JAGGED1-induced mineralization and gene expression in HBO cells:

Seven HBO cell lines were treated with growth media alone, osteogenic media alone or with Fc-Dynabeads (5.7 μM) or JAG1-Dynabeads (5.7 μM). The cells were half-fed every 5 days. On day 21 cells were fixed with 50% ethanol and thereafter, stained with 1% Alizarin Red S. (B) Alizarin Red S dye was extracted from Alizarin Red S-stained cells using a 1:10 dilution of acetic acid and water, and the absorbance was read at 420 nm. Representative image of HBO2. Data represents the mean values of three technical replicates per cell line (mean ± SD, one way ANOVA with Tukey post-hoc). (C) HBO1 primary cell line was grown in triplicate and treated with growth media alone, osteogenic media alone or with Fc-Dynabeads (5.7 μM) or JAG1-Dynabeads (5.7 μM). The cells were half-fed every 5 days and collected at 7 days, 14 days, and 21 days. qRT-PCR was performed (see Methods). Data was normalized to growth media with GAPDH as the reference gene. Data represents the mean values of three biological and two technical replicates per condition (mean ± SD, ordinary one-way ANOVA with Šídák’s multiple comparisons test, with single pooled variance).

JAG1 delivery in a PEG hydrogel stimulates bone regeneration in a critical-sized bone defect mouse model:

HBO cells alone or in the presence of JAG1-Dynabeads complex (20 μM) ± DAPT and BMP2 (2.5 µM)+ Fc-Dynabeads were incorporated in 4% PEG-MAL hydrogels and implanted into 4 mm critical-sized defects in the parietal bones of 6-8-week old NOD SCID mice (n = 4 – 6 per HBO-cell donor, 13 – 15 total) as two separate doses (Initial dose, Week four). After eight weeks, we quantified differences in regenerated bone volume within the defect and compared them between experimental groups by μCT analysis. (A) μCT reconstructions of defects. (B) Quantification of regenerated bone volume. Data are presented as mean (n = 13-15) ± SD with p-values reported (one way ANOVA with Šídák’s multiple comparisons test). (C) shows representative sections of the defect area on skulls from mice from all experimental groups stained with Masson Trichrome stain.

Transcriptional profiling reveals genes and pathways stimulated by non-canonical NOTCH signaling.

(A) RNAseq reveals clusters of genes associated with the non-canonical NOTCH pathway (rows are z-scored, side color bar identifies up- and down-regulated DEGs stimulated by the non-canonical pathway). (B) Overlapping DEGs in the JAG1-bds vs no-treatment and JAG1-bds + DAPT vs no-treatment comparisons reveal the non-canonical pathway (DEseq2). (C) Overlapping DEGs from JAG1-bds + DAPT vs no-treatment comparison. (D) Gene ontology over-representation test reveals significantly enriched up- and down-regulated pathways (FDR adjusted).

JAGGED1 induces a non-canonical NOTCH pathway in HBO cells.

HBO cells undergo mineralization through a non-canonical pathway. Luminex analysis of lysates obtained from three HBO cell lines untreated or treated with Dynabeads-bound recombinant JAG1-Fc fragment (5.7 μM) ± DAPT (15 μM), a NOTCH canonical pathway inhibitor in a time course manner (5, 10, 15 and 30 minutes), A) Heatmaps and B) Z-scores plotted on graphs. Each data point represents mean n = 3 ± SD per cell line with p-values reported.

p70 S6K is an essential target during JAGGED1-induced mineralization of HBO cells:

HBO cells were treated with growth media alone, osteogenic media alone or with Fc-Dynabeads (5.7 μM), S6K-18 alone (a p70 S6K phosphorylation inhibitor) (50 μM), and JAG1-Dynabeads (5.7 μM) alone or in combination with S6K-18 (50 μM). The cells were half-fed every 5 days. On day 21 cells are fixed with 50% ethanol and thereafter, stained with 1% Alizarin Red S. (B) Alizarin Red S dye was extracted from stained cells using a 1:10 dilution of acetic acid and water, and the absorbance was read at 420nm. Representative image of HBO7. Data represents mean n = 3 ± SD per cell line with p-values indicated.

Method of processing human bone samples to produce primary Human Bone-derived Osteoblast-like cell lines.

Human Bone-derived Osteoblast-like cells (HBO) were isolated by collagenase digestion of pediatric healthy fibular bones. See Methods.

PEG-4MAL hydrogel-encapsulated JAGGED1-induced mineralization of HBO cells.

HBO cells alone or in the presence of JAG1-Dynabeads complex (20 μM) were incorporated in 4% PEG-MAL hydrogels and grown in culture. The cells were half-fed every 5 days. On day 21 cells were fixed with 50% ethanol and thereafter, stained with 1% Alizarin Red S. (B) Alizarin Red S dye was extracted from stained cells using a 1:10 dilution of acetic acid and water, and the absorbance was read at 420nm. Data represents mean ± SD per cell line with p-values indicated, n = 3.

Visual Depiction of Calvarial Defect Studies.

HBO cells alone or in the presence of other treatments were incorporated in 4% PEG-MAL hydrogels and implanted into 4 mm critical-sized defects in the parietal bones of 6-8-week-old NOD SCID mice as two separate doses (Initial dose, Week four). After eight weeks, the regenerated bone volume, within the defect and compared them between experimental groups by μCT analysis.

Pilot study of JAG1-bds delivery in a PEG hydrogel stimulates bone regeneration in a critical-sized bone defect mouse model.

No cells (Empty Defect) or HBO cells alone or in the presence of JAG1-Dynabeads complex (20 μM) ± DAPT and BMP2 (2.5 µM)+ Fc-Dynabeads were incorporated in 4% PEG-MAL hydrogels and implanted into 4 mm critical-sized defects in the parietal bones of 6-8-week old NOD SCID mice (n = 3) as two separate doses (Initial dose, Week four). After eight weeks, we quantified differences in regenerated bone volume within the defect and compared them between experimental groups by μCT analysis. (A) μCT reconstructions of defects. (B) Quantification of regenerated bone volume. Data are presented as mean (n = 3) ± SD with p-values reported (ordinary one-way ANOVA with Tukey’s multiple comparisons test with a single pooled variance).

Immunohistochemical staining of calvarial defect tissue for Collagen 1.

FFPE tissue from the calvarial defect experiment shown in Figure 2 was sectioned and stained with an antibody for COL1A1 (Cell Signaling #72026S) and counterstained with Hematoxylin. Slides were scanned with the Olympus Nanozoomer whole-slide scanner at 20x.

Mineralization assay with DAPT inhibition of the NOTCH canonical pathway:

HBO1 cells were treated with JAG1-Dynabeads (5.7 μM) alone or in combination with increasing concentrations of DAPT in a dose response test (50 μM). The cells were half-fed every 5 days. On day 21 cells were fixed with 50% ethanol and thereafter, stained with 1% Alizarin Red S. Representative image of HBO1. (B) Alizarin Red S dye was extracted from stained cells using a 1:10 dilution of acetic acid and water, and the absorbance was read at 420nm. Data represents mean ± SD per cell line with p-values indicated, n = 9.

Inhibition of JAGGED1-induced mineralization of HBO cells with inhibitors of NOTCH and p70 S6K:

HBO cells were treated with JAG1-bds (5.7 μM) alone or in combination with DAPT (15µM), S6K-18 (a p70 S6K phosphorylation inhibitor) (50 μM) or S6K-18 (50 μM) + DAPT (15 μM). The cells were half-fed every 5 days. Mineralization assays were conducted in triplicate, and cells were collected at days 9, 14, and 21. RNA was subsequently assessed by qRT-PCR as described in Methods. Data was normalized to growth media with GAPDH as the reference gene. These charts compare normalized expression of genes with JAGGED1 stimulation and with inhibitors. Data represents the mean values of three biological and two technical replicates per condition (mean ± SD, ordinary one-way ANOVA with Šídák’s multiple comparisons test, with single pooled variance).