Many human disorders, such as Alzheimer’s
and Parkinson’s, are characterized by misfolded proteins that aggregate
(combine). It is important to understand the aggregation of proteins when creating
soluble protein-based drugs. Therefore, methods that can predict the properties
of protein aggregation are essential. Within the past decade, various tools
have been developed to anticipate protein aggregation. Ventura et. al developed
AGGRESCAN, which is the first tool that has relied on experimental patterns of
aggregation on a cellular level. Mutants of the amyloid β-peptide (Aβ) were
used to create the algorithm. Aggregation in the cytoplasm of E. coli was studied. Each mutation
affects the peptide aggregation differently. The various aggregation patterns
can be used to determine the likelihood of aggregation for the natural amino
acids. AGGRESCAN is capable of using this information to determine the tendency
of aggregation for each protein based on its relative position in the sequence.
The algorithm has been commonly used due to its accuracy in predicting in vivo aggregation.
The
majority of the methods used to predict protein aggregation analyze linear
sequences. Hence, they assume that the protein of interest is partially
unstructured. For folded globular proteins, these regions may be blocked and
therefore do not have a significant impact on aggregation. The authors
introduce AGGRESCAN3D (A3D), which has evolved from the AGGRESCAN method. A3D
allows the aggregation properties of globular proteins to be predicted because
it is able to detect spatial relationships. A3D includes a “Dynamic Mode” that
can be used, which considers the different structures that the input structure
can have in the predicted aggregation patterns. Therefore, when wild type
structures and destabilizing pathogenic mutations change in structure, A3D can
account for the variations and model the aggregation patterns that stem from
those variations in structure.
A3D is
documented online on a server and can be used to predict the likelihood of
aggregation for globular proteins. The algorithm can be used to anticipate the
effect of mutations in conformational disorders. It can also be used to design
soluble protein-based drugs that can be used as treatment for disorders
characterized by protein aggregation. Therefore, it provides the possibility of
more effective treatment for patients with disorders like Alzheimer’s and
Parkinson’s.
Reference:
R. Zambrano, M.
Jamroz, A. Szczasiuk, J. Pujols, S. Kmiecik, S. Ventura. “AGGRESCAN3D
(A3D): server for prediction of aggregation properties of protein structures”.
Nucleic Acids Research, 2015; DOI: https://doi.org/10.1093/nar/gkv359
