein A-IV enhances lipid transport in newborn swine intestinal epithelial cells, J. Biol. Chem. 277, 31929- 31937.
86. Quarfordt,S.H. and Goodman,D.S. (1966) Heterogeneity in the rate of plasma clearance of chylomicrons of different size, Biochim. Biophys. Acta 116, 382-385.
87. Martins,I.J., Mortimer,B.C., Miller,J., and Redgrave,T.G. (1996) Effects of particle size and number on the plasma clearance of chylomicrons and remnants, J. Lipid Res. 37, 2696-2705.
88. Zilversmit,D.B. (1995) Atherogenic nature of triglycerides, postprandial lipidemia, and triglyceride-rich remnant lipoproteins, Clin. Chem. 41, 153-158.
89. Jonas,A. (1991) Lecithin-cholesterol acyltransferase in the metabolism of high- density lipoproteins. [Review], Biochim. Biophys. Acta. 1084, 205-220.
90. Emmanuel,F., Steinmetz,A., Rosseneu,M., Brasseur,R., Gosselet,N., Attenot,F., Cuine,S., Seguret,S., Latta,M., Fruchart,J.C., and et al (1994) Identification of specific amphipathic alpha- helical sequence of human apolipoprotein A-IV involved in lecithin:cholesterol acyltransferase activation, J. Biol. Chem. 269, 29883-29890.
35 91. Weinberg,R.B., Cook,V.R., Jones,J.B., Kussie,P., and Tall,A.R. (1994) Interfacial properties of recombinant human cholesterol ester transfer protein, J. Biol. Chem. 269, 29588-29591.
92. Fournier,N., Atger,V., Paul,J.L., Sturm,M., Duverger,N., Rothblat,G.H., and Moatti,N. (2000) Human ApoA-IV overexpression in transgenic mice induces cAMP-stimulated cholesterol efflux from J774 macrophages to whole serum, Arterioscler. Thromb. Vasc. Biol. 20, 1283-1292.
93. Oram,J.F., Lawn,R.M., Garvin,M.R., and Wade,D.P. (2000) ABCA1 is the cAMP-inducible apolipoprotein receptor that mediates cholesterol secretion from macrophages, J. Biol. Chem. 275, 34508-34511.
94. Smith,J.D., Miyata,M., Ginsberg,M., Grigaux,C., Shmookler,E., and Plump,A.S. (1996) Cyclic AMP induces apolipoprotein E binding activity and promotes cholesterol efflux from a macrophage cell line to apolipoprotein acceptors, J. Biol. Chem. 271, 30647-30655.
95. Recalde,D., Ostos,M.A., Badell,E., Garcia-Otin,A.L., Pidoux,J., Castro,G., Zakin,M.M., and Scott-Algara,D. (2004) Human apolipoprotein A-IV reduces secretion of proinflammatory cytokines and atherosclerotic effects of a chronic infection mimicked by lipopolysaccharide, Arterioscler. Thromb. Vasc. Biol. 24, 756-761.
96. Vowinkel,T., Mori,M., Krieglstein,C.F., Russell,J., Saijo,F., Bharwani,S., Turnage,R.H., Davidson,W.S., Tso,P., Granger,D.N., and Kalogeris,T.J. (2004) Apolipoprotein A-IV inhibits experimental colitis, J. Clin. Invest 114, 260-269.
97. Fujimoto,K., Cardelli,J.A., and Tso,P. (1992) Increased apolipoprotein A-IV in rat mesenteric lymph after lipid meal acts as a physiological signal for satiation, Am. J. Physiol 262, G1002- G1006.
98. Woods,S.C., Seeley,R.J., Porte,D., Jr., and Schwartz,M.W. (1998) Signals that regulate food intake and energy homeostasis, Science 280, 1378-1383.
99. Havel,P.J. (2001) Peripheral signals conveying metabolic information to the brain: short-term and long-term regulation of food intake and energy homeostasis, Exp. Biol. Med. (Maywood. ) 226, 963-977.
100. Campfield,L.A., Smith,F.J., and Burn,P. (1998) Strategies and potential molecular targets for obesity treatment, Science 280, 1383-1387.
101. Liu,M., Doi,T., Shen,L., Woods,S.C., Seeley,R.J., Zheng,S., Jackman,A., and Tso,P. (2001) Intestinal satiety protein apolipoprotein AIV is synthesized and regulated in rat hypothalamus, Am. J. Physiol Regul. Integr. Comp Physiol 280, R1382-R1387.
102. Liu,M., Doi,T., and Tso,P. (2003) Regulation of intestinal and hypothalamic apolipoprotein A-IV, Exp. Biol. Med. (Maywood. ) 228, 1181-1189.
103. Raybould,H.E., Meyer,J.H., Tabrizi,Y., Liddle,R.A., and Tso,P. (1998) Inhibition of gastric emptying in response to intestinal lipid is dependent on chylomicron formation, Am. J. Physiol. 274, R1834-R1838.
104. Okumura,T., Taylor,I.L., Fukagawa,K., Tso,P., and Pappas,T.N. (1995) Apolipoprotein A-IV acts centrally in the brain to reduce the severity of gastric ulceration in the rat, Brain Res. 673, 153- 156.
36 105. Glatzle,J., Darcel,N., Rechs,A.J., Kalogeris,T.J., Tso,P., and Raybould,H.E. (2004) Apolipoprotein A-IV stimulates duodenal vagal afferent activity to inhibit gastric motility via a CCK1 pathway, Am. J. Physiol Regul. Integr. Comp Physiol 287, R354-R359.
106. Cohen,R.D., Castellani,L.W., Qiao,J.H., Van Lenten,B.J., Lusis,A.J., and Reue,K. (1997) Reduced aortic lesions and elevated high density lipoprotein levels in transgenic mice overexpressing mouse apolipoprotein A-IV, J. Clin. Invest 99, 1906-1916.
107. Duverger,N., Tremp,G., Caillaud,J.M., Emmanuel,F., Castro,G., Fruchart,J.C., Steinmetz,A., and Denefle,P. (1996) Protection against atherogenesis in mice mediated by human apolipoprotein A- IV, Science 273, 966-968.
108. Weinstock,P.H., Bisgaier,C.L., Hayek,T., Aalto-Setala,K., Sehayek,E., Wu,L., Sheiffele,P., Merkel,M., Essenburg,A.D., and Breslow,J.L. (1997) Decreased HDL cholesterol levels but normal lipid absorption, growth, and feeding behavior in apolipoprotein A-IV knockout mice, J. Lipid Res. 38, 1782-1794.
109. Aalto-Setala,K., Bisgaier,C.L., Ho,A., Kieft,K.A., Traber,M.G., Kayden,H.J., Ramakrishnan,R., Walsh,A., Essenburg,A.D., and Breslow,J.L. (1994) Intestinal expression of human apolipoprotein A-IV in transgenic mice fails to influence dietary lipid absorption or feeding behavior, J. Clin. Invest 93, 1776-1786.
110. Lohse,P., Kindt,M.R., Rader,D.J., and Brewer,H.B., Jr. (1991) Three genetic variants of human plasma apolipoprotein A-IV. apoA- IV-1(Thr347----Ser), apoA-IV-0(Lys167----Glu,Gln360---- His), and apoA-IV-3(Glu165----Lys), J. Biol. Chem. 266, 13513-13518.
111. Lohse,P., Kindt,M.R., Rader,D.J., and Brewer,H.B., Jr. (1990) Genetic polymorphism of human plasma apolipoprotein A-IV is due to nucleotide substitutions in the apolipoprotein A-IV gene, J. Biol. Chem. 265, 10061-10064.
112. Tenkanen,H., Lukka,M., Jauhiainen,M., Metso,J., Baumann,M., Peltonen,L., and Ehnholm,C. (1991) The mutation causing the common apolipoprotein A-IV polymorphism is a glutamine to histidine substitution of amino acid 360, Arterioscler. Thromb. 11, 851-856.
113. Weinberg,R.B., Jordan,M.K., and Steinmetz,A. (1990) Distinctive structure and function of human apolipoprotein variant ApoA-IV-2, J. Biol. Chem. 265, 18372-18378.
114. Weinberg,R.B. (2002) Apolipoprotein A-IV polymorphisms and diet-gene interactions, Curr. Opin. Lipidol. 13, 125-134.
115. Weinberg,R.B., Geissinger,B.W., Kasala,K., Hockey,K.J., Terry,J.G., Easter,L., and Crouse,J.R. (2000) Effect of apolipoprotein A-IV genotype and dietary fat on cholesterol absorption in humans, J. Lipid Res. 41, 2035-2041.
116. Hockey,K.J., Anderson,R.A., Cook,V.R., Hantgan,R.R., and Weinberg,R.B. (2001) Effect of the apolipoprotein A-IV Q360H polymorphism on postprandial plasma triglyceride clearance, J. Lipid Res. 42, 211-217.
117. Fiegenbaum,M. and Hutz,M.H. (2003) Further evidence for the association between obesity- related traits and the apolipoprotein A-IV gene, Int. J. Obes. Relat Metab Disord. 27, 484-490.
118. Fisher,R.M., Burke,H., Nicaud,V., Ehnholm,C., and Humphries,S.E. (1999) Effect of variation in the apo A-IV gene on body mass index and fasting and postprandial lipids in the European Atherosclerosis Research Study II, J. Lipid Res. 40, 287-294.
37 119. Lefevre,M., Lovejoy,J.C., DeFelice,S.M., Keener,J.W., Bray,G.A., Ryan,D.H., Hwang,D.H., and Greenway,F.L. (2000) Common apolipoprotein A-IV variants are associated with differences in body mass index levels and percentage body fat, Int. J. Obes. Relat. Metab. Disord. 24, 945-953.
120. Wong,W.M., Hawe,E., Li,L.K., Miller,G.J., Nicaud,V., Pennacchio,L.A., Humphries,S.E., and Talmud,P.J. (2003) Apolipoprotein AIV gene variant S347 is associated with increased risk of coronary heart disease and lower plasma apolipoprotein AIV levels, Circ. Res. 92, 969-975.
38 FIGURES:
N- EVSADQVATVMWDYFSQLSNNAKEAVEH LQKSELTQQLN 1-39 1 ALFQDKLGEVNTYAGDLQKKLV 40-61 2 PFATELHERLAKDSEKLKEEIG 62-83 KELEELRARLL 84-94 3 PHANEVSQKIGDNLRELQQRLE 95-116 4 PYADQLRTQVNTQAEQLRRQLT 117-138 5 PYAQRMERVLRENADSLQASLR 139-160 6 PHADELKAKIDQNVEELKGRLT 161-182 7 PYADEFKVKIDQTVEELRRSLA 183-204 8 PYAQDTQEKLNHQLEGLTFQMK 205-226 9 KNAEELKARISASAEELRQRLA 227-248 10 PLAEDVRGNLRGNTEGLQKSLA 249-248
ELGGHLDQQVEEFRRRVE 271-288 11 PYGENFNKALVQQMEQLRQKLG 289-310 12 PHAGDVEGHLSFLEKDLRDKVN 311-332 SFFSTFKEKESQDKTLSLPELEQQQEQQ QEQQQEQVQMLAPLES 333-376
Figure 1-1: The primary sequence of human apoA-IV. The amino acids are shown in
their one letter form. The amino acids are numbered from 1-376 and do not include the
pre-sequence that is removed during translation. The amino acids shown in gray are
predicted to form amphipathic helices and each helix is numbered from 1-12.
39
*
* LDL Ox-LDL
V Liver VLDL
MΦ apoA-I CM * CETP L P remnants L ABCA1 CM * LCAT * Nascent Mature HDL HDL
Figure 1-2: A general schematic of lipid metabolism and the potential roles of
apoA-IV. An asterisk is placed at each position where apoA-IV has been hypothesized to play a role in vivo. For a further description, please refer to the preceding text.
40 CHAPTER 2:
BACTERIAL EXPRESSION AND CHARACTERIZATION OF RAT
APOLIPOPROTEIN A-IV AND APOLIPOPROTEIN E
ABSTRACT/SUMMARY
Apolipoprotein A-IV is a glycoprotein synthesized in the small intestine and liver
of rodents and the small intestine of humans. This apolipoprotein is evolutionarily
related to the exchangeable apolipoproteins found in high density lipoproteins including
apoA-I and apoE. Unlike many of these, however, a distinct function for apoA-IV has not yet been unambiguously identified. Its synthesis is increased in response to lipid absorption and the formation of chylomicrons and may play a role in cholesterol efflux and lipoprotein metabolism, upper gastrointestinal function, protection against lipid oxidation and atherosclerosis and regulation of food intake.
Unfortunately, apoA-IV is a difficult protein to purify from either plasma or lymph. Various methods for its purification exist, but these methods require several tedious and time-consuming steps and result in poor apoA-IV recovery. So, studying the protein has been problematic in the past. Therefore, it was necessary to develop an expression system that would allow the production and purification of human and rat apoA-IV. Large amounts of protein can be produced and purified quickly and economically in a bacterial expression system because of the ability to incorporate an
affinity tag on the expressed protein. Furthermore, once such an expression system was
established, mutagenesis studies could be performed in order to manipulate its primary
sequence.
41 During my graduate research, I have established an expression system for human, rat and mouse apoA-IV. Also, as a side project with Dr. Min Liu, I was able to create an expression system for both rat and mouse apoE. Chapter 2 describes the development of bacterial expression systems for rat apoA-IV and apoE for use in food intake and apolipoprotein structure studies. Studies utilizing the recombinant human apoA-IV are presented in Chapters 3 and 4. Mutagenesis studies on rat apoA-IV and the regions required for control of food intake are presented in Chapter 5.
INTRODUCTION
Apolipoprotein A-IV is a major component of chylomicrons and high density lipoproteins. Rat apoA-IV is synthesized in the small intestine and liver, whereas human apoA-IV is predominantly found in small intestine (1;2). Several studies have provided evidence that apoA-IV modulates plasma cholesterol and lipoprotein metabolism (3;4), upper gastrointestinal function (5;6), control of food intake (7-9), protection against lipoprotein oxidation (10), and the risk for atherosclerosis (11;12). ApoA-IV mRNA and protein were recently identified in the hypothalamus of rats, and mRNA levels were found to be regulated physiologically (13).
Apolipoprotein E is a 34 kDa protein that was discovered in 1973 as a protein constituent of human very low density lipoprotein (14). ApoE is synthesized by many tissues, including the liver and brain, and various cell types within these tissues (15). It plays a critical role in lipid transport and cholesterol homeostasis as it is a ligand for the low density lipoprotein (LDL) receptor (16;17). ApoE also has an important role in neuroprotection (18-20) and the human apolipoprotein E4 isoform is linked to an
42 increased risk for Alzheimer’s disease (21-23) in humans. Many studies have focused
both on the structure and function of apoE, making it one of the most well studied
apolipoproteins. Recently, several groups have shown that apoE production is increased
in response to certain dietary conditions (24;25). Additionally, preliminary data from the
Dr. Min Liu has shown that centrally administered apoE is involved in the regulation of
food intake. Therefore, large amounts of apoE will be required to determine whether
peripheral administration will also exhibit an inhibitory effect on food intake. Although
several bacterial expression systems have been developed for human and mouse apoE
(26-28), there is still a need for a recombinant system that produce large amounts of pure
rat apoE for in vivo experiments in the rat model.
We adapted our efficient expression system that has been used previously for apoA-I for these proteins (29). A key feature of our design was the use of the Igase protease which we demonstrate does not cleave apolipoproteins at unwanted sites in the sequence. We found that the expression system was able to produce at least 20 milligrams of highly pure rat protein per liter of culture media. Importantly, the recombinant rat apoA-IV and apoE produced in this expression system were nearly identical to the native apoA-IV and apoE purified from plasma. These rodent expression systems will be useful for in vivo studies on the effects of apoA-IV and apoE on food intake and lipid metabolism.
43 EXPERIMENTAL PROCEDURES
Materials
Primer synthesis and DNA sequencing were performed by the University of Cincinnati
DNA Core (Cincinnati, OH). Restriction enzymes were purchased from New England
Biolabs (Beverly, MA). Rat Liver QUICK-Clone cDNA was purchased from BD
Biosciences Clontech (Palo Alto, CA). Cloned Pfu DNA polymerase was purchased from Stratagene (La Jolla, CA). SDS-PAGE gels were obtained from Bio-Rad (Hercules,
CA) or Amersham-Pharmacia (Piscataway, NJ). The low molecular weight marker was from Amersham-Pharmacia (Piscataway, NJ). Pre-stained protein ladder was from
Invitrogen Life Technologies (Carlsbad, CA). Immun-Blot PVDF membrane was from
Bio-Rad (Hercules, CA). Enhanced chemiluminescence (ECL) system was from
Amersham Life Science (Princeton, NJ). IgA protease (Igase) was purchased from
MoBiTec (Germany). BL-21 (DE3) Escherichia coli and the pET30 vector were from
Novagen (Madison, WI). Kanamycin was purchased from Calbiochem (San Diego, CA).
Isopropyl-β-D-thiogalactoside (IPTG) was from Fisher Scientific (Pittsburgh, PA). 1,2- dimyristoyl-sn-glycero-3-phosphatidylcholine (DMPC) was acquired form Avanti Polar
Lipids (Birmingham, AL). HisBind Resin was from Novagen (Madison, WI).
Centriprep centrifugal concentrators were from Millipore/Amicon Bioseparations
(Bedford, MA). Fatty acid-free bovine serum albumin (BSA) was from Calbiochem (San
Diego, CA). 8-Bromoadenosine 3’:5’-cAMP was from Sigma-Aldrich (St. Louis, MO).
Fetal bovine serum, Dulbecco’s modified Eagle medium and phosphate-buffered saline were from Invitrogen Life Technologies (Carlsbad, CA). RAW 264.7 mouse
44 macrophage cells were from American Type Culture Collection (Manassas, VA). All
chemicals reagents were of the highest quality available.
Methods
Construction of the expression vector for rat apoA-IV.
The rat apoA-IV cDNA contained in a pSP65 maintenance vector (Promega,
Madison, WI) was a gift from Dr. David Hui at the University of Cincinnati. An Afl III
restriction site was engineered immediately 5’ of the coding sequence for the mature
apoA-IV using the QuickChangeTM site-directed mutagenesis kit (Stratagene, La Jolla,
CA). As depicted in Figure 2-1, the gene was excised from the maintenance vector at the
Afl III and BamH I sites and ligated into the pET30 (Novagen, Madison, WI) expression
vector between the Nco I and BamH I sites in the pET30 multiple cloning region (Afl III
and Nco I have compatible cohesive ends). This positioned apoA-IV immediately behind a 13-kDa leader sequence containing a 6-amino acid histidine tag and an enteropeptidase cleavage site immediately upstream of the coding sequence for the protein.
Enteropeptidase was selected because it cleaves at the extreme C-terminal end of its recognition site, leaving the target protein with its natural N-terminal sequence.
However, preliminary expression experiments indicated that cleavage of the histidine tag by enterokinase would be problematic because the enterokinase cleaved apoA-IV at numerous sites within the sequence in addition to removing the histidine tag. This apparent lack of specificity of enterokinase has been observed before for human apoA-I
(30) and appears to be a common feature of bacterially-expressed apolipoproteins. To resolve this problem, we replaced the coding region for enterokinase cleavage site in pET
30 with an Igase proteolytic cleavage site. Igase is substantially more specific than
45 enterokinase and was found to cleave the histidine tag without degrading apoA-IV.
Because of the characteristics of the Igase cleavage site, the expressed apoA-IV has an additional Thr-Pro- on the N-terminus.
Cloning and mutagenesis of rat apoE.
A rat liver cDNA library was used to clone rat apoE. PCR primers were first designed to amplify the coding region of rat apoE (Forward 5’-
GCCCTGCTGTTGGTCCC-3’ and Reverse 5’-GCGGCAGGGCGTAGGTGAGGG-3’) and PCR amplification was carried out using a Perkin-Elmer thermocycler: 1 min hold at
94˚C; 30 rounds of 94˚C for 1.5 min, 55˚C for 1.5 min, 72˚C for 1.5 min; followed by a hold for 10 min at 72˚C to ensure full extension. Cloned Pfu DNA polymerase
(Stratagene, La Jolla, CA) was used for this and the subsequent reaction. Following amplification, forward and reverse flap primers were designed to contain restriction enzyme sites. The forward primer also contained the sequence encoding for a proteolytic cleavage site specific for Igase that is required for removing the histidine tag from the purified protein. The forward primer sequence is shown in Fig. 2-8 with the flap region containing the Nco I restriction site in italics and the Igase site in underlined bold lettering. The clamp region is shown in capital letters and matches the forward sequence for apoE. The reverse primer contained a clamp region and a flap region containing a
Hind III restriction enzyme site. The amplified DNA from the first PCR reaction was used as a template for these flap primers. The apoE DNA and empty pET30 vector were cut with Nco I and Hind III, ligated together and the entire construct was completely
46 sequenced. Only the DNA for the mature sequence of rat apoE was included in the expression vector.
Protein expression and purification for both apoA-IV and apoE.
To generate recombinant apoA-IV and apoE, we modified our highly efficient E. coli expression system used previously for human apoA-I (29). The expression construct was transfected into E. coli BL-21 cells and grown on Luria-Bertani (LB) agar plates that contained 30 µg/ml kanamycin for 16 h at 37˚C. A single colony from the plate was inoculated into 10 ml cultures that included 30 µg/ml kanamycin and grown overnight in a 37˚C shaking incubator (225 rpm). Next, 100 ml culture flasks containing kanamycin had 1 ml of the confluent 10 ml culture tubes added and were grown in the shaking incubator for approximately 3 h. The cells were grown to an A600 between 0.6-0.7.
Protein expression was driven by the T7 promoter upon exposure to 50 µl of 1 M IPTG for 2 h. After expression, the bacterial cells were pelleted at 9700g in a Sorvall SLA1500 for 10 min, the supernatant was removed and the cells were frozen at -20˚C overnight.
The cells were resuspended in 4 ml of 1X histidine (His)-bind buffer (5mM imidazole, 0.5 M NaCl, 20 mM Tris:HCl) per 100 ml of original culture with a final concentration of protease inhibitors; 100 µM PMSF, 20 µM leupeptin and 1 µM pepstatin A. The cells were lysed with a Model 550 Sonic Dismembrator at level 5
(Fisher, Pittsburgh, PA) 3 times for 1 min with a 1 min incubation on ice in between.
The lysed cells were then spun at 28,384g for 20 min in a Sorvall SLA600, leaving the recombinant apoA-IV or apoE in the supernatant. The presence of the histidine tag allowed for purification by using a His-bind column (Ni2+-affinity column). The
47 supernatant was passed through a 0.45 µM filter and added to a prepared His-bind
column (Novagen: #69673-3) poured with HisBind Resin. The columns were prepared
by packing with 4 ml resin by gravity flow. Using a vacuum manifold, the columns were
washed with 15 ml H2O, followed by 25 ml 1X His-charge buffer (0.05 M nickel sulfate) and finally 15 ml 1X His-bind buffer. At this step, the filtered supernatant was added and allowed to flow through. The column was then washed with 30 ml of 1X His-bind buffer and 30 ml of 1X His-wash buffer (60 mM imidazole, 0.5 M NaCl, 20 mM Tris-HCl).
Next, the protein was collected with 9 ml of 1X His-elution buffer (1 M imidazole, 0.5 M
NaCl, 20 mM Tris-HCl). The column was then regenerated by adding 20 ml of 1X Strip buffer (100 mM EDTA, 0.5 M NaCl, 20 mM Tris-HCl). The protein sample was concentrated using Centriprep centrifugal concentrators (30kDa cutoff) to a concentration of approximately 5 mg/ml and then dialyzed into standard Tris buffer (STB, 10mM Tris-
HCl, 1mM EDTA, 150mM NaCl and 0.2% NaN3).
The protein concentration was determined and the protease, Igase, was added at a mass ratio of 1:5000 (Igase/uncut protein) for 24 h at 37˚C in a shaking incubator. Igase is an IgA protease from Neisseria gonorrhoeae that recognizes and cleaves the sequence
Ala-Pro-Arg-Pro-Pro-Thr-Pro. It cleaves the leader portion along with the histidine tag of the protein, leaving the mature form of apoA-IV or apoE with an additional Thr-Pro on the N-terminus. A denaturing gel was run to ensure complete cleavage of the protein
which occurs at 24 h. In a time-course experiment, we found that 24 h was best for
removal of the His-tag from the recombinant apoE (data not shown). However, for
recombinant human apoA-I, full cleavage occurs at 16 h. The time requirement
48 difference is probably due to the overall availability of the cleavage sequence to the Igase
because of the protein’s overall conformation.
Next, 10X His-bind buffer was added to the sample to make it 1X and the sample
was applied to a prepared His-bind column a second time to separate the mature protein
from the cleaved tag. The sample was added to the column and the flowthrough was
collected, then the column was washed with 9 ml of 1X His-bind buffer and finally with
9 ml of 1X His-wash buffer (both washes were collected). The mature cut protein was located in the flowthrough, His-bind and His-wash buffer fractions. Any uncut protein could be collected by washing the column with His-elute buffer. The fractions were
combined, concentrated and dialyzed into STB or PBS for storage.
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE).
SDS-PAGE was performed using an 18% polyacrylamide gel slab manufactured
by Bio-Rad (Hercules, CA). The samples were treated with pure water and loading
buffer at 100° C for 3 min prior to electrophoresis. The gels were stained with
Coomassie Brilliant Blue.
Immunoblot analysis for apoA-IV.
Proteins (50 ng) were separated by 12% polyacrylamide gel electrophoresis,
transferred to Immun-Blot PVDM membrane (Bio-Rad) and non-specific sites were
blocked (20 mM Tris-base, 0.137 M NaCl, 0.1% (v/v) Tween-20 and 5% (w/v) powdered
milk, pH 7.6). The membrane was incubated with the primary polyclonal antibody goat
against rat apoA-IV (1:3000 dilution), followed by incubation with rabbit anti-goat IgG
49 coupled to horseradish peroxidase at a dilution of 1:2000. Bound conjugated antibodies were visualized by using an enhanced chemiluminescence detection system (Amersham
Life Science, Princeton, NJ).
Electrospray mass spectrometry.
Intact recombinant apoA-IV was dissolved in ammonium bicarbonate buffer at pH 7.8 at an approximate concentration of 1.5 mg/ml. This sample was analyzed by matrix assisted laser desorption ionization (MALDI) spectrometry on an APB Perseptive
Voyager time-of-flight instrument (located at Proctor and Gamble Pharmaceuticals,
Mason OH). The average molecular mass of the protein was calculated from the protonated ions in the mass spectrum using the manufacturer’s software. The mass spectrometer was set to detect ions in the range of 4000 to 45,000 mass units, with an approximate error of 0.2 % of the measured mass.
N-terminal amino acid sequence analysis.
N-terminal amino acid sequence was determined by automated Edman degradations using an applied ABI Procize Sequencer (also located at Proctor and
Gamble Pharmaceuticals).
Lipoprotein particle reconstitution.
The ability of recombinant apoA-IV to form reconstituted lipoprotein particles was determined by the cholate dialysis method. Briefly, rHDL particles were prepared as described previously (31) using palmitoyloleoyl phosphatidylcholine (POPC). Initial
50 lipid to protein molar ratios ranged from 50 to 150:1. The resulting particle
hydrodynamic diameters were measured by gradient native polyacrylamide electrophoresis (32). The number of apolipoprotein molecules per rHDL complex was determined by BS3 cross-linking (33).
Biological activity of recombinant apoA-IV.
Rat recombinant (1-4 µg), natural apoA-IV (4 µg) or control solution (saline) were injected into the 3rd cerebral ventricle (i3vt) of fasted male Sprague-Dawley rats using chronically implanted i3vt cannulas, as described previously (34). These rats were individually housed under stable environmental conditions with a 12-h dark/12-h light cycle with free access to pelleted food and water. All procedures were performed in accordance with the guidelines of the University Institutional Animal Care and Use
Committee at the University of Cincinnati. Following recovery from surgery as assessed by daily food intake and body weight gain, the animals were fasted for 24 h and assigned to one of five body weight-matched groups. At the onset of dark, rats received i3vt injections (4 µl) containing saline (n = 11), 1 µg (n = 8), 2 µg (n = 8), or 4 µg (n = 10) recombinant apoA-IV, or 4 µg native rat apoA-IV (n = 9), with saline as the vehicle.
Rats were visually observed, and food and water consumption was measured at 0.5, 1, 2,
4, and 24 h post-injection.
Conditioned taste aversion (CTA).
51 Rats were placed on a water restriction schedule where they could consume tap
water from either of two metal sipper tubes attached to 500-ml drinking bottles during a
one-hour test session on 10 consecutive baseline days with food available at all times.
Following this baseline period, the rats were rank ordered according to mean water
intakes over baseline trials 7-10 and randomly assigned to four groups based on
quantitatively similar water intake and mean body weight. Each group was then
randomly assigned to one of four drug treatment groups (6 rats per group). On Day 11,
the conditioning day, all rats were presented with a 0.15% saccharin solution instead of
tap water in both bottles during the one-hour drinking session. Immediately following
the one-hour access to saccharin, the bottles were removed and the rats were injected (IP)
with 0.15 M LiCl, at a dose of 20 ml/kg, that has been demonstrated to produce a strong
CTA (35), or i3vt injected with 4 µl saline, 4 µg native rat apoA-IV, or 4 µg recombinant rat apoA-IV. On Days 12-13, animals received tap water during the one-hour drinking sessions. On the test day (Day 14) the rats were given a choice of saccharin solution and water in pre-weighed bottles for one-hour. After each rat had licked one bottle for a second, that bottle was removed so that the rat could sample the second bottle for a second. Then both bottles were returned at the same time. The purpose was to give each rat an opportunity to sample the fluid from both bottles. Intake was measured after one hour, and a saccharin preference ratio was calculated as the amount of saccharin consumed divided by the total consumption of both liquids during the one-hour trial.
Statistical analysis.
52 Results are presented as mean ± SE. Data were analyzed by one-way analysis of variance (ANOVA) followed by Tukey’s test. Differences were considered significant when the probability of the difference occurring by chance was less than 5 in 100, 2- tailed (P < 0.05).
Native protein purification.
Rat plasma was obtained from abdominal aorta blood of Long Evans rats. The samples were centrifuged at a density of > 1.065 g/ml to isolate any apoB containing lipoproteins. After dialysis to remove KBr that was used to set the density, the rat apolipoproteins present in the HDL density and greater were then separated by SDS-
PAGE and purified by gel extraction as described previously (36).
Western blot analysis for apoE.
Purified proteins (100 ng) were loaded on an 18% denaturing polyacrylamide gel and separated by electrophoresis and transferred to Immun-Blot PVDF membrane. Non- specific sites were blocked overnight (20 mM Tris-base, 0.137 M NaCl, 0.1% v/v Tween-
20 and 5% w/v powdered milk, pH 7.6). The membrane was then incubated for 2 h with the primary polyclonal goat against native rat apoE at a dilution of 1:3000. The membrane was washed and incubated with rabbit anti-goat IgG coupled to horseradish peroxidase for 45 min (1:2000) and subsequently washed. An enhanced chemiluminescence (ECL) system was used to visualize the bound antibodies following manufacturer’s protocol.
53 Circular dichroism (CD) spectroscopy.
The average secondary structural content of recombinant and native rat apoE were
determined by obtaining CD spectra at room temperature using a Jasco S-720
spectropolarimeter. The samples were analyzed at 0.1 mg/ml in 20 mM phosphate
buffer. The α-helix content was calculated from the mean residual ellipticity at 222 nm,
as described (37).
Dimyristoylphosphatidylcholine (DMPC) liposome solubilization.
DMPC was dried in a glass tube and STB was added to a final concentration of 5
mg/ml. Multilamellar liposomes were formed by a brief bath sonication. The turbidity
clearance experiments were performed in an Amersham Biosciences Ultraspec 4000 at a
constant temperature of 24.5˚C maintained by a circulating water bath. The absorbance
was recorded over 30 minutes at 325 nm after protein samples were added (5:1 mass ratio
of DMPC/protein). The Y-axis of the graph is the ratio of the absorbance at any given
time point (OD) to the initial absorbance at time zero (ODo) (38). For a more detailed description of the assay, please refer to (39).
Cholesterol efflux studies.
The transformed mouse macrophage cell line RAW264.7 was maintained in
growth medium (Dulbecco’s modified Eagle medium, 10% fetal bovine serum, and 50
µg/ml gentamycin). The cells were grown to 70-80% confluency in 48-well plates. The
cells were then labeled with 1.0 µCi/ml [3H] cholesterol in efflux medium (DMEM +
0.2% BSA) with or without 0.3 mM 8-bromo-cAMP for 24 h. After 24 h, the cells were
54 washed and efflux medium containing differing concentrations of protein acceptors were
added to the wells, maintaining the cAMP concentration used in the labeling media.
After 8 h, the samples were filtered through a 0.45-µm filter to remove any floating cells
and analyzed by liquid scintillation counting. The percent cholesterol efflux was
calculated by dividing the counts in the medium by total cell counts at time zero at 8
hours (40).
RESULTS: RAT APOA-IV
The overexpression and purification of apoA-IV resulted in 40 mg of pure protein
from one liter of culture. Figure 2-2 depicts the SDS-PAGE analysis of the recombinant
protein produced at different steps of the purification process. We purposely overloaded
the gel to indicate the purity of the recombinant apoA-IV (50 µg/well). Lane 1 depicts the recombinant apoA-IV that was expressed and isolated on the His-bind column prior to His-tag cleavage. As purification proceeds, a 43 kDa, similar to the molecular weight of native rat apoA-IV, was visualized (Lanes 2 and 3).
The isolated recombinant protein was reactive with a polyclonal antibody directed against rat apoA-IV (Figure 2-3). The antibody was also able to clearly recognize the
His-tag version of the protein. To further characterize the protein, we performed electrospray mass spectrometry analysis. We found that the recombinant protein had a major mass peak with molecular mass of 42642 Da (Figure 2-4), in agreement with that predicted from the amino acid sequence of rat apoA-IV (42635, including the Thr-Pro- on the N-terminus) (41). The difference of 7 Da is well within the expected variation for a protein of this size. Analysis of the N-terminal sequence of the recombinant protein
55 yielded the amino acid sequence of Thr-Pro-Glu-Val-Thr-Ser-Asp-Gln-Val-Ala-Asn-Val-
Met-Trp-Asp, which is identical to the N-terminal sequence of the natural protein, with the exception of the Thr-Pro-on the N-terminus (see Methods) (1 to 13 amino acids).
To determine if the recombinant protein was capable of forming reconstituted lipoproteins, we performed a series of sodium cholate dialysis reconstitutions (31). The particles were reconstituted with a lipid to protein molar ratio varying from 50 to 150:1.
Figure 2-5 depicts a non-denaturing polyacrylamide gradient (8-24%) gel indicating the sizes of the resulting complexes. The lipid-free protein exhibited two distinct bands of about 7.0 nm and 7.8 nm. These corresponded to the monomeric and dimeric forms of apoA-IV, respectively. When lipid was present at a 50:1 molar ratio, there was no apparent formation of lipoprotein particles of increased diameter. However, when the lipid ratios were increased to 100 and 150:1, a band corresponding to a lipid-protein complex appeared with a diameter of about 12.0 nm. The size of this band was not affected by further increases in phospholipid:protein ratio (data not shown). These data indicate that apoA-IV can form intact lipoprotein particles similar to those found using native rat apoA-IV and recombinant human apoA-I.
The ability of recombinant apoA-IV to inhibit food intake in fasted rats was then assessed. When injected prior to the onset of the dark, recombinant apoA-IV significantly reduced 4-h food intake in a dose dependent manner. By 24 h, this effect was only observed at the higher doses (2 and 4 µg). Rat native apoA-IV at dose of 4 µg also significantly suppressed food intake over 4 h, but this effect was absent by 24 h
(Figure 2-6). Although native rat apoA-IV appeared to be slightly more potent than recombinant rat apoA-IV over 4 hours, the difference was not statistically significant.
56 The important point, however, is that these studies demonstrate that recombinant apoA-
IV is biologically active when i3vt administered.
When apoA-IV was injected into the 3rd ventricle, no significant aversive effects
were observed. Rats that received saccharin paired with i3vt injection of either natural
rat apoA-IV (4 µg) or recombinant rat apoA-IV (4 µg) did not have a significantly decreased preference for saccharin on the test day compared with controls. By contrast, animals that received LiCl paired with saccharin consumed significantly less saccharin than controls on the test day as seen in Figure 2-7.
DISCUSSION: RAT APOA-IV
In the present report, we describe the use of an E. Coli expression system for the production of a high-level apoA-IV protein (40 mg/L of culture medium). The recombinant protein expression was induced with IPTG, extracted from the cytoplasm of cells, cleaved from pET-30 vector by Igase, purified by His-bind columns and analyzed by SDS-PAGE. When the recombinant protein was probed with a polyclonal apoA-IV antibody, a band of mass corresponding to that of the rat apoA-IV standard was recognized, indicating that the recombinant protein is apoA-IV. Electrospray mass spectrometry analysis confirmed that the isolated recombinant apoA-IV is correctly expressed in E. coli because the recombinant protein only had one major mass peak with a molecular mass of 43 kDa, which was expected from the amino acid sequence of rat apoA-IV. N-terminal sequence analysis of the recombinant protein demonstrated that the amino acid sequence of recombinant apoA-IV is identical to the N-terminal sequence of
57 the natural protein. All of these results strongly indicate the successful synthesis of the rat apoA-IV.
We then assessed the biological activity of the recombinant protein using an in vivo behavioral assay, and compared its potency with native apoA-IV prepared from rat plasma. Recombinant rat apoA-IV elicited a dose-dependent reduction in food intake in fasted rats after i3vt injection, but no statistically significant reduction was observed within the first 2 hours after i3vt injection. Compared to native rat apoA-IV, recombinant rat apoA-IV appeared to elicit a slightly smaller but longer-lasting suppression of food intake. This difference may be partially explained by a different affinity to apoA-IV binding protein (or receptor) between natural and recombinant rat apoA-IV. Another possibility would be that recombinant apoA-IV may have a lower dissociation rate from the binding protein or a lower metabolism rate than natural apoA-IV, although this has yet to be demonstrated. The two amino acids (Thr-Pro) which remain at the amino terminus may not affect apoA-IV physiological function, but may alter the binding ability with binding protein and/or the stability of recombinant apoA-IV. Another implication of the present data is that the carbohydrate moiety of the circulating natural rat apoA-IV is not required for its biological activity in inhibiting food intake since this moiety is lacking on the recombinant protein.
Although the ability of exogenous apoA-IV to modulate feeding behavior is consistent with the suggestion that this protein is an endogenous regulatory agent, central administration of the protein may have aversive side effects, which could explain the anorexia. To assess this, exposure to a saccharin taste was immediately followed by central administration of apoA-IV to determine if this protein would produce a
58 conditioned taste aversion (CTA). Doses of natural and recombinant apoA-IV (4 µg) that produced comparable reductions of short-term food intake did not generate a CTA
(Figure 2-7), suggesting that malaise is not a factor in the suppressive activity of the protein on food intake. As a positive control, a dose of lithium chloride that suppresses short-term food intake comparable (35) elicited a strong CTA.
In conclusion, this is the first report that recombinant rat apoA-IV can be produced in high yield from an E. coli system. We have demonstrated that the recombinant apoA-IV appears intact and that the molecular weight is compatible with that of natural rat apoA-IV. We have further demonstrated that it is immunologically and
functionally indistinguishable from plasma-derived rat apoA-IV. The E. coli expression
system described in the present study may be a valuable tool for studying apoA-IV
structure-function relationships using site-directed mutagenesis and for generation of a
large-scale amount of apoA-IV for further biological and physiological studies of this
protein.
RESULTS: RAT APOE
Figure 2-9A shows an 18% SDS-PAGE gel stained with Coomassie blue that
summarizes the purification of rat apoE at various stages of the expression/purification.
The cell lysate from bacteria over expressing rat apoE is shown in lane 2. Recombinant
rat apoE exhibited a high degree of induction with approximately 25 mg of recombinant
apoE per L of culture media from roughly 6 grams of wet weight cells per L at harvest.
Following sonication, the cellular proteins were purified on a His-bind column and the
resulting His-tagged rat apoE is shown in lane 3. It is clear that the recombinant protein
59 was cleanly isolated from the cellular proteins resulting in high purity samples. The protein was then cleaved with Igase, removing the N-terminal His-tag and leaving the mature protein with an additional Thr-Pro on the N-terminus. The mature cut protein was then purified away from the His-tag by applying the sample to the His-bind column a second time (Lane 4). The total protein, approximate purity, estimated protein product and approximate yield are given in Table 2-1 and this shows that both the yield and the purity of the recombinant apoE are excellent. Figure 2-9B shows a western blot analysis of recombinant rat and native rat apoE in lanes 2 and 3 respectively. Both proteins were recognized by goat polyclonal native rat apoE anti-serum and generated a single band present at 34 kDa. The predicted size for the recombinant rat apoE was 34,053 Da. This was verified by MALDI mass spectrometry at the Genome Research Institute, University of Cincinnati Proteomics Core which generated essentially a single peak with a mass of
34,077 Da. The discrepancy between the theoretical value and the experimentally derived value was within the expected error of the machine.
To determine the stability of the purified recombinant rat apoE, samples of the protein were stored in STB at both 4˚C and -20˚C for 3 weeks each. These were compared to a sample that had been mixed with SDS loading buffer and stored at -20˚C from the beginning of the experiment. Figure 2-10 is an SDS gel showing the proteins under these various storage conditions. There was no difference between the three different storage conditions over the 3 week period. This demonstrates that the protein is stable in buffer and does not precipitate over time. Furthermore, this indicates that the purity is such that contaminating proteases are not present in high enough amounts to degrade the protein significantly over 3 weeks.
60 Having established that the expression system produces significant amounts of
stable protein, we next characterized the protein with respect to structure and function.
Samples of recombinant and native rat apoE were analyzed using circular dichroism (CD)
to compare average secondary structure content. Figure 2-11 shows that native and
recombinant rat apoE exhibit nearly identical far-UV CD spectra in both overall shape
and the magnitudes of the minima at 208 and 222 nm, which are characteristic of a highly
helical protein. The CD values were used to determine α-helical content for each of the proteins (42). The recombinant and native rat helical contents were very similar at 60 ±
3% and 59 ± 1 % respectively (n = 6). These data indicate that the recombinant and native apoE contained a similar overall secondary structure content.
One of the main functions of apolipoproteins is to interact with and bind to lipids.
To determine whether recombinant and native rat apoE were similar in their ability to bind and reorganize lipid, a DMPC lipid clearance assay was performed. In this assay,
DMPC multi-lamellar liposomes are generated in buffer, resulting in a turbid solution.
Protein samples are then added and the turbidity clears as the protein binds and
reorganizes the liposomes into small discoidal structures. Figure 2-12 shows that over
time, DMPC alone with no protein clears at a very slow rate. However, the addition of
apoE caused the solution to clear more rapidly. Both recombinant and native rat apoE
were able to clear lipid at a similar rate.
Similar to the related protein apolipoprotein A-I, apoE has been shown to have the
ability to promote the efflux of cholesterol via the ABCA1-pathway (43). Therefore, we
tested the ability of the recombinant and native apoE proteins to promote cholesterol
efflux from RAW264 mouse macrophages. This system was used because it has been
61 clearly established that the addition of cAMP to the cells causes an increase in ABCA1 at
the cell surface (44;45). Following 24 h treatment of the cells with labeled cholesterol
and cAMP, lipid-free protein acceptors were added for 8 hours to determine their ability
to promote cholesterol efflux by the ABCA1-mediated pathway. Figure 2-13 shows
ABCA1-mediated cholesterol efflux to recombinant and native rat apoE over a range of
concentrations. Again, the recombinant and the native rat apoE were nearly super
imposable in terms of their concentration effect on cholesterol efflux.
DISCUSSION: RAT APOE
We have developed a bacterial expression system using the pET30 vector that is
highly efficient and effective at producing rat apoE. The recombinant rat apoE compares
favorably both structurally and functionally to native rat apoE purified from plasma. This
lends confidence that future studies using the recombinant protein will faithfully
represent the functionality of the native protein. We should also note that we applied the same strategy to produce recombinant mouse apoE and its characteristics compared favorably to the native rat apoE (data not shown). As stated above, we have determined that centrally administered apoE has inhibitory effects on rat food intake. Large amounts of apoE are required for studying its effects on food intake more thoroughly. Relying solely on the purification of native protein from rat plasma, these types of experiments would not be possible on a large scale. However, this recombinant expression system will allow these tests to be performed in a timely manner. Furthermore, the expression system will allow for the convenient engineering of apoE mutants that can be used to isolate particular regions of apoE that may be involved in food intake regulation.
62 REFERENCES
1. Green,P.H., Glickman,R.M., Riley,J.W., and Quinet,E. (1980) Human apolipoprotein A-IV. Intestinal origin and distribution in plasma, J. Clin. Invest 65, 911-919.
2. Wu,A.L. and Windmueller,H.G. (1979) Relative contributions by liver and intestine to individual plasma apolipoproteins in the rat, J. Biol. Chem. 254, 7316-7322.
3. Dvorin,E., Gorder,N.L., Benson,D.M., and Gotto,A.M., Jr. (1986) Apolipoprotein A-IV. A determinant for binding and uptake of high density lipoproteins by rat hepatocytes, J. Biol. Chem. 261, 15714-15718.
4. Goldberg,I.J., Scheraldi,C.A., Yacoub,L.K., Saxena,U., and Bisgaier,C.L. (1990) Lipoprotein ApoC- II activation of lipoprotein lipase. Modulation by apolipoprotein A-IV, J. Biol. Chem. 265, 4266- 4272.
5. Okumura,T., Taylor,I.L., Fukagawa,K., Tso,P., and Pappas,T.N. (1995) Apolipoprotein A-IV acts centrally in the brain to reduce the severity of gastric ulceration in the rat, Brain Res. 673, 153-156.
6. Okumura,T., Fukagawa,K., Tso,P., Taylor,I.L., and Pappas,T.N. (1996) Apolipoprotein A-IV acts in the brain to inhibit gastric emptying in the rat, Am. J. Physiol 270, G49-G53.
7. Fujimoto,K., Fukagawa,K., Sakata,T., and Tso,P. (1993) Suppression of food intake by apolipoprotein A-IV is mediated through the central nervous system in rats, J. Clin. Invest 91, 1830- 1833.
8. Fujimoto,K., Cardelli,J.A., and Tso,P. (1992) Increased apolipoprotein A-IV in rat mesenteric lymph after lipid meal acts as a physiological signal for satiation, Am. J. Physiol 262, G1002-G1006.
9. Fujimoto,K., Machidori,H., Iwakiri,R., Yamamoto,K., Fujisaki,J., Sakata,T., and Tso,P. (1993) Effect of intravenous administration of apolipoprotein A-IV on patterns of feeding, drinking and ambulatory activity of rats, Brain Res. 608, 233-237.
10. Qin,X., Swertfeger,D.K., Zheng,S., Hui,D.Y., and Tso,P. (1998) Apolipoprotein AIV: a potent endogenous inhibitor of lipid oxidation, American Journal of Physiology 274, H1836-H1840.
11. Duverger,N., Tremp,G., Caillaud,J.M., Emmanuel,F., Castro,G., Fruchart,J.C., Steinmetz,A., and Denefle,P. (1996) Protection against atherogenesis in mice mediated by human apolipoprotein A-IV, Science 273, 966-968.
12. Ostos,M.A., Conconi,M., Vergnes,L., Baroukh,N., Ribalta,J., Girona,J., Caillaud,J.M., Ochoa,A., and Zakin,M.M. (2001) Antioxidative and antiatherosclerotic effects of human apolipoprotein A- IV in apolipoprotein E-deficient mice, Arterioscler. Thromb. Vasc. Biol. 21, 1023-1028.
13. Liu,M., Doi,T., Shen,L., Woods,S.C., Seeley,R.J., Zheng,S., Jackman,A., and Tso,P. (2001) Intestinal satiety protein apolipoprotein AIV is synthesized and regulated in rat hypothalamus, Am. J. Physiol Regul. Integr. Comp Physiol 280, R1382-R1387.
14. Shore,V.G. and Shore,B. (1973) Heterogeneity of human plasma very low density lipoproteins. Separation of species differing in protein components, Biochemistry 12, 502-507.
15. Elshourbagy,N.A., Liao,W.S., Mahley,R.W., and Taylor,J.M. (1985) Apolipoprotein E mRNA is abundant in the brain and adrenals, as well as in the liver, and is present in other peripheral tissues of rats and marmosets, Proc. Natl. Acad. Sci. U. S. A 82, 203-207.
63 16. Brown,M.S. and Goldstein,J.L. (1986) A receptor-mediated pathway for cholesterol homeostasis. [Review], Science 232, 34-47.
17. Mahley,R.W. (1988) Apolipoprotein E: cholesterol transport protein with expanding role in cell biology. [Review], Science 240, 622-630.
18. Boyles,J.K., Pitas,R.E., Wilson,E., Mahley,R.W., and Taylor,J.M. (1985) Apolipoprotein E associated with astrocytic glia of the central nervous system and with nonmyelinating glia of the peripheral nervous system, J. Clin. Invest 76, 1501-1513.
19. Ignatius,M.J., Gebicke-Harter,P.J., Skene,J.H., Schilling,J.W., Weisgraber,K.H., Mahley,R.W., and Shooter,E.M. (1986) Expression of apolipoprotein E during nerve degeneration and regeneration, Proc. Natl. Acad. Sci. U. S. A 83, 1125-1129.
20. Sheng,H., Laskowitz,D.T., Bennett,E., Schmechel,D.E., Bart,R.D., Saunders,A.M., Pearlstein,R.D., Roses,A.D., and Warner,D.S. (1998) Apolipoprotein E isoform-specific differences in outcome from focal ischemia in transgenic mice, J. Cereb. Blood Flow Metab 18, 361-366.
21. Strittmatter,W.J., Saunders,A.M., Schmechel,D., Pericak-Vance,M., Enghild,J., Salvesen,G.S., and Roses,A.D. (1993) Apolipoprotein E: high-avidity binding to beta-amyloid and increased frequency of type 4 allele in late-onset familial Alzheimer disease, Proc. Natl. Acad. Sci. USA 90, 1977-1981.
22. Corder,E.H., Saunders,A.M., Strittmatter,W.J., Schmechel,D.E., Gaskell,P.C., Small,G.W., Roses,A.D., Haines,J.L., and Pericak-Vance,M.A. (1993) Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer's disease in late onset families [see comments], Science 261, 921- 923.
23. Saunders,A.M., Strittmatter,W.J., Schmechel,D., George-Hyslop,P.H., Pericak-Vance,M.A., Joo,S.H., Rosi,B.L., Gusella,J.F., Crapper-MacLachlan,D.R., Alberts,M.J., and . (1993) Association of apolipoprotein E allele epsilon 4 with late-onset familial and sporadic Alzheimer's disease, Neurology 43, 1467-1472.
24. Knight,D.S., Mahajan,D.K., and Qiao,X. (2001) Dietary fat up-regulates the apolipoprotein E mRNA level in the Zucker lean rat brain, Neuroreport 12, 3111-3115.
25. Mazur,A., Cardot,P., Felgines,C., Remesy,C., and Rayssiguier,Y. (1991) Apolipoprotein E gene expression in the liver of genetically obese Zucker rats: the effect of high-fiber diet, Horm. Metab Res. 23, 347-348.
26. Vogel,T., Weisgraber,K.H., Zeevi,M.I., Ben Artzi,H., Levanon,A.Z., Rall,S.C., Jr., Innerarity,T.L., Hui,D.Y., Taylor,J.M., Kanner,D., and . (1985) Human apolipoprotein E expression in Escherichia coli: structural and functional identity of the bacterially produced protein with plasma apolipoprotein E, Proc. Natl. Acad. Sci. U. S. A 82, 8696-8700.
27. Morrow,J.A., Arnold,K.S., and Weisgraber,K.H. (1999) Functional characterization of apolipoprotein E isoforms overexpressed in Escherichia coli, Protein Expr. Purif. 16, 224-230.
28. Pillot,T., Barbier,A., Visvikis,A., Lozac'h,K., Rosseneu,M., Vandekerckhove,J., and Siest,G. (1996) Single-step purification of two functional human apolipoprotein E variants hyperexpressed in Escherichia coli, Protein Expr. Purif. 7, 407-414.
29. Panagotopulos,S.E., Witting,S.R., Horace,E.M., Nicholas,M.J., and Sean,D.W. (2002) Bacterial expression and characterization of mature apolipoprotein A-I, Protein Expr. Purif. 25, 353-361.
64 30. Panagotopulos,S.E., Horace,E.M., Maiorano,J.N., and Davidson,W.S. (2001) Apoliprotein A-I adopts a belt-like orientation in reconstituted high density lipoproteins, J. Biol. Chem.
31. Jonas,A. (1986) Reconstitution of high-density lipoproteins, Methods in Enzymology 128, 553-582.
32. Davidson,W.S., Rodrigueza,W.V., Lund-Katz,S., Johnson,W.J., Rothblat,G.H., and Phillips,M.C. (1995) Effects of acceptor particle size on the efflux of cellular free cholesterol, J. Biol. Chem. 270, 17106-17113.
33. Jonas,A., Kezdy,K.E., and Wald,J.H. (1989) Defined apolipoprotein A-I conformations in reconstituted high density lipoprotein discs, J. Biol. Chem. 264, 4818-4824.
34. Liu,M., Doi,T., Shen,L., Woods,S.C., Seeley,R.J., Zheng,S., Jackman,A., and Tso,P. (2001) Intestinal satiety protein apolipoprotein AIV is synthesized and regulated in rat hypothalamus, Am. J. Physiol Regul. Integr. Comp Physiol 280, R1382-R1387.
35. Thiele,T.E., Van Dijk,G., Campfield,L.A., Smith,F.J., Burn,P., Woods,S.C., Bernstein,I.L., and Seeley,R.J. (1997) Central infusion of GLP-1, but not leptin, produces conditioned taste aversions in rats, Am. J. Physiol 272, R726-R730.
36. Hayashi,H., Nutting,D.F., Fujimoto,K., Cardelli,J.A., Black,D., and Tso,P. (1990) Transport of lipid and apolipoproteins A-I and A-IV in intestinal lymph of the rat, J. Lipid Res. 31, 1613-1625.
37. Chen,Y.H., Yang,J.T., and Martinez,H.M. (1972) Determination of the secondary structures of proteins by circular dichroism and optical rotatory dispersion, Biochemistry 11, 4120-4131.
38. Weinberg,R.B., Anderson,R.A., Cook,V.R., Emmanuel,F., Denefle,P., Hermann,M., and Steinmetz,A. (2000) Structure and interfacial properties of chicken apolipoprotein A-IV, J. Lipid Res. 41, 1410-1418.
39. Pearson,K., Saito,H., Woods,S.C., Lund-Katz,S., Tso,P., Phillips,M.C., and Davidson,W.S. (2004) Structure of human apolipoprotein A-IV: a distinct domain architecture among exchangeable apolipoproteins with potential functional implications, Biochemistry 43, 10719-10729.
40. Panagotopulos,S.E., Witting,S.R., Horace,E.M., Hui,D.Y., Maiorano,J.N., and Davidson,W.S. (2002) The Role of Apolipoprotein A-I Helix 10 in Apolipoprotein-mediated Cholesterol Efflux via the ATP-binding Cassette Transporter ABCA1, J. Biol. Chem. 277, 39477-39484.
41. Boguski,M.S., Elshourbagy,N., Taylor,J.M., and Gordon,J.I. (1984) Rat apolipoprotein A-IV contains 13 tandem repetitions of a 22-amino acid segment with amphipathic helical potential, Proc. Natl. Acad. Sci. U. S. A 81, 5021-5025.
42. Chen,Y.H., Yang,J.T., and Martinez,H.M. (1972) Determination of the secondary structures of proteins by circular dichroism and optical rotatory dispersion, Biochemistry 11, 4120-4131.
43. Remaley,A.T., Stonik,J.A., Demosky,S.J., Neufeld,E.B., Bocharov,A.V., Vishnyakova,T.G., Eggerman,T.L., Patterson,A.P., Duverger,N.J., Santamarina-Fojo,S., and Brewer,H.B., Jr. (2001) Apolipoprotein specificity for lipid efflux by the human ABCAI transporter, Biochem. Biophys. Res. Commun. 280, 818-823.
44. Oram,J.F., Lawn,R.M., Garvin,M.R., and Wade,D.P. (2000) ABCA1 is the cAMP-inducible apolipoprotein receptor that mediates cholesterol secretion from macrophages, J. Biol. Chem. 275, 34508-34511.
65 45. Smith,J.D., Miyata,M., Ginsberg,M., Grigaux,C., Shmookler,E., and Plump,A.S. (1996) Cyclic AMP induces apolipoprotein E binding activity and promotes cholesterol efflux from a macrophage cell line to apolipoprotein acceptors, J. Biol. Chem. 271, 30647-30655.
46. Markwell,M.A., Haas,S.M., Bieber,L.L., and Tolbert,N.E. (1978) A modification of the Lowry procedure to simplify protein determination in membrane and lipoprotein samples, Anal. Biochem. 87, 206-210.
47. Sparks,D.L. and Phillips,M.C. (1992) Quantitative measurement of lipoprotein surface charge by agarose gel electrophoresis, J. Lipid Res. 33, 123-130.
66 TABLES:
Table 2-1: Purification of recombinant rat apolipoprotein E.
Purification Total Approximate Estimated Approximate stage protein purity of protein yield (mg)a apoE (%)b product (mg)c (%)d Total cell extracte 180 14 25 - Uncut apoE (His-tag present) 25 93 23 96 Cut apoE (His-tag removed) 22 92 20 80
a Total protein was determined by Markwell-Lowry protein assay (46). b Approximate purity was determined by densitometric scanning. c Estimated protein product was calculated by multiplying the total protein by the approximate purity. d Approximate yield was calculated by dividing the estimated protein product of both the uncut and cut apoE by the estimated protein product in the total cell extract. e One liter of culture was used for this prep and the total cell extract was collected from approximately 6 grams of wet weight cells.
67 FIGURES:
ApoA-IV
Figure 2-1: Schematic representation of the pET 30 vector. The rat apoA-IV gene was ligated into pET 30 vector utilizing Nco I and BamH I sites.
68
M 3 2 1 ApoA-IV 66 kDa with His-tag 45 kDa ApoA-IV
30 kDa
20 kDa
14 kDa His-tag
Figure 2-2: SDS polyacrylamide gel of recombinant apoA-IV produced by E.Coli at different steps of the purification process. The gel was a 12% SDS-PAGE gel stained with Coomassie blue. Lane M: low molecular weight standards with corresponding molecular weights in kDa. Lane 1: recombinant apoA-IV that was expressed and isolated on the His-bind column prior to His-tag cleavage. Lane 2: recombinant apoA-IV after
Igase cleavage of the His-tag sequence. Lane 3: recombinant apoA-IV after passage through the second His-bind column to remove the cleaved His-tag.
69
M1 2
ApoA-IV 61 kDa with His-tag 49 kDa ApoA-IV 36 kDa
Figure 2-3: Immunoblot analysis of recombinant rat apoA-IV expressed in E. coli.
The immunoblot was probed with goat polyclonal rat apoA-IV serum as described under
Materials and Methods. Lane M: molecular weight marker. Lane 1: purified
recombinant apoA-IV. Lane 2: expressed recombinant apoA-IV prior to His-tag
cleavage.
70
Figure 2-4: Electrospray mass spectrum of recombinant apoA-IV. The major mass peak (~43 Da) represents this protein although several minor peaks exist at lower and higher masses.
71
1 2 3 4 17 nm 12.2 nm
10.4 nm 8.2 nm
7.1 nm
8-24% Native PAGE
Figure 2-5: Native gradient polyacrylamide gel of reconstituted particles made with
recombinant apoA-IV. The gel depicted is an 8-24% gradient Phast gel stained with
Coomassie blue. The molecular weight markers are labeled with experimentally derived hydrodynamic diameters as reported by Sparks et al. (47). Lane 1: lipid-free recombinant apoA-IV; Lane 2: 50:1 mol PC/mol A-IV; Lane 3: 100:1; and Lane 4: 150:1.
72 30 Saline Rec. apoA-IV (1 µg) Rec. apoA-IV (2 µg) 25 Rec. apoA-IV (4 µg) * Rat apoA-IV (4 µg)
20
**
)
g
(
e 15 k
a
t
n
i
d o 10
o F * ****
5
0 0-0.5 h 0-1 h 0-2 h 0-4 h 0-24 h
Time after i3vt injection
Figure 2-6: Mean food intake following i3vt recombinant rat apoA-IV and natural rat apoA-IV. Fasted rats were injected immediately prior to lights out (Time 0) and food intake was measured at 0.5, 1, 2, 4 and 24 h after the injection of either 4 µl saline or different doses of recombinant rat apoA-IV or natural rat apoA-IV in 4 µl saline. Data are means ± S.E.M. Asterisks indicate significant differences from the saline group
(*P<0.05, **P < 0.01).
73
)
80
% (
n 70
o i
t 60 p
m
u 50 s
n 40
o
c
n 30 i r **
a 20
h c
c 10
a S 0
Saline LiCl Rat Rec. apoA-IV apoA-IV
Treatment
Figure 2-7: Conditioned taste aversion. Mean percent saccharin consumption for each
group of rats on the test day following treatment with either vehicle (saline), 0.15 M LiCl,
native rat apoA-IV (4 µg), or recombinant apoA-IV (4 µg). **P<0.01, compared with saline group.
74
Begin mature Igase rat apoE gene Nco I site cutting site 5’-at tgc tcc atg gct cca cgt cca ccg aca ccc GAG GGA
Ala-Pro-Arg-Pro-Pro-Thr-Pro-Glu-Gly Primer end GAG CTG G ag gtg aca gat cag ctc cca ggg caa agc gac…3’ Glu-Leu-G lu-Val-Thr-Asp-Gln-Leu-Pro-Gly-Gln-Ser-Asp-
Figure 2-8: Forward oligonucleotide primer sequence. The flap primer was designed so that its 3’ end clamps onto the rat apoE mature DNA sequence at its 5’ end. The clamp region is shown in capital letters. The primer also contained an Igase cleavage site for cleavage of the histidine tag located at the N-terminus of the protein. This sequence is shown in underlined bold lettering. An NcoI restriction site was located near the 5’ end of the primer and was used to ligate the resulting PCR product into the pET30 vector.
The NcoI site is shown in italics. The reverse primer was similar but contained a clamp region and a HindIII site for ligation into the pET30 vector.
75 AB
1 2 3 4 1 2 3 kDa kDa 97 66 50 45 37
26 30 19
15
20.1
14.4
Figure 2-9: SDS-PAGE analysis of recombinant rat apoE at different stages of purification and western blot analysis following purification. (A) Samples of apoE were electrophoresed on an 18% denaturing polyacrylamide gel and stained with
Coomassie blue. Lane 1: Amersham-Pharmacia low molecular weight marker. Lane 2:
Whole bacterial cell lysate expressing rat apoE. The cells were solubilized in SDS containing loading buffer and boiled for 3 minutes. Lane 3: Uncut His-tagged recombinant rat apoE after purification on His-bind column. Lane 4: Recombinant rat apoE with His-tag removed. (B) Western blot analysis of recombinant rat apoE (Lane 2) along with native rat apoE (Lane 3). The proteins were purified and 100 ng of each was loaded on an 18% denaturing polyacrylamide gel along with a protein ladder (Lane 1-
Invitrogen BenchMark Pre-stained protein ladder). The proteins were transferred to
PVDF membrane and probed with goat polyclonal native rat apoE serum as described under Material and Methods.
76
1 2 3 4 kDa 97 66
45
30
20
Figure 2-10: SDS-PAGE analysis of rat apoE under different storage conditions.
Samples of apoE were electrophoresed on an 18% denaturing polyacrylamide gel and stained with Coomassie blue following storage under different conditions. Lane 1:
Amersham-Pharmacia low molecular weight marker. Lane 2: Recombinant rat apoE stored in SDS loading buffering under denaturing conditions at -20˚C for 3 weeks. Lane
3: Recombinant rat apoE stored at 4˚C in STB for 3 weeks. Lane 4: Recombinant rat apoE stored at -20˚C in STB for 3 weeks.
77
30
Native rat apoE 20 Rec. rat apoE
) * 1000 10
-1
ol
0 dm
2
-10
(deg cm Mean Residual Ellipticity -20
200 220 240 260
Wavelength (nm)
Figure 2-11: Far UV circular dichroism spectra of apoE. The spectra of native rat apoE (solid line) and recombinant rat apoE (broken line) were recorded on a Jasco J-720 spectopolarimeter from 195-260 nm. All samples were analyzed at 0.1 mg/ml at room temperature.
78
1.0
0.8
325 o 0.6
/OD
5 32 0.4
OD Native rat apoE Rec. rat apoE 0.2 DMPC only
0.0 0 5 10 15 20 25 30
Time (min)
Figure 2-12. Dimyristoylphosphatidylcholine (DMPC) liposome solubilization by apoE. DMPC multilamellar liposomes in standard Tris buffer were maintained at 24.5˚C by water bath and monitored at 325 nm for 30 minutes after 21.3 µg protein (5:1 lipid:protein mass ratio) was added. The filled triangles represent DMPC alone. The filled circles designate native rat apoE and the open circles represent recombinant rat apoE. The Y-axis is the ratio of the absorbance at 325 nm at any given time point (OD) to the initial absorbance (ODo). The data are the mean ± SEM of at 3 experiments.
79
16
14
12 per 8 h)
(% 10
8
6 Native rat apoE terol Efflux 4 Rec. rat apoE
2
Choles
0 0 10203040
Concentration (µg/ml)
Figure 2-13: ABCA1-mediated cholesterol efflux from RAW264 macrophages.
Cells were grown to 70-80% confluence in 48-well plates and labeled for 24 h with [3H] cholesterol along with 0.3 mM 8-bromo-cAMP in order to up regulate ABCA1. Protein acceptors at different concentrations were added in DMEM media with 0.2% BSA and
0.3 mM cAMP. After 8 h, samples were counted in a scintillation counter and divided by total cell labeling to give % cholesterol efflux. Filled circles represent native rat apoE and open circles designate recombinant rat apoE.
80 Chapter 3:
THE STRUCTURE OF HUMAN APOLIPOPROTEIN A-IV: A DISTINCT
DOMAIN ARCHITECTURE AMONG EXCHANGEABLE APOLIPOPROTEINS
WITH POTENTIAL FUNCTIONAL IMPLICATIONS.
ABSTRACT
Apolipoprotein (apo) A-IV is an exchangeable apolipoprotein that shares many
functional similarities with related apolipoproteins such as apoE and apoA-I, but has also
been implicated as a circulating satiety factor. However, despite the fact that it contains
many predicted amphipathic α-helical domains, relatively little is known about its tertiary
structure. We hypothesized that apoA-IV exhibits a characteristic functional domain organization that has been proposed to define apoE and apoA-I. To test this, we created truncation mutants in a bacterial system that deleted amino acids from either the N- or C- terminal ends of human apoA-IV. We found that apoA-IV was less stable than apoA-I, but was more highly organized in terms of its cooperativity of unfolding. Deletion of the extreme N- and C-termini of apoA-IV did not significantly affect the cooperativity of unfolding but deletions past amino acid 333 on the C-terminus or 61 on the N-terminus had major destabilizing effects. Functionally, apoA-IV was less efficient than apoA-I at clearing multilamellar phospholipid liposomes and promoting ATP binding cassette transporter A1-mediated cholesterol efflux. However, deletion of a C-terminal glutamine-rich region (amino acids 333-376) of apoA-IV stimulated both of these activities dramatically. We conclude that the amphipathic α-helices in apoA-IV form a single, large domain that may be similar to the N-terminal helical bundle domains of
81 apoA-I and apoE, but that apoA-IV lacks the C-terminal lipid binding and cholesterol
efflux-promoting domain present in these apolipoproteins. In fact, the C-terminus of
apoA-IV contains a unique glutamine-rich domain that reduces the ability of apoA-IV to
interact with lipids and promote cholesterol efflux. We suggest that this sequence may have evolved to modulate the function of apoA-IV, possibly allowing it to perform distinct functions from those performed by other exchangeable apolipoproteins.
INTRODUCTION
Apolipoprotein (apo) A-IV is a 46 kDa protein discovered in 1974 in rat plasma high density lipoprotein (HDL) (1). In humans, it is synthesized by the small intestine and secreted into the mesenteric lymphatics associated with chylomicrons. Under conditions of high lipid absorption, apoA-IV secretion by the gut is up regulated and can account for up to 3% of total gut secreted protein (2;3). Upon chylomicron lipolysis in the general circulation, apoA-IV rapidly dissociates from remnant particles, primarily
residing in HDL and as lipid-poor protein (2;4). Despite this responsiveness to lipid
intake, an unambiguous function for apoA-IV has not yet been widely recognized. It has
been proposed to play many functions in vivo including: food intake regulation (5),
gastrointestinal motility (6), structural constituent of lipoproteins (7), protection against
lipid oxidation and atherosclerosis (8), and can also mimic many of the roles of apoA-I
including cholesterol efflux and activation of lecithin:cholesterol acyl transferase (9;10).
ApoA-IV shares several general features with other exchangeable
apolipoproteins, especially apoA-I and apoE. The primary sequences of all three proteins
are dominated by multiple 22-amino acid repeats which are predicted to form
82 amphipathic alpha helices (11). These repeats are likely responsible for the ability of
these proteins to associate with lipids. ApoA-IV contains at least 12 such repeats of
which most are punctuated by proline residues (11). However, the majority of its repeats
are significantly more hydrophilic than the helices within apoA-I and apoE, making it the most hydrophilic of the human exchangeable apolipoproteins. As a result, its helices have been proposed to penetrate less deeply into lipid than the helices of other apolipoproteins (12), possibly accounting for the significant portion of apoA-IV that exists in a lipid dissociated state in human plasma (2;4).
In addition to similarities in predicted secondary structure, apoA-IV is also related to the other apolipoproteins at the genetic level. The human apoA-IV gene is located in
the same 15-kilobase gene complex as apoA-I and apoC-III on the long arm of
chromosome 11 (13). In fact, intraexonic duplication of the apoA-I gene has been
suggested to have led to the formation of the apoA-IV gene some 300 million years ago
(14). Therefore, one might expect that apoA-IV would exhibit tertiary structure
similarities to the more extensively studied apoA-I and apoE. Both of these
apolipoproteins have been proposed to exhibit a general structural organization that is
composed of two functional domains. Composed of clusters of amphipathic helical
repeats, these domains play different functional roles within the protein. The best
information obtained thus far has come from an X-ray crystallography study of an N-
terminal fragment of apoE. This work shows that the N-terminal 22 kDa of apoE can
form a four-helix bundle structure that is relatively stable in the absence of lipid (15).
The remaining C-terminal domain exists in a less defined structure that is poised to
interact with lipids to trigger an unfolding of the N-terminal domain as the protein
83 associates with lipid. This unfolding event exposes a binding site for the low density
lipoprotein receptor in the N-terminal domain (for a review, see (16)). Without the
benefit of a monomeric, lipid-free crystal structure, it has also been suggested that apoA-I
exhibits a similar functional domain organization in solution. Deletion mutagenesis
studies have indicated that the N-terminal half of apoA-I is responsible for most of the
helical structure and stability in the lipid-free form and that the C-terminal half is less
organized. Upon lipid binding, the C-terminal domain undergoes a dramatic increase in
helicity and becomes the major stabilizing domain in the lipid-bound state (17;18). More recent work has further defined these domains and proposed a two-step model for the binding of apoA-I to lipid (19). The weight of the evidence indicates that the N-terminal domain of apoA-I (about residues 1-187) forms a helical bundle structure that is in a molten globular state (20;21). By contrast, the C-terminus (188-243) forms a less organized domain that is required for lipid binding events (17;19;22-25).
Given the similarities between apoA-IV and the relatively well-studied apoA-I and apoE molecules, we hypothesized that a two functional domain architecture may be a common theme among the exchangeable apolipoproteins and that a similar model would apply to apoA-IV. On this basis, we expected to find: a) a N-terminal domain composed of associated α-helices and a relatively disordered C-terminal region and, b) the lipid- binding functionally of apoA-IV to be localized to the C-terminus. In addition, our recent studies have indicated that the class Y character of the extreme C-terminal helix in apoA-
I is critical for its interaction with the ABCA1 transporter (25;26). We speculated that this helix might act as a tether to specific lipid domains on the cell surface, facilitating its interaction with ABCA1 (26). Given that apoA-IV contains a number of class Y helices
84 in its C-terminal half, we further hypothesized that one or more of these helices would be critical for apoA-IV’s ability to promote cellular cholesterol efflux via ABCA1.
To test these ideas, we took the approach of deleting discrete sequences from each end of apoA-IV and assessing the effects on the structural organization as well as the ability to associate with lipids and promote ABCA1-mediated cholesterol efflux. We found that apoA-IV indeed contains a large domain that resembles the N-terminal helical bundle structure of apoA-I. However, apoA-IV lacks the strong C-terminal lipid-binding domain. In fact, it contains a unique Gln-rich domain at the C-terminus that may inhibit the molecule from interacting with lipids and promoting cholesterol efflux.
85 EXPERIMENTAL PROCEDURES
Materials
SDS-PAGE gels were obtained from Bio-Rad (Hercules, CA) or Amersham-Pharmacia
(Piscataway, NJ). Primer synthesis and DNA sequencing were performed by the
University of Cincinnati DNA Core (Cincinnati, OH). Restriction enzymes were
purchased from New England Biolabs (Beverly, MA). Human small intestine QUICK-
Clone cDNA was purchased from BD Biosciences Clontech (product #7176-1) (Palo
Alto, CA). IgA protease (Igase) was purchased from Mobi-Tech (Marco Island, FL).
BL-21 (DE3) Escherichia coli and the pET30 vector were from Novagen (Madison, WI).
Isopropyl-β-D-thiogalactoside (IPTG) was from Fisher Scientific (Pittsburgh, PA).
Bis[sulfosuccinimidyl] suberate (BS3) was purchased from Pierce (Rockford, IL). 1,2-
dimyristoyl-sn-glycero-3-phosphatidylcholine (DMPC) was acquired form Avanti Polar
Lipids (Birmingham, AL). Fatty acid-free bovine serum albumin (BSA) was from
Calbiochem (San Diego, CA). 8-Bromoadenosine 3’:5’-cAMP was from Sigma-Aldrich
(St. Louis, MO). Fetal bovine serum, Dulbecco’s modified Eagle medium and
phosphate-buffered saline were from Invitrogen Life Technologies (Carlsbad, CA).
RAW 264.7 mouse macrophage cells were from American Type Culture Collection
(Manassas, VA). All chemicals reagents were of the highest quality available.
Methods
Cloning and mutagenesis of human apoA-IV.
A human small intestine cDNA library was used to clone human apoA-IV. PCR primers were designed to amplify the coding region of human apoA-IV. Following amplification, forward and reverse flap primers were designed containing an NcoI and
86 HindIII cleavage site respectively. The forward primer also contained the sequence
encoding for a proteolytic cleavage site specific for Igase that is required for removing
the histidine tag from the purified protein (27). The apoA-IV DNA was then ligated into
the pET30 vector (Novagen, Madison, WI) and completely sequenced.
Most of the deletion mutants used WT apoA-IV in the pET30 vector as the
template for PCR based mutagenesis. The ∆1-39 mutant and ∆1-61 mutant forward
primers were designed similarly to the WT, but instead of the clamp region nucleotides
encoding the N-terminal amino acids, they matched the 40th and 62nd amino acids
respectively. The C-terminal mutations were created by performing PCR-based site-
directed mutagenesis (QuickChange, Novagen, Madison, WI) in order to insert stop
codons at specific locations. The N- and C-terminal double mutations were created by inserting stop codons into the ∆1-39 mutant as their template.
Protein expression and purification.
To generate WT human apoA-IV, we modified our highly efficient E. coli expression system used previously for human apoA-I (27). The expression construct was transfected into E. Coli BL-21 cells and grown in Luria-Bertani (LB) culture media that included kanamycin for selection of the pET30 transformants. Protein expression was driven by the T7 promoter upon exposure to IPTG for 2h. After expression, the bacterial cells were pelleted and lysed by sonication, leaving the recombinant apoA-IV in the supernatant after centrifugation. The presence of the histidine tag allowed for easy purification by using a histidine-binding column (Ni2+-affinity column). The protease
Igase was used to cleave the leader portion (with the histidine tag) of the protein, thus
87 leaving the mature form of apoA-IV with an additional Thr-Pro on the N-terminus.
Finally, the sample was applied to a His-bind column a second time to separate the
mature protein from the cleaved tag. ApoA-IV exists in humans as two major isoforms
differing at amino acid position 360 (Gln, Type I; His, Type II) (28). The protein studied
in this work was the Type II isoform. However, we have generated the point mutants
necessary to convert Type II to Type I for both the WT protein and the ∆1-61 deletion
mutants (the C-terminal deletions do not contain this site) and compared them using all
the structural and functional assays described below. We found no significant differences
between the two isoforms (data not shown).
Circular dichroism (CD) spectroscopy.
The average secondary structures of the apoA-IV mutants were determined by
obtaining CD spectra at room temperature using a Jasco J-600 spectropolarimeter. The
α-helix content was calculated from the molar ellipticity at 222 nm, as described (29).
The thermal denaturation was monitored from the change in molar ellipticity at 222 nm
over the temperature range 20-80˚C, as described (30). The cooperativity index, n,
describing the sigmoidicity of the thermal denaturation curve and the van’t Hoff
enthalpy, ∆Hv, were calculated as described previously (19).
Fluorescence Measurements.
Fluorescence measurements were carried out with a Hitachi F-4500 fluorescence spectrophotometer at 25˚C. 8-anilino-1-naphthalenesulfonic acid (ANS) fluorescence
88 spectra were collected from 400 to 600 nm at an excitation wavelength of 395 nm in the
presence of 50 µg/ml protein and an excess of ANS (250 µM) (19).
Self-association determination.
Lipid-free WT apoA-IV and various mutants were dialyzed into 20 mM sodium
phosphate buffer, pH 7.5 containing 20 mM NaCl at a protein concentration of 1 mg/ml.
The proteins were then cross-linked by adding 50 mM BS3 stock in DMSO to a final
concentration of 10 mM BS3 for 1 h at room temperature. The reactions were quenched by adding 1 M Tris•HCl, pH 7.5 to a final concentration of 50 mM for 15 minutes at room temperature. After quenching, 4 µg of each protein sample were then run on 4-25% gradient polyacrylamide denaturing gels and stained with Coomassie blue (31).
Dimyristoylphosphatidylcholine (DMPC) liposome solubilization.
DMPC in chloroform was dried in a glass tube and brought up in 1 ml of degassed standard Tris buffer (10 mM Tris, 0.15 M NaCl, 1 mM EDTA, 0.2% NaN3) at a concentration of 5 mg/ml. Multilamellar liposomes were formed by sonication for 30 s with a model 550 sonic dismembrator at level 5 with a microtip (Fisher). The experiments were performed in an Amersham Biosciences Ultraspec 4000 at a constant temperature of 24.5˚C, as maintained by circulating water bath. Each protein sample was then added to the liposomes at a mass ratio of 10:1 (wt. DMPC:protein) in separate cuvettes and the absorbance was recorded over 30 minutes at 325 nm. The Y-axis of the graph is the ratio of the absorbance at any given time point (OD) to the initial absorbance at time zero (ODo) (32).
89
Cholesterol efflux studies.
The transformed mouse macrophage cell line RAW264.7 was maintained in
growth medium (Dulbecco’s modified Eagle medium, 10% fetal bovine serum, and 50
µg/ml gentamycin). The cells were grown to 70-80% confluency in 48-well plates and
the growth medium was removed. The cells were then labeled with 1.0 µCi/ml [3H] cholesterol in efflux medium (DMEM + 0.2% BSA) with or without 0.3 mM 8-bromo-
cAMP for 24 h. After 24 h, the cells were washed twice with PBS containing 0.2% BSA and once with efflux medium. Efflux medium containing differing concentrations of protein acceptors was added to the wells with or without cAMP for 8 h. After 8 h, 150 µl sample of efflux medium was removed and filtered through 0.45-µm filter to remove any floating cells. The amount of [3H] cholesterol in 100 µl of each sample was then measured by liquid scintillation counting. The percent cholesterol efflux was calculated by dividing the counts in the medium by total cell counts at time zero minus cholesterol efflux to media alone at 8 hours (26). For the experiments studying the dependence of
cholesterol efflux on acceptor protein concentration, estimated “Km” and “Vmax” values
were obtained by fitting the data to the following equation: Y=Vmax * (X/(X+Km)), where Y represents the % of total cellular cholesterol effluxed per hour and X represents the concentration of acceptor protein present in terms of µg/ml (33).
RESULTS
As discussed above, we hypothesized apoA-IV would have a modular domain
structure based on its sequence similarities to apoE and apoA-I. To test this, we created
90 deletion mutants by cutting from the N- and C-terminal ends of the apoA-IV molecule and then subjected the mutants to a battery of structural and functional assays designed to determine: a) the region(s) of highest structural organization and, b) region(s) involved in lipid binding and ABCA1-mediated cholesterol efflux. Figure 3-1 is a linear diagram of human apoA-IV showing its predicted 22-residue repeats and punctuating proline residues (34;35). The arrows indicate sites of deletion from both ends of the molecule.
For the C-terminal mutations, stop codons were introduced to delete the C-terminal Gln- rich domain (∆333-376) and then each predicted helical domain was deleted at the helical repeat junctions up to amino acid 249. Regions were deleted from the N-terminus to remove a predicted non-amphipathic helical domain (∆1-39) and then the first predicted
N-terminal helical segment (∆1-61).
After expression and purification, the proteins were analyzed by SDS-PAGE.
The protein preparations were routinely greater than 95% pure as illustrated by the subset of proteins in Figure 3-2. The majority of the proteins migrated to positions near their predicted molecular weight as determined by amino acid sequence. However, the ∆1-39 mutant appeared to be larger than expected (about 43,000 daltons as opposed to 39,166 daltons) on the gel. We confirmed that the mass of this mutant was indeed correct by
MALDI mass spectrometry (39,175 daltons). Similar inconsistencies in the apparent molecular weight of apoA-IV deletion mutants have been noted previously and may be attributed to decreased SDS binding to particular mutants causing them to migrate slower on the gel (36).
Structural Studies.
91 We next performed a series of structural studies on the proteins. The α-helical
content was determined by circular dichroism (Table 3-1). The α-helical content of WT
apoA-IV was 40 % or roughly 151 helical residues. Removal of the C-terminal amino
acids down to residue 271 resulted in little change in the overall helical content.
However, for apoA-IV molecules containing deletions past residue 271, we noted a
marked decrease in helicity. Interestingly, removing the N-terminal 39 amino acids
resulted in a marked decrease in helicity that persisted in all mutants that had this domain
deleted.
We next performed thermal denaturation experiments that monitored the helical
content of the proteins by circular dichroism as a function of temperature. Figure 3-3A
shows WT apoA-IV remained in a relatively native conformation until about 43˚C where
it underwent a rapid denaturation followed by a plateau. This is compared to a
denaturation profile for recombinant WT apoA-I. The Tm for apoA-IV is 51˚C compared
to 60˚C for apoA-I indicating that, on average, the alpha-helices in apoA-IV are
significantly less stable. The sigmoidicity of the apoA-I denaturation curve is
significantly less than for apoA-IV. A cooperativity index that describes the sigmoidal
nature of the denaturation is listed in Table 3-1 along with the midpoint temperature and
the van’t Hoff enthalpy of the denaturation. Removal of C-terminal residues 333-376
stabilized the protein slightly (Tm =54 vs. 51˚C), and had relatively minimal effects on the cooperativity index and van’t Hoff enthalpy. The properties of ∆333-376 were similar to the proposed N-terminal, single domain helix bundle structure formed by residues 1-190 of apoA-I. Removal of the 22-residue helix spanning residues 311-333 disrupted the organization of the apoA-IV molecule, as indicated by much larger
92 reductions in the cooperativity and van’t Hoff enthalpy. As more extensive deletions of
the C-terminus were made (Figure 3-3A and B), the cooperativity decreased to a
minimum of about 6 with similar reductions for the Tm and ∆Hv. At the N-terminus,
deleting amino acids 1-39 or 1-61 had little effect on cooperativity, Tm and van’t Hoff
enthalpy as compared to WT (Figure 3-3C, Table 3-1). Thus it appeared that removing
the extreme ends of the molecule (1-61 or 333-376) preserved the general structure
adopted by the WT protein, however, further deletions into the C-terminal end clearly
disrupted the protein organization. It should be noted that we attempted to make further
deletions into the N-terminus (∆1-83, ∆1-94, and ∆1-138). However, these deletions
resulted in mutant proteins that were quickly degraded during the purification from our
bacterial expression system (data not shown). We believe that these mutations caused
catastrophic perturbations of the tertiary structure exposing numerous sites that were
normally buried within the protein to proteolytic degradation.
Double mutants were also created with deletions at both the N- and C-termini.
The first mutant was designed in order to remove amino acids 1-39 and 333-376 which
are not thought to form class A or class Y amphipathic helices (Figure 3-1). A second mutant was then designed to include amino acids 40 to 270. Figure 3-4 and Table 3-1 show ∆1-39,333-376 exhibited similar, if not increased, stability parameters to WT, but the structure of ∆1-39,271-376 was sufficiently perturbed that it did not undergo cooperative unfolding. Thus, the highest cooperativity of unfolding was consistently noted for intact domain represented by residues 40-332.
As an additional measure of structural organization, we examined the interaction of ANS with the mutants. ANS binds to exposed hydrophobic surfaces resulting in
93 increased fluorescence emission (37). Therefore, the more exposed hydrophobic
domains, the higher the degree of ANS binding and fluorescence. Figure 3-5 shows the
effects of interaction with WT apoA-IV, WT apoA-I and the various deletion mutants on
ANS fluorescence. WT apoA-IV induced much less ANS fluorescence per amino acid as
compared to apoA-I. Similar to the thermal denaturation studies, removing amino acids
1-61 or amino acids 333-376 had relatively minor effects on ANS fluorescence as
compared to WT apoA-IV. ∆311-376 also exhibited low ANS fluorescence. However,
further deletions into the C-terminal end resulted in dramatic increases in ANS
fluorescence. Consistent with the results from the denaturation studies, the double
mutant, ∆1-39,333-376, exhibited similar ANS fluorescence to WT, but ∆1-39,271-376
was markedly increased. In general, the ANS fluorescence results suggest that the
domain spanning residues 40-332 can fold into a structure that exhibits minimal exposure
of hydrophobic residues.
Functional Studies.
We next determined the effects of the deletions on several indices of apoA-IV
function in vitro. Since self-association is a common feature of lipid-free
apolipoproteins, the ability of apoA-IV and its deletion mutants to oligomerize was
determined by crosslinking with BS3. Figure 3-6 is an SDS-PAGE analysis of protein samples crosslinked at 1 mg/ml. In the case of WT apoA-IV, approximately 70% of the protein was present as a dimer with the remaining 30% as monomer. Removal of the N- terminal amino acids 1-61 and removal of C-terminal amino acids up to 311-376 did not affect the oligomerization pattern as compared to WT. However, further deletions into
94 the C-terminus resulted in an even distribution between monomer, dimer and trimer.
Both double mutants followed their C-terminal mutant counterpart with ∆1-39,333-376 forming only monomer and dimer and ∆1-39,271-376 forming monomer, dimer and trimer (data not shown).
In order to determine the regions of apoA-IV that are important in lipid binding and reorganization, a DMPC lipid clearance assay was performed. In this assay, DMPC multi-lamellar liposomes are generated in buffer, resulting in a turbid solution. Protein samples are then added and the turbidity clears as the protein binds and reorganizes the liposomes into discoidal structures. Figure 3-7 shows that WT apoA-IV cleared the liposomes more slowly than WT apoA-I but clearly was able to reorganize the lipid to some extent. Surprisingly, removal of regions from the C-terminus dramatically increased the rate of clearance, contrary to the effect of C-terminal deletion on the ability of apoA-I to clear DMPC liposomes (26). In fact, the increase in lipid clearance was so profound that we were forced to modify our usual protocol from a 2.5:1 (wt. phospholipid to wt. protein) to a 10:1 ratio in order to slow the reaction to a point at which the kinetics could be measured accurately. The highly increased ability of the C-terminal truncation mutants persisted when the proteins were compared on an equal protein molar basis as well as on a mass basis (data not shown). On the other hand, removal of N-terminal amino acids 1-39 or 1-61 did not affect lipid clearance in the apoA-IV mutants.
Interestingly, the N- and C-terminal double mutants did not differ from WT, despite lacking the same C-terminal amino acids that resulted in the dramatic increase in the C- terminus only mutants.
95 Similar to apoA-I, apoA-IV has been shown to have the ability to promote the
efflux of cholesterol via the ABCA1-pathway (9). We took advantage of our deletion
mutants to determine if a particular domain in apoA-IV is responsible for interaction with
ABCA1. We used RAW264 macrophages in these studies because it has been clearly
established that the addition of cAMP to the cells causes an increase in ABCA1 at the cell
surface (38;39). Following 24 h treatment of the cells with labeled cholesterol and
cAMP, lipid-free protein acceptors were added for 8 hours to determine their ability to
promote cholesterol efflux by the ABCA1-mediated pathway. Figure 3-8 shows
ABCA1-mediated cholesterol efflux over a range of concentrations of WT apoA-IV, the
deletion mutants and WT apoA-I. The apparent Km and Vmax values for each of the lipid- free protein acceptors are also given in Table 3-2. These results show that all proteins tested exhibited generally similar maximum capacities (Vmax) for ABCA1-mediated
cholesterol efflux under these conditions. However, large differences in apparent Km
existed between the proteins; much more WT apoA-IV was required to reach its
maximum effect compared to apoA-I. However, removal of any of the C-terminal
regions resulted in a markedly lower Km than for WT apoA-IV and a value that was similar to that of WT apoA-I. In contrast, removal of the N-terminal amino acids 1-39 or
1-61 caused smaller reductions in Km. The double mutant, ∆1-39,271-376 exhibited a lower Km than WT and was similar to the rest of the C-terminal mutants. In general, the
presence of the region 333-376 dramatically diminished apoA-IV’s ability to promote
cholesterol efflux via the ABCA1 pathway.
96 DISCUSSION
Based on previous studies of apoE and apoA-I, we anticipated that apoA-IV
might exhibit a two-domain structure comprising an N-terminal domain folded into a
bundle of 22-residue amphipathic alpha-helices and a relatively disordered C-terminal domain responsible for lipid interactions. This was reasonable considering the evolutionary and predicted secondary structural similarities between these proteins and the fact that the most C-terminal amphipathic repeat in apoA-IV is a class Y helix with a similar mean residue hydrophobic moment (34) to the lipid binding/cholesterol-effluxing helical domain identified in apoA-I (26). While we found evidence that apoA-IV contains at least one large domain (residues 40-332) comprising about 77% of the amino acid sequence, the region(s) of the apoA-IV molecule primarily involved in DMPC clearance and in ABCA1-mediated cholesterol efflux were not located in the C-terminal
1/3 of the protein. Indeed, removal of the polar Gln-rich C-terminus domain (residues
333-376) of apoA-IV resulted in increased DMPC clearance and cholesterol efflux. The effects of the deletions performed in this study on apoA-IV structure and function are summarized graphically in Figure 3-9. We discuss these observations in detail below.
ApoA-IV structure. Our data indicate that apoA-IV exists as a single cooperative domain
between residues 40-332. Inspection of Figure 3-1 shows that this region encompasses
about 9 of the potential 22-mer helical repeats that are predicted for the protein.
However, our CD data in Table 3-1 for the mutant ∆1-39,333-376 indicates that the
domain is only about 1/3 helical, suggesting that the domain contains an average of about
91 helical amino acids – or roughly four 22-mer helices or perhaps a higher number of
97 shorter helices. Based on the structures of other lipid-free apolipoproteins, it follows that the hydrophobic faces of these helical domains are likely clustered together in the interior of the protein. Such a four helical bundle structure has been established for the N- terminal domain of apoE (15). In this case, the four-helix bundle has well defined secondary and tertiary structure and is stabilized by strong hydrophobic forces and intrahelical salt-bridge interactions, explaining its relatively high free energy of stabilization of about 10 kcal/mol (15). By contrast, the helical bundle in apoA-I is significantly less stable with a free energy of stabilization of about 2-3 kcal/mol (40).
This domain likely differs from a classical four helix bundle in that it is more dynamic with secondary structural elements that do not maintain stable tertiary interactions. This melted tertiary structure is characteristic of molten globule proteins (20). The conformational dynamics of lipid-free apoA-I are evident when it is visualized on a native polyacrylamide gel; it runs as a diffuse or smeared band that is much larger than predicted by its molecular weight (data not shown). ApoA-IV is even less stable than apoA-I as shown in Table 3-1 and by the fact that it is completely denatured in about 0.5
M guanidine HCl with a calculated free energy of stabilization of less than 1 kcal/mol
(41). Despite this low stability, our data indicate that the structure of the domain is highly cooperative and organized in such a way as to efficiently sequester hydrophobic residues within the interior of the protein molecule. This high level of organization is consistent with the fact that monomeric lipid-free apoA-IV runs as a tightly focused band on a native polyacrylamide gel (data not shown). The precise reason for the dichotomy between low stability and high structural organization cannot be directly addressed by our studies. However, we propose that, compared to apoA-I and apoE, the stabilizing forces
98 within apoA-IV may be dominated less by hydrophobic forces acting upon sequestered nonpolar residues and more by hydrophilic interactions among exposed polar residues.
Indeed, the helical repeats within apoA-IV are, on average, less hydrophobic and have a smaller hydrophobic helical wedge angle than the helices in the other apolipoproteins
(12). Such polar interactions on the surface may result in a delicate, but defined, tertiary structure for apoA-IV.
We should point out that our results differ from previous studies of internal deletion mutants of apoA-IV. Studies by Weinberg et al. (32;42) and Emmanuel et al.(36) demonstrated that all internal deletions of various regions of apoA-IV, including areas that are outside of the 40-332 domain proposed here, resulted in major reductions in the thermal stability of apoA-IV. However, the mutants utilized in these studies contained a hydrophobic decapeptide tag on the N-terminus that was demonstrated to affect both structural and functional aspects of the protein (42). It may be that the presence of the tag region made the mutants much more sensitive to destabilizations from deletions than seen in the current study in which the purification tag had been removed.
In any case, the general result from both studies is consistent with the notion that lipid- free apoA-IV is composed of a large domain that is relatively organized, but unstable.
Functional Studies. An important function of the exchangeable apolipoproteins is to bind to lipid to form lipoproteins. We compared WT apoA-IV’s ability to clear DMPC liposomes to that of apoA-I. This assay measures the end result of a complex, multi-step interaction between proteins and lipids (see the reaction scheme in Segall et al.(43)).
Although the assay does not speak directly to lipid binding affinity per se, it does offer a
99 general index of the overall ability of a protein to solubilize a planar lipid surface into lipoprotein-like particles. In agreement with studies by Weinberg et al. (32), apoA-IV was less efficient than apoA-I at clearing lipid in the DMPC assay. However, when we removed the polar Gln-rich region (amino acids 333-376), apoA-IV’s ability to clear
DMPC increased strikingly. Further deletions into the C-terminal end, including our most extreme mutation ∆249-376, did not alter this increased clearance. This is in direct contrast to the case of apoA-I where deletion of the final helix resulted in dramatically decreased DMPC clearance efficiency (17;19). Removing amino acids 1-39 from WT apoA-IV had no effect on lipid clearance (Figure 3-7C) but dramatically decreased the ability of mutants lacking residues 271-336 or 333-376 to solubilize DMPC liposomes
(Figure 3-7B and D). These results suggest that residues 1-39 within apoA-IV can mediate DMPC clearance efficiently but, in the presence of the C-terminal Gln-rich sequence, this site is masked (Figure 3-9). Removal of C-terminal domain may either unmask this site or allow a conformational change that brings it to bear on the target lipid surface. Whether this N-terminal region is the actual site of lipid binding or if perturbations within this domain affect a more distally located functional domain is unclear and will require further study.
Both apoA-IV and apoA-I are acceptors in ABCA1-mediated cholesterol efflux in vitro. Figure 3-8 and Table 3-2 show that apoA-IV and apoA-I have similar Vmax values
for ABCAI-mediated cholesterol efflux but apoA-IV is less efficient, exhibiting a higher
Km value. However, similar to DMPC binding, the removal of the C-terminal Gln-rich
region (333-376) increased apoA-IV’s ability to promote cholesterol efflux by reducing
the Km to a value comparable to that of apoA-I. Further C-terminal deletions had similar
100 effects, presumably because they caused disruption of the helix bundle exposing 22-
residue amphipathic alpha-helices (see the increased ANS fluorescence – Figure 3-5). In
terms of the identity of a recognition site within apoA-IV, it has been suggested that a
particular array of acidic residues across two amphipathic helices in the C-terminus of
apoA-I is required to promote ABCA1-mediated cholesterol efflux (44). Our analysis
shows that the helical domains between amino acids 150 and 193 in apoA-IV appear to
fit this criterion whereas those closer to the C-terminus do not. Future studies will be
aimed at determining if this domain is responsible for mediating the interaction of apoA-
IV in the absence of the C-terminus. Intriguingly, the double mutant ∆1-39, 271-376 that
cleared lipid slowly (Figure 3-7D), was still efficient at promoting cholesterol efflux
(Figure 3-8D). This suggests that different sites within the apoA-IV molecule may be required for the two functions.
We would emphasize two implications of this work concerning the functionality of apoA-IV. First, we have demonstrated that there are sequences within apoA-IV that are clearly capable of mediating apoA-I- like functions such as lipid binding and cholesterol efflux. However, the Gln-rich C-terminal domain can attenuate these functions. Since there is no homolog of this sequence in any of the other exchangeable apolipoproteins,
Weinberg et al. have proposed that this region may be an evolutionary innovation that imparted a role for apoA-IV in chylomicron assembly and secretion around the time of the divergence of mammals and avians (45). Based on our data, we would propose that this domain may also have evolved to allow apoA-IV to perform additional functions that are distinct from the other exchangeable apolipoproteins by suppressing its apoA-I-like properties. Indeed, the protein has been implicated as a circulating satiety factor along
101 with its roles in lipid metabolism (46). This may also account for observations that apoA-
IV may exist in the lipid-free form to a much greater extent in plasma vs. apoA-I or apoE
(4). Second, it is clear that apoA-IV does not possess the C-terminal lipid-binding functional domain that is present in apoA-I and apoE. This is the case even in the absence the inhibitory Gln-rich region. Although apoA-IV may have evolved from gene duplication events of ancestral apolipoproteins and shares the basic amphipathic helical building blocks, the overall localization of functional domains within the sequence is quite different from apoA-I and apoE. We suggest that this is further evidence that apoA-IV evolved from the exchangeable apolipoproteins to perform a distinct function that may not be directly related to lipoprotein formation and/or integrity. Clearly more work is warranted to fully understand the functional consequences of this distinct structural organization.
102 REFERENCES
1. Swaney,J.B., Reese,H., and Eder,H.A. (1974) Polypeptide composition of rat high density lipoprotein: characterization by SDS-gel electrophoresis, Biochem. Biophys. Res. Commun. 59, 513-519. 2. Hayashi,H., Nutting,D.F., Fujimoto,K., Cardelli,J.A., Black,D., and Tso,P. (1990) Transport of lipid and apolipoproteins A-I and A-IV in intestinal lymph of the rat, J. Lipid Res. 31, 1613-1625. 3. Tso,P., Liu,M., Kalogeris,T.J., and Thomson,A.B. (2001) The role of apolipoprotein A-IV in the regulation of food intake, Annu. Rev. Nutr. 21, 231-254. 4. Green,P.H., Glickman,R.M., Riley,J.W., and Quinet,E. (1980) Human apolipoprotein A-IV. Intestinal origin and distribution in plasma, J. Clin. Invest 65, 911-919. 5. Fujimoto,K., Fukagawa,K., Sakata,T., and Tso,P. (1993) Suppression of food intake by apolipoprotein A-IV is mediated through the central nervous system in rats, J. Clin. Invest 91, 1830-1833. 6. Okumura,T., Fukagawa,K., Tso,P., Taylor,I.L., and Pappas,T.N. (1996) Apolipoprotein A-IV acts in the brain to inhibit gastric emptying in the rat, Am. J. Physiol 270, G49-G53. 7. Lefevre,M. and Roheim,P.S. (1984) Metabolism of apolipoprotein A-IV. [Review] [46 refs], J. Lipid Res. 25, 1603-1610. 8. Ostos,M.A., Conconi,M., Vergnes,L., Baroukh,N., Ribalta,J., Girona,J., Caillaud,J.M., Ochoa,A., and Zakin,M.M. (2001) Antioxidative and antiatherosclerotic effects of human apolipoprotein A- IV in apolipoprotein E-deficient mice, Arterioscler. Thromb. Vasc. Biol. 21, 1023-1028. 9. Remaley,A.T., Stonik,J.A., Demosky,S.J., Neufeld,E.B., Bocharov,A.V., Vishnyakova,T.G., Eggerman,T.L., Patterson,A.P., Duverger,N.J., Santamarina-Fojo,S., and Brewer,H.B., Jr. (2001) Apolipoprotein specificity for lipid efflux by the human ABCAI transporter, Biochem. Biophys. Res. Commun. 280, 818-823. 10. Steinmetz,A. and Utermann,G. (1985) Activation of lecithin: cholesterol acyltransferase by human apolipoprotein A-IV, J. Biol. Chem. 260, 2258-2264. 11. Boguski,M.S., Elshourbagy,N., Taylor,J.M., and Gordon,J.I. (1985) Comparative analysis of repeated sequences in rat apolipoproteins A-I, A-IV, and E, Proc. Natl. Acad. Sci. USA 82, 992-996. 12. Weinberg,R.B. (1987) Differences in the hydrophobic properties of discrete alpha-helical domains of rat and human apolipoprotein A-IV, Biochim. Biophys. Acta 918, 299-303. 13. Karathanasis,S.K. (1985) Apolipoprotein multigene family: tandem organization of human apolipoprotein AI, CIII, and AIV genes, Proc. Natl. Acad. Sci. U. S. A 82, 6374-6378. 14. Karathanasis,S.K., Oettgen,P., Haddad,I.A., and Antonarakis,S.E. (1986) Structure, evolution, and polymorphisms of the human apolipoprotein A4 gene (APOA4), Proc. Natl. Acad. Sci. U. S. A 83, 8457-8461. 15. Wilson,C., Wardell,M.R., Weisgraber,K.H., Mahley,R.W., and Agard,D.A. (1991) Three-dimensional structure of the LDL receptor-binding domain of human apolipoprotein E, Science 252, 1817-1822. 16. Weisgraber,K.H. (1994) Apolipoprotein E: structure-function relationships. [Review] [201 refs], Adv. Protein Chem. 45, 249-302. 17. Davidson,W.S., Hazlett,T., Mantulin,W.W., and Jonas,A. (1996) The role of apolipoprotein AI domains in lipid binding, Proc. Natl. Acad. Sci. USA 93, 13605-13610. 18. Oda,M.N., Forte,T.M., Ryan,R.O., and Voss,J.C. (2003) The C-terminal domain of apolipoprotein A-I contains a lipid-sensitive conformational trigger, Nat. Struct. Biol 10, 455-460. 19. Saito,H., Dhanasekaran,P., Nguyen,D., Holvoet,P., Lund-Katz,S., and Phillips,M.C. (2003) Domain structure and lipid interaction in human apolipoproteins A-I and E, a general model, J. Biol. Chem. 278, 23227-23232.
103 20. Gursky,O. and Atkinson,D. (1996) Thermal unfolding of human high-density apolipoprotein A-1: implications for a lipid-free molten globular state, Proc. Natl. Acad. Sci. USA 93, 2991-2995. 21. Rogers,D.P., Roberts,L.M., Lebowitz,J., Engler,J.A., and Brouillette,C.G. (1998) Structural analysis of apolipoprotein A-I: effects of amino- and carboxy-terminal deletions on the lipid-free structure, Biochemistry 37, 945-955. 22. Segrest,J.P., Jones,M.K., De Loof,H., Brouillette,C.G., Venkatachalapathi,Y.V., and Anantharamaiah,G.M. (1992) The amphipathic helix in the exchangeable apolipoproteins: a review of secondary structure and function. [Review], J. Lipid Res. 33, 141-166. 23. Schmidt,H.H., Remaley,A.T., Stonik,J.A., Ronan,R., Wellmann,A., Thomas,F., Zech,L.A., Brewer,H.B., Jr., and Hoeg,J.M. (1995) Carboxyl-terminal domain truncation alters apolipoprotein A-I in vivo catabolism, J. Biol. Chem. 270, 5469-5475. 24. Ji,Y. and Jonas,A. (1995) Properties of an N-terminal proteolytic fragment of apolipoprotein AI in solution and in reconstituted high density lipoproteins, J. Biol. Chem. 270, 11290-11297. 25. Gillotte,K.L., Zaiou,M., Lund-Katz,S., Anantharamaiah,G.M., Holvoet,P., Dhoest,A., Palgunachari,M.N., Segrest,J.P., Weisgraber,K.H., Rothblat,G.H., and Phillips,M.C. (1999) Apolipoprotein-mediated plasma membrane microsolubilization. Role of lipid affinity and membrane penetration in the efflux of cellular cholesterol and phospholipid, J. Biol. Chem. 274, 2021-2028. 26. Panagotopulos,S.E., Witting,S.R., Horace,E.M., Hui,D.Y., Maiorano,J.N., and Davidson,W.S. (2002) The role of apolipoprotein A-I helix 10 in apolipoprotein-mediated cholesterol efflux via the ATP- binding cassette transporter ABCA1, J. Biol. Chem. 277, 39477-39484. 27. Panagotopulos,S.E., Witting,S.R., Horace,E.M., Nicholas,M.J., and Sean,D.W. (2002) Bacterial expression and characterization of mature apolipoprotein A-I, Protein Expr. Purif. 25, 353-361. 28. Kamboh,M.I. and Ferrell,R.E. (1987) Genetic studies of human apolipoproteins. I. Polymorphism of apolipoprotein A-IV, Am. J. Hum. Genet. 41, 119-127. 29. Sparks,D.L., Phillips,M.C., and Lund-Katz,S. (1992) The conformation of apolipoprotein A-I in discoidal and spherical recombinant high density lipoprotein particles. 13C NMR studies of lysine ionization behavior, J. Biol. Chem. 267, 25830-25838. 30. Acharya,P., Segall,M.L., Zaiou,M., Morrow,J., Weisgraber,K.H., Phillips,M.C., Lund-Katz,S., and Snow,J. (2002) Comparison of the stabilities and unfolding pathways of human apolipoprotein E isoforms by differential scanning calorimetry and circular dichroism, Biochim. Biophys. Acta 1584, 9- 19. 31. McGuire,K.A., Davidson,W.S., and Jonas,A. (1996) High yield overexpression and characterization of human recombinant proapolipoprotein A-I, J. Lipid Res. 37, 1519-1528. 32. Weinberg,R.B., Anderson,R.A., Cook,V.R., Emmanuel,F., Denefle,P., Hermann,M., and Steinmetz,A. (2000) Structure and interfacial properties of chicken apolipoprotein A-IV, J. Lipid Res. 41, 1410- 1418. 33. Davidson,W.S., Gillotte,K.L., Lund-Katz,S., Johnson,W.J., Rothblat,G.H., and Phillips,M.C. (1995) The effect of high density lipoprotein phospholipid acyl chain composition on the efflux of cellular free cholesterol, J. Biol. Chem. 270, 5882-5890. 34. Segrest,J.P., Garber,D.W., Brouillette,C.G., Harvey,S.C., and Anantharamaiah,G.M. (1994) The amphipathic alpha helix: a multifunctional structural motif in plasma apolipoproteins. [Review], Adv. Protein Chem. 45, 303-369. 35. Li,W.H., Tanimura,M., Luo,C.C., Datta,S., and Chan,L. (1988) The apolipoprotein multigene family: biosynthesis, structure, structure-function relationships, and evolution, J. Lipid Res. 29, 245-271. 36. Emmanuel,F., Steinmetz,A., Rosseneu,M., Brasseur,R., Gosselet,N., Attenot,F., Cuine,S., Seguret,S., Latta,M., Fruchart,J.C., and et al (1994) Identification of specific amphipathic alpha-helical sequence of human apolipoprotein A-IV involved in lecithin:cholesterol acyltransferase activation, J. Biol. Chem. 269, 29883-29890.
104 37. Stryer,L. (1965) The interaction of a naphthalene dye with apomyoglobin and apohemoglobin. A fluorescent probe of non-polar binding sites, J. Mol. Biol. 13, 482-495. 38. Oram,J.F., Lawn,R.M., Garvin,M.R., and Wade,D.P. (2000) ABCA1 is the cAMP-inducible apolipoprotein receptor that mediates cholesterol secretion from macrophages, J. Biol. Chem. 275, 34508-34511. 39. Smith,J.D., Miyata,M., Ginsberg,M., Grigaux,C., Shmookler,E., and Plump,A.S. (1996) Cyclic AMP induces apolipoprotein E binding activity and promotes cholesterol efflux from a macrophage cell line to apolipoprotein acceptors, J. Biol. Chem. 271, 30647-30655. 40. Sparks,D.L., Davidson,W.S., Lund-Katz,S., and Phillips,M.C. (1993) Effect of cholesterol on the charge and structure of apolipoprotein A-I in recombinant high density lipoprotein particles, J. Biol. Chem. 268, 23250-23257. 41. Weinberg,R.B. and Spector,M.S. (1985) Structural properties and lipid binding of human apolipoprotein A-IV, J. Biol. Chem. 260, 4914-4921. 42. Weinberg,R.B., Anderson,R.A., Cook,V.R., Emmanuel,F., Denefle,P., Tall,A.R., and Steinmetz,A. (2002) Interfacial exclusion pressure determines the ability of apolipoprotein A-IV truncation mutants to activate cholesterol ester transfer protein, J. Biol. Chem. 277, 21549-21553. 43. Segall,M.L., Dhanasekaran,P., Baldwin,F., Anantharamaiah,G.M., Weisgraber,K.H., Phillips,M.C., and Lund-Katz,S. (2002) Influence of apoE domain structure and polymorphism on the kinetics of phospholipid vesicle solubilization, J. Lipid Res. 43, 1688-1700. 44. Natarajan,P., Forte,T.M., Chu,B., Phillips,M.C., Oram,J.F., and Bielicki,J.K. (2004) Identification of an apolipoproteinA-I structural element that mediates cellular cholesterol efflux and stabilizes ABCA1, J. Biol. Chem. 45. Jin,L., Shieh,J.J., Grabbe,E., Adimoolam,S., Durbin,D., and Jonas,A. (1999) Surface plasmon resonance biosensor studies of human wild-type and mutant lecithin cholesterol acyltransferase interactions with lipoproteins, Biochemistry 38, 15659-15665. 46. Rodriguez,M.D., Kalogeris,T.J., Wang,X.L., Wolf,R., and Tso,P. (1997) Rapid synthesis and secretion of intestinal apolipoprotein A-IV after gastric fat loading in rats, Am. J. Physiol 272, R1170-R1177. 47. Chen,Y.H., Yang,J.T., and Martinez,H.M. (1972) Determination of the secondary structures of proteins by circular dichroism and optical rotatory dispersion, Biochemistry 11, 4120-4131.
105 TABLES:
Table 3-1: Conformation and stability properties of WT apoA-IV and the various deletion mutants.
c d e α- Cooperativity Tm ∆Hv Self association Heli Index b (°C) (kcal/mol) x (%) a WT apoA-IV 40 13.4 51 58 1/2 ∆ 333-376 39 12.5 54 52 1/2 ∆ 311-376 39 8.8 51 40 1/2 ∆ 289-376 44 5.9 46 33 1/2/3/4.. ∆ 271-376 41 6.4 47 34 1/2/3/4.. ∆ 249-376 28 5.6 44 33 1/2/3/4.. ∆ 1-39 31 14.5 54 60 1/2 ∆ 1-61 35 13.2 52 56 1/2 ∆ 1-39, 333-376 31 16.3 52 68 1/2 ∆ 1-39, 271-376 20 5.5 56 28 1/2/3/4.. WT apoA-I 46 7.8 60 33 ND ∆ 190-243 57 12.0 56 48 ND (apoA-I)
a Approximate α-helical contents determined from the molar ellipticity at 222 nm at 25°C as calculated according to Chen et al. (47). Each value represents the average of at least 2 observations. Typical standard deviations for these experiments are ± 5%. b Describes the sigmoidicity of the thermal denaturation curve (see Methods). Each value represents the average of at least 2 observations. c The midpoint temperature of denaturation. Typical reproducibility for these experiments is ± 2.0 °C (19). d The estimated error on these experiments is ± 0.5 kcal/mol. e Qualitative estimation of the oligomerization behavior of the mutants as assessed by BS3 cross-linking studies. 1/2 indicates the presence of monomers and dimers (typically about 70% dimer, 30% monomer). 1/2/3/4.. indicates the presence of a relatively even distribution of multiple oligomers.
106 Table 3-2: Kinetic parameters for cholesterol efflux from RAW264 macrophages to lipid-free apoA-IV and its deletion mutants.
a b Km Vmax WT apoA-IV 54 3.0 ∆ 333-376 11 3.2 ∆ 311-376 8 3.0 ∆ 289-376 5 2.4 ∆ 271-376 7 3.0 ∆ 249-376 8 3.2 ∆ 1-39 26 2.4 ∆ 1-61 23 2.5 ∆ 1-39, 333-376 NAc NA ∆ 1-39, 271-376 13 3.4 WT apoA-I 7 3.1
a Concentration of acceptor apolipoprotein at 50% of the maximal rate of cholesterol efflux (Vmax). The units are (µg apolipoprotein / ml media). Typical error is approximately 14% of the mean value. b Maximal rate of cholesterol efflux (% total cellular cholesterol label per h ) calculated as described in Methods). Typical error is approximately 36% of the mean value. c This mutant did not exhibit saturable cholesterol efflux kinetics and could not be subjected to analysis.
107 FIGURES:
Figure 3-1: Linear diagrams of WT apoA-IV and truncation mutants. The shaded boxes represent predicted class Y amphipathic α-helices and the hatched boxes represent class A helices. Most repeats are punctuated by a proline residue (P) (22). The arrows are placed at sites where deletions were made either in the N- or C-terminus. Each deletion mutant used in this study is represented by a solid black line showing the sequence that was left intact.
108
Figure 3-2: SDS-PAGE analysis of WT apoA-IV and representative mutant forms.
After expression and purification as described in Experimental Procedures, 4 µg samples of apolipoprotein were electrophoresed on an 18% denaturing polyacrylamide gel and stained with Coomassie blue. Lane 1: Amersham-Pharmacia low molecular weight marker. Lane 2: WT apoA-IV. Lane 3: ∆333-376. Lane 4: ∆311-376. Lane 5: ∆1-61.
Lane 6: ∆1-39.
109
AB 0 0 0
-2 -6 -2 ) A-IV -1 -4 -12 -4 A-I -6 -6 -18 20 40 60 80
dmole -8 2 -8 -10 -10 -12 -12 WT apo A-IV Molar Ellipticity -14 WT apo A-IV -14 ∆289-376 (mdeg cm ∆333-376 ∆271-376 -16 ∆311-376 -16 ∆249-376
-18 -18 20 30 40 50 60 70 80 20 30 40 50 60 70 80
C 0 -2 -4 -6 -8 -10 -12 WT apo A-IV -14 ∆1-39 ∆1-61 -16 -18 20 30 40 50 60 70 80
Temperature (°C)
Figure 3-3: Thermal unfolding of N- and C-terminal mutants of apoA-IV. Thermal unfolding was monitored by ellipticity at 222 nm. Panel A: WT apoA-IV (circles), ∆333-
376 (triangles) and ∆311-376 (squares). The inset shows WT apoA-IV compared to human apoA-I. Panel B: WT apoA-IV (circles), ∆289-376 (triangles), ∆271-376
(squares) and ∆249-376 (hexagons). Panel C: WT apoA-IV (black), ∆1-39 (triangles) and
∆1-61 (squares).
110
0
) -2 -1 e
l -4 ity
c -6 pti dmo
2 -8 -10 cm
lar Elli -12 WT apo A-IV Mo -14 ∆1-39, 271-376 (mdeg -16 ∆1-39, 333-376 -18 20 30 40 50 60 70 80 Temperature (°C)
Figure 3-4: Thermal unfolding of combined N- and C-terminal mutants. WT apoA-
IV (circles), ∆1-39,271-376 (triangles) and ∆1-39,333-376 (squares).
111
Figure 3-5: Effects of apoA-I, WT apoA-IV and mutant forms of apoA-IV on ANS fluorescence. 50 µg/ml protein was added to an excess of ANS (250 µM) and the fluorescence spectra were collected from 400-600 nm at an excitation wavelength of 395 nm. The data are reported as detector counts per amino acid measured at the wavelength of maximum fluorescence (484 nm). The values for human apoA-I are shown for comparison.
112
Figure 3-6: SDS-PAGE analysis of the self-association of WT apoA-IV and representative mutant forms. Lipid-free samples were cross-linked in 20 mM PBS with 10 mM BS3 for 60 minutes at room temperature. After quenching, 4 µg protein
samples were run on 4-25% gradient polyacrylamide denaturing gel and stained with
Coomassie blue. Lane 1: Amersham-Pharmacia low molecular weight marker. Lane 2:
WT apoA-IV. Lane 3: ∆1-39. Lane 4: ∆1-61. Lane 5: WT apoA-IV. Lane 6: ∆333-376.
Lane 7: ∆311-376. Lane 8: ∆289-376. Lane 9: ∆271-376.
113
Figure 3-7: Dimyristoylphosphatidylcholine (DMPC) liposome solubilization by apoA-I, apoA-IV and deletion mutants of apoA-IV. DMPC multilamellar liposomes in standard Tris buffer were maintained at 24.5˚C by water bath and monitored at 325 nm
for 30 minutes after protein was added (see Methods). Panel A: WT apoA-IV versus
apoA-I. Panel B: WT apoA-IV versus all of the C-terminal deletion mutants. Panel C:
WT apoA-IV versus the N-terminal deletion mutants. Panel D: WT apoA-IV versus the
combined N- and C-terminal deletion mutants. The Y-axis is the ratio of the absorbance
at 325 nm at any given time point (OD) to the initial absorbance (ODo) (32).
114
Figure 3-8: ABCA1-mediated cholesterol efflux from RAW264 macrophages. Cells were grown to 70-80% confluency in 48-well plates and labeled for 24 h with [3H] cholesterol along with 0.3 mM 8-bromo-cAMP in order to upregulate ABCA1. Protein acceptors at different concentrations were added in DMEM media with 0.2% BSA and
0.3 mM cAMP. After 8 h, samples were counted in a scintillation counter. Panel A compares WT apoA-IV and apoA-I. Panel B compares WT apoA-IV, ∆333-376, ∆311-
376, ∆289−376, ∆271-376 and ∆249-376. Panel C compares WT apoA-IV, ∆1-39 and
∆1-61. Panel D compares WT apoA-IV and the double mutant ∆1-39,271-376. Note: the concentration ranges differ between panels A and C vs. B and D in order to more effectively show the differing apparent Km values among the mutants. The data were fit
to a simple Michaelis-Menten equation (see Methods) and are shown by the solid lines.
115
Figure 3-9: A generalized model for the effects of deletions on the structure and function of lipid-free apoA-IV. The hatched square represents the correctly folded, large structural domain (a bundle of 22-mer amphipathic alpha-helices punctuated by prolines) that is postulated to exist in lipid-free apoA-IV. The irregular shape with the wavy lines represents a major disruption in the stability and cooperativity of this domain. The small white square on the N-terminus represents a putative lipid binding domain that is masked by the C-terminal Gln-rich domain in the WT protein (see text). The relative rates of lipid clearance refer to DMPC clearance rates relative to the WT apoA-IV (defined as “slow”) vs. the ∆333-376 deletion mutant (defined as
“fast”), see Figure 3-7.
116 Chapter 4:
A POTENTIAL INTERACTION BETWEEN THE N- AND C-TERMINI
OF APOLIPOPROTEIN A-IV AND ITS ROLE IN LIPID ASSOCIATION
ABSTRACT
Apolipoprotein (apo) A-IV is a 376 residue exchangeable apolipoprotein that may play a number of important roles in lipid metabolism, including chylomicron assembly, reverse cholesterol transport, and appetite regulation. In vivo, apoA-IV exists in both lipid-poor and lipid-associated forms and the balance between these states may determine its function. We examined the structural elements that modulate apoA-IV lipid binding by producing a series of recombinant human apoA-IV deletion mutants and determining their ability to interact with phospholipid liposomes. We found that the deletion of residues 333 to 343 strongly increased the lipid association rate vs. native apoA-IV.
Additional mutagenesis revealed that the sequence from residues 333-337 mediated the lipid-binding inhibitory effect, and that an intact N-terminus was required for the enhanced lipid affinity induced by deletion of the C-terminal sequence. We propose a structural model in which an interaction between specific N- and C-terminal sequences acts as a conformational switch that modulates the rate at which apoA-IV associates with lipid.
INTRODUCTION
Human apolipoprotein (apo) A-IV is a 46kD glycoprotein that is the largest member of the exchangeable apolipoprotein family, which includes apoA-I and apoE. It is synthesized by enterocytes of the small intestine in response to lipid absorption and is
117 secreted into circulation on the surface of chylomicrons. As chylomicrons undergo
lipolysis in the plasma compartment, apoA-IV rapidly dissociates from their surface and
thereafter circulates as a lipid-free protein and in association with HDL (1). It has been
proposed that apoA-IV evolved to play a role in chylomicron assembly or catabolism (2).
However, several other additional functions have been proposed, including inhibition of
lipid oxidation (3) and inflammatory processes (4), participation in reverse cholesterol
transport (5), and regulation of food intake (6). Because plasma apoA-IV can exist as
both a component of plasma lipoproteins and also as a lipid-free protein, it is possible that
alternate conformations perform distinct functions. However, the structural features that
mediate the conversion between these states are largely unknown.
ApoA-IV shares many features with the other exchangeable apolipoproteins,
especially apoA-I and apoE. Indeed, intra-exonic duplication of a primordial apoA-I
gene may have led to the origin of the apoA-IV gene some 300 million years ago (7). A
major feature of the primary sequences of the exchangeable apolipoproteins is a variable
number of 22-residue amphipathic α-helical repeats, which are thought to confer the
ability to bind to the surface of lipoprotein particles (8). ApoA-IV (376 residues) contains 13 such repeats, most punctuated by proline residues, in a cluster between residues 40 and 332. The first 39 amino acids of apoA-IV are encoded by a separate exon and contain a weakly amphipathic helical domain that is similar to those found in globular proteins (9). By contrast, the C-terminus (from 333-376) is predicted to be devoid of ordered secondary structure. Residues 354-367 comprise a glutamine-rich sequence that is not found in any other apolipoprotein (10).
118 Lipid-free forms of apoA-I and apoE have been shown to exhibit a
compartmentalized architecture characterized by a well organized N-terminal domain and
a relatively unstable C-terminal domain (11-13), which contains lipid binding sequences.
Given the similarities among these apolipoproteins, we expected that apoA-IV would
exhibit a similar organization. However, we found that the α-helices in apoA-IV are
organized around a single, large domain (14). Furthermore, the C-terminal third of apoA-
IV not only lacked a lipid-binding domain, it appeared to actually inhibit lipid
interactions. Removing the C-terminal 44 amino acids of apoA-IV (∆333-376) resulted
in a mutant that reorganized liposomes significantly faster than both WT apoA-IV and
apoA-I (14). The goal of the present study was to identify the sequence in the C-terminal
region responsible for this lipid binding inhibitory effect.
EXPERIMENTAL PROCEDURES
Materials
SDS-PAGE gels were obtained from Bio-Rad (Hercules, CA) or Amersham-Pharmacia
(Piscataway, NJ). Primer synthesis and DNA sequencing were performed by the
University of Cincinnati DNA Core (Cincinnati, OH). Restriction enzymes were purchased from New England Biolabs (Beverly, MA). IgA protease (Igase) was purchased from MoBiTec (Germany). BL-21 (DE3) Escherichia coli and the pET30 vector were from Novagen (Madison, WI). Isopropyl-β-D-thiogalactoside (IPTG) was
from Fisher Scientific (Pittsburgh, PA). HisBind Resin was from Novagen (Madison,
WI). Centriprep centrifugal concentrators were from Millipore/Amicon Bioseparations
(Bedford, MA). 1,2-dimyristoyl-sn-glycero-3-phosphatidylcholine (DMPC) was
119 acquired form Avanti Polar Lipids (Birmingham, AL). Fatty acid-free bovine serum
albumin (BSA) was from Calbiochem (San Diego, CA). 8-Bromoadenosine 3’:5’-cAMP
was from Sigma-Aldrich (St. Louis, MO). Fetal bovine serum, Dulbecco’s modified
Eagle medium and phosphate-buffered saline were from Invitrogen Life Technologies
(Carlsbad, CA). RAW 264.7 mouse macrophage cells were from American Type Culture
Collection (Manassas, VA). All chemicals reagents were of the highest quality available.
Methods
Mutagenesis of human apoA-IV.
Human apoA-IV DNA was ligated into the pET30 expression vector using the
NcoI and HindIII sites. The C-terminal mutations were created by performing PCR-
based site-directed mutagenesis (Quick-Change, Novagen) (14). For the ∆333-376,
∆344-376 and ∆352-376 mutants, a stop codon was inserted in place of the first amino
acid to be deleted. For internal deletion mutants, complementary forward and reverse
primers were designed with clamp regions on the 5’ and 3’ ends of the sequence
removed. The ∆333-343 mutant DNA contained in the pET30 vector was used as the
template for the double mutants (∆1-10, 333-343, ∆1-20, 333-343, ∆1-30, 333-343 and
∆1-39, 333-343). These mutants used a forward primer that had a 3’ clamp region in
which the nucleotides encoded the amino acids at position 11, 21, 31 and 40,
respectively. The primer also consisted of a 5’ flap region that encoded for the NcoI site
and an Igase (IgA protease) cleavage site which is used to remove the N-terminal histidine tag (as described below). For further details, please refer to (14).
Protein expression and purification.
120 Our protocol was similar to previous work with human apoA-I and more recently
with rat apoE (15). The expression vector was transformed into E. coli BL-21 competent
cells and plated overnight at 37˚C. Kanamycin (30 µg/ml) was used as the selective
agent. Cell colonies were picked and grown in Luria-Bertani (LB) culture media
overnight in 10 ml culture tubes in a shaking incubator at 37˚C. The cells in culture
media were then transferred to fresh 100 ml culture and grown to an A600 of 0.6-0.7. At this point, IPTG was added for 2 h to induce overexpression of the protein. Cell pellets were collected by centrifugation and brought up in His-bind buffer along with protease inhibitors. Cells were disrupted by sonication, centrifuged, and the supernatant was filtered through a Millex 0.45 µm filter and added to His-bind columns according to manufacturer’s instructions. The protein was eluted, concentrated and cut using Igase, a protease that recognizes the sequence Ala-Pro-Arg-Pro-Pro-Thr-Pro and cleaves before the Thr. This removes the His-tag leaving a Thr-Pro at the N-terminus. Next, the sample was passed over the His-bind column a second time to remove the cleaved tag. The His- bind column buffers in the purification of apoA-IV contained 3 M guanidine. The samples were then concentrated and dialyzed into standard Tris buffer (STB) (10 mM
Tris-HCl, 1 mM EDTA, 150 mM NaCl and 0.2 % NaN3) for storage.
DMPC liposome solubilization.
DMPC in chloroform was dried in a borosilicate tube with a stream of nitrogen, solubilized in degassed STB, and bath sonicated for 30 s to form multilamellar liposomes. Liposomes were placed in cuvettes at 24.5˚C, proteins were added to the liposomes in separate cuvettes at a final DMPC/protein mass ratio of 2.5:1, and the
121 absorbance was monitored at 325 nm in an Amersham Biosciences Ultraspec 4000 at
24.5˚C. Rate constants, k, were calculated for each mutant by fitting the first 5 min of data to a single exponential decay curve as a function of time (t): f(t) = a-kt using Sigma
Plot 2000 (16;17).
Interfacial Behavior at the Oil/water Interface.
The binding of apoA-IV at the oil/water interface was examined using a Tracker®
automatic tensiometer (IT Concept, France) (28). Drops of pure triolein were injected
into a sample chamber containing 25 µg/ml protein in PBS buffer, and protein adsorption
to the surface of the drop was monitored as the time-dependent decrease in interfacial
tension.
Thermal denaturation studies.
The thermal denaturation of the lipid-free apoA-IV mutants was monitored as the
change in molar ellipticity at 222 nm over a temperature range of 20-80˚C (18). The
van’t Hoff enthalpy, ∆Hv, was calculated as described previously (13).
Fluorescence spectroscopy.
Protein samples were studied at 0.1 mg/ml in STB on a Photon Technology
International Quantamaster spectrometer in photon counting mode at room temperature.
Excitation wavelength was 295 nm, excitation and emission band-passes were 3.0 nm,
and emission wavelength was monitored from 302-375 nm. Buffer blanks were
subtracted from each of the samples. Quenching studies were performed by addition of
122 0-0.2 M acrylamide; Stern-Volmer quenching constants, Ksv were determined as described (19).
RESULTS
To study the lipid association properties of human apoA-IV, we developed an efficient bacterial expression system that enables systematic deletion mutagenesis.
Figure 4-1 summarizes the N- and C-terminal deletion mutants that were generated for this study. The mutants were greater than 95% pure and had a size consistent with the engineered deletions (Figure 4-2A).
To determine if the deletions altered the oligomerization state of the mutants, we examined the mutant proteins by non-denaturing gel PAGE (Figure 4-2B). As reported previously, WT apoA-IV forms a mixture of monomer and dimer in solution with an approximate distribution of 70 % dimer to 30 % monomer (20). None of the deletions significantly affected the quaternary structure of apoA-IV. Interestingly, some mutants migrated more slowly than predicted. For example, the ∆333-376, ∆333-343, ∆333-337, and ∆1-39 mutants exhibited larger hydrodynamic diameters than WT apoA-IV, despite their shorter sequence length, suggesting that these deletions had induced a conformational change that altered the compactness of protein folding.
Consistent with other studies (2;21), WT apoA-IV was only marginally able to reorganize the DMPC multilamellar liposomes into micellar particles that scatter less light (Figure 4-3). However, as noted in our previous study (14), apoA-IV ∆333-376 cleared DMPC vesicles at a dramatically faster rate than WT apoA-IV, even surpassing
123 the lipid binding ability of apoA-I, which is well documented to bind lipid with higher
affinity than apoA-IV (2;22;23).
As the DMPC assay measures the end result of a complex process that involves
both an initial apolipoprotein-lipid interaction and a subsequent lipid reorganization (see
reaction scheme in (24)), we also examined the lipid affinity of WT and ∆333-376 apoA-
IV using oil drop tensiometry (25). This technique measures the rate at which
apolipoproteins in solution bind to and lower the surface tension of the more hydrophobic
triolein/water interface. As such, this technique provides a biophysical model for the
binding of apoA-IV to the nascent triglyceride-rich chylomicron surface. As seen in Fig.
4-3B, apoA-IV ∆333-376 caused a much more rapid decrease in interfacial tension compared to WT protein, in accord with the data from the DMPC clearance assay. In fact, for each of the mutants studied, oil drop tensiometry yielded qualitatively similar results as the DMPC clearance assay (data not shown).
We hypothesized that a specific sequence in the C-terminal 44 a.a. of apoA-IV mediated the lipid binding inhibitory effect. Structure prediction algorithms suggest that the C-terminus of apoA-IV contains predominantly random coil structure (8). We thus designed a series of candidate truncation mutants to localize the active inhibitory sequence. Deletion of residues ∆352-376 had no effect on the rate of lipid binding compared to WT (data not shown). Likewise, internal deletion of the unique EQQQ domain from ∆354-367, which generates a "pig-like" apoA-IV (26), also had no effect on lipid binding (Figure 4-4A). We next examined apoA-IV ∆344-376, because serine-343 is the terminal amino acid in chicken apoA-IV (2). Once again, this mutant was similar to WT in its lipid binding behavior (Fig. 4-4A).
124 These data suggested that the lipid binding inhibitory sequence resided between residues 333-343. To confirm this, we examined the behavior of an internal deletion mutant apoA-IV ∆333-343. As shown in Figure 4-4B, apoA-IV ∆343-343 disrupted
DMPC liposomes as fast as apoA-IV ∆333-376. A consideration of the sequence conservation of residues 333-343 across 7 species revealed that residues 333-337 are very highly conserved, whereas, residues 338-342 are much less so (Fig. 4-5A). We therefore generated two shorter internal deletion mutants, apoA-IV ∆333-337, which lacked the highly conserved sequence, and apoA-IV ∆338-342, which lacked the more variable interval. Figure 4-5B shows that apoA-IV ∆338-342 exhibited an intermediate binding rate. However, apoA-IV ∆333-337 was extremely efficient. In fact, this mutant was one of the fastest lipid-binding apolipoproteins we have observed. This data suggests that the most potent lipid binding inhibitory sequence is located between residues 333-337, but may overlap into the N-terminal regions of 338-342.
We previously noted the intriguing finding that the increases in lipid binding rate displayed by ∆333-376 and ∆271-376 deletion mutants were effaced by simultaneous deletion of the first 39 residues from the N-terminus (14), which suggested an interaction between a C-terminal inhibitory sequence and the N-terminus. To determine whether this phenomenon maintained with the ∆333-343 mutant, we prepared a series of C- and N- terminal double deletion mutants. Deletion of the N-terminal 39 residues on the background of the apoA-IV ∆333-343 mutant (∆1-39,333-343) converted the rapid binding ∆333-343 mutant to a slow lipid-binding protein, with kinetics similar to WT apoA-IV (Fig. 4-6). Similarly, the ∆1-20,333-343 deletion mutant also displayed slow binding. However, when only the first ten residues were deleted from the N-terminus,
125 ∆1-10,333-343, the mutant maintained the rapid binding profile of the ∆333-343 single mutant. We then generated an internal deletion mutant, apoA-IV ∆11-20,333-343, that contained an intact N-terminus. Figure 4-6B shows that this mutant was also a slow lipid binder. These data support the existence of an interaction between localized sequences in the amino and carboxy termini of apoA-IV that modulates its lipid affinity.
The DMPC clearance rate constants for all apoA-IV mutants, calculated as described in Experimental Procedures, are listed in Table 1. The clearance rate constants for apoA-IV ∆333-376, ∆333-343, ∆333-337 and ∆1-10,333-343 were statistically
different compared to WT apoA-IV. There was no difference between the clearance rate
for WT apoA-IV and the ∆354-367, ∆344-376, ∆338-342, ∆1-39,333-343 and ∆1-
20,333-343 mutants. Although there was no statistical difference between the clearance
rate for WT apoA-IV and apoA-IV ∆338-342 over the first 5 minutes, it is evident from
Figure 4-5 that this deletion mutant was better able to reorganize lipid at longer
incubation times.
To examine the impact of the various deletions on protein structure, we
determined their α-helical content and thermodynamic stability using circular dichroism
spectroscopy, and also examined the fluorescence properties of Trp-12, which has proved
to be a useful measure of apoA-IV conformation (27). Thermal denaturation studies
determined that the ∆344-376, ∆333-376, and internal ∆333-343 deletions did not
significantly alter global protein stability, as measured by the van’t Hoff enthalpy of
denaturation (Table 2). However, mutants that lacked the inhibitory sequence exhibited
enthalpies that were slightly lower than WT apoA-IV. The wavelength of maximum
fluorescence (λmax) of Trp-12 in WT apoA-IV was 335 nm, indicative of a relatively
126 hydrophobic environment (28). ApoA-IV ∆344-376 had a similar λmax., however, deletion of ∆333-376 or the inhibitory sequence in ∆333-343 caused a significant red shift in Trp emission, indicating that the amino terminus had relocated to a more polar environment. Acrylamide fluorescence quenching studies further indicated that the change in the polarity of the N-terminus was accompanied by a parallel increase in aqueous accessibility to the neutral quenching agent. These observations, together with the DMPC binding data, suggest that an interaction with the C-terminal inhibitory sequence may maintain the amino terminal globular domain in a more compact conformation that shields Trp-12 from the aqueous milieu.
DISCUSSION
It is well established that apoA-IV binds to lipid with much lower affinity than all the other members of the exchangeable apolipoprotein family (21;23). It has been postulated that its distinctively weak lipid binding behavior is due to the relative hydrophilicity and low amphipathic moment of its constitutive α-helices (29), the fact that most of these helices are of the Y-type, which may not be capable of deeply penetrating lipid surfaces (30), and/or the possibility that these helices are organized such that their hydrophobic regions are inaccessible for interaction with hydrophobic interfaces (2;23;25).
These hypotheses for the weak lipid affinity of apoA-IV notwithstanding, our data establish that human apoA-IV contains a short sequence in a region of random coil structure near the C-terminus that dramatically constrains its ability to associate with lipid. Removal of as few as five amino acids between residues 333-337 transforms
127 apoA-IV into one of the fastest lipid binding apolipoproteins that we have measured. Our
data further suggest that this lipid binding inhibition is mediated by an interaction between this short sequence and a domain in the N-terminus of apoA-IV in a manner that affects its global conformation. Clearly, apoA-IV has the innate capability to bind lipid with high efficiency, but this potential appears to be attenuated by conformational factors
in its lipid-free state. The mechanisms by which this occurs, and their possible
physiological consequences, are discussed below.
Based upon a series of fluorescence studies, Weinberg proposed that apoA-IV
exists as a loosely organized confluence/bundle of amphipathic alpha helices in which the
hydrophobic faces are oriented inwards towards the center of the bundle (2;23;25). Using
a mutagenesis approach, we subsequently proposed that the conformation of lipid-free
apoA-IV is dominated by a large helical bundle-like domain that encompasses residues
40 to 332 (14). The structural organization of such helical bundles has been extensively
documented for apoA-I, apoE, and lipophorin III (LpIII) (11;13;31;32), and it has been
proposed that before apolipoproteins can bind to lipid, they must first undergo a
conformational transformation in which the helical bundle domain "opens", thereby
reorienting the amphipathic faces of the constituent helices towards the hydrophobic lipid
interface (32;33). In apoE, intramolecular interactions between residues in distant
domains stabilize the lipid-free "closed" conformation (11). In apoA-IV, we have
observed that deletion of the N-terminus increases both its stability and the cooperativity
of interactions between its α-helices. This region is predicted to exist as a globular
domain, containing a G-type helix which is not known for mediating lipid binding (30).
128 Thus, the N-terminal domain may be able to destabilize the helical packing or global conformation of apoA-IV, perhaps contributing to its marginal stability in solution (23).
As a unifying interpretation of our data we propose that an interaction between the N-terminus (possibly between residues 11-20) and residues 333-337 in the C- terminus tethers or otherwise confines the globular domain, thereby attenuating its destabilizing impact. This would maintain the helical bundle in a more tightly packed,
"closed" conformation which, in turn, would result in slow binding to lipid interfaces.
Deletion of residues 333-337 would release the globular domain, and thus destabilize or
"loosen" the packing of the multi-helical bundle. This is supported by the observations that deletion of the inhibitory sequence increased the aqueous exposure of the N-terminal
Trp residue, as documented by fluorescence spectroscopy, and also increased the protein hydrodynamic radius on non-denaturing PAGE. Such a destabilized apoA-IV would require less energy to unfold into a lipid binding configuration, and would thus display rapid binding to lipid interfaces. This is consistent with the observations that deletion of the inhibitory sequence: 1) reduced the van’t Hoff enthalpy (albeit by only 7-15% compared to the 40-45% reduction seen with deletions that disrupt the helical bundle domain (14)); 2) dramatically increased the rate of DMPC vesicle clearance. However, if the N-terminal globular domain were also disrupted by mutagenesis, it would no longer function as a destabilizing factor, and apoA-IV would revert to its native slow binding behavior.
What could be the biological function of this inhibition of lipid binding in native apoA-IV? One possibility is that the varied metabolic functions of apoA-IV may require differential but adaptable lipid affinity. For example, it has been postulated that apoA-IV
129 plays a role in chylomicron assembly (1;34). However, once lipolysis of chylomicrons
begins, apoA-IV is rapidly shed from their surface and thereafter circulates predominantly as a lipid-free protein. Perhaps the N/C-terminal interaction evolved to
"lock" the protein in a closed conformation with relatively low lipid affinity, to maintain it in a free form required for other metabolic functions once its role in chylomicron assembly was completed. In this regard, although apoA-IV is present at a lower concentration than apoA-I in plasma, a much higher fraction exists in a lipid-free form and consequently its partitioning into the interstitial space may be much greater than apoA-I (35). Since lipid-free apolipoproteins are required for interaction with the
ABCA1 transporter, it is possible that this initial low lipid affinity of apoA-IV may allow
it to play a role in apolipoprotein-mediated cholesterol efflux in the periphery. It is also
possible that the lipid-free form of apoA-IV is required for its antioxidant and anorexic
roles as well (3;36). Finally, it is possible that localized intracellular or extracellular
environmental factor, such as pH or ionic concentration, could induce a conformational
change in apoA-IV that could modulate its lipid affinity.
The powerful impact of the C-terminal residues on the structure and function of
apoA-IV is evidenced by the fact that the most common polymorphisms of apoA-IV are
two SNP's that encode a T347S (apoA-IV-S) and a Q360H (apoA-IV-2) substitution in
the apoA-IV molecule (37). These substitutions have significant biophysical
consequences: the H-isoprotein displays increased lipid affinity (38), whereas the S-
isoprotein binds lipid with lower affinity (25) compared to the WT protein. Most likely
as a direct result of these changes in lipid binding, individuals carrying an H-allele
display more rapid post-prandial triglyceride-rich lipoprotein clearance than Q-allele
130 homozygotes, whereas individuals carrying an S-allele display delayed clearance (39).
Thus, it appears that even the most subtle modifications of the C-terminus in apoA-IV can have important physiological consequences.
In conclusion, we have established the exact location of a C-terminal sequence in human apoA-IV that dramatically constrains its ability to bind to lipid interfaces. Given the importance of lipid affinity in the biological function of apoA-IV, this finding provides a guide to the design of informative apoA-IV mutants, which when introduced into cell and transgenic mouse models will be a powerful tool to delineate the role of apoA-IV lipid binding on its function in chylomicron assembly and in lipoprotein metabolism in general.
131 REFERENCES
1. Green,P.H., Glickman,R.M., Riley,J.W., and Quinet,E. 1980. Human apolipoprotein A-IV. Intestinal origin and distribution in plasma. J.Clin.Invest 65:911-919.
2. Weinberg,R.B., Anderson,R.A., Cook,V.R., Emmanuel,F., Denefle,P., Hermann,M., and Steinmetz,A. 2000. Structure and interfacial properties of chicken apolipoprotein A-IV. J.Lipid Res. 41:1410- 1418.
3. Qin,X., Swertfeger,D.K., Zheng,S., Hui,D.Y., and Tso,P. 1998. Apolipoprotein AIV: a potent endogenous inhibitor of lipid oxidation. American Journal of Physiology 274:H1836-H1840.
4. Vowinkel,T., Mori,M., Krieglstein,C.F., Russell,J., Saijo,F., Bharwani,S., Turnage,R.H., Davidson,W.S., Tso,P., Granger,D.N. et al. 2004. Apolipoprotein A-IV inhibits experimental colitis. J.Clin.Invest 114:260-269.
5. Remaley,A.T., Stonik,J.A., Demosky,S.J., Neufeld,E.B., Bocharov,A.V., Vishnyakova,T.G., Eggerman,T.L., Patterson,A.P., Duverger,N.J., Santamarina-Fojo,S. et al. 2001. Apolipoprotein specificity for lipid efflux by the human ABCAI transporter. Biochem.Biophys.Res.Commun. 280:818-823.
6. Fujimoto,K., Cardelli,J.A., and Tso,P. 1992. Increased apolipoprotein A-IV in rat mesenteric lymph after lipid meal acts as a physiological signal for satiation. Am.J.Physiol 262:G1002-G1006.
7. Karathanasis,S.K., Oettgen,P., Haddad,I.A., and Antonarakis,S.E. 1986. Structure, evolution, and polymorphisms of the human apolipoprotein A4 gene (APOA4). Proc.Natl.Acad.Sci.U.S.A 83:8457- 8461.
8. Boguski,M.S., Elshourbagy,N., Taylor,J.M., and Gordon,J.I. 1985. Comparative analysis of repeated sequences in rat apolipoproteins A-I, A-IV, and E. Proc.Natl.Acad.Sci.USA 82:992-996.
9. Elshourbagy,N.A., Walker,D.W., Paik,Y.K., Boguski,M.S., Freeman,M., Gordon,J.I., and Taylor,J.M. 1987. Structure and expression of the human apolipoprotein A-IV gene. J.Biol.Chem. 262:7973- 7981.
10. Weinberg,R.B. 1994. Identification of functional domains in the plasma apolipoproteins by analysis of inter-species sequence variability. J.Lipid Res. 35:2212-2222.
11. Wilson,C., Wardell,M.R., Weisgraber,K.H., Mahley,R.W., and Agard,D.A. 1991. Three-dimensional structure of the LDL receptor-binding domain of human apolipoprotein E. Science 252:1817-1822.
12. Davidson,W.S., Hazlett,T., Mantulin,W.W., and Jonas,A. 1996. The role of apolipoprotein AI domains in lipid binding. Proc.Natl.Acad.Sci.USA 93:13605-13610.
13. Saito,H., Dhanasekaran,P., Nguyen,D., Holvoet,P., Lund-Katz,S., and Phillips,M.C. 2003. Domain structure and lipid interaction in human apolipoproteins A-I and E, a general model. J.Biol.Chem. 278:23227-23232.
14. Pearson,K., Saito,H., Woods,S.C., Lund-Katz,S., Tso,P., Phillips,M.C., and Davidson,W.S. 2004. Structure of human apolipoprotein A-IV: a distinct domain architecture among exchangeable apolipoproteins with potential functional implications. Biochemistry 43:10719-10729.
15. Pearson,K., Liu,M., Shen,L., Tso,P., and Davidson,W.S. 2005. Bacterial expression and characterization of rat apolipoprotein E. Protein Expr.Purif. 41:447-453.
132 16. McLean,L.R. and Hagaman,K.A. 1993. Kinetics of the interaction of amphipathic alpha-helical peptides with phosphatidylcholines. Biochim.Biophys.Acta. 1167:289-295.
17. Panagotopulos,S.E., Witting,S.R., Horace,E.M., Hui,D.Y., Maiorano,J.N., and Davidson,W.S. 2002. The Role of Apolipoprotein A-I Helix 10 in Apolipoprotein-mediated Cholesterol Efflux via the ATP-binding Cassette Transporter ABCA1. J.Biol.Chem. 277:39477-39484.
18. Acharya,P., Segall,M.L., Zaiou,M., Morrow,J., Weisgraber,K.H., Phillips,M.C., Lund-Katz,S., and Snow,J. 2002. Comparison of the stabilities and unfolding pathways of human apolipoprotein E isoforms by differential scanning calorimetry and circular dichroism. Biochim.Biophys.Acta 1584:9- 19.
19. Lehrer,S.S. 1971. Solute perturbation of protein fluorescence. The quenching of the tryptophyl fluorescence of model compounds and of lysozyme by iodide ion. Biochemistry 10:3254-3263.
20. Weinberg,R.B. and Spector,M.S. 1985. The self-association of human apolipoprotein A-IV. Evidence for an in vivo circulating dimeric form. J.Biol.Chem. 260:14279-14286.
21. Weinberg,R.B., Ibdah,J.A., and Phillips,M.C. 1992. Adsorption of apolipoprotein A-IV to phospholipid monolayers spread at the air/water interface. A model for its labile binding to high density lipoproteins. J.Biol.Chem. 267:8977-8983.
22. Weinberg,R.B. and Spector,M.S. 1985. Human apolipoprotein A-IV: displacement from the surface of triglyceride-rich particles by HDL2-associated C-apoproteins. J.Lipid Res. 26:26-37.
23. Weinberg,R.B. and Spector,M.S. 1985. Structural properties and lipid binding of human apolipoprotein A-IV. J.Biol.Chem. 260:4914-4921.
24. Segall,M.L., Dhanasekaran,P., Baldwin,F., Anantharamaiah,G.M., Weisgraber,K.H., Phillips,M.C., and Lund-Katz,S. 2002. Influence of apoE domain structure and polymorphism on the kinetics of phospholipid vesicle solubilization. J Lipid Res. 43:1688-1700.
25. Weinberg,R.B., Cook,V.R., Beckstead,J.A., Martin,D.D., Gallagher,J.W., Shelness,G.S., and Ryan,R.O. 2003. Structure and interfacial properties of human apolipoprotein A-V. J.Biol.Chem. 278:34438-34444.
26. Navarro,M.A., Acin,S., Iturralde,M., Calleja,L., Carnicer,R., Guzman-Garcia,M.A., Gonzalez- Ramon,N., Mata,P., Isabel,B., Lopez-Bote,C.J. et al. 2004. Cloning, characterization and comparative analysis of pig plasma apolipoprotein A-IV. Gene 325:157-164.
27. Weinberg,R.B. 1988. Exposure and electronic interaction of tyrosine and tryptophan residues in human apolipoprotein A-IV. Biochemistry 27:1515-1521.
28. Davidson,W.S., Arnvig-McGuire,K., Kennedy,A., Kosman,J., Hazlett,T.L., and Jonas,A. 1999. Structural organization of the N-terminal domain of apolipoprotein A-I: studies of tryptophan mutants. Biochemistry 38:14387-14395.
29. Weinberg,R.B. 1987. Differences in the hydrophobic properties of discrete alpha-helical domains of rat and human apolipoprotein A-IV. Biochim.Biophys.Acta 918:299-303.
30. Segrest,J.P., Garber,D.W., Brouillette,C.G., Harvey,S.C., and Anantharamaiah,G.M. 1994. The amphipathic alpha helix: a multifunctional structural motif in plasma apolipoproteins. [Review]. Advances in Protein Chemistry 45:303-369.
133 31. Silva,R.A., Hilliard,G.M., Fang,J., Macha,S., and Davidson,W.S. 2005. A Three-Dimensional Molecular Model of Lipid-Free Apolipoprotein A-I Determined by Cross-Linking/Mass Spectrometry and Sequence Threading. Biochemistry 44:2759-2769.
32. Wang,J., Sykes,B.D., and Ryan,R.O. 2002. Structural basis for the conformational adaptability of apolipophorin III, a helix-bundle exchangeable apolipoprotein. Proc.Natl.Acad.Sci.U.S.A 99:1188- 1193.
33. Weisgraber,K.H. 1994. Apolipoprotein E: structure-function relationships. [Review] [201 refs]. Advances.in Protein Chemistry. 45:249-302.
34. Kalogeris,T.J., Rodriguez,M.D., and Tso,P. 1997. Control of synthesis and secretion of intestinal apolipoprotein A-IV by lipid. [Review] [87 refs]. Journal.of.Nutrition. 127:537S-543S.
35. Roheim,P.S., Edelstein,D., and Pinter,G.G. 1976. Apolipoproteins in rat serum and renal lymph. Proc.Natl.Acad.Sci.U.S.A 73:1757-1760.
36. Ostos,M.A., Conconi,M., Vergnes,L., Baroukh,N., Ribalta,J., Girona,J., Caillaud,J.M., Ochoa,A., and Zakin,M.M. 2001. Antioxidative and antiatherosclerotic effects of human apolipoprotein A- IV in apolipoprotein E-deficient mice. Arterioscler.Thromb.Vasc.Biol. 21:1023-1028. 37. Weinberg,R.B. 2002. Apolipoprotein A-IV polymorphisms and diet-gene interactions. Curr.Opin.Lipidol. 13:125-134.
38. Weinberg,R.B., Jordan,M.K., and Steinmetz,A. 1990. Distinctive structure and function of human apolipoprotein variant ApoA-IV-2. J.Biol.Chem. 265:18372-18378.
39. Hockey,K.J., Anderson,R.A., Cook,V.R., Hantgan,R.R., and Weinberg,R.B. 2001. Effect of the apolipoprotein A-IV Q360H polymorphism on postprandial plasma triglyceride clearance. J.Lipid Res. 42:211-217.
134 TABLES:
Table 4-1. Rate constants for WT and mutant apoA-IV in the DMPC clearance assay.
Mutant protein ka WT apoA-IV 0.039 ± 0.012 ∆354-367 0.024 ± 0.005 ∆344-376 0.042 ± 0.017 ∆333-376 0.279 ± 0.080* ∆333-343 0.171 ± 0.029* ∆333-337 0.217 ± 0.039* ∆338-342 0.094 ± 0.034 ∆1-39,333-343 0.040 ± 0.006 ∆1-20,333-343 0.051 ± 0.006 ∆1-10,333-343 0.260 ± 0.065* a k is the rate constant derived from the first 5 min of the DMPC clearance assay; means ± SD * Denotes difference of P<0.05 between WT and mutant as determined by One-way ANOVA followed by a Tukey-Kramer Multiple Comparisons Test.
135 Table 4-2. Fluorescence and thermal denaturation parameters of WT and mutant apoA-IV
a b -1 c Mutant protein λmax (nm) Ksv (M ) ∆Hv (kcal/mol) WT apoA-IV 334.7 ± 2.1 2.84 ± 0.20 47.8 ∆344-376 335.5 ± 0.7 2.81 ± 0.23 46.9 ∆333-376 342.6 ± 1.9* 8.78 ± 1.09* 44.5 ∆333-343 338.5 ± 1.0* 5.53 ± 0.02* 39.7 a The λmax is the wavelength of maximum fluorescence at 25˚C (n = 3, ±1 S.D.) b The Ksv is the Stern-Volmer quenching constant, indicating relative exposure of the amino terminal tryptophan residue to acrylamide. c The estimated error on these experiments is ± 0.5 kcal/mol (n=2). * P<0.05 between WT and mutant as determined by One-way ANOVA followed by a Tukey- Kramer Multiple Comparisons Test.
136 FIGURES:
1 39 333 376 EVSADQVATVMWDYFSQLSNNAKEAVEHLQKSELTQQLN SFFSTFKEKESQDKTLSLPELEQQQEQQQEQQQEQVQMLAPLES
∆333-376 ∆344-376 ∆354-367 ∆333-343 ∆333-337 ∆338-342 ∆1-10,333-343 ∆1-20,333-343 ∆1-39,333-343 ∆11-20,333-343
Figure 4-1. Schematic map of human apoA-IV deletion mutants used in this study.
The single letter amino acid code is shown for residues 1-39 and 333-376. A wavy line indicates intervening residues. The thick gray bars identify the location of amino acids that were removed from the C-terminus or N-terminus.
137 kDa 1234567 8910 97 66 45 A 30
20.1
14.4
kDa 12 3 4 5 678 910 11 12 669 440 B 232 140
66
Figure 4-2. SDS-PAGE and native-PAGE analysis of WT apoA-IV and selected truncation mutants. (A) SDS-PAGE. Three microgram samples of purified apoA-IV
mutants were electrophoresed on an 18 % SDS polyacrylamide gel and stained with
Coomassie blue. Lane 1: Low-molecular weight markers. Lane 2: WT apoA-IV. Lane
3: apoA-IV ∆344-376. Lane 4: apoA-IV ∆333-376. Lane 5: apoA-IV ∆333-343. Lane
6: apoA-IV ∆333-337. Lane 7: apoA-IV ∆338-342. Lane 8: Low-molecular weight
marker. Lane 9: apoA-IV ∆1-10,333-343. Lane 10: apoA-IV ∆1-20,333-343. (B)
Native-PAGE. Three microgram samples of purified apoA-IV mutants were
electrophoresed an 8-25% gradient non-denaturing polyacrylamide gel and stained with
Coomassie blue. Lane 1: High-molecular weight markers. Lane 2: WT apoA-IV. Lane
3: apoA-IV ∆344-376. Lane 4: apoA-IV ∆333-376. Lane 5: apoA-IV ∆333-343. Lane
6: High-molecular weight markers. Lane 7: WT apoA-IV. Lane 8: apoA-IV ∆333-337.
Lane 9: apoA-IV ∆338-342. Lane 10: High-molecular weight markers. Lane 11: WT
apoA-IV. Lane 12: apoA-IV ∆1-39. ApoA-IV ∆1-10, 333-343 and apoA-IV ∆1-20, 333-
343 are not shown, but gave results similar to apoA-IV ∆333-343.
138 1.0
0.8 DMPC 5
32 WT apoA-IV 0.6 o D
0.4 OD/O 0.2 A ∆333-376 0.0 0 5 10 15 20 25 30 Time (min) 35
30
25 n (mN/m) 20 io s WT apoA-IV 15 Ten B ∆333-376 10 0 50 100 150 200 250 300 Time (sec)
Figure 4-3. Lipid association of WT apoA-IV and apoA-IV ∆333-376. (A) DMPC liposome clearance assay. DMPC multilamellar liposomes in STB were added to a cuvette maintained at 24.5˚C and optical density (OD) was continuously monitored at
325 nm following addition of 85 µg WT or mutant apoA-IV. OD values are normalized to the initial absorbance at time zero (ODo). (B) Oil drop tensiometry. A Tracker® oil drop tensiometer was used to monitor the decrease in surface tension induced by the binding of WT or mutant apoA-IV to a triolein/buffer interface. Surface tension was measured at 1 second intervals following rapid injection of a 10 µl drop of pure triolein into a cuvette containing 25 µg/ml WT apoA-IV or ∆333-376.
139 1.0 1.0 ∆344-376 0.8 ∆354-367 0.8 5 WT apoA-IV 32 WT o 0.6 0.6 D
0.4 0.4 OD/O 0.2 ∆333-376 0.2 ∆333-343 0.0 A 0.0 B 0 5 10 15 20 25 30 0 5 10 15 20 25 30
Time (min)
Figure 4-4. Lipid association of WT apoA-IV and apoA-IV ∆344-376 and ∆354-367.
DMPC multilamellar liposomes in STB were maintained at 24.5˚C by a circulating water bath and monitored at 325 nm following addition of 85 µg protein. The time dependent optical density (OD) was normalized to the initial absorbance (ODo).
140 333 334 335 336 337 338 339 340 341 342 343 Human: S F F S T F K E K E S Macaque: S F F S T F K E K E S Baboon: S F F S T F K E K E S A Pig: T F F S T L K E E A S Rat: S F M S T L Q K K G S Mouse: S F M S T L E K K G S Chicken: T F L S T T E Q A E S
1.0
0.8
5 WT 32 0.6 o D B O 0.4 OD/ 0.2 ∆338-342
∆333-337 0.0 0 5 10 15 20 25 30 Time (min)
Figure 4-5. Sequence alignment of the putative “inhibitory” region from amino acid
333 to 343 and the lipid binding performance of resulting mutants. (A) Species sequence alignment of human apoA-IV. The amino acids are highlighted depending on their physical properties: Yellow, neutrally charged polar; green, hydrophobic, aqua; positively charged; red negatively charged. (B) DMPC liposome solubilization by mutants lacking these two regions. This panel is organized similarly to Fig. 4-3A.
141
1.0 1.0 ∆11-20,333-343 0.8 0.8 ∆1-39,333-343
325 ∆1-20,333-343 o 0.6 WT apoA-IV 0.6
/OD 0.4 0.4
OD ∆333-343 0.2 0.2
∆1-10,333-343 0.0 A 0.0 B 0 5 10 15 20 25 30 0 5 10 15 20 25 30 Time (min)
Figure 4-6. DMPC liposome solubilization by WT apoA-IV and N-terminal mutants also lacking the inhibitory region from 333-343. Both panels are organized similarly to Fig. 4-3A.
142
3.0 ∆333-376
2.5
F ∆333-343 / o 2.0 F
∆344-376 1.5 WT
1.0 0.00 0.05 0.10 0.15 0.20 [Acrylamide] (M)
Figure 4-7. Steady-state Stern-Volmer plot of WT apoA-IV and deletion mutants.
Fluorescence emission spectra of WT and mutant apoA-IV at concentration of 0.1 mg/ml were collected in STB at room temperature an excitation wavelength of 295 nm.
Fluorescence emission was scanned from 302-375 nm to determine Fo/F, the ratio of the maximum intensity in the absence of acrylamide to the intensity in the presence of the indicated concentration of acrylamide. Each data point is the average of three separate observations (± S.D.). The lines are linear regressions of the data; the slopes of the lines give the Stern-Volmer quenching constants.
143 CHAPTER 5:
IDENTIFICATION OF AN APOLIPOPROTEIN A-IV SEQUENCE
RESPONSIBLE FOR CENTRAL INHIBITION OF FOOD INTAKE
ABSTRACT
Apolipoprotein A-IV (apoA-IV) is secreted by the intestine as a component of lymph chylomicrons. Its expression is dramatically increased in response to lipid feeding, implying a role in triglyceride-rich lipoprotein metabolism. In addition, apoA-IV has been shown to inhibit food intake in rats when injected centrally or peripherally. This
anorectic effect is not shared by the related protein apoA-I. In the current study, we
hypothesized that a specific sequence within rat apoA-IV is responsible for mediating the
anorectic effect. We used a bacterial expression system to generate truncation mutants of
rat apoA-IV and assess the ability of various regions of the molecule to mediate satiety in
a rat food intake model. The results indicated that an active sequence exists within the N-
terminal 61 amino acids of rat apoA-IV. Synthetic peptides were used to isolate the
region to between residues 17 and 30. A 14-mer peptide encompassing this sequence
was capable of reducing food intake in a dose dependent manner with a 60% reduction
when 1 µg was injected into the 3rd cerebral ventricle of fasted rats. A peptide modeled on a more C-terminal region of apoA-IV failed to exhibit a dose dependent effect. The isolation of this sequence provides a valuable tool for future work directed at identifying the apoA-IV binding protein and is a key step for exploring the potential of therapeutic manipulation of food intake via this pathway. However, the mechanism(s) by which apoA-IV mediates this anorectic effect are unknown. So, in addition to localizing the
144 active region of apoA-IV, we determined whether apoA-IV was able to cross the blood-
brain barrier. We injected 125I-labeled recombinant rat apoA-IV along with 99mTc-labeled
BSA into the periphery of mice and then isolated brain tissue and found that apoA-IV
does not appear to cross the blood-brain barrier.
INTRODUCTION
Apolipoprotein A-IV is a lipid-binding protein first isolated from rat plasma high
density lipoprotein (HDL) (1). In humans, it is synthesized by the small intestine and
secreted into the intestinal lymphatics associated with chylomicrons. In rodents, apoA-
IV is also produced in the liver (2). Examination of the amino acid sequence indicates
the presence of amphipathic helical repeat structures that likely mediate its interactions
with lipids (3). This is a theme shared by other exchangeable apolipoproteins such as apo
A-I. Indeed, apoA-IV has been suggested to play many of the same roles as apoA-I and
may have evolved directly from a precursor apoA-I-like protein (4). For example, apoA-
IV has been shown to activate lecithin:cholesterol acyl transferase (5), cholesterol ester
transfer protein (6) and to promote cholesterol efflux from cells (7). It associates with many of the same lipoproteins as apoA-I, although some reports suggest that it is present in the lipid-free form to a greater extent, at least in human plasma (8-10). ApoA-IV clearly plays an important role in the clearance of dietary fat as it is a key component of chylomicrons and variants of the protein can affect postprandial responses in triglyceride-
rich lipoproteins (11). Mice made transgenic with either human (12) or mouse (13)
apoA-IV are significantly protected from atherosclerosis. In addition, apoA-IV has also
been demonstrated to possess antioxidant properties (14). However, at least in terms of
145 lipoprotein metabolism, there is not yet a clear function for apoA-IV that is not shared by
other apolipoproteins.
Interestingly, unlike the other exchangeable apolipoproteins, apoA-IV message
levels and protein secretion are markedly increased during lipid absorption in the gut
(15). Under conditions of high lipid load, apoA-IV can account for up to 3% of total gut
secreted protein (15;16). In addition to the clearly implied role in chylomicron assembly
and TG mobilization, several studies have suggested that levels of apoA-IV may also be
an important feedback mechanism for the regulation of food intake. Fujimoto et al. (17)
showed that mesenteric lymph from lipid infused donor rats exerted an anorectic effect
when intravenously infused into fasted rat recipients. When this lymph was
immunodepleted of apoA-IV, the effect was blocked. Importantly, the effect was not
mediated by infused apoA-I. Furthermore, it was demonstrated that apoA-IV had a direct
effect on food intake when injected into the third ventricle of the rat brain (18). By contrast, antibodies to apoA-IV injected at the same location increased food intake in the animals vs. saline. Interestingly, the dose required to achieve the anorectic effect in the brain was 50 fold lower than when apoA-IV was injected peripherally.
More experiments have bolstered the connection between apoA-IV levels and effects on satiety. Fukagawa et al. (19) have shown a relationship between the circadian rhythm of apoA-IV in circulation and periods of active feeding in rats. Additionally, apoA-IV has been shown to inhibit the rate of gastric emptying (20;21) and gastric acid secretion (22) by acting through the central nervous system. Both effects may offer potential mechanisms for the regulation of satiety by apoA-IV. Intriguingly, Liu et al.
(23) have recently reported that the hypothalamus can produce apoA-IV de novo. During
146 periods of fasting, hypothalamic apoA-IV mRNA levels decrease and then rise upon lipid
feeding (24). However, this responsiveness of brain apoA-IV levels to the diet tends to
degrade over prolonged periods of fat absorption (25), implying that a chronic high fat
diet may reduce the responsiveness of the satiety feedback pathway resulting in long-term
over eating. Indeed, the potential role of apoA-IV in the long-term effects of obesity are
illustrated by the fact that certain polymorphisms of apoA-IV are associated with a higher
body mass index than others (26).
As suggested above, apoA-IV inhibits food intake centrally. There are two main
ways that satiety peptides can signal the brain, either through peripheral nerves or
through receptors in the brain itself (27). The hypothalamic production of apoA-IV
suggests there may be a mechanism or nerve stimulation acting as a satiety signal between the gut and the brain to produce apoA-IV. On the other hand, circulating apoA-
IV, or a peptide derived from it, may cross the blood-brain-barrier (BBB) and exert its anorectic effects centrally. In fact, either nerve stimulation or apoA-IV crossing the BBB could then bind to a receptor in the brain and serving as a satiety signal.
In the current study, we hypothesized that apoA-IV mediates its effects through a molecular interaction that requires a specific sequence within the protein. To test this, we performed deletion mutagenesis on rat apoA-IV and produced synthetic peptides for various regions of the sequence. The results indicate that a domain in the N-terminus of apoA-IV is responsible for its anorectic effects. Secondly, we hypothesized that apoA-IV has the ability to cross the BBB. In order to test this, we radiolabeled apoA-IV and injected it into the periphery of mice and then excised the brains and found that apoA-IV did not cross the BBB.
147 EXPERIMENTAL PROCEDURES
Construction, expression and purification of apoA-IV truncation mutants. The development of the rat apoA-IV expression system has been described
previously (28). The wild type (WT) apoA-IV construct in the pET-30 vector
(Invitrogen) was used as the template for PCR based mutagenesis. The C-terminal
mutations were created by performing PCR-based site-directed mutagenesis
(QuickChange, Novagen, Madison, WI) in order to insert stop codons at positions 249 or
117, respectively (Figure 5-1). The PCR was performed using a Perkin Elmer
thermocycler (94°C, 1 min: 18 cycles of 94°C-1 min, 72°C-1 min, 68°C-10 min:
followed by an additional 10 min at 68°C to assure full extension). The ∆1-61 mutant
was generated by a “flap” PCR approach in which a forward primer was designed with a
clamp region that matched nucleotides beginning at amino acid 62 and a 5’ flap region
that contained the His-tag and an NcoI site for ligation back into the pET30 vector (29).
The PCR was performed using a Perkin Elmer thermocycler (94°C, 1 min: 30 cycles of
94°C-1.5 min, 55°C-1.5 min, 72°C-1.5 min: followed by an additional 10 min at 68°C to
assure full extension). The sequence of each mutant construct was verified on an Applied
Biotechnology System DNA sequencer at the University of Cincinnati DNA Core.
The expression constructs were transfected into E. Coli BL-21 cells and
grown on Luria-Bertani (LB) agar plates containing 30 µg/ml kanamycin (Calbiochem,
San Diego, CA) for 16 h at 37˚C. A single colony was inoculated into 10 ml culture
media tubes and grown overnight at 37˚C. These were then used to inoculate 100 ml LB
cultures that were grown for approximately 3 h until an A600 of between 0.6-0.7. Protein expression was driven by the T7 promoter upon exposure to 50 µl of 1 M IPTG for 2 h.
148 After expression, the bacterial cells were pelleted at 9700g in a Sorvall SLA1500 for 10
min, the supernatant was removed and the cells were frozen at -20˚C overnight. The cells
were resuspended in 4 ml 1X Histidine (His) bind buffer (5 mM imidazole, 0.5 M NaCl,
20 mM Tris:HCl) per 100ml of original culture with a final concentration of protease
inhibitors; 100 µM PMSF, 20 µM Leupeptin and 1 µM Pepstatin A. The cells were lysed
with a Model 550 Sonic Dismembrator at level 5 (Fisher, Pittsburgh, PA) 3 times for 1
min with a 1 min incubation on ice in between. The lysed cells were then spun at 28k x g
for 20 min in a Sorvall SLA600, leaving the recombinant apoA-IV in the supernatant.
The presence of the His-tag allowed for easy purification by using a His-binding column
(Ni2+-affinity column). The supernatant was passed through a 0.45 uM filter and added to a prepared His-bind column (Novagen) poured with HisBind Resin and the column was eluted according to manufacturer’s instructions. The resulting protein was concentrated by using Centricon fitration units (Millipore, Bedford, MA) to a concentration of approximately 5 mg/ml and then dialyzed into standard Tris buffer
(STB; 10 mM Tris-HCl, 1 mM EDTA, 150 mM NaCl and 0.2% NaN3). The protein concentration was determined and the protease Igase (MoBiTec, Germany) was added (1
µl of Igase/1 mg apoA-IV) for 24h at 37˚C in a shaking incubator. Igase cleaves the leader portion (with the His-tag) of the protein, thus leaving the mature form with an additional Thr-Pro on the N-terminus. After cleavage, the protein was applied to the His- bind column a second time to separate the mature protein from the cleaved tag. The flowthrough was collected, concentrated and dialyzed into STB or saline for storage.
Removal of lipopolysaccharide (LPS).
149 Bacterial LPS was removed from the fully purified apoA-IV samples by passing it down an AffinityPak Detoxi-Gel endotoxin removing gel (Pierce, Rockford, IL). The protein was diluted to a final concentration of 1.5 mg/ml in STB. The column was equilibrated in STB buffer and then 1 ml of sample was applied to each column and incubated for 1 h at R.T. Each column was then eluted with 10 ml of STB and concentrated. Post column samples were sent to Cambrex-BioWhittaker for a Kinetic
QCL colorimetric assay of endotoxin levels.
Native rat apoA-IV purification.
Rat plasma was obtained from abdominal aorta blood of rats. A series of delipidation and centrifugation steps were performed in order to isolate various apolipoproteins. The rat apolipoproteins were then separated by SDS-PAGE and were purified by gel extraction. For a more detailed description, please refer to Hayashi et al.
(15).
Peptide synthesis.
Synthetic peptides were generated by automated solid phase peptide synthesis on a Milligen 9050 Peptide Synthesizer at the Louisiana State University core facility. They were washed in diethyl ether and freeze dried for use. Electrospray mass spectrometry was performed on each peptide to determine purity and correct molecular weight. In some cases, the peptides were sequenced directly using an Applied Biosystems 477A
Protein Sequencer. The sequences of the peptides used in this study were as follows:
ApoA-IV (1-30) EVTSDQVANVMWDYFTQLSNNAKEAVEQLQ, apoA-IV (1-15)
150 EVTSDQVANVMWDYF, apoA-IV (17-30) QLSNNAKEAVEQLQ, and ApoA-IV
(211-232) QEKLNHQMEGLAFQMKKNAEEL.
Food intake studies.
Male Sprague-Dawley rats (275-300 g, Harlan, Indianapolis, IN) were individually housed under stable environmental conditions with a 12-h dark/12-h light cycle with free access to pelleted food and water. A 21-gauge cannula (Plastic One,
Roanoke, VA) was placed stereotaxically into the third cerebral ventricle using the method described previously (28). All procedures were performed in accordance with the guidelines of the University Institutional Animal Care and Use Committee at the
University of Cincinnati. Cannula placement was verified one week after surgery by i3vt injection of angiotensin II (10 ng). Only those animals consuming more than 5 ml of water in 60 min post-injection were included in the study. Two weeks after cannula implantation, rats were fasted for 24 h and assigned to weight-matched groups. At the onset of dark, rats received i3vt injections (4 µl) containing saline, or the indicated apoA-
IV mutant or peptide in saline. The rats were visually observed, and food and water consumption was measured at 0.5, 1, 2, 4 h post-injection.
Statistical analysis.
Results are presented as mean ± SE. Data were analyzed by one-way analysis of variance (ANOVA) followed by Tukey’s test. Differences were considered significant
151 when the probability of the difference occurring by chance was less than 5 in 100, 2-
tailed (P < 0.05).
BBB studies
We purchased 125I-labeled leptin and then 99mTc-labeled BSA following the protocol of Banks et al. (30;31). We then labeled recombinant apoA-IV with 125I by the
Iodo-Bead reagent (Pierce Chemicals, Rockford, IL) method and purified on a column of
Sephadex G-10. BSA was a vascular marker of a functionally intact BBB because it has no significant entry into the CNS when the BBB is intact. To confirm the methodology, we injected 125I-labeled leptin and 99mTc-labeled BSA through the exposed left jugular vein of C57B6 mice (at 1 x 106 cpm) in 200 µl of lactated Ringer’s solution. Arterial
blood was collected via the carotid artery at 1, 7.5, 15 and 30 min (n=3 for each time
point). Immediately following, the animals were sacrificed and the whole brain was
removed, rinsed with saline (to remove excess blood), and weighed. Whole blood was
centrifuged at 4000 × g at 4°C for 10 min and plasma was separated. The levels of
radioactivity in the plasma and the brain region tissue were determined in a gamma
counter which can recognize a difference between 125I-labeled leptin and 99mTc-labeled
BSA. The ratio of radioactivity of the brain tissue (cpm/g) to plasma (cpm/µl) was
calculated and plotted over time. This same experiment was repeated with 125I-labeled apoA-IV and 99mTc-labeled BSA but more frequent time points were used at 1, 2.5, 5,
7.5, 15 and 30 min (n=3 for each time point).
152 RESULTS
The approach for this study was straightforward. To test the hypothesis that apoA-IV contains a specific sequence that is responsible for its anorectic effect, we used deletion mutagenesis to arbitrarily divide the molecule into approximate thirds and evaluate the ability of the remaining protein to mediate the anorectic effect (Figure 5-1).
We previously demonstrated that the recombinant protein produced in our expression system is structurally and functionally similar to native rat apoA-IV in terms of its ability to bind lipid and inhibit food intake in rats (28). An example of this is shown in Figure 5-
2. When injected into the 3rd ventricle of the rat brain prior to the onset of dark,
recombinant apoA-IV significantly reduced 4 h food intake in a dose dependent manner.
A 4 µg dose of native rat apoA-IV also significantly suppressed food intake over 4 h with
a magnitude similar to the recombinant protein. This indicates that the recombinant form
of apoA-IV recapitulates the effects of the native protein.
However, before initiating studies with a series of mutants, we addressed a
potential complication of injecting bacterial expressed protein into a living brain. It is well known that bacterial cells contain significant levels of lipopolysaccharide (LPS), also known as endotoxin, in the cell wall. LPS is a known neurotoxin that elicits inflammatory responses in many tissues. If present in our purified protein preparations, it is possible that LPS may cause the reduction of food intake that is independent of the apoA-IV effect. To assure that the LPS did not have an effect on the rat food intake studies, we removed it by column chromatography on an endotoxin affinity column. LPS analysis showed that this treatment significantly reduced the LPS in the apoA-IV preparation. The untreated sample [LPS (+)] contained 31 endotoxin units per µg of
153 apoA-IV while the column treated sample [LPS (-)] contained approximately 1 endotoxin unit per µg of apoA-IV. We then compared the two preparations of apoA-IV in a rat food intake experiment. Figure 5-3 shows a time course of food intake over 4 h post infusion of these two samples vs. saline alone. It is clear that both recombinant apoA-IV samples significantly reduced food intake at 2 and 4 h vs. the saline control.
Furthermore, the presence of the miniscule amount of LPS in the untreated sample had no significant impact on the anorectic effect. Because of pragmatic considerations, we chose to perform the remainder of the studies without further removal of the LPS from our samples.
Confident in the food intake measurement using recombinant proteins, we proceeded to generate C-terminal truncation mutants of rat apoA-IV (Figure 5-1). A stop codon was placed at proline 249 to generate apoA-IV ∆249-371. This terminated protein translation at the junction between 22 amino acid helical repeats 9 and 10 (as numbered according to (32)). An additional mutant, apoA-IV ∆117-371, was generated by placing a stop codon at proline 117 located at the junction between helical repeats 3 and 4.
Figure 5-4 shows an SDS-PAGE analysis of the of the purified C-terminal mutants. WT rat apoA-IV (lane 4) migrated to an apparent MW of about 43 kDa in agreement with its predicted mass of 42,635 Da (including the Thr-Pro- on the N-terminus of the protein).
ApoA-IV ∆249-371 and apoA-IV ∆117-371 migrated to apparent masses of 28 kDa
(predicted 28.6 kDa) and 14 kDa (predicted 13.4 kDa), respectively. It is clear that the recombinant proteins were highly homogeneous.
The ability of the C-terminal truncation mutants of rat apo A-IV to inhibit food intake in fasted rats was then assessed. Figure 5-5A confirms that the WT recombinant
154 apoA-IV reduced food intake at both 2 and 4 h compared to the saline control. Similarly,
apoA-IV ∆249-371 inhibited food intake at the same time points. Apo A-IV ∆117-371
also demonstrated an anorectic effect at 4 h, although a trend toward inhibited food intake at 2 h did not reach statistical significance. Although this mutant was not as effective as the WT protein at inhibiting food intake at the 4 h time point, it still exhibited a clear and reproducible anorectic effect.
It follows from the results in Figure 5-5 that a sequence located in the N-terminal
116 amino acids contains a site that is responsible for food intake inhibition, as this sequence was intact in all the mutants. To probe further, we generated an N-terminal mutant that lacked the first 61 amino acids of the protein. This site was chosen because it divided the N-terminal 117 residues roughly in half and is located at the helical junction between repeats 1 and 2 of apoA-IV. Figure 5-4 shows the SDS-PAGE analysis of this mutant. A minor contaminant band is apparent at about 32 kDa that is likely a proteolytic cleavage product of the truncation mutant. Interestingly, Figure 5-6 shows that this mutant was indistinguishable from the saline control in terms of inhibiting food intake in the rat 3rd ventricle. This strongly suggests that the active sequence resides in the N- terminal 61 amino acids of rat apoA-IV.
Peptide studies. In independent experiments, we generated synthetic peptides corresponding to the rat apoA-IV sequence. Figure 5-7 shows an overlay of the N- terminal 61 amino acids of rat apoA-IV compared to the sequences of human, mouse, baboon, pig and chicken. It is clear that this domain contains significantly conserved regions, especially within the first 30 amino acids. We generated a series of three
155 peptides encompassing the first 30 amino acids of the protein (1-30, 1-15, and 17-30).
An additional peptide that contained an amphipathic helical sequence from the C-
terminal region of the protein, apoA-IV (211-232), was also included as a control. These
were injected into the 3rd ventricle of rats to assess the impact on food intake. Figure 5-
8A shows the effects of this series of peptides at two doses on rat food intake 30 minutes
after injection. Compared to saline, the apoA-IV (1-15) peptide preparations were similar
to the saline control at both doses tested. Similarly, the C-terminal peptide apoA-IV
(211-232), although it exhibited a slight anorectic effect at the low dose, failed to mediate a further effect at the higher dose. By contrast, apoA-IV (1-30) exhibited a small decrease in food intake at the low dose and a significant effect at the high dose. ApoA-
IV (17-30) exhibited the most potent anorectic effect at 0.5 µg which became more pronounced at the higher concentration. The effect of apoA-IV (17-30) is further demonstrated in Figure 5-8B with larger groups of rats at the 1.0 µg dose. Since both active peptides contain the sequence 17-30, the simplest explanation for the data is that this region contains a sequence capable of mediating the anorectic effect.
Blood-brain barrier studies. In order to gain insight into the mechanism of how apoA-IV can mediate the anorectic effect, we performed an experiment to determine if intact apoA-IV can cross the BBB. In order to verify the experimental system, we first performed the BBB studies with radiolabeled leptin to repeat the experiments done by
Banks et al. that showed leptin is able to cross the BBB (31). Figure 5-9 is a graph showing the ratio of cpm in the brain as compared to the cpm in the serum on the Y-axis over time on the X-axis. Indeed, there is an increase in the presence of leptin in the brain
156 over time. Also, we used labeled BSA as a negative control as it does not cross an intact
BBB. Figure 5-9 shows that BSA did not accumulate in the brain. Next, we determined
whether apoA-IV crossed the BBB using the same experimental design. Figure 5-10
shows the ratio of brain (cpm) to serum (cpm) over time for both apoA-IV and BSA. In
this case, there is no increase for apoA-IV over time as compared to the leptin
experiment. This suggests that unlike leptin, apoA-IV does not have the ability to cross
the BBB.
DISCUSSION
In the present report, we provide evidence from recombinant deletion mutants and
synthetic peptides that a region near the N-terminus between residues 17 and 30 of apoA-
IV is capable of mediating an anorectic effect when administered to the 3rd ventricle of the rat brain. The fact that only a single 14-mer peptide can, at least partially, recapitulate the effects of the full length protein indicates that the effect depends on the amino acid sequence and not on the functionality of the fully folded protein. The significance of these findings with respect to the mechanism of the anorectic effect of apoA-IV and its therapeutic potential are discussed below.
It is clear that apolipoprotein A-IV suppressed food intake when administered either centrally or, at much higher doses, peripherally into the rat. This effect is not shared by apoA-I. Conversely, when endogenous levels of apoA-IV in the brain are immuno-depleted, food intake increases. The most intuitive mechanism that explains the data is that apoA-IV levels in plasma increase in response to a fatty meal and somehow signals the hypothalamus to reduce food intake. From our BBB results, it does not
157 appear that apoA-IV is crossing the BBB directly. Therefore the alternative mechanism that apoA-IV may act locally in the gut to trigger afferent vagal inputs to the hypothalamus may be more likely. Regardless of the location, the initial signaling event could occur via two general mechanisms. The first could be a direct interaction, such as through a specific protein-protein interaction (i.e. a receptor). Indeed, there have been several reports of specific binding of apoA-IV to various cell types (33-35) although a specific binding protein has not yet been identified, either in the periphery or in the brain.
In the absence of a binding protein, apoA-IV might also be envisioned to work through an indirect mechanism such as manipulating the lipid content of a neuronal membrane to modify its activity. However, we suggest that the indirect mechanism is less plausible for several reasons. First, apoA-I is known to be a more effective lipid binding molecule
(36;37) and would be expected to more readily manipulate membrane lipid content than apoA-IV (29), yet it does not exhibit the anorectic effect. Second, the 14-mer peptide that we identified in this work is in a domain that does not appear to contain significant amphipathic helical character, a trait that is likely critical for lipid interactions. We suggest that the ability of the 14-mer peptide to promote the effect in the absence of the rest of the apoA-IV molecule argues for some form of a direct protein-protein interaction.
Confirmation of this hypothesis must await further studies directed at isolating the putative binding protein. The active 14-mer peptide will likely be a useful tool for such pursuits.
The results of this study may also be informative with respect to the delivery of apoA-IV to the site of activity because the full-length protein does not cross the BBB.
Ordinarily, a 43 kDa protein such as full length apoA-IV would not be expected to cross
158 the blood-brain barrier in significant amounts unless an active transport system was at
work. Our observations that a small peptide can be efficacious raises the possibility that
apoA-IV can undergo proteolytic cleavage in the circulation resulting in a small
biologically active peptide that would be more likely to diffuse across the blood-brain
barrier. Further studies will be required to determine if apoA-IV is susceptible to such a
cleavage both in vitro and in vivo. Again, if no form of apoA-IV is crossing the BBB, it
will be very important to look at whether vagus nerve stimulation has any effect on food
intake. The afferent fibers of the vagus nerve have been shown to transmit signals for several meal-related metabolites as well as various mechanical and chemical stimuli
(38;39). Therefore, it will be informative to perform food intake studies in normal rats injected peripherally with apoA-IV and also in rats that have had a vagotomy. This experiment has been done quite successfully with other satiety peptides (40;41).
In terms of the structure of the anorectic domain itself, the sequence is relatively well conserved among the species with known apoA-IV sequences. An exception to this is the chicken sequence which contains only 4 identical residues among the 14 (Figure 5-
7). Chicken apoA-IV is a more hydrophobic protein than that from mammals and it lacks a unique glutamine rich domain present at the C-terminus (42). It has been proposed that the addition of the C-terminal domain evolved in mammals to assist in chylomicron formation with the advent of apoB mRNA editing (43). Based on structural arguments, we have suggested that this C-terminal domain may have imparted dramatically different lipid binding characteristics on mammalian apoA-IV vs. the chicken, perhaps facilitating a change in apoA-IV function from a lipid binding structural protein to a signaling protein with a role in food intake (29). Following this logic, it may be possible that the
159 anorectic sequence in the N-terminus also evolved as a binding site for this signaling function in mammals.
Finally, we would point out that this work has several implications for the treatment of obesity in humans. The identification of the anorectic domain within apoA-
IV gives rise to the possibility that stable derivatives of the peptide could be used for pharmaceutical manipulation of satiety. In addition, this work strongly suggests that there is a specific binding protein for apoA-IV that mediates its effects on the nervous system. Identification of such mediators will be important for further understanding the complex and redundant set of pathways that govern the critical function of food intake.
160 REFERENCES
1. Swaney,J.B., Reese,H., and EDER,H.A. (1974) Polypeptide composition of rat high density lipoprotein: characterization by SDS-gel electrophoresis, Biochem. Biophys. Res. Commun. 59, 513- 519.
2. Elshourbagy,N.A., Boguski,M.S., Liao,W.S., Jefferson,L.S., Gordon,J.I., and Taylor,J.M. (1985) Expression of rat apolipoprotein A-IV and A-I genes: mRNA induction during development and in response to glucocorticoids and insulin, Proc. Natl. Acad. Sci. U. S. A 82, 8242-8246.
3. Boguski,M.S., Elshourbagy,N., Taylor,J.M., and Gordon,J.I. (1985) Comparative analysis of repeated sequences in rat apolipoproteins A-I, A-IV, and E, Proc. Natl. Acad. Sci. USA 82, 992-996.
4. Karathanasis,S.K., Oettgen,P., Haddad,I.A., and Antonarakis,S.E. (1986) Structure, evolution, and polymorphisms of the human apolipoprotein A4 gene (APOA4), Proc. Natl. Acad. Sci. U. S. A 83, 8457-8461.
5. Emmanuel,F., Steinmetz,A., Rosseneu,M., Brasseur,R., Gosselet,N., Attenot,F., Cuine,S., Seguret,S., Latta,M., Fruchart,J.C., and et al (1994) Identification of specific amphipathic alpha- helical sequence of human apolipoprotein A-IV involved in lecithin:cholesterol acyltransferase activation, J. Biol. Chem. 269, 29883-29890.
6. Main,L.A., Ohnishi,T., and Yokoyama,S. (1996) Activation of human plasma cholesteryl ester transfer protein by human apolipoprotein A-IV, Biochim. Biophys. Acta 1300, 17-24.
7. Remaley,A.T., Stonik,J.A., Demosky,S.J., Neufeld,E.B., Bocharov,A.V., Vishnyakova,T.G., Eggerman,T.L., Patterson,A.P., Duverger,N.J., Santamarina-Fojo,S., and Brewer,H.B., Jr. (2001) Apolipoprotein specificity for lipid efflux by the human ABCAI transporter, Biochem. Biophys. Res. Commun. 280, 818-823.
8. Bisgaier,C.L., Sachdev,O.P., Megna,L., and Glickman,R.M. (1985) Distribution of apolipoprotein A-IV in human plasma, J. Lipid Res. 26, 11-25.
9. DeLamatre,J.G., Hoffmeier,C.A., Lacko,A.G., and Roheim,P.S. (1983) Distribution of apolipoprotein A-IV between the lipoprotein and the lipoprotein-free fractions of rat plasma: possible role of lecithin:cholesterol acyltransferase, J. Lipid Res. 24, 1578-1585.
10. Ghiselli,G., Crump,W.L., Musanti,R., Sherrill,B.C., and Gotto,A.M.J. (1989) Metabolism of apolipoprotein A-IV in rat, Biochim. Biophys. Acta 1006, 26-34.
11. Ostos,M.A., Lopez-Miranda,J., Ordovas,J.M., Marin,C., Blanco,A., Castro,P., Lopez-Segura,F., Jimenez-Pereperez,J., and Perez-Jimenez,F. (1998) Dietary fat clearance is modulated by genetic variation in apolipoprotein A-IV gene locus, J. Lipid Res. 39, 2493-2500.
12. Duverger,N., Tremp,G., Caillaud,J.M., Emmanuel,F., Castro,G., Fruchart,J.C., Steinmetz,A., and Denefle,P. (1996) Protection against atherogenesis in mice mediated by human apolipoprotein A-IV, Science 273, 966-968.
13. Cohen,R.D., Castellani,L.W., Qiao,J.H., Van Lenten,B.J., Lusis,A.J., and Reue,K. (1997) Reduced aortic lesions and elevated high density lipoprotein levels in transgenic mice overexpressing mouse apolipoprotein A-IV, J. Clin. Invest 99, 1906-1916.
14. Qin,X., Swertfeger,D.K., Zheng,S., Hui,D.Y., and Tso,P. (1998) Apolipoprotein AIV: a potent endogenous inhibitor of lipid oxidation, American Journal of Physiology 274, H1836-H1840.
161 15. Hayashi,H., Nutting,D.F., Fujimoto,K., Cardelli,J.A., Black,D., and Tso,P. (1990) Transport of lipid and apolipoproteins A-I and A-IV in intestinal lymph of the rat, J. Lipid Res. 31, 1613-1625.
16. Tso,P., Liu,M., Kalogeris,T.J., and Thomson,A.B. (2001) The role of apolipoprotein A-IV in the regulation of food intake, Annu. Rev. Nutr. 21, 231-254.
17. Fujimoto,K., Cardelli,J.A., and Tso,P. (1992) Increased apolipoprotein A-IV in rat mesenteric lymph after lipid meal acts as a physiological signal for satiation, Am. J. Physiol 262, G1002-G1006.
18. Fujimoto,K., Fukagawa,K., Sakata,T., and Tso,P. (1993) Suppression of food intake by apolipoprotein A-IV is mediated through the central nervous system in rats, J. Clin. Invest 91, 1830- 1833.
19. Fukagawa,K., Gou,H.M., Wolf,R., and Tso,P. (1994) Circadian rhythm of serum and lymph apolipoprotein AIV in ad libitum-fed and fasted rats, Am. J. Physiol 267, R1385-R1390.
20. Okumura,T., Fukagawa,K., Tso,P., Taylor,I.L., and Pappas,T.N. (1996) Apolipoprotein A-IV acts in the brain to inhibit gastric emptying in the rat, Am. J. Physiol 270, G49-G53.
21. Raybould,H.E., Meyer,J.H., Tabrizi,Y., Liddle,R.A., and Tso,P. (1998) Inhibition of gastric emptying in response to intestinal lipid is dependent on chylomicron formation, Am. J. Physiol. 274, R1834-R1838.
22. Okumura,T., Fukagawa,K., Tso,P., Taylor,I.L., and Pappas,T.N. (1996) Apolipoprotein A-IV acts in the brain to inhibit gastric emptying in the rat, Am. J. Physiol 270, G49-G53.
23. Liu,M., Doi,T., Shen,L., Woods,S.C., Seeley,R.J., Zheng,S., Jackman,A., and Tso,P. (2001) Intestinal satiety protein apolipoprotein AIV is synthesized and regulated in rat hypothalamus, Am. J. Physiol Regul. Integr. Comp Physiol 280, R1382-R1387.
24. Liu,M., Shen,L., Liu,Y., Tajima,D., Sakai,R., Woods,S.C., and Tso,P. (2004) Diurnal rhythm of apolipoprotein A-IV in rat hypothalamus and its relation to food intake and corticosterone, Endocrinology 145, 3232-3238.
25. Liu,M., Shen,L., Liu,Y., Woods,S.C., Seeley,R.J., D'Alessio,D., and Tso,P. (2004) Obesity induced by a high-fat diet downregulates apolipoprotein A-IV gene expression in rat hypothalamus, Am. J. Physiol Endocrinol. Metab 287, E366-E370.
26. Lefevre,M., Lovejoy,J.C., DeFelice,S.M., Keener,J.W., Bray,G.A., Ryan,D.H., Hwang,D.H., and Greenway,F.L. (2000) Common apolipoprotein A-IV variants are associated with differences in body mass index levels and percentage body fat, Int. J. Obes. Relat. Metab. Disord. 24, 945-953.
27. Woods,S.C., Seeley,R.J., Porte,D., Jr., and Schwartz,M.W. (1998) Signals that regulate food intake and energy homeostasis, Science 280, 1378-1383.
28. Liu,M., Maiorano,N., Shen,L., Pearson,K., Tajima,D., Zhang,D.M., Woods,S.C., Seeley,R.J., Davidson,W.S., and Tso,P. (2003) Expression of biologically active rat apolipoprotein AIV in Escherichia coli, Physiol Behav. 78, 149-155.
29. Pearson,K., Saito,H., Woods,S.C., Lund-Katz,S., Tso,P., Phillips,M.C., and Davidson,W.S. (2004) Structure of human apolipoprotein A-IV: a distinct domain architecture among exchangeable apolipoproteins with potential functional implications, Biochemistry 43, 10719-10729.
30. Banks,W.A. and Kastin,A.J. (1993) Physiological consequences of the passage of peptides across the blood-brain barrier, Rev. Neurosci. 4, 365-372.
162 31. Banks,W.A., Kastin,A.J., Huang,W., Jaspan,J.B., and Maness,L.M. (1996) Leptin enters the brain by a saturable system independent of insulin, Peptides 17, 305-311.
32. Segrest,J.P., Jones,M.K., De Loof,H., Brouillette,C.G., Venkatachalapathi,Y.V., and Anantharamaiah,G.M. (1992) The amphipathic helix in the exchangeable apolipoproteins: a review of secondary structure and function. [Review], J. Lipid Res. 33, 141-166.
33. Weinberg,R.B. and Patton,C.S. (1990) Binding of human apolipoprotein A-IV to human hepatocellular plasma membranes, Biochim. Biophys. Acta 1044, 255-261.
34. Savion,N. and Gamliel,A. (1988) Binding of apolipoprotein A-I and apolipoprotein A-IV to cultured bovine aortic endothelial cells, Arteriosclerosis 8, 178-186.
35. Dvorin,E., Gorder,N.L., Benson,D.M., and Gotto,A.M., Jr. (1986) Apolipoprotein A-IV. A determinant for binding and uptake of high density lipoproteins by rat hepatocytes, J. Biol. Chem. 261, 15714-15718.
36. Rifici,V.A., EDER,H.A., and Swaney,J.B. (1985) Isolation and lipid-binding properties of rat apolipoprotein A-IV, Biochim. Biophys. Acta 834, 205-214.
37. Weinberg,R.B., Ibdah,J.A., and Phillips,M.C. (1992) Adsorption of apolipoprotein A-IV to phospholipid monolayers spread at the air/water interface. A model for its labile binding to high density lipoproteins, J. Biol. Chem. 267, 8977-8983.
38. Moran,T.H., Baldessarini,A.R., Salorio,C.F., Lowery,T., and Schwartz,G.J. (1997) Vagal afferent and efferent contributions to the inhibition of food intake by cholecystokinin, Am. J. Physiol 272, R1245-R1251.
39. Williams,D.L., Grill,H.J., Cummings,D.E., and Kaplan,J.M. (2003) Vagotomy dissociates short- and long-term controls of circulating ghrelin, Endocrinology 144, 5184-5187.
40. Smith,G.P., Jerome,C., Cushin,B.J., Eterno,R., and Simansky,K.J. (1981) Abdominal vagotomy blocks the satiety effect of cholecystokinin in the rat, Science 213, 1036-1037.
41. Garlicki,J., Konturek,P.K., Majka,J., Kwiecien,N., and Konturek,S.J. (1990) Cholecystokinin receptors and vagal nerves in control of food intake in rats, Am. J. Physiol 258, E40-E45.
42. Steinmetz,A., Hermann,M., Nimpf,J., Aebersold,R., Ducret,A., Weinberg,R.B., and Schneider,W.J. (1998) Expression and conservation of apolipoprotein AIV in an avian species, J. Biol. Chem. 273, 10543-10549.
43. Weinberg,R.B., Anderson,R.A., Cook,V.R., Emmanuel,F., Denefle,P., Hermann,M., and Steinmetz,A. (2000) Structure and interfacial properties of chicken apolipoprotein A-IV, J. Lipid Res. 41, 1410-1418.
163 FIGURES:
1 50 100 150 200 250 300 350 371 N 12 345678 910 1112 C P P P P P P P P P
WT
∆ 249-371
∆ 117-371
∆ 1-61
Figure 5-1: Linear diagram of WT rat apoA-IV and recombinant deletion mutants.
The sequence is numbered according to the mature apoA-IV sequence (not including its
20 amino acid pre-peptide). The white boxes represent putative 22 amino acid amphipathic alpha helices. Most of the repeats are punctuated by proline residues (P).
The arrows are placed at sites where deletions were made either in the C- or N-terminus.
Each deletion mutant used in this study is represented by a solid black line showing the sequence that was left intact.
164
Figure 5-2: Food intake measurements in i3vt cannulated rats. 24 h fasted rats were injected with either saline alone or the indicated amount of apoA-IV (recombinant or native rat plasma) immediately before lights out and voluntary, accumulated food intake was measured for 4 h. The data are means ± S.E.M. Asterisks indicate significant differences from the saline group (*P< 0.05, **P< 0.01).
165
Figure 5-3: The effect of residual LPS on suppression of rat food intake by
recombinant apoA-IV. The rat food intake assay was performed as in Figure 5-2.
Accumulated food intake is plotted as a function of time post injection of 4 µg
recombinant apoA-IV. LPS (+) refers to a preparation of recombinant apoA-IV as
derived from the His-binding column described in Experimental Procedures prior to
injection into the 3rd ventricle of the rat brain. LPS (-) refers to a similar sample that had been passed down an endotoxin affinity column. The LPS (+) sample contained 31 endotoxin units per µg of apoA-IV and the LPS (-) sample contained approximately 1 endotoxin unit per µg of apoA-IV. The data are means ± S.E.M.
166
Figure 5-4: SDS-PAGE analysis of WT rat apoA-IV and recombinant deletion mutants. After expression and purification as described in Experimental Procedures, 4
µg samples of apolipoprotein were electrophoresed on an 18% denaturing
polyacrylamide gel and stained with Coomassie blue. Lane 1: Amersham low molecular
weight protein standards; Lane 2: apoA-IV ∆249-371; Lane3: apoA-IV ∆117-371; Lane
4: WT rat apoA-IV; Lane 5: standards; Lane 6: ∆1-61.
167
Figure 5-5: Effect of C-terminal truncation mutants of apoA-IV on rat food intake when injected centrally. The rat food intake assay was performed as in Figure 5-2.
Accumulated food intake is plotted as a function of time post injection of 4 µg recombinant apoA-IV. The data are means ± S.E.M.. Asterisks indicate significant differences from the saline group (*P< 0.05, **P< 0.01).
168
Figure 5-6: Effect of an N-terminal deletion of apoA-IV on rat food intake when injected centrally. The rat food intake assay was performed as in Figure 5-2.
Accumulated food intake is plotted as a function of time post injection of 4 µg recombinant apoA-IV. The data are means ± S.E.M..
169
Figure 5-7: Multi-species sequence alignment of the N-terminal 61 amino acids of apoA-IV. Amino acids highlighted in black are completely conserved across all species.
Those highlighted in dark gray are conserved across all but one species and those in light gray are conserved in at least 3 of 6 species. The solid lines above the sequences indicate the N-terminal peptides used in this study.
170
Figure 5-8: Effect of N-terminal and control synthetic peptides of rat apoA-IV on food intake. The rat food intake assay was performed as in Figure 5-2. Panel A:
Accumulated food intake for 30 minutes is plotted as a function of infusion dose of the relevant peptide in saline. The number of animals at each experimental point is 4. Panel
B: Direct comparison of food intake at 30 min after i3vt injection between saline and 1
µg of apoA-IV peptide (17-30), n = 7. All data are means ± S.E.M.
171
25
) 20 l/g
µ ( 15 m
seru
/ 10
Brain 5 Leptin BSA
0 0 5 10 15 20 25 30
Exposure time (min)
Figure 5-9: Leptin has the ability to cross the BBB. 125I-labeled leptin and 99mTc- labeled BSA were injected into the left jugular vein of mice in 200 µl of lactated Ringer’s solution. Arterial blood was collected via the carotid artery at 1, 7.5, 15 and 30 min.
Immediately following, the animals were sacrificed and the whole brain was removed, rinsed and weighed. Plasma and brain tissue were then counted and plotted as the ratio of radioactivity of brain tissue to plasma.
172
12
10
l/g) µ 8
m ( 6
seru / 4
Brain Apo A-IV 2 BSA
0 0 5 10 15 20 25 30
Exposure time (min)
Figure 5-10: ApoA-IV does not appear to cross the BBB. 125I-labeled recombinant
apoA-IV and 99mTc-labeled BSA were injected into the left jugular vein of mice in 200 µl
of lactated Ringer’s solution. Arterial blood was collected via the carotid artery at 1, 2.5,
5, 7.5, 15 and 30 min. Immediately following, the animals were sacrificed and the whole brain was removed, rinsed and weighed. Plasma and brain tissue were then counted and plotted as the ratio of radioactivity of brain tissue to plasma.
173 CHAPTER 6:
GENERAL DISCUSSION
The general goal of this research was to test the hypothesis that specific regions of
apoA-IV are responsible for its ability to bind lipid and independently, inhibit food
intake. The first aim of this research (Chapters 3 and 4) was to determine the region of
apoA-IV that is responsible for its ability to bind lipid as well as to determine its general
structural organization. The second aim of this research (Chapter 5) was to determine the
region of apoA-IV that is responsible for its role in the inhibition of food intake.
Aim 1: Despite predictions that apoA-IV would be similar to apoA-I and apoE, it was found instead that apoA-IV was organized quite differently. Structurally, apoA-IV was found to consist of a single domain structure from residues 40-332 instead of the two domain structure as suggested for apoA-I and apoE. Also, the lipid binding region was clearly not found at the C-terminus like both of these apolipoproteins. Rather, the C- terminus had an ‘inhibitory’ region that caused slower lipid clearance than seen in apoA-I or apoE. Indeed, removing this inhibitory region resulted in a mutant that was much more efficient at clearing liposome than the WT or apoA-I. However, removing the region from 11-20 as well as the ‘inhibitory’ region slowed down the lipid clearance similar to WT. This lipid binding data suggests that there is an interaction between the
N- and C-terminus and that was supported by further structural studies.
174 Discussion Aim 1: These studies were important because they have shown that all
apolipoproteins do not share a similar two domain structure. This was significant
because it is well accepted that structure affects function so this could have far reaching
effects in vivo. The different tertiary conformation may allow apoA-IV to perform functions that are unrelated to the other apolipoproteins such as its role in food intake or
inflammation.
It has been shown previously that apoA-IV was able to bind to lipid, however, at a
much lower affinity than all of the other apolipoproteins tested. Importantly, my work
has shown, for the first time, that apoA-IV has an ‘inhibitory’ region located at the C- terminus between residues 333-343. Unexpectedly, the N-terminal amino acids 11-20 were also required for its rapid lipid binding ability. There was a significant decrease in lipid clearance in the ∆1-20,333-343 (similar to WT) as compared to the ∆333-343 mutant alone. This was not anticipated because removing the N-terminal amino acids from 1-39 had no effect compared to WT. This data can be interpreted one of two ways.
One is that the amino acids from 11-20 are in fact responsible for its ability to bind lipid and our assay is not sensitive enough to distinguish between the ∆1-39 mutant and the
WT. Another possible explanation is that an interaction may be occurring between amino acids 11-20 and a third region in the protein that causes the lipid binding site to be exposed or put in favorable conformation for lipid binding. This data suggests the
‘inhibitory’ region is interacting with the N-terminus and this interaction does not allow the conformational change in the lipid binding site. The region that is actually responsible for the lipid clearance is probably found between amino acids 62-248.
175 Further evidence for the interaction between the N- and C-terminus was provided by the
fluorescence quenching experiments and native gel electrophoresis.
The presence of the ‘inhibitory’ sequence could act as a switch-type mechanism
for apoA-IV. The C-terminus in the WT conformation decreases the ability of the protein
to interact with lipid. However, in vivo, many different conditions may exist that could
cause a conformational change that would free the N-terminus. One of these conditions
could be lipid content of a cell or possibly lipoprotein levels in the plasma. Another possibility could be that the ‘inhibitory’ region may interact with another protein that would in turn free the N-terminus as well. This switch would be important because it would allow the protein to perform other functions when the demands for its lipid binding abilities are low.
Future studies for Aim 1: Future studies will be performed to better understand the tertiary structure of apoA-IV and also the lipid binding region of the protein. As stated earlier in Chapter 3, X-ray crystallography and other techniques that show a protein’s structure have not been possible with apoA-IV. Instead, the structure of the protein will be investigated by performing crosslinking studies with the monomer and dimer of the lipid-free and lipid-bound protein. The amino end of the protein and then lysines throughout the protein will be crosslinked with BS3. Only lysines that are within the
length of the linker arm will be crosslinked. Then the crosslinked protein will be digested
with trypsin (an endopeptidase) and then analyzed using mass spectrometric techniques
similar to what has been done with apoA-I (1;2). The peptides can then lead to a further
understanding of the structure of the protein because the lysines must be in close
176 proximity in order to be crosslinked. These studies will likely result in a molecular model of the protein.
Site-directed mutagenesis will be performed in order to further characterize the
‘inhibitory’ region in apoA-IV. These studies are important because point mutations should result in more subtle changes to the protein. A series of point mutants will be made to determine which amino acids from 333-343 are responsible for the ‘inhibitory’ effect. Once these amino acids are identified, the same mutations will be made in mouse apoA-IV. Finally, apoA-IV KO mice will be made transgenic with WT mouse apoA-IV or with the fast lipid binding mutant. These two groups of animals will then be compared to determine what effects the fast lipid binding mutant has on lipid clearance and other functions. These types of studies may yield much new information on the function of apoA-IV that might not be possible from studies that simply knock out apoA-IV. There are many examples where mouse physiology has compensated for the complete removal of a protein from the genome. However, using protein engineering to specifically target a particular function of an existing protein may allow one to dissect out the various roles of the protein in a model where the protein is fully expressed and nearly intact.
Aim 2: The role of apoA-IV in food intake has been studied for approximately 15 years
(3). However, the region of the protein involved in this role has not been established. In order to elucidate this region, several C-terminal deletion mutants were made in rat apoA-
IV. The mutants were injected into rat brains and their food intake was determined.
From this initial series of mutants, it was found that the N-terminal 116 amino acids were responsible for its role in food intake inhibition. Another mutant was made that removed
177 the N-terminal 61 amino acids. This mutant did not inhibit food intake as compared to
WT suggesting the region involved in food intake is located in the first 61 amino acids.
In separate studies, a 14 amino acid peptide from 17-30 also inhibited food intake in rats
suggesting it is the region of importance. It was also shown that full-length apoA-IV
does not have the ability to cross the blood-brain barrier suggesting its food intake effects
must be from nerve stimulation.
Discussion Aim 2: These studies were important because they have localized the region
of apoA-IV that is responsible for inhibiting food intake. Deletion mutagenesis showed
that the N-terminal 61 amino acids were responsible for this effect. In separate studies, a
14 amino acid peptide from residue 17-30 was found to have this inhibitory effect as well. This 14-mer could be used as a possible drug to target the hypothalamus in order to inhibit food intake. This aim also showed that full length apoA-IV did not cross the blood-brain barrier. Because of this, it is thought that apoA-IV may stimulate the vagus nerve which in turn could lead to hypothalamic production of apoA-IV. Another possibility is that an N-terminal piece of the protein may be cleaved and that portion could cross the blood-brain barrier. There has not been any information published on cleavage products of the protein under normal conditions. Recently, an N-terminal cleavage product of apoA-IV (approximately 70 residues) was found in an amyloid deposit in the heart and lungs of a human patient (4;5). Even though it was found in
pathologic state, this is direct evidence that it is possible for cleavage of the protein to
occur in vivo.
178 Future studies for Aim 2: Future studies will be to further elucidate the mechanism
behind apoA-IV regulating food intake. The first thing that needs to be determined is
whether a receptor exists for apoA-IV. It has been shown that apoA-IV has the ability to
bind to a variety of cell surfaces such as hepatocytes (6). Therefore, it is possible that
apoA-IV does in fact have a receptor. An apoA-IV receptor would be a potential drug
target for inhibiting food intake. His-tagged apoA-IV can be incubated with cellular
proteins and then it should bind to its receptor. Then the tagged protein can be purified
on an affinity column and the receptor should be purified along with it. The receptor can
then be identified using mass spectrometry. Other studies will be performed with the 14-
mer such that higher doses of the peptide will be injected intravenously to see whether
the inhibitory effect exists if it is injected peripherally. If so, the peripheral peptide
injections will be performed on rodents that have had vagotomies to determine whether
food intake is still inhibited.
Final thoughts: As discussed in this thesis, it is clear that apoA-IV has many functions.
These functions include interacting with lipids, inhibiting food intake and also acting as
an anti-inflammatory agent. Atherosclerosis and obesity are two of the largest killers in
industrialized countries. However, apoA-IV may be protective against these deadly
diseases. It is important to note the region of the N-terminus required for rapid lipid binding (amino acids 11-20) overlaps a portion of the region involved in food intake inhibition (amino acids 17-30). Therefore, it is possible that a peptide is freed by proteolysis (such as the one seen in the amyloid plaque) allowing the rest of the protein to perform some function in lipoprotein metabolism, perhaps in reverse cholesterol transport
179 (refer to Chapters 3 and 4) while the peptide crosses the BBB to inhibit food intake. All these functions could occur after the intact form of apoA-IV executes its roles in chylomicron metabolism. Many different possibilities exist in order to elucidate the potentially important physiological functions and their relationships with each other. For example, it is important to find cleavage products of apoA-IV in normal plasma and determine whether they function as predicted from in vitro experiments. Finally, in the next several years, apoA-IV will be studied more thoroughly and will become one of the most fascinating proteins in scientific history.
180 REFERENCES
1. Davidson,W.S. and Hilliard,G.M. (2003) The spatial organization of apolipoprotein A-I on the edge of discoidal high density lipoprotein particles: A mass spectrometry study, J Biol Chem 278, 27199- 27207.
2. Silva,R.A., Hilliard,G.M., Fang,J., Macha,S., and Davidson,W.S. (2005) A Three-Dimensional Molecular Model of Lipid-Free Apolipoprotein A-I Determined by Cross-Linking/Mass Spectrometry and Sequence Threading, Biochemistry 44, 2759-2769.
3. Fujimoto,K., Cardelli,J.A., and Tso,P. (1992) Increased apolipoprotein A-IV in rat mesenteric lymph after lipid meal acts as a physiological signal for satiation, Am. J. Physiol 262, G1002-G1006.
4. Bergstrom,J., Murphy,C., Eulitz,M., Weiss,D.T., Westermark,G.T., Solomon,A., and Westermark,P. (2001) Codeposition of apolipoprotein A-IV and transthyretin in senile systemic (ATTR) amyloidosis, Biochem. Biophys. Res. Commun. 285, 903-908.
5. Bergstrom,J., Murphy,C.L., Weiss,D.T., Solomon,A., Sletten,K., Hellman,U., and Westermark,P. (2004) Two different types of amyloid deposits--apolipoprotein A-IV and transthyretin--in a patient with systemic amyloidosis, Lab Invest 84, 981-988.
6. Weinberg,R.B. and Patton,C.S. (1990) Binding of human apolipoprotein A-IV to human hepatocellular plasma membranes, Biochim. Biophys. Acta 1044, 255-261.
181