Phc2 controls hematopoietic stem and progenitor cell mobilization from bone marrow by repressing Vcam1 expression

The timely mobilization of hematopoietic stem and progenitor cells (HSPCs) is essential for maintaining hematopoietic and tissue leukocyte homeostasis. Understanding how HSPCs migrate between bone marrow (BM) and peripheral tissues is of great significance in the clinical setting, where therapeutic strategies for modulating their migration capacity determine the clinical outcome. Here, we identify an epigenetic regulator, Phc2, as a critical modulator of HSPC trafficking. The genetic ablation of Phc2 in mice causes a severe defect in HSPC mobilization through the derepression of Vcam1 in bone marrow stromal cells (BMSCs), ultimately leading to a systemic immunodeficiency. Moreover, the pharmacological inhibition of VCAM-1 in Phc2-deficient mice reverses the symptoms. We further determine that Phc2-dependent Vcam1 repression in BMSCs is mediated by the epigenetic regulation of H3K27me3 and H2AK119ub. Together, our data demonstrate a cell-extrinsic role for Phc2 in controlling the mobilization of HSPCs by finely tuning their bone marrow niche.

Serial competitive BM repopulation assays were performed using WT and KO mice. For primary transplantation, LSK cells (5 × 10 4 ) from WT and KO mice (CD45.2) were mixed with 5 × 10 4 competitor BM cells (CD45.1), and intravenously injected into lethally irradiated WT CD45.1 recipient mice. n = 5 per group. Secondary transplantation was performed at 12 weeks after primary engraftment. LSK cells (1 × 10 5 ) from primary transplants were intravenously injected into lethally irradiated WT CD45.1 recipient mice. n = 5 per group. a Schematic representation of serial competitive transplantation strategy. b Percentage of donor chimerism in BM, PB, thymus (Thy), and spleen (SP) of primary transplantation recipients (left) and percentage of donor-derived T cells, B cells, myeloid cells in SP of primary transplantation recipients (right). c Percentage of donor chimerism in BM, PB, Thy, and SP of secondary transplantation recipients (left) and percentage of donorderived T cells, B cells, myeloid cells in SP of secondary transplantation recipients (right). Statistical significance was assessed by two-tailed Student's t test. All data are presented as means ± SEM. Source data are provided as a Source Data File. and SP (lower) of recipients at 16 h after CFSE-labeled LSK cell (WT or KO) injection into recipient mice (WT or KO). n = 5. Statistical significance was assessed by ANOVA analysis. Mean values not sharing the same superscript letter ( a-f ) differ significantly at P < 0.05. All data presented as the means ± SEM. Source data are provided as a Source Data File. Figure 6. Phc2 deficiency leads more firm adhesion between HSPCs and BMSCs in BM niches. Endothelial cells (Endo), osteoblasts (OB) or mesenchymal stem cells (MSC) in WT or KO BM were isolated by FACS sorting. Then, trans-stromal migration assays were performed to measure the capacity of WT LSK cells to transmigrate across each subpopulation of WT or KO BMSCs. a Relative ratio of migrated WT LSK cells through each subpopulation of WT or KO BMSCs. n = 9. b Relative ratio of adhered WT LSK cells to each subpopulation of WT or KO BMSCs. n = 3. Statistical significance was assessed by two-tailed Student's t test. *P < 0.05; **P < 0.01. All data are presented as means ± SEM. Source data are provided as a Source Data File.   Figure 10. Relative fold change of Vcam1 mRNA expression in LPS-stimulated peritoneal macrophages from WT or KO mice. WT or KO peritoneal macrophages were harvested on day 3 after intraperitoneal injection of thioglycolate (3%). Isolated macrophages (2 x 10 6 cells well -1 ) were seeded and stimulated with LPS (1 g ml -1 ) in 6 well plate. Three h after stimulation, total RNA of each sample was isolated to monitor Vcam1 mRNA expression by real-time qRT-PCR analysis. n = 3. The level of Vcam1 mRNA transcripts was normalized with that of Gapdh. Relative fold change of each Vcam1 mRNA expression from WT or KO macrophages was calculated relative to basal level of that in unstimulated WT or KO macrophages, respectively. Statistical significance was assessed by two-tailed Student's t test. *P < 0.05. All data are presented as means ± SEM. Source data are provided as a Source Data File.

Supplementary
Supplementary Figure 11. GSK126 inhibits cellular H3K27me3 in a dose dependent manner in BMSCs. BMSCs (OP9 cells) were treated with various concentrations of GSK126 for 48 h. After treatment, cell lysates were subjected to immunoblot analyses using anti-H3K27me3 and anti-histone H3 antibody (upper). To quantify inhibitory effect of GSK126 on H3K27me3, immunoreactive bands were normalized to total H3 levels (lower). n = 3. Statistical significance was assessed by two-tailed Student's t test. *P < 0.05; **P < 0.01. All data are presented as means ± SEM. Source data are provided as a Source Data File.
Supplementary Figure 12. In vivo administration of GSK126 into WT mice mimics the phenotype of KO mice. WT mice were scarified on 16 h after intraperitoneal injection of GSK126 (50 mg kg -1 or 100 mg kg -1 ) to monitor VCAM-1 expression in BMSCs and perform trans-stromal migration and adhesion assays. a, b Relative ratio of VCAM-1 expression in BMSCs isolated from GSK126 injected mice was analyzed by immunoblot (a) and flow cytometry (b). n = 3. EC, endothelial cells; OB, osteoblasts; MSC, mesenchymal stem cells. c Relative ratio of migrated LSK cells through BMSCs isolated from GSK126 injected mice. n = 9. d Relative ratio of adhered LSK to BMSCs isolated from GSK126 injected mice. n = 5. Statistical significance was assessed by two-tailed Student's t test. *P < 0.05; **P < 0.01. All data are presented as means ± SEM. Source data are provided as a Source Data File.

a b c d
Supplementary Figure 13. In vivo homing assay using mice administered with anti-VCAM-1 antibody. CFSE-labeled WT LSK cells were intravenously injected into lethally irradiated WT or KO recipient mice. Anti-VCAM-1 Ab or its isotype control Ab (2 mg kg -1 ) was intravenously injected into recipient mice 1 h before adoptive cell transfer. Sixteen h after adoptive cell transfer, CFU assays were performed to measure the frequencies of donor-derived clonogenic progenitors homing to the BM (femur, upper), PB (middle) and SP (lower) of recipients. n = 5. WT, WT recipient mice; KO, KO recipient mice. Isotype, isotype antibody treatment; anti-VCAM-1, anti-VCAM-1 antibody treatment. Statistical significance was assessed by ANOVA analysis. Mean values not sharing the same superscript letter ( a-e ) differ significantly at p < 0.05. All data presented as the means ± SEM. Source data are provided as a Source Data File. Figure 14. Absolute number of LSK cells in the BM, PB, and spleen (SP) of WT and KO mice on day 5 after G-CSF treatment. Anti-VCAM-1 Ab or an isotype control Ab (2 mg kg -1 day -1 ) was daily administered intravenously into WT and KO recipient mice. n = 5. "-", treatment with an isotype control Ab; "+", treatment with anti-VCAM-1 Ab. Statistical significance was assessed by two-tailed Student's t test. **P < 0.01. All data are presented as means ± SEM. Source data are provided as a Source Data File.

Supplementary
Supplementary Figure 15. Absolute number of LSK cells in the BM, PB, and spleen (SP) of WT and KO mice on day 16 after 5-FU treatment. Anti-VCAM-1 Ab or an isotype control Ab (2 mg kg -1 day -1 ) was daily administered intravenously into WT and KO recipient mice. n = 5. "-", treatment with an isotype control Ab; "+", treatment with anti-VCAM-1 Ab. Statistical significance was assessed by twotailed Student's t test. **P < 0.01. All data are presented as means ± SEM. Source data are provided as a Source Data File.
Supplementary Figure 16. Administration of anti-VCAM-1 Ab reverses the defect involving G-CSF and AMD3100-induced HSPC mobilization in KO mice. G-CSF and AMD3100-induced HSPC mobilization assay with anti-VCAM-1 Ab is described in "Methods" section. n = 3 per group. a WBC counts from experimental mice administered with G-CSF, AMD3100 and/or anti-VCAM-1 Ab. b Representative image of spleens (left) and absolute number of splenocytes (right) from experimental mice administered with G-CSF, AMD3100 and/or anti-VCAM-1 Ab. c CFU-C in the BM (femur, left), PB (middle), and spleen (SP, right) from experimental mice administered with G-CSF, AMD3100 and/or Ab. "-", treatment with vehicle or isotype control Ab; "+", treatment with G-CSF or anti-VCAM-1 Ab. Statistical significance was assessed by two-tailed Student's t test. **P < 0.01. All data are presented as means ± SEM. Source data are provided as a Source Data File.

a b c
Supplementary Figure 17 Supplementary Figure 17. G-CSF induced HSPC mobilization assay in BM chimeric mice. LSK cells (1 × 10 5 ) from WT mice (CD45.1) were injected intravenously into lethally irradiated WT or KO CD45.2 recipient mice. Twelve weeks later, G-CSF or vehicle was administered by daily subcutaneous injection to chimeric mice at a dose of 250 g kg -1 day -1 for 5 days. To block the VLA-4 and VCAM-1 interaction, anti-VCAM-1Ab or the respective isotype control (2 mg kg -1 day -1 ) was administered intravenously every day. n = 5 per group. a Schematic representation of LSK transplantation and G-CSF induced mobilization assay. b Comparison of WBC counts between BM chimeric mice on day 5 after G-CSF and/or anti-VCAM1 Ab treatment. c Representative image of spleens from BM chimeric mice (left) and absolute numbers of splenocytes for BM chimeric mice (right) on day 5 after G-CSF and/or anti-VCAM1 Ab treatment. d CFU-C in the BM (femur, left), PB (middle), and spleen (SP, right) of BM chimeric mice on day 5 after G-CSF and anti-VCAM1 Ab treatment. WT, WT recipient mice; KO, KO recipient mice. "-", treatment with vehicle or isotype control Ab; "+", treatment with G-CSF or anti-VCAM-1 Ab. Statistical significance was assessed by two-tailed Student's t test. **P < 0.01. All data are presented as means ± SEM. Source data are provided as a Source Data File.   Gating strategy for LSK cell sorting. Lineage positive cells in BM were depleted using a lineage-specific Ab cocktail coupled with anti-biotin microbeads by MACS cell separation system. After depletion, lineage negative cells were stained with Sca-1 and c-kit antibodies. Double positive (Sca-1 + c-kit + ) cells were then isolated using FACSAria Fusion Cell Sorter.