NFAT2 is a critical regulator of the anergic phenotype in chronic lymphocytic leukaemia

Abstract

Chronic lymphocytic leukaemia (CLL) is a clonal disorder of mature B cells. Most patients are characterised by an indolent disease course and an anergic phenotype of their leukaemia cells, which refers to a state of unresponsiveness to B cell receptor stimulation. Up to 10% of CLL patients transform from an indolent subtype to an aggressive form of B cell lymphoma over time (Richter´s syndrome) and show a significantly worse treatment outcome. Here we show that B cell-specific ablation of Nfat2 leads to the loss of the anergic phenotype culminating in a significantly compromised life expectancy and transformation to aggressive disease. We further define a gene expression signature of anergic CLL cells consisting of several NFAT2-dependent genes including Cbl-b, Grail, Egr2 and Lck. In summary, this study identifies NFAT2 as a crucial regulator of the anergic phenotype in CLL.NFAT2 is a transcription factor that has been linked with chronic lymphocytic leukaemia (CLL), but its functions in CLL manifestation are still unclear. Here the authors show, by analysing mouse CLL models and characterising biopsies from CLL patients, that NFAT2 is an important regulator for the anergic phenotype of CLL.

Fig. 1

NFAT2 is overexpressed in CLL cells. a NFAT2 mRNA expression in B cells from healthy volunteers (n = 3) and human CLL patients (n = 12), indolent CLL (n = 6) and aggressive CLL (n = 6) normalised to Actin assessed by RT-PCR (Welch’s t-test, Mean ± S.E.M., *P < 0.05, **P < 0.01). b NFAT2 protein expression in physiological B cells (n = 2), indolent CLL (n = 4) and aggressive CLL (n = 6) samples assessed by western blotting. One representative blot is shown. c Quantitative analysis of NFAT2 protein expression in physiological B cells (n = 10), indolent CLL (n = 7) and aggressive CLL (n = 8) samples (Welch’s t-test, mean ± S.E.M., *P < 0.05). d Ca²⁺ flux of splenic CD19⁺CD5⁺ cells from 28-week-old Nfat2 ^+/+ and TCL1 Nfat2 ^+/+ mice after stimulation with 10 µg/ml αIgM. Ca²⁺ was added after 3 min for extracellular flux. 1 µM ionomycin was added as a positive control. One representative experiment of three independent experiments is shown. e IgM surface expression on B cells from 28-week-old Nfat2 ^+/+ and TCL1 Nfat2 ^+/+ mice determined by flow cytometry (n = 5 per group) (Student’s t-test, Mean ± S.E.M., *P < 0.05). f Mouse breeding scheme to generate Eµ-TCL1 Nfat2 ^fl/fl Cd19-Cre mice. g NFAT2 protein expression in one representative CD19⁺ splenic B cell sample from 28-week-old TCL1 Nfat2 ^+/+ and TCL1 Nfat2 ^−/− mice assessed by western blotting

Fig. 2

Nfat2 deletion in Eµ-TCL1 mice leads to significant acceleration of disease. a Flow cytometric analysis for CD19⁺CD5⁺ CLL cells of the peripheral blood of one representative mouse for the indicated genotypes at an age of 28 weeks. b Statistical analysis of the flow cytometry data from a. n = 5 animals per group were analysed (Welch’s t-test, Mean ± S.E.M., *P < 0.05). c Absolute lymphocyte count of Nfat2 ^+/+ (n = 5), TCL1 Nfat2 ^+/+ (n = 10) and TCL1 Nfat2 ^−/− (n = 10) mice. Peripheral blood was harvested every 4 weeks and samples were analysed using an Advia 120 haematology analyser (Paired Wilcoxon test, Mean ± S.E.M.). d Expansion of CD19⁺CD5⁺ B cells in the peripheral blood of Nfat2 ^+/+ (n = 5), TCL1 Nfat2 ^+/+ (n = 10) and TCL1 Nfat2 ^−/− (n = 10) mice assessed by flow cytometry at the indicated time points (Paired Wilcoxon test, Mean ± S.E.M.). e Proliferation and apoptosis of CD19⁺CD5⁺ B cells in the peripheral blood of 28-week-old mice of the indicated genotypes (n = 5 per group). Mice were injected with 10 mM BrdU i.p. and peripheral blood cells were harvested after 48 h. CD19⁺CD5⁺ B cells were stained with BrdU and Annexin V antibodies and measured by flow cytometry (n = 5) (Welch’s t-test, Mean ± S.E.M., *P < 0.05). f Cell cycle analysis of CD19⁺CD5⁺ B cells in the peripheral blood of 28-week-old mice of the indicated genotypes (n = 5 per group). Mice were injected with 10 mM BrdU i.p. and peripheral blood was harvested 24 h after injection. CD19⁺CD5⁺ B cells were stained with BrdU antibody and 7-AAD and subsequently analysed by flow cytometry

Fig. 3

Nfat2 deletion in CLL cells leads to significantly compromised survival of Eµ-TCL1 mice. a Kaplan–Meier survival plot of TCL1 Nfat2 ^+/+ mice (n = 10), TCL1 Nfat2 ^−/− mice (n = 10) and Nfat2 ^+/+ (n = 5) and Nfat2 ^−/− (n = 5) controls. Statistical significance was determined using the Log-rank (Mantel–Cox) test, P = 0.0014. b Spleen size of representative animals of the indicated genotypes at an age of 36 weeks. c Average spleen weight of n = 5 animals with the indicated genotypes at an age of 36 weeks (Welch’s t-test, Mean ± S.E.M., ***P < 0.005, not significant (n.s.)). d Accumulation of CD19⁺CD5⁺ B cells in the spleen, lymph nodes and bone marrow of one representative Nfat2 ^+/+, TCL1 Nfat2 ^+/+ and TCL1 Nfat2 ^−/− mouse at an age of 28 weeks assessed by flow cytometry. e–g Accumulation of CD19⁺CD5⁺ B cells in the spleen, lymph nodes and bone marrow of TCL1 Nfat2 ^+/+ mice, TCL1 Nfat2 ^−/− and Nfat2 ^+/+ mice (n = 5 per group) assessed by flow cytometry at an age of 28 weeks (Welch’s t-test, mean ± S.E.M., *P < 0.05). h ZAP70 and CD38 expression on CD19⁺CD5⁺ B cells from animals with the indicated genotypes at an age of 28 weeks. n = 5 animals per group were analysed by flow cytometry (Welch’s t-test, Mean ± S.E.M., ***P < 0.005). i Expansion of CD19⁺CD5⁺ CLL cells in the peripheral blood of NSG mice transplanted with CLL cells from TCL1 Nfat2 ^+/+ (n = 5) or TCL1 Nfat2 ^−/− mice (n = 6) assessed by flow cytometry at the indicated time points (Student’s t-test, mean ± S.E.M., p = 0,0134). j Kaplan–Meier survival plot of NSG mice transplanted with CLL cells of the indicated genotype. Statistical significance was determined using the Log-rank (Mantel–Cox) test, P = 0.0014

Fig. 4

Nfat2 ablation leads to histologic transformation of CLL to aggressive disease. a H&E staining of paraffin-embedded spleen sections of one representative TCL1 Nfat2 ^+/+ and TCL1 Nfat2 ^−/− mouse at 12 and 28 weeks of age (upper panels) and immunohistochemical staining for CD3, B220 and Ki-67 (lower panels). For higher magnification of B220 and Ki-67 staining and CD79a staining in TCL1 Nfat2 ^−/− mice and Staining of Nfat2 ^+/+control mice see also Supplementary Fig. 6. b H&E staining of paraffin-embedded spleen sections of representative TCL1 Nfat2 ^+/+ and TCL1 Nfat2 ^−/− mice and immunohistochemical staining for CD3, B220 and CD79a at an age of 36 weeks

Fig. 5

NFAT2 regulates the expression of multiple anergy-associated genes in CLL. a, b Microarray analysis of FACS-sorted splenic CD19⁺CD5⁺ CLL cells from 20-week-old TCL1 Nfat2 ^+/+ and TCL1 Nfat2 ^−/− mice (n = 3 per group). Log₂-fold differences of gene expressions are displayed in a hierarchical normalised heat map analysis which reveals changes in the expression of multiple cell cycle genes a and genes involved in BCR signalling b. c Relative gene expression of Cbl-b, Grail, Egr2 and Lck mRNA normalised to Actin expression in ex vivo splenic CLL cells from 20-week-old TCL1 Nfat2 ^+/+ and TCL1 Nfat2 ^−/− mice (n = 5 per group) assessed by qRT-PCR (Student’s t-test, Mean ± S.E.M., *P < 0.05; ***P < 0.005). d Quantification of protein expression of splenic CLL cells from 20-week-old TCL1 Nfat2 ^+/+ and TCL1 Nfat2 ^−/− mice for CBL-B (n = 4), GRAIL (n = 3), EGR2 (n = 3) and LCK (n = 6) assessed by western blotting (Student’s t-test, Mean ± S.E.M., *P < 0.05). e–g Splenic CLL cells from 20-week-old TCL1 Nfat2 ^+/+ mice were isolated and stimulated in vitro with 10 µg/ml αIgM, 10 µg/ml αIgM + 1 µM CsA or 10 µg/ml αIgM + 1 µM FK506 for 6 h. Relative mRNA expression of Cbl-b (n = 4) f, Grail (n = 3) g and Egr2 (n = 5) h were assessed by qRT-PCR and normalised to Actin (One-way analysis of variance (ANOVA), Mean ± S.E.M., *P < 0.05, **P < 0.01, ***P < 0.005). h Splenic CLL cells from 20-week-old TCL1 Nfat2 ^+/+ mice (n = 5) were isolated and stimulated in vitro with 10 µg/ml αIgM, 10 µg/ml αIgM + 1 µM CsA, 10 µg/ml αIgM + 1 µM FK506 or 10 µg/ml αIgM + 10 µM VIVIT peptide for 1 h. Relative Lck mRNA expression normalised to Actin was assessed by qRT-PCR (One-way ANOVA, mean ± S.E.M., *P < 0.05, **P < 0.01)

Fig. 6

Lck is a direct target gene of NFAT2 and co localises with the BCR in CLL cells. a Schematic illustration of the Lck gene with a putative NFAT binding site. b Chromatin immunoprecipitation (ChIP) with CLL patient cells stimulated for 16 h with PMA/ionomycin. Cells were fixed with paraformaldehyde and DNA content was sheared by sonication. ChIP was performed with NFAT2 (7A6) and IgG control antibodies. The number of immunoprecipitated regions for each gene was calculated and normalised to the respective IgG control. One representative experiment of three independent experiments is shown. c LCK mRNA expression in untouched isolated physiological B cells from healthy volunteers (n = 6) and human CLL patients (n = 11) with indolent CLL (n = 6) and aggressive CLL (n = 5) normalised to GAPDH assessed by RT-PCR (Welch’s t-test, mean ± S.E.M., *P < 0.05, **P < 0.01, ***P < 0.001). d Representative LCK protein expression and its activating phosphorylation at Tyr394 in physiological B cells (n = 2), indolent CLL (n = 3) and aggressive CLL (n = 3) cells assessed by western blotting. e, f Quantitative analysis of LCK e and P-LCK (Tyr394) f expression on western blots with physiological B cells (n = 8), indolent CLL (n = 8) and aggressive CLL (n = 8) cells (Welch’s t-test, mean ± S.E.M., **P < 0.01). g Co localisation assay: Proteins were labelled with monoclonal antibodies from different species (1) and subsequently incubated with oligonucleotide-coupled secondary antibodies (2). In the case of close proximity of both proteins, the oligonucleotides on the secondary antibodies are able to ligate (3). A subsequent polymerisation and DNA amplification initiates the development of a red fluorescent signal (4), which can be detected in the fluorescence microscope. h Cytospins of PBMCs from 20-week-old TCL1 Nfat2 ^+/+ and TCL1 Nfat2 ^−/− mice were either prepared without stimulation or after treatment with 20 µg/ml αIgM F(ab’)₂ fragments for 10 min. Cells were stained with antibodies for CD79a and LCK. Co localisation of CD79a and LYN was used as a positive control. For negative controls, the primary antibody against CD79a was not added. DAPI was used for nuclear staining (630×)

Fig. 7

NFAT2 controls the anergic phenotype in CLL. a Ca²⁺ flux of splenic CD19⁺CD5⁺ CLL cells from 28-week-old TCL1 Nfat2 ^+/+ and TCL1 Nfat2 ^−/− mice after stimulation with 10 µg/ml αIgM. Ca²⁺ was added after 3 min for extracellular flux. 1 µM ionomycin was added as a positive control. One representative result of three independent experiments is shown. b IgM surface expression on B cells from 28-week-old Nfat2 ^+/+, TCL1 Nfat2 ^+/+ and TCL1 Nfat2 ^−/− mice determined by flow cytometry (n = 5 per group). A representative histogram plot (left panel) and a statistical analysis of all animals (right panel) are shown (Student’s t-test, mean ± S.E.M., *P < 0.05, ***P < 0.005). c Relative Prdm1 mRNA expression normalised to Actin in ex vivo splenic CLL cells from 28-week-old TCL1 Nfat2 ^+/+ and TCL1 Nfat2 ^−/− mice (n = 6 per group) assessed by qRT-PCR (Student’s t-test, mean ± S.E.M., **P < 0.01). d Splenic CD19⁺CD5⁺ CLL cells from 20-week-old TCL1 Nfat2 ^+/+ and TCL1 Nfat2 ^−/− mice were stimulated in vitro with 20 µg/ml αIgM for the time points indicated. Total protein levels and phosphorylation status of LYN (Tyr507), SYK (Tyr519/520), LCK (Tyr394 and Tyr505) and ERK1/2 (Thr202/Tyr204 Thr185/Tyr187) were assessed by western blotting. One representative result of three independent experiments is shown. e Paraffin-embedded lymph node biopsies from human patients with CLL (n = 5) and Richter’s syndrome (n = 9) were analysed for relative NFAT2 mRNA expression normalised to GAPDH using qRT-PCR (Mann–Whitney test, Mean ± S.E.M., **P < 0.01). f Paraffin-embedded lymph node biopsies from human patients with CLL (n = 5) and Richter’s syndrome (n = 9) were analysed for relative LCK mRNA expression normalised to GAPDH using qRT-PCR (Student’s t-test, mean ± S.E.M., *P < 0.05). g NFAT2 and LCK protein expression in one representative patient with indolent CLL (left) and one patient with Richter’s syndrome (right). h Immunohistochemical staining for LCK of a representative lymph node with indolent CLL and Richter’s Syndrome (400×)

Fig. 8

Putative mechanism of the role of NFAT2 in the regulation of the anergic phenotype in CLL. (Left panel) Tonic stimulation of the BCR in indolent CLL is associated with the recruitment of LCK and the activation of NFAT2, which subsequently translocates to the nucleus to induce the transcription of several anergy-associated genes including Cbl-b, Grail, Egr2 as well as Lck. (Right panel) In the absence of NFAT2, BCR stimulation leads to enhanced phosphotyrosine induction and calcium flux culminating in the activation of AKT and ERK and subsequent cell proliferation as observed in patients with aggressive forms of CLL or Richter’s syndrome