CpG DNA activation and plasma-cell differentiation of CD27- naive human B cells

Abstract

Unmethylated CpG DNA activation of naive CD27- B cells has been reported to require B-cell-receptor (BCR) cross-linking. We describe a culture system using CpG DNA with sequential steps for T-cell-independent activation of naive CD19+CD27- human peripheral blood B cells that induces efficient CD138+ plasma-cell differentiation. CD27+ and CD27- B cells were cultured in a 3-step system: (1) days 0 to 4: CpG, IL-2/10/15; (2) days 4 to 7: IL-2/6/10/15 and anti-CD40L; (3) days 7 to 10: IL-6/15, IFN-alpha, hepatocyte growth factor, and hyaluronic acid. Both CD27+ and CD27- B cells up-regulated intracytoplasmic TLR-9 following CpG DNA activation. CD27- B-cell activation required cell-cell contact. Both naive and memory B cells progressed to a plasma-cell phenotype: CD19lowCD20lowCD27+CD38+HLA-DRlow. Seventy percent of the CD27--derived CD138+ cells demonstrated productive V chain rearrangements without somatic mutations, confirming their origin from naive precursors. Plasma cells derived from CD27+ B cells were primarily IgG+, while those from CD27- B cells were IgM+. Our results indicate that under certain conditions, naive B cells increase TLR-9 expression and proliferate to CpG DNA stimulation without BCR signaling. In addition to its immunologic significance, this system should be a valuable method to interrogate the antigenic specificity of naive B cells.

Figure 1

Activation of CD27⁻ B cells with CpG DNA is a function of cell-cell contact conditions. CD27⁻ and CD27⁺ B cells were labeled with CFSE and activated with CpG DNA₂₀₀₆. After 4 days, proliferation was analyzed by flow cytometry. Activation and proliferation of CD27⁻ naive B cells required an increase in cell density to 1 × 10⁵ cells per well for maximal proliferation to occur. In contrast, CD27⁺ B cells proliferated following CpG DNA at all cell concentrations and densities, although higher cell densities led to an increase in proliferation. IL-10 improved proliferation of both CD27⁺ and CD27⁻ cells, while the combination of IL-2, IL-10, and IL-15 and cell-cell contact gave maximal proliferation. Results are representative of 15 independent experiments from 9 different blood donors.

Figure 2

Intracellular TLR-9 correlates with proliferation following CpG DNA stimulation for both naive (CD27⁻) and memory (CD27⁺) B cells. Intracytoplasmic TLR-9 was measured by flow cytometry immediately after isolating naive and memory B cells (shaded histogram) for both populations. The bar represents the isotype control measurement. Simultaneous intracytoplasmic TLR-9 and CFSE measurements were obtained following CpG DNA stimulation in the presence of IL-2, IL-10, and IL-15 at the time intervals listed for each population. Both cell types expressed a basal level of intracytoplasmic TLR-9, while memory B cells had a faster increase in intracytoplasmic levels than their CD27⁻ counterparts. Results are representative of 7 independent experiments from different blood donors.

Figure 3

Surface phenotype of CpG DNA₂₀₀₆ activated CD27⁺ and CD27⁻ B cells with CD138 induction. Surface phenotype of the CD27⁺ and CD27⁻ peripheral blood B cells beginning with the initial isolation and following each culture phase. Both CD27⁺ and CD27⁻ B cells were activated by CpG DNA and underwent plasmablast differentiation, ultimately resulting in a surface phenotype characteristic of human bone marrow–resident plasma cells: CD20⁻, CD38⁺⁺⁺, CD138⁺⁺, and HLA-DR⁻. Results are representative of 15 independent experiments from 12 different blood donors.

Figure 4

Death of CD27⁺- and CD27⁻-derived plasma cells during differentiation in vitro. (A-B) At the indicated time points, cells were stained with trypan blue, and the absolute number of live (open diamonds with black lines) and dead (open circles with gray lines) cells were counted. (C) The percentage of living cells at each culture day for CD27⁻ (open triangles with thin line) and CD27⁺ (open squares with thick line) cells. (D) Western blot analysis of bcl-2 protein expression during in vitro plasma-cell differentiation from naive (thin line) and memory (solid line) B-cell precursors. (A-D) Error bars are mean ± standard error for all plots (n = 2 experiments from 2 subjects).

Figure 5

Morphology of CpG DNA₂₀₀₆–activated CD27⁺ and CD27⁻ B cells. Wright-Giemsa staining was used to demonstrate morphologic differences between the initial CD27⁺ and CD27⁻ peripheral blood B cells as well as differentiation changes following each phase of the culture system for both populations. By the end of phase III, CD27⁺ and CD27⁻ B cells had an identical plasmablast morphology. Results are representative of 4 independent experiments, each from a different blood donor. Images were photographed using a Nikon Labophot microscope (100×/1.17 numerical aperture oil objective) with a Nikon Coolpix digital camera using the manufacturer's software (Nikon, Melville, NY). JPEG images were viewed using Adobe Photoshop CS2 (Adobe Systems, San Jose, CA), and contrast adjustments were made.

Figure 6

Induction of IgG-producing plasma cells from CD27⁺ B cells and IgM-producing plasma cells from CD27⁻ B cells. (A,C) Intracytoplasmic immunoglobulin staining was used to assess immunoglobulin production with each phase. Intracellular IgM increased in both (A) CD27⁺ and (C) CD27⁻ B cells by the end of phase II, and then declined after phase III. In contrast, CD27⁺ cells had a steady increase in intracellular IgG staining through terminal differentiation into CD138⁺ plasma cells. Figures are representative of 4 independent experiments with cells from 4 different blood donors. (B,D) IgG, IgM, and IgA were measured by ELISA from supernatants taken at the end of each culture phase for (B) CD27⁺ and (D) CD27⁻ B cells. Error bars are mean ± standard error for all plots. Experimental data are from 5 separate experiments each with different blood donors.