help button home button Biophys. J.
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
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 arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kim, J. R.
Right arrow Articles by Murphy, R. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kim, J. R.
Right arrow Articles by Murphy, R. M.
Biophysical Journal 86:3194-3203 (2004)
© 2004 The Biophysical Society

Mechanism of Accelerated Assembly of ß-Amyloid Filaments into Fibrils by KLVFFK6

Jin Ryoun Kim and Regina M. Murphy

Department of Chemical and Biological Engineering, University of Wisconsin, Madison, Wisconsin

Correspondence: Address reprint requests to Dr. Regina M. Murphy, Dept. of Chemical and Biological Engineering, University of Wisconsin, 1415 Engineering Dr., Madison, WI 53706. Tel: 608-262-1587; Fax: 608-262-5434; E-mail: murphy{at}che.wisc.edu.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
 REFERENCES
 
Extracellular senile plaques are a central pathological feature of Alzheimer's disease. At the core of these plaques are fibrillar deposits of ß-amyloid peptide (Aß). In vitro, Aß spontaneously assembles into amyloid fibrils of cross-ß sheet structure. Although it was once believed that the fibrils themselves were toxic, more recent data supports the hypothesis that aggregation intermediates, rather than fully formed fibrils, are the most damaging to neuronal tissue. In previously published work, we identified several small peptides that interact with Aß and increase its aggregation rate while decreasing its toxicity. In this work, we examined in detail the interaction between Aß and one of these peptides. Using a mathematical model of Aß aggregation kinetics, we show that the dominant effect of the peptide is to accelerate lateral association of Aß filaments into fibrils.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
 REFERENCES
 
Alzheimer's disease (AD) is an age-associated neurodegenerative disease characterized by loss of memory and language skills, damaged cognitive function, and altered behavior. A central pathological feature of AD is the presence of extracellular senile plaques, found in the hippocampus and the neocortex and associated with synaptic loss and cell death (Selkoe, 1991Go). At the core of senile plaques are proteinaceous amyloid deposits. Chemical analysis of these deposits revealed that the major protein constituent is ß-amyloid (Aß) (Glenner and Wong, 1984Go; Masters et al., 1985Go). Aß is a 39–43 residue proteolytic fragment of a larger integral membrane protein called amyloid precursor protein (APP) (Kang et al., 1987Go).

Aß undergoes aggregation spontaneously and assembles into amyloid fibrils with a cross ß-sheet structure (Kirschner et al., 1987Go). Aß assembly can be accelerated by several factors, including locally high Aß concentration, acidic pH, metal ions, osmolytes, and interaction with lipid membranes (Barrow and Zagorski, 1991Go; Hilbich et al., 1991Go; Fraser et al., 1992Go; Yang et al., 1999Go; Yip et al., 2002Go; for review see McLaurin et al., 2000Go). Electron microscopy (EM) studies demonstrated that Aß fibrils from senile plaques are straight and unbranched, 5–12 nm in diameter, and appear to be in a helical array of ß-sheet structure (Burkoth et al., 2000Go; for review see Serpell, 2000Go).

Multiple observations indicate that amyloid fibril formation is an early and required event in the AD neurodegenerative process (Pike et al., 1993Go; Selkoe, 1993Go; Simmons et al., 1994Go; Moran et al., 1995Go; Geula et al., 1998Go). The "amyloid cascade" hypothesis for AD postulates that amyloid fibril accumulation directly leads to synaptic loss, neuritic dystrophy, and the neurotransmitter deficits that are manifestations of dementia (Hardy and Higgins, 1992Go). However, more recent in vitro observations have demonstrated that small, soluble, and diffusible oligomeric Aß species are also capable of initiating pathogenic events (Roher et al., 1996Go; Lambert et al., 1998Go; Hartley et al., 1999Go). These data have motivated several researchers to postulate that Aß oligomeric intermediates, rather than fully formed fibrils, are the predominant toxic species (Kirkitadze et al., 2002Go).

Considerable research effort has focused on discovery of candidate compounds that block the toxicity of Aß, by targeting a specific step involved in Aß aggregation (Camilleri et al., 1994Go; Tomiyama et al., 1994Go; Klunk et al., 1998Go; Pappolla et al., 1998Go; Hughes et al., 2000Go). Given the hypothesis that aggregation intermediates are responsible for Aß toxicity, such compounds could theoretically prevent all aggregation, or alternatively cause further association of toxic oligomers into larger nontoxic aggregates. In previous work from our group, a strategy for designing peptidyl inhibitors against Aß toxicity was proposed (Pallitto et al., 1999Go; Lowe et al., 2001Go). Briefly, the inhibitors were envisioned as containing a "recognition domain," a short peptide sequence homologous to a fragment of full-length Aß (KLVFF, 16–20), and a "disrupting domain," a polypeptide chain with the ability to interfere with Aß aggregation. Inhibitors that protected PC-12 cells from Aß toxicity actually increased the rate of Aß aggregation (Pallitto et al., 1999Go; Lowe et al., 2001Go). Among these peptides, KLVFFK6 was the most potent at preventing Aß-associated toxicity to PC-12 cells and caused the largest change in Aß aggregation kinetics and aggregate morphology (Pallitto et al., 1999Go).

In previous work, we developed a mathematical model of Aß aggregation kinetics from in vitro experimental data (Pallitto and Murphy, 2001Go). In this article, we carefully evaluated Aß aggregation kinetics in the presence of KLVFFK6. Several questions we addressed include: 1), Does the inhibitor change the distribution of Aß between nonamyloid and amyloid pathways? 2), Which specific step in the Aß aggregation pathway is the most affected by KLVFFK6? 3), What is the mechanism of interaction between Aß and inhibitors? To answer these questions, physicochemical measurements were collected and the data analyzed using the kinetic model.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
 REFERENCES
 
Peptide synthesis
Aß(1–40) was purchased from AnaSpec (San Jose, CA). KLVFFK6 was synthesized by solid phase peptide synthesis using Fmoc-protected amino acids and purified by reverse phase high-pressure liquid chromatography on a C4 column (Vydac, Hesperia, CA) using linear gradient of acetonitrile/water with 0.1% trifluoroacetic acid. Molecular mass of KLVFFK6 was analyzed by mass spectrometry to be 1421.7 Da (theoretical molecular mass of 1421.9 Da). Peptides were stored as lyophilized powders at –70°C.

Sample preparation
Phosphate-buffered saline with azide ((PBSA) 0.01 M Na2HPO4/NaH2PO4, 0.15 M NaCl, 0.02% (w/v) NaN3, pH 7.4) was double filtered through 0.22-µm filters. Urea (8 M) was prepared in 10 mM glycine-NaOH buffer, pH 10, then filtered through 0.22-µm filters. Lyophilized Aß(1–40) was solubilized at a concentration of 2.8 mM using prefiltered 8 M urea, pH 10. After 10 min, samples were diluted to 140 µM Aß into filtered PBSA, or PBSA containing KLVFFK6. Samples were rapidly filtered through 0.45-µm filters directly into light-scattering cuvettes (for light scattering (LS)) or microtubes (for size exclusion chromatography (SEC)). Molecular-biology-grade urea was purchased from Boehringer-Mannheim (Indianapolis, IN). All other chemicals were purchased from Sigma-Aldrich (St. Louis, MO). The concentration of Aß was determined from the peak area of the sample injected onto the fast protein liquid chromatography system without the column in place, using an extinction coefficient of 0.3062 (mg/ml)–1 cm–1 (Pallitto and Murphy, 2001Go), and was 120 ± 20 µM. Residual urea was 0.4 M. Urea dissolution was required to yield highly quantitatively reproducible results and to ensure a controlled initial state. The urea may change absolute rates of aggregation, but does not change trends (Lowe et al., 2001Go).

Size exclusion chromatography
Samples were analyzed with SEC using a precision column prepacked with Superdex 75 (Pharmacia, Piscataway, NJ) on a Pharmacia fast protein liquid chromatography system, as described previously (Pallitto and Murphy, 2001Go). Briefly, the mobile phase flow rate was set at 0.05 ml/min and elution peaks were detected by ultraviolet (UV) absorbance at either 214 or 280 nm. Mobile phase buffer was matched to buffer used for dissolution of Aß. The column was calibrated using the following proteins as molecular weight standards: insulin chain B (3500), ubiquitin (8500), ribonuclease A (13,700), ovalbumin (43,000), and bovine serum albumin (67,000). To determine the distribution between small oligomers that could be resolved on the column (molecular mass 3–70 kDa), and larger species that could not be resolved, samples were injected without the column in place; the percent of nonaggregates (monomer and dimer (M+D)) was calculated by dividing the M+D peak area by the peak area without the column in place.

Laser light scattering
Samples prepared as described above were placed in a bath of the index-matching solvent decahydronaphthalene, which was temperature controlled to 25°C. Dynamic light scattering data were taken using a Coherent (Santa Clara, CA) argon ion laser operated at 488 nm and a Malvern 4700c system (Southborough, MA), as described in more detail elsewhere (Lowe et al., 2001Go).

Information on average particle molecular mass, shape, and dimensions were obtained using static light scattering measurements, as described previously in detail (Murphy and Pallitto, 2000Go). Two alternative models of particle shape, semiflexible (wormlike) chain and semiflexible branched, were used to fit the data. The semiflexible chain model describes a linear chain with total contour length Lc and Kuhn statistical segment length lk (a measure of the stiffness of the chain, equal to two times the persistence length). We have shown previously that this model is a good description of Aß fibrils (Shen et al., 1993Go; Shen and Murphy, 1995Go). The continuous semiflexible branched model describes a branched particle with a center from which extend nb (number of branches) semiflexible chains of contour length Lc,a and stiffness lk.

Circular dichroism spectroscopy
Secondary structure of Aß was determined using circular dichroism (CD), collected using an Aviv 62A DS circular spectrometer (Lakewood, NJ) in the far-UV range with 0.1 cm of pathlength of cuvette. Ellipticity of sample containing 140 µM of Aß with and without addition of KLVFFK6 at each wavelength was measured without dilution for 5 s and then averaged out. The lower wavelength boundary was limited to 211 nm due to the residual urea (0.4 M) in the sample. The spectrum of the background was measured for 5 s and averaged, followed by that of the sample. The average background was then subtracted from the average sample spectrum. The percent of each secondary structure element was calculated by nonlinear least square curve fitting of experimental ellipticities using several standard methods (Provencher and Glöckner, 1981Go; Manavalan and Johnson, 1987Go).

Mathematical modeling
A mathematical model of Aß aggregation kinetics was developed by Pallitto and Murphy (2001)Go. A schematic of the model is shown in Fig. 1. Diameters of filament and fibril were assumed to be 3 and 8 nm, respectively. Model equations were solved numerically, and parameter values were obtained by multiresponse nonlinear regression, as described previously (Pallitto and Murphy, 2001Go).



View larger version (21K):
[in this window]
[in a new window]
  		FIGURE 1  A schematic showing Aß aggregation kinetic model (Pallitto and Murphy, 2001Go).

 

   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 ACKNOWLEDGEMENTS
 REFERENCES
 
To allow for a more rigorous analysis of Aß kinetics, 8 M urea/pH 10 was used to completely dissolve Aß and render it monomeric and random coil before dilution into PBSA to initiate aggregation (Pallitto and Murphy, 2001Go). A mathematical model describing the kinetics of assembly was developed previously (Pallitto and Murphy, 2001Go). A schematic of the model is shown in Fig. 1. Briefly, denatured monomers Mu rapidly refolded into either stable monomer M or dimer D, or less stable dimeric intermediate I. I cooperatively assembled to nucleus N to initiate formation of thin filaments f. Filaments grew linearly by repeated addition of I. Cooperative lateral association of filaments resulted in formation of fibrils F. F elongated by end-to-end association. The kinetic model was in good agreement with SEC and LS experiments.

KLVFFK6 protected PC-12 cells from Aß-associated toxicity while significantly accelerating Aß aggregation (Pallitto et al., 1999Go). In those experiments, Aß was dissolved in 0.1% trifluoroacetic acid before dilution into PBSA containing test inhibitory peptide. In the work presented here, we used the mathematical model to evaluate the mechanism underlying the effect of KLVFFK6 on aggregation kinetics of Aß, starting from the urea-unfolded state. Points of interest include: a), alteration of secondary structure of Aß; b), change in initial M+D and I split; c), shift of Aß nucleation/elongation balance; d), acceleration/deceleration of filament assembly to fibril; and e), extent of incorporation of KLVFFK6 into Aß.

Effect of KLVFFK6 on Aß secondary structure
Aß aggregation is linked to conversion of the peptide from random coil to ß-sheet secondary structure. Aß in 8 M urea/pH 10 is random coil, but after dilution into PBSA, Aß spontaneously aggregates into ß-sheet fibrils. We evaluated whether KLVFFK6 increased the extent of ß-sheet structure. Circular dichroism spectra were taken on samples containing 140 µM Aß and varying concentrations of KLVFFK6 (0:1, 0.1:1, 0.5:1, and 1:1 KLVFFK6:Aß molar ratio). KLVFFK6 alone was random coil (Fig. 2). KLVFFK6 did not alter CD spectrum of Aß at any concentration, indicating no disturbance of secondary structure by KLVFFK6 (Fig. 2). Deconvolution of CD spectra revealed ~35% of ß-sheet structure regardless of presence or absence of KLVFFK6.



View larger version (14K):
[in this window]
[in a new window]
  		FIGURE 2  Circular dichroism (CD) spectra of 140 µM of KLVFFK6 alone (thick solid line), Aß alone (thin solid line), and KLVFFK6 + Aß at 1:1 (dashed line). Data for the other molar ratio of KLVFFK6:Aß we tested were indistinguishable from those for Aß alone and 1:1 ratio of KLVFFK6 + Aß (data not shown). All samples contained 140 µM of Aß and were incubated for ~18–20 h before CD measurements. For a given wavelength, data were collected for 5 s and averaged out. Due to residual urea, the minimum wavelength was limited to 211–212 nm.

 
Effect of KLVFFK6 on Aß population
Dilution of Aß from urea into PBSA produces a distribution of variously sized species. Under our experimental conditions, ~63% of total Aß was present as stable Aß monomer or dimer, whereas the remainder was aggregates of >70 kDa, consistent with previous reports (Pallitto and Murphy, 2001Go). Size exclusion chromatography experiments were performed to see whether KLVFFK6 changed the distribution of Aß species between nonamyloidogenic and amyloidogenic states. Detection was set at 280 nm, at which wavelength Aß, but not KLVFFK6, is detected. Percentages of Aß monomer or dimer and aggregates were calculated and compared to Aß alone (Table 1). Aß population was not affected by KLVFFK6, and the distribution did not change over time (data not shown).


View this table:
[in this window]
[in a new window]
  		TABLE 1  Effect of KLVFFK6 on Aß population

 
Incorporation of KLVFFK6 into Aß
If KLVFFK6 bound to Aß monomers and dimers, one might expect a shift in elution time of the M+D peak caused by a molecular mass increase. No such shift was observed in the presence of KLVFFK6 (data not shown). This result indicates that KLVFFK6 is not associated to any measurable extent with monomeric or dimeric Aß species. We repeated SEC analysis of Aß + KLVFFK6 with detection at 214 nm, where both Aß and KLVFFK6 contributes to absorbance. We observed a significant peak at 39 min elution time, matching that of KLVFFK6 alone (data not shown). Comparison of the area of this peak in the absence and presence of Aß suggested most (>90%) of KLVFFK6 eluted as free peptide, although accurate quantification was not possible due to low signal/noise ratio. This result implied very little of KLVFFK6 is incorporated into aggregated Aß.

Effect of KLVFFK6 on Aß aggregation kinetics
We evaluated Aß aggregation kinetics in the presence of varying concentrations of KLVFFK6 using light scattering. Data are reported as average hydrodynamic diameter, dsph and scattering intensity at 90° angle, Is(90°), versus time. Because small species contribute less to scattered light, dsph and Is(90°) indicate primarily the change in size and/or mass of aggregates rather than of Aß M+D. The rate of increase in dsph and Is(90°) increased, with increasing amounts of KLVFFK6. The increase in Is(90°) was approximately twofold greater than the increase in dsph (Fig. 3). Because Is(90°) scales approximately with average aggregate molecular mass, whereas dsph scales approximately with average aggregate length, these results indicate that KLVFFK6 increased the linear density (mass per unit length) of Aß aggregates.



View larger version (22K):
[in this window]
[in a new window]
  		FIGURE 3  Hydrodynamic diameter dsph (A and B) and scattering intensity at 90° angle Is(90°) (C and D) for Aß alone ({circ}), KLVFFK6 + Aß (1:10) (•), (1:1) ({triangleup}), (2.5:1) ({blacktriangleup}), (5:1) ({square}), and (10:1) ({blacksquare}). All samples contained 140 µM Aß.

 
Effect of KLVFFK6 on Aß aggregate size and shape
We wondered whether the morphology of Aß aggregates changed in the presence of KLVFFK6. To determine this, we collected static light scattering data at multiple angles; results are presented as a Kratky plot (Fig. 4). For Aß alone and KLVFFK6 + Aß (1:1 molar ratio), the Kratky plots were fairly linear, characteristic of a rodlike morphology. With increasing concentration of KLVFFK6, a shift of the aggregate morphology was apparent. At 2.5:1 molar ratio, a plateau at higher angles was observed, indicative of longer fibrils and a linear semiflexible chain morphology. At 5:1 and 10:1 molar ratio, there was a substantial change in the curve, exhibiting an intermediate "bump" that is characteristic of a branched morphology.



View larger version (19K):
[in this window]
[in a new window]
  		FIGURE 4  Effect of KLVFFK6 on scattering intensity of Aß aggregates. Data are presented in the form of Kratky plots. Data were collected 21 h after sample preparation for Aß alone ({circ}), KLVFFK6 + Aß (1:1) ({triangleup}), (2.5:1) ({blacktriangleup}), and (5:1) ({square}) or after 2 h for KLVFFK6 + Aß (10:1) ({blacksquare}). Each sample contained 140 µM Aß. Lines indicate the fitted particle shape function, P(q) for semiflexible chains ({circ}, {triangleup}, and {blacktriangleup}) or a branched structure ({square} and {blacksquare}). Kratky plot analysis of 10:1 sample after 21 h could not be made due to formation of significant amount of precipitates. c is the peptide concentration. q = 4{pi}n/{lambda}0 sin ({theta}/2), where n is the refractive index of the solvent, {lambda}0 is the wavelength of the incident beam in vacuo, and {theta} is the scattering angle. R is the Rayleigh ratio of sample. where dn/dc is the refractive index increment, and NA is Avogadro's number.

 
<M>w, Lc (or Lc,a), lk, and nb were determined by fitting the data to model equations for the appropriate particle structure function. Results are summarized in Table 2. Both <M>w and Lc increased with increasing concentration of KLVFFK6, with a greater increase in <M>w than Lc. The increase in <M>w/Lc implies greater average linear density with increasing concentration of KLVFFK6, consistent with results shown in Fig. 3.


View this table:
[in this window]
[in a new window]
  		TABLE 2  Size characteristics of Aß aggregates in the presence of KLVFFK6

 
Kinetic model of Aß aggregation in the presence of KLVFFK6
Data were analyzed based on the kinetic model developed previously (Pallitto and Murphy, 2001Go). SEC and CD data indicated that KLVFFK6 does not change the split between amyloidogenic and nonamyloidogenic Aß, implying no change in the "refolding" step; this is reflected in the model as no change in KMD, kM/kI, and kD/kI (Fig. 1). KLVFFK6 did, however, strongly affect kinetics of growth in a concentration-dependent manner. Therefore, the major change is downstream of the "refolding" step (Fig. 1). Light scattering data were used to obtain the best fit parameter values for filament initiation and elongation (kp/kn), and fibril formation by lateral association (kla) and fibril growth by end-to-end association ({delta}{omega}fib) (Fig. 1). Because the kinetic model assumes a linear aggregate morphology, kinetic data of samples exhibiting a branched morphology could not be fitted to the model. For 5:1 ratio of KLVFFK6:Aß, it was not possible to obtain fits if the entire data set was used; therefore we used only data up to ~4 h to evaluate parameters. For the same reason, no attempt was made to fit any of the data for 10-fold excess of KLVFFK6.

Our kinetic model captured growth of Aß in the presence of KLVFFK6 reasonably well (Fig. 5). Parameter values are summarized in Table 3. First, we examined the effect of KLVFFK6 on filament initiation and elongation. kn and kp are rate constants governing nucleus N initiation from unstable intermediate species I, and filament initiation and growth by addition of I, respectively. Therefore, kn/kp represents the balance between new filament formation versus growth of existing filaments. kn/kp was not substantially affected by KLVFFK6, indicating that the peptide did not disturb this balance. Then, we examined the effect of KLVFFK6 on fibril formation and growth. kla characterizes the rate of lateral association of filaments into fibrils, whereas {delta}{omega}fib characterizes the rate of growth by fibril end-to-end association. {delta}{omega}fib decreased moderately in the presence of KLVFFK6. By far the greatest effect of KLVFFK6 was on lateral association of filaments into fibrils: kla increased by approximately fourfold at 1:10 ratio of KLVFFK6:Aß, ~70-fold at 1:1 ratio, and ~300-fold at 2.5:1 ratio.



View larger version (26K):
[in this window]
[in a new window]
  		FIGURE 5  Measured average hydrodynamic diameter dsph (A) and average scattering intensity Is(90°) (B) of Aß aggregates for Aß alone ({circ}), KLVFFK6 + Aß (1:10) (•), (1:1) ({triangleup}), (2.5:1) ({blacktriangleup}), and (5:1) ({square}) are shown along with model simulations (lines) generated using parameters given in Table 3.

 

View this table:
[in this window]
[in a new window]
  		TABLE 3  Effect of KLVFFK6 on model parameters

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
ACKNOWLEDGEMENTS
 REFERENCES
 
Examination of the events by which monomeric Aß associates into oligomers and fibrils is of central importance to elucidation of the molecular mechanisms underlying AD pathogenesis. The fact that fibrils formed in vitro are identical to those in vivo as determined by EM and x-ray diffraction, and are toxic to neuron cells, gives a strong justification for the validity of in vitro Aß aggregation studies (Kirschner et al., 1987Go; Pike et al., 1993Go).

Peptides designed to interfere with Aß aggregation also inhibited Aß-associated toxicity (Pallitto et al., 1999Go; Lowe et al., 2001Go). Among these peptides, KLVFFK6 was judged one of the most active. A detailed mathematical model of Aß aggregation kinetics was previously proposed (Pallitto and Murphy, 2001Go), and we hypothesized that the model could be used to determine the specific step(s) in the aggregation pathway affected by KLVFFK6.

We considered it possible that KLVFFK6 could increase the Aß aggregation rate by changing the secondary structure or the fraction of Aß that is aggregated. However, analysis of CD spectra demonstrated that KLVFFK6 did not change the ß-sheet content of Aß preparations. From SEC analysis, we observed no disturbance of the distribution of Aß between M+D versus aggregated material. Furthermore, KLVFFK6 does not appear to bind measurably to M+D Aß (data not shown). Together these results demonstrate that KLVFFK6 does not act by associating with monomeric Aß, by changing Aß secondary structure, or by affecting the split of Aß between nonamyloidogenic and amyloidogenic species (Fig. 1). Thus, its action is felt in a later step in the aggregation pathway.

Light-scattering analysis indicated that KLVFFK6 does cause an increase in the rate of growth of aggregate size, as measured by both dsph and Is(90°). The change in Is(90°) was greater than the change in dsph, indicating that the increase in growth rate of the average molar mass of the aggregates was greater than the increase in growth rate of the average aggregate length. Closer analysis of the angular dependence of scattering showed that increasing the concentration of KLVFFK6 increased the average molar mass and contour length of the aggregates, and led to changes in morphology from rigid rod to semiflexible chain to branched structures (Table 2 and Fig. 4). A greater mass per unit length in the presence of KLVFFK6 was detected, consistent with observations from the fixed-angle scattering measurements.

The data were used to determine kinetic parameters in our model. Neither the balance between filament elongation versus initiation (kp/kn) nor end-to-end fibril growth ({delta}{omega}fib) were affected much (Table 3). Rather, the major effect of KLVFFK6 was to accelerate the rate of lateral association of filaments to fibrils, as evidenced by a large and dose-dependent increase in kla. This finding is consistent with previous qualitative observations of an increase in linear density of Aß aggregates in the presence of KLVFFK6 (Lowe et al., 2001Go; Moss et al., 2003Go).

We wondered whether we could interpret our results in light of recent findings about Aß aggregate structure reported by the Tycko group (Tycko, 2003Go). Briefly, using solid-state NMR techniques, they proposed the following model: a), each Aß molecule contains two ß-strands (residues 13–24 and 30–40) separated by a bend (25–29); b), a double-layered "cross-ß unit" is constructed from an Aß dimer, with the two strands in each Aß molecule belonging to separate ß-sheets; and c), a "protofilament" contains two cross-ß units laterally assembled into a four-layered structure. The mass-per-unit length (MPL) of a single cross-ß unit was estimated at 9 kDa/nm, and MPL of a protofilament at 18 kDa/nm. Protofilament formation is hypothesized to be driven by hydrophobic collapse occurring between the 30–40 strands in adjacent cross-ß units. Further lateral association of protofilaments, via hydrophobic and electrostatic interactions, is allowed by the Tycko model. Atomic force microscopy studies from other groups provide further evidence of the existence of a mechanism for fibril growth by lateral association (Harper et al., 1997Go; Nichols et al., 2002Go).

We calculated MPL for our Aß aggregates from <M>w, Lc, (Table 2) and the fraction of peptide that was aggregated (Table 1). Average MPL was estimated at ~6 kDa/nm for Aß alone, and increased up to ~20 kDa/nm at high KLVFFK6:Aß ratio. These estimates are in the same range as those of Tycko (2003)Go, and suggest that the aggregated species we observe are not highly laterally associated. In light of the Tycko model, we speculate that, at no or low KLVFFK6, the detected species (what we call "filaments") are predominantly elongated cross-ß units. As the KLVFFK6 dose increases, there is a shift toward lateral assembly of cross-ß units into "protofilaments," and further alignment of "protofilaments" into fibrils. (Our model does not allow for a stable population of "protofilaments"; their stability relative to elongated cross-ß units and fibrils may depend on solvent conditions.) We propose that KLVFFK6 increases the strength of the hydrophobic interaction, driving protofilament and fibril formation. We have recently reported on a chemical modification of KLVFFK6 that increases solvent surface tension and greatly accelerates Aß aggregation relative to KLVFFK6 (Kim et al., 2003Go), in agreement with this hypothesis. This result, together with those reported here, is consistent with a speculation that this class of peptides drives Aß aggregation via changes in solvent properties, therefore strengthening hydrophobically driven lateral association.

Despite the dramatic effect of KLVFFK6 on lateral association, very little (<10%) of the peptide, if any, was actually incorporated into Aß fibrils. Still, a peptide with a nonhomologous recognition element, KLVIIK6, had no effect on Aß aggregation (data not shown), indicating that some binding between KLVFFK6 and Aß must occur. We hypothesized that KLVFFK6 binds reversibly to m binding sites on the filaments;

(1)
where [f] = free filament concentration, [P] = free KLVFFK6 peptide concentration, and [f·Pm] = filament-peptide complex concentration (all in µM). In the absence of KLVFFK6, we modeled the kinetics of lateral association of filaments as a third-order process (Pallitto and Murphy, 2001Go):

(2)

Binding of KLVFFK6 changes filament-filament versus filament-solvent energetics to favor filament-filament lateral association. In the presence of KLVFFK6, we modeled lateral association of filaments with and without bound peptide also as a third-order process:

(3)
where 0 <= s <= 3, and F' is the fibrils formed in the presence of KLVFFK6.

We hypothesized that:

or

(4)
where kla is the lateral association rate constant for Aß alone, and is the lateral association rate constant in the presence of KLVFFK6 and is a function of the peptide's concentration and its affinity for binding to filaments. Because [f]total = [f] + [f·Pm]:

(5)
where [f]total = total filament concentration (in µM). To evaluate these hypotheses, we need to determine how [f·Pm] changes with total peptide concentration. We assumed that KLVFFK6 binds to m identical and noninteracting sites on Aß filaments but does not bind to stable monomer or dimer, or,

(6)
where [Pb] = concentration of peptide bound to filament. Because on average and [P] {approx} [P]total as described in Results, Eq. 5 is further reduced by combination with Eq. 6:

(7)

We plotted log vs. log [P]total from Table 3 and obtained a slope of 3 – s = 1.3, or s = 1.7. Because s < 3, this analysis implies that only half of the filaments in a fibril must contain bound KLVFFK6 to render significant change in lateral association. Using Kd ~ 40 µM for binding of KLVFFK6 to Aß (Cairo et al., 2002Go), combined with our observations that <10% of all KLVFFK6 is bound to Aß, and that the average filament is a 120-mer (Tables 1 and 2), we estimate from Eq. 6 that m {approx} 50. Therefore >60% ((120 – 50)/120) of all Aß molecules in a filament are protected, unable to bind to KLVFFK6.

From our data, we were unable to ascertain whether KLVFFK6 remained bound to fibrils ("stoichiometric" mode) or was released from fibrils and available for another round of binding to filaments ("catalytic" mode). Equation 7 suggests that stronger affinity (lower Kd) of peptide to Aß leads to larger lateral association of filaments, as long as the amount of peptide bound to filament is small relative to total peptide concentration. In the catalytic mode, however, a very strong affinity of peptide binding to Aß filaments would reduce the peptide's effectiveness at increasing lateral association. In other words, there would be an optimum binding affinity of peptide to Aß, if these peptides act catalytically rather than stoichiometrically. With a panel of peptide inhibitors, lower Kd correlated strongly with greater effect on Aß aggregation and greater potency in inhibiting toxicity (Cairo et al., 2002Go). However, in none of these peptides was the binding affinity strong enough to distinguish between catalytic and stoichiometric modes of action.

The fact that the predominant effect of KLVFFK6 is to accelerate lateral alignment, leading to a decrease in the relative fraction of filament compared to fibril, is consistent with the hypothesis that intermediate species such as the cross-ß unit or the protofilament are toxic (Kirkitadze et al., 2002Go). This is demonstrated in Fig. 6, where we use the kinetic model to calculate the filament and fibril fraction as a function of KLVFFK6 concentration. Previous work from our group suggested that intermediate Aß species with highly exposed hydrophobic patches interact most strongly with the lipid bilayer and lead to changes in membrane fluidity (Kremer et al., 2000Go). Such changes in membrane physical properties may produce harmful effects on cellular functioning. KLVFFK6 could reduce Aß toxicity by accelerating burial of these hydrophobic patches and driving formation of inert fully formed fibrils.



View larger version (13K):
[in this window]
[in a new window]
  		FIGURE 6  Weight fraction of filament (•) and fibril ({circ}) simulated by the kinetic model using parameters given in Table 3. Points correspond to KLVFFK6 + Aß at (0:1), (0.1:1), (1:1), and (2.5:1) molar ratios.

 
This analysis demonstrates the utility of a detailed mathematical model in interpreting the mode of action of putative aggregation modifiers. Specifically, for KLVFFK6 we concluded that 1), this inhibitor does not change the distribution of Aß between amyloid and nonamyloid paths; 2), this inhibitor affects most strongly the rate of lateral aggregation of Aß filaments into fibrils; and 3), this inhibitor interacts with aggregated, not monomeric, Aß, to enhance hydrophobically driven association. Such an analysis can contribute in two ways: first, to develop testable hypotheses for the mechanism of toxicity, and second, to suggest new design strategies in the search for better inhibitors of Aß toxicity.


   ACKNOWLEDGEMENTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
REFERENCES
 
We thank Todd Gibson for synthesizing KLVFFK6.

This work was supported by National Institutes of Health grant AG 14079 from the National Institute of Aging. CD data were obtained at the University of Wisconsin-Madison Biophysics Instrumentation Facility, supported by NSF grants BIR-9512577 and S10RR13790, with help from Dr. Darrell McCaslin.

Submitted on November 14, 2003; accepted for publication January 14, 2004.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGEMENTS
 REFERENCES
 
Barrow, C. J., and M. G. Zagorski. 1991. Solution structures of ß peptide and its constituent fragments: relation to amyloid deposition. Science. 253:179–182.[Abstract/Free Full Text]

Burkoth, T. S., T. L. S. Benzinger, V. Urban, D. M. Morgan, D. M. Gregory, P. Thiyagarajan, R. E. Botto, S. C. Meredith, and D. G. Lynn. 2000. Structure of the ß-amyloid (10–35) fibril. J. Am. Chem. Soc. 122:7883–7889.

Cairo, C. W., A. Strzelec, R. M. Murphy, and L. L. Kiessling. 2002. Affinity-based inhibition of ß-amyloid toxicity. Biochemistry. 41:8620–8629.[Medline]

Camilleri, P., N. J. Haskins, and D. R. Howlett. 1994. ß-Cyclodextrin interacts with the Alzheimer amyloid ß-A4 peptide. FEBS Lett. 341:256–258.[Medline]

Fraser, P. E., J. T. Nguyen, D. T. Chin, and D. A. Kirschner. 1992. Effects of sulfate ions on Alzheimer ß/A4 peptide assemblies: implications for amyloid fibril-proteoglycan interactions. J. Neurochem. 59:1531–1540.[Medline]

Geula, C., C. K. Wu, D. Saroff, A. Lorenzo, M. Yuan, and B. A. Yankner. 1998. Aging renders the brain vulnerable to amyloid ß-protein neurotoxicity. Nat. Med. 4:827–831.[Medline]

Glenner, G. G., and C. W. Wong. 1984. Alzheimer's disease: initial report of the purification and characterization of a novel cerebrovascular amyloid protein. Biochem. Biophys. Res. Commun. 120:885–890.[Medline]

Hardy, J. A., and G. A. Higgins. 1992. Alzheimer's disease: the amyloid cascade hypothesis. Science. 256:184–185.[Free Full Text]

Harper, J. D., C. M. Lieber, and P. T. Lansbury, Jr. 1997. Atomic force microscopic imaging of seeded fibril formation and fibril branching by the Alzheimer's disease amyloid-ß protein. Chem. Biol. 4:951–959.[Medline]

Hartley, D. M., D. M. Walsh, C. P. Ye, T. Diehl, S. Vasquez, P. M. Vassilev, D. B. Teplow, and D. J. Selkoe. 1999. Protofibrillar intermediates of amyloid ß-protein induce acute electrophysiological changes and progressive neurotoxicity in cortical neurons. J. Neurosci. 19:8876–8884.[Abstract/Free Full Text]

Hilbich, C., B. Kisters-Woike, J. Reed, C. L. Masters, and K. Beyreuther. 1991. Aggregation and secondary structure of synthetic amyloid ßA4 peptides of Alzheimer's disease. J. Mol. Biol. 218:149–163.[Medline]

Hughes, E., R. M. Burke, and A. J. Doig. 2000. Inhibition of toxicity in the ß-amyloid peptide fragment ß-(25–35) using N-methylated derivatives: a general strategy to prevent amyloid formation. J. Biol. Chem. 275:25109–25115.[Abstract/Free Full Text]

Kang, J., H. G. Lemaire, A. Unterbeck, J. M. Salbaum, C. L. Masters, K. H. Grzeschik, G. Multhaup, K. Beyreuther, and B. Muller-Hill. 1987. The precursor of Alzheimer's disease amyloid A4 protein resembles a cell-surface receptor. Nature. 325:733–736.[Medline]

Kim, J. R., T. J. Gibson, and R. M. Murphy. 2003. Targeted control of kinetics of ß-amyloid self-association by surface tension-modifying peptides. J. Biol. Chem. 278:40730–40735.[Abstract/Free Full Text]

Kirkitadze, M. D., G. Bitan, and D. B. Teplow. 2002. Paradigm shifts in Alzheimer's disease and other neurodegenerative disorders: the emerging role of oligomeric assemblies. J. Neurosci. Res. 69:567–577.[Medline]

Kirschner, D. A., H. Inouye, L. K. Duffy, A. Sinclair, M. Lind, and D. J. Selkoe. 1987. Synthetic peptide homologous to ß protein from Alzheimer disease forms amyloid-like fibrils in vitro. Proc. Natl. Acad. Sci. USA. 84:6953–6957.[Abstract/Free Full Text]

Klunk, W. E., M. L. Debnath, A. M. Koros, and J. W. Pettegrew. 1998. Chrysamine-G, a lipophilic analogue of Congo red, inhibits Aß-induced toxicity in PC12 cells. Life Sci. 63:1807–1814.[Medline]

Kremer, J. J., M. M. Pallitto, D. J. Sklansky, and R. M. Murphy. 2000. Correlation of ß-amyloid aggregate size and hydrophobicity with decreased bilayer fluidity of model membranes. Biochemistry. 39:10309–10318.[Medline]

Lambert, M. P., A. K. Barlow, B. A. Chromy, C. Edwards, R. Freed, M. Liosatos, T. E. Morgan, I. Rozovsky, B. Trommer, K. L. Viola, P. Wals, C. Zhang, C. E. Finch, G. A. Krafft, and W. L. Klein. 1998. Diffusible, nonfibrillar ligands derived from Aß1–42 are potent central nervous system neurotoxins. Proc. Natl. Acad. Sci. USA. 95:6448–6453.[Abstract/Free Full Text]

Lowe, T. L., A. Strzelec, L. L. Kiessling, and R. M. Murphy. 2001. Structure-function relationships for inhibitors of ß-amyloid toxicity containing the recognition sequence KLVFF. Biochemistry. 40:7882–7889.[Medline]

Manavalan, P., and W. C. Johnson. 1987. Variable selection method improves the prediction of protein secondary structure from circular-dichroism spectra. Anal. Biochem. 167:76–85.[Medline]

Masters, C. L., G. Simms, N. A. Weinman, G. Multhaup, B. L. McDonald, and K. Beyreuther. 1985. Amyloid plaque core proteins in Alzheimer disease and Down syndrome. Proc. Natl. Acad. Sci. USA. 82:4245–4249.[Abstract/Free Full Text]

McLaurin, J., D.-S. Yang, C. M. Yip, and P. E. Fraser. 2000. Review: modulating factors in amlyoid-ß fibril formation. J. Struct. Biol. 130:259–270.[Medline]

Moran, P. M., L. S. Higgins, B. Cordell, and P. C. Moser. 1995. Age-related learning deficits in transgenic mice expressing the 751-amino acid isoform of human ß-amyloid precursor protein. Proc. Natl. Acad. Sci. USA. 92:5341–5345.[Abstract/Free Full Text]

Moss, M. A., M. R. Nichols, D. K. Reed, J. H. Hoh, and T. L. Rosenberry. 2003. The peptide KLVFF-K(6) promotes ß-amyloid(1–40) protofibril growth by association but does not alter protofibril effects on cellular reduction of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT). Mol. Pharmacol. 64:1160–1168.[Abstract/Free Full Text]

Murphy, R. M., and M. M. Pallitto. 2000. Probing the kinetics of ß-amyloid self-association. J. Struct. Biol. 130:109–122.[Medline]

Nichols, M. R., M. A. Moss, D. K. Reed, W. L. Lin, R. Mukhopadhyay, J. H. Hoh, and T. L. Rosenberry. 2002. Growth of ß-amyloid(1–40) protofibrils by monomer elongation and lateral association. Characterization of distinct products by light scattering and atomic force microscopy. Biochemistry. 41:6115–6127.[Medline]

Pallitto, M. M., J. Ghanta, P. Heinzelman, L. L. Kiessling, and R. M. Murphy. 1999. Recognition sequence design for peptidyl modulators of ß-amyloid aggregation and toxicity. Biochemistry. 38:3570–3578.[Medline]

Pallitto, M. M., and R. M. Murphy. 2001. A mathematical model of the kinetics of ß-amyloid fibril growth from the denatured state. Biophys. J. 81:1805–1822. (Errata published in Biophys. J. 82:2826.)[Abstract/Free Full Text]

Pappolla, M., P. Bozner, C. Soto, H. Shao, N. K. Robakis, M. Zagorski, B. Frangione, and J. Ghiso. 1998. Inhibition of Alzheimer ß-fibrillogenesis by melatonin. J. Biol. Chem. 273:7185–7188.[Abstract/Free Full Text]

Pike, C. J., D. Burdick, A. J. Walencewicz, C. G. Glabe, and C. W. Cotman. 1993. Neurodegeneration induced by ß-amyloid peptides in vitro: the role of peptide assembly state. J. Neurosci. 13:1676–1687.[Abstract]

Provencher, S. W., and J. Glöckner. 1981. Estimation of globular protein secondary structure from circular dichroism. Biochemistry. 20:33–37.[Medline]

Roher, A. E., M. O. Chaney, Y. M. Kuo, S. D. Webster, W. B. Stine, L. J. Haverkamp, A. S. Woods, R. J. Cotter, J. M. Tuohy, G. A. Krafft, B. S. Bonnel, and M. R. Emmerling. 1996. Morphology and toxicity of Aß-(1–42) dimer derived from neuritic and vascular amyloid deposits of Alzheimer's disease. J. Biol. Chem. 271:20631–20635.[Abstract/Free Full Text]

Selkoe, D. J. 1991. The molecular pathology of Alzheimer's disease. Neuron. 6:487–498.[Medline]

Selkoe, D. J. 1993. Physiological production of the ß-amyloid protein and the mechanism of Alzheimer's disease. Trends Neurosci. 16:403–409.[Medline]

Serpell, L. C. 2000. Alzheimer's amyloid fibrils: structure and assembly. Biochim. Biophys. Acta. 1502:16–30.[Medline]

Shen, C. L., G. L. Scott, F. Merchant, and R. M. Murphy. 1993. Light scattering analysis of fibril growth from the amino-terminal fragment ß(1–28) of ß-amyloid peptide. Biophys. J. 65:2383–2395.[Abstract/Free Full Text]

Shen, C. L., and R. M. Murphy. 1995. Solvent effects on self-assembly of ß-amyloid peptide. Biophys. J. 69:640–651.[Abstract/Free Full Text]

Simmons, L. K., P. C. May, K. J. Tomaselli, R. E. Rydel, K. S. Fuson, E. F. Brigham, S. Wright, I. Lieberburg, G. W. Becker, D. N. Brems, and W. Y. Li. 1994. Secondary structure of amyloid ß peptide correlates with neurotoxic activity in vitro. Mol. Pharmacol. 45:373–379.[Abstract]

Tomiyama, T., S. Asano, Y. Suwa, T. Morita, K. Kataoka, H. Mori, and N. Endo. 1994. Rifampicin prevents the aggregation and neurotoxicity of amyloid ß protein in vitro. Biochem. Biophys. Res. Commun. 204:76–83.[Medline]

Tycko, R. 2003. Insights into the amyloid folding problem from solid-state NMR. Biochemistry. 42:3151–3159.[Medline]

Yang, D.-S., C. M. Yip, T. H. Huang, A. Chakrabartty, and P. E. Fraser. 1999. Manipulating the amyloid-ß aggregation pathway with chemical chaperones. J. Biol. Chem. 274:32970–32974.[Abstract/Free Full Text]

Yip, C. M., A. A. Darabie, and J. McLaurin. 2002. Aß 42-peptide assembly on lipid bilayers. J. Mol. Biol. 318:97–107.[Medline]





This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
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 arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kim, J. R.
Right arrow Articles by Murphy, R. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kim, J. R.
Right arrow Articles by Murphy, R. M.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Copyright © 2004 by the Biophysical Society.