Early implementation of QbD in biopharmaceutical development: a practical example.

In drug development, the "onus" of the low R&D efficiency has been put traditionally onto the drug discovery process (i.e., finding the right target or "binding" functionality). Here, we show that manufacturing is not only a central component of product success, but also that, by integrating manufacturing and discovery activities in a "holistic" interpretation of QbD methodologies, we could expect to increase the efficiency of the drug discovery process as a whole. In this new context, early risk assessment, using developability methodologies and computational methods in particular, can assist in reducing risks during development in a cost-effective way. We define specific areas of risk and how they can impact product quality in a broad sense, including essential aspects such as product efficacy and patient safety. Emerging industry practices around developability are introduced, including some specific examples of applications to biotherapeutics. Furthermore, we suggest some potential workflows to illustrate how developability strategies can be introduced in practical terms during early drug development in order to mitigate risks, reduce drug attrition and ultimately increase the robustness of the biopharmaceutical supply chain. Finally, we also discuss how the implementation of such methodologies could accelerate the access of new therapeutic treatments to patients in the clinic.

Aggregation risk prediction for antibodies and its application to biotherapeutic development.

Aggregation is a common problem affecting biopharmaceutical development that can have a significant effect on the quality of the product, as well as the safety to patients, particularly because of the increased risk of immune reactions. Here, we describe a new high-throughput screening algorithm developed to classify antibody molecules based on their propensity to aggregate. The tool, constructed and validated on experimental aggregation data for over 500 antibodies, is able to discern molecules with a high aggregation propensity as defined by experimental criteria relevant to bioprocessing and manufacturing of these molecules. Furthermore, we show how this tool can be combined with other computational approaches during early drug development to select molecules with reduced risk of aggregation and optimal developability properties.

Probing the mechanism of amyloidogenesis through a tandem repeat of the PI3-SH3 domain suggests a generic model for protein aggregation and fibril formation.

Aggregation of the SH3 domain of the PI3 kinase, both as a single domain and as a tandem repeat in which the C terminus of one domain is linked to the N terminus of another by a flexible linker of ten glycine/serine residues, has been studied under a range of conditions in order to investigate the mechanism of protein aggregation and amyloid formation. The tandem repeat was found to form amyloid fibrils much more readily than the single domain under the acidic conditions used here, and the fibrils themselves have higher morphological homogeneity. The folding-unfolding transition of the PI3-SH3 domain shows two-state behaviour and is pH dependent; at pH 3.6, which is near the pH mid-point for folding and only slightly below the isoelectric point of the protein, both the single domain and the tandem repeat spontaneously form broad distributions of soluble oligomers without requirement for nucleation. Under prolonged incubation under these conditions, the oligomers convert into thin, curly fibrils that interact with thioflavin-T, suggesting that they contain an organised beta-sheet structure. Under more acidic conditions (pH 2.0) where the proteins are fully denatured and carry a positive net charge, long, straight fibrils are formed in a process having a pronounced lag phase. The latter was found to be reduced dramatically by the addition of oligomers exceeding a critical size of approximately 20 molecules. The results suggest that the process of aggregation of these SH3 domains can take place by a variety of mechanisms, ranging from downhill formation of relatively amorphous species to nucleated formation of highly organised structures, the relative importance of which varies greatly with solution conditions. Comparison with the behaviour of other amyloidogenic systems suggests that the general mechanistic features outlined here are likely to be common to at least a wide variety of peptides and proteins.

Molecular recycling within amyloid fibrils.

Amyloid fibrils are thread-like protein aggregates with a core region formed from repetitive arrays of beta-sheets oriented parallel to the fibril axis. Such structures were first recognized in clinical disorders, but more recently have also been linked to a variety of non-pathogenic phenomena ranging from the transfer of genetic information to synaptic changes associated with memory. The observation that many proteins can convert into similar structures in vitro has suggested that this ability is a generic feature of polypeptide chains. Here we have probed the nature of the amyloid structure by monitoring hydrogen/deuterium exchange in fibrils formed from an SH3 domain using a combination of nuclear magnetic resonance spectroscopy and electrospray ionization mass spectrometry. The results reveal that under the conditions used in this study, exchange is dominated by a mechanism of dissociation and re-association that results in the recycling of molecules within the fibril population. This insight into the dynamic nature of amyloid fibrils, and the ability to determine the parameters that define this behaviour, have important implications for the design of therapeutic strategies directed against amyloid disease.

Rational design of aggregation-resistant bioactive peptides: reengineering human calcitonin.

A high propensity to aggregate into intractable deposits is a common problem limiting the production and use of many peptides and proteins in a wide range of biotechnological and pharmaceutical applications. Many therapeutic polypeptides are frequently abandoned at an early stage in their development because of problems with stability and aggregation. It has been shown recently that parameters describing the physicochemical properties of polypeptides can be used as predictors of protein aggregation. Here we demonstrate that these and similar tools can be applied to the rational redesign of bioactive molecules with a significantly reduced aggregation propensity without loss of physiological activity. This strategy has been exemplified by designing variants of the hormone calcitonin that show a significantly reduced aggregation propensity, yet maintain, or even increase, their potency when compared to the current therapeutic forms. The results suggest that this approach could be used successfully to enhance the solubility and efficacy of a wide range of other peptide and protein therapeutics.

Prediction of "aggregation-prone" and "aggregation-susceptible" regions in proteins associated with neurodegenerative diseases.

Increasing evidence indicates that many peptides and proteins can be converted in vitro into highly organised amyloid structures, provided that the appropriate experimental conditions can be found. In this work, we define intrinsic propensities for the aggregation of individual amino acids and develop a method for identifying the regions of the sequence of an unfolded peptide or protein that are most important for promoting amyloid formation. This method is applied to the study of three polypeptides associated with neurodegenerative diseases, Abeta42, alpha-synuclein and tau. In order to validate the approach, we compare the regions of proteins that are predicted to be most important in driving aggregation, either intrinsically or as the result of mutations, with those determined experimentally. The knowledge of the location and the type of the "sensitive regions" for aggregation is important both for rationalising the effects of sequence changes on the aggregation of polypeptide chains and for the development of targeted strategies to combat diseases associated with amyloid formation.

Polypeptide models to understand misfolding and amyloidogenesis and their relevance in protein design and therapeutics.

The study of amyloid polypeptide models (polypeptides able to generate amyloid structures not necessarily connected with any pathology) provides an excellent tool to increase the understanding of the generic aspects of misfolding and aggregation as well as the details of the mechanism of polypeptide deposition in disease. This knowledge can be integrated and applied to different problems in therapy and biotechnology, and in particular to re-designing bio-active polypeptides (biopharmaceuticals) with improved properties.

Prediction of the absolute aggregation rates of amyloidogenic polypeptide chains.

Protein aggregation is associated with a variety of pathological conditions, including Alzheimer's and Creutzfeldt-Jakob diseases and type II diabetes. Such degenerative disorders result from the conversion of the normal soluble state of specific proteins into aggregated states that can ultimately form the characteristic amyloid fibrils found in diseased tissue. Under appropriate conditions it appears that many, perhaps all, proteins can be converted in vitro into amyloid fibrils. The aggregation propensities of different polypeptide chains have, however, been observed to vary substantially. Here, we describe an approach that uses the knowledge of the amino acid sequence and of the experimental conditions to reproduce, with a correlation coefficient of 0.92 and over five orders of magnitude, the in vitro aggregation rates of a wide range of unstructured peptides and proteins. These results indicate that the formation of protein aggregates can be rationalised to a considerable extent in terms of simple physico-chemical parameters that describe the properties of polypeptide chains and their environment.

A highly amyloidogenic region of hen lysozyme.

Amyloid fibrils obtained after incubating hen egg-white lysozyme (HEWL) at pH 2.0 and 65 degrees C for extended periods of time have been found to consist predominantly of fragments of the protein corresponding to residues 49-100, 49-101, 53-100 and 53-101, derived largely from the partial acid hydrolysis of Asp-X peptide bonds. These internal fragments of HEWL encompass part of the beta-domain and all the residues forming the C-helix in the native protein, and contain two internal disulfide bridges Cys64-Cys80 and Cys76-Cys94. The complementary protein fragments, including helices A, B and D of the native protein, are not significantly incorporated into the network of fibrils, but remain largely soluble, in agreement with their predicted lower propensities to aggregate. Further analysis of the properties of different regions of HEWL to form amyloid fibrils was carried out by studying fragments produced by limited proteolysis of the protein by pepsin. Here, we show that only fragment 57-107, but not fragment 1-38/108-129, is able to generate well-defined amyloid fibrils under the conditions used. This finding is of particular importance, as the beta-domain and C-helix of the highly homologous human lysozyme have been shown to unfold locally in the amyloidogenic variant D67H, which is associated with the familial cases of systemic amyloidosis linked to lysozyme deposition. The identification of the highly amyloidogenic character of this region of the polypeptide chain provides strong support for the involvement of partially unfolded species in the initiation of the aggregation events that lead to amyloid deposition in clinical disease.

Short amino acid stretches can mediate amyloid formation in globular proteins: the Src homology 3 (SH3) case.

Protein misfolding and deposition underlie an increasing number of debilitating human disorders. We have shown that model proteins unrelated to disease, such as the Src homology 3 (SH3) domain of the p58alpha subunit of bovine phosphatidyl-inositol-3'-kinase (PI3-SH3), can be converted in vitro into assemblies with structural and cytotoxic properties similar to those of pathological aggregates. By contrast, homologous proteins, such as alpha-spectrin-SH3, lack the capability of forming amyloid fibrils at a measurable rate under any of the conditions we have so far examined. However, transplanting a small sequence stretch (6 aa) from PI3-SH3 to alpha-spectrin-SH3, comprising residues of the diverging turn and adjacent RT loop, creates an amyloidogenic protein closely similar in its behavior to the original PI3-SH3. Analysis of specific PI3-SH3 mutants further confirms the involvement of this region in conferring amyloidogenic properties to this domain. Moreover, the inclusion in this stretch of two consensus residues favored in SH3 sequences substantially inhibits aggregation. These findings show that short specific amino acid stretches can act as mediators or facilitators in the incorporation of globular proteins into amyloid structures, and they support the suggestion that natural protein sequences have evolved in part to code for structural characteristics other than those included in the native fold, such as avoidance of aggregation.

Exploring amyloid formation by a de novo design.

Protein deposition as amyloid fibrils underlies many debilitating human disorders. The complexity and size of disease-related polypeptides, however, often hinders a detailed rational approach to study effects that contribute to the process of amyloid formation. We report here a simplified peptide sequence successfully designed de novo to fold into a coiled-coil conformation under ambient conditions but to transform into amyloid fibrils at elevated temperatures. We have determined the crystal structure of the coiled-coil form and propose a detailed molecular model for the peptide in its fibrillar state. The relative stabilities of the two structural forms and the kinetics of their interconversion were found to be highly sensitive to small sequence changes. The results reveal the importance of specific packing interactions on the kinetics of amyloid formation and show the potential of this exceptionally favorable system for probing details of the molecular origins of amyloid disease.

Protein aggregation and amyloid fibril formation by an SH3 domain probed by limited proteolysis.

The SH3 domains are small protein modules of 60-85 amino acid residues that are found in many proteins involved in intracellular signal transduction. The SH3 domain of the p85alpha subunit of bovine phosphatidylinositol 3'-kinase (PI3-SH3) under acidic solution adopts a compact denatured state from which amyloid fibrils are readily formed. This aggregation process has been found to be modulated substantially by solution conditions. Here, we have analyzed the conformational features of the native and acid denatured states of PI3-SH3 by limited proteolysis experiments using proteinase K and pepsin, respectively. Moreover, we have analyzed the propensity of PI3-SH3 to be hydrolyzed by pepsin at different stages in the process of aggregation and amyloid formation at pH 1.2 and 2.0 and compared the sites of proteolysis under these conditions with the conformational features of both native and aggregated PI3-SH3. The results demonstrate that the denatured state of PI3-SH3 formed at low pH is relatively resistant to proteolysis, indicating that it is partially folded. The long loop connecting beta-strands b and c in the native protein is the region in this structure most susceptible to proteolysis. Remarkably, aggregates of PI3-SH3 that are formed initially from this denatured state in acid solution display enhanced susceptibility to proteolysis of the long loop, suggesting that the protein becomes more unfolded in the early stages of aggregation. By contrast, the more defined amyloid fibrils that are formed over longer periods of time are completely resistant to proteolysis. We suggest that the protein aggregates formed initially are relatively dynamic species that are able readily to reorganize their interactions to enable formation of very well ordered fibrillar structures. In addition, the disordered and non-native character of the polypeptide chains in the early aggregates could be important in determining the high cytotoxicity that has been revealed in previous studies of these species.

Protein folding and misfolding: a paradigm of self-assembly and regulation in complex biological systems.

Understanding biological complexity is one of the grand scientific challenges for the future. A living organism is a highly evolved system made up of a large number of interwoven molecular networks. These networks primarily involve proteins, the macromolecules that enable and control virtually every chemical process that takes place in the cell. Proteins are also key elements in the essential characteristic of living systems, their ability to function and replicate themselves through controlled molecular interactions. Recent progress in understanding the most fundamental aspect of polypeptide self-organization, the process by which proteins fold to attain their active conformations, provides a global platform to gain knowledge about the function of biological systems and the regulatory mechanisms that underpin their ability to adapt to changing conditions. In order to exploit such progress effectively, we are developing a variety of approaches, including procedures that use experimental data to restrain the properties of complex systems in computer simulations, to describe their behaviour under a wide variety of conditions. We believe that such approaches can lead to significant advances in understanding biological complexity, in general, and protein folding and misfolding in particular. These advances would contribute to: a more effective exploitation of the information from genome sequences; more rational therapeutic approaches to diseases, particularly those associated with ageing; the responsible control of our own evolution; and the development of new technologies based on mimicking the principles of biological self-assembly, for instance in nanotechnology. More fundamentally, we believe that this research will result in a more coherent understanding of the origin, evolution and functional properties of living systems.

Competing intrachain interactions regulate the formation of beta-sheet fibrils in bovine PrP peptides.

At the heart of the pathogenesis of transmissible spongiform encephalopathies (TSEs), such as BSE, scrapie, and Creutzfeldt-Jakob disease, lies a poorly understood structural rearrangement of PrP, an abundant glycoprotein of the nervous and lymphoid systems. The normal form (PrP(C)), rich in alpha-helix, converts into an aberrant beta-sheet-dominated form (PrP(Sc)), which seems to be at the center of the pathotoxic symptoms observed in TSEs. To understand this process better at a molecular level, we have studied the interactions between different peptides derived from bovine PrP and their structural significance. We show that two unstructured peptides derived from the central region of bovine PrP, residues 115-133 and 140-152, respectively, interact stoichiometrically under physiological conditions to generate beta-sheet-dominated fibrils. However, when both peptides are incubated in the presence of a third peptide derived from an adjoining alpha-helical region (residues 153-169), the formation of beta-sheet-rich fibrils is abolished. These data indicate that native PrP(C) helix 1 might inhibit the strong intrinsic beta-sheet-forming propensity of sequences immediately N-terminal to the globular core of PrP(C), by keeping in place intrachain interactions that would prevent these amyloidogenic regions from triggering aggregation. Moreover, these results indicate new ways in which PrP(Sc) formation could be prevented.

Transition from natively unfolded to folded state induced by desiccation in an anhydrobiotic nematode protein.

Late embryogenesis abundant (LEA) proteins are associated with desiccation tolerance in resurrection plants and in plant seeds, and the recent discovery of a dehydration-induced Group 3 LEA-like gene in the nematode Aphelenchus avenae suggests a similar association in anhydrobiotic animals. Despite their importance, little is known about the structure of Group 3 LEA proteins, although computer modeling and secondary structure algorithms predict a largely alpha-helical monomer that forms coiled coil oligomers. We have therefore investigated the structure of the nematode protein, AavLEA1, in the first such analysis of a well characterized Group 3 LEA-like protein. Immunoblotting and subunit cross-linking experiments demonstrate limited oligomerization of AavLEA1, but analytical ultracentrifugation and gel filtration show that the vast majority of the protein is monomeric. Moreover, CD, fluorescence emission, and Fourier transform-infrared spectroscopy indicate an unstructured conformation for the nematode protein. Therefore, in solution, no evidence was found to support structure predictions; instead, AavLEA1 seems to be natively unfolded with a high degree of hydration and low compactness. Such proteins can, however, be induced to fold into more rigid structures by partner molecules or by altered physiological conditions. Because AavLEA1 is associated with desiccation stress, its Fourier transform-infrared spectrum in the dehydrated state was examined. A dramatic but reversible increase in alpha-helix and, possibly, coiled coil formation was observed on drying, indicating that computer predictions of secondary structure may be correct for the solid state. This unusual finding offers the possibility that structural shifts in Group 3 LEA proteins occur on dehydration, perhaps consistent with their role in anhydrobiosis.

De novo designed peptide-based amyloid fibrils.

Identification of therapeutic strategies to prevent or cure diseases associated with amyloid fibril deposition in tissue (Alzheimer's disease, spongiform encephalopathies, etc.) requires a rational understanding of the driving forces involved in the formation of these organized assemblies rich in beta-sheet structure. To this end, we used a computer-designed algorithm to search for hexapeptide sequences with a high propensity to form homopolymeric beta-sheets. Sequences predicted to be highly favorable on this basis were found experimentally to self-associate efficiently into beta-sheets, whereas point mutations predicted to be unfavorable for this structure inhibited polymerization. However, the property to form polymeric beta-sheets is not a sufficient requirement for fibril formation because, under the conditions used here, preformed beta-sheets from these peptides with charged residues form well defined fibrils only if the total net charge of the molecule is +/-1. This finding illustrates the delicate balance of interactions involved in the formation of fibrils relative to more disordered aggregates. The present results, in conjunction with x-ray fiber diffraction, electron microscopy, and Fourier transform infrared measurements, have allowed us to propose a detailed structural model of the fibrils.

Inherent toxicity of aggregates implies a common mechanism for protein misfolding diseases.

A range of human degenerative conditions, including Alzheimer's disease, light-chain amyloidosis and the spongiform encephalopathies, is associated with the deposition in tissue of proteinaceous aggregates known as amyloid fibrils or plaques. It has been shown previously that fibrillar aggregates that are closely similar to those associated with clinical amyloidoses can be formed in vitro from proteins not connected with these diseases, including the SH3 domain from bovine phosphatidyl-inositol-3'-kinase and the amino-terminal domain of the Escherichia coli HypF protein. Here we show that species formed early in the aggregation of these non-disease-associated proteins can be inherently highly cytotoxic. This finding provides added evidence that avoidance of protein aggregation is crucial for the preservation of biological function and suggests common features in the origins of this family of protein deposition diseases.