Nonredundant roles for B cell-derived IL-10 in immune counter-regulation

Abstract

IL-10 plays a central role in restraining the vigor of inflammatory responses, but the critical cellular sources of this counter-regulatory cytokine remain speculative in many disease models. Using a novel IL-10 transcriptional reporter mouse, we found an unexpected predominance of B cells (including plasma cells) among IL-10-expressing cells in peripheral lymphoid tissues at baseline and during diverse models of in vivo immunological challenge. Use of a novel B cell-specific IL-10 knockout mouse revealed that B cell-derived IL-10 nonredundantly decreases virus-specific CD8(+) T cell responses and plasma cell expansion during murine cytomegalovirus infection and modestly restrains immune activation after challenge with foreign Abs to IgD. In contrast, no role for B cell-derived IL-10 was evident during endotoxemia; however, although B cells dominated lymphoid tissue IL-10 production in this model, myeloid cells were dominant in blood and liver. These data suggest that B cells are an underappreciated source of counter-regulatory IL-10 production in lymphoid tissues, provide a clear rationale for testing the biological role of B cell-derived IL-10 in infectious and inflammatory disease, and underscore the utility of cell type-specific knockouts for mechanistic limning of immune counter-regulation.

Figure 1. Generation of Vert-X (fr. green, 10) mice: Targeting of an IRES-eGFP reporter into the il10 locus

(a) Maps of the il10 locus (top), the targeting construct (middle), and the mutated il10 locus (bottom). A genomic 6.8 kb EcoRI fragment of the wildtype C57BL/6 locus (top), isolated by recombineering techniques from a BAC clone, was mutated by: (i) the introduction of a loxP-flanked (filled triangles) neomycin cassette (Neo), an internal ribosomal entry site (IRES) element, enhanced green fluorescent protein (EGFP), and a polyadenylation signal (pA), all cloned into a HindIII site between the 5^th exon and the endogenous pA of il10; and (ii) the 5′ addition of an HSV-derived thymidine kinase (HSV-tk) cassette. After verification (by PCR and Southern blotting) of successful targeting into C57BL/6 embryonic stem cells, positive clones were injected into blastocysts of C57BL/6 albino females. The floxed Neo was deleted by Cre-mediated recombination using C57BL/6 Zp3 Cre deletor mice (Jackson), generating the targeted il10 locus depicted at bottom. Numbered boxes represent exons. (b) Southern blot analysis of DNA from tail biopsies of the progeny of heterozygous breeding. The novel NsiI site introduced by the targeting construct results in a 3 kb fragment in the correctly targeted allele, in place of the 10 kb fragment of the wildtype allele. +, wildtype; Vert-X, targeted allele.

Figure 2. Fidelity and bi-allelic expression of the targeted locus

(a) Splenic leukocytes from age-and sex-matched Vert-X mice were stimulated in vitro with 2 μM CpG (ODN 1826). 18 h after stimulation, cells were immunostained for CD19 and sorted by flow cytometry into CD19⁺GFP⁺ and CD19⁺GFP^- populations. IL-10 and eGFP mRNA expression was quantified by real-time RT-PCR in purified subsets, and normalized for levels of β-actin mRNA expression. Data represent means ± SE from a single mouse (representative of an experimental of N = 2). (b) Splenic cells from Vert-X mice were stimulated and sorted as above, followed by ELISPOT analysis of IL-10-secreting cell numbers in purified subsets. Data represent means (± SEM) of triplicate samples from a single mouse, representative of an experimental N = 5. Similar data were obtained with CD11c⁺ cells (data not shown). (c,d) Splenic leukocytes from heterozygous and homozygous Vert-X mice were stimulated as above. At the indicated time-points, cells were harvested, stained with antibodies to CD19 and B220, and analyzed by flow cytometry. Splenic leukocytes from wildtype C57BL/6 mice, treated identically, were used to correct for baseline autofluorescence. (c) Percentage of CD19⁺B220⁺ cells that were GFP⁺. (d) Mean fluorescence intensity (MFI) of GFP⁺CD19⁺B220⁺ cells. Data represent means (± SEM) of 3 different mice, representative of data from 2 independent experiments.

Figure 3. Baseline eGFP expression by splenic leukocytes

Splenic leukocytes were harvested from mock- (endotoxin free PBS-) stimulated Vert-X mice, immunostained with the indicated lineage-specific cell surface markers, and GFP expression was quantified by flow cytometry. Of the 1.3% (± 0.1%) of splenic leukocytes GFP⁺ at baseline, approximately 90% were accounted for by the depicted cellular subsets. (a) CD19⁺B220⁺ and CD19⁺B220^low/- cells. (b) CD19⁺CD138⁺ cells. (c) TCR⁺CD4⁺CD25⁺ and TCR⁺CD4⁺CD25^- cells. Data depicted are representative of 7 mice from 2 separate experiments. Similar results were seen with 8 unstimulated mice.

Figure 4. IL-10 competence of plasma cells

Flow cytometric quantification of plasma cell GFP expression 48 h after challenge with LPS (i.v.), 6 d after challenge with goat anti-mouse IgD (i.p.), and 7 d after challenge with MCMV Smith strain (s.c.)— or mock-challenge with endotoxin-free PBS. Data depicted represent the mean percentage (± SEM) of total GFP⁺ cells represented by CD138⁺ splenic (PBS [l], LPS, anti-IgD) or lymph node (PBS [r], MCMV) leukocytes]), in 5-10 mice. In the case of challenge with MCMV K181 strain, CD138⁺ cells represented 57.3 ± 2.66% of total GFP⁺ cells in draining lymph nodes (not shown). **P < 0.005 (Student’s t-test).

Figure 5. B cell-specific deletion of Il10 in Il10^flox/floxCD19-Cre mice

Southern blot analysis of DNA extracted from the indicated, FACS-sorted cell populations and tissues after PstI digest. CD19⁺ B cells and CD3⁺ T cells were sorted from splenocyte suspensions. F4/80⁺ macrophages were sorted from peritoneal lavage fluid. FL or D designate the loxP-flanked or deleted allele, respectively.

Figure 6. Non-redundant counter-regulatory roles for B cell-derived IL-10 during MCMV infection

(a) Restraint of the antigen-specific CD8⁺ T cell response in MCMV infection. Seven days after mice were infected with MCMV (Smith strain; 1 × 10⁵ pfu injected subcutaneously in the foot pad), splenic leukocytes were harvested, restimulated in vitro with immunodominant MCMV peptides, and MCMV-specific CD8⁺ T cells were quantified by flow cytometric quantification of intracellular IFN-γ⁺. The data depict the percentage (± SEM) of CD8⁺IFN-γ⁺ cells, as a percentage of the total number of splenic CD8⁺ T cells (17-21 mice/genotype; 5-21 mice/data point, from 3 separate experiments). (b) Restraint of plasma cell expansion. Flow cytometric quantification of plasma cell numbers in inguinal lymph nodes of mice infected as in “a”. The data depicted show the percentage (± SEM) of CD138⁺ IgM^- B cells. 5-6 mice were studied in a single experiment. ** P< 0.01, * P < 0.05; 2-tailed T-test.

Figure 7. Modest role of B-cell derived IL-10 in restraining immune activation after challenge with foreign antibodies to IgD

Eight d after challenge of wild type, IL-10^-/- and Il10^flox/floxCD19-Cre mice, splenic leukocyte subsets were quantified by flow cytometry. (a) Total splenic leukocytes; (b) B cell subsets. (c) T cell subsets. (d) NK1.1⁺, CD11b⁺ and CD11c⁺ cells. (e) Systemic IL-10 production and (f) goat IgG-specific IgG1 antibody were quantified by ELISA. Data represent the mean percentage (± SEM) of splenic leukocytes in 5 mice. (In the case of the comparison between IL-10^-/- and wild type mice, the data are representative of 3 separate experiments.) *P < 0.05, **P ≤ 0.002, ^‡P = 0.055 (Student’s t-test).

Figure 8. Lack of a role of B-cell derived IL-10 in restraining immune activation after challenge with LPS

Wild type (closed black square), IL-10^-/- (open square), Il10^flox/flox (filled diamond) and Il10^flox/floxCD19-Cre (open diamond) mice were challenged i.p. with 20 mg/kg E. coli LPS. (a) Endotoxicity score (comprised of lethargy, piloerection, ocular discharge and diarrhea (23)) and (b) Survival. Data represent the mean score (± SEM) (a) and mean percentage (b) in 7 WT, 12 IL-10^-/-, 14 Il10^flox/flox and 15 Il10^flox/floxCD19-Cre mice. Due to the early mortality in IL-10^-/- mice, endotoxicity scores are only available for the 6 h time point with these mice.