Cell and Tissue Research

, Volume 328, Issue 1, pp 137–151

Contributions of matrix metalloproteinases toward Meckel’s cartilage resorption in mice: immunohistochemical studies, including comparisons with developing endochondral bones

Authors

    • Department of Oral Anatomy, School of DentistryHealth Sciences University of Hokkaido
  • Yoichiro Hosokawa
    • Department of Dental Radiation, School of DentistryHealth Sciences University of Hokkaido
  • Eichi Tsuruga
    • Department of Oral Anatomy, School of DentistryHealth Sciences University of Hokkaido
  • Kazuharu Irie
    • Department of Oral Anatomy, School of DentistryHealth Sciences University of Hokkaido
  • Masanori Nakamura
    • Department of Oral AnatomyShowa University School of Dentistry
  • Toshihiko Yajima
    • Department of Oral Anatomy, School of DentistryHealth Sciences University of Hokkaido
Regular Article

DOI: 10.1007/s00441-006-0329-7

Cite this article as:
Sakakura, Y., Hosokawa, Y., Tsuruga, E. et al. Cell Tissue Res (2007) 328: 137. doi:10.1007/s00441-006-0329-7
  • 183 Views

Abstract

The middle portion of Meckel’s cartilage (one of four portions that disappear with unique fate) degrades via hypertrophy and the cell death of chondrocytes and via the resorption of cartilage by chondroclasts. We have examined the immunolocalization of matrix metalloproteinase-2 (MMP-2), MMP-9, MMP-13, and MMP-14 (members of the MMP activation cascade) and galectin-3 (an endogenous substrate for MMP-9 and an anti-apoptotic factor) during resorption of Meckel’s cartilage in embryonic mice and have compared the results with those of developing endochondral bones in hind limbs. MMP immunoreactivity, except for MMP-2, is present in nearly all chondrocytes in the middle portion of Meckel’s cartilage. On embryonic day 15 (E15), faint MMP-2-immunoreactive and intense MMP-13-immunoreactive signals occur in the periosteal bone matrix deposited by periosteal osteoblasts on the lateral surface, whereas MMP-9 and MMP-14 are immunolocalized in the peripheral chondrocytes of Meckel’s cartilage. The activation cascade of MMPs by face-to-face cross-talk between cells may thus contribute to the initiation of Meckel’s cartilage degradation. On E16, immunopositive signaling for MMP-13 is detectable in the ruffled border of chondroclasts at the resorption front, whereas immunostaining for galectin-3 is present at all stages of chondrocyte differentiation, especially in hypertrophic chondrocytes adjacent to chondroclasts. Galectin-3-positive hypertrophic chondrocytes may therefore coordinate the resorption of calcified cartilage through cell-to-cell contact with chondroclasts. In metatarsal specimens from E16, MMPs are detected in osteoblasts, young osteocytes, and the bone matrix of the periosteal envelope, whereas galectin-3 immunoreactivity is intense in young periosteal osteocytes. In addition, intense MMP-9 and MMP-14 immunostaining has been preferentially found in pre-hypertrophic chondrocytes, although galectin-3 immunoreactivity markedly decreases in hypertrophic chondrocytes. These results indicate that the degradation of Meckel’s cartilage involves an activation cascade of MMPs that differs from that in endochondral bone formation.

Keywords

Matrix metalloproteinasesGalectin-3Meckel’s cartilageEndochondral ossificationImmunohistochemistryMouse (ddY strain)

Introduction

The development of endochondral bones requires the differentiation, maturation, and cell death of chondrocytes, the replacement of cartilage with bone, and the remodeling of bone matrix by osteoblasts and osteoclasts. In these processes, extracellular matrix (ECM) molecules change their structure and composition through a proteolytic process. Members of the matrix metalloproteinase (MMP) family have been implicated in ECM degradation (Vu and Werb 2000). Many studies of mice deficient in MMPs have demonstrated that endochondral bone formation is spatiotemporally regulated by the expression of MMP genes, translation of protein, and activation (Vu et al. 1998; Holmbeck et al. 1999; Engsig et al. 2000; Zhou et al. 2000; Inada et al. 2004; Stickens et al. 2004).

During the embryonic stage, Meckel’s cartilage provides supporting bars for the first branchial arch and then disappears during morphogenesis of the mandible. Early studies have reported the unique fate of Meckel’s cartilage in rodents (Bhaskar et al. 1953; Bernick and Patek 1969; Frommer and Margolies 1971; Savostin-Asling and Asling 1973). In the middle portion, one of four portions in Meckel’s cartilage, membranous bone is formed by deposits of osteoblasts differentiated from perichondral cells on the lateral surface of Meckel’s cartilage closest to the growing incisor teeth. Coincidentally, chondrocytes of Meckel’s cartilage become hypertrophic and die, after which the calcified cartilage is resorbed by osteoclasts/chondroclasts (Savostin-Asling and Asling 1973; Ishizeki et al. 1999), thereby replacing the middle portion of Meckel’s cartilage with bone tissue. A series of these phenomena has been frequently noted in endochondral ossification, with Meckel’s cartilage being thought to contribute directly to mandibular bone formation (Bhaskar et al. 1953; Ishizeki et al. 1999). However, some studies have reported that the middle portion of Meckel’s cartilage disappears without giving rise to ossification, and that degenerating Meckel’s cartilage differs from cartilage undergoing endochondral ossification (Savostin-Asling and Asling 1973; Shimada et al. 2003; Sakakura et al. 2005). Concerning the expression of MMPs and their tissue inhibitors (TIMPs), Chin and Werb (1997) have demonstrated that mature chondrocytes of Meckel’s cartilage do not express MMP-2 (gelatinase A), MMP-9 (gelatinase B), and MMP-13 (collagenase-3) or TIMP-1, TIMP-2, and TIMP-3. In addition, our knowledge of the expression of MMP-1 (collagenase-1), MMP-9, and MMP-13 is fragmentary, because of the different developmental stages and portions of Meckel’s cartilage examined (Mattot et al. 1995; Ishizeki and Nawa 2000; Sasano et al. 2002; Shimo et al. 2004). Thus, whether MMPs contribute to the degradation of Meckel’s cartilage by mechanisms similar to those in endochondral ossification remains unknown.

More than 25 members of the MMP family have been identified so far (Vu and Werb 2000). MMP activity is regulated at the level of transcription, translation, and proenzyme activation and via TIMPs, which bind to MMPs (Nagase and Woessner 1999). TIMP-2 can bind to the hemopexin-like domain of pro-MMP-2 via its C-terminal region, whereas it binds to the active site of MMP-14 (membrane type 1-MMP) via its N-terminal inhibitory domain, implying the formation of an MMP-14/TIMP-2/pro-MMP-2 ternary complex (Fernandez-Catalan et al. 1998; Morgunova et al. 2002). Bound pro-MMP-2 is activated by neighboring TIMP-2-free MMP-14, and active MMP-2 is released into the extracellular space but can also promote the activation of pro-MMP-9 (Bernardo and Fridman 2003; Toth et al. 2003). In addition, MMP-2 participates in the activation of pro-MMP-13 (Knäuper et al. 1996b; Cowell et al. 1998); thus MMP-2, MMP-9, MMP-13, and MMP-14 form part of an activation cascade that involves TIMP-2.

Galectin-3, on the other hand, belongs to the family of β-galactoside-binding proteins and is a soluble protein found intracellularly in the nucleus or cytoplasm and extracellularly at the cell surface or in the matrix, depending on the cell type and differentiation stage (Barondes et al. 1994). Intracellular galectin-3 has been established as an anti-apoptotic factor that contains the anti-death domain characteristic of the Bcl-2 protein family, whereas extracellular galectin-3 negatively regulates chondrocyte differentiation, possibly by acting as an anti-apoptotic matricellular protein that maintains ECM anchorage (Akahani et al. 1997; Ortega et al. 2005). Galectin-3 null mice have a reduced number of hypertrophic chondrocytes and an increase in cells with characteristic signs of cell death in the endochondral bones (Colnot et al. 2001). Interestingly, MMP-2, MMP-9, and MMP-13 are capable of cleaving the amino-terminal domain of galectin-3 at a common site (Ochieng et al. 1994; Guévremont et al. 2004). Recently, Ortega et al. (2005), using mice deficient in MMP-9, have demonstrated that MMP-9 is the dominant MMP cleaving galectin-3 in and around the hypertrophic chondrocytes of endochondral bones. However, whether an activation cascade of MMPs is involved in the degradation of Meckel’s cartilage by cleaving galectin-3 is unknown.

To assess whether an activation cascade of MMPs participates in cartilage degradation, we have examined the immunolocalization of MMP-2, MMP-9, MMP-13, and MMP-14 in Meckel’s cartilage and compared the results with those in the developing hind limbs of embryonic mice. We have also explored whether galectin-3 contributes toward the degradation of cartilage by controlling the terminal differentiation and cell death of chondrocytes.

Materials and methods

Animals

In this study, we used embryonic mice (ddY strain) on days 14–17 of gestation. After mating during a 3-h period (09:00–12:00 h), the presence of a vaginal plug was considered to mark embryonic day zero (E0). At 13:00 h on the designated day, embryos were obtained by laparotomy from the pregnant mice, which were killed by cervical dislocation. The experimentation protocol used in this study was previously approved by the Animal Ethics and Research Committee and was conducted in accordance with the Guidelines for the Care and Use of Laboratory Animals of the Heath Sciences University of Hokkaido.

Whole-mount staining of the embryonic mandible

At E15, mandibles were fixed and stained in a mixture of 95% ethanol and acetic acid (4:1) containing 0.01% Alcian blue 8GX (Merck, Darmstadt, Germany) for 20–24 h. The mandibles were rinsed in ethanol and distilled water and then stained in 0.1% potassium hydroxide with 0.01% Alizarin red S for 24 h. Thereafter, the specimens were rinsed with 0.1% potassium hydroxide until translucent.

Tissue preparation

Mandibles removed on E14–E17 and hind limbs dissected on E16 were fixed in 4% paraformaldehyde (PFA) buffered with 0.1 M sodium cacodylate (pH 7.2) for several hours at 4°C and then re-fixed overnight in PFA. The specimens were decalcified with EDTA for 3 weeks and subsequently embedded in paraffin. The mandibles and limbs were horizontally sectioned on a plane that corresponded to the distal-proximal longitudinal axis of Meckel’s cartilage and the endochondral bones. Serial sections (6 μm thick) were stained with hematoxylin-eosin for the determination of histological features, toluidine blue for evidence of the presence of sulfated proteoglycans, and immunostained with anti-MMP and galectin-3 antibodies, as described below. To determine calcium deposition in undecalcified sections, E15 mandibles fixed with 10% neutral formalin were frozen in OCT compound (Sakura Finetechnical, Tokyo, Japan), whereas E17 mandibles were fixed with PFA buffered with 0.1 M sodium phosphate (pH 7.2), and then embedded in Technovit 8100 (Heraeus Kulzer, Wehrheim, Germany) according to the manufacturer’s protocol. To detect tartrate-resistant acid phosphatase (TRAP) and alkaline phosphatase (ALP) activities, E15 and E17 mandibles were fixed in PFA, embedded in paraffin, or frozen in embedding medium.

Histochemistry of sulfated proteoglycans

Deparaffinized sections of E14–E16 mandibles were re-hydrated in a decreasing ethanol series (100%–70%) and distilled water. The specimens were then stained with 0.05% toluidine blue (pH 2.5) for 5 min, rinsed briefly in distilled water and ethanol, cleared with xylene, and covered in mounting medium. Metachromasia (change from blue to reddish purple) indicated the presence of sulfated proteoglycans in the cartilaginous matrix.

Von Kossa’s reaction

Undecalcified frozen sections of E15 mandibles and resin sections of E17 mandibles were incubated with 5% silver nitrate for 60 min under a fluorescent lamp. Thereafter, silver deposits in the specimens were fixed with 5% sodium thiosulfate for 3 min, and the specimens were rinsed with tap water and counterstained.

Enzyme histochemistry for TRAP and ALP activities

For the detection of TRAP activity, horizontal sections of E15 mandibles were de-paraffinized and stained with reaction mixture consisting of 50 ml 0.1 M acetate buffer (pH 5.0), 5 mg naphthol AS-MX phosphate (Sigma, St Louis, Mo., USA), and 30 mg fast red violet LB salt (Sigma) in the presence of 100 mM L(+)-tartaric acid (sodium free; Wako, Osaka, Japan) for 15 min at 37°C, after which they were counterstained with Delafield’s hematoxylin. For ALP activity, frontal sections of E17 mandibles were cut at a thickness of 20 μm, so that activity could be easily detected in freshly prepared medium after a short incubation time. The specimens were incubated in reaction mixture containing 0.1 M TRIS-HCl (pH 8.5), naphthol AS-MX phosphate, and fast red violet LB for 10 min at 37°C and then counterstained with methyl green.

Immunoperoxidase staining of MMPs and galectin-3

A purified mouse monoclonal anti-MMP-2 antibody (F-68, Fuji Chemical Industries, Takaoka, Japan) was raised against a peptide mapped at the carboxyl terminus of human MMP-2. Affinity-purified goat polyclonal antibodies against MMP-9 (C-20, sc-6840) and MMP-14 (V-16, sc-12366) were purchased from Santa Cruz Biotechnology (Santa Cruz, Calif., USA). Goat anti-MMP-13 polyclonal antibody (AB8120) was obtained from Chemicon International (Temecula, Calif., USA). Supernatants of a hybridoma clone (M3/38) were used as the anti-mouse galectin-3 antibody, whose specificity and characterization were reported in previous studies (Nakamura et al. 1997; Soga et al. 1997).

De-paraffinized sections of E14-E16 mandibles and E16 hind limbs were rinsed in 0.01 M phosphate-buffered saline (PBS, pH 7.2, 3 × 10 min), followed by incubation in 0.01% hydroxyl oxygen/methanol for 30 min to block endogenous peroxidase activity. For the detection of MMP-2, sections were incubated in cold PBS containing 10% bovine serum albumin (BSA) for 30 min and reacted overnight with anti-MMP-2 antibody (diluted 1:50 in PBS) at 4°C. After being rinsed with PBS, the sections were incubated with horseradish-peroxidase-linked sheep anti-mouse Ig, F(ab’)2 fragment (NA 9310, diluted 1:100 in PBS, Amersham Biosciences, Piscataway, N.J., USA) for 90 min. For the detection of MMP-9, MMP-13, and MMP-14, sections were reacted overnight with their antibodies (diluted 1:50 in 0.1% BSA/PBS) at 4°C and incubated with a Vectastain Elite ABC kit for goat immunoglobulin (Vector Laboratories, Burlingame, Calif., USA) according to the manufacturer’s protocol. For the immunohistochemistry of galectin-3, sections were treated by microwave in 10 mM sodium citrate (pH 6.0) according to the procedure reported by Ortega et al. (2005). The specimens were incubated overnight with galectin-3 antibody (diluted 1:10) at 4°C, and subsequent immunoreaction with the secondary antibody was performed with a Vectastain kit prepared for rat immunoglobulin. Immunoreactivity was visualized with 0.004% 3,3-diaminobenzidine tetrahydrochloride (DAB; Dojindo, Kumamoto, Japan) in 0.01% H2O2 in 0.05 M TRIS-HCl (pH 7.6) for 10 min. After being washed in distilled water, the sections were counterstained with methyl green. To confirm the specificity of the immunostaining, the primary antibody was omitted or substituted with PBS and normal goat or rat serum or with an irrelevant antibody, at the same dilution, with 0.1% BSA/PBS being used for antibody dilution. No MMP immunoreactivity was detected in any of the control sections.

Results

Area of interest in Meckel’s cartilage

Whole-mount staining of E15 mandibles revealed Meckel’s cartilage stained with Alcian blue and mandibular bone stained with Alizarin red S (Fig. 1). The distal portion of Meckel’s cartilage protruded in an anterior (rostral) direction, whereas the middle portion was surrounded by developing bone, and the posterior portion extended straight toward the proximal end to be converted to inner ear ossicles. In an overview of the mandible, Meckel’s cartilage was formed within symmetrical V-shaped rods fused at the base of the distal portion. In this study, we focused on the middle portion of Meckel’s cartilage, viz. the region that would undergo the hypertrophy and death of chondrocytes, plus the resorption of cartilage.
https://static-content.springer.com/image/art%3A10.1007%2Fs00441-006-0329-7/MediaObjects/441_2006_329_Fig1_HTML.gif
Fig. 1

Whole-mount staining of an E15 mandible specimen with toluidine blue and Alizarin red S. The blue-stained Meckel’s cartilage has a symmetrical V-shape. The distal portion (Dist) protrudes in an anterior direction, whereas the posterior portion (Post) extends toward the end of the cartilage. At this stage, the middle portion (Mid) has already become surrounded by Alizarin-red-stained mandibular bone (Mb). Bar 1 mm

Histology of E14–E16 mandibles

In horizontal sections of E14 mandibles, Meckel’s cartilage had a V-shape with a protrusion consisting of immature chondrogenic cells, whereas incisor teeth were seen on both sides of the protrusion in the distal portion (Fig. 2a). By E15, the embryonic mandible had grown in an anterior direction, whereas incisor teeth at the bell-stage had grown in a proximal direction (Fig. 2b). Meckel’s cartilage increased in width and was slightly curved at this stage. In E16 mandibles, incisor tooth growth was advanced in the antero-posterior direction with development of the mandible, and Meckel’s cartilage was bent in the middle portion (Fig. 2c). In E14 mandibles, Meckel’s cartilage consisted of mature chondrocytes in the middle portion and was surrounded by perichondrium (Fig. 2d). Nearly all of the chondrocytes became hypertrophied on E15, although early hypertrophic chondrocytes were found at the outermost periphery on the lateral side (Fig. 2e). On the lateral surface, osteoblasts were found differentiated from perichondral cells, and a thin layer of appositional bone matrix was deposited under the cells. Eosinophilic multinuclear cells were often observed on the lateral side. In contrast, on the medial side, osteoblasts covering the cartilage participated in the formation of bone that expanded from the ossification center of the mandible, whereas multinuclear cells were not found on this side. In E16 mandibles, the cartilage had degraded, and osteoblasts and bone matrix had disappeared on the lateral side (Fig. 2f). At this stage, a number of eosinophilic multinuclear cells were seen in the resorption area of Meckel’s cartilage.
https://static-content.springer.com/image/art%3A10.1007%2Fs00441-006-0329-7/MediaObjects/441_2006_329_Fig2_HTML.gif
Fig. 2

Histological examination of horizontal sections of E14–E16 mandibles with hematoxylin and eosin. a E14 mandible shows fusion of both Meckel’s cartilage bars at the distal portion, whereas cap-stage incisor teeth are seen on both sides of the distal protrusion. b In the E15 mandible, the incisor teeth are at the bell-stage, and hypertrophy of chondrocytes is found in the middle portion facing the incisor teeth. c In the E16 mandible, Meckel’s cartilage has eroded on the lateral side facing the incisor teeth in the middle portion. d In the middle portion, chondrocytes are mature and the perichondrium (Pc) surrounds Meckel’s cartilage of the E14 mandible. e On E15, hypertrophic chondrocytes occupy the central area of the middle portion, whereas the peripheral chondrocytes on the medial side remain mature. Perichondral cells differentiated into osteoblasts (arrow) and a thin layer of appositional bone matrix are found on the lateral surface. At this stage, multinuclear cells are often observed on the lateral surface (arrowhead). f A number of eosinophilic multinuclear cells are seen in the resorption front of E16 cartilage (arrowheads). Bars 500 μm (a–c), 100 μm (d–f)

Histochemistry of sulfated proteoglycans, von Kossa’s reaction, TRAP activity, and ALP activity

In the horizontal sections of E14 mandibles, toluidine-blue-stained cartilaginous matrix was increased in the middle portion and consisted of mature chondrocytes, as shown by metachromasia (Fig. 3a). On E15, hypertrophic chondrocytes occupied the middle portion and were clearly surrounded by stained cartilaginous matrix (Fig. 3b). In E16 mandibles, the hypertrophic region was expanded in the antero-posterior direction, whereas the stained matrix surrounding the hypertrophic chondrocytes had become thin and had disappeared from the resorption front on the lateral side nearest to the growing incisor tooth (Fig. 3c). By E15, the stained matrix clearly outlined hypertrophic chondrocytes and showed a slightly irregular periphery of the lateral surface (Fig. 3d). The intensity of toluidine blue staining particularly decreased in the outermost layer of the cartilaginous matrix. At this stage, von Kossa’s reaction showed calcium deposition in bone tissue that expanded from the ossification center of the mandible, whereas the cartilaginous matrix and appositional bone matrix covering the cartilage were not yet calcified (Fig. 3e). Concurrently, TRAP-positive cells frequently appeared in the area in which calcium deposition was observed by von Kossa’s reaction; however, these cells had not yet begun to participate in the degeneration of Meckel’s cartilage at this stage (Fig. 3f).
https://static-content.springer.com/image/art%3A10.1007%2Fs00441-006-0329-7/MediaObjects/441_2006_329_Fig3_HTML.gif
Fig. 3

Histochemical demonstration of sulfated proteoglycan by toluidine blue staining (a–d), calcium deposition by von Kossa’s reaction (e), and TRAP activity (f). a On E14, the cartilaginous matrix shows clear metachromasia with the maturation of chondrocytes. b At E15, Meckel’s cartilage has become thick with the hypertrophy of chondrocytes, and abundant cartilaginous matrix surrounds the hypertrophic chondrocytes. c On E16, the cartilaginous matrix is emaciated, and opened chondrocytic lacunae are found in the resorption area. d At a higher magnification of Meckel’s cartilage on E15, the peripheral cartilaginous matrix shows a decreased intensity of metachromasia and an irregular contour with a slightly concave outline (arrowhead). e As shown by von Kossa’s reaction, calcium deposits are seen in the bone matrix that extends from the ossification center of the mandible, but not in the cartilaginous matrix and appositional bone matrix (arrowhead) on the lateral surface of Meckel’s cartilage at E15. f At this stage, TRAP-positive cells are often observed on the lateral side, whereas resorption of cartilage by the cells is never seen. Bars 100 μm (a–c), 100 μm (d–f)

In the frontal sections of Meckel’s cartilage stained with von Kossa’s reaction at E17, remnants of calcified cartilaginous matrix were disconnected from membranous bone newly developed from the mandible (Fig. 4a). Only part of the cartilage matrix surrounding the chondrocytes had calcified, as shown by the reaction products. At this stage, intense ALP activity was detected in osteoblasts that actively formed bone matrix, whereas little or no activity was observed in cells within the resorption area of Meckel’s cartilage (Fig. 4b).
https://static-content.springer.com/image/art%3A10.1007%2Fs00441-006-0329-7/MediaObjects/441_2006_329_Fig4_HTML.jpg
Fig. 4

Histochemical examination of calcium deposition and ALP activity in frontal sections of Meckel’s cartilage at E17. a Remnants of calcified cartilaginous matrix (arrows) are disconnected from membranous bone that has newly expanded from the mandible (asterisk). b ALP activity (red) is detected in osteoblasts that are actively forming bone matrix, whereas little or none is seen in the resorption area (asterisk). Bars 50 μm (a, b)

Immunolocalization of MMPs and galectin-3 in the middle portion of Meckel’s cartilage

Based on our histological and histochemical observations, we investigated the localization of MMPs in the middle portion of Meckel’s cartilage in E15 and E16 mandibles. On E15, immunolocalization of MMP-2 was observed in the bone matrix on the lateral surface of the middle portion (Fig. 5a). In E16 mandibles, faint immunoreactivity was seen in the opened chondrocytic lacunae and in connective tissue, but not in hypertrophic chondrocytes surrounded with cartilaginous matrix (Fig. 5b). At E15, intense MMP-9 immunoreactivity was localized in hypertrophic chondrocytes, especially in cells positioned at the lateral periphery of Meckel’s cartilage (Fig. 5c). In addition, immunoreactivity was scant in the bone matrix on the lateral surface. By E16, intense immunostaining for MMP-9 was predominantly found in ECM of the resorption area and in chondroclasts that were actively resorbing cartilaginous matrix (Fig. 5d). Hypertrophic chondrocytes exhibited moderate MMP-9 immunoreactivity. Furthermore, intense MMP-13 immunoreactivity was observed in hypertrophic chondrocytes and bone matrix on the lateral side in E15 mandibles (Fig. 5e). In E16 mandibles, MMP-13 immunoreactivity was localized in the resorption area on the lateral side of Meckel’s cartilage (Fig. 5f). On E15, immunolocalization for MMP-14 was found in chondrocytes located at the lateral periphery of the cartilage, whereas it was scarcely detected in periosteal osteoblasts and bone matrix (Fig. 5g). Moderate MMP-14 immunoreactivity was observed in the resorption area in E16 mandibles (Fig. 5h), whereas higher magnification showed chondroclasts that resorbed cartilaginous matrix and osteoclasts that were attached to the edge of bone trabeculae (not shown).
https://static-content.springer.com/image/art%3A10.1007%2Fs00441-006-0329-7/MediaObjects/441_2006_329_Fig5_HTML.gif
Fig. 5

MMP-2 (a, b), MMP-9 (c, d), MMP-13 (e, f), and MMP-14 (g, h) immunoreactivity on the lateral side of the middle portion in E15 mandibles (a, c, e, g), and in the resorption area of E16 mandibles (b, d, f, h). a MMP-2 immunoreactivity is faint in the appositional bone matrix covering the lateral surface of Meckel’s cartilage (arrowhead). b Although the immunoreactivity is faint, opened chondrocytic lacunae (arrowheads) and connective tissue are positive in the resorption area, whereas the hypertrophic chondrocytes (arrows) are negative. c Intense MMP-9 immunoreactivity is found in the peripheral chondrocytes of Meckel’s cartilage. d The immunoreactivity is intense in the resorption area and especially in the chondroclasts (arrowhead), whereas hypertrophic chondrocytes show moderate reactivity. e MMP-13 immunoreactivity is intense in hypertrophic chondrocytes and in appositional bone matrix covering the lateral surface of Meckel’s cartilage. f On E16, the resorption area exhibits intense MMP-13 immunoreactivity. g Intense MMP-14 immunoreactivity is found in the peripheral chondrocytes, whereas scant immunoreactivity is seen in the periosteal osteoblasts on the lateral side of Meckel’s cartilage. h Moderate MMP-14 immunoreactivity is observed in the resorption area and some of the hypertrophic chondrocytes are intense. Bar 100 μm

Of the MMPs examined in this study, MMP-13 showed prominent immunolocalization during the degeneration of Meckel’s cartilage. In E15 mandibles, intense MMP-13 immunoreactivity was located in the bone matrix located under osteoblasts that had differentiated from perichondral cells (Fig. 6a). Immunoreactivity was found in some multinuclear cells, probably the ruffled border of chondroclasts, adjacent to chondrocytes in E16 mandibles (Fig. 6b). In developing mandibular bone, MMP-13 immunostaining was also observed in bone matrix under some osteoblasts and in the ruffled borders of osteoclasts attached to the edge of bone trabeculae (not shown).
https://static-content.springer.com/image/art%3A10.1007%2Fs00441-006-0329-7/MediaObjects/441_2006_329_Fig6_HTML.gif
Fig. 6

Higher magnification of MMP-13 and galectin-3 immunoreactivity in E15 (a, c) and E16 (b, d) mandibles. a At this stage, when bone matrix deposits are found, MMP-13 immunoreactivity has become localized under the osteoblasts differentiated from perichondral cells (arrowhead). b Intense MMP-13 immunoreactivity is observed in a portion of the chondroclasts, probably the ruffled border (arrow). Note the chondroclast (arrowheads). c Galectin-3 immunoreactivity is detected in osteoclasts (arrowhead) and in chondrocytes, but not in periosteal osteoblasts, on the lateral surface of Meckel’s cartilage. d On E16, a number of chondroclasts (arrowheads) show intense immunoreactivity for galectin-3. Further, hypertrophic chondrocytes have galectin-3-immunopositive vacuolated cytoplasm and round nuclei (arrows). A vacuolated chondrocyte (asterisk) is often located closest to the chondroclast. Bars 10 μm (a, b), 50 μm (c, d)

Intense immunoreactivity for galectin-3 was found at all stages of chondrocyte differentiation. In E15 mandibles, periosteal osteoblasts covering the lateral surface of Meckel’s cartilage were negative for immunoreactivity, whereas chondrocytes displayed intense signals (Fig. 6c). In addition, moderate immunoreactivity for galectin-3 was seen in the multinuclear osteoclasts closest to the periosteal osteoblasts on the lateral surface. On E16, a number of chondroclasts located in the resorption front were immunopositive for galectin-3 (Fig. 6d). At this stage, nearly all of the hypertrophic chondrocytes had vacuolated cytoplasm and round nuclei, and the cytoplasm presented intense immunoreactivity for galectin-3. Some vacuolated chondrocytes were located close to the chondroclasts.

Immunolocalization of MMPs and galectin-3 in developing endochondral bones

For a comparison of MMP localization in the middle portion of Meckel’s cartilage, we examined the immunolocalization of periosteal envelope and diaphyseal chondrocytes in the developing metatarsals of E16 hind limbs. MMP-2 immunoreactivity was preferentially restricted to loose connective tissue in the hind limbs (Fig. 7a). The periosteal envelope covering the diaphysis exhibited a distinct boundary line with the area of immunoreactivity (Fig. 7b). Faint immunostaining for MMP-2 was often found in pre-hypertrophic chondrocytes and young osteocytes of the periosteal bone envelope. Furthermore, intense MMP-9 immunoreactivity was predominantly detected in pre-hypertrophic chondrocytes; however, it greatly decreased in hypertrophic chondrocytes in the terminal stage of cytodifferentiation (Fig. 7c). Immunoreactivity was moderate in the osteoblasts and young osteocytes of the periosteal envelope, whereas faint immunostaining was found in the connective tissue (Fig. 7d). Moderate MMP-13 immunoreactivity was observed in hypertrophic chondrocytes (Fig. 7e). When observed at higher magnification, intense immunoreactivity was preferentially seen in osteoblasts and young osteocytes of the periosteal bone envelope at the diaphysis, whereas only faint immunoreactivity was found in pre-hypertrophic chondrocytes (Fig. 7f). Immunolocalization of MMP-14 resembled that of MMP-9 and was found in both extremities of the hypertrophic core (Fig. 7g). In addition, some periosteal osteoblasts, young osteocytes, and the surrounding matrix were faintly immunopositive for MMP-14 (Fig. 7h).
https://static-content.springer.com/image/art%3A10.1007%2Fs00441-006-0329-7/MediaObjects/441_2006_329_Fig7_HTML.gif
Fig. 7

MMP-2 (a, b), MMP-9 (c, d), MMP-13 (e, f), and MMP-14 (g, h) immunoreactivity in developing metatarsals of E16 hind limb. The cartilaginous anlage shows hypertrophic chondrocytes and the periosteal envelope in the diaphysis. a MMP-2 immunoreactivity is preferentially seen in the connective tissue of the hind limb. b At higher magnification, faint immunoreactivity is found in some of the pre-hypertrophic chondrocytes (arrow) and young osteocytes (arrowheads) of the periosteal bone envelope. c Intense MMP-9 immunoreactivity is seen at both extremities of the hypertrophic zone, whereas the hypertrophic cells show moderate reactivity. d Pre-hypertrophic chondrocytes are intensely immunopositive, and osteoblasts and young osteocytes (arrowheads) in the periosteal envelope are moderately immunostained. e At lower magnification, moderate MMP13 immunoreactivity is observed in the hypertrophic chondrocyte zone. f Immunoreactivity is intense in the periosteal osteoblasts and young osteocytes but faint in pre-hypertrophic chondrocytes. g Intense MMP-14 immunoreactivity is preferentially detected in the pre-hypertrophic chondrocytes and is also observed in proliferating chondrocytes. h In the periosteal envelope, faint immunoreactivity is seen in the osteoblasts, young osteocytes, and bone matrix. Bars 100 μm (a, c, e, g), 50 μm (b, d, f, h)

Galectin-3 immunoreactivity was extremely intense in the epidermal cells of E16 hind limbs (Fig. 8a) and was also found in the periosteal envelope at the diaphysis of cartilaginous anlage and in proliferating cells located at both metatarsal extremities. At higher magnification, galectin-3 immunoreactivity was detected in chondrocytes at all stages of differentiation, especially in the cytoplasm of pre-hypertrophic chondrocytes, whereas it decreased in hypertrophic chondrocytes (Fig. 8b). Furthermore, intense immunoreactivity was prominent in young osteocytes of the periosteal bone envelope. The epiphyseal cartilage of femurs showed immunoreactive patterns for MMPs and galectin-3 similar to those in metatarsals (see Table 1).
https://static-content.springer.com/image/art%3A10.1007%2Fs00441-006-0329-7/MediaObjects/441_2006_329_Fig8_HTML.gif
Fig. 8

Galectin-3 immunoreactivity in the cartilaginous primordia of E16 metatarsals. a Immunoreactivity is predominantly localized in the periosteal envelope. b Immunoreactivity is found in chondrocytes at all stages of differentiation and is higher in the cytoplasm of pre-hypertrophic cells (arrows). Note that young osteocytes (arrowheads) embedded in the periosteal bone matrix exhibit intense immunoreactivity. Bars 100 μm (a), 50 μm (b)

Table 1

Summary of profiles for immunoreactivity against MMP2, MMP-9, MMP-13, and MMP-14 ( not detectable, + weak or faint, ++ moderate, +++ intense)

Mandible

MMP-2

MMP-9

MMP-13

MMP-14

Limb

MMP-2

MMP-9

MMP-13

MMP-14

E15

    

Metatarsal

    

Periosteal

    

Periosteal

    

 Osteoblastsa

+

 Osteoblasts

++

+++

+

     

 Osteocytes

+

++

+++

+

 Matrixa

+

+

+++

 Matrix

+

±

±

+

Chondrocytes

    

Chondrocytes

    

 Mature

++

++

++

 Mature

±

+

 Pre-hypertrophic

++

++

++

 Pre-hypertrophic

±

+++

+

+++

 Hypertrophic

+++

+++

+/++

 Hypertrophic

+/++

++

++

E16

    

Femur

    

Chondrocytes

    

Chondrocytes

    

 Mature

++

++

++

 Mature

+

±

+

 Pre-hypertrophic

++

+++

++

 Pre-hypertrophic

++

+

++

 Hypertrophic

++

+++

++

 Hypertrophic

+

±

+

Opened lacunae

+

++

++

++

Opened lacuna wall

+/++

+/++

Chondroclasts

+++

+++

±

Trabecular bone

    
     

 Osteoblastsb

+

+

++

+

     

 Osteocytesb

+

+

++

±

     

 Osteoclastsb

++

±

+

aOsteoblasts and matrix derived from perichondral cells on the lateral surface of Meckel’s cartilage

bCells seen in trabecular bone of E16 femurs

Discussion

We have studied Meckel’s cartilage resorption by histological examination, mainly to determine the immunolocalization of MMP, and have compared the results with those in developing hind limb bones in mice. The difference in the immunoreactivity of MMP between the middle portion of Meckel’s cartilage and endochondral bones is summarized in Table 1.

Role of MMPs in resorption in Meckel’s cartilage

As shown by our histochemical data, the middle portion of Meckel’s cartilage exhibits a partial decrease in sulfated proteoglycan and an irregular contour at the periphery of uncalcified cartilage on the lateral surface of E15 mandibles. These results indicate that uncalcified cartilage in this restricted area disintegrates via a proteolytic process prior to the calcification of the cartilaginous matrix and the subsequent resorption by osteoclasts/chondroclasts.

Faint MMP-2 immunoreactivity has been detected in the bone matrix under periosteal osteoblasts, similar to reports showing that signals for MMP-2 mRNA are intense in the perichondrium around Meckel’s cartilage in mice (Reponen et al. 1992; Chin and Werb 1997; Werb and Chin 1998). On the other hand, MMP-13 immunoreactivity is preferentially localized in bone matrix deposited under periosteal osteoblasts; this indicates that MMP-13 expression is restricted to the active phase of morphological change (Hayami et al. 2000; Bae et al. 2003). This expression might also be associated with the multiple functions of MMP-13, including the degradation of types I, II, III, and X collagen and the cartilage proteoglycan aggrecan, and with the activation of MMP-9 (Fosang et al. 1996; Mitchell et al. 1996; Knäuper et al. 1996a, 1997a,b). However, MMP-9 and MMP-14 immunoreactivity is scarcely detectable in the deposited bone matrix and periosteal osteoblasts, whereas proteinases show intense immunolocalization in the peripheral chondrocytes of Meckel’s cartilage. Since TIMP-2 mRNA transcripts have been detected in the perichondrium of Meckel’s cartilage (Chin and Werb 1997), an activation cascade of MMPs by an MMP-14/TIMP-2/pro-MMP-2 ternary complex may be involved in the disintegration of uncalcified bone matrix covering the lateral surface through face-to-face cross-talk between periosteal osteoblasts and peripheral chondrocytes of Meckel’s cartilage (Fig. 9a).
https://static-content.springer.com/image/art%3A10.1007%2Fs00441-006-0329-7/MediaObjects/441_2006_329_Fig9_HTML.gif
Fig. 9

Schematic diagrams of our working hypotheses for the roles of MMPs during the degradation in Meckel’s cartilage and endochondral ossification (RANKL receptor activator of nuclear factor-κB ligand, OPG osteoprotegerin, RANK receptor for RANKL, TGF-β transforming growth factor-β, L-TGF-β latent TGF-β, filled arrows activation of MMPs and TGF-β, open curved arrows release of MMPs, dotted arrows actions of MMPs and TGF-β). a Role of MMPs in the initiation of appositional bone matrix and Meckel’s cartilage degradation. b Role of MMPs in pericellular ECM degradation and resorption of Meckel’s cartilage by chondroclasts. c Role of MMPs in periosteal bone collar formation (thick arrow conversion of osteoblasts to osteocytes). d Role of MMPs in pericellular ECM degradation and chondrocyte cell death in endochondral bone

Trans-membranous MMP-14 bound to peripheral chondrocytes acts on pro-MMP-2 released from periosteal osteoblasts, which in turn activates MMP-9 from chondrocytes and MMP-13 from osteoblasts. MMP-13 efficiently degrades native fibrillar collagens, generating fragments approximately 3/4 and 1/4 the size of intact molecules, whereas MMP-2 and MMP-9 are gelatinases capable of degrading denatured collagen fragments following the cleavage of native fibrillar collagens by other MMPs, such as MMP-13 (Freije et al. 1994; Mackay et al. 1990). During these processes, the MMPs may release and activate a latent form of transforming growth factor-β (TGF-β) from ECM-bound stores (Yu and Stamenkovic 2000; D’Angelo et al. 2001; Dallas et al. 2002; Karsdal et al. 2002). TGF-β might be involved in the differentiation and recruitment of osteoclasts through the regulation of RANK (a receptor for RANKL, which is a receptor activator of nuclear factor-κB ligand), RANKL, and osteoprotegerin (a decoy receptor against RANKL) on the lateral surface of E15 Meckel’s cartilage (Takai et al. 1998; Thirunavukkarasu et al. 2001; Karsdal et al. 2003; Sakakura et al. 2005).

MMPs expressed by osteoclasts and chondroclasts clearly play a role in E16 Meckel’s cartilage. MMP-9 has been shown to be highly restricted to osteoclasts in mandibles and metatarsals during mouse development (Reponen et al. 1994; Blavier and Delaissé 1995; Chin and Werb 1997; Vu et al. 1998; Werb and Chin 1998). MMP-14 has been immunolocalized in osteoclasts and chondroclasts in our previous investigation (Irie et al. 2001) and in this study (not shown). An interesting finding in this study is the intense immunostaining for a collagenase, MMP-13, in some chondroclasts, probably the ruffled border, adjacent to chondrocytes. An electron-microscopic study of rat proximal tibiae has demonstrated MMP-13 immunoreactivity on the bone surface under the clear zone and ruffled border of resorbing osteoclasts, and in osteocytes adjacent to osteoclasts and bone-lining cells (Nakamura et al. 2004). MMP-13 has also been reported to play a pivotal role in the migration of osteoclasts over MMP-13-positive bone surfaces by removing non-mineralized collagen fibrils from the bone surface prior to osteoclastic bone resorption; the enzyme is derived from osteocytes adjacent to osteoclasts and is translocated onto bone surfaces under osteoclasts through the osteocytic lacunae-canaliculi channel. However, as seen in frontal sections of E17 mandibles in this study (Fig. 4), ALP-positive cells, probably osteoblasts, are not or only rarely detected in the resorption area of Meckel’s cartilage and in the remnants of calcified cartilaginous matrix disconnected from the calcified bone matrix that newly develops from the ossification center of the mandible. Thus, Meckel’s cartilage does not participate in mandibular bone formation by endochondral ossification, and MMP-13-positive hypertrophic chondrocytes might act as a substitute for osteoblast lineage cells (Fig. 9b). In addition, since MMP-2 immunoreactivity has been detected in opened chondrocytic lacunae and in connective tissue, MMP-2 might be associated with the activation of other MMPs regulated by MMP-14 from chondroclasts. In addition, activated MMP-2 might act on the activation of MMP-9 secreted by chondroclasts and MMP-13 from vacuolated hypertrophic chondrocytes. Thus, chondrocytes adjacent to chondroclasts appear to be an indispensable partner for the resorption of calcified Meckel’s cartilage by chondroclasts.

Role of MMPs and galectin-3 in chondrocyte differentiation and in cell death in Meckel’s cartilage

In this study, no MMP-2 has been found in Meckel’s cartilage chondrocytes at any stage of differentiation, in agreement with the report of a lack of MMP-2 expression in mature chondrocytes of Meckel’s cartilage (Chin and Werb 1997). MMP-9, MMP-13, and MMP-14 have been co-localized in nearly all chondrocytes with intense and moderate immunoreactivity. Mice deficient in membrane-type MMP-14 display extensive ghosting of Meckel’s cartilage accompanied by complete proteoglycan depletion, numerous empty chondrocyte lacunae, and the cell death of residual chondrocytes (Holmbeck et al. 2003). MMP-14 not only functions as a collagenase capable of cleaving types I, II, and III collagen, but also mediates pericellular proteolysis of the ECM, such as cartilage proteoglycan, fibronectin, vitronectin, and laminin-1 (Ohuchi et al. 1997). In addition, MMP-14 in chondrocytes can activate pro-MMP-13, which in turn cleaves pro-MMP-9 during irreversible cartilage destruction (Dreier et al. 2004). Although gelatinase MMP-9 is capable of cleaving extracellular galectin-3, which is a direct substrate of MMP-9 and an anti-apoptotic factor (Akahani et al. 1997; Colnot et al. 2001; Ortega et al. 2005), it may greatly contribute to chondrocyte maturation and hypertrophy rather than to cell death through the disintegration of the pericellular ECM, including extracellular galectin-3. Therefore, membrane-type MMP-14 might coordinate chondrocyte maturation and differentiation through pericellular ECM degradation and/or via the activation of MMP-13 and MMP-9 (Fig. 9b).

Although chondrocytes are generally accepted to die by apoptosis, a recent study of rabbit growth plates indicates that chondrocyte death does not involve classical apoptosis (Roach and Clarke 2000). Results from galectin-3 null mice strongly suggest the involvement of galectin-3 in chondrocyte differentiation and cell death (Colnot et al. 2001). On the other hand, during the degeneration of Meckel’s cartilage, nuclear DNA fragmentation, an apoptotic phenomenon, is occasionally observed. However, dying chondrocytes with fragmented DNA appear independently of the differentiation stage and localization of chondrocytes, indicating that chondrocyte apoptosis is not a primary prerequisite for the disappearance of Meckel’s cartilage (Harada and Ishizeki 1998). Furthermore, cell-to-cell contact of chondroclasts with intact hypertrophic chondrocytes in the opened chondrocyte lacuna has been reported (Harada and Ishizeki 1998). We have found intense immunoreactivity for galectin-3 in chondrocytes at all stages of differentiation. Moreover, in E16 mandibles, intense immunoreactivity for galectin-3 has been detected even in the vacuolated chondrocytes closest to chondroclasts, implying the continuation of high galectin-3 expression in chondrocytes of Meckel’s cartilage. Therefore, intracellular galectin-3 might permit chondrocytes to survive during differentiation and maturation, and consequently, galectin-3-positive vacuolated hypertrophic chondrocytes appear to coordinate Meckel’s cartilage resorption through cell-to-cell contact with chondroclasts.

Role of MMPs in developing endochondral bones

Unlike the lateral surface of Meckel’s cartilage, the periosteal envelope shows faint immunostaining for MMP-2 and MMP-14 and intense immunoreactivity for MMP-9 and MMP-13 in the diaphysis of developing metatarsals. Several studies have reported the expression of these MMPs in the periosteum of endochondral bones (Blavier and Delaissé 1995; Kinoh et al. 1996; Chin and Werb 1997; Ståhle-Bäckdahl et al. 1997). In particular, MMP-9 in the periosteal envelope appears to play an essential role in the invasion of osteoclasts and endothelial cells from the periosteum into calcified cartilage and has an enhancing effect on focal collagenolysis by MMP-13 (Engsig et al. 2000). Furthermore, TIMP-2 co-localizes with MMP-2 and MMP-14 in the periosteum of endochondral bone in mouse embryos (Kinoh et al. 1996). Notably, MMP-14, a membrane-type MMP and a key proteinase of the activation cascade, controls the conversion of osteoblasts into osteocytes by blocking osteoblast apoptosis through the activation of latent TGF-β (Karsdal et al. 2002). Indeed, we have detected intense galectin-3 immunoreactivity in young osteocytes embedded within the periosteal bone matrix (Fig. 8b). This observation is in accordance with previous reports that galectin-3 expression is significantly increased at the matrix maturation stage of osteoblasts (Aubin et al. 1996; Stock et al. 2003). In addition, MMP-14 appears to contribute actively to the establishment and/or maintenance of a communicating network of osteocytic processes (Holmbeck et al. 2005). Thus, an activation cascade of MMPs might play an important role in periosteal bone collar formation leading to the molding of the bone shaft during endochondral osteogenesis (Fig. 9c). In particular, the expression of MMP-14 in periosteal osteoblasts appears to be indispensable for transdifferentiation into osteocytes, comparable with Meckel’s cartilage.

In the process of chondrocyte hypertrophy, the MMPs examined in this study have been found in chondrocytes of developing metatarsals at various levels of immunoreactivity. Immunostaining for MMP-14 and TIMP-2 exhibits their co-localization in hypertrophic chondrocytes (Haeusler et al. 2005), and thus the regulation of the MMP activation cascade by the MMP-14/TIMP-2/pro-MMP-2 ternary complex appears to be involved in the orderly differentiation and maturation of chondrocytes through pericellular ECM degradation (Fig. 9d). In particular, intense immunoreactivities for MMP-9 and MMP-14 are exclusively co-localized in pre-hypertrophic chondrocytes. Concurrently, galectin-3 immunoreactivity is moderate in pre-hypertrophic chondrocytes, whereas it is markedly decreased in hypertrophic cells. Targeted inactivation of the MMP-9 gene induces a delay in terminal hypertrophic chondrocyte apoptosis, and MMP-9 null mice show an abnormal accumulation of galectin-3 in an increased number of hypertrophic chondrocytes (Vu et al. 1998; Ortega et al. 2005). Furthermore, metatarsals from E16.5 wild-type mice cultured in the presence of full-length recombinant galectin-3 exhibit an embryonic MMP-9 null phenotype with an increased hypertrophic zone (Ortega et al. 2005). On the other hand, MMP-14 not only functions as a collagenase, but also mediates the pericellular proteolysis of non-collagenous ECM and adhesion molecules (Ohuchi et al. 1997). Thus, MMP-9 and MMP-14 strongly expressed in the restricted zone might be involved in the rapid transition from pre-hypertrophic to hypertrophic chondrocytes, and in the induction of cell death through rapid changes in the pericellular microenvironment, such as galectin-3 cleavage and the disintegration of other adhesion molecules (Fig. 9d).

Concluding remarks

Our immunohistochemical results suggest that MMPs contribute to resorption and chondrocyte cell death in Meckel’s cartilage by an activation cascade different from the disintegration of cartilage undergoing endochondral ossification. However, the activities of the MMPs in the restricted region remain to be investigated by using methods such as in situ zymography.

Copyright information

© Springer-Verlag 2006