Limits...
Fibril growth kinetics reveal a region of beta2-microglobulin important for nucleation and elongation of aggregation.

Platt GW, Routledge KE, Homans SW, Radford SE - J. Mol. Biol. (2008)

Bottom Line: We describe the kinetics of seeded and spontaneous (unseeded) fibril growth of wild-type beta(2)m and 12 variants at pH 2.5, targeting specifically an aromatic-rich region of the polypeptide chain (residues 62-70) that has been predicted to be highly amyloidogenic.The results reveal the importance of aromatic residues in this part of the beta(2)m sequence in fibril formation under the conditions explored and show that this region of the polypeptide chain is involved in both the nucleation and the elongation phases of fibril formation.No direct correlation was observed, however, between the extent of non-random structure in the unfolded state and the rates of fibril nucleation and elongation, suggesting that the early stages of aggregation involve significant conformational changes from the initial unfolded state.

View Article: PubMed Central - PubMed

Affiliation: Astbury Centre for Structural Molecular Biology and Institute of Molecular and Cellular Biology, University of Leeds, Leeds LS2 9JT, UK.

ABSTRACT
Amyloid is a highly ordered form of aggregate comprising long, straight and unbranched proteinaceous fibrils that are formed with characteristic nucleation-dependent kinetics in vitro. Currently, the structural molecular mechanism of fibril nucleation and elongation is poorly understood. Here, we investigate the role of the sequence and structure of the initial monomeric precursor in determining the rates of nucleation and elongation of human beta(2)-microglobulin (beta(2)m). We describe the kinetics of seeded and spontaneous (unseeded) fibril growth of wild-type beta(2)m and 12 variants at pH 2.5, targeting specifically an aromatic-rich region of the polypeptide chain (residues 62-70) that has been predicted to be highly amyloidogenic. The results reveal the importance of aromatic residues in this part of the beta(2)m sequence in fibril formation under the conditions explored and show that this region of the polypeptide chain is involved in both the nucleation and the elongation phases of fibril formation. Structural analysis of the conformational properties of the unfolded monomer for each variant using NMR relaxation methods revealed that all variants contain significant non-random structure involving two hydrophobic clusters comprising regions 29-51 and 58-79, the extent of which is critically dependent on the sequence. No direct correlation was observed, however, between the extent of non-random structure in the unfolded state and the rates of fibril nucleation and elongation, suggesting that the early stages of aggregation involve significant conformational changes from the initial unfolded state. Together, the data suggest a model for beta(2)m amyloid formation in which structurally specific interactions involving the highly hydrophobic and aromatic-rich region comprising residues 62-70 provide a complementary interface that is key to the generation of amyloid fibrils for this protein at acidic pH.

Show MeSH

Related in: MedlinePlus

Comparison of fibril formation rates with structural properties of the initial unfolded state. (a) Relative apparent elongation rates (taken from 1/t[50%] value) determined using seeded growth assays normalised to the data for wild-type β2m. The fastest rates are shown in red, intermediate rates are shown in green and the slowest rates are indicated in blue. (b) Normalised 1/lag times determined using spontaneous fibril growth experiments, coloured as in (a). (*) Not all of the reactions containing L65R resulted in fibril formation over the 75-h period studied. In these cases, the end point of the experiment (75 h) was taken as the lag time. The data shown, therefore, are an underestimate of the true lag time. (c) Rates of T2 relaxation for the hydrophobic cluster involving residues 62–70 calculated using an average of the values obtained for residues 61, 68, 73 and 75 for each variant. The bars are coloured according to the scheme used in (a) and (b).
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC2627305&req=5

fig5: Comparison of fibril formation rates with structural properties of the initial unfolded state. (a) Relative apparent elongation rates (taken from 1/t[50%] value) determined using seeded growth assays normalised to the data for wild-type β2m. The fastest rates are shown in red, intermediate rates are shown in green and the slowest rates are indicated in blue. (b) Normalised 1/lag times determined using spontaneous fibril growth experiments, coloured as in (a). (*) Not all of the reactions containing L65R resulted in fibril formation over the 75-h period studied. In these cases, the end point of the experiment (75 h) was taken as the lag time. The data shown, therefore, are an underestimate of the true lag time. (c) Rates of T2 relaxation for the hydrophobic cluster involving residues 62–70 calculated using an average of the values obtained for residues 61, 68, 73 and 75 for each variant. The bars are coloured according to the scheme used in (a) and (b).

Mentions: The structure in the acid-unfolded state of β2m (reflected by the R2 value for the 62–70 cluster) and the apparent rates of fibril nucleation and elongation are compared in Fig. 5. Whilst the relationship between these parameters is complex, qualitatively clear features emerge. Thus, all of the variants that have an average R2 value in the cluster 62–70 similar to that of wild-type β2m result in nucleation and apparent elongation rates akin to those of the wild-type protein (Fig. 5, red bars). For variants with reduced R2 values in this cluster, however, no clear correlation is observed between the R2, the lag time and the apparent elongation rate. For instance, Y66E and F70A display similar relaxation rates; however, the apparent rate of fibril elongation is far greater for Y66E. Similarly, F62A has one of the slowest rates of elongation and nucleation, but this substitution has only a moderate effect on the R2 values. Finally, whilst L65R has a dramatic effect on the ability of β2m to form amyloid fibrils, the R2 of this variant is only moderately perturbed. Similar results have been obtained previously for other proteins, wherein aggregation has been shown to be favoured by solution conditions that promote stable intermolecular interactions rather than inducing specific conformational properties in a partially unfolded state.35,36 Whilst the retention of a hydrophobic cluster in the region 62–70 in β2m is commensurate with rapid fibril growth, the precise effects of an amino acid substitution thus appear to depend critically on the type of substitution made and the location of the amino acid substituted, suggesting that the context of each amino acid in the polypeptide sequence and in the structure of the early aggregating species plays a critical role in determining the aggregation potential of each variant. Consistent with this conclusion, algorithms developed to predict aggregation potential or elongation rates, assuming that self-assembly involves an unstructured precursor in which every amino acid contributes equally and is dependent only upon its physicochemical properties, result in a poor prediction of the aggregation rates of the 12 variants created here (data not shown).5,29,30


Fibril growth kinetics reveal a region of beta2-microglobulin important for nucleation and elongation of aggregation.

Platt GW, Routledge KE, Homans SW, Radford SE - J. Mol. Biol. (2008)

Comparison of fibril formation rates with structural properties of the initial unfolded state. (a) Relative apparent elongation rates (taken from 1/t[50%] value) determined using seeded growth assays normalised to the data for wild-type β2m. The fastest rates are shown in red, intermediate rates are shown in green and the slowest rates are indicated in blue. (b) Normalised 1/lag times determined using spontaneous fibril growth experiments, coloured as in (a). (*) Not all of the reactions containing L65R resulted in fibril formation over the 75-h period studied. In these cases, the end point of the experiment (75 h) was taken as the lag time. The data shown, therefore, are an underestimate of the true lag time. (c) Rates of T2 relaxation for the hydrophobic cluster involving residues 62–70 calculated using an average of the values obtained for residues 61, 68, 73 and 75 for each variant. The bars are coloured according to the scheme used in (a) and (b).
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC2627305&req=5

fig5: Comparison of fibril formation rates with structural properties of the initial unfolded state. (a) Relative apparent elongation rates (taken from 1/t[50%] value) determined using seeded growth assays normalised to the data for wild-type β2m. The fastest rates are shown in red, intermediate rates are shown in green and the slowest rates are indicated in blue. (b) Normalised 1/lag times determined using spontaneous fibril growth experiments, coloured as in (a). (*) Not all of the reactions containing L65R resulted in fibril formation over the 75-h period studied. In these cases, the end point of the experiment (75 h) was taken as the lag time. The data shown, therefore, are an underestimate of the true lag time. (c) Rates of T2 relaxation for the hydrophobic cluster involving residues 62–70 calculated using an average of the values obtained for residues 61, 68, 73 and 75 for each variant. The bars are coloured according to the scheme used in (a) and (b).
Mentions: The structure in the acid-unfolded state of β2m (reflected by the R2 value for the 62–70 cluster) and the apparent rates of fibril nucleation and elongation are compared in Fig. 5. Whilst the relationship between these parameters is complex, qualitatively clear features emerge. Thus, all of the variants that have an average R2 value in the cluster 62–70 similar to that of wild-type β2m result in nucleation and apparent elongation rates akin to those of the wild-type protein (Fig. 5, red bars). For variants with reduced R2 values in this cluster, however, no clear correlation is observed between the R2, the lag time and the apparent elongation rate. For instance, Y66E and F70A display similar relaxation rates; however, the apparent rate of fibril elongation is far greater for Y66E. Similarly, F62A has one of the slowest rates of elongation and nucleation, but this substitution has only a moderate effect on the R2 values. Finally, whilst L65R has a dramatic effect on the ability of β2m to form amyloid fibrils, the R2 of this variant is only moderately perturbed. Similar results have been obtained previously for other proteins, wherein aggregation has been shown to be favoured by solution conditions that promote stable intermolecular interactions rather than inducing specific conformational properties in a partially unfolded state.35,36 Whilst the retention of a hydrophobic cluster in the region 62–70 in β2m is commensurate with rapid fibril growth, the precise effects of an amino acid substitution thus appear to depend critically on the type of substitution made and the location of the amino acid substituted, suggesting that the context of each amino acid in the polypeptide sequence and in the structure of the early aggregating species plays a critical role in determining the aggregation potential of each variant. Consistent with this conclusion, algorithms developed to predict aggregation potential or elongation rates, assuming that self-assembly involves an unstructured precursor in which every amino acid contributes equally and is dependent only upon its physicochemical properties, result in a poor prediction of the aggregation rates of the 12 variants created here (data not shown).5,29,30

Bottom Line: We describe the kinetics of seeded and spontaneous (unseeded) fibril growth of wild-type beta(2)m and 12 variants at pH 2.5, targeting specifically an aromatic-rich region of the polypeptide chain (residues 62-70) that has been predicted to be highly amyloidogenic.The results reveal the importance of aromatic residues in this part of the beta(2)m sequence in fibril formation under the conditions explored and show that this region of the polypeptide chain is involved in both the nucleation and the elongation phases of fibril formation.No direct correlation was observed, however, between the extent of non-random structure in the unfolded state and the rates of fibril nucleation and elongation, suggesting that the early stages of aggregation involve significant conformational changes from the initial unfolded state.

View Article: PubMed Central - PubMed

Affiliation: Astbury Centre for Structural Molecular Biology and Institute of Molecular and Cellular Biology, University of Leeds, Leeds LS2 9JT, UK.

ABSTRACT
Amyloid is a highly ordered form of aggregate comprising long, straight and unbranched proteinaceous fibrils that are formed with characteristic nucleation-dependent kinetics in vitro. Currently, the structural molecular mechanism of fibril nucleation and elongation is poorly understood. Here, we investigate the role of the sequence and structure of the initial monomeric precursor in determining the rates of nucleation and elongation of human beta(2)-microglobulin (beta(2)m). We describe the kinetics of seeded and spontaneous (unseeded) fibril growth of wild-type beta(2)m and 12 variants at pH 2.5, targeting specifically an aromatic-rich region of the polypeptide chain (residues 62-70) that has been predicted to be highly amyloidogenic. The results reveal the importance of aromatic residues in this part of the beta(2)m sequence in fibril formation under the conditions explored and show that this region of the polypeptide chain is involved in both the nucleation and the elongation phases of fibril formation. Structural analysis of the conformational properties of the unfolded monomer for each variant using NMR relaxation methods revealed that all variants contain significant non-random structure involving two hydrophobic clusters comprising regions 29-51 and 58-79, the extent of which is critically dependent on the sequence. No direct correlation was observed, however, between the extent of non-random structure in the unfolded state and the rates of fibril nucleation and elongation, suggesting that the early stages of aggregation involve significant conformational changes from the initial unfolded state. Together, the data suggest a model for beta(2)m amyloid formation in which structurally specific interactions involving the highly hydrophobic and aromatic-rich region comprising residues 62-70 provide a complementary interface that is key to the generation of amyloid fibrils for this protein at acidic pH.

Show MeSH
Related in: MedlinePlus