Maintenance of the plasma cell pool is independent of memory B cells

Abstract

Humoral memory to an antigen (Ag) is maintained for several decades in the form of memory B cells and serum Ab. In fact, plasma cells (PCs) that secrete Ab are known to be long-lived and could be solely responsible for maintaining the long-lived Ab titers. Alternatively, it has been proposed that the PC compartment is maintained for long periods by the differentiation of memory cells into long-lived PCs as a result of nonspecific stimulation. This model predicts accelerated decay of PC numbers in the absence of memory cells for the same Ag. To address this prediction, we have developed a mouse model system that combined the ability to deplete B cells with the ability to detect Ag-specific memory and PCs. After establishing an immune response, we depleted Ag-specific memory B cells with an anti-hCD20 mAb and determined the effect on the PC compartment over 16 weeks. Using a combination of surface markers, we demonstrated that memory B cells remained depleted over the course of the experiment. However, despite this absence of memory cells for an extended duration, PC numbers in spleen and bone marrow did not decline, which indicates that the PC compartment does not require a significant contribution from memory B cells for its maintenance and instead that PCs are sufficiently long-lived to maintain Ab titers over a long period without renewal. This observation settles an important controversy in B cell biology and has implications for the design of vaccines and for B cell depletion therapy in patients.

Fig. 1.

Experimental design for determining the effect of memory B cell depletion on PCs. The hCD20 × B1-8 mice were immunized with NP-CGG. After 15–19 weeks, the immunized mice were treated with anti-hCD20 mAb 2H7 for 2 weeks. During the recovery period of 16 weeks, the AFCs were allowed to decay in the absence of any input from memory B cells. Instead of anti-hCD20 mAb, the control-treated mice (Ctrl-Rx) were given mouse gamma globulin. Alum-treated and unimmunized mice received no further treatment. BrdU was injected in a cohort of unimmunized, alum-treated, and NP-CGG-treated mice every day from day 9 through day 15 (d9-d15) after immunization. Cohorts of mice were analyzed 15–19 weeks after immunization (predepletion cohort), immediately after anti-hCD20 therapy (depletion cohort), and 16 weeks after termination of therapy (recovery cohort) for numbers of total B cells, memory B cells, and PCs.

Fig. 2.

Depletion of memory B cells immediately after treatment with anti-hCD20 mAb 2H7 (depletion cohort). (a) Representative FACS plots showing the gating strategy for identification of memory B cells in alum-treated (Alum, n = 5; Top); immunized, control mouse gamma globulin-treated (Ctrl-Rx, n = 5; Center), and immunized, 2H7-treated (Rx, n = 5; Bottom) mice. The parent gates are shown on Top. (b) Total numbers of B cells, NIP⁺/Kappa⁻ cells, NIP⁺/IgG1⁺ cells, and NIP⁺/IgG1⁻ cells in spleens of Ctrl-treated (open symbols) and 2H7-treated (filled symbols) mice. (c) Representative FACS plots of NIP⁺ cells from alum-treated, Ctrl-treated, and 2H7-treated mice stained with IgG1 and BrdU. The percentages of total BrdU⁺ (gate with thin line) and BrdU⁺/IgG1⁺ (gate with thick line) are shown. (d) Comparison of total numbers of NIP⁺/BrdU⁺ and NIP⁺/BrdU⁺/IgG1⁺ memory subsets in spleens of Ctrl-treated (open symbols) and 2H7-treated (filled symbols) mice. (b and d) Horizontal lines represent median. **, P < 0.01, Mann–Whitney U test.

Fig. 3.

Short-term effect of depletion of B cells on PCs (depletion cohort). (a) Comparison of the number of NP-specific ELISPots per million splenocytes from Ctrl-treated mice (open symbols, n = 5) and 2H7-treated mice (filled symbols, n = 5). (b) Representative FACS plots showing phenotyping of PCs as CD138⁺/intracellular NIP^hi/surface NIP^int/CD45⁻ in the Ctrl-Rx (Upper) and 2H7 Rx (Lower) immunized mice. (c) Total number of PCs, identified by FACS, in spleens of Ctrl-treated (open symbols) and 2H7-treated (filled symbols) mice. (a and c) Horizontal lines represent median.

Fig. 4.

Long-term effect of depletion of B cells on memory B cells and PCs (recovery cohort). (a–c) Total numbers of B cells and NIP⁺/Kappa⁻ cells (a), NIP⁺/IgG1⁺ cells and NIP⁺/IgG1⁻ cells (b), and NIP⁺/BrdU⁺ cells and NIP⁺/BrdU⁺/IgG1⁺ cells (c) in spleens of alum-treated (open triangles, n ≥ 3), Ctrl-treated (open symbols, n ≥ 5), and 2H7-treated (filled symbols, n ≥ 5) mice. (d) Comparison of the number of NP-specific ELISPots per million splenocytes from alum-treated (open triangles, n = 7), Ctrl-treated (open symbols, n = 16), and 2H7-treated (filled symbols, n = 15) mice. (e) Total number of PCs, identified by FACS, in spleens of alum-treated (open triangles, n = 7), Ctrl-treated (open symbols, n = 16), and 2H7-treated (filled symbols, n = 15) mice. (f) Comparison of the number of NP-specific ELISPots per million BM cells from alum-treated (open triangles, n = 7), Ctrl-treated (open symbols, n = 16), and 2H7-treated (filled symbols, n = 15) mice. (g) Percentage of PCs, identified by FACS, in the BM of alum-treated (open triangles, n = 3), Ctrl-treated (open symbols, n = 5), and 2H7-treated (filled symbols, n = 6) mice. (a–g) Horizontal lines represent median. **, P < 0.01; ***, P ≤ 0.0005, Mann–Whitney U test.