Isotype-specific selection of high affinity memory B cells in nasal-associated lymphoid tissue

Abstract

Mucosal immunoglobulin (Ig)A dominance has been proposed to be associated with preferential class switch recombination (CSR) to the IgA heavy chain constant region, Calpha. Here, we report that B cell activation in nasal-associated lymphoid tissue (NALT) upon stimulation with the hapten (4-hydroxy-3-nitrophenyl)acetyl (NP) coupled to chicken gamma globulin caused an anti-NP memory response dominated by high affinity IgA antibodies. In the response, however, NP-specific IgG(+) B cells expanded and sustained their number as a major population in germinal centers (GCs), supporting the view that CSR to IgG heavy chain constant region, Cgamma, operated efficiently in NALT. Both IgG(+) and IgA(+) GC B cells accumulated somatic mutations, indicative of affinity maturation to a similar extent, suggesting that both types of cell were equally selected by antigen. Despite the selection in GCs, high affinity NP-specific B cells were barely detected in the IgG memory compartment, whereas such cells dominated the IgA memory compartment. Taken together with the analysis of the V(H) gene clonotype in GC and memory B cells, we propose that NALT is equipped with a unique machinery providing IgA-specific enrichment of high affinity cells into the memory compartment, facilitating immunity with high affinity and noninflammatory secretory antibodies.

Figure 1.

Flow cytometric analysis of NP-specific B cells. (A) Spleen cells were obtained from both nonimmunized C57BL/6 mice (panel a) and mice immunized intraperitoneally with 2 μg of NP-CG and 2 μg of CT 9 d earlier (panels c and e). Pooled NALT cells (n = 20–25 mice) were also obtained from nonimmunized C57BL/6 mice (panel b) as well as from mice immunized i.n. with 2 μg of NP-CG and 2 μg of CT 7 d earlier (panels d and f). Cells were stained with anti-B220^TX and anti-CD38^APC mAbs and with NIP-BSA^PE, PNA^FITC, and PI. Viable cells (PI⁻) were selected under a lymphocyte gate based on forward and side light scatter. Thereafter, the NIP-binding cells were gated, and the expression of B220 and CD38 was analyzed (panels c and d). Finally, the CD38^dull population was gated, and binding to PNA was analyzed among the NIP-binding cells (panels e and f). Figures in each region represent the number of CD38^dull/NIP-binding/B220⁺ or B220^dull cells (panels a–d) and that of CD38^dull/NIP-binding/B220⁺/PNA⁺ or CD38^dull/NIP-binding/B220^dull/PNA⁻ cells (panels e and f) per 100,000 total viable cells. (B–D) Cells were prepared from pooled NALT (B, n = 20–25 mice) and PCLNs (C and D, n = 5–20) at various times after i.n. immunization (as above) as well as from nonimmunized mice. To monitor NP-specific/λ⁺ GC and memory B cells, cells were first stained with biotinylated mAbs against IgM, IgD, CD43, CD5, and CD90. This was followed by staining with NIP-BSA^PE and anti-CD38^TX, anti-λ^FITC, and anti-B220^APC mAbs. After washing, cells were stained with streptavidin^TC and PI to exclude from the analysis dead cells and the cells recognized by the biotinylated mAbs. To track NP-specific/λ⁺ IgA⁺ or IgG2b⁺ GC and memory B cells, cells were stained with anti-CD38^TX mAb, anti-λ^FITC mAb, NIP-BSA^PE, and with biotinylated anti-IgA or IgG2b mAb, followed by incubation with APC-conjugated streptavidin and PI. Viable cells were selected under a lymphocyte gate based on forward and side light scatter. Shown are the frequencies of NIP-binding/λ⁺ GC B cells (□) and NIP-binding/λ⁺ IgG2b⁺ (•) or IgA⁺ (○) GC B cells in NALT (B) and PCLNs (C). Also shown are the frequencies of NIP-binding/λ⁺ memory B cells (□) and NIP-binding/λ⁺ IgG2b⁺ (•) or IgA⁺ (○) memory B cells in PCLN (D). Each symbol represents the result from a single experiment. Frequency was estimated by dividing the number of cells with each phenotype by the total number of viable cells. At least 10⁵ events were collected for each frequency determination.

Figure 2.

Accumulation of somatic mutations and affinity maturation in NP-specific GC B cells (A, B, E, F) and memory B cells (C, D, G, H) in NALTs (A–D) and PCLNs (E–H) of C57BL/6 mice after i.n. immunization with NP-CG and adjuvant CT. NIP-binding/λ⁺ GC B cells (CD38^dull/B220⁺/IgM⁻/IgD⁻/CD43⁻/CD5⁻/CD90⁻) and NIP binding/λ⁺ memory B cells (CD38⁺/B220⁺/IgM⁻/IgD⁻/CD43⁻/ CD5⁻/CD90⁻) were purified from the pooled NALTs (n = 80–100) or PCLNs (n = 10–30) of NP-CG primed C57BL/6 mice at various times after immunization. Messenger RNA was purified from sorted cells, and cDNA synthesized by reverse transcriptase with specific primers for either Cγ (A, C, E, G) or Cα (B, D, F, H). Thereafter, the rearranged V_H genes were amplified from the cDNA by PCR. The frequency of artificial substitutions was calculated to be 1.210⁻⁴ base pair in control experiments using messenger RNA from the hybridoma B1–8 (see Materials and Methods). V_H genes encoded by V186.2 were dominant among the recovered clones and were selected for further analysis (see Table I). Shown are the numbers of somatic mutations in individual V186.2 genes with (•) or without (○) a Trp-to-Leu substitution at position 33 (Leu33) and genes with Gly at amino acid 99 in CDR3 (▵) recovered at the indicated date. The average number of somatic mutations of V186.2 genes (white columns) at the indicated date was determined by dividing the total number of somatic mutations observed by the number of clones. The sequence data are available from GenBank/EMBL/DDBJ under accession nos. AB057817-AB058198 and AB067398, AB067399, AB067400, AB067401, AB067402, AB067403, AB067404, AB067405, AB067406, AB067407, AB067408, AB067409, AB067410, AB067411, AB067412, AB067413, AB067414, AB067415, AB067416, AB067417, AB067418, AB067419, AB067420, AB067421, AB067422, AB067423, AB067424, AB067425, AB067426, AB067427, AB067428, AB067429, AB067430, AB067431, AB067432, AB067433, AB067434, AB067435, AB067436, AB067437, AB067438, AB067439, AB067440.

Figure 3.

Dominance of high affinity IgA in the secondary response after i.n. immunization. C57BL/6 mice were primed i.n. with 2 μg of NP-CG and CT, followed by intraperitoneal challenge with 20 μg of either NP-CG (a) or CG (c) 5 wk later. As a control, mice were i.n. immunized with 2 μg of CG plus CT, followed by intraperitoneal challenge with NP-CG (b). 7 d after the intraperitoneal challenge, immune serum, spleen, and bone marrow were obtained from individual mice (n = 4). All data are representative of three independent experiments. (A) The numbers of total (○) and high affinity (•) NP-binding/λ₁ ⁺ ASCs with IgA, IgG2b, and IgG1 subclasses were determined by ELISPOT. Columns represent the average number of total (white columns) and high affinity ASCs (black columns). (B) Total (○) and high affinity (•) NP-specific antibody titers of IgA, IgG2b, and IgG1 subclass in serum from individual mice (n = 4) were determined by ELISA. (C) Isotype-specific serum antibody affinity maturation was estimated for the individual mice (n = 4) shown in group a of B by dividing the high affinity NP-specific antibody titer by the total NP-specific antibody titer. Serum affinity maturation of NP-specific antibodies between IgA and IgG2b or IgG1 was statistically evaluated by the Mann-Whitney nonparametric tests (two-tailed).

Figure 4.

High affinity IgA memory cells generated in NALT migrate into nonmucosal lymphoid tissues such as spleen. C57BL/6 were immunized i.n. with 2 μg of CT and 2 μg of either NP-CG (a) or CG (b). 4 wk later, single cell suspensions were prepared from pooled spleens (n = 4) and transferred into RAG-1^−/− mice on a C57BL/6 background by intravenous injection (6 × 10⁷ cells per recipient). 12–24 h after the transfer, the reconstituted RAG-1^−/− mice were immunized intraperitoneal with 50 μg of NP-CG. All data are representative of three independent experiments. (A) 7 d later, the levels of NP-specific antibodies (○) and high affinity NP-specific antibodies (•) of IgA, IgG2b, and IgG1 subclasses in individual sera (n = 4) were determined by ELISA. White and black columns represent the mean titer of total and high affinity anti-NP antibodies, respectively. (B) Isotype-specific serum antibody affinity maturation was estimated for the individual mice (n = 4) shown in group (a) of A by dividing the high affinity NP-specific antibody titer by the total NP-specific antibody titer. Serum affinity maturation of NP-specific antibodies between IgA and IgG2b or IgG1 was statistically evaluated by the Mann-Whitney nonparametric tests (two-tailed).