Mechanisms of Functional Amyloid Formation

Pmel17原纤维是人类皮肤和眼睛中黑色素沉积所需的结构性支架。黑色素是在黑素小体中合成的，黑素小体是与内小体和溶酶体相关的细胞器，黑色素储存在黑素细胞中，黑素细胞负责着色。尽管黑素小体的成熟过程包括四个不同的阶段，并在超微结构水平上用透射电子显微镜(TEM)进行了详细的表征，但在每个阶段中，腔内PMel17纤维的分子性质尚不清楚。此外，PMel17细丝的淀粉样蛋白核心部分或全部由哪个多肽结构域组成还有待确定。 1.RPT淀粉样蛋白的形成依赖于pH 我们研究了重复结构域(RPT，残基315-444)作为一个模型系统，从可溶性和非结构化单体到聚集的含有β-折叠的纤维的构象变化。RPT主要氨基酸序列由10个不完全的13个残基重复组成，富含Pro、Ser、Thr和Glu。RPT含有16种羧酸，强调了它容易经历pH诱导的构象变化。由于pH和蛋白质结构在体内是联系在一起的，我们详细研究了RPT的局部和宏观构象随pH的变化。 由于Trp发射对溶剂极性、局部构象变化和蛋白质-蛋白质相互作用高度敏感，我们利用唯一的内在W423作为淀粉样蛋白结构和聚集动力学的位点特异性荧光探针。确定了纤维形成的临界pH范围(4.5到5.5)，这表明至少有三个羧酸参与了聚集所需的结构重排。W423在RPT聚集过程中的高响应性指向C-末端区域在纤维组装中的关键作用。为了研究黑素小体pH的变化与RPT纤维形成之间的直接关系，进行了pH滴定分析。将预制的RPT聚集体(pH 4.0)滴定至pH 7.0。在pH值为5.0时，高度弯曲的小聚集体变成长条纹状的纤维，这让人想起在第一阶段和第二阶段观察到的黑素小体的原纤维转变。进一步中和到pH 7.0时，RPT纤维完全解体。这种独特的聚集/解聚过程与疾病相关的淀粉样蛋白形成了鲜明对比，后者以抵抗最严厉的治疗而臭名昭著。 我们提出，在高度酸性的黑素小体中(阶段I)，只有在隔室溶液达到最佳pH值5(阶段II)后，蛋白质才会随着纤维的伸长而开始聚集。当特定的Glu残基质子化时，多肽链内的静电电荷排斥力减少，从而导致形成紧凑的结构，促进纤维形成所需的关键相互作用。此外，我们观察到RPT将很容易聚合(约在优化的pH(5.0)下，可能只稳定了原纤维，而不是潜在的有毒低聚物。虽然我们的数据显示，在第三阶段和第四阶段的黑素小体中发现的近中性条件下，纤维会溶解，但尚不清楚在黑色素沉积时，聚合物材料是否可以将纤维从溶液中隔离出来，从而保护它们免受溶解。然而，如果被释放并暴露在黑素小体外的中性环境中，纤维将很容易分解，从而保持其良性性质。 2.探测RPT原纤解离 为了研究纤维的解离，我们分别使用原子力显微镜(AFM)和核磁共振(核磁共振)光谱作为超微结构和分子探针。具体地说，我们询问与疾病相关的淀粉样蛋白相关的中间体在溶解过程中是否被绕过。 为了监测纤维的分解，将预制的RPT淀粉样蛋白沉积在云母上，并在潮湿的缓冲条件下用AFM观察。在pH为5.0时，观察到长的、直的、无分支的纤维，这与透射电子显微镜所见的相似。用pH 6.5缓冲液洗涤这些原纤维后，这些原纤维开始溶解。实时监测显示，大纤维碎裂后，较小的碎片在几分钟内完全消失。 为了获得残基特异性的洞察力，制备了用于核磁共振光谱的同位素标记的纤维RPT。在pH 5.0时，没有观察到残基378-444的主链酰胺共振，表明该区域含有淀粉样蛋白核心。这与我们的断言一致，即C-末端区域对纤维形成是重要的。溶解动力学和光谱数据表明，没有证据表明有稳定的中间体。大小排阻层析也证实了中间体的存在。此外，单个Glu主链酰胺的共振显示出类似的动力学趋势，这表明原纤维的展开是一个涉及许多去质子化事件的全球性事件。

Pmel17 fibrils serve as the structural scaffolding required for melanin deposition in human skin and eyes. Melanin is synthesized in melanosomes, organelles related to both endosomes and lysosomes, and stored in melanocytes, cells responsible for pigmentation. While the melanosome maturation process has been shown to involve four distinct stages that have been characterized in detail at the ultrastructural level by transmission electron microscopy (TEM), the molecular nature of the intralumenal Pmel17 fibrils during each of these stages is not known. Moreover, which polypeptide domain solely or partly constitutes the amyloid core of the Pmel17 filaments also remains to be defined. 1. RPT Amyloid Formation Is pH Dependent We have studied the repeat domain (RPT, residues 315-444) as a model system of conformational change from soluble and unstructured monomer to aggregated, beta-sheet-containing fibrils. The RPT primary amino acid sequence is comprised of 10 imperfect 13 residue repeats that are rich in Pro, Ser, Thr, and Glu. RPT contains 16 carboxylates underscoring its propensity to undergo pH induced conformational changes. Because pH and protein structure are linked in vivo, we studied the local and macroscopic RPT conformation as a function of pH in detail. Since Trp emission is highly sensitive to solvent polarity, local conformational changes, and protein-protein interactions, we exploited the only intrinsic W423 as a site-specific fluorescent probe of amyloid structure and aggregation kinetics. A critical pH regime (4.5 to 5.5) was identified for fibril formation suggesting the involvement of at least three carboxylic acids in the structural rearrangement necessary for aggregation. The high responsiveness of W423 during RPT aggregation points towards a key role for the C-terminal region in fibril assembly. To investigate a direct correlation between changes in melanosomal pH and formation of RPT fibrils, a pH titration assay was performed. Preformed RPT aggregates (pH 4.0) were titrated to pH 7.0. At pH 5.0, small, highly curved aggregates change into long striated fibrils, reminiscent of the fibril transition observed in stage I and II melanosomes. Further neutralization to pH 7.0 resulted in complete disassembly of RPT fibrils. This unique aggregation/disaggregation process is in contrast to disease-related amyloids, which are notorious for resisting the harshest treatments. We propose in the highly acidic melanosome (stage I), protein aggregation is initiated with fibril elongation occurring only after the compartment solution reaches an optimized pH 5 (stage II). Upon protonation of specific Glu residues, the electrostatic charge repulsion within the polypeptide chain reduces, thereby leading to formation of compact structures that promote key interactions required for fibril formation. In addition, our observation that RPT will readily aggregate (approx. 2 microM) at the optimized pH (5.0) could suggest that only fibrils are stabilized in lieu of potentially toxic oligomers. While our data show that fibrils would dissolve in the near neutral conditions found in stage III and IV melanosomes, it is unclear whether upon melanin deposition, the polymeric material could sequester the fibrils from solution and hence protect them from dissolution. Nevertheless, if released and exposed to the neutral environments outside the melanosome, fibrils will readily disintegrate and thus maintain their benign nature. 2. Probing RPT Fibril Disassembly To investigate fibril disassembly, we employed atomic force microscopy (AFM) and nuclear magnetic resonance (NMR) spectroscopy as ultrastructural and molecular probes, respectively. Specifically, we asked whether intermediates associated with disease-related amyloids are circumvented during dissolution. To monitor fibril disassembly, preformed RPT amyloid was deposited on mica and visualized by AFM under wet buffer conditions. At pH 5.0, long, straight and unbranched fibrils were observed, reminiscent to those seen by TEM. Upon washing these fibrils with pH 6.5 buffer, the fibrils begin to dissolve. Real-time monitoring reveals fragmentation of large fibrils followed by complete disappearance of smaller fragments on the order of minutes. To obtain residue specific insight, isotopically labeled fibrillar RPT was prepared for NMR spectroscopy. At pH 5.0, no backbone amide resonances were observed for residues 378-444, suggesting this region contains the amyloidogenic core. This is consistent with our assertion that the C-terminal region is important for fibril formation. Dissolution kinetics and spectra data showed no evidence of stable intermediates. The absence of intermediates also was verified by size exclusion chromatography. Furthermore, individual Glu backbone amide resonances exhibited similar kinetic trends, suggesting fibril unfolding is a global event involving many deprotonation events.