Calado lab

Immunity & Cancer Laboratory

: Interplay between immunity and cancer development

A spleen tissue cross-section.

Standard content

A major interest of the Immunity and Cancer Laboratory is to understand how healthy cells become cancerous. To address this question we study B cells, as they provide an ideal model system: first, their normal development is well defined; second, the most common types of human haematological malignancies are derived from the transformation of healthy B cells into cancer cells.

B cells, essential cellular components of the immune system develop from hematopoietic stem cells in the bone marrow through a highly regulated process that results in the production of functional mature B cells. During the course of a T cell dependent immune response to infection, mature B cells develop further in peripheral tissues, such as spleen and lymph nodes, into memory B cells or terminally differentiate into terminally differentiated plasma cells in a reaction called germinal centre (GC). Memory B cells allow the rapid immune response re-call upon re-infection. Plasma cells produce high affinity antibodies essential for the elimination of the infectious agent.

However, during the development of both memory B cells and plasma cells, healthy B cells undergo in the GC microenvironment processes of DNA mutation and recombination while rapidly dividing, and infidelity in these processes may lead to oncogenic DNA lesions. Although cancer cells may arise from healthy B cells at several stages of development, cancers derived from GC B cells or from cells that have passed the GC reaction are the most common amongst haematological malignancies in adults overall. These cancers include the vast majority of human Hodgkin and non-Hodgkin lymphomas (e.g. the aggressive diffuse large B cell and Burkitt lymphomas), and multiple myeloma (a plasma cell cancer), which is currently incurable.

Using an integrative approach to immunology and cancer biology our laboratory aims to understand the mechanisms by which healthy GC B cells become cancerous. For that we primarily perform studies on three complementary research themes:

(i) Identify and study sub-populations within healthy GC or post GC B cells, which due to their physiology are at a particularly high risk of becoming cancer cells, that is the so-called 'cells of origin in cancer'.

An example of such work is the identification of c-Myc+ B cell subpopulations in immature and mature GCs, playing indispensable roles in the formation and maintenance of GCs (Calado et al., Nat Immunol. 2012 [Abstract]). The identification of these functionally critical cellular subsets has important implications for human B cell lymphomagenesis, given that it frequently involves MYC chromosomal translocations. As these translocations are generally dependent on transcription of the recombining partner loci, the c-Myc+ GC subpopulations may be at a particularly high risk for malignant transformation, see Figure 1.

c-Myc+ B cell subpopulations in immature and mature germinal centers.

Figure 1: c-Myc+ B cell subpopulations in immature and mature germinal centers. (a) Schematic representation of immature and mature germinal centers (GC)s and of c-Myc expression in B cells at these stages of GC development. Immature GCS, day 4 after immunization; mature GCS day 10 after immunisation. Genetic ablation of Myc, demonstrates that c-l\/lyc+ B cell subpopulation share indispensable roles in the formation and maintenance of GCS. DZ→LZ, migration of GC B cells between dark zone (DZ) and light zone (LZ); LZ→DZ migration of GC B cells between LZ and DZ._(b) Histological immunofluorescence microscopy of T cells (CD3; green), FDCs (CD35; blue), Bcl-6 (white) and c-Myc (red) in spleens from wild-type C57BL/6 mice 4 d (immature GCS) after immunisation with SRBCS, with a cutoff of ‹20 (left panel) and ›20 (right panel) Bcl-6+ cells per FDC area. Scale bars, 25μm. (c)Upper panel, histological immunofluorescence microscopy of a spleen from a wild-type C57BL/6 mouse obtained 10 d (mature GCs) after immunisation with SRBCS and stained to detect FDCs (CD35;blue), Bcl-6 (white) and c-Myc (red). Bottom panel, enlargement of the image in c showing the boundary between the DZ and LZ (white dashed line) and c-Myc+ cells (white arrows). Scale bars 25μm. (d) Fraction of Bcl-6+c-Myc+ cells among Bcl-6+ cells per FDC area, for GCS from wild-type C57BL/6 mice 4d or 10d after immunisation with SRBCs. Each red symbol represents analysis of one FDC area; red line represents nonlinear regression curve fit with the assumption of a one-phase decay (dashed lines indicate the 95% confidence interval). Adapted from Calado et al., Nat Immunol. 2012 [Abstract].

(ii) Investigate the causative mutations, responsible for the progressive process of how a healthy B cell develops into a cancer cell.

High-throughput sequencing of human cancer genomes, including those of haematological malignancies, has produced large datasets. However, in human cancers, the causative/driver genetic mutations are sometimes masked by their inclusion on large amplifications or deletions or by others so-called passenger/bystander that do not contribute to the phenotype or progression of the cancer. Strategies aiming to discriminate these mutations are required to prioritize further functional validation. If truly important causative genes and their associated mechanisms are evolutionary conserved then one possible solution may be to model human cancers in the mouse. Mouse versus human inter-species oncogenomic comparisons may serve as a very powerful tool to identify causal mutations and to interrogate their role in a defined genetic context.

Exemplifying well this concept are studies done using cutting-edge genetic techniques in vivo in the mouse where gain-of-function and/or loss-of-function mutagenesis is done specifically in the GC B cells. A first study had as objective to determine if two recurrent genetic events in human diffuse large B cell Lymphoma (DLBCL), namely constitutive NF-κB activity and abrogation of terminal B cell differentiation through Blimp1 disruption were causal for the disease (Calado et al., Cancer Cell. 2010;18(6):580-9). Not only did the combination of these events led to lymphoma development in the mouse but these lymphomas recapitulated many of the features of the most aggressive and with poorer clinical prognosis human DLBCL subtype, the so called Activated B Cell DLBCL (ABC-DLBCL). Thus this work suggests that both NF-κB signalling, as an oncogenic event, and BLIMP1, as a tumour suppressor, play causal roles in the pathogenesis of human ABC-DLBCL, see Figure 2.

Constitutive canonical NF-KB activation cooperates with disruption of Blimp1 in the pathogenesis of activated B cell-like diffuse large cell lymphoma.

Figure 2: Constitutive canonical NF-KB activation cooperates with disruption of Blimp1 in the pathogenesis of activated B cell-like diffuse large cell lymphoma. (a) Survival of mice carrying the Cy1-cre transgene (expressing cre recombinase specifically in GC B cells) together with a conditional B/imp1 allele (B/imp1F) and/or a conditional R26 allele expressing a constitutively active IKK2 kinase IKK2castopFL, leading to NFκB activation). Mice carrying the Cy1-cre transgene together with a conditional R26 allele expressing eYFP (eYFPstopFL) served as controls. Kaplan-Meier survival graph of mice of the indicated genotypes. Med. surv, median survival; undefined. (b) Representative pictures of spleens of compound mutant and control mice. (c) Left panels, representative H&E staining of spleens of compound mutant and control mice. Scale bar, 1000μm; inset, 100μm. Right panels, representative immunohistochemical staining for MUM1/IRF4 (characteristic of the ABC-DLBCL subtype but not of the GC-DLBCL subtype) of spleens of compound mutant and control mice. Scale bar, 1000 μm; inset, 100 μm. (d) Heat map showing the relative transcript levels of candidate genes in lymphoma samples compared to normal GC B cells. For heat map generation the Hprt-normalized value of each candidate gene was normalized to have mean zero, variance one. Adapted from Calado et al., Cancer Cell 2010.

In another study it was found that in the development of Burkitt lymphoma (BL), which carries MYC chromosomal translocations in the vast majority of cases, requires activation of the phosphoinositide-3-kinase (PI3K) pathway to counterbalance the pro-apoptotic properties of c-Myc. PI3K synergised with c-Myc for the development of BL-like tumours in the mouse which fully phenocopied human BL with regard to histology, cellular markers, and gene expression profile (Sander et al., Cancer Cell. 2012;22(2):167-79). The identification of the tertiary mutational events in these lymphomas, led to the observation that some of these were also recurrently seen in human BL, indeed demonstrating that mouse versus human oncogenomic comparisons are a powerful tool to identify causal mutations in cancer.

(iii) Assess the processes of evasion of cancer cells from immune control by determining the nature of the interplay between a developing B cell cancer and cells of the immune system.

Host immunity plays a fundamental role in tumour formation and progression and cancer immune escape is an emerging 'hallmark of cancer'. For that reason the study of cancer cannot be confined to the intrinsic cellular mechanisms of oncogenic transformation, and has to take into account the interaction of pre-cancer and cancer cells with their microenvironment. However, limitations and inherent weaknesses of existing mouse models of cancer are clear. In a bona-fide cancer model, the introduction of oncogenic mutations must be tissue specific, temporally regulated and restricted to a small fraction of cells, requirements that are not fulfilled by most of the current cancer models. We are currently developing novel genetic approaches in the mouse, which allows us to introduce oncogenic mutations in a small fraction of GC B cells, mimicking the sporadic nature of cancer, and in a temporally regulated manner, such that it allows the follow up of mutant cells as cancer develops and progresses.