The phosphatase of regenerating liver (PRL) family of phosphatases, consisting of PRL1, PRL2, and PRL3, represents an intriguing group of proteins being validated as biomarkers and therapeutic targets in human cancer. We have been investigating the role of PRL2 in normal / malignant hematopoiesis and found that PRL2 is important for HSC self-renewal (Kobayashi et al., Stem Cells, 2014). The receptor tyrosine kinase KIT can balance quiescence for HSC maintenance and proliferation for progeny supply. The defects seen in the PRL2-deficient hematopoietic and testis cells recapitulate the phenotype of c-Kit mutant mice, suggesting that the SCF/KIT signaling may be impaired in the absence of PRL2 (Kobayashi et al., Stem Cells, 2014; Dong et al., JBC, 2013). Given that KIT also plays critical role in maintaining postnatal T-lymphopoiesis in thymus, we hypothesized that PRL2 is important for T cell development.
Here we report that loss of PRL2 impairs T-lymphopoiesis both in vitro and in vivo. PRL2 deficiency resulted in marked reduction of splenocyte and thymocyte counts compared to wild type (WT) mice. While we observed modest increase in the frequency of early T cell progenitor (ETP), DN2, and DN3 cells in PRL2 deficient thymus, T-cell reconstitution was dramatically decreased after HSC transplantation. T-cell number in the peripheral blood (PB) of recipient mice repopulated with PRL2-null HSCs was 30 times less than that of the WT HSCs (WT: 2288.6±579.8/µl vs PRL2 null: 69.5±22.1/µl, p<0.00001). Although the frequency of donor-derived thymocytes in recipient thymus was 91±6.1% in WT, PRL2 null HSCs contributed only 7.1±4.9% (p<0.00001) in the recipient thymus. By detailed fractionation, surprisingly, chimerism in ETP was comparable between WT and PRL2 null cells (WT: 91.8±10.1% vs PRL2 null: 59.6±13.5%, p<0.01). Importantly, the chimerism of PRL2-null thymocytes fell down to 10% in gated DN2, whereas WT HSCs consistently contributed around 90%, suggesting that the DN1-to-DN2 transition requires PRL2. Next, we evaluated the in vitro T-cell generation by utilizing the Delta-Like1 (DLL1) expressing OP9 (DL-OP9) stromal cells. While wild type KSLs produced massive amount of T-cells (fold increase: 33,000±3371) 22 days following plating onto the DL-OP9, PRL2 null KSLs only generated limited amount of T-cells (fold increase: 1765±665, p<0.0001), demonstrating that PRL2 is important for T-cell proliferation. We also monitored the generation of ETPs from KSLs in DL-OP9 cultures and observed significant expansion of ETPs derived from WT KSLs compared to that of the PRL2 null KSLs (fold increase: 183.8±14.4 vs 12.5±4.3, p<0.001). However, when sorted DN3 cells from WT and PRL2 KO thymus were plated onto DL-OP9, we saw similar increase in cell expansion, suggesting PRL2 regulate early T-cell development.
WhilePRL2 is a dual specificity protein phosphatase, its substrates are unknown. To identifyPRL2 substrates in hematopoietic cells, we performed a protein phosphatase substrate trap assay. We utilized a GST-tagged PRL2/CS-DA mutant, in which the catalyticsite cysteine was mutated to serine, so that PRL2 binds to its substrates better, but is unable todephosphorylate them. We found that the mutant PRL2/CS-DA showed enhanced association with KIT than WT PRL2 in Kasumi-1 cells, suggesting that KIT is a potential PRL2 substrate. The PRL2 and KIT interaction was further confirmed by the Immunoprecipitation (IP) assay in 293T cells expressing KIT. We also detected the association of PRL2 with SHP2, CBL and PLC-g in Kasumi-1 cells, which are important regulators of KIT activation and stability. Moreover, PRL2 KO hematopoietic progenitor cells show decreased KIT phosphorylation at tyrosine 703 following SCF stimulation, suggesting that PRL2 may modulate KIT activation in these cells. To evaluate the impact of SCF signal strength on T-cell proliferation, we cultured sorted lympho-primed multipotent progenitor cells (LMPPs) from WT and KO mice onto DLL-Fc coated plates with graded doses of SCF (0.2, 1, 5, 25 ng/ml). The total number of cells generated from SCF treated WT LMPPs was significantly higher than that of the KO LMPPs in a dosage dependent manner. KO exhibited 6 times less sensitive to SCF than WT, indicating that PRL2 fine-tunes SCF signal intensity in early T-cell.
Taken together, we have identified a critical role for PRL2 in T-cell proliferation and maintenance through fine-tuning SCF/KIT signaling.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.