Nuclear factor kappa B is required for the development of marginal zone B lymphocytes

Abstract

Although immunoglobulin (Ig)M(hi)IgD(lo/-)CD21(hi) marginal zone B cells represent a significant proportion of naive peripheral splenic B lymphocytes, few of the genes that regulate their development have been identified. This subset of peripheral B cells fails to emerge in mice that lack nuclear factor (NF)-kappa Bp50. Less drastic reductions in marginal zone B cell numbers are also seen in the spleens of recombination activating gene (Rag)-2(-/-) mice reconstituted with NF-kappa Bp65(-/-) fetal liver cells and in c-Rel(-/-) mice. In contrast, steady-state levels of IgD(hi) splenic follicular B cells are not significantly reduced in the absence of NF-kappa Bp50, NF-kappa Bp65, or c-Rel. Reconstitution of B cells in Rag-2(-/-) mice with a mixture of p50(-/-)/p65(-/-) fetal liver cells and Rag-2(-/-) bone marrow cells revealed that the generation of marginal zone B cells requires the expression of NF-kappa B in developing B cells, as opposed to supporting cells.

Figure 1

NF-κBp50 is required for MZ B cell generation. (a) Immunofluorescence performed on splenic cryosections with anti-IgM–FITC and MOMA-1–PE revealed a marked reduction in MZ B cells in the p50^−/− mouse (right). Multiple sections revealed similar results. WT, wild-type. (b) Splenic IgM^hiIgD^hi and IgD^hiIgM^lo follicular B cell populations appear grossly normal in the p50^−/− mouse. (c) IgM^hiIgD^loCD21^hi MZ B cells are markedly reduced in p50^−/− mice. Three wild-type and three p50^−/− mice were analyzed in the above studies. The results shown are representative. (d) IgM^hiIgD^loCD1d^hi and IgM^hiIgD^loCD38^hi MZ B cells are markedly reduced in p50^−/− mice. Three wild-type and three p50^−/− mice were analyzed in the above studies. The results shown are representative.

Figure 1

NF-κBp50 is required for MZ B cell generation. (a) Immunofluorescence performed on splenic cryosections with anti-IgM–FITC and MOMA-1–PE revealed a marked reduction in MZ B cells in the p50^−/− mouse (right). Multiple sections revealed similar results. WT, wild-type. (b) Splenic IgM^hiIgD^hi and IgD^hiIgM^lo follicular B cell populations appear grossly normal in the p50^−/− mouse. (c) IgM^hiIgD^loCD21^hi MZ B cells are markedly reduced in p50^−/− mice. Three wild-type and three p50^−/− mice were analyzed in the above studies. The results shown are representative. (d) IgM^hiIgD^loCD1d^hi and IgM^hiIgD^loCD38^hi MZ B cells are markedly reduced in p50^−/− mice. Three wild-type and three p50^−/− mice were analyzed in the above studies. The results shown are representative.

Figure 1

NF-κBp50 is required for MZ B cell generation. (a) Immunofluorescence performed on splenic cryosections with anti-IgM–FITC and MOMA-1–PE revealed a marked reduction in MZ B cells in the p50^−/− mouse (right). Multiple sections revealed similar results. WT, wild-type. (b) Splenic IgM^hiIgD^hi and IgD^hiIgM^lo follicular B cell populations appear grossly normal in the p50^−/− mouse. (c) IgM^hiIgD^loCD21^hi MZ B cells are markedly reduced in p50^−/− mice. Three wild-type and three p50^−/− mice were analyzed in the above studies. The results shown are representative. (d) IgM^hiIgD^loCD1d^hi and IgM^hiIgD^loCD38^hi MZ B cells are markedly reduced in p50^−/− mice. Three wild-type and three p50^−/− mice were analyzed in the above studies. The results shown are representative.

Figure 1

NF-κBp50 is required for MZ B cell generation. (a) Immunofluorescence performed on splenic cryosections with anti-IgM–FITC and MOMA-1–PE revealed a marked reduction in MZ B cells in the p50^−/− mouse (right). Multiple sections revealed similar results. WT, wild-type. (b) Splenic IgM^hiIgD^hi and IgD^hiIgM^lo follicular B cell populations appear grossly normal in the p50^−/− mouse. (c) IgM^hiIgD^loCD21^hi MZ B cells are markedly reduced in p50^−/− mice. Three wild-type and three p50^−/− mice were analyzed in the above studies. The results shown are representative. (d) IgM^hiIgD^loCD1d^hi and IgM^hiIgD^loCD38^hi MZ B cells are markedly reduced in p50^−/− mice. Three wild-type and three p50^−/− mice were analyzed in the above studies. The results shown are representative.

Figure 2

A T cell–independent type II antigen is captured in vivo by wild-type (WT) MZ B cells but not by any of the B cell populations in p50^−/− spleens. Splenocytes were isolated from wild-type and p50^−/− mice that were or were not immunized with TNP-Ficoll. In immunized mice, cells that could be stained with anti-TNP antibodies were readily visualized within the CD45R⁺CD21^hiCD23^lo MZ B cell gate. Significant numbers of TNP-staining cells were not seen in any of the splenic B cell pools in p50^−/− mice.

Figure 3

NF-κBp65 and c-Rel contribute to MZ B cell development. (a) IgM^hiIgD^loCD21^hi MZ B cells are markedly reduced in Rag-2^−/− mice reconstituted with p65^−/− fetal liver and Rag-2^−/− bone marrow cells (bottom panels). Splenic IgD^hiIgM^lo follicular B cell numbers appear grossly normal (top panels). Two p65^−/− chimeras were analyzed in these studies. The results shown are representative. WT, wild-type. (b) IgM^hiIgD^loCD21^hi MZ B cells are markedly reduced in c-Rel^−/− mice (bottom panels), but splenic IgD^hiIgM^lo follicular B cell numbers appear grossly normal (top panels). Two control and two c-Rel^−/− mice were analyzed. The results shown are representative.

Figure 3

NF-κBp65 and c-Rel contribute to MZ B cell development. (a) IgM^hiIgD^loCD21^hi MZ B cells are markedly reduced in Rag-2^−/− mice reconstituted with p65^−/− fetal liver and Rag-2^−/− bone marrow cells (bottom panels). Splenic IgD^hiIgM^lo follicular B cell numbers appear grossly normal (top panels). Two p65^−/− chimeras were analyzed in these studies. The results shown are representative. WT, wild-type. (b) IgM^hiIgD^loCD21^hi MZ B cells are markedly reduced in c-Rel^−/− mice (bottom panels), but splenic IgD^hiIgM^lo follicular B cell numbers appear grossly normal (top panels). Two control and two c-Rel^−/− mice were analyzed. The results shown are representative.

Figure 4

NF-κBp50 must be expressed in developing lymphocytes for MZ B cell generation. Rag-2^−/− mice were reconstituted with fetal liver from p50^−/−/p65^−/− mice, and bone marrow from Rag-2^−/− mice. Reconstitution results in clearly identifiable IgM^hiIgD^lo, IgM^hiIgD^hi, and IgD^hi IgM^lo subpopulations (top panels). In the IgM^hiIgD^lo fraction, CD21^hi cells were markedly reduced in the p50^−/−/p65^−/− chimeras (bottom panels). Six separate p50^−/−/p65^−/− chimeric mice were analyzed in these studies. The results shown are representative. WT, wild-type.

Figure 5

Absolute numbers of splenic B cells in wild-type (WT) and mutant mice. (a) Absolute numbers of newly formed, follicular, and MZ B cells in wild-type, p50^−/−, c-Rel^−/−, and Rag-2^−/− mice reconstituted with mixtures including wild-type, p65^−/−, and p50^−/−/p65^−/− fetal liver cells, respectively, in combination with Rag-2^−/− bone marrow cells. (b) Absolute numbers of MZ B cells in the above mice depicted using an expanded scale.

Figure 5

Absolute numbers of splenic B cells in wild-type (WT) and mutant mice. (a) Absolute numbers of newly formed, follicular, and MZ B cells in wild-type, p50^−/−, c-Rel^−/−, and Rag-2^−/− mice reconstituted with mixtures including wild-type, p65^−/−, and p50^−/−/p65^−/− fetal liver cells, respectively, in combination with Rag-2^−/− bone marrow cells. (b) Absolute numbers of MZ B cells in the above mice depicted using an expanded scale.