Disease-related amyloidogenic variants of human lysozyme trigger the unfolded protein response and disturb eye development in Drosophila melanogaster.

We have created a Drosophila model of lysozyme amyloidosis to investigate the in vivo behavior of disease-associated variants. To achieve this objective, wild-type (WT) protein and the amyloidogenic variants F57I and D67H were expressed in Drosophila melanogaster using the UAS-gal4 system and both the ubiquitous and retinal expression drivers Act5C-gal4 and gmr-gal4. The nontransgenic w(1118) Drosophila line was used as a control throughout. We utilized ELISA experiments to probe lysozyme protein levels, scanning electron microscopy for eye phenotype classification, and immunohistochemistry to detect the unfolded protein response (UPR) activation. We observed that expressing the destabilized F57I and D67H lysozymes triggers UPR activation, resulting in degradation of these variants, whereas the WT lysozyme is secreted into the fly hemolymph. Indeed, the level of WT was up to 17 times more abundant than the variant proteins. In addition, the F57I variant gave rise to a significant disruption of the eye development, and this correlated to pronounced UPR activation. These results support the concept that the onset of familial amyloid disease is linked to an inability of the UPR to degrade completely the amyloidogenic lysozymes prior to secretion, resulting in secretion of these destabilized variants, thereby leading to deposition and associated organ damage.

Engineering a camelid antibody fragment that binds to the active site of human lysozyme and inhibits its conversion into amyloid fibrils.

A single-domain fragment, cAb-HuL22, of a camelid heavy-chain antibody specific for the active site of human lysozyme has been generated, and its effects on the properties of the I56T and D67H amyloidogenic variants of human lysozyme, which are associated with a form of systemic amyloidosis, have been investigated by a wide range of biophysical techniques. Pulse-labeling hydrogen-deuterium exchange experiments monitored by mass spectrometry reveal that binding of the antibody fragment strongly inhibits the locally cooperative unfolding of the I56T and D67H variants and restores their global cooperativity to that characteristic of the wild-type protein. The antibody fragment was, however, not stable enough under the conditions used to explore its ability to perturb the aggregation behavior of the lysozyme amyloidogenic variants. We therefore engineered a more stable version of cAb-HuL22 by adding a disulfide bridge between the two beta-sheets in the hydrophobic core of the protein. The binding of this engineered antibody fragment to the amyloidogenic variants of lysozyme inhibited their aggregation into fibrils. These findings support the premise that the reduction in global cooperativity caused by the pathogenic mutations in the lysozyme gene is the determining feature underlying their amyloidogenicity. These observations indicate further that molecular targeting of enzyme active sites, and of protein binding sites in general, is an effective strategy for inhibiting or preventing the aberrant self-assembly process that is often a consequence of protein mutation and the origin of pathogenicity. Moreover, this work further demonstrates the unique properties of camelid single-domain antibody fragments as structural probes for studying the mechanism of aggregation and as potential inhibitors of fibril formation.

The Argonaute protein TbAGO1 contributes to large and mini-chromosome segregation and is required for control of RIME retroposons and RHS pseudogene-associated transcripts.

The protist Trypanosoma brucei possesses a single Argonaute gene called TbAGO1 that is necessary for RNAi silencing. We previously showed that in strain 427, TbAGO1 knock-out leads to a slow growth phenotype and to chromosome segregation defects. Here we report that the slow growth phenotype is linked to defects in segregation of both large and mini-chromosome populations, with large chromosomes being the most affected. These phenotypes are completely reversed upon inducible re-expression of TbAGO1 fused to GFP, demonstrating their link with TbAGO1. Trypanosomes that do not express TbAGO1 show a general increase in the abundance of transcripts derived from the short retroposon RIME (Ribosomal Interspersed Mobile Element). Supplementary large RIME transcripts emerge in the absence of RNAi, a phenomenon coupled to the disappearance of short transcripts. These fluctuations are reversed by inducible expression of GFP::TbAGO1. Furthermore, we use a combination of Northern blots, RT-PCR and sequencing to reveal that RNAi controls expression of transcripts derived from RHS (Retrotransposon Hot Spot) pseudogenes (RHS genes with retro-element(s) integrated within their coding sequence). Absence of RNAi also leads to an increase of steady-state transcripts from regular RHS genes (those without retro-element), indicating a role for pseudogene in control of gene expression. However, analysis of retroposon abundance and arrangement in the genome of multiple clonal cell lines of TbAGO1-/- failed to reveal movement of mobile elements despite the increased amounts of retroposon transcripts.

A nanobody binding to non-amyloidogenic regions of the protein human lysozyme enhances partial unfolding but inhibits amyloid fibril formation.

We report the effects of the interaction of two camelid antibody fragments, generally called nanobodies, namely cAb-HuL5 and a stabilized and more aggregation-resistant variant cAb-HuL5G obtained by protein engineering, on the properties of two amyloidogenic variants of human lysozyme, I56T and D67H, whose deposition in vital organs including the liver, kidney, and spleen is associated with a familial non-neuropathic systemic amyloidosis. Both NMR spectroscopy and X-ray crystallographic studies reveal that cAb-HuL5 binds to the α-domain, one of the two lobes of the native lysozyme structure. The binding of cAb-HuL5/cAb-HuL5G strongly inhibits fibril formation by the amyloidogenic variants; it does not, however, suppress the locally transient cooperative unfolding transitions, characteristic of these variants, in which the ß-domain and the C-helix unfold and which represents key early intermediate species in the formation of amyloid fibrils. Therefore, unlike two other nanobodies previously described, cAb-HuL5/cAb-HuL5G does not inhibit fibril formation via the restoration of the global cooperativity of the native structure of the lysozyme variants to that characteristic of the wild-type protein. Instead, it inhibits a subsequent step in the assembly of the fibrils, involving the unfolding and structural reorganization of the α-domain. These results show that nanobodies can protect against the formation of pathogenic aggregates at different stages in the structural transition of a protein from the soluble native state into amyloid fibrils, illustrating their value as structural probes to study the molecular mechanisms of amyloid fibril formation. Combined with their amenability to protein engineering techniques to improve their stability and solubility, these findings support the suggestion that nanobodies can potentially be developed as therapeutics to combat protein misfolding diseases.

Characterization of oligomeric species on the aggregation pathway of human lysozyme.

The aggregation process of wild-type human lysozyme at pH3.0 and 60 degrees C has been analyzed by characterizing a series of distinct species formed on the aggregation pathway, specifically the amyloidogenic monomeric precursor protein, the oligomeric soluble prefibrillar aggregates, and the mature fibrils. Particular attention has been focused on the analysis of the structural properties of the oligomeric species, since recent studies have shown that the oligomers formed by lysozyme prior to the appearance of mature amyloid fibrils are toxic to cells. Here, soluble oligomers of human lysozyme have been analyzed by a range of techniques including binding to fluorescent probes such as thioflavin T and 1-anilino-naphthalene-8-sulfonate, Fourier transform infrared spectroscopy, and controlled proteolysis. Oligomers were isolated after 5 days of incubation of the protein and appear as spherical particles with a diameter of 8-17 nm when observed by transmission electron microscopy. Unlike the monomeric protein, oligomers have solvent-exposed hydrophobic patches able to bind the fluorescent probe 1-anilino-naphthalene-8-sulfonate. Fourier transform infrared spectroscopy spectra of oligomers are indicative of misfolded species when compared to monomeric lysozyme, with a prevalence of random structure but with significant elements of the beta-sheet structure that is characteristic of the mature fibrils. Moreover, the oligomeric lysozyme aggregates were found to be more susceptible to proteolysis with pepsin than both the monomeric protein and the mature fibrils, indicating further their less organized structure. In summary, this study shows that the soluble lysozyme oligomers are locally unfolded species that are present at low concentration during the initial phases of aggregation. The nonnative conformational features of the lysozyme molecules of which they are composed are likely to be the factors that confer on them the ability to interact inappropriately with a variety of cellular components including membranes.