Preneoplastic Alterations Define CLL DNA Methylome and Persist through Disease Progression and Therapy

Mouse models of CLL: In vivo modeling of disease initiation, progression, and transformation to Richter transformation

Chronic lymphocytic leukemia (CLL) is a highly complex disease characterized by the proliferation of CD5+ B cells in lymphoid tissues. Current modern treatments have brought significant clinical benefits to CLL patients. However, there are still unmet needs. Patients relapse on Bruton's tyrosine kinase inhibitors and BCL2 inhibitors and often develop more aggressive diseases including Richter transformation (RT), an incurable complication of up to ∼10% patients. This evidence underscores the need for improved immunotherapies, combination treatment strategies, and predictive biomarkers. A mouse model that can recapitulate human CLL disease and certain components of the tumor immune microenvironment represents a promising preclinical tool for such purposes. In this review, we provide an overview of CRISPR-engineered and xenograft mouse models utilizing either cell lines, or primary CLL cells suitable for studies of key events driving the disease onset, progression and transformation of CLL. We also review how CRISPR/Cas9 established mouse models carrying loss-of-function lesions allow one to study key mutations driving disease progression. Finally, we discuss how next generation humanized mice might improve to generation of faithful xenograft mouse models of human CLL.

The molecular map of CLL and Richter's syndrome

Clonal expansion of B-cells, from the early stages of monoclonal B-cell lymphocytosis through to chronic lymphocytic leukemia (CLL), and then in some cases to Richter's syndrome (RS) provides a comprehensive model of cancer evolution, notable for the marked morphological transformation and distinct clinical phenotypes. High-throughput sequencing of large cohorts of patients and single-cell studies have generated a molecular map of CLL and more recently, of RS, yielding fundamental insights into these diseases and of clonal evolution. A selection of CLL driver genes have been functionally interrogated to yield novel insights into the biology of CLL. Such findings have the potential to impact patient care through risk stratification, treatment selection and drug discovery. However, this molecular map remains incomplete, with extant questions concerning the origin of the B-cell clone, the role of the TME, inter- and intra-compartmental heterogeneity and of therapeutic resistance mechanisms. Through the application of multi-modal single-cell technologies across tissues, disease states and clinical contexts, these questions can now be addressed with the answers holding great promise of generating translatable knowledge to improve patient care.

GraphCpG: imputation of single-cell methylomes based on locus-aware neighboring subgraphs

MOTIVATION

Single-cell DNA methylation sequencing can assay DNA methylation at single-cell resolution. However, incomplete coverage compromises related downstream analyses, outlining the importance of imputation techniques. With a rising number of cell samples in recent large datasets, scalable and efficient imputation models are critical to addressing the sparsity for genome-wide analyses.

RESULTS

We proposed a novel graph-based deep learning approach to impute methylation matrices based on locus-aware neighboring subgraphs with locus-aware encoding orienting on one cell type. Merely using the CpGs methylation matrix, the obtained GraphCpG outperforms previous methods on datasets containing more than hundreds of cells and achieves competitive performance on smaller datasets, with subgraphs of predicted sites visualized by retrievable bipartite graphs. Besides better imputation performance with increasing cell number, it significantly reduces computation time and demonstrates improvement in downstream analysis.

AVAILABILITY AND IMPLEMENTATION

The source code is freely available at https://github.com/yuzhong-deng/graphcpg.git.

Monoclonal B-cell lymphocytosis, monoclonal gammopathy of undetermined significance, and T-cell clones of uncertain significance: are these premalignant conditions sharing a common identity?

Monoclonal B-cell lymphocytosis, monoclonal gammopathy of undetermined significance, and T-cell clones of uncertain significance are three premalignant conditions characterised by the presence of small clonal cell expansions in individuals without symptoms or signs that distinguish the related overt malignancies (chronic lymphocytic leukaemia, multiple myeloma, and T-cell large granular lymphocytic leukaemia). As most individuals with these precursor states never progress to malignancies, considerable interest has arisen in comprehending the steps involved in the progression to malignancy, providing more accurate models to investigate potential mechanisms of early blood cancer identification, prevention, and, possibly, intervention. Single-cell technologies and recent progress in high-throughput sequencing and multiomics approaches have contributed to a better definition of the pathophysiological mechanisms of these premalignant conditions, moving our knowledge in the field forward. In this Viewpoint, we analyse the seemingly shared biological trajectories in these precursor haematological malignancies in search of common pathogenetic events. In particular, we address the issue of interactions between expanding clones and their immune ecosystem, offering new clues that might prompt innovative ideas and inspire further investigations to understand the cellular and molecular dynamics entailing progression into overt malignant disease. The relationships between the non-leukaemic microenvironmental cells and the leukaemic counterpart, and the primary drivers of their initial clonal expansion, represent shared biologies that suggest a common identity among the premalignant conditions considered in this Viewpoint.

A critical appraisal of the relative contribution of tissue architecture, genetics, epigenetics and cell metabolism to carcinogenesis

Here we contrast several carcinogenesis models. The somatic-mutation-theory posits mutations as main causes of malignancy. However, inconsistencies led to alternative explanations. For example, the tissue-organization-field-theory considers disrupted tissue-architecture as main cause. Both models can be reconciled using systems-biology-approaches, according to which tumors hover in states of self-organized criticality between order and chaos, are emergent results of multiple deviations and are subject to general laws of nature: inevitable variation(mutation) explainable by increased entropy(second-law-of-thermodynamics) or indeterminate decoherence upon measurement of superposed quantum systems(quantum mechanics), followed by Darwinian-selection. Genomic expression is regulated by epigenetics. Both systems cooperate. So cancer is neither just a mutational nor an epigenetic problem. Rather, epigenetics links environmental cues to endogenous genetics engendering a regulatory machinery that encompasses specific cancer-metabolic-networks. Interestingly, mutations occur at all levels of this machinery (oncogenes/tumor-suppressors, epigenetic-modifiers, structure-genes, metabolic-genes). Therefore, in most cases, DNA mutations may be the initial and crucial cancer-promoting triggers.

Dynamic antagonism between key repressive pathways maintains the placental epigenome

DNA and Histone 3 Lysine 27 methylation typically function as repressive modifications and operate within distinct genomic compartments. In mammals, the majority of the genome is kept in a DNA methylated state, whereas the Polycomb repressive complexes regulate the unmethylated CpG-rich promoters of developmental genes. In contrast to this general framework, the extra-embryonic lineages display non-canonical, globally intermediate DNA methylation levels, including disruption of local Polycomb domains. Here, to better understand this unusual landscape's molecular properties, we genetically and chemically perturbed major epigenetic pathways in mouse trophoblast stem cells. We find that the extra-embryonic epigenome reflects ongoing and dynamic de novo methyltransferase recruitment, which is continuously antagonized by Polycomb to maintain intermediate, locally disordered methylation. Despite its disorganized molecular appearance, our data point to a highly controlled equilibrium between counteracting repressors within extra-embryonic cells, one that can seemingly persist indefinitely without bistable features typically seen for embryonic forms of epigenetic regulation.

Genetics and epigenetics of CLL

Chronic lymphocytic leukemia (CLL) has a heterogeneous biological behavior, which is highly influenced by its immunogenetic, epigenetic, and genomic properties. The remarkably variable clinical course of the disease has been associated with genetic features such as chromosomal abnormalities, the presence of either high or low numbers of somatic hypermutations (SHM) in the variable region of the immunoglobulin heavy chain locus (IGHV), and somatic mutations of several specific driver genes. Next-generation sequencing (NGS) technologies have provided a comprehensive characterization of the genomic and epigenomic landscape in CLL, elucidating important underlying mechanisms of the disease's biology. The scope of this review is to summarize the most recent discoveries about novel genetic and epigenetic alterations, discussing their impact on clinical outcomes and response to currently available therapy.

The end of the beginning: application of single-cell sequencing to chronic lymphocytic leukemia

Single-cell analysis has emerged over the past decade as a transformative technology informative for the systematic analysis of complex cell populations such as in cancers and the tumor immune microenvironment. The methodologic and analytical advancements in this realm have evolved rapidly, scaling from but a few cells at its outset to the current capabilities of processing and analyzing hundreds of thousands of individual cells at a time. The types of profiling attainable at individual cell resolution now range from genetic and transcriptomic characterization and extend to epigenomic and spatial analysis. Additionally, the increasing ability to achieve multiomic integration of these data layers now yields ever richer insights into diverse molecular disease subtypes and the patterns of cellular circuitry on a per-cancer basis. Over the years, chronic lymphocytic leukemia (CLL) consistently has been at the forefront of genomic investigation, given the ready accessibility of pure leukemia cells and immune cells from circulating blood of patients with this disease. Herein, we review the recent forays into the application of single-cell analysis to CLL, which are already revealing a new understanding of the natural progression of CLL, the impact of novel therapies, and the interactions with coevolving nonmalignant immune cell populations. As we emerge from the end of the beginning of this technologic revolution, CLL stands poised to reap the benefits of single-cell analysis from the standpoints of uncovering fresh fundamental biological knowledge and of providing a path to devising regimens of personalized diagnosis, treatment, and monitoring.

Memory B-cell like chronic lymphocytic leukaemia is associated with specific methylation profile of WNT5A promoter and undetectable expression of WNT5A gene

Genome methylation profiles define naïve-like (n-CLL), memory-like (m-CLL), and intermediate (i-CLL) subsets of chronic lymphocytic leukaemia (CLL). The profiles can be easily determined by the analysis of the five-CpG signature. m-CLL, i-CLL, and n-CLL with the good, intermediate, and poor prognoses, respectively, differ by the somatic hypermutation status of the immunoglobulin heavy chain variable gene (IGHV), a widely used prognostic predictor in CLL. We have previously shown that the expression of WNT5A, encoding a ROR1 ligand, distinguishes patients with the worse outcome within the prognostically favourable IGHV-mutated subgroup. To analyse the mechanisms controlling WNT5A expression, we investigated the methylation status of 54 CpG sites within the WNT5A promoter and its relation to the WNT5A gene expression. In a cohort of 59 CLL patients balanced for combinations of IGHV and WNT5A statuses, we identified three promoter CpG sites whose methylation level correlated with the WNT5A expression within the IGHV-mutated subgroup. Further, we complemented our data with the methylation status of the five-CpG signature. IGHV-mutated/WNT5A-negative and IGHV-mutated/WNT5A-positive cases overlapped with m‑CLL and i‑CLL methylation subgroups, respectively, while most IGHV‑unmutated samples were assigned to n-CLL. Median methylation levels of all the three CpG sites in the WNT5A promoter were lowest in i-CLL. Finally, a detailed analysis of m-CLL and i-CLL showed that undetectable WNT5A expression predicts longer treatment-free survival with higher statistical significance than the classification according to the five-CpG signature. To conclude, a favourable m-CLL subgroup is associated with mutated IGHV and undetectable WNT5A expression due to its promoter methylation.

Clonal Evolution of High-Risk Chronic Lymphocytic Leukemia: A Contemporary Perspective

Clonal evolution represents the natural process through which cancer cells continuously search for phenotypic advantages that enable them to develop and expand within microenvironmental constraints. In chronic lymphocytic leukemia (CLL), clonal evolution underpins leukemic progression and therapeutic resistance, with differences in clonal evolutionary dynamics accounting for its characteristically diverse clinical course. The past few years have witnessed profound changes in our understanding of CLL clonal evolution, facilitated by a maturing definition of high-risk CLL and an increasing sophistication of next-generation sequencing technology. In this review, we offer a modern perspective on clonal evolution of high-risk CLL, highlighting recent discoveries, paradigm shifts and unresolved questions. We appraise recent advances in our understanding of the molecular basis of CLL clonal evolution, focusing on the genetic and non-genetic sources of intratumoral heterogeneity, as well as tumor-immune dynamics. We review the technological innovations, particularly in single-cell technology, which have fostered these advances and represent essential tools for future discoveries. In addition, we discuss clonal evolution within several contexts of particular relevance to contemporary clinical practice, including the settings of therapeutic resistance to CLL targeted therapy and immunotherapy, as well as Richter transformation of CLL to high-grade lymphoma.

Activation of Notch and Myc Signaling via B-cell-Restricted Depletion of Dnmt3a Generates a Consistent Murine Model of Chronic Lymphocytic Leukemia

Chronic lymphocytic leukemia (CLL) is characterized by disordered DNA methylation, suggesting these epigenetic changes might play a critical role in disease onset and progression. The methyltransferase DNMT3A is a key regulator of DNA methylation. Although DNMT3A somatic mutations in CLL are rare, we found that low DNMT3A expression is associated with more aggressive disease. A conditional knockout mouse model showed that homozygous depletion of Dnmt3a from B cells results in the development of CLL with 100% penetrance at a median age of onset of 5.3 months, and heterozygous Dnmt3a depletion yields a disease penetrance of 89% with a median onset at 18.5 months, confirming its role as a haploinsufficient tumor suppressor. B1a cells were confirmed as the cell of origin of disease in this model, and Dnmt3a depletion resulted in focal hypomethylation and activation of Notch and Myc signaling. Amplification of chromosome 15 containing the Myc gene was detected in all CLL mice tested, and infiltration of high-Myc-expressing CLL cells in the spleen was observed. Notably, hyperactivation of Notch and Myc signaling was exclusively observed in the Dnmt3a CLL mice, but not in three other CLL mouse models tested (Sf3b1-Atm, Ikzf3, and MDR), and Dnmt3a-depleted CLL were sensitive to pharmacologic inhibition of Notch signaling in vitro and in vivo. Consistent with these findings, human CLL samples with lower DNMT3A expression were more sensitive to Notch inhibition than those with higher DNMT3A expression. Altogether, these results suggest that Dnmt3a depletion induces CLL that is highly dependent on activation of Notch and Myc signaling. SIGNIFICANCE: Loss of DNMT3A expression is a driving event in CLL and is associated with aggressive disease, activation of Notch and Myc signaling, and enhanced sensitivity to Notch inhibition.

CpG Transformer for imputation of single-cell methylomes

MOTIVATION

The adoption of current single-cell DNA methylation sequencing protocols is hindered by incomplete coverage, outlining the need for effective imputation techniques. The task of imputing single-cell (methylation) data requires models to build an understanding of underlying biological processes.

RESULTS

We adapt the transformer neural network architecture to operate on methylation matrices through combining axial attention with sliding window self-attention. The obtained CpG Transformer displays state-of-the-art performances on a wide range of scBS-seq and scRRBS-seq datasets. Furthermore, we demonstrate the interpretability of CpG Transformer and illustrate its rapid transfer learning properties, allowing practitioners to train models on new datasets with a limited computational and time budget.

AVAILABILITY AND IMPLEMENTATION

CpG Transformer is freely available at https://github.com/gdewael/cpg-transformer.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Epigenetic Trajectories of the Premalignant-to-Malignant Transition of Chronic Lymphocytic Leukemia

Kretzmer and colleagues show that the transition to altered methylome occurs very early in chronic lymphocytic leukemia, and once acquired, it is a clonal and extremely stable change. However, the precise time point when the leukemic clone starts deviating significantly from the normal B-cell differentiation trajectory is still elusive. See related article by Kretzmer et al., p. 54.

Potential Relevance of B-cell Maturation Pathways in Defining the Cell(s) of Origin for Chronic Lymphocytic Leukemia

Chronic lymphocytic leukemia (CLL) is a common, incurable disease of undefined cause. Notably, the normal cell equivalents of CLL cells remain elusive, and it is possible that the disease emanates from several normal B-cell subsets. This article reviews the literature relating to this issue, focusing on recent findings, in particular made through epigenetic analyses that strongly support the disease developing from a normal Ag-experienced and memory cell-like B lymphocyte. It also reports the known pathways whereby normal B lymphocytes mature after antigenic challenge and proposes that this information is relevant in defining the cells of origin of this disease.