Cases will be selected based on their assignment to major subsets #1, #2 and #8 and their satellites in such a way to ensure that both genders are equally represented. The study group will comprise 10 cases from each individual major subset and 10 cases from each individual satellite subset, respectively; regarding the latter, selection will be based on the assignment to subsets #99 (subset #1 satellite), #169 (subset #2 satellite) and #8B (subset #8 satellite), respectively. CLL cells will be negatively selected from all cases using a magnetic negative selection kit (EasySep™ Human B Cell Enrichment Kit). DNA and RNA will be extracted (AllPrep DNA/RNA Kit) and cDNA will be synthesized from all samples (SuperScript™ II).
BcR IG heavy and light chain gene rearrangements will be amplified in all samples of the study group (n=60 samples) using protocols developed by the Euroclonality-NGS Working Group for NGS immunogenetic studies. The quantity and quality of PCR products will be assessed with Qubit® (ThermoFisher Scientific) and Fragment Analyzer (Automated CE System), respectively. Library preparation will be performed with the NEBNext® Ultra™ II DNA Library Prep Kit for Illumina (NewEngland Biolabs). Sequencing will be implemented on the MiSeq platform (2x300bp) (Illumina). NGS BcR IG gene data will be annotated with IMGT/HighV-QUEST. Metadata will be analyzed with tripr, providing information about the clonal composition of each individual sample. Intraclonal diversification will be analyzed with IgIDivA, an innovative approach based on the assessment of graph network metrics. Three-dimensional modelling will be also applied to assess the impact of the immunogenetic properties at the structural and, thus, functional level of the clonotypic BcR IG, using an established workflow.
The whole exome of selected CLL patient samples will be targeted; in particular, the study group for this WP will comprise 5 cases from each individual major subset and 5 cases from each satellite, respectively. Sample selection will be based on the results from WP2 in terms of immunogenetic homogeneity. For each patient, WES will be performed on DNA from CLL and buccal (control) cells (n= 60 samples), with an average autosomal coverage of around 100X. In more detail, the Twist Comprehensive Exome Panel will be used, offering a coverage of greater than 99% of protein coding genes (Twist Bioscience). In terms of gDNA library preparation, the Twist Library Preparation Kit will be used. Analysis of raw reads will be performed with a bioinformatics pipeline, previously applied by our group. Sequencing reads will be aligned to the hg19 reference genome using BWA. BAM-files will be locally realigned around small insertions and deletions (indels) using GATK. Single nucleotide variants (SNV) and indels will be identified with FreeBayes. Somatic copy number alterations (sCNA) will be analyzed with Control-FREEC. Analysis of the mutational spectra will be performed using the MutationalPatterns package.
Library preparation for RNA samples will be also performed in representative samples from each stereotyped subset and its satellite (the same that will be subjected to WES) using the NEBNext ® Ultra™ II Directional RNA Library Prep Kit (NE Biolabs) and quantification with the KAPA SYBR FAST Universal qPCR kit (KAPA Biosystems), respectively. Paired-end sequencing will take place on a NextSeq 2000 platform (Illumina). Sequencing reads will be quality-controlled using the FastQC software (Babraham Institute, Cambridge, UK). Subsequently, reads will be trimmed and filtered using the Trim Galore! Suite and aligned against the UCSC hg19 reference genome using HISAT2. Differential gene expression analysis will be performed in R v4.2.0 using the limma package. Comparison will be performed between cases belonging to each major stereotyped subset and its satellite along with publicly available data from CLL patients. Finally, gene enrichment analysis will be performed with Enrichr using the KEGG 2021 gene database.
The reactivity profiles of BcRs from representative cases (as mentioned in WP3, n= 30 samples) will be examined. Antigens from CLL-associated classes (pathogens and autoantigens) will be included in the analysis. Cloning, expression, and purification of CLL rmAbs will be performed according to established methods, extensively used by our group. In detail, 293A human embryonic kidney fibroblasts will be cultured in DMEM supplemented with 10% ultra-low IgG FCS and co-transfected with plasmid DNA encoding the BcR IG heavy and light chain gene rearrangements. Transfected cells will be washed with serum-free DMEM after 8 hours and then will be cultured in DMEM supplemented with 1% Nutridoma SP (Roche). After 8 days of culture, the supernatants containing the rmAbs will be collected. BcR IG reactivity will be determined with the use of protein assays (GeneCopoeia's OmicsArray™). Primary rmAbs will be incubated with the antigen arrays. Bound antibodies will be measured with cy5-conjugated anti-human IgM (1:2000, Jackson ImmunoResearch Laboratories), using a GenePix® 4000B scanner. The fluorescent signals from the array will be analyzed using the GenePix Pro v7.0 software. Results will be validated with the use of Enzyme-linked immunosorbent assays (ELISA) using a TECAN Spark microplate reader. CLL-associated pathogens and autoantigens will be used at a concentration of 5 μg/mL. CLL mAbs will be used as primary antibodies at a concentration of 20 μg/mL or substituting patient serum at 50 μg/mL in commercial ELISA kits.
Big biodata from all WPs, including omics data (immunogenetics, genomics and transcriptomics) and BcR reactivity data will be deposited in a purpose-built database. To ensure scalability of the database, an object-oriented schema will be applied, allowing for easy integration and mapping across the multi-type data, even in cases of missing/non-available data points. Data generated from the same patient will be integrated towards building the complete biological profile of each major stereotyped subset and its satellite. Finally, biomarkers with prognostic relevance will be identified through association mining between the integrated multi-type data and established clinicobiological parameters collected from all cases. Patient profiles will be scalable towards adopting other clinical or digital biomarkers. Both traditional methods as well as Deep Learning approaches will be considered for the discovery of insights within individual patients and group associations.