Liver-type of tissue non-specific alkaline phosphatase is induced during mouse bone and tooth cell differentiation.

BACKGROUND AND OBJECTIVE: Tissue non-specific alkaline phosphatase (TNSALP) contains two types-bone- and liver-type-which are produced from the same gene due to differences in splicing. These two differ in their promoter, but the amino acid sequences of the mature proteins are identical. In this study, we examined the relationship between the two types of TNSALP expression and osteoblast differentiation. DESIGN: Gene expression of the two types of TNSALP was observed by reverse transcription-polymerase chain reaction. MC3T3-NM4 was sub-cloned from an established mouse osteoblastic cell line in which osteoblast characters do not appear without dexamethasone. The C2C12 mouse myoblastic cell line, which can be induced to osteoblasts with bone morphogenic protein 2, and organ-cultured tooth germs were also used in this work. RESULTS: The gene expression of liver-type TNSALP was observed in only MC3T3-NM4 activated by dexamethasone. For C2C12, the gene expression of bone-type TNSALP was observed even in non-induced conditions where myotubes were formed, whereas the liver-type TNSALP mRNA was only expressed when C2C12 differentiated into osteoblasts by bone morphogenic protein 2. Furthermore, in the organ-cultured tooth germs, the liver-type TNSALP mRNA was expressed according to differentiation of tooth germs. CONCLUSION: These results suggest that the liver-type TNSALP mRNA is induced according to differentiation of bone and tooth.

Essential roles of G9a in cell proliferation and differentiation during tooth development.

Teeth develop through interactions between epithelial and mesenchymal tissues mediated by a signaling network comprised of growth factors and transcription factors. However, little is known about how epigenetic modifiers affect signaling pathways and thereby regulate tooth formation. We previously reported that the histone 3 lysine 9 (H3K9) methyltransferase (MTase) G9a is specifically enriched in the tooth mesenchyme during mouse development. In this study, we investigated the functions of G9a in tooth development using G9a conditional knockout (KO) mice. We used Sox9-Cre mice to delete G9a in the tooth mesenchyme because Sox9 is highly expressed in the mesenchyme derived from the cranial neural crest. Immunohistochemical analyses revealed that G9a expression was significantly decreased in the mesenchyme of Sox9-Cre;G9afl/fl (G9a cKO) mice compared with that in Sox9-Cre;G9a fl/+(control) mice. Protein levels of the G9a substrate H3K9me2 were also decreased in the tooth mesenchyme. G9a cKO mice showed smaller tooth germ after embryonic day (E) 16.5 and E17.5, but not at E15.5. The developing cusp tips, which were visible in control mice, were absent in G9a cKO mice at E17.5. At 3 weeks after birth, small first molars with smaller cusps and unseparated roots were formed. Organ culture of tooth germs derived from E15.5 cKO mouse embryos showed impaired tooth development, suggesting that tooth development per se is affected independently of skull development. BrdU labeling experiments revealed that the proliferation rates were decreased in the mesenchyme in G9a cKO mice at E17.5. In addition, the proliferation rates in the tooth inner enamel epithelium were also decreased. In situ hybridization revealed altered localization of genes associated with tooth development. In cKO mice, intensively localized expression of mRNAs encoding bone morphogenic protein (Bmp2 and Bmp4) was observed in the tooth mesenchyme at E17.5, similar to the expression patterns observed in control mice at E15.5. Localization of Shh and related signaling components, including Gli1, Ptch1, and Ptch2, in the tooth mesenchyme of cKO mice was generally similar to that at earlier stages in control mice. In addition, expression of Fgf3 and Fgf10 in the mesenchyme was decreased in G9a cKO mice at P0. Expression levels of Fgf9 and p21, both of which were expressed in the secondary enamel-knot, were also decreased. Thus, the expression of genes associated with tooth development was delayed in cKO mice. Our results suggest that H3K9MTase G9a regulates cell proliferation and timing of differentiation and that G9a expression in the tooth mesenchyme is required for proper tooth development.

Expression of transcripts for fibroblast growth factor 18 and its possible receptors during postnatal dentin formation in rat molars.

Fibroblast growth factors (FGFs) regulate the proliferation and differentiation of various cells via their respective receptors (FGFRs). During the early stages of tooth development in fetal mice, FGFs and FGFRs have been shown to be expressed in dental epithelia and mesenchymal cells at the initial stages of odontogenesis and to regulate cell proliferation and differentiation. However, little is known about the expression patterns of FGFs in the advanced stages of tooth development. In the present study, we focused on FGF18 expression in the rat mandibular first molar (M1) during the postnatal crown and root formation stages. FGF18 signals by RT-PCR using cDNAs from M1 were very weak at postnatal day 5 and were significantly up-regulated at days 7, 9 and 15. Transcripts were undetectable by in situ hybridization (ISH) but could be detected by in situ RT-PCR in the differentiated odontoblasts and cells of the sub-odontoblastic layer in both crown and root portions of M1 at day 15. The transcripts of FGFR2c and FGFR3, possible candidate receptors of FGF18, were detected by RT-PCR and ISH in differentiated odontoblasts throughout postnatal development. These results suggest the continual involvement of FGF18 signaling in the regulation of odontoblasts during root formation where it may contribute to dentin matrix formation and/or mineralization.

Polyglucosan neurotoxicity caused by glycogen branching enzyme deficiency can be reversed by inhibition of glycogen synthase.

Uncontrolled elongation of glycogen chains, not adequately balanced by their branching, leads to the formation of an insoluble, presumably neurotoxic, form of glycogen called polyglucosan. To test the suspected pathogenicity of polyglucosans in neurological glycogenoses, we have modeled the typical glycogenosis Adult Polyglucosan Body Disease (APBD) by suppressing glycogen branching enzyme 1 (GBE1, EC 2.4.1.18) expression using lentiviruses harboring short hairpin RNA (shRNA). GBE1 suppression in embryonic cortical neurons led to polyglucosan accumulation and associated apoptosis, which were reversible by rapamycin or starvation treatments. Further analysis revealed that rapamycin and starvation led to phosphorylation and inactivation of glycogen synthase (GS, EC 2.4.1.11), dephosphorylated and activated in the GBE1-suppressed neurons. These protective effects of rapamycin and starvation were reversed by overexpression of phosphorylation site mutant GS only if its glycogen binding site was intact. While rapamycin and starvation induce autophagy, autophagic maturation was not required for their corrective effects, which prevailed even if autophagic flux was inhibited by vinblastine. Furthermore, polyglucosans were not observed in any compartment along the autophagic pathway. Our data suggest that glycogen branching enzyme repression in glycogenoses can cause pathogenic polyglucosan buildup, which might be corrected by GS inhibition.

Histological demonstration of glucose transporters, fructose-1,6-bisphosphatase, and glycogen in gas gland cells of the swimbladder: is a metabolic futile cycle operating?

Luminal surface of the swimbladder is covered by gas gland epithelial cells and is responsible for inflating the swimbladder by generating O(2) from Root-effect hemoglobin that releases O(2) under acidic conditions. Acidification of blood is achieved by lactic acid secreted from gas gland cells, which are poor in mitochondria but rich in the glycolytic activity. The acidic conditions are locally maintained by a countercurrent capillary system called rete mirabile. To understand the regulation of anaerobic metabolism of glucose in the gas gland cells, we analyzed the glucose transporter expressed there and the fate of ATP generated by glycolysis. The latter is important because the ATP should be immediately consumed otherwise it strongly inhibits the glycolysis rendering the cells unable to produce lactic acid anymore. Expression analyses of glucose transporter (glut) genes in the swimbladder of fugu (Takifugu rubripes) by RT-PCR and in situ hybridization demonstrated that glut1a and glut6 are expressed in gas gland cells. Immunohistochemical analyses of metabolic enzymes demonstrated that a gluconeogenesis enzyme fructose-1,6-bisphosphatase (Fbp1) and a glycolytic enzyme glyceraldehyde-3-phosphate dehydrogenase (Gapdh) are highly expressed in gas gland cells. The simultaneous catalyses of glycolysis and gluconeogenesis reactions suggest the presence of a futile cycle in gas gland cells to maintain the levels of ATP low and to generate heat that helps reduce the solubility of O(2).

[Analysis and classification of Latin anatomical names of skeletal fossa in Terminologia Anatomica, and comparison with corresponding Japanese anatomical names].

For a better understanding of the structures comprising the human body and in view of the possible need for future revision of anatomical nomenclature, Latin anatomical names of skeletal fossa in Terminologia Anatomica were analyzed and classified, and compared with the corresponding Japanese anatomical names. The words following Fossa indicated: 1) the form of the fossa, 2) the structure to which the fossa belongs, 3) the position of the fossa, 4) the structure that exists near the fossa, 5) the structure that the fossa contains, 6) the structure attached to the fossa, 7) the structure that transmits the fossa, or 8) the structure with which the fossa articulates. The analysis of Latin names and comparison with Japanese names clarified some characteristics of both names and revealed some problems in them.

[Analysis and classification of Latin anatomical names of skeletal canals in Terminologia anatomica, and comparison with corresponding Japanese anatomical names].

For a better understanding of the structures comprising the human body and in view of the possible need for future revision of anatomical nomenclature, Latin anatomical names of skeletal canals (including canaliculi) in Terminologia Anatomica were analyzed and classified, and compared with the corresponding Japanese anatomical names. The words following Canalis or Canales indicated: (1) the structure to which the canal belongs, (2) the structure to which the opening of the canal belongs, (3) the position of the canal, (4) the structure to which the canal leads, (5) the structure that exists near the opening of the canal, (6) the structure that transmits the canal, (7) the structure that is a component of the canal, or (8) the function of the canal. The analysis of Latin names and comparison with Japanese names clarified some characteristics of both names and revealed some problems in them.

[Analysis and classification of Latin anatomical names of skeletal foramina in Terminologia Anatomica, and comparison with corresponding Japanese anatomical names].

For better understanding of the structures comprising the human body and in view of the possible need for future revision, Latin anatomical names of skeletal foramina in Terminologia Anatomica were analyzed and classified, and compared with the corresponding Japanese anatomical names. The words following Foramen or Foramina indicated:1) the form of the foramen, 2) the state of the foramen, 3) the absolute size of the foramen, 4) the structure to which the foramen belongs, 5) the position of the foramen in the skeletal system, 6) the part of the human body in which the foramen exists, 7) the structure that transmits the foramen, or 8) the function of the foramen. Analysis of Latin names and comparison with Japanese names clarified some characteristics of both names and revealed some problems in them.

Identification of mycobacterial alpha-glucan as a novel ligand for DC-SIGN: involvement of mycobacterial capsular polysaccharides in host immune modulation.

Mycobacterium tuberculosis possesses a variety of immunomodulatory factors that influence the host immune response. When the bacillus encounters its target cell, the outermost components of its cell envelope are the first to interact. Mycobacteria, including M. tuberculosis, are surrounded by a loosely attached capsule that is mainly composed of proteins and polysaccharides. Although the chemical composition of the capsule is relatively well studied, its biological function is only poorly understood. The aim of this study was to further assess the functional role of the mycobacterial capsule by identifying host receptors that recognize its constituents. We focused on alpha-glucan, which is the dominant capsular polysaccharide. Here we demonstrate that M. tuberculosis alpha-glucan is a novel ligand for the C-type lectin DC-SIGN (dendritic cell-specific ICAM-3-grabbing nonintegrin). By using related glycogen structures, we show that recognition of alpha-glucans by DC-SIGN is a general feature and that the interaction is mediated by internal glucosyl residues. As for mannose-capped lipoarabinomannan, an abundant mycobacterial cell wall-associated glycolipid, binding of alpha-glucan to DC-SIGN stimulated the production of immunosuppressive IL-10 by LPS-activated monocyte-derived dendritic cells. By using specific inhibitors, we show that this IL-10 induction was DC-SIGN-dependent and also required acetylation of NF-kappaB. Finally, we demonstrate that purified M. tuberculosis alpha-glucan, in contrast to what has been reported for fungal alpha-glucan, was unable to activate TLR2.

Development and terminal differentiation of pulp and periodontal nerve elements in subcutaneous transplants of molar tooth germs and incisors of the rat.

Ectopic tooth transplants are known to receive rich innervation of local neurons, but the precise location and structural features of neurites in the pulp and periodontal ligament (PDL) of such transplants are unclear. In this experiment, the molar tooth germs of rat embryos and incisors of young rats were subcutaneously transplanted into the dorsal regions of rats and processed, at various time intervals, for immunohistochemical demonstration of neural elements. Teeth with periodontal tissue elements developed in most of the molar transplants in 6 or 8 wk and received rich innervation, including some autonomic fibres, in the pulp. Nerve elements were also confirmed to be present in the PDL of these transplants, including specialized nerve ending-like structures reminiscent of the periodontal Ruffini endings. Mechanoreceptor-like structures were also induced in the regenerated PDL of similarly transplanted incisors, although the success rate was low. We conclude that rich and highly ordered innervation of the pulp, and occasional development of mechanoreceptors in the regenerated PDL of ectopic dental transplants, imply a high probability of successful induction of teeth with both nociceptive and mechanical sensations in the ectopic tooth and/or tooth germ transplant systems, although differentiation of mechanoreceptor-like nerve endings occurred in only a few rare cases.

Aggregation and amyloid fibril formation of the prion protein is accelerated in the presence of glycogen.

Prion diseases like Creutzfeldt-Jakob disease in humans or scrapie in sheep and goats are infectious neurodegenerative diseases. Their infectious agent, called prion, is composed mainly of aggregated and misfolded prion protein and non-proteinaceous components. An example of such a common non-proteinaceous secondary component of natural prions is the polysaccharide scaffold. We studied the influence of such a polysaccharide on the conformational transition of PrP applying an in vitro conversion system. Here we report that glycogen supports and accelerates PrP amorphous aggregation similar to seeded aggregation and leads to co-aggregates. Furthermore, PrP fibril formation was highly accelerated in the presence of glycogen.

"Fluorescent glycogen" formation with sensibility for in vivo and in vitro detection.

There are presently many methods of detecting complex carbohydrates, and particularly glycogen. However most of them require radioisotopes or destruction of the tissue and hydrolysis of glycogen to glucose. Here we present a new method based on the incorporation of 2-NBDG (2-{N-[7-nitrobenz-2-oxa-1, 3-diazol 4-yl] amino}-2-deoxyglucose), a D-glucose fluorescent derivative, into glycogen. Two kinds of approaches were carried out by using Clone 9 rat hepatocytes as a cellular model; (1) Incubation of cell lysates with 2-NBDG, carbohydrate precipitation in filters and measurement of fluorescence in a microplate reader (2) Incubation of living hepatocytes with 2-NBDG and recording of fluorescence images by confocal microscopy. 2-NBDG labeled glycogen in both approaches. We confirmed this fact by comparison to the labeling obtained with a specific monoclonal anti-glycogen antibody. Also drugs that trigger glycogen synthesis or degradation induced an increase or decrease of fluorescence, respectively. This is a simple but efficient method of detecting glycogen with 2-NBDG. It could be used to record changes in glycogen stores in living cells and cell-free systems and opens the prospect of understanding the role of this important energy reserve under various physiological and pathophysiological conditions.

Formation of acellular cementum-like layers, with and without extrinsic fiber insertion, along inert bone surfaces of aging c-Src gene knockout mice.

To investigate the long-term effects of c-src deficiency on skeletal and dental tissues, we examined the lower jaws and long bones of c-src gene knockout (c-src KO) mice by histological and histochemical methods. Numerous multinucleated osteoclasts were distributed throughout the mandible in 5-wk-old c-src KO mice, but by 14 wk they had almost completely disappeared from the alveolar bone, leaving tartrate-resistant acid phosphatase (TRAP)-positive layers along the bone surface. Deposition of osteopontin-positive mineralized tissue, reminiscent of acellular afibrillar cementum (AAC), was confirmed along the TRAP-positive bone surface at 14 wk. The layer progressively thickened up to 21 months. A comparable mineralized layer was noted along the trabeculae of long bones as thickened cement lines. In the periostin-rich areas of jaw bones, but not in the long bones, portions of AAC-like mineralized layers were often replaced with and/or covered by acellular extrinsic fiber cementum (AEFC)-like tissue. These data suggest that the deposition of AAC-like mineralized tissue is a general phenomenon that may occur along inert or slowly remodeling bone surfaces under conditions characterized by reduced bone-resorbing activity, whereas the induction of AEFC-like tissue seems to be associated with the expression of certain molecules that are particularly abundant in the microenvironment of the periodontal ligament.

Proteolysis on maturing enamel surface, as shown by gel-coating methods.

Degradation of enamel matrix proteins, and their removal during early maturation, is critical for the growth of large enamel crystals in the subsequent processes of enamel maturation. In this study, we sought to demonstrate, using in vivo zymography, the exact sites of proteolysis in maturing enamel and its relationship to the overlying ameloblasts. The maturing enamel surfaces of rat and bovine incisors were exposed and painted either with pre-exposed autoradiographic emulsion or with densely fluorescein-conjugated (DQ) gelatin. After a few hours, photographic development of the emulsion revealed alternate black and white banding patterns over the maturing enamel surface. DQ gelatin also revealed similar banding patterns of fluorescent and non-fluorescent regions. White, powdery areas of emulsion and fluorescent bands of DQ gelatin both corresponded to the areas of ruffle-ended ameloblasts, at least up to the mid stages of enamel maturation, implicating a predominant contribution of ruffle-ended ameloblasts in the degradation of enamel matrix proteins. Powdery white bands in autoradiographic emulsion shifted from the areas of ruffle-ended to smooth-ended ameloblasts in late maturation in both bovine and rat incisors and were not influenced by proteinase inhibitors or heat inactivation, implicating non-enzymatic interactions. DQ gelatin, in fact, did not generate any fluorescence in such smooth-ended ameloblast regions.

Identification of genes preferentially expressed in periodontal ligament: specific expression of a novel secreted protein, FDC-SP.

Gene expression in human periodontal ligament (PDL) was examined by suppression subtractive hybridization to identify genes that are preferentially expressed in tissue compared to cultured PDL fibroblasts. The most enriched genes in a subtracted cDNA library are primarily genes for extracellular matrix components, types I and III collagen, lumican, periostin, and asporin, among others, whose expression conveys unique mechanical properties to the PDL. Also within this group is the gene for follicular dendritic cell secreted protein (FDC-SP), a small protein like statherin in saliva, not previously found in PDL. FDC-SP's presence in PDL was confirmed by in situ hybridization in mouse which also showed that it was definitely present in the parotid gland, but, surprisingly, not in the other salivary glands: submandibular and sublingual. Since only normal tissue was examined, these findings suggest that FDC-SP plays an important but previously unsuspected role within oral connective tissue.

Constitutive activation of GSK3 down-regulates glycogen synthase abundance and glycogen deposition in rat skeletal muscle cells.

The effects of inhibition or constitutive activation of glycogen synthase kinase-3 (GSK3) on glycogen synthase (GS) activity, abundance, and glycogen deposition in L6 rat skeletal muscle cells were investigated. GS protein expression increased approximately 5-fold during differentiation of L6 cells (comparing cells at the end of day 5 with those at the beginning of day 3). However, exposure of undifferentiated myoblasts (day 3) to 50 microM SB-415286, a GSK3 inhibitor, led to a significant elevation in GS protein that was not accompanied by changes in the abundance of GLUT4, another late differentiation marker. In contrast, stable expression of a constitutively active form of GSK3beta (GSK3S9A) led to a significant reduction (approximately 80%) in GS protein that was antagonized by SB-415286. Inhibition of GSK3 or expression of the constitutively active GSK3S9A did not result in any detectable changes in GS mRNA abundance. However, the increase in GS protein in undifferentiated myoblasts or that seen following incubation of cells expressing GSK3S9A with GSK3 inhibitors was blocked by cycloheximide suggesting that GSK3 influences GS abundance possibly via control of mRNA translation. Consistent with the reduction in GS protein, cells expressing GSK3S9A were severely glycogen depleted as judged using a specific glycogen-staining antibody. Inhibiting GSK3 in wild-type or GSK3S9A-expressing cells using SB-415286 resulted in an attendant activation of GS, but not that of glucose transport. However, GS activation alone was insufficient for stimulating glycogen deposition. Only when muscle cells were incubated simultaneously with insulin and SB-415286 or with lithium (which stimulates GS and glucose transport) was an increase in glycogen accretion observed. Our findings suggest that GSK3 activity is an important determinant of GS protein expression and that while glycogen deposition in muscle cells is inherently dependent upon the activity/expression of GS, glucose transport is a key rate-determining step in this process.

Expression of dentin matrix protein 1 (DMP1) in nonmineralized tissues.

Dentin matrix protein 1 (DMP1) is an Arg-Gly-Asp-containing acidic phosphoprotein that was originally identified from a rat incisor cDNA library and was thought to be a dentin-specific protein. DMP1 was later shown to express in a number of hard tissue-forming cells, including osteoblasts, osteocytes, ameloblasts, and cementoblasts, and was considered to play important roles in mineralization. Further, DMP1 gene expression was also detected in fetal bovine brain and in newborn mouse brain. These findings indicate the possibility of DMP1 expression in other soft tissues. In the present study, to clarify the significance of DMP1 expression in nonmineralized tissues, we made a specific antibody to mouse DMP1 peptides and demonstrated that DMP1 protein was localized in mouse brain, pancreas, and kidney by immunohistochemistry. Further DMP1 mRNA was detected in nonmineralized mouse tissues including liver, muscle, brain, pancreas, and kidney by RT-PCR. Based on the evidence that the localization and the expression of DMP1 are not restricted to mineralized tissues, we assume that DMP1 may have functions other than the regulation of mineralization.

The induction of enamel and dentin complexes by subcutaneous implantation of reconstructed human and murine tooth germ elements.

Tooth induction by xenogenic graft of reconstructed human tooth germ components has never been attempted. Here we report our first attempt at a transplantation of human tooth germ components, heterologously recombined with mouse dental epithelia, into immunocompromised animals. Human third molar tooth germs enucleated from young patients as prophylactic treatment for orthodontic reasons were collected. The whole or minced human dental papilla was reconstructed with human- or mouse molar enamel epithelium, and transplanted in the dorsal aspect of C.B-17/Icr-scid Jcl mice. The transplant of human dental papilla reconstructed with human enamel epithelium formed thin dentin and immature enamel layers by 3 to 4 weeks, but remained extremely small in quantity due to a shortage of epithelial components in the graft. The addition of E16 mouse molar enamel organs (n=10-12) to each graft augmented the formation of tooth germ-like structures, but the differentiation of mouse molar ameloblasts was suppressed. However, once a solid layer of mineralized dentin was established, mouse ameloblasts accelerated their differentiation, and completed the enamel matrix formation and maturation within the following 4 weeks, whereas human ameloblasts, which had interacted with human dental papilla, remained in the stage of matrix formation during the same period. These data imply that, in reconstructed transplants, the differentiation of mouse dental epithelia is restrained by putative suppressive factors derived from human dental papilla until they are separated by mineralized dentin layers that serve as a diffusion barrier. The mouse enamel organ nevertheless retains its own phenotypic characteristics and intrinsic timing of cell differentiation and function.

A novel domain in AMP-activated protein kinase causes glycogen storage bodies similar to those seen in hereditary cardiac arrhythmias.

The AMP-activated protein kinase (AMPK) is an alphabetagamma heterotrimer that is activated by low cellular energy status and affects a switch away from energy-requiring processes and toward catabolism. While it is primarily regulated by AMP and ATP, high muscle glycogen has also been shown to repress its activation. Mutations in the gamma2 and gamma3 subunit isoforms lead to arrhythmias associated with abnormal glycogen storage in human heart and elevated glycogen in pig muscle, respectively. A putative glycogen binding domain (GBD) has now been identified in the beta subunits. Coexpression of truncated beta subunits lacking the GBD with alpha and gamma subunits yielded complexes that were active and normally regulated. However, coexpression of alpha and gamma with full-length beta caused accumulation of AMPK in large cytoplasmic inclusions that could be counterstained with anti-glycogen or anti-glycogen synthase antibodies. These inclusions were not affected by mutations that increased or abolished the kinase activity and were not observed by using truncated beta subunits lacking the GBD. Our results suggest that the GBD binds glycogen and can lead to abnormal glycogen-containing inclusions when the kinase is overexpressed. These may be related to the abnormal glycogen storage bodies seen in heart disease patients with gamma2 mutations.

Possible role of dentin matrix in region-specific deposition of cellular and acellular extrinsic fibre cementum.

The mechanism whereby a region-specific deposition of the two types of cementum (cellular cementum and acellular extrinsic fibre cementum) is regulated on the growing root surface was tested using bisphosphonate-affected teeth of young rats and guinea pigs. The animals were injected subcutaneously with 8 or 10 mg P x kg body weight(-1) x day(-1) of 1-hydroxyethylidene-1,1-bisphosphonate (HEBP) for 1 or 2 weeks. In rat molars, HEBP prevented mineralization of newly formed root dentin matrix and totally inhibited de novo deposition of acellular extrinsic fibre cementum. Instead, thick cellular cementum was induced on the non-mineralized root dentin surface, irrespective of the position of the root. In both animals, cellular cementum was also induced on the non-mineralized surface of root analogue dentin in HEBP-affected incisors, where only acellular extrinsic fibre cementum is deposited under normal conditions. In normal rat molars, dentin sialoprotein (DSP) was concentrated along the dentin-cellular cementum border, but not that of dentin and acellular extrinsic fibre cementum. In HEBP-affected rat incisors, DSP was shown to penetrate through the non-mineralized dentin into the surrounding tissues, but not through the mineralized portions. These data suggest that, at the site of cellular cementum formation, putative inducing factors for cellular cementum might diffuse into the periodontal space through the newly deposited mantle dentin matrix before it is mineralized. At earlier stages of root formation, mantle dentin might mineralize more promptly not to allow such diffusion. The timing of mineralization of mantle dentin matrix might be the key determinant of the types of the cementum deposited on the growing root surface.