Advertisement
JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

J. Biol. Chem., Vol. 280, Issue 6, 4154-4165, February 11, 2005
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Wang, L.
Right arrow Articles by Small, D. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wang, L.
Right arrow Articles by Small, D. M.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

The Interfacial Properties of ApoA-I and an Amphipathic {alpha}-Helix Consensus Peptide of Exchangeable Apolipoproteins at the Triolein/Water Interface*

Libo Wang, David Atkinson, and Donald M. Small{ddagger}

From the Department of Physiology and Biophysics, Boston University School of Medicine, Boston, Massachusetts 02118

Received for publication, October 12, 2004 , and in revised form, November 12, 2004.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Apolipoprotein A-I (apoA-I) is the major protein in high density lipoprotein (HDL). During lipid metabolism, apoA-I moves among HDL and triacylglycerol-rich lipoproteins. The main structure and the major lipid binding motif of apoA-I is the amphipathic {alpha}-helix. To understand how apoA-I behaves at hydrophobic lipoprotein interfaces, the interfacial properties of apoA-I and an amphipathic {alpha}-helical consensus sequence peptide (CSP) were studied at the triolein/water (TO/W) interface. CSP ((PLAEELRARLRAQLEELRERLG)2-NH2) contains two 22-residue tandem repeat sequences that form amphipathic {alpha}-helices modeling the central part of apoA-I. ApoA-I or CSP added into the aqueous phase surrounding a triolein drop lowered the interfacial tension ({gamma}) of TO/W in a concentration- and time-dependent fashion. The {gamma}TO/W was lowered ~16 millinewtons (mN)/m by apoA-I at 1.4 x 10–6 M and ~15 mN/m by CSP at 2.6 x 10–6 M. At equilibrium {gamma}, both apoA-I and CSP desorbed from the interface when compressed and readsorbed when expanded. The maximum surface pressure CSP could withstand without being ejected ({Pi}MAX) was 16 mN/m. The {Pi}MAX of apoA-I was only 14.8 mN/m, but re-adsorption kinetics suggested that only part of the apoA-I desorbed at {Pi} between 14.8 and 19 mN/m. However, above ~19 mN/m ({Pi}OFF) the entire apoA-I molecule desorbed into the water. ApoA-I was more flexible at the TO/W interface than CSP and showed more elasticity at oscillation periods 4–128 s even at high compression, whereas CSP was elastic only at faster periods (4 and 8 s) and moderate compression. Flexibility and surface pressure-mediated desorption and re-adsorption of apoA-I probably provides lipoprotein stability during metabolic-remodeling reactions in plasma.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Apolipoprotein A-I (apoA-I)1 is the major protein of high density lipoprotein (HDL) (1). It promotes cholesterol efflux from tissues to the liver for excretion and is a cofactor for lecithin:cholesterol acyltransferase (LCAT), which is responsible for the formation of most plasma cholesterol esters. ApoA-I exchanges off HDL onto other lipoproteins like very low density lipoprotein (VLDL) and moves back and forth among lipoprotein classes in the course of its metabolism. Thus the conformational adaptability of apoA-I to the polar environment in plasma and the apolar environment on lipoprotein particles is essential for their function. The predominant secondary structure feature of apoA-I is the 11/22-mer tandem repeat amino acid segments with high propensity to form amphipathic {alpha}-helices (1–3). Many of the amphipathic {alpha}-helices in apoA-I are class A amphipathic {alpha}-helices, which have a large (30–50%) apolar face with the positive residues distributed in the polar-nonpolar interface and are highly suited to lipid binding. This structure is thought to serve as the lipid association motif of apoA-I and to stabilize the lipoprotein surface.

ApoA-I exists in at least three states during metabolism: lipid-poor or free in solution, bound to a discoidal nascent HDL assembled with phospholipids and free cholesterol in a bilayer, or at the surface of a spherical lipoprotein particle (HDL or VLDL) with an apolar core of cholesterol esters and triglycerides, and a surface of phospholipids and free cholesterol (4). It is believed that apoA-I interacts primarily with the phospholipid hydrocarbon chains on discoidal HDL, and it is possible that it also interacts with the hydrophobic core of HDL particles and with the hydrophobic core of VLDL. Intensive studies (1, 5–11) have been carried out on how apoA-I interacts with phospholipids and combinations of phospholipids together with hydrophobic core lipids (triglycerides and cholesterol esters). ApoA-I can spontaneously penetrate and solubilize multilamellar liposomes of dimyristoyl phosphatidylcholine (DMPC) to form discrete small bilayered discoidal lipoproteins (5–8). Sonication or chaotropic agents are required to form quasi-spherical HDL-like particles with apoA-I, phospholipid, and cholesterol ester (9–11). Deletion mutations of apoA-I (12–14), synthetic peptides encompassing specific regions in apoA-I (15), and "idealized" designed synthesized model peptides of apoA-I (16) have been used in studies to estimate the lipid binding function of different regions of apoA-I. It is generally agreed that the C-terminal region of apoA-I (185–243 amino acids) has the highest affinity for lipid and plays a critical role in initiating the lipid binding. The N-terminal amphipathic region (residues 44–65) also has a higher affinity for lipids (17), whereas the middle region of the protein has a significant but lower affinity (15).

To understand how apolipoproteins behave at the lipoprotein surface, extensive studies have been carried out on apolipoproteins interacting with phospholipid monolayers spread at the air/water (A/W) interface (15, 18–26). As one could expect, apolipoproteins (apoA-I, apoA-II, apoC-I, apoC-II, apoA-IV, and apoE) are all surface-active and lower the surface energy. Exclusion pressures ({Pi}e) at which the apoproteins could no longer penetrate into the phospholipids monolayer have been measured. For instance, apoA-I has an {Pi}e of ~33 mN/m (18) at the egg phosphatidylcholine (PC) A/W interface, whereas the {Pi}e of apoA-II is 34 mN/m and that of apoA-IV is 29 mN/m at the egg PC A/W interface. The {Pi}e of synthesized 8 tandem-repeating 22-mer domains of apoA-I at egg PC A/W interface has also been measured (15). Among the 22-mers, the N- and C-terminal peptides (44–65 amino acids and 220–241 amino acids) exhibited the highest {Pi}e (28 mN/m and 30 mN/m), whereas other central peptides exhibited lower {Pi}e values (<23 mN/m). A number of consensus sequences of amphipathic {alpha}-helices modeling different sequences of apolipoproteins (7, 8) have been investigated using the {Pi}e technique as well. Some of these peptides, as short as 18 amino acids, have high {Pi}e in phospholipid monolayers, and others exhibit lower {Pi}e depending upon the specific amino acid arrangement of the amphipathic helical segment. For instance, upon substituting phenylalanine (Phe) for other hydrophobic residues in an A-type 18-residue {alpha}-helix (18A), {Pi}e rose from 30 mN/m for 18A-1F to 46 mN/m for 18A-6F (27).

To date only a few studies (26, 28) have concerned the surface behavior of apolipoproteins or their consensus peptides on oil/water interfaces. In a previous report (28), we have compared the interfacial properties of a synthesized consensus sequence peptide (CSP) of exchangeable apolipoproteins at the dodecane/water (DD/W) and the A/W interface. CSP consists of two 22-residue tandem repeat sequences derived from apoA-I, apoA-IV, and apoE-3. It is expected to comprise two antiparallel 20-residue amphipathic {alpha}-helices linked by a 4-residue proline-containing turn and is an idealized fundamental structural motif of exchangeable apolipoproteins (16, 29). Our results indicated that CSP binds strongly to the DD/W interface and the A/W interface to lower the interfacial tension ({gamma}) and the free energy (28). CSP binds more tightly to the DD/W interface than the A/W interface, but it lies flat on both interfaces at saturation with an average area per residue of 14–16 Å2. It can be compressed 6–12% while remaining on the surface, but it is ejected from the surface above a critical surface pressure ({Pi}MAX). {Pi}MAX on the A/W interface is 20.3 mN/m, but more than 10 mN/m higher on DD/W (31.7 mN/m). We suggest that surface pressure-mediated desorption and re-adsorption of amphipathic {alpha}-helices provide lipoprotein stability during remodeling reactions in plasma.

We have also studied the interfacial properties of two amphipathic {beta} strand (A{beta}S) consensus peptides of apoB (P27 and P12) on DD/W and triolein/water (TO/W) interfaces (30). Unlike CSP peptide, A{beta}S peptides show stronger binding to these oil/water interfaces, and the bound peptides are difficult to force off the surface under compression. Furthermore, A{beta}S peptides are almost purely elastic on DD/W, TO/W, and A/W interfaces when oscillated at periods from 4 to 128 s and from 6 to 25% deformation.

In this report, we examined the interfacial properties of native apoA-I and CSP at a more physiological TO/W interface. We measured and compared the adsorption isotherms, desorption behaviors, and the elasticity properties of both the protein and the peptide. Based on our results, we propose a probable mechanism for how apoA-I behaves at a hydrophobic interface, partly detaching at lower surface pressure and fully detaching at high pressures. Such information will be helpful to understand the structure of apolipoproteins on lipoprotein surfaces and the structural changes that occur during remodeling of lipoproteins in plasma.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Materials
The CSP peptide ((PLAEELRARLRAQLEELRERLG)2-NH2), which contains two 22-amino acid tandem repeats derived from the consensus sequence of the tandem repeats of human apoA-I, apoA-IV, and apoE (28, 29), was synthesized at Quality Controlled Biochemical Inc., using a Biosearch9050 Plus Continuous Flow Synthesizer, and purified >97+%. Stock solution of CSP (9–10 mg/ml) was prepared in ultra-filtered water obtained from a Hydro (Research Triangle Park, NC) Picosystem. The peptide has been shown to exhibit ~90% {alpha}-helical conformation in solution (16, 28, 29) and to interact with DMPC to form well defined phospholipid bilayer discoidal complexes. To measure the interfacial tension of the TO/W interface with CSP, varied amounts of peptide stocks were added to the aqueous phase to obtain different peptide concentrations (from 7 x 10–9 M to 5 x 10–6 M). The pH of the aqueous phase was kept at pH 7.4 with phosphate buffer (2 mM). Triolein (>99% pure) was purchased from NU-CHEK PREP, Inc. (Elysian, MN), and its interfacial tension was 32 mN/m. All other reagents were of analytical grade.

Source and Refolding of Human ApoA-I
Human apoA-I was isolated and purified as described previously (31). Unfolded human plasma apoA-I (0.1 mg/ml) was dialyzed against 6 M guanidine in the cold room overnight, then dialyzed against 6 M urea, 4 M guanidine, and 2 M guanidine overnight separately, and then dialyzed against pH 7.4 phosphate buffer (0.02 M) for 2 days. An Amicon was used to concentrate the apoA-I solution. Its purity was assessed by running a SDS-PAGE mini-gel; its folded state was checked by CD spectroscopy, and the apoA-I concentration was measured by protein assay. For interfacial tension measurements on the TO/W interface, varied amounts of apoA-I stocks were added to the aqueous phase to obtain different protein concentrations (from 1.7 x 10–8 M to 1.4 x 10–6 M). The pH of the aqueous phase was kept at pH 7.4 with phosphate buffer (2 mM).

Interfacial Tension ({gamma}) Measurement
The interfacial tension of the TO/W interface in the presence of different amounts of CSP or apoA-I in the aqueous phase was measured with an I. T. CONCEPT (Longessaigne, France) Tracker oil-drop tensiometer (32). 8-µl triolein drops were formed in gently stirred pH 7.4 phosphate buffer (6.0 ml) containing a given amount of apoA-I or CSP peptide. The interfacial tension was recorded continuously until it approached an equilibrium. The surface pressure {Pi} was obtained from {gamma} of the interface without CSP or apoA-I ({gamma}0) minus the surface tension of the interface with CSP or apoA-I ({gamma}), i.e. {Pi} = {gamma}0 – {gamma}. All experiments were carried out at 25 ± 0.1 °C in a thermostated system.

Compression and Expansion of the Interfaces
Once the interfacial tension curve approached an equilibrium, the oil drop (8 µl) was compressed by rapidly decreasing the volume by about 25% (2 µl), 12% (1 µl), or 6% (0.5 µl). In some experiments with apoA-I, larger compression was achieved by decreasing the oil drop volume by up to 5 µl (62%). The sudden decrease in volume instantaneously decreased the drop surface area and resulted in a sudden compression causing the tension to decrease abruptly. The oil drop was held at this reduced volume for about 5 or 10 min, and the surface tension was recorded continuously. If peptide or protein molecules readily desorbed from the surface, the surface tension rose back toward the equilibrium value, and this is called the desorption curve. If peptide or protein molecules stayed on the interface, {gamma} would not change. To expand the interface, the decreased volume of the oil drop was rapidly increased by about 25% (2 µl), 12% (1 µl), or 6% (0.5 µl) back to its initial volume (8 µl), respectively. In cases following larger compression of apoA-I, the decreased volume of the oil drop was increased by up to 5 µl back to its initial volume (8 µl). As a result the surface area increased, and the tension abruptly increased. If molecules adsorbed from the bulk phase to adhere to the newly formed extra surface, then the surface tension would drop back toward the equilibrium, and this is called the readsorption curve. The 25% desorption and adsorption curves were fitted by a bi-exponential equation: y = y0 + A1exp(–x/t1) + A2exp(–x/t2). The time constant t1 and t2 are related to desorption and adsorption processes (28).

Value of [Pokoj]MAX
To estimate the maximum pressure ({Pi}MAX) that the peptide or protein molecule could withstand without being ejected from the surface, a series of experiments were carried out in which the oil drop volume was decreased abruptly to increase the surface pressure (i.e. decrease the {gamma}) to a given value, {Pi}0, and then the interfacial tension {gamma} was followed for 3–10 min. The change in tension ({Delta}{gamma}) was then plotted against {Pi}0. A positive {Delta}{gamma} indicated that the bound peptide or protein molecules had desorbed from the surface. If there was no change in {gamma}, it indicated that the peptide or protein molecule on the surface remained there without desorption. In addition, a negative {Delta}{gamma} indicated that the peptide or protein molecules in the bulk solution were still able to adsorb onto the surface; that is, that {Pi}0 was not quite at equilibrium. The plot of the {Pi}0 versus {Delta}{gamma} was fitted to a straight line. The intercept at {Delta}{gamma} = 0 gave the {Pi} at which peptide or protein molecules showed no net adsorption or desorption, this is the so-called {Pi}MAX.

Value of [Pokoj]OFF of ApoA-I
Because apoA-I consists of 10 putative tandem-repeating amphipathic {alpha}-helical domains, it may partially desorb from the interface under compression. {Pi}MAX of apoA-I gives the estimated pressure at which bound apoA-I starts to be pushed off the interface. To estimate the pressure at which the whole molecule of apoA-I is ejected from the interface, we used the same method used to estimate {Pi}MAX except that we used larger (>50%) compressions and re-expansions of the interface. We then compared the re-adsorption curve (after the desorption process) with the initial adsorption curve and identified all cases where the curves were almost identical under the same conditions. When the re-adsorption curve was identical to the initial adsorption curve (see Fig. 6B for example) in the same interfacial tension range, it indicated that the compression used had displaced the entire molecule of apoA-I into the aqueous phase. For these cases we plotted {Delta}{gamma} against {Pi}0 and fitted the data to a straight line. The intercept at {Delta}{gamma} = 0 gave the {Pi} at which the whole apoA-I molecule started to be ejected from the interface (see "Results" for detail), which we call {Pi}OFF. When the compression was less and gave rise to a lower {Pi}0, the re-adsorption curve was much faster than the initial adsorption curve in the same {gamma} range (see Fig. 6A for example). This indicated that only part of apoA-I was displaced from the surface and could re-adsorb very rapidly when the surface was expanded. The data from those cases formed a separate line with a different slope (Fig. 7), where {Pi}MAX was the lowest {Pi} needed to displace part of apoA-I from the interface.



View larger version (26K):
[in this window]
[in a new window]
  		FIG. 6.
Comparison of "re-adsorption" curves after compression and re-expansion with original adsorption curves of apoA-I from aqueous solution. A, at smaller compression and expansion of the interface (25%), an 8-µl triolein drop was decreased by 2 µl, the decreased volume was held for several minutes then increased by 2 µl. The re-adsorption curve falls much faster than the original adsorption curve, which indicates that under smaller compression (25%), only part of apoA-I molecule was pushed off the interface. B, at larger compression and expansion of the interface (62%), an 8-µl triolein drop was decreased by 5 µl, and the decreased volume was held for several minutes then increased by 5 µl. The re-adsorption curve is identical to the original adsorption curve, which indicates that under higher compression the whole apoA-I molecule was pushed off the interface. The curves of adsorption and re-adsorption superimpose in B but not A. The concentration of apoA-I in bulk solution was 2.1 x 10–7 M.

 



View larger version (18K):
[in this window]
[in a new window]
  		FIG. 7.
The estimation of {Pi}MAX and {Pi}OFF of apoA-I at the TO/W interface. Changes of {gamma} ({Delta}{gamma}) during compression plotted against the instant pressure ({Pi}0) generated right after compression. Open circles (fast re-adsorption): only part of apoA-I was pushed off the interface. The intercept of the fitted line at {gamma} = 0 gives the value of {Pi}MAX of apoA-I; crosses (slow re-adsorption): whole apoA-I was pushed off the interface and had to re-adsorb from solution. The intercept of the fitted line at {gamma} = 0 gives the threshold value of {Pi}OFF of apoA-I. Therefore, when compressed to pressures above 14.8 mN/m, part of apoA-I was pushed off the interface, whereas only at pressures above 18.9 mN/m did whole apoA-I begin to be ejected from the interface.

 
Oscillation of the Interface and the Elasticity Analysis
Equilibrium Oscillations—When the tension approached an equilibrium value, the volume of the triolein drop was sinusoidally oscillated at different periods (frequencies) ranging from 4 to 128 s (0.25–0.008Hz) and different amplitudes of about ±25%, ±12% or ±6%. As the volume V was oscillated in a sinusoidal fashion, interfacial area A and surface tension {gamma} were recorded continuously and the phase angle {phi} between compression and expansion computed. The interfacial elasticity modulus {epsilon} was derived ({epsilon} = d{gamma}/d ln A). The elasticity real part {epsilon}' and the elasticity imaginary part {epsilon}'' were obtained ({epsilon}'= |{epsilon}| cos {phi}, {epsilon}''= |{epsilon}| sin {phi}) (33, 34).

Continuous Oscillation—Continuous oscillations were carried out 10 s after an 8-µl oil drop was formed in CSP or apoA-I solution. Allowing 10 s to form the drop allowed the system to stabilize. Then the volume was oscillated in separate experiments at different periods (4, 8, and 16 s) and different amplitude of about ±25%, ±12%, or ±6% continuously until the mean interfacial tension reached an equilibrium. This gave a continuous measurement of the mean surface tension (and surface pressure), {epsilon}, {epsilon}', {epsilon}'', and {phi}.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
Adsorption of CSP on Triolein/Water (TO/W) Interface—Fig. 1 shows a typical set of interfacial tension ({gamma}) curves of the TO/W interface with different amounts of CSP in the aqueous phase. The {gamma} of TO/W interface was about 32 mN/m. When adding different amounts of CSP into the aqueous phase (7 x 10–9 M to 4.9 x 10–6 M), the CSP molecule adsorbed onto the TO/W interface and lowered {gamma} to approach an equilibrium value.



View larger version (20K):
[in this window]
[in a new window]
  		FIG. 1.
Examples of interfacial tension, {gamma}, against time curves of CSP at TO/W interface. An 8-µl triolein drop formed in 2 mM, pH 7.4 phosphate buffer with different amounts of CSP. Line a, 7 x 10–9 M; line b, 3 x 10–8 M; line c, 1.3 x 10–7 M; and line d, 2.6 x 10–6 M. All experiments were carried out at 25 ± 0.1 °C.

 
At the most dilute concentration (7 x 10–9 M, Fig. 1, curve a), {gamma} showed a very slow decrease over about 7000 s (lag period), then a sharp decrease, and then a very gradual decrease approaching an equilibrium value at about 19 mN/m, which generated 13-mN/m pressure on the surface. As the CSP concentration increased, the lag period shortened, and the steepness of the rapid fall of {gamma} increased. At the highest concentration (2.6 x 10–6 M) shown in Fig. 1, {gamma} did not show the lag period and started falling rapidly at a very early stage and reached 16.7 mN/m at 1 h, which generated 15.3-mN/m surface pressure. In our previous study (27), we have shown that CSP can generate up to ~31-mN/m pressure on the DD/W interface and up to ~24.5-mN/m pressure on the A/W interface. CSP generated less pressure at the TO/W interface than at DD/W and A/W interfaces, because the starting {gamma} of TO/W (32 mN/m) is much lower than DD/W (52 mN/m) or A/W (72 mN/m).

At lower concentrations, the {gamma}-time curves (Fig. 1) showed a discontinuity in the lag period. For example, at 7 x 10–9 M, it started at about 5000 s and resulted in a retardation of the fall in {gamma} with time. We suggest this might be related to a rearrangement in the triolein surface in the presence of CSP, because we did not see this discontinuity in the {gamma}-time curves of CSP adsorbed onto DD/W or A/W interfaces (28), nor in the apoA-I {gamma}-time curves (see below).

Adsorption of ApoA-I on the TO/W Interface—Fig. 2 shows examples of {gamma}-time curves of apoA-I as it adsorbed onto the TO/W interface with varied amounts of apoA-I in the bulk solution. Without apoA-I, the TO/W interface showed a constant tension at ~32 mN/m. While adding different amounts of apoA-I into the bulk solution, apoA-I adsorbed onto the interface, lowering the tension toward an equilibrium value. At the lowest concentration (1.7 x 10–8 M) {gamma} fell to 18.6 mN/m, while at the highest concentration studied (1.4 x 10–6 M), {gamma} fell to about 15.6 mN/m, which generated about 16.4 mN/m surface pressure. The tension curves of apoA-I also showed three regions: first a lag period where {gamma} fell slowly with time; second, a much faster fall in {gamma}; and third, a gradually decreasing {gamma} moving toward an equilibrium level. As the apoA-I concentration increased, the lag period shortened and the steepness of the second region increased. At the highest concentration (7.1 x 10–7 M) shown in Fig. 2, {gamma} did not show the lag period and started falling very early.



View larger version (24K):
[in this window]
[in a new window]
  		FIG. 2.
Examples of interfacial tension, {gamma}, against time curves of apoA-I at the TO/W interface. An 8-µl triolein drop formed in 2 mM, pH 7.4 phosphate buffer with different amounts of apoA-I. Line a, no apoA-I; line b, 1.7 x 10–8 M; line c, 3.5 x 10–8 M; line d, 8.9 x 10–8 M; line e, 1.8 x 10–7 M; and line f, 7.1 x 10–7 M. All experiments were carried out at 25 ± 0.1 °C.

 
Desorption of CSP from TO/W Interface When the Interface Was Compressed—After CSP lowered the {gamma} to an equilibrium level, we studied desorption and re-adsorption of CSP by compressing and expanding the TO/W interface.

Fig. 3A shows an example of the desorption and the readsorption of CSP under compression and expansion by 25%, 12, and 6% in volume, V (8 µl) at 5.9 x 10–7 M CSP in bulk solution. For ~25% compression (the first volume decrease in Fig. 3A, –2 µl), when {gamma} of the TO/W interface approached an equilibrium value (18.1 mN/m), the V (8.1 µl) was suddenly decreased to 6.2 µl. Thus the surface area (A) was decreased, and the surface concentration ({Gamma}) of CSP rose accordingly, decreasing {gamma} to 9.9 mN/m ({Pi} = 21.1 mN/m). The decreased V (6.2 µl) was held constant for about 6 min. During this period, {gamma} rose gradually back to a value of 11.4 mN/m. The change of {gamma} for the 6 min was 1.5 mN/m, and this indicated that some bound CSP was pushed off the interface by the compression. The decreased V was then expanded from 6.2 µl back to 8.1 µl, {gamma} immediately rose to 28.2 mN/m. The fact that {gamma} rose about 10 mN/m above the starting equilibrium {gamma} before compression (18.1 mN/m) clearly showed that compression to 9.9 mN/m had pushed some CSP off the surface, i.e. decreased the surface concentration, {Gamma}. In the following few minutes after expansion, {gamma} fell back to an equilibrium value (18.9 mN/m), which indicated that CSP in the bulk solution re-adsorbed to the interface. For 12 and 6% compression (the second and third volume decreases), the changes of {gamma} were 1.4 and 0.9 mN/m, respectively, whereas for 12 and 6% expansion (the second and third volume increases), re-adsorption of CSP occurred, and the values of {gamma} fell from 25.0 to 19.2 mN/m and from 19.8 to 18.3 mN/m, respectively.



View larger version (19K):
[in this window]
[in a new window]
  		FIG. 3.
Examples of desorption and re-adsorption curves of CSP (A) and apoA-I (B) on the TO/W interface. The interface was compressed from 8 to 6 µl (~25%), and then after a few minutes re-expanded to 8 µl. The same procedure was done with 12 and 6% compression, respectively. A, [CSP] = 5.9 x 10–7 M; B, [apoA-I] = 3.5 x 10–8 M.

 
To find the maximum surface pressure ({Pi}MAX) at which CSP could remain on the interface without being pushed off the interface, we did a series of compression and expansion experiments as shown in Fig. 3 under varied conditions, i.e. different concentration of CSP, different compression or expansion ratios. As illustrated in Fig. 3, when the interface was compressed, {gamma} fell to a given value {gamma}0, which generated the pressure {Pi}0, and changed while the decreased volume was held for a period of time. We then plotted the changes in {gamma} ({Delta}{gamma}) versus {Pi}0 and fitted the data with a straight line (Fig. 4). The intercept of the line at {Delta}{gamma} = 0 gave the {Pi}MAX of CSP at TO/W interface. In our study, 65 measurements varying {Pi}0 from 16 mN/m to 23 mN/m gave {Pi}MAX = 16.0 mN/m for CSP at the TO/W interface. We have shown (28) that {Pi}MAX of CSP is 31.7 mN/m at DD/W interface and is 21.3 mN/m at A/W interfaces. That {Pi}MAX for CSP at TO/W interface is smaller than that at DD/W and A/W interfaces might suggest that CSP has lower affinity for the TO/W interface. However, CSP lowered the interfacial {gamma} to a lower level (~16 mN/m) than it did at the other interfaces (~46.5 mN/m at A/W and ~20 mN/m at DD/W), which suggests that the surface free energy is the lowest at the CSP-saturated TO/W interface.



View larger version (16K):
[in this window]
[in a new window]
  		FIG. 4.
The {Pi}MAX of CSP at the TO/W interface. The interface was suddenly compressed to pressure {Pi}0, and the change in interfacial tension ({Delta}{gamma}) followed for several minutes. {Delta}{gamma} was plotted against {Pi}0. Linear regression was made to the plot, r = 0.84975, p = 1.136 x 10–23. The intercept of the fit line at {gamma} = 0 gives the {Pi}, at which peptide or protein molecule shows no net adsorption or desorption; this is the so-called {Pi}MAX. The value of {Pi}MAX of CSP at TO/W is 16 mN/m.

 
Behavior of ApoA-I on the TO/W Interface When the Interface Was Compressed and Re-expanded—Fig. 3B shows an example of desorption and re-adsorption of apoA-I under compression and expansion at 3.5 x 10–8 M apoA-I in bulk solution. The interface was compressed by decreasing the drop volume (8 µl) by 25% (–2 µl), 12% (–1 µl), or 6% (–0.5 µl), respectively. The decreased volume was held for several minutes and then the surface was re-expanded to the original volume (8 µl). ApoA-I showed desorption phenomena similar to CSP. Note that at 6% compression (the third volume decrease), apoA-I showed very little desorption with changes of {gamma} ~ 0.2 mN/m, whereas at 6% expansion (the third volume increase), apoA-I also showed very little re-adsorption with {gamma} falling from 18.8 to 18.0 mN/m. Therefore, around equilibrium pressure a 6% compression caused very little desorption.

Compared with CSP, apoA-I is much larger and has 10 tandem-repeating amphipathic {alpha}-helix domains. Furthermore, the differences in the behavior of the domains on a phospholipid/water monolayer surface showed that the N and C termini have greater exclusion pressures {Pi}e than the central part of apoA-I (15). We noted that, during moderate amounts of compression, the {gamma}-time curve on the expansion appeared to be much more rapid than the {gamma}-time segment of the same initial {gamma}-time curve of the same experiment (Fig. 6A). We speculated that perhaps the protein was not coming completely off the surface but only partly detaching and then during expansion the part that was desorbed rapidly re-adsorbed. We thus speculated that we needed greater pressure to eject the whole apoA-I molecule off the interface. Therefore, in addition to the small compression and expansion (6–25%) of the interface (Fig. 3B), we increased compression and expansion to >60% of the interface. Fig. 5 shows an example in which we compressed and expanded the triolein drop (8 µl) by –/+2 µl (25%), –/+4 µl (50%, twice), and –/+5 µl (>60%, twice). We compared the re-adsorption {gamma}-time curves with the initial adsorption curves over the same range of {gamma} from the same experiment, we found that at small compression and expansion (6–25%), the readsorption curves showed a more rapid fall of {gamma} from the same {gamma} than the initial adsorption {gamma}-time curves. Examples of readsorption curves at 8–/+2 µl and 8–/+5 µl compression are shown in Fig. 6. After large (62%) compression, the re-adsorption {gamma}-time curve was almost identical with the initial adsorption curve (Fig. 6B), whereas after small (~25%) compression, the {gamma}-time re-adsorption curve fell rapidly. The initial adsorption curve showed the {gamma}-time changes as the whole apoA-I protein adsorbed onto the interface from the bulk solution. The re-adsorption curve showed the {gamma}-time changes when the interface was expanded after a compression process. During the compression process, bound apoA-I either partly or fully desorbed from the interface as shown in Figs. 3B and 5. When the interface was expanded after the desorption process, new space was generated. If only part of the bound apoA-I molecule was pushed off during the compression process, those ejected parts would rapidly re-adsorb to the interface when expanded. Because some part of apoA-I was still anchored on the interface, the re-adsorption of the ejected part should be much faster than the initial adsorption of the whole protein from the bulk solution. Rapid re-adsorption, i.e. fast {gamma}-time decreases occurring under small compressions are shown in Fig. 6A. On the other hand, if the whole apoA-I molecule was ejected from the interface during the compression process, then new apoA-I molecules re-adsorbed onto the interface from the bulk solution following the similar {gamma}-time changes as shown in the initial adsorption curve. This happened when compression was large (Fig. 6B).



View larger version (19K):
[in this window]
[in a new window]
  		FIG. 5.
Examples of larger amplitude compression followed by re-expansion of apoA-I at the TO/W interface. When an 8-µl triolein drop got saturated with apoA-I on the surface, the drop was rapidly compressed by 25% (–2 µl) to 6 µl, and the {gamma} followed for several minutes, then the drop was re-expanded to 8 µl (once). Similar processes were done to reduce the drop volume by 50% (–4 µl), then expand back to 8 µl (twice) and reduce the drop volume by 62% (–5 µl), and then expand back to 8 µl (twice). The concentration of apoA-I in bulk solution was 5.2 x 10–8 M.

 
Based on the two kinds of desorption, i.e. part-off (rapid re-adsorption) and all-off (slow re-adsorption) of apoA-I, we derived two critical surface pressures: {Pi}MAX, the pressure at which part of apoA-I started to detach, and {Pi}OFF, the pressure at which apoA-I was completely expelled. Fig. 7 shows the {Pi}MAX and {Pi}OFF values of apoA-I. 22 compressions of "part-off" desorption varying {Pi}0 from 14.5 to 22 mN/m gave {Pi}MAX = 14.8 mN/m for apoA-I at the TO/W interface. 21 compressions of "all-off" desorption varying {Pi}0 from 20.5 mN/m to 23 mN/m gave {Pi}OFF = 18.9 mN/m for apoA-I at the TO/W interface. Thus under small compression which generated lower surface pressure (14.8 mN/m ~ 18.9 mN/m), bound apoA-I could be partially pushed off the interface; while only under compression which generated higher surface pressure (>18.9 mN/m), could the whole apoA-I molecule be ejected from the interface. There was a zone of intermediate compressions giving rise to {Pi}0 from 18.9 to 21.5 mN/m in which the re-adsorption curve showed a more rapid {gamma}-time curve than that of the adsorption curve where we surmised that both part-off and all-off occurred.

The Kinetics of {gamma}-Time Curves following Compression and Expansion—To estimate the kinetics of the {gamma} changes in the {gamma}-time curves of CSP following compression and expansion, we fitted the 25% compression and expansion curves at varied CSP bulk concentration with a two exponential equation from which a rapid (t1) and slow (t2) time constants were derived. The time constants after compression are related to desorption, whereas those after expansion are related to re-adsorption. During desorption the rapid time constant t1 was plotted against the ln molar concentration of CSP in the bulk solution and showed that the desorption of CSP is not a concentration-dependent process (Fig. 8). The desorption t1 for all the concentrations are ~2 s. On the other hand, the adsorption process depends on CSP concentration, as t1 is faster at higher CSP concentration. Similar behavior was observed for CSP on DD/W and A/W interfaces (30). We studied the kinetics of 25% desorption and re-adsorption of apoA-I on the TO/W interface at three apoA-I concentrations, 5.2 x 10–8 M, 3.3 x 10–7 M, and 8.4 x 10–7 M (data not shown). The desorption t1 for apoA-I was rapid (~2s) like CSP and not concentration-dependent. The re-adsorption curve could not be adequately fitted to the bi-exponential equations, and no meaningful t1 data were obtained. These results show that, like CSP, the desorption of apoA-I is not concentration-dependent, which indicates that desorption is a strictly surface phenomenon.



View larger version (21K):
[in this window]
[in a new window]
  		FIG. 8.
Desorption and re-adsorption t1 for CSP at the TO/W interface plotted against the ln molar concentration of CSP in the aqueous phase. An 8-µl triolein drop was compressed and expanded by 25%. The number of experiments at each CSP concentration is given in parentheses. The t1 for desorption is constant at ~2 s, whereas the t1 for re-adsorption after expansion is concentration-dependent. The desorption t1 of apoA-I at three concentrations were also studied at compression of 25% and was ~2 s (data not shown).

 
Viscous and Elastic Properties of CSP and ApoA-I at the TO/W Interface—To investigate the elasticity of bound CSP at the TO/W interface, we carried out oscillations after {gamma} approached an equilibrium. The drop volume V was oscillated in a sinusoidal fashion, and the area (A) and the interfacial tension {gamma} were monitored. The phase angle {phi} between {gamma} and A, the elasticity modulus {epsilon}, the real part (elastic part) of {epsilon} ({epsilon}'), and the imaginary part (viscous part) of {epsilon} ({epsilon}'') were calculated accordingly (see "Materials and Methods").

At 5.9 x 10–7 M bulk CSP concentration, the {gamma} of the TO/W interface was allowed to reach an equilibrium {gamma} of about 16.5 mN/m, and then the volume of the oil drop was oscillated at different periods (4, 8, 16, 32, 64, and 128 s) and three amplitudes (8 ± 0.5, 8 ± 1, and 8 ± 2 µl). Table I gives the values of {phi}, {epsilon}, {epsilon}', and {epsilon}'' at each condition, and Fig. 9 shows the changes of {phi}, {epsilon}, and {epsilon}' plotted against the oscillation period.


View this table:
[in this window]
[in a new window]
  		TABLE I
Dynamic interfacial properties of CSP at TO/W interface

All oscillation experiments were carried out on the TO/W interface in pH 7.4 phosphate buffer (2 mM) at 25 ± 0.1 °C. The concentration of CSP in the aqueous phase was 5.9 x 10-7 M.

 



View larger version (23K):
[in this window]
[in a new window]
  		FIG. 9.
A, changes of phase angle ({phi}), and B, elasticity modulus ({epsilon}), and elasticity real part ({epsilon}') against the oscillation period of CSP at the TO/W interface. The concentration of CSP in aqueous phase was 5.9 x 10–7 M. Solid lines represent the oscillations at 8 ± 2 µl; dashed lines represent the oscillations at 8 ± 1 µl; dotted lines represent the oscillations at 8 ± 0.5 µl. Triangles, {phi}; filled circles, {epsilon}; and rectangles, {epsilon}'.

 
For 8 ± 0.5 µl oscillations (dotted lines in Fig. 9) at the 4-s period, {phi} was almost zero (Fig. 9A), and {epsilon} and {epsilon}' were virtually the same (Fig. 9B); at the 8-s period {phi} increased slightly but {epsilon} and {epsilon}' were very close. At 16 s and longer, {phi} rose from 13 to 25° and {epsilon}' gradually became less than {epsilon}, which indicated that there was more and more viscous component added into this system as the period lengthened. Similar results were shown at 8 ± 1 and 8 ± 2 µl oscillations (Fig. 9), but the effects were larger. Oscillations at bigger amplitude had a larger {phi}, which indicated a less elastic interface. Thus we conclude that the bigger the amplitude and the longer the period, the larger the {phi}, and the smaller the {epsilon} and {epsilon}', and the more viscous the interface. Fig. 10A shows examples of {Pi}-A plots of CSP at the TO/W interface at three different amplitudes and periods from 4 to 128 s. At 4 s, the hysteresis of the {Pi}-A plot was very small, whereas at longer periods, the divergence on compression and expansion was much bigger. Note that even when the compression was greater than {Pi}MAX, little hysteresis was present if the compression was fast (4 s).



View larger version (30K):
[in this window]
[in a new window]
  		FIG. 10.
A, surface pressure ({Pi}) versus area (A) plots of CSP at the TO/W interface. 8-µl triolein drops were oscillated at different amplitudes and periods. Dotted straight line shows the {Pi}MAX level. The concentration of CSP in aqueous phase was 5.9 x 10–7 M. B, surface pressure ({Pi}) versus area (A) plots of apoA-I at the TO/W interface. 8-µl triolein drops were oscillated at different amplitudes and periods. Dotted straight lines show the {Pi}MAX and {Pi}OFF levels. The concentration of apoA-I in aqueous phase was 6.1 x 10–7 M.

 
The viscous behavior at larger amplitude and longer period is probably due to the CSP exchanging between the interface and the bulk solution. We showed earlier (Figs. 3 and 4), when the CSP interface was compressed above {Pi}MAX, bound CSP desorbed from the interface and {gamma} rose; then we re-expanded the interface and CSP re-adsorbed from the aqueous phase back to the interface and {gamma} fell. The oscillations generated a higher {Pi} often more than {Pi}MAX (Fig. 10A). When compression generated a pressure greater than {Pi}MAX, bound CSP should be pushed off the interface during the compression. A larger amplitude created more vacant space, and thus more peptide re-adsorbed onto the interface. Enhanced desorption and adsorption at large amplitude indicates more exchange of the peptide between the interface and the bulk solution, which relaxes the stress more and makes the interface appear more viscous. However, the longer periods allowed the peptide enough time to desorb and then re-adsorb thus amplifying a more viscous surface. A 4-s period did not give the peptide enough time to desorb and re-adsorb effectively, and thus the surface appeared nearly elastic. Combining the effect of amplitude and period, oscillations at the smallest period (4 s) and the smallest amplitude (±0.5 µl) gave the smallest {phi} (3.2°) and the highest {epsilon} (150.9 mN/m), i.e. the most elastic interface. Oscillations at the largest amplitude (±2 µl) and the longest period (128 s) gave the biggest {phi} (50.9°) and the lowest {epsilon} (37.0 mN/m), i.e. the most viscous interface (Table I and Fig. 10A).

Systemic study of the equilibrium oscillations of apoA-I at the TO/W interface at 6.1 x 10–7 M bulk apoA-I concentration with different periods (4, 8, 16, 32, 64, and 128 s) and different amplitudes (8 ± 0.5, 8 ± 1, and 8 ± 2 µl) are listed in Table II, the changes of {phi}, {epsilon}, and {epsilon}' at different amplitudes were plotted against the period in Fig. 11, and the examples of {Pi}-A curves given in Fig. 10B. We saw similar trends as in the oscillations of CSP in that the smaller the period and amplitude the smaller the {phi} and the larger the {epsilon}. Except that the {phi} for all periods (4–128 s) were all relatively small with the largest {phi} being at 8 ± 2 µl and 128 s was 18.8°, which is much different from CSP. Furthermore, the difference between {epsilon} and {epsilon}' was so small that it was hard to distinguish (Fig. 11B). This suggests that apoA-I is more flexible at the TO/W interface than CSP and is quite elastic at all conditions. The {Pi}-A plots (Fig. 10B) of the 128-s period showed minor hysteresis with a small {phi} between the compression and the expansion. At fast oscillation (4 s) the {phi} was negligible. This indicates that apoA-I forms nearly elastic interfaces. Note that, at all oscillations, the highest {Pi} generated was greater than {Pi}MAX but barely reached {Pi}OFF (Fig. 10B), and this probably was a factor in maintaining elasticity.


View this table:
[in this window]
[in a new window]
  		TABLE II
Dynamic interfacial properties of apoA-1 at the TO/W interface

All oscillation experiments were carried out on the TO/W interface in pH 7.4 phosphate buffer (2 mM) at 25 ± 0.1 °C. The concentration of apoA-1 in the aqueous phase was 6.1 x 10-7 M.

 



View larger version (19K):
[in this window]
[in a new window]
  		FIG. 11.
A, changes of phase angle ({phi}), and B, elasticity modulus ({epsilon}), and elasticity real part ({epsilon}') against the oscillation period of apoA-I at the TO/W interface. The concentration of apoA-I in aqueous phase was 6.1 x 10–7 M. Solid lines represent the oscillations at 8 ± 2 µl; dashed lines represent the oscillations at 8 ± 1 µl; dotted lines represent the oscillations at 8 ± 0.5 µl. Triangles, {phi}; filled circles, {epsilon}; and rectangles, {epsilon}'.

 
We carried out continuous oscillations to study the changes of the elasticity of the interface as a function of {Pi}. Elastic modulus, {epsilon} versus {Pi} plots of CSP at the TO/W interface at different amplitudes, and periods are shown in Fig. 12. There were three states of the interface as the {Pi} rose. At low {Pi} up to about 4–5 mN/m, {epsilon} was low and changed little. When {Pi} > 4–5 mN/m, {epsilon} increased sharply and increased until it reached a pressure at 12.5–14.5 mN/m. The slopes (d{epsilon}/d{Pi}) of the 8 ± 2 oscillations vary greatly with the period, being 6 at 16 s oscillations and ~21 at 4 s oscillations. Finally, at high {Pi} (>13.5–15.5 mN/m) {epsilon} decreased. Similar behavior had been found on the DD/W interface but not on the A/W interface (28). This phenomenon is characteristic of many proteins at the oil/water interfaces (33, 34) and suggests marked changes in the behavior of the interface at high pressures.



View larger version (25K):
[in this window]
[in a new window]
  		FIG. 12.
Elasticity modulus ({epsilon}) against surface pressure ({Pi}) of CSP at the TO/W interface. The concentration of CSP in aqueous phase was 5.9 x 10–7 M. Data of oscillations at 8 ± 1 µl fell in between these curves (not shown).

 
Fig. 13 shows {epsilon}-{Pi} plots of apoA-I at the TO/W interface at different amplitudes and periods. At {Pi}> 2.5 mN/m, {epsilon} increased at a 5:1 {epsilon}/{Pi} ratio until the pressure reached ~9 mN/m on all curves. However, at smaller amplitude (8 ± 0.5 µl) {epsilon}/{Pi} stayed the same until the pressure reached ~14 mN/m then flattened. Although, at larger amplitude (8 ± 2 µl) and the slowest period (16 s), {epsilon} slightly decreased between 13 and 15 mN/m for {Pi}.



View larger version (26K):
[in this window]
[in a new window]
  		FIG. 13.
Elasticity modulus ({epsilon}) against surface pressure ({Pi}) of apoA-I at the TO/W interface. The concentration of apoA-I in aqueous phase was 1.8 x 10–7 M. Data of oscillations at 8 ± 1 µl fell in between these curves (not shown).

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
ApoA-I is the major apolipoprotein of HDL and a critical factor in the formation of nascent phospholipid/cholesterol HDL through interaction with ABCA1 and perhaps ABCG-1 (ATP-binding cassette (ABC) transporter G-1) and G-4 (35) transporters. It is a cofactor for the lecithin:cholesterol acyltransferase reaction to generate cholesterol ester and form spherical HDL and a ligand for the selective cholesterol ester transporter SRB1. It therefore plays a critical role in formation in the metabolism of HDL and in reverse cholesterol transport.

The lipid binding motif used by apoA-I, as well as other exchangeable apolipoproteins, is the amphipathic {alpha}-helix (2) formed by tandem repeated amino acid sequences of 11/22 residues. The current concept of the structure of lipid-free apoA-I is that a molten globule (36), with a relatively compact hydrophobic core (though with little specific tertiary structure) that is stabilized through interaction with the N and C termini (37). Upon lipid binding, the helical content of apoA-I increases from 40–50% to ~75% (40–43). The transition from lipid-free apoA-I to a lipid-bound form must involve a structural rearrangement of the helical segments. The C-terminal region is thought to initiate binding prior to the remainder of the protein binding in a two-step fashion (14). Lipid-free apoA-I can spontaneously form discoidal complexes with DMPC (5, 42–44) implying that apoA-I has a natural tendency to solubilize lipid. All the helical repeats in apoA-I may not have the same capacity to bind lipid, based upon the distribution of their charged side chains around the helical axis (45). Amphipathic {alpha}-helices in general are categorized into A (amphipathic), L (lytic), H (hormone), M (transmembrane), G (globular), and Y (having specific charge distribution), with the majority of apolipoprotein helixes being placed in the class A, characterized by the clustering of positively charged side chains at the polar/apolar interface, with negatively charged side chains being predominately located in the center of the polar face (3, 45). However, three helical repeats in apoA-I are not characterized as class A helices, being amino acids 8–33 in class G and amino acids 88–120 and 209–241 in class Y. The relevancy of these classifications to the lipid binding properties of the exchangeable apolipoprotein helical repeats has not been clearly established.

Repeat five (residues 121–142) in the lipid-free form is thought to be unfolded, and this region may play a role in helix formation and structural rearrangement upon lipid binding (13). Thus the structural organization of apoA-I seems to be that of discreet folded repeats that function in a concerted fashion.

The interfacial behavior of apoA-I has been studied at the A/W interface by a number of investigators since the pioneering studies of Phillips et al. (18) and Shen and Scanu (20). When spread on a Langmuir surface balance on low salt buffers, the protein has a gas to liquid transition at low pressure and about 24–30 Å2 per amino acid depending on the study. Weinberg (26) found that the protein films on a 0.1 M NaC1 sub-phase collapsed at a pressure of about 17 to 18 mN/m and an area of about 21 Å2 per amino acid. When apoA-I was spread on a 2 M KCl sub-phase, the isotherm was similar, but the collapse pressure was raised to ~24 mN/m and the area was 20 Å2 per amino acid. Bolaños-Garcia (46) studied the isotherms of several apolipoproteins, including apoA-I on a 3.5 M KCl sub-phase and found that the gas/liquid transition was roughly at 27–30 Å2 per amino acid, and the collapse was at ~18 Å2 per amino acid and 30 mN/m. The isotherm did not appear to have any major transitions between very low pressure (1 mN/m) and the collapse pressure, although a slight inflection occurred at about 8 mN/m. This is unlike a similar apolipoprotein (apoA-IV) where Weinberg et al. (26) found that when spread on a 0.1 M NaCl sub-phase there was a transition occurring at about 15 mN/m. Bolaños-Garcia (49) also carried out Brewster angle microscopy and glancing incidence x-ray diffraction at ust below and above the collapse pressure (46). Brewster angle microscopy was monotonously gray below the transition but showed very slight striations above it, and glancing incidence x-ray diffraction showed no diffraction indicating a fairly unordered surface of apoA-I.

The interactions of exchangeable apolipoproteins with phospholipids monolayers have been studied widely (15, 18–26). In these studies a surface layer of phospholipids was spread and the apolipoproteins were inected under the surface. If the protein penetrated the surface and pushed the phospholipids together, the pressure increased. When an initial phospholipid pressure was reached where there was no increase in pressure on inection of protein, this is called the exclusion pressure {Pi}e. {Pi}e for apoA-I at an egg PC A/W interface is about 32–33 mN/m (18).

Although much information exists for apoA-I and some peptides at the A/W interface, very little data has accumulated concerning the behavior of apoA-I at a more physiological interface such as triglyceride/water or an oil/water interface. Weinberg (26) has studied the adsorption of apoA-I (10 mg/ml in bulk) on the TO/W interface using the drop volume technique. By expressing the droplets at different rates they showed that apoA-I approached an equilibrium {gamma} of about 16 mN/m. To explore further the behavior of apoA-I at the TO/W interface, we carried out extensive studies looking at the fall of {gamma} versus time at different apoA-I bulk concentrations, the compression to displace apoA-I from the surface or to partly push apoA-I off the surface, the desorption rates of apoA-I from the surface, and the elastic properties of apoA-I at the interface. We have compared apoA-I to a consensus peptide called CSP, which models the central part of apoA-I. This peptide has been previously studied at the DD/W and A/W interfaces and was shown to decrease the {gamma} appreciably as a function of time and concentration, to occupy about 14–16Å2 per amino acid at saturation, to be minimally compressible after reaching equilibrium and to be forced off the interface by further compression (28). The present study showed that CSP behaved in a similar fashion and that it could lower the {gamma} ~15 mN/m at its highest bulk concentration studied (2.6 x 10–6 M); the compression beyond a few percent forced the peptide off into the aqueous phase, and expansion allowed it to re-adsorb. It was elastic only at rapid rates of compression and small amplitude. Larger amplitudes or slower rates of compression allowed the peptide to desorb from the surface and re-adsorb when the surface was expanded. The maximum pressure ({Pi}MAX) at which CSP could exist on the surface without being ejected was estimated to be 16 mN/m. ApoA-I also depressed the {gamma} as a function of time and concentration and lowered the {gamma} ~16 mN/m at its highest bulk concentration studied (1.4 x 10–6 M) in 1 h. When apoA-I was rapidly compressed at the surface, small compressions up to about 20–25% ejected part of the protein from the surface, which then rapidly re-adsorbed when surfaces were expanded. The re-adsorption rate was much faster under these conditions than the initial adsorption rate of apoA-I from the aqueous phase. However, large compressions, i.e. >50%, caused the desorption of the complete protein from the surface and when the surface was expanded the re-adsorption rate was almost identical to the initial adsorption rate. Therefore, two processes could be distinguished at the apoA-I/triolein interface, first, a partial expulsion of some parts of the protein from the surface and second, at higher pressure an expulsion of the entire protein from the surface. A {Pi}MAX for partial desorption was estimated to be about 14.8 mN/m where the threshold for complete desorption of the whole peptide ({Pi}OFF) was estimated to be 18.9 mN/m. Furthermore, from the {Pi}-A curves the viscoelastic properties of apoA-I were assessed at different rates. When the rate was very fast, the protein appeared elastic. Whereas, when the period was very slow there was a small viscous component added into the {epsilon} due to the desorption and re-adsorption processes.

The {Pi}MAX of CSP is a little higher (16 mN/m) than that of apoA-I (14.8 mN/m) but lower than the {Pi}OFF of apoA-I. This makes sense, because apoA-I has 10 tandem-repeating amphipathic {alpha}-helix domains, some of which probably have varied affinity to the hydrophobic interface. The more loosely binding domains may desorb at lower pressure than the tightly bound domains. CSP is an"ideal"amphipathic {alpha}-helix consisting of two 22-mer tandem repeats. So it may need a little higher pressure to be pushed off the interface than the loosely binding part of apoA-I. But a much higher pressure still is required to reach the threshold (~19 mN/m) to eject the whole apoA-I off the interface.

The penetration of soluble apolipoproteins and peptides into phospholipid monolayers (15) at the A/W interface using the exclusion pressure ({Pi}e) technique has been well documented (15, 18–26). However, the actual meaning of {Pi}e is quite different from that of {Pi}MAX or {Pi}OFF. {Pi}e is the measurement of the ability of the peptide in the aqueous phase to penetrate between the polar head groups and presumably the aliphatic chains of phospholipids at a given surface concentration of phospholipid. The lower the initial surface concentration, i.e. the lower the pressure exerted by the phospholipids, the greater the amount of the apolipoprotein that can penetrate and increase the surface pressure. The process implies both penetration phenomena between the polar heads and into the interfacial region and probable interaction of hydrophobic domains of the peptide with both the hydrocarbon of the phospholipid chains and perhaps with air. On the other hand, in the case of CSP, {Pi}MAX represents the maximum pressure that that peptide, once established at the interface, can withstand before being ejected back into the aqueous phase.

Nevertheless, there should be some correspondence between the ability of a peptide to penetrate phospholipid monolayers and the pressure at which it is ejected from an interface. Peptides that weakly penetrate a phospholipid monolayer (low {Pi}e) would be expected to have a lower {Pi}MAX than those that had high {Pi}e. Thus we might expect the C and N termini to be those parts of apoA-I that remain anchored to the triolein surface above the {Pi}MAX, i.e. because they have the highest {Pi}e among apoA-I domains (Fig. 14). What actual parts of apoA-I desorbed at {Pi}MAX are not clear, but we would speculate that they were some of the more central helical domains, and perhaps the domain between amino acids 120 and 140, which are only induced to form an {alpha}-helix on binding to phospholipids (13).



View larger version (18K):
[in this window]
[in a new window]
  		FIG. 14.
A schematic model for the surface behavior of apoA-I at the TO/W interface (1). Lipid-free apoA-I adsorbs to the TO/W interface in a time- and concentration-dependent manner. Once the interface reached an equilibrium, it was saturated with apoA-I molecules (upper left). When apoA-I on a triolein drop exerts a surface pressure of less than 14.8 mN/m, the protein is fully spread on the surface with most, if not all of the amphipathic {alpha}-helices engaged in the TO/W interface. Small decreases in the area give rise to small increases in surface pressure (2). When the surface pressure goes above a threshold pressure of 14.8 mN/m ({Pi}MAX) conformational changes occur in apoA-I, which allows some of the middle (M) portion of the apoA-I to disassociate from the TO/W interface leaving the rest of the protein, including the N and C termini (N and C) attached. If the area is suddenly restored to the original area, then these portions very rapidly snap back onto the surface (3). If the volume of the droplet is decreased by about 50–60% (4), which sharply decreases the area and results in the surface pressure rising to above 18.9 mN/m, then the entire protein disassociates into the aqueous phase. Therefore, the threshold pressure for desorption of the entire protein is ~18.9 mN/m. The protein left on the surface at this pressure is probably still partly desorbed from the surface and anchored presumably by the N and/or C termini. When this system is rapidly re-expanded to the original volume (5), the proteins on the surface rapidly revert to the original conformation, but the loss of apoA-I from the surface creates space (6). Aqueous apoA-I re-adsorbs onto the new space. Therefore, the re-adsorption kinetics, as reflected in the fall of {gamma} with time (see Fig. 6B), is slow and similar to that of the original adsorption of apoA-I from the aqueous phase.

 
The concept that different parts of soluble apolipoproteins might bind with different affinities to lipid or lipoprotein surfaces has been nicely covered in two recent reviews (47, 48). The original idea was suggested by Brouillette et al. in 1984 (49) when they noted that DMPC-apoA-I discoidal particles have only two apoA-I molecules ranged in diameter from 126 to 145 Å. To account for apoA-I conformation on the small discs, they suggested that two or three helices of apoA-I came off the DMPC disc and self-associated. Recently, Saito et al. (50) studied the adsorption of apoE4 onto egg yolk phosphatidylcholine/TO emulsions and showed that with nonsaturating amounts of apoE4 on the interface both N and C termini bound. However, when saturating amounts of the apoE4 were used, only the C terminus was bound to the emulsion particle, while the N terminus formed a four-helical bundle in the adacent aqueous phase. Although this study was not correlated with surface pressures, it is worthwhile considering the changes in surface pressure that occur at an emulsion surface as peptide binds. Small and Phillips (51) have estimated that the peptide-free egg yolk phosphatidylcholine/TO emulsions (~100 nm diameter) have a surface pressure of about 17 mN/m. However, because apolipoproteins such as apoA-I or apoE bound to the interface, the surface pressure would increase proportionally to the amount of bound protein. Thus, we suggest that when the first few molecules of apoE adsorbed onto the emulsion interface both the N and C termini bind. As more protein binds and the surface pressure increases, at a certain pressure the N-terminal is pushed off the surface. At saturation only the C-terminal remains bound. We estimate the surface pressure at saturation would be near {Pi}e for apoE. Thus, the consideration of the surface pressure on emulsions is consistent with that of the entire apoE molecule bound at low surface pressures, i.e. low and nonsaturating amounts of protein on the interface, whereas at high surface pressures near {Pi}e the N terminus desorbed. It would be instructive to study the adsorption of apoE4 onto the TO/W interface and determine the {Pi}MAX and {Pi}OFF of apoE and compare them to the {Pi}MAX of the N and C termini.


   FOOTNOTES
 
* This work was supported by NHLBI, National Institutes of Health Grant 2P01-HL26335-21. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

{ddagger} To whom correspondence should be addressed: Dept. of Physiology and Biophysics, Boston University School of Medicine, 715 Albany St., W-302, Boston, MA 02118. Tel.: 617-638-4001; Fax: 617-638-4041; E-mail: dmsmall{at}bu.edu

1 The abbreviations used are: apoA-I, apolipoprotein A-I; HDL, high density lipoprotein; TAG, triacylglycerol; TO/W, triolein/water; VLDL, very low density lipoprotein; DMPC, dimyristoyl phosphatidylcholine; A/W, air/water; PC, phosphatidylcholine; DD/W, dodecane/water; apoA-IV/A-IV, apolipoprotein A-IV; apoE-3, apolipoprotein E-3; A{beta}S, amphipathic {beta} strands; CSP, consensus sequence peptide; mN, millinewton(s). Back


   ACKNOWLEDGMENTS
 
We thank Donna Ross for manuscript preparation.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Atkinson, D., and Small, D. M. (1986) Annu. Rev. Biophys. Biophys. Chem. 15, 403–456[CrossRef][Medline] [Order article via Infotrieve]
  2. Segrest, J. P., Jackson, R. L., Morrisett, J. D., and Gotto, A. M., Jr. (1974) FEBS Lett. 38, 247–253[CrossRef][Medline] [Order article via Infotrieve]
  3. Segrest, J. P., Jones, M. K., DeLoof, H., Brouilette, C. G., Venkatachalapathi, Y. V., and Anantharamaiah, G. M. (1992) J. Lipid Res. 33, 141–166[Abstract]
  4. Small, D. M. (1992) in Current Issues in Endocrinology and Metabolism: Plasma Lipoproteins and Coronary Artery Disease (Kreisberg, R. A., and Segrest, J. P., eds) pp. 57–91, Blackwell Scientific Publications, Boston, MA
  5. Tall, A. R., Small, D. M., Deckelbaum, R. J., and Shipley, G. G. (1977) J. Biol. Chem. 252, 4701–4717[Free Full Text]
  6. Pownall, H. J., Massey, J. B., Kusserow, S. K., and Gotto, A. M., Jr. (1979) Biochemistry 181, 574–579
  7. Brouillette, C. G., and Anantharamaiah, G. M. (1995) Biochim. Biophys. Acta 1256, 103–129[Medline] [Order article via Infotrieve]
  8. Brouillette, C. G., Anantharamaiah, G. M., Engler, J. A., and Borhani, D. W. (2001) Biochim. Biophys. Acta 1531, 4–46[Medline] [Order article via Infotrieve]
  9. Scanu, A. M. (1972) Biochim. Biophys. Acta 265, 471–508
  10. Ritter, M. C., and Scanu, A. M. (1977) J. Biol. Chem. 252, 1200–1216
  11. Pittman, R. C., Glass, C. A., Atkinson, D. A., and Small, D. M. (1987) J. Biol. Chem. 262, 2435–2442[Abstract/Free Full Text]
  12. Fang, Y., Gursky, O., and Atkinson, D. (2003) Biochemistry 42, 13260–13268[CrossRef][Medline] [Order article via Infotrieve]
  13. Gorshkova, I. N., Liu, T., Zannis, V. I., and Atkinson, D. (2002) Biochemistry 41, 10529–10539[CrossRef][Medline] [Order article via Infotrieve]
  14. Saito, H. S., Dhanasekaran, P., Nguyen, D., Holvoet, P., Lund-Katz, S., and Phillips, M. C. (2003) J. Biol. Chem. 278, 23227–23232[Abstract/Free Full Text]
  15. Palgunachari, M. N., Mishra, V. K., Lund-Katz, S., Phillip, M. C., Adeyeye, S. O., Alluri, S., Anantharamaiah, G. M., and Segrest, J. P. (1996) Arterioscler. Thromb. Vasc. Biol. 16, 328–338[Abstract/Free Full Text]
  16. Chao, Y. (2002) Conformational Studies of a Consensus Sequence Peptide (CSP) and a Real Sequence Peptide (RSP) of Apolipoproteins by Circular Dichroism and X-ray Crystallography. Ph.D. thesis, Department of Physiology and Biophysics, Boston University School of Medicine, Boston, MA
  17. Zhu, H., and Atkinson, D. (2004) Biochemistry 43, 13156–13164[CrossRef][Medline] [Order article via Infotrieve]
  18. Phillips, M. C., and Krebs, K. E. (1986) Methods Enzymol. 128, 387–403[Medline] [Order article via Infotrieve]
  19. Krebs, K. E., Phillips, M. C., and Sparkes, C. E. (1983) Biochim. Biophys. Acta 751, 470–473[Medline] [Order article via Infotrieve]
  20. Shen, B. W., and Scanu, A. M. (1980) Biochemistry 19, 3643–3650[CrossRef][Medline] [Order article via Infotrieve]
  21. Krebs, K. E., Ibdah, J. A., and Phillips, M. C. (1988) Biochim. Biophys. Acta 959, 229–237[Medline] [Order article via Infotrieve]
  22. Ibdah, J. A., and Phillips, M. C. (1988) Biochemistry 27, 7155–7162[CrossRef][Medline] [Order article via Infotrieve]
  23. Ibdah, J. A., Lund-Katz, S., and Phillips, M. D. (1989) Biochemistry 28, 1126–1133[CrossRef][Medline] [Order article via Infotrieve]
  24. Ibdah, J. A., Krebs, K. E., and Phillips, M. C. (1989) Biochim. Biophys. Acta 1004, 300–308[Medline] [Order article via Infotrieve]
  25. Weinberg, R. B., Ibdah, J. A., and Phillips, M. C. (1992) J. Biol. Chem. 267, 8977–8983[Abstract/Free Full Text]
  26. Weinberg, R. B., Cook, V. R., DeLozier, J. A., and Shelness, G. S. (2000) J. Lipid Res. 41, 1419–1427[Abstract/Free Full Text]
  27. Datta, G., Chaddha, M., Hama, S., Navab, M., Fogelman, A. M., Garber, D. W., Mishra, V. K., Epand, R. M., Epand, R. F., Lund-Katz, S., Phillips, M. C., Segrest, J. P., and Anantharamaiah, G. M. (2001) J. Lipid Res. 42, 1096–1104[Abstract/Free Full Text]
  28. Wang, L., Atkinson, D., and Small, D. M. (2003) J. Biol. Chem. 278, 37480–37491[Abstract/Free Full Text]
  29. Nolte, R. T., and Atkinson, D. (1992) Biophys. J. 63, 1221–1239[Medline] [Order article via Infotrieve]
  30. Wang, L., and Small, D. M (2004) J. lipid Res. 45, 1704–1715[Abstract/Free Full Text]
  31. Wetterau, J. R., and Jonas, A. (1982) J. Biol. Chem. 257, 10961–10966[Abstract/Free Full Text]
  32. Labourdenne, S., Gaudry-Rolland, N., Letellier, S., Lin, M., Cagna, A., Esposito, G., Verger, R., and Rivière, C. (1994) Chem. Phys. Lipids 71, 163–173[CrossRef]
  33. Benamins, J., and Lucassen-Reynders, E. H. (1998) in Proteins at Liquid Interfaces, (Möbius, D., and Miller, R., eds) pp. 341–384, Elsevier, Amsterdam
  34. Benamins, J., Cagna, A., and Lucassen-Reynders, E. H. (1996) Coll. and Surf. 114, 245–254[CrossRef]
  35. Wang, N., Lan, D., Chen, W., Matsuura, F., and Tall, A. R. (2004) Proc. Natl. Acad. Sci. U. S. A. 101, 9774–9779[Abstract/Free Full Text]
  36. Gursky, O., and Atkinson, D. (1996) Proc. Natl. Acad. Sci. U. S. A. 93, 2991–2995[Abstract/Free Full Text]
  37. Fang, Y., Gursky, O., and Atkinson, D. (2003) Biochemistry 39, 15910–15919
  38. Jonas, A., Kezdy, K. E., and Wald, J. H. (1989) J. Biol. Chem. 264, 4818–4824[Abstract/Free Full Text]
  39. Brasseur, R., De Meutter, J., Vanloo, B., Goormaghtigh, E., Ruysschaert, J. M., and Rosseneu, M. (1990) Biochim. Biophys. Acta 1043, 245–252[Medline] [Order article via Infotrieve]
  40. Jonas, A., Wald, J. H., Toohill, K. L., Krul, E. S., and Kezdy, K. E. (1990) J. Biol. Chem. 265, 22123–22129[Abstract/Free Full Text]
  41. Sparks, D. L., Lund-Katz, S., and Phillips, M. C. (1992) J. Biol. Chem. 267, 25839–25847[Abstract/Free Full Text]
  42. Tall, A. R., Small, D. M., Shipley, G. G., and Lees, R. S. (1975) Proc. Natl. Acad. Sci. U. S. A. 72, 4940–4942[Abstract/Free Full Text]
  43. Tall, A. R., and Small, D. M. (1977) Nature 265, 163–164[CrossRef][Medline] [Order article via Infotrieve]
  44. Atkinson, D., Smith, H. M., Dickson, J., and Austin, J. P. (1976) Eur. J. Biochem. 64, 541–547[Medline] [Order article via Infotrieve]
  45. Segrest, J. P., Garber, D. W., Brouillette, C. G., Harvey, S. C., and Anantharamaiah, G. M. (1994) Adv. Prot. Chem. 45, 303–369[Medline] [Order article via Infotrieve]
  46. Bolanos-Garcia, V. M., Ramos, S., Castillo, R., Xicohtencatl-Cortes, J., and Mas-Oliva, J. (2001) J. Phys. Chem. B 105, 5757–5765[CrossRef]
  47. Narayanaswami, V., and Ryan, R. O. (2000) Biochim. Biophys. Acta 1483, 15–36[Medline] [Order article via Infotrieve]
  48. Saito, H., Lund-Katz, S., and Phillips, M. C. (2004) Prog. Lipid Res. 43, 350–380[CrossRef][Medline] [Order article via Infotrieve]
  49. Brouillette, C. G., Jones, J. L., Ng, T. C., Kercret, H., Chung, B. H., and Segrest, J. P. (1984) Biochemistry 23, 359–367[CrossRef][Medline] [Order article via Infotrieve]
  50. Saito, H., Dhanasekaran, P., Baldwin, F., Weisgraber, K. H., Lund-Katz, S., and Phillips, M. C. (2001) J. Biol. Chem. 276, 40949–40954[Abstract/Free Full Text]
  51. Small, D. M., and Phillips, M. C. (1992) Adv. Coll. Interf. Sci. 41, 1–8[CrossRef][Medline] [Order article via Infotrieve]

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles:


Home page
 J. Lipid Res.Home page
L. Wang, D. D. O. Martin, E. Genter, J. Wang, R. S. McLeod, and D. M. Small
Surface study of apoB1694-1880, a sequence that can anchor apoB to lipoproteins and make it nonexchangeable
J. Lipid Res., July 1, 2009; 50(7): 1340 - 1352.
[Abstract] [Full Text] [PDF]

Home page
 J. Lipid Res.Home page
D. M. Small, L. Wang, and M. A. Mitsche
The adsorption of biological peptides and proteins at the oil/water interface. A potentially important but largely unexplored field
J. Lipid Res., April 1, 2009; 50(Supplement): S329 - S334.
[Abstract] [Full Text] [PDF]

Home page
 Proc. Natl. Acad. Sci. USAHome page
L. Wang, M. T. Walsh, and D. M. Small
Apolipoprotein B is conformationally flexible but anchored at a triolein/water interface: A possible model for lipoprotein surfaces
PNAS, May 2, 2006; 103(18): 6871 - 6876.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Wang, L.
Right arrow Articles by Small, D. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wang, L.
Right arrow Articles by Small, D. M.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 All ASBMB Journals   Molecular and Cellular Proteomics 
 Journal of Lipid Research   ASBMB Today 
Copyright © 2005 by the American Society for Biochemistry and Molecular Biology.
Advertisement
spacer
Advertisement
Advertisement