Role of mir-15a/16-1 in early B cell development in a mouse model of chronic lymphocytic leukemia

Abstract

In both human chronic lymphocytic leukemia (CLL) and the New Zealand Black (NZB) murine model of CLL, decreased levels of microRNAs miR-15a/16 play an important role in the disease. Here we investigate the effects of this microRNA on early steps of B cell development and the capacity of miR-15a-deficient hematopoietic stem cells (HSC) and B1 progenitor cells (B1P) to reproduce CLL-like phenotype both in vitro and in vivo. Our results demonstrate that both miR-15a deficient HSC and B1P cells are capable of repopulating irradiated recipients and produce higher numbers of B1 cells than sources with normal miR-15a/16 levels. Furthermore, induced pluripotent stem (iPS) cells derived for the first time from NZB mice, provided insights into the B cell differentiation roadblock inherent in this strain. In addition, exogenously delivered miR-15a into the NZB derived B cell line provided valuable clues into novel targets such as Mmp10 and Mt2. Our data supports the hypothesis that miR-15a/16 deficient stem cells and B1Ps experience a maturation blockage, which contributes to B1 cells bias in development. This work will help understand the role of miR-15a in early events of CLL and points to B1P cells as potential cells of origin for this incurable disease.

Keywords: B1 progenitors; cancer stem cells; chronic lymphocytic leukemia; induced pluripotent stem cells; microRNAs.

Figure 1. Comparison of splenic phenotype in control DBA, NZB and DBA congenic (DBA^−/−) mice

A. TaqMan real-time PCR quantification of miR-15a and miR-150 levels in sorted B1a subpopulations from spleen; N = 3, columns represent mean RQs, bars are SEMs; B. Flow cytometry staining for B1 CD5+, B220^dull cells (gated on CD3-CD19+). This population was sorted and used for analysis in panels C and D. C. Quantification of B1 (left) and B2 (right) cells in DBA, NZB and DBA−/− congenic mice spleen; n = 3. D. TaqMan PCR levels of miR-15a and miR-150 in sorted B2 subpopulation from spleen; RQ is relative quantification normalized to snRNA U6 expression; E. Flow cytometry quantitative analysis of immature IgM+IgDlow B cells; columns are means, bars are SEMs; n≥3. F. Flow cytometry quantitative analysis of CD19-CD3+ T cells in spleen, age = 12 months; columns are means, bars are SEMs; n≥3. G. Representative pictures of single molecule RNA-FISH with Pax5, Dleu2 and PU.1 probes; H. Quantitative analysis of Pax5, PU.1 and Dleu2 transcripts in splenocytes derived from DBA, NZB and DBA−/− mice by RNA-FISH (only Pax5+ B cells were counted). At least 50 individual cells from each animal were counted from at least 3 (n = 3) animals of the same strain and analyzed. Columns represent average mRNA molecules per cell and bars are SEMs. Asterisks represent statistical difference (p < 0.05).

Figure 2. B1 progenitors (B1P) and colony forming units (CFU) analysis

A. Pre-B CFU assay for bone marrow cells derived from the indicated mouse strains; B. CFU-GM assay of bone marrow cells derived from the indicated mouse strains. Columns are means and bars are SEMs. Asterisks represent statistical difference (p < 0.05). C. Flow cytometry quantitative analysis of B1 progenitors in the spleen of the indicated mouse strains (age = 12 months, n = 3); columns represent means and bars are SEMs. D. Flow cytometry quantitative analysis of B1 progenitors in the bone marrow of the indicated mouse strains (age = 12 months, n = 3); no statistical difference observed. B1 progenitors were defined as Lin⁻AA4.1⁺CD19⁺B220^neg/dull population.

Figure 3. In vitro differentiation of bone marrow progenitors from miR-15a-deficient mice

A. The levels of miR-15a and miR-150 in LSK cells co cultured with OP9 stromal cells at day 11 measured by TaqMan qPCR; B. Representative flow cytometry analysis of DBA, NZB and DBA−/− LSK derived immature (top box) and mature (bottom box) B cell progenitors at day 11 of OP9 co-culture. C. Quantitative flow cytometry analysis of mature AA4.1⁻B220⁺ (right) and immature (left) B cells progenitors. N = 3, columns represent means and bars are SEMs; Transcript expression analysis of D. PU.1, E. Pax5, and F. cMyb measured by qPCR assay are shown. MicroRNA levels were assessed using TaqMan assays with custom probes for has-miR-15a and has-miR-150. Protein-coding genes transcripts measurements were done by qPCR with the use of SYBR green master mix and specific primers; n = 3, columns represent means and bars are SEMs; G. The ratio of PU.1 to Pax5 gene expression in LSK cells at day 11 of co culture with OP9 cells. Asterisks represent statistically significant difference (p < 0.05).

Figure 4. Repopulation potential of DBA, NZB and DBA^−/− congenic mice derived HSC and B1 progenitor cells in immunodeficient NSG recipients at 48 days post injection

A. Mir-15a, B. cMyb and C. PU.1 expression levels in DBA, NZB and DBA^−/− HSC and B1Ps. MicroRNA levels were assessed using TaqMan assays with custom probes for hsa-miR-15a. Quantitative PCR for protein coding genes was performed with the use of SYBR green master mix and specific primers; n = 3, columns represent means and bars are SEMs. Flow cytometry analysis of D. bone marrow, E. spleen and F. peritoneal cavity cells from NSG recipients. Recipient NSG mice were analyzed 48 days post injection for the presence of donor cells (H2D^d+) which also expressed markers for B1 progenitors as a percentage of the total donor cell population in the bone marrow and spleen (mean ± SEM, n ≥ 3, p < 0.05). Cells were gated on donor H2D^d+IgM⁺CD19⁺ mature B cells and analyzed for CD5 and B220 expression. The B1 population (CD5⁺B220^dull) is indicated in the circle and the B2 population (CD5⁻B220^hi) is indicated in the box. Quantitative analysis of B1 cells is shown as mean ± SEM on the right side of each panel. Open bars represent DBA, black bars are NZB and grey bars are DBA^−/− source. Asterisks indicate significant difference from control DBA donors.

Figure 5. Quantitative PCR analysis of in vitro differentiated NZB ES/iPS cells

The expression levels of A. Pax5, B. PU.1, C. cMyb, D. miR-15a, E. miR-150 in ES (left panels) and iPS (right panels) cells during in vitro differentiation. F. The calculated ratios of PU.1 to Pax5 expression in ES (left) and iPS (right) cells. Relative quantification (RQ) and log10 RQ are indicated on the axes; n = 3, p < 0.05 is considered significantly different from the previous time point and marked with an asterisk. ES and iPS cells were pre-differentiated as embryoid bodies for 7 days and plated in semisolid pre-B methocult medium for 12 days. qPCR assays were performed using SYBR green master mix and custom primers for protein coding gene transcripts. First strand cDNA synthesis was performed by using SuperScript First-Strand Synthesis System (Invitrogen) using oligo-d(T)n. MicroRNA measurements were done using TaqMan assays for mmu-miR-15a and mmu-miR-150. TaqMan assay for U6 was used as a housekeeping control.

Figure 6. Gene expression analysis of NZB B cell line (LNC) transduced with miR-15a lentiviral construct

Four cell lines including the non-NZB B cell line A20, the NZB cell line, LNC and two sublines of LNC which had been transduced with a lentivirus containing GFP (LNC-GFP) or LNC transduced with a lentivirus containing both GFP and miR-15a/16 (LNC-miR-GFP) were studied:. A. Quantitative PCR TaqMan assay for mir-15a levels in LNC cells lentivirally transduced with wild type mir-15a construct in comparison to non-transduced and non-NZB cells; B. Representative RNA FISH microscopy images of individual transcripts. The slides were imaged using probes specific to Pax5, PU.1, Dleu2 and IL-10 RNA molecules; C. Quantitative analysis of RNA-FISH data to obtain average number of specific mRNAs per cell. At least 100 cells were analyzed; D. Calculated PU.1 to Pax5 expression levels ratio. Each cell was simultaneously probed for both PU.1 and Pax5 and the ratio was determined; n = 3, asterisks represent statistically significant difference (p < 0.05).