Advertisement
Indian hedgehog stimulates periarticular chondrocyte differentiation to regulate growth plate length independently of PTHrP
Tatsuya Kobayashi,1 Desi W. Soegiarto,2 Yingzi Yang,3 Beate Lanske,2,4 Ernestina Schipani,1 Andrew P. McMahon,5 and Henry M. Kronenberg1
1Endocrine Unit, Massachusetts General Hospital, Boston, Massachusetts, USA. 2Molecular Endocrinology, Max-Planck-Institute for Biochemistry, Martinsried, Germany. 3Developmental Genetics Section, Genetic Disease Research Branch, National Human Genome Research Institute, Bethesda, Maryland, USA. 4Department of Oral and Developmental Biology, The Forsyth Institute, Boston, Massachusetts, USA. 5Department of Molecular and Cellular Biology, Harvard University, Cambridge, Massachusetts, USA.
Address correspondence to: Henry M. Kronenberg, Endocrine Unit, Massachusetts General Hospital, 50 Blossom Street, Boston, Massachusetts 02114, USA. Phone: (617) 726-3966; Fax: (617) 726-7543; E-mail: kronenberg.henry@mgh.harvard.edu.
Find articles by Kobayashi, T. in: JCI | PubMed | Google Scholar
1Endocrine Unit, Massachusetts General Hospital, Boston, Massachusetts, USA. 2Molecular Endocrinology, Max-Planck-Institute for Biochemistry, Martinsried, Germany. 3Developmental Genetics Section, Genetic Disease Research Branch, National Human Genome Research Institute, Bethesda, Maryland, USA. 4Department of Oral and Developmental Biology, The Forsyth Institute, Boston, Massachusetts, USA. 5Department of Molecular and Cellular Biology, Harvard University, Cambridge, Massachusetts, USA.
Address correspondence to: Henry M. Kronenberg, Endocrine Unit, Massachusetts General Hospital, 50 Blossom Street, Boston, Massachusetts 02114, USA. Phone: (617) 726-3966; Fax: (617) 726-7543; E-mail: kronenberg.henry@mgh.harvard.edu.
Find articles by Soegiarto, D. in: JCI | PubMed | Google Scholar
1Endocrine Unit, Massachusetts General Hospital, Boston, Massachusetts, USA. 2Molecular Endocrinology, Max-Planck-Institute for Biochemistry, Martinsried, Germany. 3Developmental Genetics Section, Genetic Disease Research Branch, National Human Genome Research Institute, Bethesda, Maryland, USA. 4Department of Oral and Developmental Biology, The Forsyth Institute, Boston, Massachusetts, USA. 5Department of Molecular and Cellular Biology, Harvard University, Cambridge, Massachusetts, USA.
Address correspondence to: Henry M. Kronenberg, Endocrine Unit, Massachusetts General Hospital, 50 Blossom Street, Boston, Massachusetts 02114, USA. Phone: (617) 726-3966; Fax: (617) 726-7543; E-mail: kronenberg.henry@mgh.harvard.edu.
Find articles by Yang, Y. in: JCI | PubMed | Google Scholar
1Endocrine Unit, Massachusetts General Hospital, Boston, Massachusetts, USA. 2Molecular Endocrinology, Max-Planck-Institute for Biochemistry, Martinsried, Germany. 3Developmental Genetics Section, Genetic Disease Research Branch, National Human Genome Research Institute, Bethesda, Maryland, USA. 4Department of Oral and Developmental Biology, The Forsyth Institute, Boston, Massachusetts, USA. 5Department of Molecular and Cellular Biology, Harvard University, Cambridge, Massachusetts, USA.
Address correspondence to: Henry M. Kronenberg, Endocrine Unit, Massachusetts General Hospital, 50 Blossom Street, Boston, Massachusetts 02114, USA. Phone: (617) 726-3966; Fax: (617) 726-7543; E-mail: kronenberg.henry@mgh.harvard.edu.
Find articles by Lanske, B. in: JCI | PubMed | Google Scholar
1Endocrine Unit, Massachusetts General Hospital, Boston, Massachusetts, USA. 2Molecular Endocrinology, Max-Planck-Institute for Biochemistry, Martinsried, Germany. 3Developmental Genetics Section, Genetic Disease Research Branch, National Human Genome Research Institute, Bethesda, Maryland, USA. 4Department of Oral and Developmental Biology, The Forsyth Institute, Boston, Massachusetts, USA. 5Department of Molecular and Cellular Biology, Harvard University, Cambridge, Massachusetts, USA.
Address correspondence to: Henry M. Kronenberg, Endocrine Unit, Massachusetts General Hospital, 50 Blossom Street, Boston, Massachusetts 02114, USA. Phone: (617) 726-3966; Fax: (617) 726-7543; E-mail: kronenberg.henry@mgh.harvard.edu.
Find articles by Schipani, E. in: JCI | PubMed | Google Scholar
1Endocrine Unit, Massachusetts General Hospital, Boston, Massachusetts, USA. 2Molecular Endocrinology, Max-Planck-Institute for Biochemistry, Martinsried, Germany. 3Developmental Genetics Section, Genetic Disease Research Branch, National Human Genome Research Institute, Bethesda, Maryland, USA. 4Department of Oral and Developmental Biology, The Forsyth Institute, Boston, Massachusetts, USA. 5Department of Molecular and Cellular Biology, Harvard University, Cambridge, Massachusetts, USA.
Address correspondence to: Henry M. Kronenberg, Endocrine Unit, Massachusetts General Hospital, 50 Blossom Street, Boston, Massachusetts 02114, USA. Phone: (617) 726-3966; Fax: (617) 726-7543; E-mail: kronenberg.henry@mgh.harvard.edu.
Find articles by McMahon, A. in: JCI | PubMed | Google Scholar
1Endocrine Unit, Massachusetts General Hospital, Boston, Massachusetts, USA. 2Molecular Endocrinology, Max-Planck-Institute for Biochemistry, Martinsried, Germany. 3Developmental Genetics Section, Genetic Disease Research Branch, National Human Genome Research Institute, Bethesda, Maryland, USA. 4Department of Oral and Developmental Biology, The Forsyth Institute, Boston, Massachusetts, USA. 5Department of Molecular and Cellular Biology, Harvard University, Cambridge, Massachusetts, USA.
Address correspondence to: Henry M. Kronenberg, Endocrine Unit, Massachusetts General Hospital, 50 Blossom Street, Boston, Massachusetts 02114, USA. Phone: (617) 726-3966; Fax: (617) 726-7543; E-mail: kronenberg.henry@mgh.harvard.edu.
Find articles by Kronenberg, H. in: JCI | PubMed | Google Scholar
Published July 1, 2005 - More info
J Clin Invest. 2005;115(7):1734–1742. https://doi.org/10.1172/JCI24397.
© 2005 The American Society for Clinical Investigation
Received: January 5, 2005; Accepted: April 5, 2005
-
Results
Upregulation of Ihh action contributes to growth-plate phenotypes of mice with mosaic ablation of PPR signaling. We examined the effect of upregulation of Ihh action on growth-plate length in mice with PPR-mosaic ablation. In PPR-mosaic mice in which the floxed PPR genes are disrupted in some proliferating chondrocytes (Cre driven by an osteocalcin promoter:PPRfl/– mice [Ost-Cre:PPRfl/– mice]) (14), chondrocytes missing PPR signaling prematurely differentiate into hypertrophic chondrocytes in the columnar region, as shown by type X collagen (ColX) and Ihh expression (Figure 1A). This ectopic expression of Ihh causes upregulation of Ihh action in the periarticular region, as shown by the increase in Ptc and PTHrP expression (Figure 1A and data not shown). The extent of growth plate elongation of these PPR-mosaic mice (shown by Alcian blue staining) was reduced in an Ihh+/– background (100% ± 0.78% Ihh+/+ vs. 93% ± 1.38% Ihh+/–; n = 3 in each group; P < 0.05) (Figure 1A). We previously found that upregulation of Ihh action in the periarticular region was associated with accelerated differentiation of periarticular chondrocytes into columnar chondrocytes. This finding was based on an increase in the BrdU-labeling ratio despite smaller than normal sizes of periarticular regions in mice with PPR-mosaic ablation and in PPR-hypomorphic mice (14). We hypothesized, therefore, that BrdU-labeling ratios in periarticular chondrocytes might be lower in Ihh+/– than in Ihh+/+ backgrounds. We were, however, unable to show statistical differences either in BrdU-labeling ratio or in the size of the periarticular region in Ihh+/+ compared with Ihh+/– backgrounds, even when littermates were compared (data not shown). In the columnar region as well, we did not find differences in BrdU-labeling ratio between Ihh+/+ and Ihh+/– mice (data not shown). It appears that the effects of Ihh heterozygosity on chondrocyte proliferation are too small to measure accurately at any one moment in time, given current methods, but that these small changes might, over time, lead to substantial morphologic changes.
Figure 1Effect of Ihh reduction in mice with conditional mosaic ablation of the PPR. (A) The growth plate expansion in conditional PPR-mosaic mice (Ost-Cre:PPRfl/–) was partially reduced by Ihh heterozygosity. Ectopic hypertrophic differentiation due to loss of the functional PPR is shown by ectopic expression of the ColX mRNA (ColX; Col10a; MGI) and Ihh. Ihh heterozygosity reduced the Ihh expression level in both ectopic (arrowhead) and eutopic (double arrowhead) hypertrophic chondrocytes. The length of the growth plate (distance between the articular end and the interface between cartilage and bone marrow) was significantly different between Ost-Cre:PPRfl/–:Ihh+/+ and Ost-Cre:PPRfl/–:Ihh+/– mice (100% ± 0.78% Ihh+/+ versus 93% ± 1.38% Ihh+/– ; littermates were compared, n = 3 each group; P < 0.05). There was also modest shortening of the growth plate in PPR+/+:Ihh+/– compared with PPR+/+:Ihh+/+ mice. Vertical lines represent the periarticular and columnar regions. ALN, Alcian blue staining. (B) Conditional double mosaic ablation of the PPR and Ihh genes. The PPR floxed gene and Ihh-floxed gene were combined with the Ost-Cre transgene. In both PPR single mosaic (Ost-Cre:PPRfl/–:Ihhfl/+) and PPR/Ihh–double-mosaic (Ost-Cre:PPRfl/-:Ihhfl/–) growth plates, ectopic hypertrophic differentiation indicated by ColX mRNA (ColX) expression occurred whereas the level of ectopic Ihh mRNA (arrowheads) was reduced compared with that of the eutopic Ihh mRNA expression (double arrowheads). This reduction in Ihh mRNA expression accompanied reduction of growth plate expansion.
We further ablated Ihh simultaneously with PPR in a mosaic fashion using a floxed Ihh conditional allele (Figure 1B). The elongation of the columnar region in PPR–single-mosaic mice (Ost-Cre:PPRfl/–:Ihhfl/+) was no longer observed in the double mosaic mice (Ost-Cre:PPRfl/–:Ihhfl/–), demonstrating that Ihh was responsible for the elongation of the columnar region. Because we were unable to obtain multiple double mosaic mice in single litters, statistical analysis of the BrdU-labeling index in the periarticular region was not possible. Although the identity of the mechanism whereby upregulated Ihh action in PPR-mosaic mice causes growth plate elongation is not clear from this experiment, these results demonstrate the causative role of ectopic Ihh in growth plate elongation in the mosaic mice.
Ihh overexpression causes acceleration of periarticular chondrocyte differentiation. In order to investigate further the effect of Ihh overexpression in chondrocytes, chick-Ihh (cIhh) was overexpressed in the growth plate using the GAL4/UAS binary system. In this system, a type II collagen (Col2) promoter indirectly drives expression of cIhh (9, 15). These mice exhibit diverse developmental defects, including acranium, facial malformations, and fusion of knee joints (Figure 2B). Because of the distortion and fusion of growth plates of the proximal tibia and distal femur, we examined instead the distal tibia, in which normal morphology was relatively well preserved. At E17.5, Ihh bigenic (Ihh-Bg) mice showed longitudinal expansion of the growth plate while the width of the bone was diminished (Figure 2A). Upregulation of Ptc and PTHrP expression in the periarticular region indicated an increase in hedgehog signaling (Figure 2C). The BrdU-labeling assay revealed that the proliferation of periarticular chondrocytes (Figure 2E) was significantly increased in Ihh-Bg mice despite a decrease in the number of periarticular chondrocytes (Figure 2, D and E). The increase in BrdU-labeling ratio and decrease in cell number in the periarticular regions were observed in multiple stages of late fetal development (E16.5, E17.5, and E18.5; data not shown). In contrast, the proliferation rate in the columnar region (Figure 2E) did not differ statistically between the genotypes, whereas the number of columnar chondrocytes was significantly greater in Ihh-Bg mice than in control mice (Figure 2E).
Figure 2Ihh overexpression in the growth plate causes growth plate expansion and acceleration of periarticular chondrocyte differentiation. (A) Alcian blue staining showed that Ihh overexpression in the growth plate by the Col2-GAL4:UAS-cIhh bigenic system (Ihh-Bg) caused longitudinal elongation of the growth plate of the E17.5 distal tibia. Col2-GAL4 single-transgenic mice were used as the control (Ctrl). (B) The distal growth plate of the tibia was analyzed due to the knee joint fusion of Ihh-Bg. (C) Ptc and PTHrP upregulation in the periarticular region of the Ihh-Bg growth plate (arrowheads). (D and E) Cell number was reduced despite increased proliferation in the periarticular region of Ihh-Bg mice. Three mice in each group were analyzed. Sections representing the mid–cross-plane of the tibia were selected for BrdU staining (D) and cell number counting (D and E). Cell number in each region was counted on BrdU/H&E-stained sections and expressed as BrdU-positive cells/total cells in the periarticular (P) or columnar (C) regions of each section. Red and magenta dotted lines indicate the periarticular and columnar regions, respectively. (E) Despite an approximately 50% increase in BrdU-labeling ratio in the periarticular region of Ihh-Bg mice, the cell number was smaller than that in control mice by approximately 30%. In contrast, the cell number in the columnar (D) region of Ihh-Bg mice was approximately 30% greater, whereas the BrdU labeling ratio was similar to that of control mice. n = 3 in each group; *P < 0.05.
These observations strongly suggest that periarticular chondrocyte differentiation (periarticular to columnar) is accelerated in Ihh-Bg mice and that this acceleration may contribute to the elongation of the growth plate.
Effect of heterozygosity of PTHrP on the length of the growth plate with upregulated Ihh action. In Ost-Cre:PPRfl/– and Ihh-Bg mice, longitudinal expansion of the growth plate is primarily caused by an increase in the length of the columnar region. In these mouse models, since the rate of columnar cell proliferation was similar to that in control mice (Figure 2, D and E), the expansion of the growth plate could be due to an increase of periarticular to columnar differentiation, a decrease of columnar to hypertrophic differentiation, or both. In the mouse models with upregulation of Ihh action in the periarticular region, this increase of Ihh led to an increase in PTHrP expression. Since PTHrP inhibits premature hypertrophic differentiation in the late fetal growth plate, it is possible that the upregulation of PTHrP in the present mouse models suppresses hypertrophic differentiation and thereby contributes to the lengthening of the columnar region.
To clarify the role of PTHrP upregulation in the growth plate expansion in these mouse models, we compared degrees of growth plate expansion in PTHrP+/+and PTHrP+/– backgrounds in Ost-Cre:PPRfl/– and Ihh-Bg mice. In the genetic background used in these experiments, PTHrP heterozygosity alone caused modest shortening of the growth plate, approximately 8% (Figure 3, A and B). PTHrP heterozygosity caused only modest growth plate shortening of Ost-Cre:PPRfl/– and Ihh-Bg mice in which the basal PTHrP expression was upregulated (Figure 3, A and B). Reduction of PTHrP expression in PTHrP+/– mice was confirmed (Figure 3B and data not shown). The extent of reduction of the growth plate length in the PTHrP+/– background was approximately 8% in each of these models. The observation that there was only modest growth plate shortening even in Ost-Cre:PPRfl/– and Ihh-Bg mice, which had high basal PTHrP expression, suggests that growth plate elongation caused by Ihh overactivity may not be mediated by PTHrP upregulation alone.
Figure 3Effect of PTHrP heterozygosity on the length of the growth plate. Alcian blue–stained growth plates of E17.5 proximal tibiae from wild-type and E16.5 proximal tibiae from Ost-Cre:PPRfl/– mice (A) and E16.5 distal tibiae from control (Col2-GAL4) and Col2-GAL4:UAS-cIhh (Ihh-Bg) mice (B) in PPR+/+ and PPR+/– backgrounds are shown. In situ hybridization confirmed a reduction in PTHrP expression in Ihh-Bg:PTHrP+/– mice. Littermates were compared. Growth plates of PPR+/– mice were shorter than those of controls by approximately 8% regardless of other genetic modifications.
Ectopic hypertrophic chondrocytes cause columnar cell differentiation independently of PTHrP upregulation. These observations prompted us to investigate the independent contributions of Ihh and PTHrP in regulating the columnar cell mass. Because of the mutual regulation of PTHrP and Ihh actions in the growth plate, the feedback loop between these 2 ligands had to be disrupted in order to demonstrate each molecule’s distinct effect. We therefore tried to manipulate the PPR and Ihh genes while we maintained PTHrP signaling constant in cells in which the PPR was intact.
We used the mouse model with mosaic PPR ablation in the growth plate to learn what effects ectopic hypertrophic chondrocytes had on periarticular chondrocyte differentiation under conditions of fixed PTHrP expression. To achieve this, PTHrP–/– pups were rescued by expression of an exogenous PTHrP gene controlled by a Col2 promoter (3). Mosaic PPR ablation was then induced by Cre/loxP recombination on this background. The rescued growth plates of PTHrP–/–:PTHrP-Tg mice (mice with exogenously provided PTHrP/exo-PTHrP) were similar to those of PTHrP-Tg mice with wild-type PTHrP genes (Figure 4B and data not shown): the tibia was short, and bone and bone marrow formation were suppressed. This rescued growth plate lacked hypertrophic chondrocytes and columnar cells. When mosaic Cre/loxP recombination was induced on the floxed PPR gene (exo-PTHrP:mosaic), ectopically differentiated hypertrophic chondrocytes appeared (Figure 4C). These ectopic hypertrophic chondrocytes were accompanied by an increase in the length of the tibia, by bone and marrow formation (Figure 4, D and E), and by column formation (Figure 4, F and G). The ectopic hypertrophic chondrocytes expressed Ihh (Figure 4, H and I). Interestingly, mosaic PPR ablation stimulated eutopic hypertrophic differentiation as well (Figure 4C). This was confirmed by eutopic Ihh mRNA (Figure 4I) and ColX mRNA expression (data not shown). This result indicates that ectopic hypertrophic chondrocytes that express Ihh induce columnar chondrocyte differentiation independently of upregulation of PTHrP and stimulate subsequent differentiation of chondrocyte with intact PPRs toward hypertrophy.
Figure 4Ectopic hypertrophic chondrocytes induced column formation in the growth plate with exogenously controlled PTHrP. (A–C) H&E-stained growth plates from postnatal day 0.5 (P0.5) pups with the indicated genotypes. (B and C) The endogenous PTHrP was replaced by exogenous PTHrP expressed from a transgene driven by a Col2 promoter (exo-PTHrP). (C) The PPR was ablated in a mosaic fashion in this backgound (exo-PTHrP:mosaic). (D and E) Low-power views of B and C. The exo-PTHrP:mosaic tibia showed a more mature appearance than the exo-PTHrP tibia with elongation of the bone (vertical lines), presence of hematopoietic marrow, and mineralized bone formation (arrowhead). F and G are magnified views of boxed areas shown in B and C, respectively. Column formation was induced by the mosaic PPR ablation. (H and I) In situ hybridization for Ihh in exo-PTHrP (H) and exo-PTHrP:mosaic (I) growth plates. Mosaic PPR ablation induced Ihh expression in ectopic hypertrophic (arrowhead) and eutopic hypertrophic chondrocytes (double arrowhead).
Hedgehog directly stimulates differentiation of periarticular chondrocytes into columnar chondrocytes. In order to further determine whether Ihh directly acts on periarticular chondrocytes to stimulate differentiation into columnar chondrocytes, we used primary chondrocytes from E18.5 PTHrP-Tg tibiae that comprised almost exclusively cells with characteristics of periarticular chondrocytes (See Figure 4B). We used bagpipe homeobox gene 1 homolog (Bapx1) expression as a marker of a gene preferentially expressed in columnar chondrocytes. Bapx1 is predominantly expressed in the columnar region of wild-type mice (Figure 5A) whereas PTHrP-Tg tibiae lack columnar chondrocytes that express Bapx1 (Figure 5B). Primary chondrocytes from PTHrP-Tg embryos were treated with recombinant sonic hedgehog (Shh) to stimulate hedgehog signaling in these cells in vitro. Treatment with Shh significantly increased Bapx1 expression (Figure 5C), suggesting a direct stimulatory effect of hedgehog signaling on differentiation of periarticular chondrocytes. A similar result was also obtained in an in vivo model in which Ihh was overexpressed in the growth plate overexpressing PTHrP. Compared with the PTHrP-Tg littermate, which had few Bapx1-expressing cells, the Ihh-Bg:PTHrP-Tg compound mouse showed the appearance of cells expressing Bapx1 (Figure 5B).
Figure 5Hedgehog overactivation in periarticular chondrocytes stimulates differentiation into columnar chondrocytes. (A) Bapx1 is expressed predominantly in columnar chondrocytes. H&E staining and in situ hybridization for Bapx1 were performed on sections of the proximal tibial growth plate of an E16.5 wild-type mouse embryo. (B) Overexpression of Ihh in PTHrP-expressing cartilage induces Bapx1-expressing cells. In contrast, the PTHrP-Tg tibia, comprising almost exclusively periarticular chondrocytes, lacks the Bapx1 domain. In situ hybridization was performed on distal tibiae of E16.5 embryos. Dotted lines indicate bone contours. (C) Induction of Bapx1 after Shh treatment. Chondrocytes were isolated from E18.5 PTHrP-Tg tibiae, comprising almost exclusively periarticular chondrocytes, dispersed, and cultured in vitro with or without stimulation with 1 mg/ml recombinant Shh for indicated periods. The expression of Bapx1 mRNA and Gapdh mRNA was assessed by quantitative RT-PCR. The expression of Bapx1 mRNA, a marker for columnar chondrocytes, was significantly increased after stimulation with recombinant Shh, whereas Gapdh mRNA was unchanged. The expression level of Bapx1 mRNA was normalized to that of Gapdh mRNA. Expression levels are shown as relative to mean values of control (100%). White bars and gray bars represent control and Shh-treated cells, respectively. Standard deviations are shown. n = 3 in each group; *P < 0.01.
Ihh increases the length of the columnar region independently of PTHrP. To directly assess the role of Ihh in regulating the length of the columnar region separately from PTHrP action, we overexpressed Ihh in the growth plate in which PTHrP signaling was maintained so that it remained constant. To achieve Ihh-independent PTHrP signaling in chondrocytes, caPPR was expressed under the control of a Col2 promoter in the PTHrP-null background (caPPR:PTHrP–/–/constant growth plate PPR signaling [caPPR:PTHrP–/–/c-PPR]). In this setting, PPR signaling is maintained solely by the basal activity of the caPPR transgene. The growth plate of the distal tibia of c-PPR mice at E17.5 showed relatively normal appearance except for mild shortening of the columnar region and delayed blood vessel invasion (Figure 6A). Overexpression of Ihh increased the length of the columnar region even in this background with constant PPR signaling (c-PPR:Ihh-Bg) (Figure 5, A and B). Since the expression level of caPPR was too low to enable direct detection by in situ hybridization (data not shown), we examined the expression level of the endogenous Col2 mRNA (Col2/Col2a1, MGI), which was expected to correlate with the caPPR expression level. We found comparable levels of Col2 expression in c-PPR and c-PPR:Ihh-Bg mice (Figure 6B). Elongation of the columnar region was confirmed by the expansion of the domain expressing Bapx1, a gene encoding a transcription factor expressed in columnar cells (Figure 6B). Thus, independently of PTHrP, Ihh regulates the length of the columnar region.
Figure 6Growth plate elongation was induced by Ihh overexpression without upregulation of PTHrP. PTHrP–/– mice were rescued by transgenic expression of caPPR to achieve Ihh-independent PPR signaling (c-PPR). Ihh was overexpressed in the growth plate with constantly maintained PPR signaling (c-PPR:Ihh-Bg). (A) H&E-stained growth plates of the distal tibia of caPPR:PTHrP–/– (c-PPR) and caPPR:PTHrP–/–:GAL4:UAS-Ihh (c-PPR:Ihh-Bg). Vertical lines represent the periarticular and columnar regions. (B) Alcian blue staining and in situ hybridization assay. Col2 expression was comparable between c-PPR and c-PPR:Ihh-Bg mice. Expansion of the columnar region in c-PPR:Ihh-Bg mice was confirmed by Bapx1 expression. Brackets indicate the proliferating zone (the periarticular and columnar regions). Dashed lines indicate bone contours.
-
Discussion
In the growth plate, chondrocytes go through sequential differentiation steps. Mechanisms regulating each step are poorly understood. Differentiation of round periarticular chondrocytes into flat, column-forming columnar chondrocytes is an important step in the longitudinal growth of bones since column formation provides a clear asymmetry in the bone that allows increase in the length of the bone without proportional increase in width. Further, chondrocytes in the columnar region generally proliferate more vigorously than periarticular chondrocytes. In the present study, we provide evidence that Ihh regulates this step in a PTHrP-independent fashion.
Changes in levels of Ihh action in multiple models have demonstrated the role of Ihh in determining the length of the growth plate. First, expression of Ihh abnormally close to the periarticular region led to growth plate elongation in previously reported chimeric and mosaic growth plates (13, 14). This elongation is dependent on the ectopic expression of Ihh in these models; the expansion did not occur with removal of ectopic Ihh expression in chimeric growth plates comprising wild-type and PPR–/–:Ihh–/– double null cells (12) or in the growth plate with double mosaic ablation of the PPR and Ihh genes in the present study. Second, we observed that Ihh heterozygosity reduced the degree of growth plate elongation in mice with mosaic PPR ablation. This effect of Ihh heterozygosity was also observed in PPR-hypomorphic mice (14), in which Ihh expression was also increased (data not shown). This is consistent with our previously proposed model that upregulation of Ihh action contributed to the growth plate phenotypes in these mice. Third, when Ihh was overexpressed, growth plate elongation occurred. These data demonstrate the causative role of the upregulated Ihh action in the growth plate expansion in these mice.
We previously showed a correlation between Ihh action in periarticular chondrocytes and the rate of differentiation of periarticular to columnar chondrocytes in several mouse models with PPR abnormalities (14). The acceleration of differentiation of periarticular to columnar chondrocytes was deduced from the finding that, despite an increase in BrdU-labeling ratio in the periarticular region, the size of the periarticular region was smaller in these mice. We hoped to see these changes partially reversed by Ihh heterozygosity. As expected, Ihh heterozygosity in these models with upregulation of Ihh action had dramatic effects on the length of the growth plate. Because we could not detect changes in BrdU labeling associated with Ihh heterozygosity, we could not determine whether there were changes in periarticular chondrocyte differentiation. In Ihh-Bg mice overexpressing Ihh in chondrocytes, however, we were able to detect the contrasting changes in the BrdU-labeling ratio and the number of periarticular chondrocytes at multiple developmental stages. Ihh overexpression caused a decrease in cell number despite an increase in proliferation in the periarticular region whereas it caused an increase in cell number without an increase in proliferation in the columnar region. This strongly suggests that Ihh overexpression stimulates periarticular chondrocyte differentiation.
The relationship between hedgehog signaling and cell proliferation has been shown in various organisms in various contexts; hedgehog signaling promotes cell division and cell growth directly through N-myc (16, 17), cyclin D and E expression (18, 19), and secondary molecules such as Wnts (20, 21) and IGF2 (22). In the mouse growth plate, increased hedgehog signaling was associated with increases in chondrocyte proliferation in zone 1 (corresponding to the periarticular region) and an increase in cyclin D expression (9). However, in our models, increases in Ihh action are associated not only with increases in proliferation of periarticular chondrocytes but also with decreases in the size of the periarticular region. This suggests that stimulation of periarticular into columnar differentiation in these models is not a mere passive consequence of the increase in proliferation of the periarticular chondrocytes but is independent of the proliferative effect of Ihh.
Hedgehog overactivation in the context of PTHrP overexpression stimulated expression of BapX1, used as a marker for columnar chondrocytes, in vivo and in vitro. Since tibiae of PTHrP-Tg mice at the ages used in the present study are comprised almost solely of periarticular chondrocytes and devoid of both columnar chondrocytes and hypertrophic chondrocytes, this result is consistent with the hypothesis that hedgehog signaling in periarticular chondrocytes directly stimulates their differentiation into columnar chondrocytes. This is also consistent with the finding from our previous study of cartilage-specific PPR ablation, in which increased Ihh action in periarticular chondrocytes was associated with stimulation of chondrocyte differentiation even though few columnar chondrocytes were found in these bones (14). These results, taken together, suggest that Ihh can act directly on periarticular chondrocytes to accelerate their further differentiation.
The precise mechanisms whereby chondrocytes form stacks of flat cells in orderly arrays are incompletely understood. Studies of mice missing β1 integrin in chondrocytes have shown that this integrin is required for the formation of flat chondrocytes but also that such chondrocytes still exhibit features that distinguish them from periarticular chondrocytes (23). This finding demonstrates that changes in the cell attachment machinery are included in the multiple changes needed for functional maturation of columnar chondrocytes.
The length of the columnar region is determined by the rate of differentiation of periarticular into columnar chondrocytes (entrance into the pool of columnar chondrocytes), the rate of columnar cell proliferation, and the rate of differentiation of columnar to hypertrophic chondrocytes (exit from the pool of columnar chondrocytes). The latter process appears to be regulated by multiple signaling molecules, including PTHrP. PTHrP, regulated by Ihh, is crucial for normal chondrocyte differentiation. PTHrP-null mice exhibit premature terminal differentiation (1). Further, PPR-null and PPR-hypomorphic mice show acceleration of hypertrophic differentiation (2, 14). In contrast, overexpression of PTHrP or a constitutively active PPR suppresses the initial differentiation of hypertrophic chondrocytes (3, 4) (Figure 4). These findings suggest that PTHrP signaling regulates differentiation of columnar to hypertrophic chondrocytes in late fetal and perinatal growth plates, thereby regulating columnar chondrocyte mass. In this model, the length of the columnar region is determined by the stimulatory effect of Ihh on periarticular chondrocyte differentiation and the inhibitory effect of PTHrP on hypertrophic differentiation. However, the findings that PTHrP heterozygosity had only modest effects on growth plates with upregulated Ihh action and that the growth plate elongation caused by Ihh overexpression did not require upregulation of PTHrP suggest that Ihh substantially regulates the length of the columnar region independently of PTHrP.
From these observations, we propose that Ihh directly acts on periarticular chondrocytes to stimulate their differentiation to columnar chondrocytes to regulate the length of late fetal growth plates independently of PTHrP. PTHrP, on the other hand, is required for prevention of premature differentiation of chondrocytes into hypertrophic chondrocytes and thereby suppresses premature expression of Ihh (Figure 7).
Figure 7Model for the regulation of growth plate chondrocyte differentiation by Ihh and PTHrP at distinct steps. In addition to regulating PTHrP expression and chondrocyte proliferation, Ihh directly acts on periarticular chondrocytes to stimulate their differentiation to columnar chondrocytes (a), which regulates the length of the columnar region independently of PTHrP. PTHrP, on the other hand, is required for prevention of premature differentiation of chondrocytes into prehypertrophic and hypertrophic chondrocytes, and thereby suppresses premature expression of Ihh (b). P → C, differentiation of periarticular chondrocytes into columnar chondrocytes; C → H, differentiation of columnar chondrocytes into hypertrophic chondrocytes.
Copyright © 2024 American Society for Clinical Investigation
ISSN: 0021-9738 (print), 1558-8238 (online)