Stromal Cell-Contact Dependent PI3K and APRIL Induced NF-κB Signaling Prevent Mitochondrial- and ER Stress Induced Death of Memory Plasma Cells

Abstract

The persistence of long-lived memory plasma cells in the bone marrow depends on survival factors available in the bone marrow, which are provided in niches organized by stromal cells. Using an ex vivo system in which we supply the known survival signals, direct cell contact to stromal cells, and the soluble cytokine a proliferation-inducing ligand (APRIL), we have elucidated the critical signaling pathways required for the survival of long-lived plasma cells. Integrin-mediated contact of bone marrow plasma cells with stromal cells activates the phosphatidylinositol 3-kinase (PI3K) signaling pathway, leading to critical inactivation of Forkhead-Box-Protein O1/3 (FoxO1/3) and preventing the activation of mitochondrial stress-associated effector caspases 3 and 7. Accordingly, inhibition of PI3K signaling in vivo ablates bone marrow plasma cells. APRIL signaling, by the nuclear factor κB (NF-κB) pathway, blocks activation of the endoplasmic-reticulum-stress-associated initiator caspase 12. Thus, stromal-cell-contact-induced PI3K and APRIL-induced NF-κB signaling provide the necessary and complementary signals to maintain bone marrow memory plasma cells.

Keywords: APRIL; BCMA; FoxO; IRF4; PI3K/AKT; bone marrow; caspase 12; caspase 3; caspase 7; long-lived memory PCs; stromal cell.

Graphical abstract

Figure 1

Survival of Bone Marrow Memory PCs Is Dependent on Direct Cell Contact with Stromal Cells and the Presence of APRIL (A) Survival of primary murine bone marrow PCs cultured ± ST2 cells and ± APRIL for up to 6 days at 4.2% O₂. Viable plasma cells (CD138⁺⁺/DAPI⁻) were counted by flow cytometry. Median of at least 5 pooled independent experiments with at least n = 14 technical replicates for each group. Statistics: Kruskal-Wallis test. (B) Isolated PCs treated with or without pan-caspase inhibitor when cultured ± ST2 cells and ± APRIL. Viable PCs were counted on day 1 of culture (pooled from two independent experiments with a minimum of n = 7 technical replicates for each group). Statistics: ordinary one-way ANOVA. (C) Survival of PCs in the presence of APRIL on day 1 and day 3, when cultured in transwell or directly contacting ST2 cells (pooled from two independent experiments with n = 4 technical replicates for each group). Statistics: t test. (D) Survival of PCs on day 1 and day 3 treated with specific siRNA directed against ITGB1 and scrambled controls (pooled from three independent experiments with n = 9 technical replicates for each group). Statistics: ordinary one-way ANOVA.

Figure 2

Stromal-Cell-Contact-Induced PI3K Signaling Is Essential for Survival of Memory Bone Marrow PCs Ex Vivo and In Vivo (A–C) Survival of PCs preincubated with different concentrations of the irreversible PI3K-inhibitor Wortmannin (A), directly treated during culture with the inhibitor LY294002 (B), or with combinations of three subunit-specific PI3K inhibitors determined by counting viable CD138⁺⁺/DAPI⁻ PCs by flow cytometry (C) (pooled from two independent experiments with n = 6 technical replicates for each group). Statistics: ordinary one-way ANOVA. (D) Experimental design: C57BL/6J were primed and boosted twice (days 21 and 42) with CGG-NP/IFA and treated with the PI3K-inhibitor Wortmannin on days 90, 92, and 94. On day 95, the mice were analyzed. (E) Representative plot of B220⁻/CD138⁺⁺/CD19⁻ PCs in the bone marrow from control and Wortmannin-treated mice gated on DAPI⁻ viable cells. (F) Absolute cell counts of total bone marrow cells and memory PCs, in the bone marrow in control and Wortmannin-treated mice. Median of two pooled independent experiments with n = 5 biological replicates. Statistics: t test.

Figure 3

Stromal-Cell-Contact-Induced PI3K Signaling Downregulates FoxO1/3 Protein Expression and Is Essential for Survival of Memory Bone Marrow PCs (A and B) FoxO1 (A) and FoxO3 (B) expression of CD138⁺ PCs, as determined by flow cytometry (geometric mean fluorescence intensity, pooled from two independent experiments with n = 8 technical replicates for each group). Statistics: ordinary one-way ANOVA. (C) Survival of PCs treated with siRNA specific for FoxO1/FoxO3 or scrambled control on day 1 and 3 of culture with APRIL alone or ST2 and APRIL counted by flow cytometry. Median of at least three pooled independent experiments with n = 3 technical replicates for each group. Statistics: ordinary one-way ANOVA.

Figure 4

Contact with Stromal Cells Inhibits Activation of Caspase 3 and 7, whereas APRIL Inhibits Activation of Caspase 12 Expression of activated caspase 3 (A), caspase 7 (B), and caspase 12 (C) of CD138⁺ PCs determined by flow cytometry, shown as geometric mean fluorescence intensity, on day 1 in PCs cultured under the indicated conditions (pooled from a minimum of two independent experiments with n = 6–12 technical replicates for each group). Statistics: Kruskal-Wallis test (activated Casp3), ordinary one-way ANOVA (activated Casp7/Casp12).