TBL1XR1 Mutations Drive Extranodal Lymphoma by Inducing a Pro-tumorigenic Memory Fate

Abstract

The most aggressive B cell lymphomas frequently manifest extranodal distribution and carry somatic mutations in the poorly characterized gene TBL1XR1. Here, we show that TBL1XR1 mutations skew the humoral immune response toward generating abnormal immature memory B cells (MB), while impairing plasma cell differentiation. At the molecular level, TBL1XR1 mutants co-opt SMRT/HDAC3 repressor complexes toward binding the MB cell transcription factor (TF) BACH2 at the expense of the germinal center (GC) TF BCL6, leading to pre-memory transcriptional reprogramming and cell-fate bias. Upon antigen recall, TBL1XR1 mutant MB cells fail to differentiate into plasma cells and instead preferentially reenter new GC reactions, providing evidence for a cyclic reentry lymphomagenesis mechanism. Ultimately, TBL1XR1 alterations lead to a striking extranodal immunoblastic lymphoma phenotype that mimics the human disease. Both human and murine lymphomas feature expanded MB-like cell populations, consistent with a MB-cell origin and delineating an unforeseen pathway for malignant transformation of the immune system.

Keywords: ABC-DLBCL; BACH2; BCL6; TBL1XR1; cell fate; cell of origin; extranodal lymphoma; germinal center; memory B cells.

Figure 1.. Tbl1xr1 mutation impairs GC development.

A, TBL1XR1 mutations in DLBCL (Arthur et al., 2018; Ma et al., 2019; Reddy et al., 2017) and FL (Krysiak et al., 2017; Ma et al., 2019; Ortega-Molina et al., 2015). TBL1XR1-SMRT interacting region (Zhang et al., 2002), and PPI mutated positions are indicated. See also Data S1A–C and Table S1. B, Exposed (>25% accessible surface) or buried residues affected by missense DLBCL mutations in TBL1XR1 WD40 domain (4LG9; DOI: 10.2210/pdb4lg9/pdb). C-D, FC analysis of splenic (C) total B-cells or (D) GCB. See also Data S1D–F. E, Spleen sections H&E from animals treated as in C. Insets show zoom of outlined areas. Scale = 500μm (top), 100μm (bottom). F, B220 or PNA IHC in consecutive spleen sections from E. Scale = 100μm. G-H, (G) Number of GC per spleen section or (H) GC size as (left) number of cells or (right) area, based on PNA staining. Dots represent individual (G) animals or (H) GCs. Results for 5 animals per genotype. I, FC analysis of D370Y/WT and WT/WT relative contribution to total B-cells and GCB, based on CD45 allelic frequencies. See also Data S1H–J. J, Use of the Rosa26YFPstop reporter. K, FC analysis of splenic GCB. Left to right: n = 4, 5, 5, 4 per genotype. See also Data S1K–L. L, FC analysis of splenic GCB. Values represent mean ± SEM. Data reproducible with three repeats. NS, not significant; *P < 0.05; **P < 0.01; ***P < 0.001, using unpaired (C,D,K) or paired (I) two-tailed Student’s t-test; or Mann-Whitney U-test (G,H); or one-way ANOVA with Tukey’s post-test (L).

Figure 2.. Tbl1xr1 mutation impairs GC proliferation.

A, FC analysis of AnnexinV/DAPI staining of splenic GCB. See also Data S1P. B, IHC staining in consecutive spleen sections. Contour delineates GC area based on PNA stain. Insets show zoom of outlined areas. Arrows point representative TUNEL⁺ GCB. Graph shows frequency of TUNEL⁺ GCB per GC, where each dot is a GC. Results for 5 animals per genotype. Scale = 50μm (left and center), 10μm (right). C, FC analysis of splenic PCNA⁺ GCB. NB from a WT mouse illustrate non-proliferating cells. D, FC analysis of EdU incorporation by splenic GCB. NB from a WT mouse illustrate non-proliferating cells. E, FC analysis of splenic GCB cell cycle in D, based on EdU/DAPI staining. Data for n=5 per genotype. F, Use of the R26-Fucci2aR reporter. G-H, FC analysis of cell cycle distribution of (G) CC or (H) CB. See also Data S1Q. Values represent mean ± SEM. Data reproducible with two repeats. P-values calculated using unpaired (A,C,D,G,H) or paired (D,E) two-tailed Student’s t-test; or Mann-Whitney U-test (B).

Figure 3.. Tbl1xr1 mutation induces preMB expansion.

A, Differentially expressed genes in splenic D370Y/WT GCB. Genes of interest are highlighted. See also Figures S3A–C and Table S2. B, Pathway enrichment for genes in A. See also Table S2. C-D, FC analysis of (C) CCR6⁺ or (D) IL-9R⁺ splenic GCB. E, GSEA of D370Y/WT GCB against preMB (EFNB1⁺S1PR2^lo) GCB (GSE89897). See also Figures S3G–H. F-G, FC analysis of EFNB1⁺ and preMB populations in splenic GCB. See also Figures S3I–K and Data S1R. H, Experimental scheme and timeline for I-K. I, FC analysis of preMB in splenic GCB. J-K, FC analysis of splenic (J) total B-cells or (K) GCB. Values represent mean ± SEM. Data reproducible with two repeats. P-values calculated using unpaired two-tailed Student’s t-test (C,D,F,G), or one-way ANOVA with Tukey’s post-test (I-K).

Figure 4.. Tbl1xr1 mutations bias cell fate towards MB.

A-B, GSEA of antigen-specific (A) MB or (B) PC, relative to GCB (GSE4142), against D370Y/WT GCB. C, Experimental scheme and timeline for D-G. D-E, FC analysis of (D) total YFP⁺ splenocytes in IgG(Ctrl Ab) treated mice, and (E) the relative cell type composition of this population. Results for 4 animals. See also Figure S4C and Data S1U. F-G, FC analysis of (F) total YFP⁺ splenocytes in anti-CD40L treated mice, and (G) the relative cell type composition of this population. Results for 4 animals. H, FC analysis of cell type composition of antigen-specific splenocytes. Results for 5 animals. See also Data S1V. I, FC analysis of (top) total or (bottom) GC-derived BM LLPC. Values represent mean ± SEM. Data reproducible with two repeats. P-values calculated using unpaired (H,I) or paired (D-G) two-tailed Student’s t-test.

Figure 5.. TBL1XR1 mutations introduce a BCL6-to-BACH2 switch.

A, TBL1XR1^WT interacting proteins. Shown are hits with SAINT>0.75 and 2-fold spectral counts (averaged between biological duplicates) over eGFP-BirA. See also Figure S5C and Table S3. B-C, Comparison between TBL1XR1 WT and Y446S BioID. Results as average of biological duplicates for each condition. D, Co-IP results for V5 pulldown in cells inducibly expressing WT or mutant TBL1XR1-V5 fusion proteins. See also Figures S5D–E. E, GSEA of D370Y/WT GCB against siBCL6-treated OCI-Ly1 (GSE29282). F, FC analysis of splenic GCB. See also Data S1W. G, IHC staining of spleen sections from F. Graph shows GC area based on PNA staining. Dots are individual GCs. Results for 4 animals per genotype. Scale = 100μm. H-I, GSEA of IμBcl6-Tbl1xr1^D370Y/WT GCB against (H) BACH2^hi GCB (GSE77319) or (I) siBCL6-treated OCI-Ly1. J-K, GSEA of KO/KO GCB against (J) siBCL6-treated OCI-Ly1 or (K) BACH2^hi GCB. L, FC analysis of splenic GCB. Values represent mean ± SEM. Data reproducible with two repeats. P-values calculated using one-way ANOVA with Tukey’s post-test (F,L), or Kruskal–Wallis H test with Dunn’s post-test (G).

Figure 6.. Tbl1xr1 mutant MB cells preferentially become GCB upon recall.

A-B, FC analysis of total or antigen-specific MB population in (A) spleen or (B) BM. See also Data S1BB–CC. C-D, FC analysis of IgG1⁺ or IgM⁺ splenic (C) GCB or (D) MB. See also Figure S6B and Data S1DD. E, FC profiling of donor-derived antigen-specific splenic MB. See also Figure S6L. F, Experimental scheme and timeline for G-J. G-I, FC analysis of (G) total B220+, (H) GCB or (I) PC donor-derived antigen-specific cells, in the spleens of MB-recipient animals. See also Figure S6M. J, ELISA for NP-specific serum Ig in MB-recipients in F. Graphs show difference between 0 and 3.5d post-immunization. Values represent mean ± SEM. Data reproducible with two repeats. P-values calculated using unpaired two-tailed Student’s t-test.

Figure 7.. Tbl1xr1 alterations lead to extranodal immunoblastic-like lymphomas.

A, Experimental scheme and timeline for B-K and N-O. B, Incidence and representative images of macroscopic tumors found in VavP-Bcl2-CD19Cre-Tbl1xr1^KO/KO mice. Left to right: tumors associated to intestine, lungs and para-aortic tissues. Numbers represent different animals. C, Spleens at time of necropsy; scale = 500mm. Graph shows spleen to body weight ratios. Dashed line represents average for non-VavP-Bcl2 age-matched controls (n=10). D-H, FC analysis of splenic (D) B/T-cell populations, (E) GCB, (F) preMB, or (G) MB (B220⁺CD138⁻IgD⁻FAS^−/loGL7⁻CD38⁺), or (H) CD138⁺ cells. See also Data S1HH–II. I, H&E of spleen sections from animals in A. Scale = 500μm (left) or 15μm (center and right). See also Figure S7M. J-K, H&E of (J) liver or (K) kidney sections from animals in A. Scale = 500μm (left), 100μm (center), 15μm (right). L, Experimental scheme and timeline for M. M, Survival curve for VavP-Bcl2;CD19Cre-Tbl1xr1^KO/KO and VavP-Bcl2;CD19Cre-Tbl1xr1^WT/WT mice. N, SHM burden at Ig intron J_H4 in extranodal tumors from A. Numbers in square brackets are total numbers of clones, pooled for 3 animals. Aicda^KO/KO GCB were used as control. O, Mutation burden at Pim1 locus in tumors from A. Numbers in square brackets represent total clones, pooled for n=5 per genotype. P, Representative images and quantification of CD38 IHC for MCD/C5 and non-MCD/C5 ABC-DLBCL human specimens. Scale = 20μm. Q, (Left) Surface marker contribution to PCA loading and (right) PCA plots for MB and GC-B from reactive LN, used as reference in CyTOF experiments. R-S, PCA for CyTOF profiling of human DLBCL specimens carrying canonical MCD/C5 lesions and (R) TBL1XR1 mutations or (S) TBL1XR1 WT alleles. See also Table S5. Values represent mean ± SEM. Data reproducible with two repeats. P-values calculated using unpaired two-tailed Student’s t-test (D-H), or log-rank test (M), or Wilcoxon rank sum test with (O) or without (N) BH correction, or two-sided Chi-square test (P).