Tfl deletion induces extraordinary Cxcl13 secretion and cachexia in VavP- Bcl2 transgenic mice

Abstract

Loss of TFL, found in several types of lymphoma, induces excessive CXCL13 secretion through RNA dysregulation contributing to body weight loss and early death in lymphoma model mice. Follicular lymphoma (FL) is associated with overexpressed BCL-2 and other genetic aberrations, including 6q-. We identified a novel gene on 6q25, "Transformed follicular lymphoma (TFL)," from a transformed FL. TFL regulates several cytokines via mRNA degradation, which has been suggested to underlie resolving inflammation. Fluorescence in situ hybridization revealed a deletion of TFL occurred in 13.6% of various B-cell lymphoma samples. We developed VavP-bcl2 transgenic, TFL deficit mice (Bcl2-Tg/Tfl ^-/-) to seek how TFL affects disease progression in this lymphoma model. While Bcl2-Tg mice developed lymphadenopathy and died around 50 weeks, Bcl2-Tg/Tfl ^-/- mice lost body weight around 30 weeks and died about 20 weeks earlier than Bcl2-Tg mice. Furthermore, we found a unique B220^-IgM⁺ cell population in the bone marrow of Bcl2-Tg mice. cDNA array in this population revealed that Cxcl13 mRNA in Bcl2-Tg/Tfl ^-/- mice expressed significantly higher than Bcl2-Tg mice. In addition, bone marrow extracellular fluid and serum showed an extremely high Cxcl13 concentration in Bcl2-Tg/Tfl ^-/- mice. Among bone marrow cells, the B220^-IgM⁺ fraction was the main producer of Cxcl13 in culture. A reporter assay demonstrated TFL regulates CXCL-13 via induction of 3'UTR mRNA degradation in B lineage cells. These data suggest Tfl regulates Cxcl13 in B220^-IgM⁺ cells in the bone marrow, and a very high concentration of serum Cxcl13 arising from these cells may contribute to early death in lymphoma-bearing mice. Since several reports have suggested the association of CXCL13 expression with lymphoma, these findings provide new insights into cytokine regulation via TFL in lymphoma.

Keywords: Bcl-2; MCPIP-4; RNA regulation; Regnase-4; cachexia.

Figure 1

Loss of TFL locus in mature B cell neoplasms. (A) TFL DNA probe used for FISH analysis was derived from RP1-281H8 (BACPAC Resources Center). It covered 110kbp of the 6q25 region, which contains part of MAP3KIP2 and TFL. (B) D6Z1 was used as a centromere probe (CEP). When the number of CEP is equal to that of a local specific identifier (LSI; here, TFL probe), TFL is not considered to be deleted. When the number of LSI is less than that of CEP, TFL is considered to be deleted. Typical normal staining is shown. (C) The results of FISH analysis for bone marrow samples. (D) The results of FISH analysis for lymph nodes (LN) or tissue samples. (E) RNA blot analysis for various lymphoma tissue samples. TFL: Transformed follicular lymphoma, EtBr: Ethidium bromide, DLBCL: Diffuse large B cell lymphoma, DLBCL/FL: Follicular lymphoma transformed into DLBCL, FL III/II/I: Follicular lymphoma grade III/II/I, CLL: Chronic lymphocytic leukemia, LPL: Lymphoplasmacytic lymphoma, PTCL: Peripheral T cell lymphoma. (F) Quantification of RNA blot analysis is shown in (E). The image was measured and normalized by the value of EtBr.

Figure 2

Survival of Bcl2-Tg/Tfl ^-/- mice. (A) A Kaplan-Meier plot for wild type (N=14), Bcl2-Tg (N=28), Bcl2-Tg/Tfl ^+/- (N=23) and Bcl2-Tg/Tfl ^-/- mice (N=39). (B) Histological analysis of Bcl2-Tg and Bcl2-Tg/Tfl ^-/- mice. Hematoxylin and eosin (HE) staining was shown for the spleen (upper), lymph nodes (middle), or bone marrow (lower panel). (C) Ki-67 immunostaining was shown for the spleen (upper), lymph nodes (middle), or bone marrow (lower panel). For each section, at least three pairs of mice were analyzed. The representative pictures were shown. The percentage of positive cells was shown. Positive cells were counted for 5 independent areas in each organ. The p-value is shown as **<0.01 and ****<0.0001.

Figure 3

Bcl2-Tg/Tfl ^-/- mice decrease their body weight. Peanut agglutinin lectin (PNA) immunohistochemical staining (A) and NFκB1 immunostaining (B) in the spleen of Bcl2-Tg/Tfl ^+/+ and Bcl2-Tg/Tfl ^-/- mice were shown (Left). The percentage of positive cells was shown (Right). Positive cells were counted for 5 independent areas. (C) Body weight change of male Bcl2-Tg (N=17), and Bcl2-Tg/Tfl ^-/- mice (N=18). The p-value was shown as *<0.05, **<0.01, and ***<0.001.

Figure 4

Increased Cxcl13 secretion in Bcl2-Tg/Tfl^-/- mice. (A) Representative dot plots (Left panel) of cells from bone marrow or spleen of wild type, Tfl ^-/-, Bcl2-Tg, or Bcl2-Tg/Tfl ^-/- mice. Each B220⁺, B220^-IgM^-, and B220^-IgM⁺ fraction is shown in rectangles. The percentage of each fraction in wild-type, Tfl ^-/-, Bcl2-Tg, or Bcl2-Tg/Tfl ^-/- mice are shown for bone marrow (Right upper panel) and spleen (Right lower panel). (B) Real-time PCR analysis for Il-21, Cxcl13, Cdk6, Cdk1, and Cxcr7 in the B220^-IgM⁺ bone marrow cells between Bcl2-Tg (N=6) and Bcl2-Tg/Tfl ^-/- (N=5) mice. P values are shown in the figure. Cxcl13 measurement of bone marrow extra fluid (C) or plasma (D) at 40-44 weeks from Bcl2-Tg (N=7-8) and Bcl2-Tg/Tfl ^-/- (N=7-8) mice. Data from wild-type (N=9) and Tfl ^-/- (N=6) mice are shown as references. The course of Cxcl13 measurement of plasma (E) from Bcl2-Tg (N=38) and Bcl2-Tg/Tfl ^-/- (N=28) mice. The p-value is shown as ns (not significant), *<0.05, **<0.01, ***<0.001, and ****<0.0001.

Figure 5

B220^-IgM⁺cells are the main producers of Cxcl13. (A) ELISA assay to detect Cxcl13 of supernatant from cultured cells. Bone marrow (BM) and splenocytes from wild-type (WT) or Bcl2-Tg mice were cultured, and the supernatant of each cultured cell was provided for the measurement. Medium only was put as a reference. (B) Cxcl13 from BM and spleen cells of Tfl ^-/-, or Bcl2-Tg/Tfl ^-/- mice (N=2). The assay was performed similarly to (A). (C) Representative plot for B220⁺, IgM^-, or IgM⁺ fraction. The fraction gated as a red rectangle was provided for the following analysis. (D) CXCL-13 measurement from the supernatant of cultured cells of wild-type (WT) and Bcl2-Tg mice. B220+, IgM-, or IgM+ cells were provided with cell sorting indicated as (C). (E) Cxcl13 measurement from the supernatant of cultured cells of Tfl ^-/- or Bcl2-Tg/Tfl ^-/- mice. Two independent data are shown (Exp.1 and Exp. 2).

Figure 6

TFL degrades CXCL13 via 3’UTR. GFP mock (control) or GFP conjugated human TFL (TFL) vectors were transfected to several cell lines with Renilla luciferase conjugated with or without 3’UTR of CXCL13 or IL-2. Results of reporter assay of Renilla luciferase conjugated 3’UTR of CXCL13 (A) or IL-2 (B) were shown. Obtained luminescence data (Renilla luciferase/CXCL13 3’UTR and IL-2 3’UTR) were normalized by firefly luciferase as an internal control. The data were standardized by the data with the control vector, and the average of control data was arbitrarily adjusted to 1 for comparison. Experiments were repeated 3 times, and the p-value is shown as *<0.05 and **<0.01.