B-cell tumor development in Tet2-deficient mice

Abstract

The TET2 gene encodes an α-ketoglutarate-dependent dioxygenase able to oxidize 5-methylcytosine into 5-hydroxymethylcytosine, which is a step toward active DNA demethylation. TET2 is frequently mutated in myeloid malignancies but also in B- and T-cell malignancies. TET2 somatic mutations are also identified in healthy elderly individuals with clonal hematopoiesis. Tet2-deficient mouse models showed widespread hematological differentiation abnormalities, including myeloid, T-cell, and B-cell malignancies. We show here that, similar to what is observed with constitutive Tet2-deficient mice, B-cell-specific Tet2 knockout leads to abnormalities in the B1-cell subset and a development of B-cell malignancies after long latency. Aging Tet2-deficient mice accumulate clonal CD19⁺ B220^low immunoglobulin M⁺ B-cell populations with transplantable ability showing similarities to human chronic lymphocytic leukemia, including CD5 expression and sensitivity to ibrutinib-mediated B-cell receptor (BCR) signaling inhibition. Exome sequencing of Tet2^-/- malignant B cells reveals C-to-T and G-to-A mutations that lie within single-stranded DNA-specific activation-induced deaminase (AID)/APOBEC (apolipoprotein B messenger RNA editing enzyme, catalytic polypeptide-like) cytidine deaminases targeted motif, as confirmed by the lack of a B-cell tumor in compound Tet2-Aicda-deficient mice. Finally, we show that Tet2 deficiency accelerates and exacerbates T-cell leukemia/lymphoma 1A-induced leukemogenesis. Together, our data establish that Tet2 deficiency predisposes to mature B-cell malignancies, which development might be attributed in part to AID-mediated accumulating mutations and BCR-mediated signaling.

Graphical abstract

Figure 1.

Tet2 deficiency leads to the accumulation of clonal B220^lowB-cell population in a cell-autonomous fashion. (A) Monthly monitoring of blood cell count in Tet2^CD19-Cre− and Tet2^CD19-Cre+ mice. ***P < .001. (B) Monthly monitoring of CD19⁺ B220^low abnormal cell population in the blood of Tet2^CD19-Cre− and Tet2^CD19-Cre+ mice (left panel). Significance was tested using an unpaired Student t test. Time-dependent representation of the proportion of the B220^low population within the circulating B cells (CD19⁺) in the same mice as in panel A (right panel). Significance was tested using an unpaired Student t test. (C) Spleen weight of sacrificed mice at various time points was measured for both types of mice. Significance was tested using an unpaired Student t test. (D) Immunostaining of total splenocytes (polychromatic dot plot) from a representative sick Tet2^CD19-Cre+ mouse (same genotypes as above). Splenocytes were first stained with B220 and CD19 to define the B220⁺ and B220^low population of CD19⁺ cells (left polychromatic dot plot). Other dot plots show B220 expression together with the indicated antibody on the x-axis of the overlaid B220⁺CD19⁺ population (green) and B220^lowCD19⁺ population (violet). (E) Analyses of BCR rearrangement clonality in abnormal Tet2^CD19-Cre+ B cells. V-J junctions were PCR-amplified from cDNA of sorted B cells using VH family primers (VH1, VH2, VH3, VH5, VH6, and VH7, as indicated) and conjugated consensus JH primer to assess the IgH CDR3 diversity. Top panels show Results shown are obtained from sorted Tet2^CD19Cre− B splenocytes (B220⁺CD19⁺) (top) and from age-matched Tet2^CD19-Cre+ sorted abnormal B cells (B220^lowCD19⁺) (bottom). (F) Transcription of selected genes important for B-cell differentiation differs between normal WT and Tet2-deficient tumor B cells. Histograms show quantitative reverse transcription (qRT)–PCR analysis results ΔΔCT-normalized by Abl1 expression levels. Results are an average of 3 independent duplicate experiments. The differences were validated in an extension cohort for 8 of the 11 genes. Significance was tested using the Mann-Whitney U test. *P < .05; **P < .01; ***P < .001. ns, not significant.

Figure 2.

Normal steady-state early B-cell development in Tet2-deficent mice. (A) qRT-PCR analysis transcription levels of the 3 murine Tet genes normalized by Abl1 expression. Results are the average of 3 independent experiments performed in duplicate. (B) Bone marrow B-cell development analysis. Total BM cells were analyzed for B220 and IgM expression. Immature B cells B220^lowIgM⁺ and mature B cells B220⁺IgM⁺ are indicated. Then CD43 expression was evaluated in gated B220⁺IgM⁻ cells to define pre-B (CD43⁻) and pro-B (CD43⁺) populations; finally, the more immature Hardy fractions (Fr.A CD24⁻Bp1⁻; Fr.B: CD24⁺Bp1⁻; and Fr.CC′: CD24⁺Bp1⁺) were analyzed in the pro-B-cell gate. The genotypes of the analyzed mice are indicated on the left side. (C) Average percentage of cell populations. (Left) Total B220⁺ cells (each dot represents an independent mouse). (Middle) Average percentage of the indicated population in the B-cell compartment for each genotype. (Right) Hardy fraction percentage inside the pro-B-cell compartment. Genotypes are indicated (Tet2^+/+ mice, n = 6; Tet2^−/− mice, n = 8). Im-B, immature B cells B220^lowIgM⁺; Mat-B, mature B cells B220⁺IgM⁺.

Figure 3.

Tet2 deficiency induces alterations of the proportions of peripheral B-cell populations in old mice. (A) Analysis of splenic B-cell populations in 4-month-old mice. (Left) B220 and CD19 expression analysis on total splenocytes. (Middle) CD23 and CD21 expression in the B220⁺CD19⁺ (B220⁺) gate showing the follicular (FO:CD23⁺CD21/35^medium), marginal zone (MZ: CD23^lowCD21/35^high), and newly formed (CD23⁻CD21/35⁻) B-cell populations. (Right) IgM and IgD staining gated in B220⁺CD19⁺ are also shown. Mice genotypes are indicated on the left-hand side. Histograms show the percentages of populations. (Top left) Average percentage ± standard error of the mean (SEM) of B220⁺ and B220^low populations in total splenocytes. (Bottom) Average percentage ± SEM of FO, MZ, and NF. (Top right) IgM⁺IgD⁻ and IgM⁺IgD⁺ populations in the B220⁺ B-cell compartment (Tet2^+/+ mice, n = 5; Tet2^−/− mice, n = 8). (B) Analysis of B-cell populations in the peritoneum. Peritoneal cell suspensions were stained with B220 and CD19 to define B2 (B220⁺CD19⁺) and B1 (B220^lowCD19⁺) populations. In the B1-cell gate, CD19 and CD5 expressions allow us to define B1a (CD19⁺CD5⁺) and B1b (CD19⁺CD5⁻) subsets. (Tet2^+/+, n = 4; Tet2^−/−, n = 5). The histograms show the average percentage ± SEM of the peritoneum populations. (C) Venn diagram illustrating comparison of upregulated genes in Tet2^−/− abnormal B cells and B1a cells, both with respect to splenic wild-type B cells. The deregulated pathways, as identified using gene set enrichment analyses, are shown at the bottom. SPB1a: B1a population from spleen. PCB1a: B1a population from peritoneum. (Bottom) List of signatures enriched in genes specifically upregulated in Tet2 samples. (D) Same as in panel C but for downregulated genes. (Bottom) List of signatures enriched in genes specifically downregulated in Tet2 samples. *P < .05.

Figure 4.

Tet2-deficient abnormal B cells share features with human B-cell tumors. (A) Coding sequence somatic mutations. Frequencies of the type of mutations observed in the exome analysis of abnormal B-cell populations obtained from 6 independent tumors. The type of point mutations is indicated as well as insertions and deletions (indels). Transversion mutations (green) and transitions (red) are indicated. Note the statistically significant high frequency of C > T and G > A mutations. (B) Survival curves of Aid^+/+Tet2^+/+ (control [CTR]), Aicda-deficient mice: Aid^−/− Tet2^+/+ (Aid^−/−); Tet2-deficient mice: Aid^+/+Tet2^−/− (Tet2^−/−); and double-deficient mice: Aid^−/−Tet2^−/− strains. Only mice sacrificed because of B-cell tumors are scored. (C) Venn diagram of exome mutations from the literature of human CLL^,,- and DLBCL^,- in comparison with exome mutations from the Tet2-deficient sorted abnormal B cells. The name of the mutated genes in the overlaying zones is indicated. *P < .05.

Figure 5.

Tet2-deficient abnormal B-cells are BCR-inhibition sensitive. (A) Abnormal B-cell phenotype and frequencies in the peripheral blood of mice transplanted with TCL1A transduced Tet2^+/+ or Tet2^−/− hematopoietic progenitors. Dot plots show a representative example of B220 and CD19 expression analysis in the GFP⁺ total leukocytes 4 months after transplantation. Diagrams depict the average percentages of the B220^low population in the blood of n = 3 mice per group 4 months after transplantation. Error bars represent means ± standard deviation (SD). (B) Tet2 deficiency enhances the tumor phenotype of TCL1A (GFP⁺) overexpressing B cells, as is shown by abnormal cell numbers and CD5 membrane expression level. (Upper left) B220 CD19 expression in gated GFP⁺ cells expressing TCL1A in primary and secondary recipient mice engrafted with BM cells overexpressing TCL1A in a Tet2^+/+ or Tet2^−/− background. (Lower left) Average ± SD of the median fluorescence intensity of the B220 antigen at the cell surface of CD19⁺ B cells in each group. (Upper right) Expression of CD5 and IgM on the gated B cells as in the upper-left contour plots. A representative contour plot is shown for each group. (Lower right) Average ± SD of the median fluorescence intensity of the CD5 antigen at the cell surface of CD19⁺ B cells in each group. (C) Ibrutinib-mediated BTK inhibition results in growth reduction and increase of surface IgM expression on Tet2-deficient B cells. Dose effects of ibrutinib on the B-cell population of Tet2-deficient or wild-type cell lines are derived from independent primary mice: 2 tumoral B-cell lines, one obtained from a spontaneous B-cell malignancy development in a Tet2^−/− mouse, and the other obtained from a mouse from a BMT experiment of wild-type Tet2^+/+ BM cells transduced by the TCL1A oncogene; and a T-cell line obtained from a BMT experiment of wild-type Tet2^+/+ BM cells transduced by the BRAFG469R oncogene. Diagrams depict variable cell proportion in comparison with the nontreated cells. Cells were counted 3 days after treatment with the indicated inhibitor concentration. (D) Surface expression level (mean fluorescence intensity) of IgM on B-cell population (IgM⁺B220⁺) observed after staining of the 2 B-cell lines cultivated as in panel C. (E) Splenocytes from Tet2^−/− mice were treated with ibrutinib and fludarabine for 18 hours ex vivo. Cells were stained with CD19-BV510, B220-allophycocyanin (APC)-Cy7, annexin V–APC, and 7-aminoactinomycin D (7-AAD). Gated B220^lowCD19⁺ splenic cells were analyzed for annexin V and 7-AAD staining. Histograms show the mean ± SD of the percentages of viable cells (annexin V and 7-AAD double negative) in gated B220^lowCD19⁺ populations from spleens of Tet2^−/− mice (n = 2). *P < .05 (unpaired Student t test). MFI, mean fluorescence intensity.