Archs oral Eiol. Vol. 35, No. 5, pp. 337-346, 1990 Printed in Great Britain. All rights reserved
Copyright
0
0003-9969/90 $3.00 + 0.00 1990 Pergamon Press plc
IMMUNOHISTOCHEMICAL LOCALIZATION OF SPARC (OSTEONECTIN) AND DENATURED COLLAGEN AND THEIR RELATIONSHIP TO REMODELLING IN RAT DENTAL TISSUES J. SALONEN,* C.
DOMENICUCCI,
H. A.
GOLDBERG
and J. SorxKt
Medical Research Council Group in Periodontal Physiology, Faculty of Dentistry, University of Toronto, Ontario, Canada MSS lA8 (Accepted
24 November 1989)
Summary-To study this relationship, specific antibodies were used to determine the distribution of these proteins in mature rat dental tissues. Staining for SPARC with affinity-purified polyclonal antibodies was prominent throughout molar and incisor ligaments, endosteal tissue, dental pulp and muscle. More moderate staining was observed in other soft tissues including the lamina propria of gingiva, whereas the staining of demineralized bone was weak and in dentine was barely detectable. A monoclonal antibody (MBP 322), raised against a denatured form of a small collagenous bone protein, reacted strongly with osteoblastic cells but more moderately with alveolar bone. A strong reaction, indicative of unfolded collagen, was also evident throughout the dental pulp and molar ligament, whereas in the incisor ligament staining was largely restricted to the tooth-related half. Moderate staining with this antibody was also observed in other soft tissues and in dentine. The monoclonal antibody also stained the nuclei of certain cells: notably, whereas most of the fibroblasts in the tooth-related half of the incisor ligament were stained strongly, only occasional nuclei of fibroblasts in the molar ligament and in the bone-related half of the incisor ligament showed immunoreactivity. The differential staining of nuclei provides evidence for phenotypic differences between fibroblast populations within these tissues. The prominence of SPARC in the ligament tissues is consistent with their embryonic characteristics, whereas unfolded collagen recognized b’y the MBP 322 antibody may indicate sites of rapid collagen remodelling. Key words: SPARC protein, osteonectin, periodontal tissues, immunolocalization.
SCAB proteins, denatured collagen, collagen remodelling,
INTRODUCTION
SPARC is a 32 kD calcium-binding protein (Termine et al., 198la, 198lb: Romberg et al., 1985;‘Engel et al., 1987; Domenicucci et al., 1988) that was originally
isolated from bone and named osteonectin (Termine et al., 1981a, 198lb). Although its precise function is not known, its synthesis by many cell types and its occurrence in a variety of connective tissues (Wasi et al., 1984; Mason et al., 1986) and in basement membranes (Mann et al., 1987) indicate that it has a fundamental biologjcal role which may be of particular importance in development. Notably, the synthesis of SPARC is induced in the differentiation of parietal endoderm (Mason et al., 1986), is stimulated by shock in cultured endothelial cells (Sage, Johnson and Bornstein, 1984) and appears to be abundant in fetal tissues (Tung et al., 1985; Mason et al., 1986; *Present address: Institute of Dentistry, University of Turku, 20520 Turku, Finland. TAddress correspondence to: Dr J. Sodek, MRC Group in Periodontal Physiology, 4384 Medical Science Building, University of Toronto, Toronto, Ontario, Canada MSS lA8 Abbreviations : BSA, bovine serum albumin; PBS, phosphate-buffered saiine; SCAB, small collagenous apatite binding; SPARC, secreted protein, acidic, rich in cysteine.
Nomura et al., 1988). In adult tissues, SPARC is concentrated in bone and to a lesser extent in dentine (Termine et al., 198la; Domenicucci et al., 1988) and it is also found within human platelets (Stenner et al., 1986). In earlier studies we have shown that SPARC is synthesized by periodontal ligament and gingival fibroblasts (Wasi et al., 1984; Zung et al., 1986; Otsuka et al., 1988) and that SPARC protein is enriched in porcine periodontal ligament (Wasi et al., 1984; Tung et al., 1985). However, in contrast to the situation in larger mammals, there is relatively little SPARC in rat bone and it is barely detectable in rat incisor dentine (Zung et al., 1986). The relationship between collagen remodelling and the presence of unfolded collagen has been investigated by using polyclonal antibodies to collagen which were affinity-purified against gelatin to select antibodies that recognized only denatured collagen (Rucklidge, Riddoch and Robbins, 1986). With such antibodies, tissues were identified that were perceived to be undergoing significant remodelling of collagen. During the preparation of monoclonal antibodies to bone proteins, we isolated a hybridoma that produced an antibody (MBP 322) against an apparently novel SCAB protein (Kuwata et al., 1987). This antibody was also found to recognize an epitope in the unfolded tl chains of interstitial collagens I-III and the ctl chain of type V collagen. However, it 337
J. SALONEN el al.
338
did not recognize the native forms of these collagens. The MBP 322 antibody, therefore, appeared to be useful in examining the relationship between the presence of denatured/unfolded collagen and collagen remodelling. Remodelling of collagen in periodontal tissues is rapid, especially in periodontal ligament (Sodek and Ferrier, 1988; Sodek, 1989), which also exhibits characteristics of embryonic and wound healing tissues. Therefore, we have used specific antibodies to SPARC and the MBP 322 monoclonal antibody to study the distribution of SPARC and the occurrence of unfolded collagen molecules in these and other dental tissues in the rat, to determine whether or not there is a relationship between these proteins and tissue remodelling.
MATERlALS
AND METHODS
Preparation and characterization of antibodies
Polyclonal antisera to purified porcine SPARC were raised in a New Zealand white rabbit (Domenicucci et al., 1988). This antiserum showed strong cross-reactivity to both human and rat SPARC (Otsuka et al., 1988). Although the antiserum specifically recognized SPARC in radiolabelled cultures of connective tissue cells, including rat cells, when used initially for immunocytochemical staining, non-specific reaction was observed in cell nuclei. Consequently, the antiserum in 200 ~1 portions was affinity-purified using pure porcine SPARC (100 pg) Sepharose 4B immobilized to CNBr-activated (100 ~1, Pharmacia, Uppsala, Sweden); these procedures were described in detail by Tung et al. (1985). The affinity-purified antibodies prepared in this way were used at dilutions of between 1: 40 and 1: 200 with respect to the original antiserum and were judged to be specific by using immunoprecipitations and immunoblots of tissue extracts. The monoclonal antibody, MBP 322, was produced by a hybridoma clone obtained from fusion of myeloma cells and spleen cells from a mouse immunized with an impure preparation of porcine osteonectin. These hybridoma cells were used to produce ascites fluid from which the antibodies were prepared and characterized as described by Kuwata et al. (1987). Affinity-purified antibodies to pig type I and III collagens were prepared and used as described by Rao et al. (1979) and Wang et al. (1980)
Immunohistochemical procedures
Lower jaws from 250 g adult Wistar rats were fixed in 10% neutral formalin or 3% glutaraldehyde for 24 h, then demineralized for 3 weeks in a 10% solution of sodium citrate in 22.5% formic acid. The tissues were then dehydrated and embedded in paraffin. Before staining, 6pm sections attached to microscope slides were de-paraffinized in toluene, and then hydrated in graded alcohol and PBS. To block endogenous peroxidase, the sections were soaked in 0.5% hydrogen peroxide for 5 min. After washing in PBS, the tissue sections were incubated overnight at 4°C with primary antibody diluted with PBS containing 2% BSA. After again washing with PBS, the sections were incubated for 1 h at 22°C with peroxidase-conjugated antibody directed at the primary antibody. For rabbit anti-SPARC antibodies, peroxidase-conjugated sheep anti-rabbit IgG (12.0 mg/ml protein); for MBP 322 mouse antibodies, peroxidaselinked rabbit anti-mouse IgG (12.4 mg/ml protein); and for sheep anti-porcine collagen types I and III antibodies, peroxidase-conjugated rabbit anti-sheep IgG was used. All second antibodies were obtained from Cappel (Organon Teknika N.V., Belgium) and were used at a dilution of 1: 50 in PBS containing 2% BSA. The peroxidase activity was revealed with a solution of 25 mg 3,3’-diaminobenzidine tetrahydrochloride in 50 ml of 50 mM tris-HCl buffer, pH 7.6, containing 100 ~1 of 3% hydrogen peroxide. The stained sections were dehydrated and mounted in Prot-Texx mounting medium (Lerner Laboratories, New Haven, CT, U.S.A.) and examined and photographed under a Leitz Orthoplan microscope. The specificity of the immunoreactions was controlled by substituting primary antibody with the equivalent fractions from the respective affinity columns upon which normal or pre-immune sera were fractionated, and in other cases by omitting the second antibody in the reaction sequence. RESULTS
Immunostaining for SPARC protein with affinitypurified antibodies demonstrated a relatively strong reaction in a number of soft tissues including periodontal ligament, pulp, endosteal tissues, marrow spaces and muscle, as shown in the sagittal section through a rat mandible (Fig. 1). More moderate staining was apparent in the gingival connective tissues with weak staining in the demineralized bone
Plate I Figs 1-6. Tissues Fig. 1. Sagittal
fixed in glutaraldehyde
to SPARC
section of lower jaw from a 250 g adult rat displaying immunoreactivity in tissues of molar and incisor teeth and alveolus.
Fig. 2. Cross-section Fig. 3. Enlarged
through
section
Fig. 5. Section
the first molar
through
Fig. 4. Section
the periodontal
as shown
through
Fig. 6. Section Ml,
and stained with specific antibodies collagen type I.
and associated antibodies. ligament
in Fig. 3 stained
a bundle as shown
periodontal
showing
in Fig. 5, but stained
M2 and M3, molar teeth 1, 2 and 3, respectively; ES, endosteal space; PL, periodontal
for SPARC stained
immunostaining
for SPARC
of muscle fibres stained
tissues
protein
and
protein
with SPARC
for collagen
type I.
distribution.
with haemotoxyhn for SPARC
protein
and eosin.
protein.
AB, alveolar bone; D, dentine; ligament; P, dental pulp.
C, cementurn;
Denatured
collagen
and SPARC
in rat dental
tissues
339
340
J. SALONENet al.
Plate 2
Denatured collagen and SPARC in rat dental tissues
tissues. No immunoreaction above background was evident in either the dentine, cementum or the enamel matrix at this resolution. The strong staining in the molar periodontal ligament and dental pulp shown at higher magnification in Fig. 2 contrasts with the unstained enamel matrix and cementum and also with the weak staining in the alveolar bone. A slight, diffuse immunoreaction appeared to be present in the demineralized dentine matrix at this magnification. However, this may be the result of reagents being trapped in the dentinal tubules. At even higher magnification, the fibrous pattern of staining that was evident for type I collagen (Fig. 3) was similar to the SPARC staining in the molar ligament (Fig. 4). However, SPARC stained more strongly and was more uniformly distributed in the soft tissues of the endosteal spaces. The diffuse and patchy staining of the dentine and alveolar bone with SPARC antibodies contrasted with the more uniform staining for type I collagen in these tissues. However, it was clear that type I staining in these tissues was compromised by the demineralization of these tissues, as has been found before (Rao et al., 1979). The distribution of SPARC in muscle was evaluated at higher magnification with the aid of successive tissue sections stained with haematoxylin and eosin (Fig. 5). The SPAFLC protein (Fig. 6) was revealed both on muscle fibre surfaces and between fibres, where it produced a floccular appearance. No particular association of the SPARC with muscle cells was apparent. The similarity in the distribution of SPARC and type III collagen was evident in the incisor ligament, which is shown in a series of successive sagittal sections that included an adjacent molar tooth root in Figs 7-12. The histological features of the tissues at medium and high magnification (Figs 7 and 8, respectively) are shown in sections stained with haematoxylin and ‘eosin. Type III collagen was revealed to be more or less uniformly distributed in both molar and incisor ligaments with circumferential staining in the dental pulp and somewhat stronger staining of the incisor ligament adjacent to the tooth in some regions (Figs 9 and 10). The SPARC showed a similar distribution to type III collagen in the ligaments but was stained more intensely and uniformly in the molar pulp (Figs 11 and 12). An interrupted band of staining for SPARC was also apparent between the dentine and cementurn. The mineralized tissues of bone and dentine were diffusely stained, whereas the soft tissue in the endosteal spaces stained strongly. At high magnification of the incisor ligament, SPARC protein had a fibrous appearance. In the region shown (Fig. 12), particularly strong staining was observed in material adjacent to the tooth, which appeared to run parallel to its length, whereas in the middle region of the ligament the
341
staining material appeared to follow a more oblique distribution. The most notable feature observed with the MBP 322 antibodies was the intense staining of cell nuclei in the tooth-related half of the incisor ligament. Also, the matrix in this region, which had fibre bundles lying almost parallel with the tooth and which stained strongly for collagen III and SPARC, was stained more intensely with MBP 322 antibody than was the matrix in the rest of the ligament. These features were revealed more clearly at higher magnifications shown in Figs 13-16. Notably, the number of nuclei stained with MBP 322 decreased distally from this region and was more clearly reduced in the more proximal regions of the ligament where the demarcation between the tooth-related and bone-related halves of the tissue is lost. The MBP 322 monoclonal antibody reacted fairly evenly with both hard and soft dental tissues. The immunostaining with MBP 322 was particularly strong throughout the molar periodontal ligament, endosteal soft tissue and pulp, with more moderate staining in the dentine and bone (Fig. 13). Notably, the enamel matrix was devoid of any staining. This distribution was similar to that of SPARC (Fig. 2), although MBP 322 stained the mineralized connective tissues more strongly. Considerably stronger staining of matrix was observed with tissues fixed with glutaraldehyde than in formalin-fixed tissue and undecalcified frozen sections. However, the relative staining intensities for the matrix in the different tissues was similar, indicating the antigen was more accessible in the glutaraldehyde-fixed tissues. At the highest magnification used, the staining in the molar ligament was revealed in fibrous elements traversing between the cementum and alveolar bone (Fig. 14). More intense staining was observed in the nuclei of a few cells (arrows), contrasting with the majority of cell nuclei, which showed no apparent immunoreactivity. In the incisor periodontal ligament the immunoreactivity was more clearly divided into a more heavily stained tooth-related half and a lighter stained bone-related half (Fig. 15). As in the molar ligament, relatively few nuclei were stained in the bone-related half of the ligament, whereas most of the nuclei of cells in the tooth-related half of the ligament were intensely stained. At even higher magnification, the antibody was seen to have stained fibrous structures in the tissue matrix; the nuclei had a granular staining pattern, with some nuclei staining more strongly than others (Fig. 16). DISCUSSION
In earlier immunohistochemical studies we described the distribution of SPARC protein (osteonectin) in porcine dental tissues (Wasi et al., 1984; Tung et al., 1985). Although it was already known
Plate 2 Figs 7-12. Sag&al sections through the mandibular incisor, including the distal root of the second molar, (Figs 7, 9, 11) and enlarged for the incisor ligament (Figs 8, 10, 12). Tissues were fixed in glutaraldehyde and sections stained with haematoxylin and eosin (Figs 7 and 8), with antibodies to collagen type III (Figs 9 and 10) and SPARC protein (Figs 11 and 12). AB, alveolar bone; D, molar dentine; ES, endosteal space; ID, incisor dentine; IL, incisor periodontal ligament; P, molar dental pulp; PL, molar periodontal ligament.
J. SALONEN et al.
342
from biochemical studies that the protein is abundant in porcine bone and dentine, those studies revealed that it was also present in a number of soft connective tissues, with a particularly strong reaction in periodontal ligament. Those observations supported biochemical studies in which SPARC protein has been extracted from various soft tissues including gingiva and periodontal ligament (Wasi et al., 1984) and in vitro studies, in which SPARC was shown to be synthesized by fibroblasts at levels similar to those of bone cells (Wasi et al., 1984; Otsuka et al., 1984). Subsequently, SPARC was found in platelets (Stenner et al., 1986) and basement membranes (Mann et al., 1987); it was produced by parietal endoderm (Mason et al., 1986) endothelial (Sage et al., 1984) and epithelial cells (Dziadek et al., 1986). Northern blot analysis (Mason et al., 1986) and in situ hybridization studies (Nomura et al., 1988; Holland et al., 1987) have indicated that the SPARC mRNA is expressed in a variety of tissues, with a particular abundance in placenta. In the fetal mouse, tissues associated with developing teeth showed relatively strong hybridization, indicating higher levels of SPARC expression in these tissues. Our immunostaining of adult rat dental tissues has now shown that SPARC is a prominent protein in several soft tissues including periodontal ligament, the soft tissue in endosteal spaces, dental pulp and muscle. The staining of ligament and endosteal tissue is consistent with our findings in porcine tissues (Tung et al., 1985). However, pulp was more strongly and uniformly stained in the rat, whereas the weak staining of rat alveolar bone and the even weaker staining of rat dentine contrasts with the conditions in porcine tissues. The staining for SPARC in porcine bone and dentine appears to be decreased in adult tissues (Tung et al., 1985), but it was also less than anticipated from biochemical analysis. Because SPARC is bound to the hydroxyapatite crystals (Domenicucci et al., 1988) it is possible that some of the SPARC was lost from the mineralized tissues during demineralization even though this was carried out in the presence of fixative. However, in rat bone, SPARC is present in substantially lower amounts than in the larger mammals such as man, pig and cow, and is difficult to detect in the incisor dentine (Zung er al., 1986; Sodek et al., 1986). It has been suggested that the low content of SPARC in rat bone could be attributed to rat SPARC having a lower affinity for hydroxyapatite because of differences in the structure of the putative hydroxyapatite-binding region (Domeniccuci et al., 1988). Further, SPARC appears to be concentrated in peritubular dentine (Tung et nl., 1985) and the lack of such dentine in rat incisors may explain the virtual absence of SPARC in that tissue.
The preponderance of SPARC in fetal tissues (Tung et al., 1985; Mason et al., 1986; Nomura et al., 1988) and its stimulation in response to trauma (Sage et al., 1984; Sage, Tupper and Bramson, 1986) and wound-healing hormones such as transforming growth factor-p (Wrana et al., 1988; Sodek et al., 1988) indicate a function for this protein in developmentally related processes. The relatively high content of SPARC in periodontal ligament is consistent with the embryonic features of this tissue, which include rapid remodelling, higher content of type III collagen and the type of collagen crosslinks (reviewed by Shuttleworth and Smalley, 1986). Although the immunostaining for SPARC protein in the periodontal ligament has a fibrous appearance similar to that of collagen, a direct association between these proteins appears unlikely, as SPARC protein is quantitatively extracted from bone by 0.5 M EDTA in the absence of denaturants and porcine SPARC does not bind to collagen in either the presence or absence of calcium ions (Domenicucci et al., 1988). It is interesting, however, that the majority of the SPARC protein in porcine periodontal ligament can only be extracted in the presence of 0.5 M EDTA (Wasi et al., 1984). A similar observation was made for basement membrane SPARC protein (Mann et al., 1987). Changes in secondary structure occur on calcium binding (Engel et al., 1987; Domenicucci et al., 1988) and it has been conformational suggested that calcium-induced changes may be responsible for SPARC binding to tissue matrices (Mann et al., 1987). The monoclonal antibody MBP 322 displayed a more general staining than the SPARC protein antibodies. The staining of bone and of osteoblastic cells adjacent to bone is consistent with the antibody being raised against a SCAB protein (Kuwata et al., 1987) which has yet to be fully characterized (Goldberg et al., 1988). Previous studies have indicated that the SCAB protein is present only in bone (Kuwata et al., 1987) and the staining of other tissues observed in our study would appear to be due to the recognition of the unfolded c(chains of interstitial collagens by these antibodies. That denatured forms of collagen exist in connective tissues has been indicated by Rucklidge et al. (1986) who have suggested that the presence of denatured collagen reflects collagen breakdown that occurs in association with tissue remodelling. The strong staining of molar periodontal ligament, in which the collagen has a half-life between 1 and 3 days (Sodek and Ferrier, 1988), appears to support this suggestion, as does the strong staining of pulp, in which collagen also appears to remodel rapidly (Orlowski and Doyle, 1976). The collagen in the incisor ligament also remodels rapidly, albeit with a slightly longer half-life of 3-6 days (Sodek and Ferrier, 1988). In this tissue, the tooth-related
Plate 3 Figs 13-14.
Immunostaining
Fig. 13. Cross-section Fig.
14. Molar
AB, alveolar
through
of glutaraldehyde-fixed second molar showing
tissues with MPB-322 strong,
uniform
staining
periodontal ligament at higher magnification showing occasional nuclei stained with the MPB 322 antibody
bone; B, bone above incisor ligament; C, cementum; periodontal ligament; IL, incisor periodontal ligament;
monoclonal
antibody.
in the periodontal
fibrous staining (arrows).
D, dentine; E, enamel; P, molar dental pulp.
ligament.
pattern
and
PL, molar
Denatured
collagen
and SPARC
Plate 3
in rat dental
tissues
343
J. SALONEN ef al.
344
.
’
Plate 4
Denatured collagen and SPARC in rat dental tissues ligament was stained more strongly than the bonerelated tissue. This varied from one-third to one-half of the width of th’: ligament depending upon the location along the incisor. Based on these observations, the rapid remodelling of collagen in the incisor ligament would appear to be uniform across the tooth-related part of the ligament rather than being localized to the intermediate plexus, where in the mouse, collagen phagocytosis (Beer&en and Everts, 1977) predominates. However, because of the effects of tissue preparation on the intensity of reaction with this antibody and the variability in the staining along the length of the incisor ligament, these findings need to be substantiated further before firm conclusions can be drawn. The staining of nuclei of cells that reside in the tooth-related half of the incisor ligament with MBP 322 is an interesting phenomenon for which an explanation cannot be given at present. Other studies have shown that n4BP 322 stains the nuclei of a variable number of cells in several different connective tissue populations. Further, there appears to be a reciprocal relationship between nuclear staining and the more frequent cytoplasmic staining that localizes secreted protein (Kuwata et al., 1987). It is possible that the epltope recognized by the antibody is present in a regulatory protein within the nucleus. However, there is no apparent correlation between this nuclear staining and cells involved in the collagen turnover because the nuclei of fibroblasts in the molar ligament are rarely stained. The most notable feature of the fibroblasts in the incisor ligament whose nuclei stain with MBP 322 is that these cells migrate with the eruption of the tooth (Beertsen, 1975), whereas the fibroblasts in the tooth-related part of the incisor ligament and those in molar periodontal ligament do not migrate. While such correlations are speculative, it is clear nevertheless that there are phenotypic differences between fibroblasts both within and between these tissues. Although the precise mechanism of collagen degradation in periodontal tissues is not known, it appears that collagen is normally degraded by a phagocytic pathway (Everts, Bi:ertsen and Trichelagaar-Gutter, 1985; Sodek and Overall, 1988), which may not involve collagenase (Everts and Beertsen, 1988). How the collagen that is to be degraded is identified and how it becomes internalized is currently unknown. The denatured collagen that is recognized by MBP 322 may be important in the identification process or it may represent collagen that has been partially digested, or both. Immuno-electron microscopic studies with this antibody may prove useful in this regard.
345
We have thus shown that SPARC protein and unfolded collagen are typically associated with rapidly remodelling soft connective tissues. Differences in staining patterns for these proteins can be attributed to the more general role of SPARC protein in developmental processes that involve tissues other than connective tissues alone. In contrast, the presence of denatured collagen is restricted to connective tissues, where it appears to reflect sites of collagen degradation. Acknowledgemenls -We are grateful for the expertise of Brigitte Hawrylyshyn in preparing antibodies for this study and Mr D. Wagner for his advice on tissue preparation and staining techniques. This study was supported by an MRC (Canada) Group Grant. C.D. is a recipient of an I’Anson Foundation Fellowship.
REFERENCES
Beertsen W. (1975) Migration of fibroblasts in the periodontal ligament of the mandibular incisor of the mouse: a preliminary report. Archs oral Biol. 18, 61-66. Beertsen W. and Everts V. (1977) The site of remodelling of collagen in the periodontal ligament of the mouse incisor. Anat. Rec. 189, 479-497. Domenicucci C., Goldberg H. A., Hofmann T., Isenman D., Wasi S. and Sodek J. (1988) Characterization of porcine osteonectin extracted from foetal calvariae. Biochem. J. 253, 139-151. Dziadek M., Paulsson M., Aumailley M. and Timpl R. (1986) Purification and tissue distribution of a small protein (BM-40) extracted from basement membrane tumor. Eur. J. Biochem. 161, 455-464. Engel J., Taylor W., Paulsson M., Sage H. and Hogan B. (1987) Calcium binding domains and calcium-induced conformational transition of SPARC/BM_40/osteonectin, an extracellular glycoprotein expressed in mineralized and non-mineralized tissues. Biochemistry 26, 6958-6965. Everts V. and Beertsen W. (1988) The cellular basis of tooth eruption; the role of collagen phagocytosis. In: The Biological Mechanism of Tooth Eruption and Root Resorp tion (Edited by Davidovitch Z.) pp. 237-242. EBSCO Media, Birmingham, AL. Everts V., Beertsen W. and Trichelagaar-Gutter W. (1985) The role of microtubules in the phagocytosis of collagen by fibroblasts. Coil. Rel. Res. I, 1-15. Goldberg H. A., Maeno M., Domenicucci C., Zhang Q. and Sodek J. (1988) Identification of small collagenous proteins with properties of procollagen G(l(1) pN-propetide in fetal porcine calvarial bone. Coil. Rel. Res. S, 187-197. Holland P. W. H., Harper S. J., McVey J. H. and Hogan B. L. M. (1987) In uiuo expression of mRNA for the Ca’+-binding protein SPARC (osteonectin) revealed by in situ hybridization. J. Cell Biol. 105, 473-482. Kuwata F., Maeno M., Yao K-L., Domenicucci C., Goldberg H. A., Wasi S., Aubin J. E. and Sodek J. (1987)
Plate 4 Figs 15-16. Fig. 15. Sagittal the tooth-related Fig. 16. Higher AB, alveolar
Immunostaining
of glutaraldehyde-fixed
tissues with MPB-322
monoclonal
antibody.
section through the incisor tooth beneath the first molar showing stronger staining in half of the periodontal ligament and the strong staining of cell nuclei. Strong staining of cells adjacent to bone (B) shown by arrows. magnification of the tooth-related half of the incisor ligament showing (arrows) and poorly stained nuclei (crossed arrows).
bone; B, bone above incisor ligament; C, cementurn; periodontal ligament; IL, incisor periodontal ligament;
the stained
D, dentine; E, enamel; P, molar dental pulp.
nuclei
PL, molar
J. SALONEN et al.
346
Characterization of a monoclonal antibody recognizing small collagenous proteins in fetal bone. Coil. Rel. Res. 7. 39-55.
-
-
Mann K., Deutzmann R., Paulsson M. and Timpl R. (1987) Solubilization of protein BM-40 from a basement membrane tumor with chelating agents and evidence for its identity with osteonectin and SPARC. FEES Left. 218, 167-172. Mason I. J., Taylor A., Williams J. G., Sage H. and Hogan B. L. M. (1986) Evidence from molecular cloning that SPARC, a major product of mouse embryo parietal endoderm, is related to an endothelial cell ‘culture shock protein. EMBO J. 5, 1465-1472. Nomura S., Wills A. J., Edwards D. R., Heath J. K. and Hogan B. L. M. (1988) Developmental expression of 2ar (osteopontin) and SPARC (osteonectin) RNA as revealed by in situ hybridization. J. Cell Biol. 106, 44-450. Orlowski W. and Doyle J. (1976) Collagen metabolism in the pulps of rat teeth. Archs oral Biol. 2.4, 391-394. Otsuka K.. Pitaru S.. Overall C. M.. Aubin J. E. and Sodek J. (1988) Biochemical comparison of fibroblast populations from different periodontal tissues: characterization of matrix protein and collagenolytic enzyme synthesis. Biochem. Cell Biol. 66, 167-176.
Otsuka K., Yao K-L., Wasi S., Tung P. S., Aubin J. E., Sodek J. and Termine J. D. (1984) Biosynthesis of osteonectin by fetal porcine calvarial cells in vitro. J. biol. Chem. 259, 9805-9812.
Rao L. G., Wang H-M., Kalliecharan R., Heersche J. N. M. and Sodek J. (1979) Specific immunohistochemical localization of type I collagen in porcine periodontal tissues using the peroxidase-labelled antibody technique. Histochem. J. 11, 73-82. Romberg R. W., Wemess P. G., Lollar P., Riggs B. L. and Mann- K. G. (1985) Isolation and characterization of native adult osteonectin. J. biol. Chem. 260. 2728-2736. Rucklidge G. H., Riddoch G. I. and Robbins’S. P. (1986) Immunocytochemical staining of rat tissues with antibodies to denatured type I collagen: a technique for localizing areas of collagen degradation. Coil. Rel. Res. 6, 41-49.
Sage H., Johnson C. and Bomstein P. (1984) Characterization of a novel serum albumin-binding glycoprotein secreted by endothelial ceils in culture. J. biol. Chem. 259, 3993-4007.
Sage H., Tupper J. and Bramson R. (1986) Endothelial cell injury in vitro is associated with increased secretion of an Mr 43,000 glycoprotein ligand. J. CeN Physiol. 127, 373-387.
Shuttleworth C. A. and Smalley J. W. (1986) Periodontal Ligament. In: International &views of Connective Tissue Research (Edited bv Hall D. A. and Jackson D. S.) Vol. 10, pp. 211-248. Academic Press, London. ’ Sodek J. (1989) Collagen turnover in periodontal ligament. In: Biology of Tooth Movement (Edited by Norton A. L.
and Burstone C. J.) pp. 157-181. CRC Press, Boca Raton, FL. Sodek J. and Ferrier J. M. (1988) Collagen remodeling in rat periodontal tissues: compensation for precursor reutilization confirms rapid turnover of collagen. Co/l. Rel. Res. 8, 11-21.
Sodek J. and Overall C. M. (1988) Collagen remodelling in hard and soft connective tissues. In: h4echanisms of Tooth Movement and Root Resorption (Edited by Davidovitch Z.) pp. 303-312. EBSCO Media, Birmingham, AL. Sodek J., Domenicucci C., Zung P., Kuwata F. and Wasi S. (1986) Biosynthesis, properties and distribution of osteonectin. In: Cell Mediaied Calcification and Matrix Vescicles (Edited by Ali S. Y.) pp. 135-141. Elsevier, Amsterdam. Sodek J., Overall C. M., Wrana J. L., Maeno M. and Kubota T. (1988) Molecular mechanisms of remodelling in the periodontium: regulation by transforming growth factor-b. In: Recent Advances in Clinical Periodontology (Edited by Ishikawa J.) pp. 63-78. Elsevier, Amsterdam. Stenner D. D., Tracy R. P., Riggs L. B. and Mann K. G. (1986) Human platelets contain and secrete osteonectin, a major protein of mineralized bone. Proc. natn. Acad. Sci. U.S.A. 83, 6892-6896.
Termine J. D., Belcourt A. B., Conn K. M. and Kleimnan H. K. (1981a) Mineral and collagen-binding proteins of fetal calf bone. J. biol. Chem. 256, 10,403-10,408. Termine J. D., Kleinman H K., Whitson S. W., Conn K. M., McGarvey M. L. and Martin G. R. (1981b) Osteonectin, a bone-specific protein linking mineral to collagen. Cell 26, 99-105. Tung P. S., Domenicucci C., Wasi S. and Sodek J. (1985) Localization of osteonectin and collagen types I and III in fetal and adult porcine dental tissues J. Histochem. Cytochem. 33, 531-540. Wang H-M., Nanda V., Rao L. G., Melcher A. H., Heersche, J. N. M. and Sodek J. (1980) Specific immunohistochemical localization of type III collagen in porcine periodontal tissues using the peroxidase anti-peroxidase method. J. Histochem. Cyrochem. 28, 121551223. Wasi S., Otsuka K., Yao K-L., Tung P. S., Aubin J. E., Sodek J. and Termine J. D. (1984) An osteonectin-like protein in porcine periodontal ligament and its synthesis by periodontal ligament fibroblasts. Can. J. Biochem. Cell Biol. 62, 470-478.
Wrana J. L., Maeno M., Yao K-L., Hawrylyshyn B., Domenicucci C. and Sodek J. (1988) Differential effects of TGF-fi on the synthesis of extracellular matrix proteins by normal fetal rat calvarial bone cell populations. J. Cell Biol. 106, 915-924. Zung P., Domenicucci C., Wasi S., Kuwata F. and Sodek J. (1986) Osteonectin is a minor component of mineralized connective tissues in rat. Biochem. Cell Biol. 64, 356-362.
Copyright
0
0003-9969/90 $3.00 + 0.00 1990 Pergamon Press plc
IMMUNOHISTOCHEMICAL LOCALIZATION OF SPARC (OSTEONECTIN) AND DENATURED COLLAGEN AND THEIR RELATIONSHIP TO REMODELLING IN RAT DENTAL TISSUES J. SALONEN,* C.
DOMENICUCCI,
H. A.
GOLDBERG
and J. SorxKt
Medical Research Council Group in Periodontal Physiology, Faculty of Dentistry, University of Toronto, Ontario, Canada MSS lA8 (Accepted
24 November 1989)
Summary-To study this relationship, specific antibodies were used to determine the distribution of these proteins in mature rat dental tissues. Staining for SPARC with affinity-purified polyclonal antibodies was prominent throughout molar and incisor ligaments, endosteal tissue, dental pulp and muscle. More moderate staining was observed in other soft tissues including the lamina propria of gingiva, whereas the staining of demineralized bone was weak and in dentine was barely detectable. A monoclonal antibody (MBP 322), raised against a denatured form of a small collagenous bone protein, reacted strongly with osteoblastic cells but more moderately with alveolar bone. A strong reaction, indicative of unfolded collagen, was also evident throughout the dental pulp and molar ligament, whereas in the incisor ligament staining was largely restricted to the tooth-related half. Moderate staining with this antibody was also observed in other soft tissues and in dentine. The monoclonal antibody also stained the nuclei of certain cells: notably, whereas most of the fibroblasts in the tooth-related half of the incisor ligament were stained strongly, only occasional nuclei of fibroblasts in the molar ligament and in the bone-related half of the incisor ligament showed immunoreactivity. The differential staining of nuclei provides evidence for phenotypic differences between fibroblast populations within these tissues. The prominence of SPARC in the ligament tissues is consistent with their embryonic characteristics, whereas unfolded collagen recognized b’y the MBP 322 antibody may indicate sites of rapid collagen remodelling. Key words: SPARC protein, osteonectin, periodontal tissues, immunolocalization.
SCAB proteins, denatured collagen, collagen remodelling,
INTRODUCTION
SPARC is a 32 kD calcium-binding protein (Termine et al., 198la, 198lb: Romberg et al., 1985;‘Engel et al., 1987; Domenicucci et al., 1988) that was originally
isolated from bone and named osteonectin (Termine et al., 1981a, 198lb). Although its precise function is not known, its synthesis by many cell types and its occurrence in a variety of connective tissues (Wasi et al., 1984; Mason et al., 1986) and in basement membranes (Mann et al., 1987) indicate that it has a fundamental biologjcal role which may be of particular importance in development. Notably, the synthesis of SPARC is induced in the differentiation of parietal endoderm (Mason et al., 1986), is stimulated by shock in cultured endothelial cells (Sage, Johnson and Bornstein, 1984) and appears to be abundant in fetal tissues (Tung et al., 1985; Mason et al., 1986; *Present address: Institute of Dentistry, University of Turku, 20520 Turku, Finland. TAddress correspondence to: Dr J. Sodek, MRC Group in Periodontal Physiology, 4384 Medical Science Building, University of Toronto, Toronto, Ontario, Canada MSS lA8 Abbreviations : BSA, bovine serum albumin; PBS, phosphate-buffered saiine; SCAB, small collagenous apatite binding; SPARC, secreted protein, acidic, rich in cysteine.
Nomura et al., 1988). In adult tissues, SPARC is concentrated in bone and to a lesser extent in dentine (Termine et al., 198la; Domenicucci et al., 1988) and it is also found within human platelets (Stenner et al., 1986). In earlier studies we have shown that SPARC is synthesized by periodontal ligament and gingival fibroblasts (Wasi et al., 1984; Zung et al., 1986; Otsuka et al., 1988) and that SPARC protein is enriched in porcine periodontal ligament (Wasi et al., 1984; Tung et al., 1985). However, in contrast to the situation in larger mammals, there is relatively little SPARC in rat bone and it is barely detectable in rat incisor dentine (Zung et al., 1986). The relationship between collagen remodelling and the presence of unfolded collagen has been investigated by using polyclonal antibodies to collagen which were affinity-purified against gelatin to select antibodies that recognized only denatured collagen (Rucklidge, Riddoch and Robbins, 1986). With such antibodies, tissues were identified that were perceived to be undergoing significant remodelling of collagen. During the preparation of monoclonal antibodies to bone proteins, we isolated a hybridoma that produced an antibody (MBP 322) against an apparently novel SCAB protein (Kuwata et al., 1987). This antibody was also found to recognize an epitope in the unfolded tl chains of interstitial collagens I-III and the ctl chain of type V collagen. However, it 337
J. SALONEN el al.
338
did not recognize the native forms of these collagens. The MBP 322 antibody, therefore, appeared to be useful in examining the relationship between the presence of denatured/unfolded collagen and collagen remodelling. Remodelling of collagen in periodontal tissues is rapid, especially in periodontal ligament (Sodek and Ferrier, 1988; Sodek, 1989), which also exhibits characteristics of embryonic and wound healing tissues. Therefore, we have used specific antibodies to SPARC and the MBP 322 monoclonal antibody to study the distribution of SPARC and the occurrence of unfolded collagen molecules in these and other dental tissues in the rat, to determine whether or not there is a relationship between these proteins and tissue remodelling.
MATERlALS
AND METHODS
Preparation and characterization of antibodies
Polyclonal antisera to purified porcine SPARC were raised in a New Zealand white rabbit (Domenicucci et al., 1988). This antiserum showed strong cross-reactivity to both human and rat SPARC (Otsuka et al., 1988). Although the antiserum specifically recognized SPARC in radiolabelled cultures of connective tissue cells, including rat cells, when used initially for immunocytochemical staining, non-specific reaction was observed in cell nuclei. Consequently, the antiserum in 200 ~1 portions was affinity-purified using pure porcine SPARC (100 pg) Sepharose 4B immobilized to CNBr-activated (100 ~1, Pharmacia, Uppsala, Sweden); these procedures were described in detail by Tung et al. (1985). The affinity-purified antibodies prepared in this way were used at dilutions of between 1: 40 and 1: 200 with respect to the original antiserum and were judged to be specific by using immunoprecipitations and immunoblots of tissue extracts. The monoclonal antibody, MBP 322, was produced by a hybridoma clone obtained from fusion of myeloma cells and spleen cells from a mouse immunized with an impure preparation of porcine osteonectin. These hybridoma cells were used to produce ascites fluid from which the antibodies were prepared and characterized as described by Kuwata et al. (1987). Affinity-purified antibodies to pig type I and III collagens were prepared and used as described by Rao et al. (1979) and Wang et al. (1980)
Immunohistochemical procedures
Lower jaws from 250 g adult Wistar rats were fixed in 10% neutral formalin or 3% glutaraldehyde for 24 h, then demineralized for 3 weeks in a 10% solution of sodium citrate in 22.5% formic acid. The tissues were then dehydrated and embedded in paraffin. Before staining, 6pm sections attached to microscope slides were de-paraffinized in toluene, and then hydrated in graded alcohol and PBS. To block endogenous peroxidase, the sections were soaked in 0.5% hydrogen peroxide for 5 min. After washing in PBS, the tissue sections were incubated overnight at 4°C with primary antibody diluted with PBS containing 2% BSA. After again washing with PBS, the sections were incubated for 1 h at 22°C with peroxidase-conjugated antibody directed at the primary antibody. For rabbit anti-SPARC antibodies, peroxidase-conjugated sheep anti-rabbit IgG (12.0 mg/ml protein); for MBP 322 mouse antibodies, peroxidaselinked rabbit anti-mouse IgG (12.4 mg/ml protein); and for sheep anti-porcine collagen types I and III antibodies, peroxidase-conjugated rabbit anti-sheep IgG was used. All second antibodies were obtained from Cappel (Organon Teknika N.V., Belgium) and were used at a dilution of 1: 50 in PBS containing 2% BSA. The peroxidase activity was revealed with a solution of 25 mg 3,3’-diaminobenzidine tetrahydrochloride in 50 ml of 50 mM tris-HCl buffer, pH 7.6, containing 100 ~1 of 3% hydrogen peroxide. The stained sections were dehydrated and mounted in Prot-Texx mounting medium (Lerner Laboratories, New Haven, CT, U.S.A.) and examined and photographed under a Leitz Orthoplan microscope. The specificity of the immunoreactions was controlled by substituting primary antibody with the equivalent fractions from the respective affinity columns upon which normal or pre-immune sera were fractionated, and in other cases by omitting the second antibody in the reaction sequence. RESULTS
Immunostaining for SPARC protein with affinitypurified antibodies demonstrated a relatively strong reaction in a number of soft tissues including periodontal ligament, pulp, endosteal tissues, marrow spaces and muscle, as shown in the sagittal section through a rat mandible (Fig. 1). More moderate staining was apparent in the gingival connective tissues with weak staining in the demineralized bone
Plate I Figs 1-6. Tissues Fig. 1. Sagittal
fixed in glutaraldehyde
to SPARC
section of lower jaw from a 250 g adult rat displaying immunoreactivity in tissues of molar and incisor teeth and alveolus.
Fig. 2. Cross-section Fig. 3. Enlarged
through
section
Fig. 5. Section
the first molar
through
Fig. 4. Section
the periodontal
as shown
through
Fig. 6. Section Ml,
and stained with specific antibodies collagen type I.
and associated antibodies. ligament
in Fig. 3 stained
a bundle as shown
periodontal
showing
in Fig. 5, but stained
M2 and M3, molar teeth 1, 2 and 3, respectively; ES, endosteal space; PL, periodontal
for SPARC stained
immunostaining
for SPARC
of muscle fibres stained
tissues
protein
and
protein
with SPARC
for collagen
type I.
distribution.
with haemotoxyhn for SPARC
protein
and eosin.
protein.
AB, alveolar bone; D, dentine; ligament; P, dental pulp.
C, cementurn;
Denatured
collagen
and SPARC
in rat dental
tissues
339
340
J. SALONENet al.
Plate 2
Denatured collagen and SPARC in rat dental tissues
tissues. No immunoreaction above background was evident in either the dentine, cementum or the enamel matrix at this resolution. The strong staining in the molar periodontal ligament and dental pulp shown at higher magnification in Fig. 2 contrasts with the unstained enamel matrix and cementum and also with the weak staining in the alveolar bone. A slight, diffuse immunoreaction appeared to be present in the demineralized dentine matrix at this magnification. However, this may be the result of reagents being trapped in the dentinal tubules. At even higher magnification, the fibrous pattern of staining that was evident for type I collagen (Fig. 3) was similar to the SPARC staining in the molar ligament (Fig. 4). However, SPARC stained more strongly and was more uniformly distributed in the soft tissues of the endosteal spaces. The diffuse and patchy staining of the dentine and alveolar bone with SPARC antibodies contrasted with the more uniform staining for type I collagen in these tissues. However, it was clear that type I staining in these tissues was compromised by the demineralization of these tissues, as has been found before (Rao et al., 1979). The distribution of SPARC in muscle was evaluated at higher magnification with the aid of successive tissue sections stained with haematoxylin and eosin (Fig. 5). The SPAFLC protein (Fig. 6) was revealed both on muscle fibre surfaces and between fibres, where it produced a floccular appearance. No particular association of the SPARC with muscle cells was apparent. The similarity in the distribution of SPARC and type III collagen was evident in the incisor ligament, which is shown in a series of successive sagittal sections that included an adjacent molar tooth root in Figs 7-12. The histological features of the tissues at medium and high magnification (Figs 7 and 8, respectively) are shown in sections stained with haematoxylin and ‘eosin. Type III collagen was revealed to be more or less uniformly distributed in both molar and incisor ligaments with circumferential staining in the dental pulp and somewhat stronger staining of the incisor ligament adjacent to the tooth in some regions (Figs 9 and 10). The SPARC showed a similar distribution to type III collagen in the ligaments but was stained more intensely and uniformly in the molar pulp (Figs 11 and 12). An interrupted band of staining for SPARC was also apparent between the dentine and cementurn. The mineralized tissues of bone and dentine were diffusely stained, whereas the soft tissue in the endosteal spaces stained strongly. At high magnification of the incisor ligament, SPARC protein had a fibrous appearance. In the region shown (Fig. 12), particularly strong staining was observed in material adjacent to the tooth, which appeared to run parallel to its length, whereas in the middle region of the ligament the
341
staining material appeared to follow a more oblique distribution. The most notable feature observed with the MBP 322 antibodies was the intense staining of cell nuclei in the tooth-related half of the incisor ligament. Also, the matrix in this region, which had fibre bundles lying almost parallel with the tooth and which stained strongly for collagen III and SPARC, was stained more intensely with MBP 322 antibody than was the matrix in the rest of the ligament. These features were revealed more clearly at higher magnifications shown in Figs 13-16. Notably, the number of nuclei stained with MBP 322 decreased distally from this region and was more clearly reduced in the more proximal regions of the ligament where the demarcation between the tooth-related and bone-related halves of the tissue is lost. The MBP 322 monoclonal antibody reacted fairly evenly with both hard and soft dental tissues. The immunostaining with MBP 322 was particularly strong throughout the molar periodontal ligament, endosteal soft tissue and pulp, with more moderate staining in the dentine and bone (Fig. 13). Notably, the enamel matrix was devoid of any staining. This distribution was similar to that of SPARC (Fig. 2), although MBP 322 stained the mineralized connective tissues more strongly. Considerably stronger staining of matrix was observed with tissues fixed with glutaraldehyde than in formalin-fixed tissue and undecalcified frozen sections. However, the relative staining intensities for the matrix in the different tissues was similar, indicating the antigen was more accessible in the glutaraldehyde-fixed tissues. At the highest magnification used, the staining in the molar ligament was revealed in fibrous elements traversing between the cementum and alveolar bone (Fig. 14). More intense staining was observed in the nuclei of a few cells (arrows), contrasting with the majority of cell nuclei, which showed no apparent immunoreactivity. In the incisor periodontal ligament the immunoreactivity was more clearly divided into a more heavily stained tooth-related half and a lighter stained bone-related half (Fig. 15). As in the molar ligament, relatively few nuclei were stained in the bone-related half of the ligament, whereas most of the nuclei of cells in the tooth-related half of the ligament were intensely stained. At even higher magnification, the antibody was seen to have stained fibrous structures in the tissue matrix; the nuclei had a granular staining pattern, with some nuclei staining more strongly than others (Fig. 16). DISCUSSION
In earlier immunohistochemical studies we described the distribution of SPARC protein (osteonectin) in porcine dental tissues (Wasi et al., 1984; Tung et al., 1985). Although it was already known
Plate 2 Figs 7-12. Sag&al sections through the mandibular incisor, including the distal root of the second molar, (Figs 7, 9, 11) and enlarged for the incisor ligament (Figs 8, 10, 12). Tissues were fixed in glutaraldehyde and sections stained with haematoxylin and eosin (Figs 7 and 8), with antibodies to collagen type III (Figs 9 and 10) and SPARC protein (Figs 11 and 12). AB, alveolar bone; D, molar dentine; ES, endosteal space; ID, incisor dentine; IL, incisor periodontal ligament; P, molar dental pulp; PL, molar periodontal ligament.
J. SALONEN et al.
342
from biochemical studies that the protein is abundant in porcine bone and dentine, those studies revealed that it was also present in a number of soft connective tissues, with a particularly strong reaction in periodontal ligament. Those observations supported biochemical studies in which SPARC protein has been extracted from various soft tissues including gingiva and periodontal ligament (Wasi et al., 1984) and in vitro studies, in which SPARC was shown to be synthesized by fibroblasts at levels similar to those of bone cells (Wasi et al., 1984; Otsuka et al., 1984). Subsequently, SPARC was found in platelets (Stenner et al., 1986) and basement membranes (Mann et al., 1987); it was produced by parietal endoderm (Mason et al., 1986) endothelial (Sage et al., 1984) and epithelial cells (Dziadek et al., 1986). Northern blot analysis (Mason et al., 1986) and in situ hybridization studies (Nomura et al., 1988; Holland et al., 1987) have indicated that the SPARC mRNA is expressed in a variety of tissues, with a particular abundance in placenta. In the fetal mouse, tissues associated with developing teeth showed relatively strong hybridization, indicating higher levels of SPARC expression in these tissues. Our immunostaining of adult rat dental tissues has now shown that SPARC is a prominent protein in several soft tissues including periodontal ligament, the soft tissue in endosteal spaces, dental pulp and muscle. The staining of ligament and endosteal tissue is consistent with our findings in porcine tissues (Tung et al., 1985). However, pulp was more strongly and uniformly stained in the rat, whereas the weak staining of rat alveolar bone and the even weaker staining of rat dentine contrasts with the conditions in porcine tissues. The staining for SPARC in porcine bone and dentine appears to be decreased in adult tissues (Tung et al., 1985), but it was also less than anticipated from biochemical analysis. Because SPARC is bound to the hydroxyapatite crystals (Domenicucci et al., 1988) it is possible that some of the SPARC was lost from the mineralized tissues during demineralization even though this was carried out in the presence of fixative. However, in rat bone, SPARC is present in substantially lower amounts than in the larger mammals such as man, pig and cow, and is difficult to detect in the incisor dentine (Zung er al., 1986; Sodek et al., 1986). It has been suggested that the low content of SPARC in rat bone could be attributed to rat SPARC having a lower affinity for hydroxyapatite because of differences in the structure of the putative hydroxyapatite-binding region (Domeniccuci et al., 1988). Further, SPARC appears to be concentrated in peritubular dentine (Tung et nl., 1985) and the lack of such dentine in rat incisors may explain the virtual absence of SPARC in that tissue.
The preponderance of SPARC in fetal tissues (Tung et al., 1985; Mason et al., 1986; Nomura et al., 1988) and its stimulation in response to trauma (Sage et al., 1984; Sage, Tupper and Bramson, 1986) and wound-healing hormones such as transforming growth factor-p (Wrana et al., 1988; Sodek et al., 1988) indicate a function for this protein in developmentally related processes. The relatively high content of SPARC in periodontal ligament is consistent with the embryonic features of this tissue, which include rapid remodelling, higher content of type III collagen and the type of collagen crosslinks (reviewed by Shuttleworth and Smalley, 1986). Although the immunostaining for SPARC protein in the periodontal ligament has a fibrous appearance similar to that of collagen, a direct association between these proteins appears unlikely, as SPARC protein is quantitatively extracted from bone by 0.5 M EDTA in the absence of denaturants and porcine SPARC does not bind to collagen in either the presence or absence of calcium ions (Domenicucci et al., 1988). It is interesting, however, that the majority of the SPARC protein in porcine periodontal ligament can only be extracted in the presence of 0.5 M EDTA (Wasi et al., 1984). A similar observation was made for basement membrane SPARC protein (Mann et al., 1987). Changes in secondary structure occur on calcium binding (Engel et al., 1987; Domenicucci et al., 1988) and it has been conformational suggested that calcium-induced changes may be responsible for SPARC binding to tissue matrices (Mann et al., 1987). The monoclonal antibody MBP 322 displayed a more general staining than the SPARC protein antibodies. The staining of bone and of osteoblastic cells adjacent to bone is consistent with the antibody being raised against a SCAB protein (Kuwata et al., 1987) which has yet to be fully characterized (Goldberg et al., 1988). Previous studies have indicated that the SCAB protein is present only in bone (Kuwata et al., 1987) and the staining of other tissues observed in our study would appear to be due to the recognition of the unfolded c(chains of interstitial collagens by these antibodies. That denatured forms of collagen exist in connective tissues has been indicated by Rucklidge et al. (1986) who have suggested that the presence of denatured collagen reflects collagen breakdown that occurs in association with tissue remodelling. The strong staining of molar periodontal ligament, in which the collagen has a half-life between 1 and 3 days (Sodek and Ferrier, 1988), appears to support this suggestion, as does the strong staining of pulp, in which collagen also appears to remodel rapidly (Orlowski and Doyle, 1976). The collagen in the incisor ligament also remodels rapidly, albeit with a slightly longer half-life of 3-6 days (Sodek and Ferrier, 1988). In this tissue, the tooth-related
Plate 3 Figs 13-14.
Immunostaining
Fig. 13. Cross-section Fig.
14. Molar
AB, alveolar
through
of glutaraldehyde-fixed second molar showing
tissues with MPB-322 strong,
uniform
staining
periodontal ligament at higher magnification showing occasional nuclei stained with the MPB 322 antibody
bone; B, bone above incisor ligament; C, cementum; periodontal ligament; IL, incisor periodontal ligament;
monoclonal
antibody.
in the periodontal
fibrous staining (arrows).
D, dentine; E, enamel; P, molar dental pulp.
ligament.
pattern
and
PL, molar
Denatured
collagen
and SPARC
Plate 3
in rat dental
tissues
343
J. SALONEN ef al.
344
.
’
Plate 4
Denatured collagen and SPARC in rat dental tissues ligament was stained more strongly than the bonerelated tissue. This varied from one-third to one-half of the width of th’: ligament depending upon the location along the incisor. Based on these observations, the rapid remodelling of collagen in the incisor ligament would appear to be uniform across the tooth-related part of the ligament rather than being localized to the intermediate plexus, where in the mouse, collagen phagocytosis (Beer&en and Everts, 1977) predominates. However, because of the effects of tissue preparation on the intensity of reaction with this antibody and the variability in the staining along the length of the incisor ligament, these findings need to be substantiated further before firm conclusions can be drawn. The staining of nuclei of cells that reside in the tooth-related half of the incisor ligament with MBP 322 is an interesting phenomenon for which an explanation cannot be given at present. Other studies have shown that n4BP 322 stains the nuclei of a variable number of cells in several different connective tissue populations. Further, there appears to be a reciprocal relationship between nuclear staining and the more frequent cytoplasmic staining that localizes secreted protein (Kuwata et al., 1987). It is possible that the epltope recognized by the antibody is present in a regulatory protein within the nucleus. However, there is no apparent correlation between this nuclear staining and cells involved in the collagen turnover because the nuclei of fibroblasts in the molar ligament are rarely stained. The most notable feature of the fibroblasts in the incisor ligament whose nuclei stain with MBP 322 is that these cells migrate with the eruption of the tooth (Beertsen, 1975), whereas the fibroblasts in the tooth-related part of the incisor ligament and those in molar periodontal ligament do not migrate. While such correlations are speculative, it is clear nevertheless that there are phenotypic differences between fibroblasts both within and between these tissues. Although the precise mechanism of collagen degradation in periodontal tissues is not known, it appears that collagen is normally degraded by a phagocytic pathway (Everts, Bi:ertsen and Trichelagaar-Gutter, 1985; Sodek and Overall, 1988), which may not involve collagenase (Everts and Beertsen, 1988). How the collagen that is to be degraded is identified and how it becomes internalized is currently unknown. The denatured collagen that is recognized by MBP 322 may be important in the identification process or it may represent collagen that has been partially digested, or both. Immuno-electron microscopic studies with this antibody may prove useful in this regard.
345
We have thus shown that SPARC protein and unfolded collagen are typically associated with rapidly remodelling soft connective tissues. Differences in staining patterns for these proteins can be attributed to the more general role of SPARC protein in developmental processes that involve tissues other than connective tissues alone. In contrast, the presence of denatured collagen is restricted to connective tissues, where it appears to reflect sites of collagen degradation. Acknowledgemenls -We are grateful for the expertise of Brigitte Hawrylyshyn in preparing antibodies for this study and Mr D. Wagner for his advice on tissue preparation and staining techniques. This study was supported by an MRC (Canada) Group Grant. C.D. is a recipient of an I’Anson Foundation Fellowship.
REFERENCES
Beertsen W. (1975) Migration of fibroblasts in the periodontal ligament of the mandibular incisor of the mouse: a preliminary report. Archs oral Biol. 18, 61-66. Beertsen W. and Everts V. (1977) The site of remodelling of collagen in the periodontal ligament of the mouse incisor. Anat. Rec. 189, 479-497. Domenicucci C., Goldberg H. A., Hofmann T., Isenman D., Wasi S. and Sodek J. (1988) Characterization of porcine osteonectin extracted from foetal calvariae. Biochem. J. 253, 139-151. Dziadek M., Paulsson M., Aumailley M. and Timpl R. (1986) Purification and tissue distribution of a small protein (BM-40) extracted from basement membrane tumor. Eur. J. Biochem. 161, 455-464. Engel J., Taylor W., Paulsson M., Sage H. and Hogan B. (1987) Calcium binding domains and calcium-induced conformational transition of SPARC/BM_40/osteonectin, an extracellular glycoprotein expressed in mineralized and non-mineralized tissues. Biochemistry 26, 6958-6965. Everts V. and Beertsen W. (1988) The cellular basis of tooth eruption; the role of collagen phagocytosis. In: The Biological Mechanism of Tooth Eruption and Root Resorp tion (Edited by Davidovitch Z.) pp. 237-242. EBSCO Media, Birmingham, AL. Everts V., Beertsen W. and Trichelagaar-Gutter W. (1985) The role of microtubules in the phagocytosis of collagen by fibroblasts. Coil. Rel. Res. I, 1-15. Goldberg H. A., Maeno M., Domenicucci C., Zhang Q. and Sodek J. (1988) Identification of small collagenous proteins with properties of procollagen G(l(1) pN-propetide in fetal porcine calvarial bone. Coil. Rel. Res. S, 187-197. Holland P. W. H., Harper S. J., McVey J. H. and Hogan B. L. M. (1987) In uiuo expression of mRNA for the Ca’+-binding protein SPARC (osteonectin) revealed by in situ hybridization. J. Cell Biol. 105, 473-482. Kuwata F., Maeno M., Yao K-L., Domenicucci C., Goldberg H. A., Wasi S., Aubin J. E. and Sodek J. (1987)
Plate 4 Figs 15-16. Fig. 15. Sagittal the tooth-related Fig. 16. Higher AB, alveolar
Immunostaining
of glutaraldehyde-fixed
tissues with MPB-322
monoclonal
antibody.
section through the incisor tooth beneath the first molar showing stronger staining in half of the periodontal ligament and the strong staining of cell nuclei. Strong staining of cells adjacent to bone (B) shown by arrows. magnification of the tooth-related half of the incisor ligament showing (arrows) and poorly stained nuclei (crossed arrows).
bone; B, bone above incisor ligament; C, cementurn; periodontal ligament; IL, incisor periodontal ligament;
the stained
D, dentine; E, enamel; P, molar dental pulp.
nuclei
PL, molar
J. SALONEN et al.
346
Characterization of a monoclonal antibody recognizing small collagenous proteins in fetal bone. Coil. Rel. Res. 7. 39-55.
-
-
Mann K., Deutzmann R., Paulsson M. and Timpl R. (1987) Solubilization of protein BM-40 from a basement membrane tumor with chelating agents and evidence for its identity with osteonectin and SPARC. FEES Left. 218, 167-172. Mason I. J., Taylor A., Williams J. G., Sage H. and Hogan B. L. M. (1986) Evidence from molecular cloning that SPARC, a major product of mouse embryo parietal endoderm, is related to an endothelial cell ‘culture shock protein. EMBO J. 5, 1465-1472. Nomura S., Wills A. J., Edwards D. R., Heath J. K. and Hogan B. L. M. (1988) Developmental expression of 2ar (osteopontin) and SPARC (osteonectin) RNA as revealed by in situ hybridization. J. Cell Biol. 106, 44-450. Orlowski W. and Doyle J. (1976) Collagen metabolism in the pulps of rat teeth. Archs oral Biol. 2.4, 391-394. Otsuka K.. Pitaru S.. Overall C. M.. Aubin J. E. and Sodek J. (1988) Biochemical comparison of fibroblast populations from different periodontal tissues: characterization of matrix protein and collagenolytic enzyme synthesis. Biochem. Cell Biol. 66, 167-176.
Otsuka K., Yao K-L., Wasi S., Tung P. S., Aubin J. E., Sodek J. and Termine J. D. (1984) Biosynthesis of osteonectin by fetal porcine calvarial cells in vitro. J. biol. Chem. 259, 9805-9812.
Rao L. G., Wang H-M., Kalliecharan R., Heersche J. N. M. and Sodek J. (1979) Specific immunohistochemical localization of type I collagen in porcine periodontal tissues using the peroxidase-labelled antibody technique. Histochem. J. 11, 73-82. Romberg R. W., Wemess P. G., Lollar P., Riggs B. L. and Mann- K. G. (1985) Isolation and characterization of native adult osteonectin. J. biol. Chem. 260. 2728-2736. Rucklidge G. H., Riddoch G. I. and Robbins’S. P. (1986) Immunocytochemical staining of rat tissues with antibodies to denatured type I collagen: a technique for localizing areas of collagen degradation. Coil. Rel. Res. 6, 41-49.
Sage H., Johnson C. and Bomstein P. (1984) Characterization of a novel serum albumin-binding glycoprotein secreted by endothelial ceils in culture. J. biol. Chem. 259, 3993-4007.
Sage H., Tupper J. and Bramson R. (1986) Endothelial cell injury in vitro is associated with increased secretion of an Mr 43,000 glycoprotein ligand. J. CeN Physiol. 127, 373-387.
Shuttleworth C. A. and Smalley J. W. (1986) Periodontal Ligament. In: International &views of Connective Tissue Research (Edited bv Hall D. A. and Jackson D. S.) Vol. 10, pp. 211-248. Academic Press, London. ’ Sodek J. (1989) Collagen turnover in periodontal ligament. In: Biology of Tooth Movement (Edited by Norton A. L.
and Burstone C. J.) pp. 157-181. CRC Press, Boca Raton, FL. Sodek J. and Ferrier J. M. (1988) Collagen remodeling in rat periodontal tissues: compensation for precursor reutilization confirms rapid turnover of collagen. Co/l. Rel. Res. 8, 11-21.
Sodek J. and Overall C. M. (1988) Collagen remodelling in hard and soft connective tissues. In: h4echanisms of Tooth Movement and Root Resorption (Edited by Davidovitch Z.) pp. 303-312. EBSCO Media, Birmingham, AL. Sodek J., Domenicucci C., Zung P., Kuwata F. and Wasi S. (1986) Biosynthesis, properties and distribution of osteonectin. In: Cell Mediaied Calcification and Matrix Vescicles (Edited by Ali S. Y.) pp. 135-141. Elsevier, Amsterdam. Sodek J., Overall C. M., Wrana J. L., Maeno M. and Kubota T. (1988) Molecular mechanisms of remodelling in the periodontium: regulation by transforming growth factor-b. In: Recent Advances in Clinical Periodontology (Edited by Ishikawa J.) pp. 63-78. Elsevier, Amsterdam. Stenner D. D., Tracy R. P., Riggs L. B. and Mann K. G. (1986) Human platelets contain and secrete osteonectin, a major protein of mineralized bone. Proc. natn. Acad. Sci. U.S.A. 83, 6892-6896.
Termine J. D., Belcourt A. B., Conn K. M. and Kleimnan H. K. (1981a) Mineral and collagen-binding proteins of fetal calf bone. J. biol. Chem. 256, 10,403-10,408. Termine J. D., Kleinman H K., Whitson S. W., Conn K. M., McGarvey M. L. and Martin G. R. (1981b) Osteonectin, a bone-specific protein linking mineral to collagen. Cell 26, 99-105. Tung P. S., Domenicucci C., Wasi S. and Sodek J. (1985) Localization of osteonectin and collagen types I and III in fetal and adult porcine dental tissues J. Histochem. Cytochem. 33, 531-540. Wang H-M., Nanda V., Rao L. G., Melcher A. H., Heersche, J. N. M. and Sodek J. (1980) Specific immunohistochemical localization of type III collagen in porcine periodontal tissues using the peroxidase anti-peroxidase method. J. Histochem. Cyrochem. 28, 121551223. Wasi S., Otsuka K., Yao K-L., Tung P. S., Aubin J. E., Sodek J. and Termine J. D. (1984) An osteonectin-like protein in porcine periodontal ligament and its synthesis by periodontal ligament fibroblasts. Can. J. Biochem. Cell Biol. 62, 470-478.
Wrana J. L., Maeno M., Yao K-L., Hawrylyshyn B., Domenicucci C. and Sodek J. (1988) Differential effects of TGF-fi on the synthesis of extracellular matrix proteins by normal fetal rat calvarial bone cell populations. J. Cell Biol. 106, 915-924. Zung P., Domenicucci C., Wasi S., Kuwata F. and Sodek J. (1986) Osteonectin is a minor component of mineralized connective tissues in rat. Biochem. Cell Biol. 64, 356-362.
