Aggregation Properties Of A Short Peptide Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Short peptides have been identified from amyloidogenic proteins that form amyloid fibrils in isolation. These peptides are believed to play crucial role in the self-assembly of full length proteins by templating, facilitating, or stabilizing the amyloid fold. The hexapeptide stretch, 21DIDLHL26, has been shown to be important in the self-assembly of PI3-SH3 domain. SH3 domain of chicken brain α-spectrin, which is otherwise non-amyloidogenic, is rendered amyloidogenic if 22EVTMKK27 is replaced by DIDLHL. The fact, that replacing the residues 25-26 in PI3-SH3 by the consensus KK sequence makes it non-amyloidogenic without affecting its stability, make DIDLHL an interesting sequence to study in isolation. Studies with the C-terminal acid and C-terminal amidated forms of the peptide suggest that the peptide has the propensity to aggregate into spherical and fibrillar structures at pH 5 and 6. The aggregates are unstable and are easily modulated by the presence of mica and salt. This study suggests that the peptide need not be amyloidogenic, in itself, to facilitate the self-assembly of the full length protein. The propensity to form non-amyloid structures appear to be important in potentiating the self-assembly of full length protein into amyloid fibrils.


Peptides with highly variable lengths and amino acid sequences have been observed to have the ability to form amyloid fibrils (1-10). While many of these peptides are segments of proteins that form fibrils in vivo and are related to pathogenesis (1, 4, 5, 6, 8, 11), others are not related to pathogenesis (2, 9, 10). The conditions, under which peptides form fibrils, are highly variable with respect to pH, ionic strength, temperature, aggregation times, as well as the presence of organic solvents (6, 8, 12, 13, 14). Amyloid formation of peptides has also been observed on solid substrates such as mica or quartz (15, 16).

The amyloid forming ability of peptides, that are segments of amyloid forming proteins, has been the subject of extensive investigation (8, 17, 18, 19, 20). It appears that the peptide segments in isolation do have the ability to form fibrils which are very similar to the fibrils formed by the parent proteins. These peptides, in isolation, appear to have the conformational features of β-structure observed in protein fibrils (8, 21, 22, 23, 24).

The SH3 domain of p85α subunit of bovine phosphatidylinositol-3-kinase (PI3-SH3) forms amyloid fibrils at acidic pH (pH 2.0) while SH3 domain of chicken brain -spectrin (SPC-SH3), which shares same fold and 24% sequence identity with PI3-SH3, does not form fibrils under these conditions (25, 26). At pH 2.0, PI3-SH3 domain is partially denatured following which the domain self-assembles into amyloid-like fibrils. Presence of extensive native-like interactions in SPC-SH3 under acidic pH, as suggested by solution NMR studies (27), might account for its inability to form amyloid fibrils under these conditions. Replacing a six amino acid stretch (22EVTMKK27) in SPC-SH3 by PI3-SH3 hexapeptide, 21DIDLHL26, comprising of residues in the diverging turn and adjacent RT loop, confers amyloidogenicity to SPC-SH3 (26). Replacement of residues 25 and 26 in PI3-SH3, by the consensus KK, renders the domain non-amyloidogenic, indicating the importance of the DIDLHL sequence in conferring amyloidogenicity to the domain. Interestingly, both amyloidogenic and non-amyloidogenic forms of PI3-SH3 are denatured to similar extent at pH 2.0. This suggests that the sequence DIDLHL does not affect the stability of SH3 domain in conferring amyloidogenicity to it. It would be interesting to study the self-assembly of this hexapeptide stretch in isolation. In this study, we have investigated the aggregation properties of NH2DIDLHLCONH2 (D-am) and NH2DIDLHLCOOH (D-ac) under various conditions.


Fmoc amino acids were purchased from Novabiochem AG (Switzerland) and Advanced ChemTech (Louisville, KY). Peptide synthesis resins, HMP-Resin (p-Hydroxymethylphenoxymethyl polystyrene resin) and PAL resin (5-(4-Aminomethyl-3,5-dimethoxyphenoxy)valeric acid resin) were purchased from Applied Biosystems (Foster City, CA) and Advanced ChemTech (Louisville, KY), respectively. Thioflavin T was purchased from Sigma.

Peptide Synthesis

The peptides, DIDLHLam (D-am), DIDLHLac (D-ac) and AcVQIVYKam (AcPHF6) were synthesized using standard Fmoc chemistry (28). The synthesized peptides were cleaved from the resin and deprotected using a mixture containing 82.5% TFA, 5% phenol, 5% H2O, 5% thioanisole, and 2.5% ethanedithiol for 12-15 hours at room temperature (29). The peptides were precipitated in ice-cold diethyl ether. The peptides were dissolved in deionized water and purified on Hewlett Packard 1100 series HPLC instrument on a reversed-phase C18 Bio-Rad column using a linear gradient of acetonitrile. 0.1% TFA was used for ion pairing. Purified peptides were characterized using matrix-assisted laser desorption ionization-time-of-flight (MALDI-TOF) mass spectrometry on a Voyager DE STR mass spectrometer (PerSeptive Biosystems, Foster city, CA). The spectra of the peptides showed m/z values of 724.26 for D-am (calculated mass: 723.80 Da), 725.24 for D-ac (calculated mass: 724.80 Da), and 790.56 for AcPHF6 (calculated mass: 789.93 Da). Stock solutions of the peptides were prepared in deionized water and concentrations were determined using Waddell's method of protein concentration estimation (30, 31) for D-am and D-ac while AcPHF6 concentrations were calculated using a molar absorption coefficient of 1280 M-1cm-1 at λ = 280 nm.

Aggregation Reactions

Aggregation reactions with 100µM peptide concentration (DIDLHL) were set up in 10 mM buffers of pH values 2, 3, 4, 5, 6, and 7 and kept at 37 oC for 5 days.

Further aggregation reactions (Peptide concentration ≥ 200 μM) were set up for both the free acid and amidated form of the hexapeptide in "10 mM acetate buffer, pH 5" and "10mM phosphate buffer, pH 6" with/without 100 mM NaCl and with/without freshly cleaved mica piece at 37 oC for different time periods. Aggregation was studied using Atomic Force Microscopy.

Fibril growth reactions of AcPHF6 were set up incubating 200 μM solution in 20 mM MOPS buffer, pH 7.2 + 150 mM NaCl at 37 oC for one week. The presence of amyloid fibrils was confirmed using ThT fluorescence and Atomic Force Microscopy.

Atomic Force Microscopy

Aggregation reaction samples were diluted in deionized water and immediately deposited on freshly cleaved surface of mica sheets ( 1 µg peptide deposited) and allowed to dry in air. Images were acquired using tapping mode AFM (Multimode, Digital Instruments, Santa Barbara, CA). A silicon nitride probe was oscillated at 275-310 KHz and images were collected at an optimized scan rate. Analysis was done using Nanoscope ® III 5.30 r1.

Thioflavin T fluorescence

Thioflavin T fluorescence assays were performed using a modification of the method described by Naiki et al (32). Peptide solutions were diluted in "10 μM ThT in 50 mM phosphate buffer, pH 7.0" to a final concentration of 10 μg/ml. Fluorescence spectra were recorded on Fluorolog-3 Model FL3-22 spectrofluorometer (Horiba Jobin Yvon, Park Avenue Edison, NJ). The excitation wavelength was set at 450 nm, slit width at 2 nm, and emission slit width at 5 nm.

Circular Dichroism

Circular dichroism (CD) spectra were recorded on Jasco J-715 spectropolarimeter. Far-UV (195-250 nm) spectra were recorded in "10 mM actetate buffer, pH 5" and "10 mM phosphate buffer, pH 6" with/without 100 mM NaCl, and in trifluoroethanol (TFE) in 0.1 cm path length cell using a step size of 0.2 nm, band width of 1 nm, and scan rate of 100 nm/min. Peptide concentrations were 60 µM for CD spectra in TFE and 120 µM for spectra recorded in acetate and phosphate buffers. The spectra were recorded by averaging 10 scans and corrected by subtracting the solvent/buffer spectra.


The peptide segment DIDLHL has been shown to promote the formation of amyloid fibrils in PI3-SH3 and α-spectrin-SH3 domains. Two peptides, DIDLHL-ac (D-ac) and DIDLHL-am (D-am) were examined for their ability to form aggregates under different conditions such as pH, salt concentration, and in the presence of mica. The aggregates were studied using AFM and the arrows and boxes shown in the AFM images are referred to in Table 1.

Preliminary aggregation reactions set up at pH 2, 3, 4, 5, 6, and 7 suggested ordered structures only for the samples prepared at pH 5 and 6, when observed using AFM. The images shown in Figure 1 were recorded for 100 μM peptides at pH 5 and 6. In the case of D-am, a fine mesh network is seen at pH 5 with filaments of ~ 3 nm height (panel A). The network is less evident at pH 6.0 (panel B). The morphology of D-ac aggregates is considerably different. Although long fibrils are observed (indicated by arrows), majority of the aggregates are spherical in nature (panels C and D).

The aggregates of D-ac at different pH values, in the presence of salt and mica were examined (Figure 2). At pH 6, largely spherical aggregates are observed (heights ~ 1 - 6 nm). Few linear fibrils (~ 2.2 - 3 nm, panels B and C), alone and in association with spherical aggregates (indicated by arrows) are also observed. In one case (panel D), fractal shaped structures are also seen (~ 3 - 5 nm in thickness). In the presence of mica at pH 6, only spherical aggregates (~ 2 - 12 nm) are observed (panels E and F). At pH 5 also, in the presence of 100 mM NaCl, only spherical aggregates (~ 1.5-15 nm) are observed; presence of mica does not cause significant change in the aggregates (data not shown).

The aggregates formed by D-am under different conditions are shown in Fig. 3 - 6. At pH 5, D-am forms spherical aggregates (Figure 3) while in the presence of mica at pH 5 (Figure 4), widely differing morphologies are observed. In panels A-D, thin films are formed over which more films (Panels A and C) and spherical aggregates (Panels B and C) are present. Thickness of the films in panel A is ~ 0.5 - 1 nm, films in panels B and C are ~ 1.5 - 3 nm while those in panel D are even thicker than 3 nm. Panel E shows presence of small fibrillar structures (2 - 6 nm in thickness). In panel F, a mixture of very small aggregates, thin fibrils (< 1 - 3 nm), and thicker fibrillar structures (3 - 6 nm, indicated by black arrows) is observed. Panel G shows very small (2 nm) to fairly large (35 nm) spherical particles arranged in a pattern. When NaCl is also present along with mica, a variety of aggregates are observed (Figure 5). Fractal-like pattern (panel A) and short fibrillar aggregates (panels B-O) are observed. Fibrils are observed more often as compared to the samples without salt. The dimensions of the fibers vary considerably (~ 1 - 10 nm). Panels D and H show the fibers which appear to vary in diameter from very thin to those typical of amyloid protofilaments and amyloid fibrils (~ 1 - 10 nm) while the aggregates in panels F, G, I, and J are similar to amyloid protofibrils in dimensions (heights ≤ 3.5 nm, lengths < 1 μm). Panels K - P show presence of thin films also (~ 1 - 6 nm, details presented in Table 1).

The D-am aggregates obtained at pH 6 are shown in Figure 6. The aggregates appear to be entirely different from those obtained at pH 5. Fibril formation was no longer discernible. Individual spheres (Panels A, B, and E), very small to very large (~ 3 - 50 nm) or large oval egg-shaped structures (Panels C and D) were observed. Peptide films (panels F and H) and fractal-like aggregates were also observed (panel G). In some cases, the spheres arrayed in rod like structures (panels I - K), ~ 1.5 - 5 nm thick, were seen which can align to give sheet like appearances (Panel K). The organized structures were absent when either mica or NaCl was present during aggregation process (Figure 7). Mostly spherical aggregates are observed with very few small fibrils or peptide films. The heights of the aggregates are tabulated in Table 1. The morphologies are clearly different from the aggregates formed by the amyloidogenic peptide, AcPHF6 which forms highly ordered fibrils (Figure 8A) that cause increase in ThT fluorescence (Figure 8B) confirming the amyloid nature of these fibrils. While aggregates with different morphologies were detected for D-am and D-ac by AFM, none of them caused any increase in ThT fluorescence indicating that the peptide sequences do not form mature amyloid like fibrils under the conditions studied.

The CD spectra of D-ac and D-am at pH 5, 6, and in TFE are shown in Fig. 9. Although salt makes an appreciable difference in the morphology of the aggregates, the CD spectra show a minimum ~ 200 nm with crossover at < 195 nm. The spectra indicate that a large fraction of the molecules populate unordered conformation. Even in TFE, the peptides are largely unordered.

In vitro, proteins that form fibrils under disease conditions as well as peptides with widely differing sequences form fibrils and other aggregated structures with highly variable morphologies (7, 10, 33, 34, 35, 36, 37, 38, 39). Even fibrils produced under single growth conditions displayed considerable variation in their morphologies (34, 40). In the case of insulin and Aβ, spherical structures appear to precede fibril formation (36, 40, 41).

Thus, the process of mature fibril formation appears to proceed via several intermediate aggregates which are not necessarily fibrillar. In fact, the initial structure need not even be a β-structure (5, 9, 42). The most amyloidogenic peptide derived from the Aβ sequence has a polyproline II like structure (43).

Our results with DIDLHL peptides indicate the propensities to form spherical aggregates as well as filamentous structures. However, we have not been able to observe the transformation of these soluble aggregates into insoluble amyloid fibril structures. Under the conditions of low pH, at which PI3-SH3 forms fibrils, it is likely that this sequence facilitates the formation of fibrils with other region of the protein acting as facilitator (5, 44). Distal peptide sequences have been shown to facilitate amyloid formation in proteins such as β2- microglobulin (45). The ability to potentiate the formation of aggregates could be the reason for the induction of aggregates in α-spectrin-SH3 domain which, on its own, does not form fibrils (27). It has been observed that a single mutation can cause α-spectrin-SH3 to form fibrillar aggregates indicating that the protein could be prone to destabilization resulting in fibril formation (46). However, replacement of 25HL26 in PI3-SH3 by the consensus KK sequence does not affect it's stability but renders it non-amyloidogenic. This suggests that the sequence, DIDLHL might not be required for the stability of the SH3 domain but is important for conferring amyloidogenicity to it, further suggesting that the stretch might act as a facilitator in amyloid formation by interacting with other aggregation-prone stretches present in the domain.

The formation of fibrils by only certain short segments of amyloid forming proteins highlights the role of such sequences in the formation of fibrils by the mature protein (8, 45, 46). In several cases, changes in the sequence of the peptides abrogate fibril formation suggesting that the requirement may be stringent (4). Our results with DIDLHL suggest that it may not be necessary in all cases for the peptide presumed to play a role in amyloid formation to form mature fibrils in isolation. The ability to aggregate particularly to form aggregates of various morphologies could be sufficient to induce fibril formation in the protein.