TSLP or IL-7 provide an IL-7Rα signal that is critical for human B lymphopoiesis

Abstract

Thymic stromal lymphopoietin (TSLP) and IL-7 are cytokines that signal via the IL-7 receptor alpha (IL-7Rα) to exert both overlapping and unique functions during early stages of mouse B-cell development. In human B lymphopoiesis, the requirement for IL-7Rα signaling is controversial and the roles of IL-7 and TSLP are less clear. Here, we evaluated human B-cell production using novel in vitro and xenograft models of human B-cell development that provide selective IL-7 and human TSLP (hTSLP) stimulation. We show that in vitro human B-cell production is almost completely blocked in the absence of IL-7Rα stimulation, and that either TSLP or IL-7 can provide a signal critical for the production and proliferation of human CD19(+) PAX5(+) pro-B cells. Analysis of primary human bone marrow stromal cells shows that they express both IL-7 and TSLP, providing an in vivo source of these cytokines. We further show that the in vivo production of human pro-B cells under the influence of mouse IL-7 in a xenograft scenario is reduced by anti-IL-7 neutralizing antibodies, and that this loss can be restored by hTSLP at physiological levels. These data establish the importance of IL-7Rα mediated signals for normal human B-cell production.

Keywords: Human B lymphopoiesis; IL-7; Pro-B cell; TSLP; Xenograft.

Figure 1. TSLP can replace IL-7 in supporting the in vitro production and proliferation of human CD19+ PAX5+ B cell precursors

(A) Human-only cultures were produced by plating primary human CB CD34+ cells on primary human BM stroma. Selective cytokine stimulation was achieved by supplementing cultures with exogenous IL-7 (5 ng/ml) or TSLP (10 ng/ml) or neutralizing antibodies to TSLP (anti-) or IL-7 (anti-7) at 1 ug/ml and 10 ng/ml respectively, as indicated. Cultures without a particular neutralizing antibody were supplemented, as a control, with non-specific isotype-matched antibodies (ctrl Ig). All culture conditions included IL-3 and Flt-3 ligand at 1 ng/ml. Cultures were maintained for 3 weeks, then harvested and stained for flow cytometry. (B) Dot plots of CD19 vs. PAX5 staining in each culture condition are shown (representative of n=3 independent experiments). PAX5 fluorescence minus one (FMO) control is displayed in inset. (C) Graphed are the relative numbers of CD19+ cells (mean± SEM) generated in vitro under indicated conditions at three weeks (n=14). Relative cell number is defined as the number of CD19+ cells/HSC generated in each condition divided by the sum of CD19+ cells/HSC in all conditions. (D) Representative dot plots of CD19 vs BrdU staining in indicated conditions. BrdU FMO control is displayed in the inset of the Ctrl dot plot. (E) Graphed is the percent BrdU+ cells (mean± SEM) in the total CD19+ population generated under indicated conditions at three weeks (n=8). Additional gating is shown in Fig. S1 in online supporting data. Statistical analysis was performed using one-tailed, paired t-test, *p<0.05, **p<0.001. Each comparison is versus the control condition. Error bars represent mean ± SEM.

Figure 2. IL-7 and TSLP are produced by human BM stroma

(A) RT-PCR, was used to detect TSLP, IL-7 or beta-2 microglobulin (β₂M, control) transcripts in primary human BM stromal cells from different human donors – pediatric (BM #1 and BM #2) and adult (BM #3) or culture medium as a negative control. Each patient sample was assessed in two or more different PCR reactions. (B) Supernatant from confluent BM stromal cell cultures were assessed by ELISA for TSLP and IL-7 protein production. Data are expressed as mean ± SEM of triplicate values for TSLP, and duplicate values for IL-7. Dashed lines (---) represent ELISA threshold of detection.

Figure 3. IL-7 and TSLP increase the in vivo production of human B cell precursors in human-mouse xenografts

(A) Immune deficient NSG mice were engineered to express physiological levels of hTSLP (+T mice) or without hTSLP (−T mice) as described [14]. CB CD34+ cells were injected by tail vein into −T and +T mice. Five weeks later, −T mice and +T mice were treated for two weeks with anti-human/mouse IL-7 antibody or isotype-matched control antibody to generate −T−7 mice (no hTSLP and reduced IL-7), −T+7 mice (no hTSLP and normal IL-7), and +T−7 mice (physiological hTSLP and reduced IL-7). At 7 weeks post-transplant, mice were euthanized and BM harvested and stained for human specific markers to identify hematopoietic subsets (for gating see Fig S1 in online Supporting Information.) Graphed are the absolute numbers of cells in (B) progenitor populations, (C) B cell subsets, and (D) non-B cells in the BM of xenograft mice. Data shown were obtained from two independent transplantations of two different CB donors with a total of four mice per group. Statistical analysis was performed using one-tailed, unpaired t-test, *p<0.05 compared to −T−7 mice. Error bars represent mean ± SEM.