Genetic and epigenetic determinants of diffuse large B-cell lymphoma

Abstract

Diffuse large B-cell lymphoma (DLBCL) is the most common type of lymphoma and is notorious for its heterogeneity, aggressive nature, and the frequent development of resistance and/or relapse after treatment with standard chemotherapy. To address these problems, a strong emphasis has been placed on researching the molecular origins and mechanisms of DLBCL to develop effective treatments. One of the major insights produced by such research is that DLBCL almost always stems from genetic damage that occurs during the germinal center (GC) reaction, which is required for the production of high-affinity antibodies. Indeed, there is significant overlap between the mechanisms that govern the GC reaction and those that drive the progression of DLBCL. A second important insight is that some of the most frequent genetic mutations that occur in DLBCL are those related to chromatin and epigenetics, especially those related to proteins that "write" histone post-translational modifications (PTMs). Mutation or deletion of these epigenetic writers often renders cells unable to epigenetically "switch on" critical gene sets that are required to exit the GC reaction, differentiate, repair DNA, and other essential cellular functions. Failure to activate these genes locks cells into a genotoxic state that is conducive to oncogenesis and/or relapse.

Fig. 1. Comparison of normal and DLBCL-infiltrated lymph node histology.

(Left) Normal lymph node after hematoxylin and eosin (H&E) staining. Note the complex and varied architecture. Arrow on the left points to a germinal center within a follicle; both are in the cortex (outer region). Arrowhead indicates the medulla (inner region). Bottom arrow shows the hilum, where blood and efferent lymph vessels are connected (Image Source: https://www.pathpedia.com/education/eatlas/histology/lymph_node/images.aspx?6 (Slide 1)). (Right) H&E staining of a lymph node that has been infiltrated by DLBCL. Note the glassy, uniform surface and complete loss of normal structures (Image Source: https://www.webpathology.com/image.asp?case=822&n=3 (Slide 3)).

Fig. 2. The germinal center (GC) reaction.

The GC reaction is the foundation of humoral immunity. Its end products are memory B-cells and plasma cells that encode high-affinity antibodies. However, it also is the source of many types of B-cell lymphoma, including DLBCL. The time-lapse panels at the top of this figure depict the sequential steps of the GC reaction, which take place within lymph node follicles. The large panel at the bottom zooms in to show the mechanisms behind, and outcomes of, the selection of GC B-cells by T_FH cells in the light zone of the germinal center (Figure Source: Victora. See this reference for a detailed review).

Fig. 3. Functional interactions between proteins relevant to GC B-cell and DLBCL physiology.

BCOR BCL6 corepressor; FBXO11 F-box only protein 11; BCL2 B-cell CLL/lymphoma 2; TP53 tumor protein 53; CDKN1A cyclin-dependent kinase inhibitor 1A; CDKN1B cyclin-dependent kinase inhibitor 1B; MEF2B myocyte enhancer factor 2B; BCL6 B-cell CLL/lymphoma 6; PRDM1 PR/SET domain 1; MYC myelocytomatosis; ATM Ataxia telangiectasia mutated; ATR Ataxia telangiectasia and Rad3-related protein; CHEK1 checkpoint kinase 1; MDM2 murine double minute 2; IRF4 interferon regulatory factor 4; AICDA activation-induced cytidine deaminase (Figure prepared using: STRING Protein-Protein Interactions Network (https://string-db.org)).

Fig. 4. Functional interactions between proteins relevant to GC B-cell and DLBCL epigenetics.

SET SET nuclear proto-oncogene; EP300 E1A binding protein 300; DNMT3A DNA methyltransferase 3A; NCOR2 nuclear receptor corepressor 2; ARID1A AT-rich interaction domain 1A; CREBBP CREB-binding protein; EZH2 Enhancer of zeste 2; DICER1 Dicer 1, ribonuclease III; CBX8 Chromobox 8; CBX1 Chromobox 1; HIST1H1E histone cluster 1 H1 family member E; TET2 Tet methylcytosine deoxygenase 2; KMT2D lysine methyltransferase 2D (Figure prepared using: STRING Protein-Protein Interactions Network (https://string-db.org)).

Fig. 5. Epigenetic switches at promoters and enhancers in GC B-cells.

Starting, maintaining, and exiting from the GC reaction requires rapid and coordinated changes in the expression of specific subsets of genes in response to cell signals. This is achieved by using epigenetic switches at the promoters and enhancers of these genes. (A) H3K4me3 at a promoter signifies that it is “on” (green). The addition of H3K27me3 by EZH2 switches it to a “poised” (yellow) state of transient repression. The Y641 EZH2 mutation increases H3K27me3 deposition and turns the promoter “off” permanently (red). (B) H3K27Ac at an enhancer means that it is active. Removal of H3K27Ac by HDAC3 (complexed with BCL6-SMRT) leaves only H3K4me1 marks behind and poises the enhancer. Inactivation of CREBBP switches enhancers off by preventing their reactivation via H3K27 (and BCL6) acetylation, leaving HDAC3 unopposed. (C) H3K4me1 is also present at active enhancers. The lysine demethylases KDM1 and KDM5 are thought to remove H3K4 methylation and poise enhancers. KMT2D inactivation silences enhancers by preventing the addition of H3K4me1. (D) TET2 demethylates cytosines at enhancers, first by converting 5-methylcytosine (5mC; repressive) to 5-hydroxymethylcytosine (5hmC; active). TET2 inactivation switches enhancers off by preventing demethylation and, instead, causing hypermethylation. Genes whose promoter and/or enhancer cannot be reactivated makes them unresponsive to important cell signals. This locks cells into the GC reaction, which can lead to lymphomagenesis (Figure based on work by Mlynarczyk et al.).

Fig. 6. DNA repair and de-methylation pathways.

Cytosine (C) methylation to 5-methyl cytosine (5-mC) is mediated by DNA-methyl transferases (DNMT) through direct transfer of CH3. De-methylation is a far more complex process, involving ten-eleven translocation (TET) enzymes, base excision repair (BER), and thymine DNA glycolase (TDG). Transition from C to uracil (U) or thymine (T) can also be involved in this complex process, through using activation-induced cytidine deaminase (AID) and apolipoprotein B mRNA-editing enzyme catalytic (APOBEC), and mismatch repair (MMR) (adapted from Bhutani et al. and Dominguez and Shaknovich). 5-hmC: 5-hydroxy methyl cytosine; 5-fC: 5-formyl cytosine; 5-caC: 5-carboxyl cytosine; 5-hmU: 5-hydroxy methyl uridine.

Fig. 7. Activation-induced cytidine deaminase (AID), DNA methylation and epigenetic heterogeneity in DLBCL and other types of lymphoma.

Activation of naïve B-cells (NBC) transitioning to normal germinal center B-cells (NGBC). Changes in DNA methylation patterns during differentiation (GC reaction) involve AID (see Fig. 6), which strongly contributes to creating epigenetic heterogeneity ultimately leading to disease progression and potentially aggressivity. Gray rectangle: promoter region; broken arrow; transcription start site (TSS); open circles: CG hypo-methylation; closed circles: CG hyper-methylation; gray circles: intermediate CG methylation state: 5-hmC, 5-fC, 5-caC, brown oval: CTCF, green oval: methyl-DNA binding proteins; ++: high expression; +: medium expression; +/−: Low expression: +/− −: very low expression; X: repressed.

Fig. 8. Epigenetic heterogeneity promotes the acquisition of aggressive traits in cancer.

Classically, cancer has been thought of as a disease that is primarily genetic in nature. However, it is now known that epigenetic dysregulation in cancer can also be functionally and clinically relevant. One clear example is its contribution to tumor heterogeneity. When the epigenome is disrupted, either independently of genetic mutations (e.g., AID-related DNA hypomethylation) or as a result of them (e.g., in histone-modifying enzymes), tumor cells can start to evolve based on selection for favorable epigenetic states. This can lead to the production of tumor subclones that are genetically identical but, in reality, are expressing different combinations of genes and/or have altered the level at which certain genes are expressed. In addition to producing more aggressive characteristics, increased tumor heterogeneity decreases the likelihood that any one treatment will be able to kill every subclone, which can lead to chemoresistance and relapse (Based on work by Easwaran et al.).