My personal site...

Proteins prone to pathological aggregation:

Since 2007 I have been focusing my research on proteins prone to pathologic aggregation. The main type of protein aggregation is amyloid aggregation and it is responsible for an important class of misfolding diseases. The molecular bases of all such pathologies are very similar: the protein responsible for the diseases is loosing its native structure and starts to aggregate leading to (often toxic) oligomers, eventually forming mature amyloid fibrils. Amyloid aggregation is a very complex process with many steps from a native monomeric protein to oligomers, protofibrils and mature fibrils.

Aggregation Schema

The study and the understanding of the molecular bases of the protein aggregation are of paramount importance to design future treatments against such severe pathologies.

I am particularly interested in studying the aggregation process structurally and biophysically: the structural determinants of aggregation propensity, the structure of oligomers and of mature fibrils are the main focus of my research.

Light Chain Amyloidosis.

Systemic amyloidoses are protein-misfolding diseases characterized by widespread deposition of amyloid fibrils with severe dysfunction of the affected organs. Light chain (AL) amyloidosis is the most common systemic amyloidosis: fibrils originate from the aggregation of misfolding-prone immunoglobulin light chains (LCs); heart involvement is common and it is the main cause of death. Current therapies for AL amyloidosis are based on suppression of LCs production in the bone marrow by chemotherapy; however, severe heart involvement precludes the use of the most aggressive and most effective schemes. Solid clinical and experimental evidence, in AL as well as in other amyloidoses, indicates that cell and organ dysfunction is not only due to fibrils, but also to soluble, pre-fibrillar amyloidogenic precursors. In particular, soluble LCs that are cardiotoxic in patients were also shown to be toxicants for cardiac cells and model animals. However, the molecular features determining the ability of a subset of LCs to target the heart and to be toxic for cardiac cells are still largely undefined. Our aim is to understand the molecular properties determining the toxicity of specific LCs. In order to clarify this issue, we are working on a pool of toxic and non-toxic LCs and we characterise their structure, fold stability, flexibility and hydrophobicity. By site-directed mutagenesis and more in general by protein modification, we aim to pinpoint the toxic species and the biophysical and biochemical properties of LCs, which correlate with the toxicity in patients.

Light Chain Amyloidosis

Human Beta-2 Microglobulin.

Beta-2 microglobulin (β2m) is the light chain of the major histocompatibility complex I. In patients with kidney failure, β2m catabolism is impaired and the protein tends to accumulate in the blood to high concentrations, triggering the formation of amyloid plaques in bones and skeletal joints. Such disorder is known as Dialysis Related Amyloidosis. β2m is a seven-stranded β-sandwich, stabilized by a disulphide bridge (Cys25 - Cys80) that locks the two β-sheets.

Click here for the full list of publications on β2m.

The D76N mutant of Beta-2 microglobulin is responsible for a novel genetic disorder:

Four persons belonging to the same family were affected by chronic intestinal disorders which resulted in a dramatic loss of weight and in the death of one patient. In such patients, abundant amyloid deposits of a mutated form of Beta-2 microglobulin (Asp76 to Asn) were found in several organs and this b2m variant is held responsible for the symptoms. The D76N variant is very amyloidogenic in vivo and in vitro and can also trigger wt b2m aggregation. The D76N variant have been - and still is structurally - characterized as monomer and in the context of the Major Histocompatibility Complex in order to understand the effects of the mutation, which results in such an increased aggregation propensity.

The DE loop:

We have shown that the β2m loop between the D and E strands (residues 57-60) is critical for β2m amyloid propensity. In wild type the geometry of the DE loop is strained and it is source of instability for the overall protein fold. To study the role of such loop in determining b2m biochemical and biophysical properties we have studied three mutants. The conserved residue Trp60 was mutated to Gly, the mutation results in a remarkably lower aggregation propensity under mild conditions, and in an increased fold stability compared to w.t. β2m. The crystal structure of the W60G mutant shows that all residues of the DE loop fall in the favoured regions of the Ramachandran plot, suggesting that the mutation to Gly confers higher overall stability to β2m, thanks to its unique conformational properties that help release stereochemical strain of the DE loop. Conversely, the mutation of Asp59 to Pro leads to a more strained DE loop, resulting in diminished thermal stability and increased propensity to form amyloid fibrils compared to w.t. β2m. Comparative analysis of the DE loop conformations shows that w.t. β2m and the D59P mutant display an irregular DE loop, while the W60G, W60V and W60C mutants host a regular β-turn. The properties of these mutants are extensively studied to understand the biophysical and structural role of the DE loop in β2m under native and aggregating conditions.

(Ramachandran plots)

Ramachandran plots of (left) β2m in the MHC class I, (middle) of monomeric β2m  and (right) of the W60G β2m mutant.

Animation showing the loop DE conformational change when β2m is complexed or monomeric. The W60G and W60V mutants share the same conformation of the complexed β2m.

β2m oligomers:

The oligomerisation occurring during amyloid formation is an important intermediate step for several reasons: growing evidence suggest that the most toxic aggregated species are the oligomers and not the mature fibrils, and oligomerisation would be the ideal step of amyloid formation to block in order to avoid the insurgence of the disease. In order to overcome the transient nature of oligomers and to structurally characterise b2m oligomers we recently designed a strategy by which we showed that it is possible to obtain stable oligomers suitable to crystallographic studies. The aggregation propensity and the crystal structures of these artificial oligomers provide interesting clues on the first steps of b2m oligomerisation.

Tetrameric crystal structure

Tetrameric crystal structure of the SS-linked dimer of E50C mutant of β2m

β2m amyloid fibrils:

Determining the fine structure of amyloid fibrils as well as understanding their processes of nucleation and growth remains a difficult yet essential challenge, directly linked to our current poor insight into protein misfolding and aggregation diseases. Thanks to the collaboration with Dr. G. Pintacuda we are working to elucidate the structure of β2m amyloids using ssNMR. So far we have used solid-state NMR to probe the structural features of fibrils formed by full-length β2m (99 residues) at pH 2.5 and pH 7.4 and crystalline β2m. Among other observations the variations in the chemical shifts of the key Pro32 residue suggest the involvement of a cis-trans isomerization in the process of β2m fibril formation.

Acylphosphatase from S. Solfataricus

Acylphosphatase from Sulfolobus solfataricus (Sso AcP) is a protein that has been largely exploited as a model to study native-like aggregation. Sso AcP is a 101-residue enzyme that catalyses the hydrolysis of the phosphoanhydride bond in acylphosphates. Structural analysis revealed that Sso AcP hosts the typical ferredoxin-like topology shared by other acylphosphatases, consisting of two a-helices facing a five-stranded b-sheet; strands 4 and 5 (B4 and B5, respectively) are located at the opposite sheet edges. Studying the structural determinants of Sso AcP aggregation helps the understanding of the basic principles of native-like protein aggregation.

Click here for the full list of publications on Sso AcP.

Human Neuroserpin

Human neuroserpin (hNS) belongs to the serpin superfamily. Serpins build a large and evolutionary widespread protein superfamily hosting members that are mainly Ser-protease inhibitors. Typically, serpins display a conserved core domain composed of three main -sheets and 9-10 -helices, for a total of about 350 amino acids. hNS is mostly expressed in neurons and in the central and peripheral nervous system, where it targets tissue-type plasminogen activator. hNS activity is relevant for axogenesis, synaptogenesis and synaptic plasticity. Five (single amino acid) hNS mutations are associated to a severe neurodegenerative disease in man, leading to early onset dementia, epilepsy and neuronal death. The functional aspects of hNS protease inhibition are linked to the presence of a long exposed loop (reactive center loop, RCL) that acts as bait for the incoming partner protease. Large hNS conformational changes, associated to the cleavage of the RCL, trap the protease in an acyl-enzyme complex with hNS. Contrary to other serpins, such complex has a half-life of about 10 min for hNS. Conformational flexibility is held to be at the bases of hNS polymerisation leading to Collins bodies intracellular deposition and neuronal damage in the pathological hNS variants.

Click here for the full list of publications on neuroserpin.

Crystallography on hNS:

hNS medical interest and the lack of structural information on hNS prompted us to study the crystal structure of native and cleaved hNS, reported here at 3.15 and 1.85 β resolution, respectively. In the light of the three-dimensional structures, we focus on the hNS reactive centre loop in its intact and cleaved conformations relative to the current serpin polymerization models and discuss the protein sites hosting neurodegenerative mutations. On the basis of homologous serpin structures, we suggest the location of a protein surface site that may stabilize the hNS native (metastable) form.


Ribbon representation of the crystallographic pentamer of native hNS (left), zoom on intermolecular beta interaction between three hNS molecules in the crystal (top right) and the superposition of the five molecules present in the asymmetric unit displaying a remarkable conformational variability of the RCL in the five molecules (bottom right)

Polymers and latent forms:

hNS polymerization is held to follow mechanisms different from those of amyloid aggregation, a general agreement on serpin polymer organisation is still lacking. We have been working on the biophysical characterisation of native hNS and of the conformational changes occurring upon temperature changes. The biophysical characterisation of the several hNS conformers and the polymerisation kinetics are crucial to unravel the mechanism of pathologic polymer formation.

Transmission electron microscopy pictures of hNS polymer-45 and polymer-85

Transmission electron microscopy pictures of hNS polymer-45 and polymer-85