FIGURE SUMMARY
Title

Proteoglycan inhibition of canonical BMP-dependent cartilage maturation delays endochondral ossification

Authors
Koosha, E., Brenna, C.T.A., Ashique, A.M., Jain, N., Ovens, K., Koike, T., Kitagawa, H., Eames, B.F.
Source
Full text @ Development

Mutant Fam20b proteins have severely hypomorphic kinase activities. (A) Model illustrating that Fam20b transiently phosphorylates xylose, the first sugar added to a serine residue of the core protein during synthesis of a chondroitin sulfate PG (Koike et al., 2009). 2-Phosphoxylose phosphatase (PXYLP) removes the phosphate, thus promoting efficient glycosaminoglycan side chain outgrowth (Koike et al., 2014). (B) Culture medium from COS-1 cells transfected with secreted forms of zebrafish Fam20b or vector alone was incubated with IgG-Sepharose, and proteins purified from the medium were subjected to SDS-PAGE. Expression of each protein A-tagged protein was examined with anti-mouse IgG antibody. Lane 1: Fam20bwt; lane 2: Fam20bb1125; lane 3: Fam20bb1127; lane 4: protein marker; lane 5: vector alone. (C) After normalization from immunoblots in B, evaluation of kinase activity demonstrated that the mutant kinases Fam20bb1125 and Fam20bb1127 both have a significant decrease in activity compared with Fam20bwt, with Fam20bb1127 displaying significantly less activity than Fam20bb1125. Both mutant kinases, however, showed more activity than the negative, empty-vector control. Data are mean+s.e.m. (n=3, Tukey's multiple comparison test, *P<0.05). CS, chondroitin sulfate; Gal, galactose; GalNAc, N-acetylgalactosamine; GlcUA, glucuronic acid; P, phosphate; Ser, serine; Xyl, xylose.

Histological and molecular markers show similar timing of early cartilage formation between wild types and fam20b mutants. (A-B′) Lack of Safranin O staining demonstrated that developing ceratohyal cells of wild-type and fam20b mutant zebrafish did not secrete cartilage PGs at 48 hpf. (C-F) Cells in the mesenchymal condensation of both wild types (C,C′,E) and fam20b mutants (D,D′,F) were specified to form cartilage, as evidenced by col2a1a (C-D′) and sox9a (E,F) expression. (G-H′) The spacing between chondrocytes suggested that cartilage PG secretion had occurred in ceratohyal chondrocytes of both wild-type and fam20b mutant zebrafish at 72 hpf, but light Safranin O staining was only observed in wild types. Scale bars: 50 μm. ch, ceratohyal; oc, oral cavity; SafO, Safranin O.

Smad-mediated BMP signalling is increased in fam20b−/− chondrocytes. (A-D′) Fluorescence immunostaining showed no clear differences between wild types (A,C,C′) and fam20b mutants (B,D,D′) in the levels of p-Smad1/5/9 in nuclei and non-nuclear regions of condensing ceratohyal mesenchyme at 48 hpf. (E-H′) p-Smad1/5/9 immunoreactivity appeared higher in non-nuclear regions of fam20b−/− ceratohyal chondrocytes (F,H,H′) at 72 hpf, compared with wild types (E,G,G′). (I-J′) p-Smad1/5/9 immunoreactivity appeared increased in both the nuclei and non-nuclear regions of fam20b−/− chondrocytes (J,J′), compared with wild types (I,I′), at 84 hpf. (K) Quantitative image analyses (n=3 for each group) demonstrated significant increases in p-Smad1/5/9 levels of fam20b−/− chondrocytes (*P<0.05, one-way ANOVA and paired Student's t-test). Scale bars: 50 μm. a.u., arbitrary units; ch, ceratohyal; oc, oral cavity; wt, wild type.

p38-mediated BMP signalling does not differ in fam20b−/− chondrocytes. (A-B′) Condensing ceratohyal mesenchyme of wild types (A,A′) and fam20b mutants (B,B′) did not show p-p38 immunoreactivity at 48 hpf. (C-D′) p-p38 immunoreactivity appeared similar in ceratohyal chondrocytes of wild types (C,C′) and fam20b mutants (D,D′) at 72 hpf. (E) Quantitative image analyses (n=6 for each group) revealed no significant differences in p-p38 levels of fam20b−/− chondrocytes. Scale bars: 50 μm. a.u., arbitrary units; wt, wild type.

Smad-mediated BMP signalling does not increase in the fam20b−/− perichondrium region. (A-B′) p-Smad1/5/9 immunoreactivity did not show a clear difference in fam20b−/− perichondral cells (B,B′) at 72 hpf, compared with wild types (A,A′). (C) Quantitative image analyses (n=3 for each group) revealed no significant differences in p-Smad1/5/9 levels of fam20b−/− perichondral cells. Scale bars: 50 μm. a.u., arbitrary units; ColII, Collagen type II; pc, perichondral cells; wt, wild type.

Inhibition of BMP signalling by DMH1 in wild types reduces cartilage maturation gene expression and perichondral bone formation. (A-D′) Immunostaining of wild-type chondrocytes at 72 hpf (i.e. after 24 h of treatment) demonstrated that DMH1 treatment reduced p-Smad1/5/9 (B,B′) and p-p38 (D,D′) immunoreactivity, compared with DMSO-treated control chondrocytes (A,A′,C,C′). (E) Quantitative image analyses (n=3 for each group) revealed significant decreases in p-Smad1/5/9 and p-p38 levels in nuclei, as well as a significant decrease in p-Smad1/5/9 in non-nuclear regions, of DMH1-treated chondrocytes, compared with DMSO-treated control chondrocytes (*P<0.05, one-way ANOVA and paired Student's t-test). (F-I) Compared with DMSO-treated control chondrocytes (F,G), DMH1 treatment decreased expression of the chondrocyte maturation genes ihha and col10a1a at 4 dpf (i.e. after 2 days of treatment; H,I). These are representative images from at least 12 samples for each group (at least six samples each from two clutches). (J-K″) Compared with DMSO-treated controls (J-J″), DMH1 treatment appeared to decrease perichondral bone at 7 dpf (i.e. 3 days after treatment ended; K-K″). (L) Quantitation of 100 larvae (20 larvae each from five clutches) for each experimental group confirmed a significant decrease in perichondral bone in DMH-treated wild types (*P<0.05, one-way ANOVA and paired Student's t-test). Scale bars: 50 μm (A-D′,F-I); 200 μm (J,K). A, anterior; AB/AR, Alcian Blue/Alizarin Red; a.u., arbitrary units; Bsr, branchiostegal ray; Ch, ceratohyal cartilage; Chb, ceratohyal bone; Hm, hyomandibular bone; Hs, hyosymplectic cartilage; L, lateral; Op, opercle; wt, wild type.

Smad-mediated BMP signalling is decreased in chondrocytes of heat-shocked, dnBMPR zebrafish embryos. (A-D′) Condensing ceratohyal mesenchyme of heat-shocked dnBMPR zebrafish embryos (D,D′) showed reduced levels of p-Smad1/5/9 immunoreactivity at 48 hpf (i.e. 24 after 20-min heat shock), compared with non heat-shocked wild-type (A,A′) and dnBMPR (B,B′) embryos and heat-shocked wild-type embryos (C,C′). (E) Quantitative image analyses (n=3 for each group) revealed significant decreases in p-Smad1/5/9 levels in nuclei of condensing ceratohyal mesenchyme of heat-shocked dnBMPR zebrafish, compared with each control group (*P<0.05, one-way ANOVA and paired Student's t-test). (F-I′) Chondrocytes of heat-shocked dnBMPR zebrafish embryos (I,I′) showed reduced levels of p-Smad1/5/9 immunoreactivity at 72 hpf (i.e. 48 h after 20-min heat shock), compared with control groups (F-H′). (J) Quantitative image analyses (n=3 for each group) revealed significant decreases in p-Smad1/5/9 levels in nuclei, but not in non-nuclear regions, of chondrocytes of heat-shocked dnBMPR zebrafish at 72 hpf, compared with each control group (*P<0.05, one-way ANOVA and paired Student's t-test). Scale bars: 50 μm. a.u., arbitrary units; HS, heat-shocked; wt, wild type.

Smad-mediated BMP signalling is decreased in the perichondrium region of heat-shocked, dnBMPR zebrafish embryos. (A-D′) Levels of p-Smad1/5/9 immunoreactivity appeared reduced in perichondral cells of heat-shocked dnBMPR ceratohyals at 72 hpf (i.e. 48 h after 20-min heat shock). (E) Quantitative image analyses (n=3 for each group) revealed significant decreases in p-Smad1/5/9 levels in both nuclei and non-nuclear regions of perichondral cells in heat-shocked dnBMPR zebrafish, compared with each control group (*P<0.05, one-way ANOVA and paired Student's t-test). Scale bars: 50 μm. a.u., arbitrary units; ColII, Collagen type II; HS, heat-shocked; pc, perichondral cells; wt, wild type.

p38-mediated BMP signalling does not differ in chondrocytes of heat-shocked, dnBMPR zebrafish embryos. (A-D′) Condensing ceratohyal mesenchyme did not show much p-p38 immunoreactivity at 48 hpf in non-heat-shocked wild-type (A,A′) and dnBMPR (B,B′) embryos and heat-shocked wild-type (C,C′) and dnBMPR (D,D′) embryos (i.e. 24 h after 20-min heat shock). (E) Quantitative image analyses (n=3 for each group) revealed no significant differences in p-p38 levels in condensing ceratohyal mesenchyme among experimental groups. One-way ANOVA and paired Student's t-test. (F-I′) Chondrocytes of heat-shocked dnBMPR zebrafish embryos (I,I′) showed similar levels of p-p38 immunoreactivity at 72 hpf (i.e. 48 h after heat shock) as control groups (F-H′). (J) Quantitative image analyses (n=3 for each group) revealed no significant differences in p-p38 levels in chondrocytes of heat-shocked dnBMPR zebrafish, compared with each control group. One-way ANOVA and paired Student's t-test. Scale bars: 50 μm. a.u., arbitrary units; HS, heat-shocked; wt, wild type.

Inhibition of BMP signalling by heat shocking dnBMPR zebrafish embryos reduces cartilage maturation gene expression and perichondral bone formation. (A-H) Compared with chondrocytes in non-heat-shocked wild-type (A,B) and dnBMPR (C,D) embryos and heat-shocked wild-type (E,F) embryos, heat-shocked dnBMPR (G,H) chondrocytes had decreased expression of the chondrocyte maturation genes ihha and col10a1a at 4 dpf (i.e. 3 days after heat shock). These are representative images from at least 12 samples for each group (at least six samples each from two clutches). (I-L″) Compared with control groups (I-K″), perichondral bone appeared to decrease in heat-shocked dnBMPR embryos (L-L″) at 7 dpf (i.e. 6 days after heat shock). (M) Quantitation of 100 embryos (20 embryos each from five clutches) for each experimental group confirmed a significant decrease in perichondral bone in heat-shocked dnBMPR embryos (*P<0.05, one-way ANOVA and paired Student's t-test). Scale bars: 50 μm (A-H); 200 μm (I-L″). A, anterior; AB/AR, Alcian Blue/Alizarin Red; Ch, ceratohyal cartilage; Chb, ceratohyal bone; Hm, hyomandibular bone; Hs, hyosymplectic cartilage; HS, heat-shocked; L, lateral; wt, wild type.

Inhibition of BMP signalling by DMH1 or by heat shocking dnBMPR zebrafish embryos rescues the increased BMP signalling in fam20b−/− chondrocytes. (A-D′) Compared with DMSO-treated (A,A′) or heat-shocked (C,C′) fam20b−/− chondrocytes, levels of p-Smad1/5/9 at 72 hpf appeared decreased in DMH1-treated fam20b−/− (i.e. after 24 h of treatment; B,B′) or heat-shocked dnBMPR; fam20b−/− (i.e. 2 days after 20-min heat shock; D,D′) chondrocytes. (E) Quantitative image analyses (n=3 for each group) revealed significant decreases in p-Smad1/5/9 levels in both nuclei and non-nuclear regions of DMH1-treated or heat-shocked fam20b−/− chondrocytes, compared with each control group (*P<0.05, one-way ANOVA and paired Student's t-test). (F-G′) Compared with DMSO-treated fam20b−/− chondrocytes (F,F′), levels of p-p38 at 72 hpf (i.e. after 24 h of treatment) appeared decreased in DMH1-treated fam20b−/− chondrocytes (G,G′). (H) Quantitative image analyses (n=3 for each group) revealed significant decreases in p-p38 levels in nuclei of DMH1-treated fam20b−/− chondrocytes (*P<0.05, one-way ANOVA and paired Student's t-test). Scale bars: 50 μm. a.u., arbitrary units; HS, heat-shocked.

Inhibition of BMP signalling by DMH1 or by heat shocking dnBMPR zebrafish embryos rescues early cartilage maturation gene expression and perichondral bone formation in fam20b mutants. (A-H) Compared with DMSO-treated (A,B) or heat-shocked (E,F) fam20b−/− chondrocytes, DMH1-treated fam20b−/− (C,D) or heat-shocked dnBMPR; fam20b−/− (G,H) chondrocytes had decreased expression of the chondrocyte maturation genes ihha and col10a1a at 72 hpf (i.e. after 24 h of DMH1 treatment, or 2 days after heat shock). These are representative images from at least 12 samples for each group (at least six samples each from two clutches). (I-L″) Compared with respective control groups (I-I″,K-K″), perichondral bone appeared to decrease in DMH1-treated fam20b−/− (i.e. 1 day after end of 48 h treatment; J-J″) or heat-shocked dnBMPR; fam20b−/− (i.e. 4 days after heat shock; L-L″) embryos at 5 dpf. (M) Quantitation of 60 embryos (20 embryos each from three clutches) for each experimental group of the DMH1 experiment and of 80 embryos (20 embryos each from four clutches) for each experiment group of the dnBMPR experiment confirmed a significant decrease in perichondral bone of DMH1-treated fam20b−/− or heat-shocked dnBMPR; fam20b−/− embryos (*P<0.05, one-way ANOVA and paired Student's t-test). Scale bars: 50 μm (A-H); 200 μm (I-L″). A, anterior; AB/AR, Alcian Blue/Alizarin Red; Ch, ceratohyal cartilage; Chb, ceratohyal bone; Hm, hyomandibular bone; Hs, hyosymplectic cartilage; HS, heat-shocked; L, lateral.

Inhibition of BMP signalling by early DMH1 treatment reduces perichondral bone formation in fam20b mutants down to wild-type levels at 10 dpf. (A-H″) Compared with DMSO-treated wild-type controls (A-B″), perichondral bone appeared to decrease in DMH1-treated wild types at 10 dpf (i.e. 6 days after end of 48 h treatment; E-F″). Similarly, perichondral bone at 10 dpf appeared to decrease in DMH1-treated fam20b mutants (G-H″), compared with DMSO-treated fam20b mutants (C-D″), with levels similar to that seen in DMSO-treated wild types (A-B″). (I) Quantitation of 20 embryos for each experimental group confirmed significant decreases (*P<0.05, one-way ANOVA and paired Student's t-test) in perichondral bone of DMH1-treated wild-type or fam20b−/− embryos. Remarkably, the levels of perichondral bone in DMH1-treated fam20b mutants was statistically indistinguishable from DMSO-treated wild-type controls. Scale bars: 200 μm (A-H). A, anterior; AB/AR, Alcian Blue/Alizarin Red; Ch, ceratohyal cartilage; Chb, ceratohyal bone; Hm, hyomandibular bone; Hs, hyosymplectic cartilage; HS, heat-shocked; L, lateral; n.s., not significant; wt, wild type.

Acknowledgments
This image is the copyrighted work of the attributed author or publisher, and ZFIN has permission only to display this image to its users. Additional permissions should be obtained from the applicable author or publisher of the image. Full text @ Development