Systematic comparison of gene expression between murine memory and naive B cells demonstrates that memory B cells have unique signaling capabilities

Abstract

Memory B cells play essential roles in the maintenance of long-term immunity and may be important in the pathogenesis of autoimmune disease, but how these cells are distinguished from their naive precursors is poorly understood. To address this, it would be important to understand how gene expression differs between memory and naive B cells to elucidate memory-specific functions. Using model systems that help overcome the lack of murine memory-specific markers and the low frequency of Ag-specific memory and naive cells, we undertook a global comparison of gene expression between memory B cells and their naive precursors. We identified genes with differential expression and confirmed the differential expression of many of these by quantitative RT-PCR and of some of these at the protein level. Our initial analysis revealed differential expression patterns of genes that regulate signaling. Memory B cells have increased expression of genes important in regulating adenosine signaling and in modulating cAMP responses. Furthermore, memory B cells up-regulate receptors that are essential for embryonic stem cell self-renewal. We further demonstrate that one of these, leukemia inhibitory factor receptor, can initiate functional signaling in memory B cells whereas it does not in naive B cells. Thus, memory and naive B cells are intrinsically wired to signal differently from one another and express a functional signaling pathway that is known to maintain stem cells in other lineages.

FIGURE 1

Isolation of memory and naive B cells for the first-generation Affymetrix screen. A, A schematic depicting the strain of mice and immunization strategy used to generate naive and memory B cells. Naive B cells were purified from spleens of unimmunized 6- to 8-wk-old mV_H186.2 Tg J_H KO mice. Memory B cells were isolated from spleens of mice 12–28 wk postsecondary immunization with 50 μg of NP₂₅-CGG in alum i.p. B, Representative plots, gating strategies, and population frequencies for memory (top) and naive (bottom) FACS sorts. Splenic B cells were first purified by negative selection using anti-Thy1.2 and -CD43 by AutoMACS (Miltenyi Biotec) and subsequently stained and sorted with a FACSVantage (BD Biosciences). Live lymphocytes were gated by FSC and SSC profiles and by PI exclusion (data not shown) then B220⁺ (left) and subsequently NIP-binding κ^low/neg (right) B cells were sorted. C, Postsort purities of memory (top) and naive (bottom) cells among all live cells are shown. D, An overview of the microarray approach. Differences between the first and second Affymetrix screens are highlighted in bold.

FIGURE 2

Isolation of memory and naive B cells for the second-generation Affymetrix screen. A, Schematic of use of cell transfer to generate Ag-specific memory B cells in the absence of newly generated naive precursors. B, Sorting strategy. Splenocytes were stained and single, live B lymphocytes were gated by FSC and SSC profiles and PI exclusion (data not shown). B220⁺ NIP-binding B cells lacking IgM^b (the recipient allotype) were gated as shown. Percentages of the parental population are indicated. Presort, live B cells are shown; postsort, all live cells are shown. For comparison, plots are shown from a control animal given donor mV_H186.2 splenocytes and immunized with alum alone and in a nontransferred recipient immunized with NP₂₅-CGG. NP-binding B cells were negligible in control mice. C, Naive B cells were purified from spleens of unimmunized 8-wk-old female mV_H186.2 Tg J_H KO Jκ KO mice. Single, live B lymphocytes were gated by FSC and SSC profiles, PI exclusion, and anti-B220-binding on a FACSAria. AA4.1⁻ NIP-binding cells were sorted. D, Overview of the microarray approach. Differences between this and the first Affymetrix screens (Fig. 1) are shown in bold.

FIGURE 3

Correlation between Affymetrix and qPCR determinations of differential gene expression. Log₂ values of fold differences in mRNA levels (memory/naive) determined by the first Affymetrix screen and by qPCR are compared. Each point represents an individual gene that was identified in the first Affymetrix screen as differentially expressed between memory and naive B cells with a Welch t test p ≤ 0.05 and a fold difference ≥3. The Pearson's correlation coefficient was 0.835.

FIGURE 4

Expression of the adenosine receptor A2A protein is elevated among memory B cells. Surface expression of the adenosine receptor A2A was compared by flow cytometry in memory and naive B cells. Naive splenic B cells were from unimmunized mV_H186.2 Tg J_H KO Jκ KO mice. In the top histogram, memory cells were from mV_H186.2 Tg J_H KO mice ≥12-wk postsecondary immunization with NP₂₅-CGG (system 1). In the bottom histogram, memory B cells were derived from NP-specific B cells from mV_H186.2 J_H KO Jκ KO donors after transfer into AM14 Tg Vκ8R Tg CB.17 recipients and subsequent immunization (system 2). Shown are live lymphocyte-gated B220⁺ NP-binding κ^low/neg B lymphocytes. Expression among memory B cells is shown in dark black and naive in gray. Isotype control stain is shaded. Shown are representative plots from four (system 1) and two (system 2) mice.

FIGURE 5

Expression of the protein kinase A regulatory subunit IIβ (PKA RIIβ) is up-regulated in memory B cells. Protein expression of PKA RIIβ and actin were compared in memory and naive B cells by Western blot analysis. Naive B cells were generated and FACS purified as described in Fig. 2. Memory B cells were generated and purified as in Fig. 1, omitting the MACS presorting step. Shown is a representative blot of two mice, from a total of five each of naive (N) and memory (M) mice.

FIGURE 6

Expression of the protein kinase C ζ (PKC ζ) is up-regulated in memory B cells. Protein expression of PKC ζ and actin were compared in memory (M) and naive (N) B cells by Western blot analysis. Naive B cells were purified from spleens of unimmunized mV_H186.2 Tg J_H KO Jκ KO mice. To analyze memory B cells, total B cells, of which ~15% are NP-binding memory B cells, were purified from spleens of mV_H186.2 Tg J_H KO mice >12-wk postsecondary immunization with NP₂₅-CGG in alum i.p. B cells were isolated by negative selection using the EasySep magnetic separation system. Thus, this level of expression underrepresents by as much as a factor of 6 the expression that would be seen in 100% pure memory cells. Each lane represents cells purified from an individual mouse.

FIGURE 7

Memory B cells up-regulate Lifr and phosphorylate Stat3 in response to Lif stimulation more robustly than do naive B cells. A, Protein expression of Lifr and actin were compared in memory and naive B cells by Western blot analysis, as described in Fig. 5. Each lane represents cells from an individual mouse. Data are representative of findings in five mice of each type. B–D, Splenocytes from 6- to 10-wk-old naive mV_H186.2 J_HD KO Jκ KO mice or mV_H186.2 Tg J_HD KO mice 15–20 wk postsecondary immunization with NP-CGG in alum i.p. were rested in culture for 2 h at 37°C in standard medium of RPMI 1640 with 10% FetalPlex and then stimulated with 1 ng/ml (B), 0.1, 1, or 10 ng/ml (C) or 10 ng/ml (D) mouse Lif for 0, 15, or 30 min. Cells were then fixed with 1.6% PFA, methanol permeabilized, and stained with PE-anti-phospho-Stat3 or PE-mouse IgG2a isotype control. Shown are data from live-gated (by FSC and SSC) B220⁺ λ⁺ B cells. B, Representative FACS plots of changing P-Stat3 levels. C, Lif-induced P-Stat3 dose-response curves of memory and naive B cells. P-Stat3 levels are plotted as isotype control-corrected median fluorescent intensities (MFIs). D, For memory (Mem) and naive (Nve) B cells, change in phosphorylated Stat3 or isotype control (IC) staining intensity between 0 and 15 min is shown.

FIGURE 8

Venn diagram of genes identified by Bhattacharya et al. (24) and this study. The Bhattacharya et al. (24) lists of genes expressed ≥2-fold differently between naive and memory B cells were taken directly from the published manuscript. The lists labeled “Tomayko et al.” were generated from the second-generation microarray screen as described in Table I, using a 2-fold memory vs naive cutoff. Bolded numbers indicate numbers of Affymetrix probe sets found in one vs both lists. Select specific genes discussed in either manuscript are included. Italicized genes in the Bhattacharya et al. only list had a trend toward similar expression in our data set (Ccr6, Rgs18) or have been previously demonstrated by us to be differentially expressed among memory and naive B cells at the protein level by FACS (CD62L) (3, 25). For genes found only in the Bhattacharya et al. list, corresponding average fold-differences and Student's t test p values for the Tomayko et al. data are listed (raw data are detailed in supplemental Table Ib). For genes with multiple probe sets, all fold values were averaged, except for Bcl2, Mll3, and Tcf4, in which case only probe sets with p values <0.05 were considered.