Additional supporting information available online.

STEM_52_sm_suppinfofigure1.tif164KSupporting Information Figure S1. Ruling out the possible contribution of non-hematopoietic populations in the 5-FU treated bone marrow cells. (A) Flowcytometric profiles for non-hematopoietic (CD45NegTer-119Neg) populations in the 5-FU treated bone marrows. Shown are the profiles obtained from two independent mice. (B) Comparisons for gene expression between the unsorted and sorted hematopoietic cells. Whole unsorted 5-FU bone marrow cells (whole) and 5-FU bone marrow cells sorted for CD45(+) cells (sorted) were compared for gene expression of each indicated gene.
STEM_52_sm_suppinfofigure2.tif57KSupporting Information Figure S2. Retroviral vector encoding stable form β-catenin and functional verification. (A) Schematic illustration of parental retroviral vector (MPG) encoding a stable form of β-catenin tagged with HA (β-catenin*). (B) Transactivation of the TCF/LEF binding sites by retroviral construct. 500 ng each of retroviral vectors were transfected into 293 T cells along with 1 ug of TOPFlash (containing 8 TCF/LEF binding sites) or FOPFlash (containing mutant binding sites) reporters. The ratio of transactivation between the TOPFlash and FOPFlash reporters was calculated 24 hrs after transfection (n=3).
STEM_52_sm_suppinfofigure3.tif268KSupporting Information Figure S3. Phenotypic and functional characterization of cultured mesenchymal stromal cells. (A) Ability to support HSC engraftment was examined by co-culture of 5-FU BMCs with established mesenchymal stroma. 5-FU BMCs were cultured in stroma-free condition or on mesenchymal stroma (MSC-stroma) for 5 days and transplanted into irradiated recipients (1×105 input cells per mouse). Shown are the mean ± SEM engraftment levels of donor-derived cells in the recipient at 16 weeks after transplantation (n=3). Hematopoietic engraftment was not detected in mice transplanted with MSC alone in the same experiment. (B) Reconstitution of stroma with in-vitro established MSCs. MSCs transduced with MPG were sorted (GFP+) and injected into irradiated recipient mice by intra-femoral injection (1×105 cells per mouse). Control mice were injected with PBS. Eight weeks after injection, mice bone marrows were harvested and cultured for stromal cells. Shown are representative flow cytometry plots of harvested bone marrow cells after a 2nd passage of adherent cells indicating reconstitution of stroma by injected stromal cells (GFP+CD45-). (C) Responsiveness of cultured stromal cells to Wnt/β-catenin signal. MSCs were stimulated in the diluted conditioned media for 3 hrs and induction and nuclear localization of β-catenin were examined. Shown are representative images (400×) of the stimulated cells immunostained using antibody against total form β-catenin and counterstained for the nucleus with Hoechst 33342. (D) Induction of β-catenin in MSCs by Wnt 3a-CM as determined in a Western blot using antibody against total form β-catenin.
STEM_52_sm_suppinfofigure4.tif617KSupporting Information Figure S4. Characterization of β-catenin/stroma in comparison to MPG/stroma. (A) Surface phenotypes of mesenchymal stromal cells transduced with MPG (MPG) or β-catenin (β-cat). Flow cytometry plots for each surface markers indicated are shown. (B, C) Comparison of the osteogenic and adipogenic differentiation potential of the transduced cells. Engineered stromal cells were subjected to chemical induction for osteogenic and adipogenic differentiation and examined by Alizarin-red staining (osteogenic) and Oil-red staining (adipogenic). Shown are the representative images of osteogenic (B) and adipogenic (C) differentiation of the stromal cells at low magnification (40×). (D-G) Comparison of the osteoblastic composition during ex-vivo culture of transduced stromal cells. (D) RT-PCR analysis of the expression of osteogenic genes in the stromal cells. OP represents osteopontin, ALP, alkaline phosphatase, OC, osteocalcin. (E) Immunohistochemical staining for osteopontin in transduced stromal cells. (F) Comparisons of ALP activity in transduced stromal cells as assessed by a p-nitrophenyl phosphate liquid substrate system normalized against total protein concentration. (G) Expression levels of N-cadherin in transduced stromal cells (MPG or β-catenin). Cells were stained with antibody against N-cadherin by intracellular staining. Shown are the flow cytometry plots for isotype control (white) and N-cadherin (gray).
STEM_52_sm_suppinfofigure5.tif52KSupporting Information Figure S5. Effect of Wnt/β-catenin stroma on the cell proliferation of mitotic activity. 5-FU BMCs were co-cultured in the β-catenin/stroma or MPG/stroma for 5 days as described in Figure 3. Shown are the total number of cells obtained from experiments described for Figure 3 indicating comparable numbers of total cells after culture (3 experiments p>0.05) with error bars representing SEM.
STEM_52_sm_suppinfofigure6.tif684KSupporting Information Figure S6. Induction of dlk-1 in the stroma by Wnt-3a stimulation. (A) Induction of dlk-1 in the stromal cells after stimulation with Wnt-3a CM. Stromal cells were stimulated with Wnt-3a CM or control-CM for 12 hrs and expression of dlk-1 in the cell was examined by real-time PCR. Shown are the representative plots with the gray line representing Wnt3a-CM treated cell and the brown line representing control-CM treated cells. (B) In-vivo induction of dlk-1 in the endosteum of trabecular bone marrows of mice stimulated with Wnt 3a. Mice were intravenously injected with Wnt3a-CM or control CM and their bone marrows were examined 24 hrs after for dlk-1 by immunohistochemistry. Shown are images at low (200×) and higher magnification (400×) of dotted inlets visualized by vector blue staining (blue) and nuclear counterstaining by Nuclear Fast-Red (red). Arrows indicate positive staining for dlk-1 in the trabecular endosteum.
STEM_52_sm_suppinfofigure7.tif666KSupporting Information Figure S7. Wnt/β-catenin signals target stromal cells of the peri-sinusodial microenvironment of bone marrow. The stromal cells in the peri-sinusoidal (vascular) component of the bone marrow niche were examined for stromal targeting of Wnt/β-catenin signals by in-situ bone marrow examination (A-B), in-vivo effects (C) and in-vitro examination of purified cells (D-F). Mice were intravenously injected with Wnt 3a-CM 24hrs before (12hr × 2 times) and bone marrows were double immunostained with indicated antibodies. (A) Identification of β-catenin (+) or CD146(+) cells in the vicinity of MECA-32+ sinusoid of murine bone marrow. Bone marrows were examined by double immunostaining for β-caten(+) cells or peri-sinusoidal CD146(+) cells, a cell population previously shown to represent the peri-sinusoidal niche (Sacchetti et al., 2007). Shown are representative images (magnification 1000×) showing β-catenin (+) cells or CD146(+) cells (brown, DAB) in the vicinity of the sinusoidal endothelium (MECA-32, blue, vector blue). (C) Co-localization of β-catenin and CD146 in the peri-sinusoidal cells of murine bone marrow. Wnt 3a-CM injected bone marrows were double stained with antibody against β-catenin (total form) and murine CD146. Left; Co-localization of β-catenin (brown, DAB) and CD146 (blue, vector blue) in the peri-sinusoidal (S) cells of trabecular endosteal region. Arrows indicate positive staining of each reaction. Right; Co-localization of β-catenin (blue, vector blue) and CD146 (brown, DAB) in the peri-sinusoidal cells of the metaphysis marrow cavity, as indicated by arrows. (C) In-vivo response of peri-sinusoidal stromal cells to Wnt-3a stimulation. Bone marrows of mice intravenously injected with Wnt 3a-CM (2×, 12hr interval) were examined for β-catenin accumulation in peri-sinusoidal stromal cell (CD45-CD31-CD146+) populations by flowcytometry. Shown are representative flowcytometry profiles for intra-cellular staining of β-catenin (total form) in the gated CD45(-)CD31(-)CD146(+) population of bone marrows (left) and the mean fluorescence intensity of β-catenin staining (3 Expts, *p<0.05). The colored region of the histograms represents the isotype control, the dotted line represents bone marrows injected with Wnt 3a-CM and the black line bone marrows injected with control-CM. (D-F) In vitro response of purified peri-sinusoidal stromal cells to Wnt-3a stimulation. (D). Representative flowcytometry profiles for sorting CD45-CD31-CD146+ cells from lineage-depleted bone marrow cells. (E) Immunofluorescence staining of β-catenin in purified CD45(-)CD31(-)CD146(+) cells stimulated in vitro for 6hrs with control-CM or Wnt 3a-CM. Shown are representative images showing induction and nuclear localization of β-catenin obtained from two independent experiments. (F) Immunoblot analysis of β-catenin from purified CD45(-)CD31(-)CD146(-) or CD45(-)CD31(-)CD146(+) cells stimulated for 12 hrs in vitro with control-CM or Wnt 3a-CM. Each lane contains protein lysates from 2×105 cells.
STEM_52_sm_suppinfotable.doc108KSupporting Information Table S1. Extracellular genes induced in β-catenin transduced MSCs. Illumina BeadChip array hybridization analysis was performed to screen the genes induced in β-catenin/stroma compared to the control/stroma. 134 genes were shown to be consistently induced in β-catenin/stroma in four independent experiments with significant difference in the expression levels between the two group (>3.0 folds, p<0.005). Shown is a subset of these genes that are localized in the extracellular region as categorized by gene ontology term using DAVID, an online tool for identification of enriched groups within gene lists (

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.