Immunoglobulin light chain (AL) amyloidosis is a hematologic disorder caused by a generally small, poorly proliferating plasma cell (PC) clone which resides in the bone marrow and produces immunoglobulin (Ig) light chains (LCs) that circulate in the bloodstream, misfold and aggregate as extracellular amyloid fibrils. The process of amyloid formation and deposition can lead to cell death, subversion of tissue architecture, organ dysfunction and death. The primary sequence of the disease-causing LC is patient-unique and dictates the protein’s propensity to aggregate and possibly its organ tropism and toxicity. Since the PC clone is small and difficult to detect, the advent of efficient next-generation sequencing (NGS) techniques to sequence Ig genes may improve our understanding of the molecular mechanisms of the disease as well as the disease management. Our laboratory has recently established an NGS-based technique to sequence the complete variable (V) region of Ig genes from multiple bulk samples in parallel. Here, we established a pipeline of bioinformatics analyses to identify and characterize the V region of both the clonal Ig gene produced by a bone marrow PC clone and of by standing polyclonal Ig genes present in the same biological sample. As a proof of concept, we have applied it first in the context of a validation data set exploring the accuracy, sensitivity and reproducibility of this approach. Sensitivity was evaluated by serial dilutions of RNA extracted from two human cell lines (one lambda- and one kappa-expressing LCs) in the RNA extracted from a healthy donor lacking PC clones. Reproducibility was evaluated by the realization of five replicates of two AL patients (one lambda and one kappa) and duplicates from three additional AL lambda patients. These studies demonstrated that our newly established NGS-based technique coupled with bioinformatics analyses can accurately identify the complete sequence of the V region of dominant, clonal Ig genes, with the obtained sequences showing 100% identity with sequences identified by conventional methods, with a sensitivity that is dictated by the number of reads achieved during sequencing and with high reproducibility also in terms of defining the molecular clonal size, as a measure of the relative frequency of clonal reads in a given sample. Next, we applied the NGS technique to a cohort of 83 patients affected by AL amyloidosis. We determined the molecular clonal size and the dominant Ig LC sequence in these patients, assigned the V germline gene used through immunogenetics analyses and predicted the occurrence of N-glycosylation in silico. These analyses confirmed published knowledge on the relative frequency of specific Ig germline genes in AL amyloidosis, as well as a high rate of putative N-glycosylation of amyloidogenic kappa LCs. We confirmed known associations of specific LC sequence features with clinical parameters in AL (including association of IGLV1-44 with heart involvement) and found novel associations (including association of putative N-glycosylation with higher clonal Ig LC levels or liver involvement). We also explored germline gene usage and N-glycosylation pattern within the by standing, polyclonal LC repertoire in these patients. Our novel NGS approach may facilitate obtaining clonal Ig gene sequence(s) at diagnosis even from small, but dangerous B cell clones. This will significantly expand our currently limited knowledge of molecular determinants of LC pathogenicity. Additionally, knowledge of the underlying clonal Ig gene sequence may facilitate the sensitive detection of residual tumoral PCs resisting chemotherapy during minimal residual disease assessment. Besides AL amyloidosis, this strategy may be applicable to sequence and detect Ig involved in other PC dyscrasias or B cell disorders.
L’amiloidosi sistemica da catene leggere immunoglobuliniche (AL) è un disordine ematologico causato da un clone plasmacellulare (PC) generalmente piccolo e poco proliferante che risiede nel midollo osseo che produce catene leggere (LC) immunoglobuliniche (Ig) che circolano nel sangue, si mal ripiegano e aggregano in fibrille di amiloide extracellulari. Il processo di formazione e deposizione di amiloide può portare a morte cellulare, alterazione dell’architettura tissutale, disfunzione d’organo e morte. La sequenza primaria delle LC è paziente-specifica e ne determina la propensità ad aggregare e il suo possibile tropismo d’organo e tossicità. Poiché il clone plasmacellulare è difficile da rilevare, l’avvento di tecniche NGS per sequenziare geni Ig potrebbe migliorare la nostra comprensione dei meccanismi molecolari della malattia e il suo management. Il nostro laboratorio ha di recente sviluppato una tecnica NGS per sequenziare la regione variabile (V) completa di geni Ig di più campioni in parallelo. In questo studio abbiamo sviluppato una piattaforma bioinformatica per identificare e caratterizzare la regione V sia di geni Ig clonali da un clone PC midollare che di geni Ig policlonali presenti nello stesso campione. In primis abbiamo applicato la tecnica nel contesto di un set di dati di validazione per valutarne accuratezza, sensibilità e riproducibilità. La sensibilità è stata valutata tramite diluizioni seriali di RNA estratto da linee cellulari umane (una esprimente LC lambda, l’altra LC kappa) nell’RNA di un donatore sano privo di cloni PC, mentre la riproducibilità è stata valutata tramite la realizzazione di cinque replicati di due pazienti AL (uno lambda e uno kappa) e duplicati di altri tre pazienti AL lambda. Questi studi hanno dimostrato che la nostra tecnica NGS, insieme ad analisi bioinformatiche, può identificare accuratamente l’intera sequenza della regione V di geni Ig clonali dominanti, mostrando 100% di identità con le sequenze ottenute con metodi convenzionali, con una sensibilità data dal numero di reads ottenute dal sequenziamento e con un’alta riproducibilità anche nel determinare le dimensioni molecolari del clone, misurate come la frequenza relativa di reads clonali di un dato campione. Quindi abbiamo applicato la metodica NGS ad una coorte di 83 pazienti con amiloidosi AL. Abbiamo determinato le dimensioni molecolari del clone e la sequenza dominante della LC Ig, abbiamo assegnato il gene codificante la regione V tramite analisi immunogenetiche e abbiamo predetto la presenza di N-glicosilazione in silico. Queste analisi hanno confermato le frequenze relative di specifici geni Ig nell’amiloidosi AL già note, così come l’alto tasso di una potenziale N-glicosilazione delle LC kappa amiloidogeniche. Abbiamo confermato associazioni note tra le caratteristiche di specifiche sequenze delle LC con parametri clinici nell’AL (inclusa l’associazione di IGLV1-44 con il cuore) e abbiamo trovato nuove associazioni (incluse l’associazione tra la presenza di una potenziale N-glicosilazione con alti livelli di LC Ig clonali o con il fegato). Abbiamo anche esplorato i geni Ig espressi e la N-glicosilazione all’interno del repertorio delle LC policlonali di questi pazienti. Il nostro innovativo approccio NGS potrebbe facilitare l’identificazione di sequenze Ig clonali da tumori PC, anche piccoli. Ciò potrebbe ampliare le nostre attuali conoscenze sulle sequenze delle LC amiloidogeniche e sui meccanismi molecolari alla base della malattia. Inoltre la conoscenza della sequenza genica Ig clonale potrebbe facilitare il rilevamento, con elevata sensibilità, di cloni PC tumorali residui, resistenti alla chemioterapia, durante la valutazione della malattia residua minima. Oltre che nell’amiloidosi AL, questa strategia potrebbe essere utilizzata per sequenziare e rilevare Ig coinvolte in altre discrasie plasmacellulari e disordini delle cellule B.
Diagnostic potential of an NGS-based antibody repertoire analysis in AL amyloidosis. (Potenziale diagnostico di un'analisi del repertorio anticorpale basata su una nuova metodica NGS nell'amiloidosi AL)
PISCITELLI, MAGGIE
2019/2020
Abstract
Immunoglobulin light chain (AL) amyloidosis is a hematologic disorder caused by a generally small, poorly proliferating plasma cell (PC) clone which resides in the bone marrow and produces immunoglobulin (Ig) light chains (LCs) that circulate in the bloodstream, misfold and aggregate as extracellular amyloid fibrils. The process of amyloid formation and deposition can lead to cell death, subversion of tissue architecture, organ dysfunction and death. The primary sequence of the disease-causing LC is patient-unique and dictates the protein’s propensity to aggregate and possibly its organ tropism and toxicity. Since the PC clone is small and difficult to detect, the advent of efficient next-generation sequencing (NGS) techniques to sequence Ig genes may improve our understanding of the molecular mechanisms of the disease as well as the disease management. Our laboratory has recently established an NGS-based technique to sequence the complete variable (V) region of Ig genes from multiple bulk samples in parallel. Here, we established a pipeline of bioinformatics analyses to identify and characterize the V region of both the clonal Ig gene produced by a bone marrow PC clone and of by standing polyclonal Ig genes present in the same biological sample. As a proof of concept, we have applied it first in the context of a validation data set exploring the accuracy, sensitivity and reproducibility of this approach. Sensitivity was evaluated by serial dilutions of RNA extracted from two human cell lines (one lambda- and one kappa-expressing LCs) in the RNA extracted from a healthy donor lacking PC clones. Reproducibility was evaluated by the realization of five replicates of two AL patients (one lambda and one kappa) and duplicates from three additional AL lambda patients. These studies demonstrated that our newly established NGS-based technique coupled with bioinformatics analyses can accurately identify the complete sequence of the V region of dominant, clonal Ig genes, with the obtained sequences showing 100% identity with sequences identified by conventional methods, with a sensitivity that is dictated by the number of reads achieved during sequencing and with high reproducibility also in terms of defining the molecular clonal size, as a measure of the relative frequency of clonal reads in a given sample. Next, we applied the NGS technique to a cohort of 83 patients affected by AL amyloidosis. We determined the molecular clonal size and the dominant Ig LC sequence in these patients, assigned the V germline gene used through immunogenetics analyses and predicted the occurrence of N-glycosylation in silico. These analyses confirmed published knowledge on the relative frequency of specific Ig germline genes in AL amyloidosis, as well as a high rate of putative N-glycosylation of amyloidogenic kappa LCs. We confirmed known associations of specific LC sequence features with clinical parameters in AL (including association of IGLV1-44 with heart involvement) and found novel associations (including association of putative N-glycosylation with higher clonal Ig LC levels or liver involvement). We also explored germline gene usage and N-glycosylation pattern within the by standing, polyclonal LC repertoire in these patients. Our novel NGS approach may facilitate obtaining clonal Ig gene sequence(s) at diagnosis even from small, but dangerous B cell clones. This will significantly expand our currently limited knowledge of molecular determinants of LC pathogenicity. Additionally, knowledge of the underlying clonal Ig gene sequence may facilitate the sensitive detection of residual tumoral PCs resisting chemotherapy during minimal residual disease assessment. Besides AL amyloidosis, this strategy may be applicable to sequence and detect Ig involved in other PC dyscrasias or B cell disorders.È consentito all'utente scaricare e condividere i documenti disponibili a testo pieno in UNITESI UNIPV nel rispetto della licenza Creative Commons del tipo CC BY NC ND.
Per maggiori informazioni e per verifiche sull'eventuale disponibilità del file scrivere a: unitesi@unipv.it.
https://hdl.handle.net/20.500.14239/12079