Nonspecificity fingerprints for clinical-stage antibodies in solution.

Monoclonal antibodies (mAbs) have successfully been developed for the treatment of a wide range of diseases. The clinical success of mAbs does not solely rely on optimal potency and safety but also require good biophysical properties to ensure a high developability potential. In particular, nonspecific interactions are a key developability parameter to monitor during discovery and development. Despite an increased focus on the detection of nonspecific interactions, their underlying physicochemical origins remain poorly understood. Here, we employ solution-based microfluidic technologies to characterize a set of clinical-stage mAbs and their interactions with commonly used nonspecificity ligands to generate nonspecificity fingerprints, providing quantitative data on the underlying physical chemistry. Furthermore, the solution-based analysis enables us to measure binding affinities directly, and we evaluate the contribution of avidity in nonspecific binding by mAbs. We find that avidity can increase the apparent affinity by two orders of magnitude. Notably, we find that a subset of these highly developed mAbs show nonspecific electrostatic interactions, even at physiological pH and ionic strength, and that they can form microscale particles with charge-complementary polymers. The group of mAb constructs flagged here for nonspecificity are among the worst performers in independent reports of surface and column-based screens. The solution measurements improve on the state-of-the-art by providing a stand-alone result for individual mAbs without the need to benchmark against cohort data. Based on our findings, we propose a quantitative solution-based nonspecificity score, which can be integrated in the development workflow for biological therapeutics and more widely in protein engineering.

Surface patches induce nonspecific binding and phase separation of antibodies.

Nonspecific interactions are a key challenge in the successful development of therapeutic antibodies. The tendency for nonspecific binding of antibodies is often difficult to reduce by rational design, and instead, it is necessary to rely on comprehensive screening campaigns. To address this issue, we performed a systematic analysis of the impact of surface patch properties on antibody nonspecificity using a designer antibody library as a model system and single-stranded DNA as a nonspecificity ligand. Using an in-solution microfluidic approach, we find that the antibodies tested bind to single-stranded DNA with affinities as high as KD = 1 µM. We show that DNA binding is driven primarily by a hydrophobic patch in the complementarity-determining regions. By quantifying the surface patches across the library, the nonspecific binding affinity is shown to correlate with a trade-off between the hydrophobic and total charged patch areas. Moreover, we show that a change in formulation conditions at low ionic strengths leads to DNA-induced antibody phase separation as a manifestation of nonspecific binding at low micromolar antibody concentrations. We highlight that phase separation is driven by a cooperative electrostatic network assembly mechanism of antibodies with DNA, which correlates with a balance between positive and negative charged patches. Importantly, our study demonstrates that both nonspecific binding and phase separation are controlled by the size of the surface patches. Taken together, these findings highlight the importance of surface patches and their role in conferring antibody nonspecificity and its macroscopic manifestation in phase separation.

Diboronate Fluorophore for the Measurement of l-Glucose and Other Carbohydrates and Its Interaction with Albumin.

l-Glucose has recently been investigated as an artificial sweetener, but no facile method is established for the measurement of l-glucose. The commercial probe Eversense employs a fluorescent diboronate in a small device for the optical monitoring of d-glucose in people with diabetes. Being achiral, the Eversense probe should be able to detect l-glucose as well as native d-glucose, but the probe is designed for fixation under the skin, and our attempts to use the probe at laboratory conditions failed, as the probe was resetting when moved between compartments. We thus designed a water-soluble anthracene diboronate 8 similar to the fluorophore used in Eversense and found 8 to respond well to l-glucose and other carbohydrates and artificial sweeteners, thus enabling measurements of l-glucose with the limit of quantification of 12 µM. Notably, the fluorescent signal of diboronate 8 was largely quenched in buffers with the physiological concentration of albumin (0.5 mM), so the given analytical method would need more optimization to be useful for measuring l-glucose and other carbohydrates in blood samples. We suspect that other diboronate fluorophores from the literature may be similarly quenched if applied in the presence of albumin.

An integrated platform approach enables discovery of potent, selective and ligand-competitive cyclic peptides targeting the GIP receptor.

In any drug discovery effort, the identification of hits for further optimisation is of crucial importance. For peptide therapeutics, display technologies such as mRNA display have emerged as powerful methodologies to identify these desired de novo hit ligands against targets of interest. The diverse peptide libraries are genetically encoded in these technologies, allowing for next-generation sequencing to be used to efficiently identify the binding ligands. Despite the vast datasets that can be generated, current downstream methodologies, however, are limited by low throughput validation processes, including hit prioritisation, peptide synthesis, biochemical and biophysical assays. In this work we report a highly efficient strategy that combines bioinformatic analysis with state-of-the-art high throughput peptide synthesis to identify nanomolar cyclic peptide (CP) ligands of the human glucose-dependent insulinotropic peptide receptor (hGIP-R). Furthermore, our workflow is able to discriminate between functional and remote binding non-functional ligands. Efficient structure-activity relationship analysis (SAR) combined with advanced in silico structural studies allow deduction of a thorough and holistic binding model which informs further chemical optimisation, including efficient half-life extension. We report the identification and design of the first de novo, GIP-competitive, incretin receptor family-selective CPs, which exhibit an in vivo half-life up to 10.7 h in rats. The workflow should be generally applicable to any selection target, improving and accelerating hit identification, validation, characterisation, and prioritisation for therapeutic development.

Non-specificity as the sticky problem in therapeutic antibody development.

Antibodies are highly potent therapeutic scaffolds with more than a hundred different products approved on the market. Successful development of antibody-based drugs requires a trade-off between high target specificity and target binding affinity. In order to better understand this problem, we here review non-specific interactions and explore their fundamental physicochemical origins. We discuss the role of surface patches - clusters of surface-exposed amino acid residues with similar physicochemical properties - as inducers of non-specific interactions. These patches collectively drive interactions including dipole-dipole, π-stacking and hydrophobic interactions to complementary moieties. We elucidate links between these supramolecular assembly processes and macroscopic development issues, such as decreased physical stability and poor in vivo half-life. Finally, we highlight challenges and opportunities for optimizing protein binding specificity and minimizing non-specificity for future generations of therapeutics.

Ubiquitin Interacting Motifs: Duality Between Structured and Disordered Motifs.

Ubiquitin is a small protein at the heart of many cellular processes, and several different protein domains are known to recognize and bind ubiquitin. A common motif for interaction with ubiquitin is the Ubiquitin Interacting Motif (UIM), characterized by a conserved sequence signature and often found in multi-domain proteins. Multi-domain proteins with intrinsically disordered regions mediate interactions with multiple partners, orchestrating diverse pathways. Short linear motifs for binding are often embedded in these disordered regions and play crucial roles in modulating protein function. In this work, we investigated the structural propensities of UIMs using molecular dynamics simulations and NMR chemical shifts. Despite the structural portrait depicted by X-crystallography of stable helical structures, we show that UIMs feature both helical and intrinsically disordered conformations. Our results shed light on a new class of disordered UIMs. This group is here exemplified by the C-terminal domain of one isoform of ataxin-3 and a group of ubiquitin-specific proteases. Intriguingly, UIMs not only bind ubiquitin. They can be a recruitment point for other interactors, such as parkin and the heat shock protein Hsc70-4. Disordered UIMs can provide versatility and new functions to the client proteins, opening new directions for research on their interactome.

(Dis)similarity Index To Compare Correlated Motions in Molecular Simulations.

Molecular dynamics (MD) simulations are widely used to complement or guide experimental studies in the characterization of protein dynamics, thanks to improvements in force-field accuracy, along with in the software and hardware to sample the conformational landscape of proteins. Among the different applications of MD simulations, the study of correlated motions is largely employed for different purposes. Several metrics have been developed to describe correlated motions in the MD ensemble, such as methods based on Pearson Correlation or Mutual Information. Cross-correlation analysis of MD trajectories is indeed appealing not only to identify residues characterized by coupled fluctuations in protein structures but also since it can be used to extrapolate motions along directions in which major conformational changes should occur, for example on longer time scales than the ones that are actually simulated. Nevertheless, most of the MD studies employ average correlation maps and mostly in a qualitative way, even when different systems or different replicates of the same system are compared. The broad application of correlation metrics in the analysis of MD simulations, especially for comparative purposes, requires a step forward toward more quantitative and accurate comparisons. We thus here employed a simple but effective index, which is based on a normalized Frobenius norm of the differences between protein correlation maps, to compare correlated motions. We applied this index for a quantitative comparison of correlated motions from MD simulations of seven proteins of different size and fold. We also employed the index to assess the robustness of correlation description when multi-replicate MD simulations of a same system are used, and we compared our index to metrics for comparison of structural ensembles such as Root Mean Square Inner Product and the Bhattacharyya Coefficient.

The Toxic Effects of Pathogenic Ataxin-3 Variants in a Yeast Cellular Model.

Ataxin-3 (AT3) is a deubiquitinating enzyme that triggers an inherited neurodegenerative disorder, spinocerebellar ataxia type 3, when its polyglutamine (polyQ) stretch close to the C-terminus exceeds a critical length. AT3 variants carrying the expanded polyQ are prone to associate with each other into amyloid toxic aggregates, which are responsible for neuronal death with ensuing neurodegeneration. We employed Saccharomyces cerevisiae as a eukaryotic cellular model to better clarify the mechanism by which AT3 triggers the disease. We expressed three variants: one normal (Q26), one expanded (Q85) and one truncated for a region lying from the beginning of its polyQ stretch to the end of the protein (291Δ). We found that the expression of the expanded form caused reduction in viability, accumulation of reactive oxygen species, imbalance of the antioxidant defense system and loss in cell membrane integrity, leading to necrotic death. The truncated variant also exerted a qualitatively similar, albeit milder, effect on cell growth and cytotoxicity, which points to the involvement of also non-polyQ regions in cytotoxicity. Guanidine hydrochloride, a well-known inhibitor of the chaperone Hsp104, almost completely restored wild-type survival rate of both 291Δ- and Q85-expressing strains. This suggests that AT3 aggregation and toxicity is mediated by prion forms of yeast proteins, as this chaperone plays a key role in their propagation.

Communication routes in ARID domains between distal residues in helix 5 and the DNA-binding loops.

ARID is a DNA-binding domain involved in several transcriptional regulatory processes, including cell-cycle regulation and embryonic development. ARID domains are also targets of the Human Cancer Protein Interaction Network. Little is known about the molecular mechanisms related to conformational changes in the family of ARID domains. Thus, we have examined their structural dynamics to enrich the knowledge on this important family of regulatory proteins. In particular, we used an approach that integrates atomistic simulations and methods inspired by graph theory. To relate these properties to protein function we studied both the free and DNA-bound forms. The interaction with DNA not only stabilizes the conformations of the DNA-binding loops, but also strengthens pre-existing paths in the native ARID ensemble for long-range communication to those loops. Residues in helix 5 are identified as critical mediators for intramolecular communication to the DNA-binding regions. In particular, we identified a distal tyrosine that plays a key role in long-range communication to the DNA-binding loops and that is experimentally known to impair DNA-binding. Mutations at this tyrosine and in other residues of helix 5 are also demonstrated, by our approach, to affect the paths of communication to the DNA-binding loops and alter their native dynamics. Overall, our results are in agreement with a scenario in which ARID domains exist as an ensemble of substates, which are shifted by external perturbation, such as the interaction with DNA. Conformational changes at the DNA-binding loops are transmitted long-range by intramolecular paths, which have their heart in helix 5.

The role of salt bridges on the temperature adaptation of aqualysin I, a thermostable subtilisin-like proteinase.

Differences in salt bridges are believed to be a structural hallmark of homologous enzymes from differently temperature-adapted organisms. Nevertheless, the role of salt bridges on structural stability is still controversial. While it is clear that most buried salt bridges can have a functional or structural role, the same cannot be firmly stated for ion pairs that are exposed on the protein surface. Salt bridges, found in X-ray structures, may not be stably formed in solution as a result of high flexibility or high desolvation penalty. More studies are thus needed to clarify the picture on salt bridges and temperature adaptation. We contribute here to this scenario by combining atomistic simulations and experimental mutagenesis of eight mutant variants of aqualysin I, a thermophilic subtilisin-like proteinase, in which the residues involved in salt bridges and not conserved in a psychrophilic homolog were systematically mutated. We evaluated the effects of those mutations on thermal stability and on the kinetic parameters. Overall, we show here that only few key charged residues involved in salt bridges really contribute to the enzyme thermal stability. This is especially true when they are organized in networks, as here attested by the D17N mutation, which has the most remarkable effect on stability. Other mutations had smaller effects on the properties of the enzyme indicating that most of the isolated salt bridges are not a distinctive trait related to the enhanced thermal stability of the thermophilic subtilase.

PyInteraph: a framework for the analysis of interaction networks in structural ensembles of proteins.

In the last years, a growing interest has been gathering around the ability of Molecular Dynamics (MD) to provide insight into the paths of long-range structural communication in biomolecules. The knowledge of the mechanisms related to structural communication helps in the rationalization in atomistic details of the effects induced by mutations, ligand binding, and the intrinsic dynamics of proteins. We here present PyInteraph, a tool for the analysis of structural ensembles inspired by graph theory. PyInteraph is a software suite designed to analyze MD and structural ensembles with attention to binary interactions between residues, such as hydrogen bonds, salt bridges, and hydrophobic interactions. PyInteraph also allows the different classes of intra- and intermolecular interactions to be represented, combined or alone, in the form of interaction graphs, along with performing network analysis on the resulting interaction graphs. The program also integrates the network description with a knowledge-based force field to estimate the interaction energies between side chains in the protein. It can be used alone or together with the recently developed xPyder PyMOL plugin through an xPyder-compatible format. The software capabilities and associated protocols are here illustrated by biologically relevant cases of study. The program is available free of charge as Open Source software via the GPL v3 license at http://linux.btbs.unimib.it/pyinteraph/.

Interactions of ataxin-3 with its molecular partners in the protein machinery that sorts protein aggregates to the aggresome.

Ataxin-3 (AT3) is the protein that triggers the inherited neurodegenerative disorder spinocerebellar ataxia type 3 when its polyglutamine (polyQ) stretch close to the C-terminus exceeds a critical length. AT3 consists of the N-terminal globular Josephin domain (JD) and the C-terminal disordered one. It cleaves isopeptide bonds between ubiquitin monomers, an event involved in protein quality control mechanisms. AT3 has been implicated in the pathway that sorts aggregated protein to aggresomes via microtubules, in which dynein and histone deacetylase 6 (HDAC6) also seem to be involved. By taking advantage of small angle X-ray scattering (SAXS) and surface plasmon resonance (SPR), we have investigated the interaction of AT3 with tubulin and HDAC6. Based on SAXS results, the AT3 oligomer, consisting of 6-7 subunits, tightly binds to the tubulin hexameric oligomer in a "parallel" fashion. By SPR analysis we have demonstrated that AT3 binds to tubulin dimer with a 50nM affinity. Binding fits with a Langmuir 1:1 model and involves a single binding interface. Nevertheless, the interaction surface consists of three distinct, discontinuous tubulin-binding regions (TBR), one located in the JD, and the two others in the disordered domain, upstream and downstream of the polyQ stretch. In the absence of any of the three TBRs, the affinity is drastically reduced. By SPR we have also provided the first evidence of direct binding of AT3 to HDAC6, with affinity in the range 0.1-1µM. These results shed light on the interactions among the components of the transport machinery that sorts aggregate protein to the aggresome, and pave the way to in vivo studies aimed at further clarifying their roles.

Different ataxin-3 amyloid aggregates induce intracellular Ca(2+) deregulation by different mechanisms in cerebellar granule cells.

This work aims at elucidating the relation between morphological and physicochemical properties of different ataxin-3 (ATX3) aggregates and their cytotoxicity. We investigated a non-pathological ATX3 form (ATX3Q24), a pathological expanded form (ATX3Q55), and an ATX3 variant truncated at residue 291 lacking the polyQ expansion (ATX3/291Δ). Solubility, morphology and hydrophobic exposure of oligomeric aggregates were characterized. Then we monitored the changes in the intracellular Ca(2+) levels and the abnormal Ca(2+) signaling resulting from aggregate interaction with cultured rat cerebellar granule cells. ATX3Q55, ATX3/291Δ and, to a lesser extent, ATX3Q24 oligomers displayed similar morphological and physicochemical features and induced qualitatively comparable time-dependent intracellular Ca(2+) responses. However, only the pre-fibrillar aggregates of expanded ATX3 (the only variant which forms bundles of mature fibrils) triggered a characteristic Ca(2+) response at a later stage that correlated with a larger hydrophobic exposure relative to the two other variants. Cell interaction with early oligomers involved glutamatergic receptors, voltage-gated channels and monosialotetrahexosylganglioside (GM1)-rich membrane domains, whereas cell interaction with more aged ATX3Q55 pre-fibrillar aggregates resulted in membrane disassembly by a mechanism involving only GM1-rich areas. Exposure to ATX3Q55 and ATX3/291Δ aggregates resulted in cell apoptosis, while ATX3Q24 was substantially innocuous. Our findings provide insight into the mechanisms of ATX3 aggregation, aggregate cytotoxicity and calcium level modifications in exposed cerebellar cells.

The conformational ensemble of the disordered and aggregation-protective 182-291 region of ataxin-3.

BACKGROUND: Intrinsically disordered proteins (IDPs) are an emerging part of structural biology that has challenged the classic paradigm of structure-function relationship. Indeed, IDPs have been associated with different physiological functions and associated with several pathologies, such as polyglutamine (polyQ) related diseases. Ataxin-3 (AT3) is the smallest polyQ protein, composed by the N-terminal folded Josephin domain (JD), which is amyloidogenic on its own, and a C-terminal unstructured part. The disordered region between the polyQ and the JD, AT3182-291 plays a key role in the development of the disease. METHODS: We integrated different biophysical experimental techniques, atomistic explicit-solvent molecular dynamics (MD) simulations and graph theory to study AT3182-291 structure. RESULTS: AT3182-291 is a monomeric intrinsically disordered (ID) domain in solution and it is characterized by different conformational states, ascribable to pre-molten globule populations with different degrees of compactness. If isolated, it decreases the aggregation of the entire AT3. CONCLUSIONS: We provided the first structural description of an ID domain associated to a polyQ protein and we also showed that it exerts protective effects against AT3 aggregation. This effect is likely to be induced by intermolecular interactions between AT3 and the ubiquitin-interacting motifs of AT3182-291. Electrostatic interactions play a pivotal role in regulating the topology and tertiary propensity of the fragment and hub residues have been identified. GENERAL SIGNIFICANCE: Synergistic use of atomistic simulations and biophysical techniques should be more generally applied to the study of IDPs. Since ID domains and polyQ-proteins are intimately connected, the study here provided can be of interest for other members of the group.

Dynamics fingerprint and inherent asymmetric flexibility of a cold-adapted homodimeric enzyme. A case study of the Vibrio alkaline phosphatase.

BACKGROUND: Protein dynamics influence protein function and stability and modulate conformational changes. Such motions depend on the underlying networks of intramolecular interactions and communicating residues within the protein structure. Here, we provide the first characterization of the dynamic fingerprint of the dimeric alkaline phosphatase (AP) from the cold-adapted Vibrio strain G15-21 (VAP), which is among the APs with the highest known kcat at low temperatures. METHODS: Multiple all-atom explicit solvent molecular dynamics simulations were employed in conjunction with different metrics to analyze the dynamical patterns and the paths of intra- and intermolecular communication. RESULTS: Interactions and coupled motions at the interface between the two VAP subunits have been characterized, along with the networks of intramolecular interactions. It turns out a low number of intermolecular interactions and coupled motions, which result differently distributed in the two monomers. The paths of long-range communication mediated from the catalytic residues to distal sites were also characterized, pointing out a different information flow in the two subunits. CONCLUSIONS: A pattern of asymmetric flexibility is evident in the two identical subunits of the VAP dimer that is intimately linked to a different distribution of intra- and intermolecular interactions. The asymmetry was also evident in pairs of cross-correlated residues during the dynamics. GENERAL SIGNIFICANCE: The results here discussed provide a structural rationale to the half-of-site mechanism previously proposed for VAP and other APs, as well as a general framework to characterize asymmetric dynamics in homomeric enzymes.

The relationship between aggregation and toxicity of polyglutamine-containing ataxin-3 in the intracellular environment of Escherichia coli.

Several neurodegenerative diseases are triggered by proteins containing a polyglutamine (polyQ) stretch expanded beyond a critical threshold. Among these, ataxin-3 (AT3) is the causative agent of spinocerebellar ataxia type-3. We expressed three authentic AT3 variants in Escherichia coli: one normal (AT3-Q24), one expanded (AT3-Q55) and one truncated immediately upstream of the polyQ (AT3-291Δ). Then, based on growth rate reduction, we quantified protein toxicity. We show that AT3-Q55 and -291Δ strongly reduced the growth rate in the early stages (2-4 h), unlike AT3-Q24. This correlated well with the appearance of soluble cytosolic oligomers, but not with the amount of insoluble protein in inclusion bodies (IBs). The impact of AT3-291Δ on cell growth suggests an intrinsic toxicity of the AT3 fragment. Besides the typical Fourier Transform Infrared Spectroscopy (FTIR) signal for intermolecular ß-sheets, the expanded form displayed an additional infrared signature, which was assigned to glutamine side-chain hydrogen bonding and associated with SDS-insoluble fibrils. The elongation of the latter was monitored by Atomic Force Microscopy (AFM). This mirrors the well-known in vitro two-step aggregation pattern of expanded AT3. We also demonstrated that final aggregates of strains expressing expanded or truncated AT3 play a protective role against toxicity. Furthermore, our findings suggest that the mechanisms of toxicity are evolutionarily conserved.

Protein aggregation: mechanisms and functional consequences.

Understanding the mechanisms underlying protein misfolding and aggregation has become a central issue in biology and medicine. Compelling evidence show that the formation of amyloid aggregates has a negative impact in cell function and is behind the most prevalent human degenerative disorders, including Alzheimer's Parkinson's and Huntington's diseases or type 2 diabetes. Surprisingly, the same type of macromolecular assembly is used for specialized functions by different organisms, from bacteria to human. Here we address the conformational properties of these aggregates, their formation pathways, their role in human diseases, their functional properties and how bioinformatics tools might be of help to study these protein assemblies.

The role of the central flexible region on the aggregation and conformational properties of human ataxin-3.

Aggregation of human ataxin-3 (AT3) into amyloid fibrils is responsible for spinocerebellar ataxia type 3. This protein consists of a folded N-terminal domain (Josephin domain, residues 1-182), a central flexible region (residues 183-291), a poly-glutamine sequence of variable length and a short C-terminal flexible region. Very little is known about the influence of the central flexible region on the conformational and aggregation properties of this protein. The present study aimed to investigate the specific role of this portion of the protein (residues 183-291). Accordingly, protein fragments 1-182 (AT3/182) and 1-291 (AT3/291) were produced and compared by thioflavin-T fluorescence, Fourier transform infrared spectroscopy, CD, intrinsic fluorescence and ESI-MS. It is shown that the central flexible region enhances protein aggregation and can populate conformational states with different degrees of compactness. Both monomeric and dimeric partially-folded forms are identified for both protein fragments under denaturing conditions. Partially-folded monomers and dimers accumulate to a larger extent in AT3/291. These species represent good candidates for early intermediates of the aggregation process under the experimental conditions employed in the present study.

Molecular determinants of enzyme cold adaptation: comparative structural and computational studies of cold- and warm-adapted enzymes.

The identification of molecular mechanisms underlying enzyme cold adaptation is a hot-topic both for fundamental research and industrial applications. In the present contribution, we review the last decades of structural computational investigations on cold-adapted enzymes in comparison to their warm-adapted counterparts. Comparative sequence and structural studies allow the definition of a multitude of adaptation strategies. Different enzymes carried out diverse mechanisms to adapt to low temperatures, so that a general theory for enzyme cold adaptation cannot be formulated. However, some common features can be traced in dynamic and flexibility properties of these enzymes, as well as in their intra- and inter-molecular interaction networks. Interestingly, the current data suggest that a family-centered point of view is necessary in the comparative analyses of cold- and warm-adapted enzymes. In fact, enzymes belonging to the same family or superfamily, thus sharing at least the three-dimensional fold and common features of the functional sites, have evolved similar structural and dynamic patterns to overcome the detrimental effects of low temperatures.

A major role for side-chain polyglutamine hydrogen bonding in irreversible ataxin-3 aggregation.

The protein ataxin-3 consists of an N-terminal globular Josephin domain (JD) and an unstructured C-terminal region containing a stretch of consecutive glutamines that triggers the neurodegenerative disorder spinocerebellar ataxia type 3, when it is expanded beyond a critical threshold. The disease results from misfolding and aggregation, although the pathway and structure of the aggregation intermediates are not fully understood. In order to provide insight into the mechanism of the process, we monitored the aggregation of a normal (AT3Q24) ataxin-3, an expanded (AT3Q55) ataxin-3, and the JD in isolation. We observed that all of them aggregated, although the latter did so at a much slower rate. Furthermore, the expanded AT3Q55 displayed a substantially different behavior with respect to the two other variants in that at the latest stages of the process it was the only one that did the following: i) lost its reactivity towards an anti-oligomer antibody, ii) generated SDS-insoluble aggregates, iii) gave rise to bundles of elongated fibrils, and iv) displayed two additional bands at 1604 and 1656 cm(-1) in FTIR spectroscopy. Although these were previously observed in other aggregated polyglutamine proteins, no one has assigned them unambiguously, yet. By H/D exchange experiments we show for the first time that they can be ascribed to glutamine side-chain hydrogen bonding, which is therefore the hallmark of irreversibly SDS-insoluble aggregated protein. FTIR spectra also showed that main-chain intermolecular hydrogen bonding preceded that of glutamine side-chains, which suggests that the former favors the latter by reorganizing backbone geometry.