M17, a gene specific for germinal center (GC) B cells and a prognostic marker for GC B-cell lymphomas, is dispensable for the GC reaction in mice

Abstract

In T-cell-dependent antibody responses, antigen-specific B cells undergo a phase of secondary antibody diversification in germinal centers (GCs). Somatic hypermutation (SHM) introduces mutations into the rearranged immunoglobulin (Ig) variable (V) region genes, and class-switch recombination (CSR) alters the Ig heavy (H) chain constant region. Aberrant SHM or CSR is thought to contribute to the development of GC-derived B-cell malignancies. Diffuse large B-cell lymphomas (DLBCLs) are a heterogeneous group of such GC-derived tumors. Based on their gene expression profile, DLBCLs can be divided into activated B-cell-like and GC-like subgroups. The human gene HGAL is predominantly expressed in GCs. It is also part of the gene expression signature of GC-like DLBCL, and its high expression in DLBCL has been associated with a better clinical prognosis. We have generated mice deficient of the HGAL homologue M17 in order to investigate its functional significance. The mutant animals form normal GCs, undergo efficient CSR and SHM, and mount T-cell-dependent antibody responses similar to wild-type controls. Thus, M17 is dispensable for the GC reaction, and its potential function in the pathogenesis of DLBCL remains elusive.

Figure 1.

Generation of M17^–/– mice. (A) Schematic representation of the gene targeting in the M17 locus by homologous recombination. C57BL/6-derived ES cells were targeted with a vector containing the loxP-flanked exons 4 and 5, a frt-flanked IRES-gfp cassette, and a neomycin resistance cassette for positive selection. Negative selection of clones harboring randomly integrated vectors was mediated by a thymidine kinase gene. Only exons 3 to 5 are shown. Rectangles indicate coding DNA (black, translated region; gray, untranslated region); filled triangles, loxP sites; ovals, frt sites; RI, EcoRI; and BII, BglII. Bold lines indicate regions of homology and Southern probes are shown as thin black lines under the wild-type locus. The map is not drawn to scale. (B) Successful homologous recombination was identified by Southern blot of EcoRI-digested genomic ES cell DNA and probe A located 5′ of exon 3. The wild-type fragment migrates at 6.9 kb, while the fragment from the targeted locus migrates at 4.6 kb. Cre-mediated deletion of M17 exons 4 and 5 was confirmed by Southern blot of BglII-digested genomic DNA using probe C located 3′ of exon 3. The wild-type fragment migrates at 2.1 kb and the fragment of the deleted locus migrates at 1.0 kb. (C) Confirmation of the successful inactivation of the M17 gene by RT-PCR. RNA was isolated from sorted CD19⁺PNA^–Fas^– naive B cells (N) or CD19⁺PNA⁺Fas⁺ GC B cells from either M17^+/+ or M17^–/– mice and reverse transcribed using an oligoT primer. PCR products were amplified with primers annealing either in exons 1 and 4 (M17 1/4) or in exons 1 and 3 (M17 1/3). Intron-spanning primers annealing in the β-actin gene were used to control for equal amounts of RNA. The PCR products were consistent with the expected sizes for the 2 alternative transcripts of M17.

Figure 2.

Cell numbers and GC formation. (A) Expression of M17 cDNA in developing B cells. Bone marrow–derived lymphocytes were sorted into B220^– cells, B220⁺ IgM^– c-kit⁺ pro-B cells, B220⁺ IgM^– CD25⁺ pre-B cells, and B220⁺ IgM⁺ B cells by fluorescent-activated cell sorting (FACS), and the expression of M17 mRNA transcripts in these cell populations was determined by RT-PCR. (B) The total number of cells in peripheral lymphoid organs. SP indicates spleen; MLN, mesenteric lymph nodes; and PP, Peyer patches. (C) Generation of GC B cells. Mice were immunized with 50 μg NP-CG and analyzed for the presence of CD19⁺PNA⁺Fas⁺ GC B cells 14 days after immunization. Only CD19⁺ cells are shown. Numbers represent the mean in percent plus standard deviation. SP indicates spleen; MLN, mesenteric lymph nodes; and PP, Peyer patches. (D-E) GC architecture in M17^–/– mice. Mice were immunized with 50 μg NP-CG. Frozen splenic sections were prepared on day 14 after immunization and analyzed by immunofluorescence. (D) Sections were incubated with αCD19 mAb (green) and PNA (red) to visualize B-cell follicles and GCs, respectively. (E) Sections were incubated with αBCL-6 mAb (brown) and counterstained with hematoxylin (blue). Representative pictures are shown. SP indicates spleen; PP, Peyer patches. Numbers below pictures indicate the magnification.

Figure 3.

Antibody titers and class-switch recombination. (A) M17 is up-regulated by IL-4. Isolated splenocytes were MACS depleted of CD43⁺ cells and subsequently activated with the indicated stimuli. Following the isolation of total RNA 48 hours later, RT-PCR was performed using intron-spanning primers for the M17 and β-actin genes. Naive B indicates naive B cells in vitro; GC, GC B cells; and H₂0, water control. A representative experiment is shown. (B) Antibody titers in the serum of unimmunized wild-type and M17^–/– mice were determined in an ELISA assay. Each circle represents one mouse. Black bars indicate the geometric means. Closed circles indicate wild-type mice; open circles, M17-deficient mice. (C) In vitro stimulation of isolated B cells of M17^–/– mice and wild-type controls. B cells were induced to undergo CSR with the indicated stimuli. The percentage of class-switched cells was determined 4 days later by flow cytometry. Numbers in the graphs represent the percentages of switched cells. A representative experiment is shown. (D) Percentage of IgA⁺ GC B cells in the Peyer patches of M17^–/– mice and wild-type controls. The differences in percentages of IgA⁺ B cells were not statistically significant (P = .15).

Figure 4.

T-cell–dependent immune response of M17^–/– mice and generation of memory B cells. (A) Primary immunization with 50 μg (left panel) or 10 μg (right panel) NP-CG in alum. Each circle represents one mouse. Bars indicate the geometric means. Closed circle indicates M17^+/+ mice; open circle, M17^–/– mice. (B) Secondary immunization with 10 μg NP-CG without alum of mice previously immunized with 50 μg (left panel) or 10 μg (right panel) NP-CG. (C) Amount of splenic IgG1⁺ memory B cells 9 weeks after the secondary immunization. Numbers are in percent plus standard deviation of total B cells (n = 5). The differences were not statistically significant.