Exploring the sequence features determining amyloidosis in human antibody light chains

Abstract

The light chain (AL) amyloidosis is caused by the aggregation of light chain of antibodies into amyloid fibrils. There are plenty of computational resources available for the prediction of short aggregation-prone regions within proteins. However, it is still a challenging task to predict the amyloidogenic nature of the whole protein using sequence/structure information. In the case of antibody light chains, common architecture and known binding sites can provide vital information for the prediction of amyloidogenicity at physiological conditions. Here, in this work, we have compared classical sequence-based, aggregation-related features (such as hydrophobicity, presence of gatekeeper residues, disorderness, β-propensity, etc.) calculated for the CDR, FR or VL regions of amyloidogenic and non-amyloidogenic antibody light chains and implemented the insights gained in a machine learning-based webserver called “VLAmY-Pred” (https://web.iitm.ac.in/bioinfo2/vlamy-pred/). The model shows prediction accuracy of 79.7% (sensitivity: 78.7% and specificity: 79.9%) with a ROC value of 0.88 on a dataset of 1828 variable region sequences of the antibody light chains. This model will be helpful towards improved prognosis for patients that may likely suffer from diseases caused by light chain amyloidosis, understanding origins of aggregation in antibody-based biotherapeutics, large-scale in-silico analysis of antibody sequences generated by next generation sequencing, and finally towards rational engineering of aggregation resistant antibodies.

Widespread amyloidogenicity potential of multiple myeloma patient-derived immunoglobulin light chains

“…The tested algorithms did not identify any clear distinctions between amyloidogenic and nonamyloid forming light chains (see Additional file 4: Fig. S6), which is in accordance with a recent analysis, which could also not identify a significant difference in the amyloid propensity between amyloidogenic and non-amyloidogenic IgLCs using Tango and Waltz [60]. The need for improved prediction algorithms for the specific case of IgLCs is increasingly recognised and has led to the recent development of two new tools, LICTOR [61] and V L AmY-Pred [61].…”

Progress and challenges for the machine learning-based design of fit-for-purpose monoclonal antibodies

“…We detail the methodology used in each tool, the corresponding lower-level developability parameters, and method availability Method name Methodology/approach Main low-level parameter(s) Availability Thermal stability Spatial aggregation propensity (SAP) 229 Custom code, MD simulation Surface hydrophobicity Mathematical equation Bekker et al 230 MD simulation Fraction of native contacts (Q-value) NA ANN model 231 ML model AA composition NA Aggregation Developability Index (DI) 232 Custom code Charge, spatial aggregation propensity (SAP) Mathematical equation AbsoluRATE 233 ML model for aggregation kinetics prediction Environmental conditions, disorderness, aggregation related properties etc. https://web.iitm.ac.in/bioinfo2/absolurate-pred/ Therapeutic Antibody Profiler (TAP) 213 Developability rules as per authors CDR length, surface hydrophobicity, charge http://opig.stats.ox.ac.uk/webapps/newsabdab/sabpred/tap V L AmY-Pred 234 ML model Charge, hydrophobicity, Disorderness, β-propensity https://web.iitm.ac.in/bioinfo2/vlamy-pred/ Solubis 235 Custom code ...…”

“…Several studies on antibodies have identified potential aggregation-prone regions in the relatively exposed CDR of the VH domains. 234 , 284–286 However, a recent study observed that almost all the latest APR prediction algorithms perform poorly on identifying aggregation. 287 This can be attributed to (1) limited overall variations in the antibody sequence (except for CDRs) leading to higher sequence conservation and (2) low sensitivity of these algorithms toward similar protein sequences.…”

“…Specifically, TAP scores take into account developability factors, such as the length of the CDRs, hydrophobicity, and the presence of charge patches, which are linked to polyspecificity, aggregation, and viscosity of mAb preparations. Rawat et al 234 developed an ML-based light chain aggregation prediction method and highlighted that lambda light chains are inherently more aggregation-prone. Notably, most approved IgG mAbs harbor the kappa light chain, and the pool of human IgG in serum has about 2-fold more of kappa light chain than the lambda.…”

Understanding AL amyloidosis with a little help from in vivo models

“…However, despite the multiple LC sequences available in the literature ( 9 , 13 , 14 ) or databases like AL base ( ), it is still impossible to predict the amyloidogenicity of an LC solely based on its amino acid sequence since each LC presents with multiple combinations of mutations that can account for their aggregation propensity. Several predictive tools have been developed based on the physicochemical properties of protein sequences ( 17 – 19 ) and, more recently, have been systematized using machine learning approaches ( 20 , 21 ), but they still fail to predict all amyloid LC sequences. They also deserve to be tested with other aggregation prone LCs (from light chain deposition diseases, light chain proximal tubulopathy with crystals, etc.)…”