Figure 1.
Tnni2K175del mutant mice showed small body size.
(A) The conservation of the 175K amino acid in TNNI2 from humans to zebrafish. (B) P18 mutants exhibited smaller body sizes than wild-type littermates. (C) The comparison of weights of newborn wild-type and mutant mice. (D) The weight changes of heterozygous (n = 7) and wild-type mice littermates (n = 7) from P5 to P20. (E) Weights of heterozygous mutant mice (n = 125) and wild-type littermates (n = 111) at P12. (F) P12 heterozygous mice and wild-type littermates were sacrificed for measurement of the length of spines and limbs. Length ratio of the spines (blue), fore limbs (orange) and hind limbs (green) of wild-type mice (n = 5) to Tnni2K175del littermates (n = 5) at P12. (G) 2-month-old heterozygous mice and wild-type littermates were narcotized for measurement of the length of spines and limbs. Lengths ratio of the spines (blue) and fore limbs (orange) of heterozygous littermates (n = 5) to those of wild-type littermates (n = 5) at 2 months of age. Student t-test, mean±s.d. Postnatal day, P.
Figure 2.
Tnni2 was expressed in chondrocytes and osteoblasts.
(A) In situ hybridization analyses with antisense Tnni2 riboprobe on histological sections of wild-type radii of E15.5 embryos showed that Tnni2 mRNA was expressed at the growth plates. Pink indicated Tnni2 mRNA signal (arrow). (B–G) Immunostaining analyses of tnni2 expression in growth plate. (B–C) tnni2 was expressed at chondrocytes (B, arrow) and osteoblasts (C, arrows) of the tibia growth plates from P8 wild-type mice. (D–F) The expression of tnni2 in the growth plates of radii and ulnae exhibited a spatial and temporal change. (D) In E15.5 wild-type embryos, tnni2 was expressed in pre-hypertrophic and hypertrophic chondrocytes (arrows). (E) At E17.5 wild-type embryos, tnni2 was mainly observed in proliferating chondrocytes (arrows). (F) At P6 wild-type radii, expression of tnni2 was in pre-hypertrophic, hypertrophic, immature proliferating and rest chondrocytes (arrows), whereas no discernible tnni2 was observed in mature proliferative chondrocytes. (G) Expression of tnni2 was in articular chondrocytes of tibiae of P12 wild-type mice (arrows). (H) tnni2 was observed in calvarial osteoblasts of P6 wild-type mice (arrows). (I) Negative staining for tnni2 protein. Scale bar, 100 µm (A–I). Embryonic day, E.
Figure 3.
The Tnni2K175del mutation disturbed bone development of mice.
Skeletons staining using Alcian blue for cartilage (blue) and Alizarin Red S for calcified bone (red). (A) Decreased mineralization (arrows) in the calvarias from Tnni2K175del/K175del mice at E16.5 and from Tnni2+K175del mice at P2 and P5 compared to their wild-type littermates, respectively. (B) The primary ossification centers (arrows) in forelimb bones from Tnni2K175del/K 175del mice at E14.75 were shorter than those of their wild-type controls. (C) Reduced ossification (arrows) in P5 mutant ribs. (D and E) At P6, the secondary ossification in phalanxes and carpals (D, the boxed regions) and metatarsal bones (E, the boxed regions) of Tnni2+/K175del mice were fewer than those of wild-type littermates. (F) Growth plates (the boxed regions) in radii and ulnae of heterozygous mutant mice at P10 were longer and less ossified than those of wild-type littermates. (G) Decrease in cartilage ossification in the P2 tails of mutants (the rectangle regions). (H) Skeletal preparation of homozygous mice and wild-type littermates at P15. All experiments were repeated using at least two pairs of samples.
Figure 4.
The Hif3a expression was significantly upregulated in the Tnni2K175del mutant long bones.
(A) Heat map diagram of the 64 top ranking differentially expressed genes (Table S1) in 1-day-old homozygous mutant radii and ulnae (n = 3) compared to wild-type littermate controls (n = 3). Genes are shown in rows; samples are showed in columns. Red bar indicated higher expression and blue bar indicated lower expression. (B) GSEA plot (top) showed that differential expression of genes between the entire radii and ulnae from 1-day-old homozygous mice (n = 3) and wild-type littermates (n = 3) were significantly enriched in hypoxia induced gene set. Heat-map diagram (bottom) showed the 20 core-enriching genes in the leading edge. The number of permutation was 1000 times. (C) Relative mRNA expression of Hif1a and Hif3a (normalized to Gapdh) in our microarray dataset. (D) The relative mRNA expression of Hif-associated signaling molecules (normalized to Gapdh) in 1-day-old homozygous mutant radii and ulnae over those of wild-type littermates (Homo/WT), using qPCR analyses.
Figure 5.
The disturbed Hif associated signaling in Tnni2K175del mice resulted in reduced angiogenesis and impeded formation of ossification centers of long bones.
(A–B) Immunostaining analyses of the hif3a (A) and hif1a (B) expression in the hypertrophic cell zones (arrows) of radii and ulnae from Tnni2K175del/K175del mice and wild-type littermates at P12. Right panel of A: Quantification of hif3a expression (n = 3). (C) Immunostaining analyses revealed a decreased vegf in the hypertrophic cell zones of growth plates of radii and ulnae (arrows) from P12 heterozygous mutant mice compared to that of wild-type littermates. Right panel of C: Quantification of vegf expression (n = 3). (D) HE staining showed that the number of blood vessels surrounding trabecular bones were markedly decreased in radii from P12 mutants. Red blood cells were stained as red color. (E) Alkaline phosphatase staining exhibited a retarded formation of the primary ossification centers in radii of E15.5 homozygous mutants. Osteoblasts (blue arrows) have entered the marrow cavity and replaced the apoptotic chondrocytes in wild-type embryos but not in homozygous embryos. (F) Immunostaining revealed a decreased vegf in the hypertrophic cell zones of growth plates of radii and ulnae (arrows) from E15.5 homozygous mutant mice compared to those of wild-type littermates. (G) The relative mRNA expression of Vegf (normalized to Gapdh) in E15.5 homozygous mutant radii and ulnae over those of wild-type littermates (WT/Homo) using qPCR analyses. (H) CD31 stained the vessel endothelial cells in bone marrow of E15.5 homozygous mutant radii and wild-type controls. Insets showed magnified views of the box regions. (I) Quantification of blood vessel density in (H). (J) HE staining showed the growth plate of radii at P6. All experiments were repeated using at least three pairs of samples. Quantification analyses in A, C, G and I: Student's t-test, mean±s.d.; Scale bar, 100 µm (A–F, H, and J).
Figure 6.
Chondrocytes differentiation and osteoblasts proliferation were impaired in mutant long bones.
(A) Von Kossa stained at P6 nasal cartilage from Tnni2+/K175del mice and their wild-type littermates. Mineralization of the matrix and hypertrophic chondrocytes (arrows) were observed in the wild-type nasal cartilage, but few in mutant. (B) The relative mRNA expression of Ihh, Pthrp, Col10a1 and Comp (normalized to Gapdh) in 1-day-old wild-type radii and ulnae over those of homozygous mutant littermates (WT/Homo), using qPCR analyses. (C) Alcian blue stained growth plate of tibiae. Hypertrophic zones (HZ) were markedly shorter in the growth plate of tibiae of Tnni2K175del/K175del mice at P12 than those of their wild-type littermates. (D) Quantification of the length of hypertrophic cell zones in (C) in growth plates of tibiae at p12 wild-type (n = 6) and Tnni2K175del/K175del mice (n = 6). (E) Polarization irradiation staining showed that type I collagen (orange) at trabecular bone domains (the regions between white dotted lines) in tibia of P12 homozygous mice was sparse relative to those of control littermates. (F) Masson's trichrome staining showed that the amount of osteoid and osteoblasts. At P6, the amount of osteoid was sparse (the regions between white dotted lines), and the number of osteoblasts (arrows) surrounding trabecular bone was decreased in radii of heterozygous mice compared to wild-type littermates. (G, H) Quantification of the number of osteoblasts surrounding trabecular bones (G) and the number of trabecular bones per unit area (H) at the radii and ulnae of Tnni2+/K175del mice (n = 5) and wild-type controls (n = 4) at P12. The trabecular bone and the number of osteoblasts were decreased on average by 53% and 57% in heterozygous DA2B mice, respectively. (I) BrdU incorporation to detect proliferating osteoblasts surrounding trabecular bones at P12. The number of osteoblasts in Tnni2+/K175del mice at P12 was significantly decreased relative to their wild-type littermates. (J) Quantification analysis of BrdU-positive cells in (I). The percentage of positive cells relative to total cell count was determined (n = 3). (K) HE staining showed that vegf treatment was able to partially rescue the number and thickness of trabecular bones and the number of osteoblasts in heterozygous mutants. (L, M) Quantification of the number of osteoblasts surrounding trabecular bones (L) and the number of trabecular bones per unit area (M) in (K) (n = 4). All experiments were repeated using at least three pairs of samples. B, D, G, H, J, L and M: Student's t-test, mean±s.d., Scale bar, 100 µm (A, C, E, F, I and K).
Figure 7.
Mutant tnni2 protein increased hif3a and reduced vegf expression.
(A) Immunofluorescence staining of tnni2 (arrows) in the cytoplasm and nucleus of the primary osteoblasts under both hypoxic (1% O2) and normoxic (20% O2) conditions. tnni2 was stained as green. Nuclei were stained with DAPI (blue). (B) Upper panel: Chromatin immunoprecipitation (ChIP)-PCR assay was performed using primary osteoblasts from 1-day-old Tnni2K175del/K175del mice and control littermates. Both wild-type tnni2 and tnni2K175del mutant protein bond to the mouse Hif3a promoter by a 471 bp fragment (referred to as Part 1) in vivo; Lower panel: Nucleotide sequence of DNA of the Part 1. The eight highly conserved fragments were highlighted with Red or Blue color. (C) Schematic diagram of the Part 1. Upper panel: The position of Part 1 at Hif3a promoter. Middle panel: The Part 1 is consists of eight highly conserved fragments shown with green bars; Lower panel: The sequence feature of these repeat fragments. (D) DNA pull-down assays (DPDAs) performed with avidin-labeled dynabeads using the nuclear extracts (NE) from wild-type tnni2-GFP 293T stable cell line (WT) and NE from tnni2K175del-GFP 293 stable cell line (K175del). Biotin-labeled single repeat fragment of the Part 1 was used as DNA probe. Negative controls: wild-type or mutant NE incubated with non-labeled beads for nonspecific binding. Input: NE from wild-type TNNI2-GFP. Immunoblotting was performed using an anti-GFP antibody. (E) EMSA and Super-EMSA were performed using NE from wild-type tnni2-GFP 293T stable cell line and tnni2K175del-GFP 293T stable cell line. The DNA probe in (D) used here. (F) Luciferase assays of the truncated fragment (−1673 to 0 Hif3a promoter), the whole length fragment (−2200 to 0 Hif3a promoter) and the Part 1 fragment (−2200 to −1674 Hif3a promoter) were performed using 293T cell line under 20% O2 condition. Firefly Luciferase expression was measured and normalized by activity of Renilla Luciferase (pRP-TK). Ratio of the normalized Firefly Luciferase expression to that of control vector was used to represent relative luciferase activity. Data represents the mean±s.d. of five experiments carried out in quadruplicate. (G) Western blot analyses of tnni2, hif3a, vegf and hif1a in growth plates from P12 heterzyous mutant and wild-type littermate mice under normoxia condition. (H) Western blot analyses of tnni2, hif3a, vegf and hif1a in primary osteoblasts from newborn homozygous mutant and wild-type littermate mice. (I) MTT assay for testing proliferation of primary homozygous mutant osteoblasts and wild-type control (n = 3). (J) ELISA assay for testing secreted vegf in cell culture supernatant from primary homozygous mutant osteoblasts and wild-type control (n = 3). The secreted vegf levels of wild-type and mutant osteoblasts were normalized by cell number of the respective genotype at corresponding time point. (K) Von Kossa staining to test deposits of calcium of primary homozygous mutant osteoblasts and wild-type control. (L) Immunostaining for tnni2, hif3a, hif1a and vegf on consecutive sections of radii and ulnae of P12 homozygous mutant mice and wild-type littermate mice.