The cytokine midkine and its receptor RPTPζ regulate B cell survival in a pathway induced by CD74

Abstract

Lasting B cell persistence depends on survival signals that are transduced by cell surface receptors. In this study, we describe a novel biological mechanism essential for survival and homeostasis of normal peripheral mature B cells and chronic lymphocytic leukemia cells, regulated by the heparin-binding cytokine, midkine (MK), and its proteoglycan receptor, the receptor-type tyrosine phosphatase ζ (RPTPζ). We demonstrate that MK initiates a signaling cascade leading to B cell survival by binding to RPTPζ. In mice lacking PTPRZ, the proportion and number of the mature B cell population are reduced. Our results emphasize a unique and critical function for MK signaling in the previously described MIF/CD74-induced survival pathway. Stimulation of CD74 with MIF leads to c-Met activation, resulting in elevation of MK expression in both normal mouse splenic B and chronic lymphocytic leukemia cells. Our results indicate that MK and RPTPζ are important regulators of the B cell repertoire. These findings could pave the way toward understanding the mechanisms shaping B cell survival and suggest novel therapeutic strategies based on the blockade of the MK/RPTPζ-dependent survival pathway.

Figure 1. MIF elevates MK expression in B cells in a CD74-dependent manner

(A–C) B cells derived from C57BL/6 (A–B) or CD74 deficient (C) mice were incubated in the presence or absence of MIF (600 ng/ml) for 8h. Total RNA was then isolated. (A) RT-PCR using primers for MK or HPRT, was performed. The intensity of the MK band after each treatment was normalized by dividing by the measured intensity of the HPRT band from the same treatment. The fold activation ratio in the absence of any treatment was normalized to 1, and the ratio for each treatment was calculated as the intensity of the treatment sample relative to 1. The results presented are representative of at least six different experiments. (B–C) Quantitative Real Time PCR was performed using primers for MK and β-actin. β-actin levels were used to normalize samples for calculation of the relative expression levels of MK. Results are expressed as fold change in MK expression in stimulated cells compared to non-stimulated cells, which was defined as 1. The results presented are representative of at least five different experiments. (D) B220+ cells derived from C57BL/6 and CD74^−/−mice were incubated in the presence or absence of MIF (600 ng/ml) for 16 hr, and intracellular MK protein levels were analyzed by intracellular staining as described in Materials and Methods. Grey line- isotype control; Black line- MK staining. The graphs show an average of five independent experiments. (E) Control mice were injected with MIF (1.2 μg) or PBS. After 24 h, mice were sacrificed, splenic B cells were purified, and intracellular MK protein levels were analyzed by intracellular staining as described in Materials and Methods. Graph shows an average of three independent experiments. Grey line- isotype control; Black line- MK staining.

Figure 2. Binding of HGF to c-Met induces MK expression in B cells

(A) B cells derived from C57BL/6 and CD74 deficient mice were incubated in the presence or absence of HGF (20 ng/ml) for 8 hr, and quantitative Real Time PCR was performed using primers for MK and β-actin. β-actin levels were used to normalize samples for calculation of the relative MK expression levels. Results are expressed as fold change in MK expression in stimulated cells compared to non-stimulated cells, which was defined as 1. The results presented are representative of at least five different experiments. (B–C) B cells derived from C57BL/6 or CD74 deficient mice were incubated in the presence or absence of HGF (20 ng/ml) (B), or the c-Met inhibitor, PHA-665752 (0.3 ng/ml) (C) for 16hr, and MK intracellular protein levels were analyzed by intracellular staining as described in Materials and Methods. Grey line- isotype control; black line- MK staining. The graphs show an average of five (B) or three (C) independent experiments. (D) B cells derived from C57BL/6 mice were incubated with or without MIF (600 ng/ml) in the presence of anti-HGF (5 μg/ml) or isotype control (5 μg/ml) antibodies for 16h. MK protein levels were analyzed by intracellular staining as described in Materials and Methods. Grey line- isotype control; Black line- MK staining. The graph summarizes the results of three independent experiments.

Figure 3. MK induced cascade in B cells

B cells derived from C57BL/6 mice were incubated in the presence or absence of MK (100 ng/ml) for various periods. Immediately after stimulation, cells were washed and fast frozen in liquid N₂. The cells were lysed as described in Materials and Methods (A) Lysates were separated on 10% (w/v) SDS-PAGE, and proteins were blotted with anti-p-Akt or anti-Tubulin antibodies. The results presented are representative of at least four different experiments. (B) An aliquot from the lysates was reserved for total Syk analysis. Phosphorylated proteins from the remaining lysate were immunoprecipitated (IP) with an anti-Tyr(P) antibody. Immunoprecipitates and total lysate proteins were separated on 10% (w/v) SDS-PAGE and blotted with an anti-Syk antibody. The results presented are representative of at least three different experiments. (C) RNA was purified from control B cells after stimulation with PBS or MK (100ng/ml) for 6h. Quantitative Real time PCR was performed using primers for Bcl-2 and β-actin. β-actin levels were used to normalize samples for calculation of the relative expression levels of Bcl-2. Results are expressed as a fold change in MK expression in stimulated cells compared to non-stimulated cells, which was defined as 1. Results shown represent an average of at least eight separate experiments. (D) B220+ B cells derived from C57BL/6 mice were incubated in the presence or absence of MK (100ng/ml) for 6h. Cells were lysed, and levels of Bcl-2 and Tubulin were analyzed by western blot analysis. The results presented are representative of at least four different experiments.

Figure 4. MK stimulation induces cell survival in vitro

(A) B220+ B cells derived from C57BL/6 or CD74 deficient mice were incubated in the presence or absence of MK (100ng/ml) for 16h. Caspase activity was analyzed by Magic Red apoptosis detection kit. The graph shows the average of ten independent experiments. (B) B220⁺ B cells derived from control mice were incubated in the presence or absence of MK (100 ng/ml) for 18 hr. Cell death was then analyzed by Annexin V and PI staining of intact cells. The graph summarizes the average apoptotic population in four independent experiments. (C–D) B220+ B cells derived from C57BL/6 were incubated in the presence or absence of MK (100 ng/ml) with or without the c-Met inhibitor, PHA-665752 (0.3 ng/ml) (C), or with anti-HGF (5 μg/ml) or an isotype control (5 μg/ml) antibody (D) for 16 hr, and caspase activity was analyzed by Magic Red. The graphs show an average of five (C), or three (D) independent experiments.

Figure 5. The MK-induced survival pathway regulates survival of the mature B population in vivo

C57BL/6 or CD74 deficient mice were injected daily with MK (400 ng) or PBS for 2 days. (A) Cell survival was analyzed by Annexin V and PI staining of intact B220⁺ cells. The graphs show the average apoptotic population in five independent experiments. (B) B cell subpopulations were analyzed for CD21, CD24, and CD23 and B220⁺ expression. The graphs show the number of mature B cells in seven independent experiments.

Figure 6. RPTPζ is the receptor for MK in B cells

(A–B) B cells derived from control-129 SvEv mice and RPTPζ deficient mice were purified. (A) RT-PCR, was performed using primers for RPTPζ or HPRT. (B) RPTPζ expression was analyzed by western blot analysis. (C–E) B220+ B cells derived from control-129 SvEv or RPTPζ deficient mice were incubated in the presence or absence of MK (100 ng/ml) (C), MIF (600 ng/ml) (D), or HGF (20 ng/ml) (E) for 16h. Caspase activity was measured by Magic Red apoptosis detection kit. The graphs show an average of at least three independent experiments.

Figure 7. RPTPζ is involved in shaping the mature B cell repertoire

(A–B) Splenic cells derived from control-129 SvEv mice and RPTPζ deficient mice were purified. (A) Dot plots showing CD21 and CD24 expression on B220+ cells derived from 129SvEv and from RPTPζ deficient mice. Graphs show the average percent and numbers of the mature population in twelve mice. (B) Dot plots showing IgD and IgM expression on B220+ cells derived from the spleen of 129SvEv and RPTPζ deficient mice. (C) Dot plots showing IgD and IgM expression on B220+ cells derived from bone marrow of 129SvEv and RPTPζ deficient mice. The graphs show the average of the percent and numbers of the mature population in seven mice. (D) Dot plots showing IgD and IgM expression on B220+ cells derived from lymph nodes of 129SvEv and RPTPζ deficient mice. The graph summarizes the average number of mature B cells in lymph node derived from 129SvEv and RPTPζ deficient mice, with six mice in each group (E) Dot plots showing CD5 and MAC-1 expression on IgM^high B220+ cells derived from the peritoneal cavity of 129SvEv and RPTPζ deficient mice.

Figure 8. MK regulates survival of chronic lymphocytic leukemia cells

(A) RPTPζ expression was analyzed on human primary CD19+ B cells by FACS analysis. (B) Human primary B cells were incubated with or without MIF (600 ng/ml). After 24 h, their conditioned medium was collected and analyzed by ELISA for human MK protein levels. (C) MK levels in serum derived from healthy early and advanced CLL patients. (D) B cells derived from early and advanced CLL patients were purified. Cells were incubated in the presence or absence of MIF (100 ng/ml). Quantitative Real time PCR was performed using primers for MK and Rp2. Rp2 levels were used to normalize samples for calculation of the relative expression levels of MK. Results are expressed as fold change in MK expression in stimulated compared to non-stimulated cells, which was defined as 1. Results shown in each graph summarize at least four CLL patients. (E–F) CLL cells were incubated in the presence or absence of MK (100ng/ml) (E) Quantitative Real time PCR was performed using primers for Bcl2 and Rp2. Rp2 levels were used to normalize samples for calculation of the relative expression levels of Bcl2. Results are expressed as fold change in Bcl-2 expression in stimulated cells compared to non-stimulated cells, which was defined as 1. Results shown in each graph summarize three different CLL patients. (F) Cells were lysed, and Bcl-2 and Tubulin levels were measured by western blot analysis. The results presented are representative of four CLL patients. (G) Caspase activity was analyzed by Magic Red apoptosis detection kit. The graph shows an average of five different patients. (H) CLL cells were incubated with or without MIF (600 ng/ml) in the presence of anti-RPTPζ (2 μg/ml) or isotype control (2 μg/ml) antibodies for 24h. Caspase activity was analyzed by Magic Red apoptosis detection kit. Graph summarizes the results of three independent experiments.