Abstract
Parathyroid hormone related peptide (PTHrP) was discovered as a causative factor of humoral hypercalcemia of malignancy (HHM). In the present study using HHM model rats, the time course of odontoblastic response to PTHrP and its relation to incisal fracture were elicited. Nude rats were implanted with PTHrP-expressing tumor (LC-6) cells, mandibular incisors were collected at several time points. Microscopically 3 distinctive types of odontoblastic/dentin lesions were observed. Hypercalcfied dentin, which was reported as hypercalcemia-induced lesion in previous reports, observed in all areas of the dentin from week 5–10 samplings. Dentin niche, observed solely in week-10 sampling point, exhibited a nature identical to that of reparative odontoblast reported in the literatures of various cytotoxic agents. Since cytotoxicites were neither observed prior to the lesions nor reported as a role of PTHrP, the reparative response may have derived from highly sustained levels of PTHrP. Loss of columnar odontoblasts height was initially observed at week-5 time point in the middle section of the incisor. This primary loss of cell height prior to incisor fracture was considered to be the earliest response to the increased PTHrP levels of this model.
Introduction
Parathyroid hormone related peptide (PTHrP) was initially discovered as an essential factor causing humoral hypercalcemia of malignancy (HHM) (Roskams and Desmet, 1997; Strewler, 2000). And in the last decade, extensive researches had been carried out to reveal PTHrP distribution and its roles on developments in several tissues (Strewler, 2000).
In terms of tooth development, PTHrP is known to express in enamel epithelia (Beck et al., 1995; Lee et al., 1995; Liu et al., 1998), while PTH/PTHrP receptor 1 (PTHR1) in cementoblasts (Tenorio and Hughes, 1996), dental papilla (Lee et al., 1995; Liu et al., 1998) and odontoblast (Lundgren et al., 1998; Calvi et al., 2004). In addition, functional studies revealed the fact that PTHrP expression was required for tooth eruption (Schipani et al., 1997; Philbrick et al., 1998; Kitahara et al., 2002), and recently Calvi and colleagues revealed, with continuously PTHR1 expressing transgenic mice, that critical role of this receptor on early odontogenesis in fetus and neonatal mice (Calvi et al., 2004).
On the other hand, to explore the nature of HHM and its effective treatments, several animal models have been proposed, and all successfully manifesting symptoms similar to those of HHM patients, including hypercalcemia and related morphological and functional changes observed in the bones and kidneys (Kukreja et al., 1990; Haq et al., 1993; Sato et al., 1993; Liu et al., 1995; Yamato et al., 1995; Takaori-Kondo et al., 1998; Yaghoobian et al., 1998).
On the same line, our group was developed several HHM xenograft models by implantation of human cancer cell lines in rodents (Endo et al., 1998). And we discovered that one of the models showed incisor fractures that appeared along with other known HHM symptoms (Kato et al., 2003). Further evaluation of the incisal lesions of this model revealed the presence of uniquely and symmetrically localized odontoblastic lesions (Kato et al., 2003). These findings suggest that the high PTHrP concentration of this model have a direct effect on certain populations of odontoblasts, consequently inducing such lesions. Furthermore, these results indicate that PTHrP may have some physiological roles in the odontogenesis of mature rodents, which has yet to be investigated.
Since the aim of our previous study was to show the characteristic of the lesions of this model, histopathological evaluation was carried out at only 1 time point of 12 weeks after tumor implantation. In contrast, the present study will serve to elucidate the time-course progression of the lesions, and to determine early odontoblastic response to PTHrP and its relation to the incisal fracture. To examine these aspects, incisors of the HHM model will be evaluated at 2, 5, 8, and 10 weeks after tumor implantation. Additionally, time-course observations will allow the further discussions of the roles of PTHrP on odontoblastic differentiation of matured rats in both physiological and pathological condition.
Materials and Methods
Animals
Thirty-two 5-week-old male nude rats were purchased from Clea Japan Inc. (Tokyo, Japan), housed in sterilized cages in an animal room maintained at a temperature of 24 ± 2°C and a humidity of 55 ± 10%, with 14 to 16 ventilations per hour and a 14-hour light and 10-hour dark cycle. They were allowed free access to standard rodent chow (CE-2, Clea Japan Inc.) and to sterilized water throughout the study. The experimental protocols were approved by the Ethics Committee for the Treatment of Laboratory Animals at Chugai Pharmaceutical Co., Ltd.
HHM Model
For experiments, the cancer cell line LC-6, established from human large-cell lung cancer at the Central Institute for Experimental Animal, Kanagawa, Japan (Tashiro et al., 1989), and known to express a large amount of PTHrP (Endo et al., 1998; Kato et al., 2003), was prepared. After 1 week of acclimatization, 20 rats received subcutaneous implants of 2-mm3 fragments of LC-6 in the right flank (HHM group). Five animals from the HHM group were sacrificed at each sampling time point of 2, 5, 8, and 10 weeks after implantation, respectively. The remaining 12 rats were not implanted (control group). Three animals from the control group were sacrificed at each aforementioned sampling time points, respectively, and serving as age-matched controls. Prior to sacrifice, the incisors of animals in both HHM and control groups were monitored twice a week, and the week of fracture occurrence (if any) were recorded.
Measurements of Blood Ca2+ and Plasma PTHrP Levels
Plasma samples were collected from the abdominal artery of all animals prior to sacrifice, from which concentrations of blood Ca2+ and the human intact PTHrP were measured by the electrode method using an autoanalyzer (M-634, Chiba Corning Diagnostics Co. Ltd., Tokyo Japan) and a 2-site immunoradiometric assay kit using recombinant PTHrP (1–87) as the standard (YUKA-MEDIAS Co. Ltd., Tokyo, Japan), respectively. The detectable limit of the PTHrP assay was 1.1 pmol/mL.
Histopathology
Animals were anesthetized with ether and exsanguinated through the bleeding from the abdominal artery. Both right and left mandibular incisors and tibias were removed, fixed in 20% neutral buffered formalin, and then decalcified in ethylenediaminetetraacetic acid 4Na for 2 weeks. Both incisors were trimmed transversely at the apical, middle, and incisal regions, embedded in paraffin, sectioned at 4-microne, stained with hematoxylin and eosin (H&E) and examined microscopically. The longitudinally trimmed H&E-stained sections of both tibias were also prepared to evaluate the proximal epiphyseal area.
To determine the characteristics and distribution of the changes in dentin and odontoblasts, the changes were recorded at lingual, mesial, lateral, and labial areas in each apical, middle, and incisal section (Figure 1).
To detect subtle changes, odontoblastic cell heights and dentin thickness were measured at the most convexed point of the dentin (point mc), and at both points of lateral and medial cement-enamel junction (points l- and m-cej) at the labial area of the middle sections (Figure 1). The measuring points were selected based on the previous reports (Weinreb and Weinreb, 1986; Ohshima and Yoshida, 1992; Tabata et al., 1993; Yoshida and Ohshima, 1996), and the reported values comparable to those of the control animals of the present study.
Statistics
Statistical analysis was conducted on the data received from measuring blood Ca2+ concentrations, odontoblastic cell heights, and dentin thickness. The means and standard deviations of the measured values were calculated in both the control and HHM groups at each sampling time point, and the significance of differences were tested. Each set of data was first examined by the F-test for homogeneity of the variance among groups. To test the significance of difference between the group means, in case the variance was equal, the Student’s t-test, or otherwise Aspin-Welch’s t-test was applied to the data sets. The significance level for all of these statistical tests was 5%.
Results
Time Course Progression of Pathophysiology of the HHM Model
In the HHM group, blood Ca2+ concentrations were higher at 5, 8, and 10 weeks after tumor implantation than that of the corresponding control, while the values 2 weeks after implantation were unchanged (Figure 2). The blood PTHrP was detected at 5, 8, and 10 weeks, but not at 2 weeks after tumor implantation and controls (Table 1).
Incisor fractures were not observed at 2 and 5 weeks after tumor implantation. Right and left incisor fractures simultaneously occurred 7 weeks after tumor implantation. Histopathology of tibial epiphyseal areas, at 5 weeks after implantation, revealed an increase of activated osteoclasts on the surface of the primary trabecular bone and a decrease of calcified- and hypertrophic-chondrocytes in the epiphyseal growth plate. At 8 and 10 weeks after implantation, the number of activated osteoclasts was markedly increased and calcified- and hypertrophic-chondrocytes almost disappeared.
Histopathology of Incisor
In the animals of HHM group, 3 types of lesions were observed in the odontoblast and/or dentin, whereas other compartments of tooth remained intact. Based on their characteristics, and unique progression and distribution patterns, the 3 types of odontoblast and/or dentin lesions were classified as “hypercalcified dentin,” “dentin niche with osteodentin,” or “decrease of odontoblastic cell height and dentin thickness.” These lesions were observed in both right and left incisor symmetrically. Tables 3 to 5, composed of ata from the right incisor, depicts the time-course distribution and incidences of the lesions.
Histopathology of Hypercalcified Dentin
The histopathological characteristic of hypercalcified dentin was visualized by concentric basophilic line within the dentin, without associating changes of odontoblasts (Figure 3). The affected dentin layers were closer to the pulp at 5 weeks, observed in more outward layers at 8 and 10 weeks after implantation (Figure 3). The lesions were observed in all animals at 5, 8 and 10 weeks after implantation, and were distributed in entire areas of each apical, middle, and incisal section (Table 2).
Histopathology of Dentin Niche
Histopathological nature of dentin niche was characterized by sharply demarcated recessed area of dentin with depolarized odontoblasts (osteodentin) incorporated within, produced by existing and/or newly formed odontoblasts (Figure 4). The lesion occurred at 1 of the animals of the 10 weeks’ sampling point of HHM group, and was localized to the labial area of the apical sections of both right and left incisors (Table 3).
Histopathology and Histomorphometry of Decreased Odontoblastic Cell Height and Dentin Thickness
In the middle section, histopathological characteristic of the odontoblasts of control group was columnar or high columnar phenotypes in the labial area, and those odontoblasts were pressed together and arranged in a pseudostratified layer. In the HHM group, the heights of these cells were decreased at 5, 8, and 10 weeks after tumor implantation, and further accompanied by the thinning of dentin at 10 weeks (Figure 5). A decrease of odontoblastic cell height occurred at 3 out of 5 animals at 5 weeks and all animals of 8 and 10 weeks’ sampling points. These lesions were distributed in the labial area at 5 and 8 weeks, and in the both labial and lingual areas at 10 weeks after implantation. The lesion was accompanied by a decrease of dentin thickness in the labial areas of all animals of 10 weeks sampling point (Table 4).
In the incisal sections, since the odontoblasts of the control animals of this section were already flat in morphology, the difference of the odontoblastic cell height between the groups was not obvious at any sampling points. At 8 and 10 weeks after tumor implantation, however, the thickness of labial dentin of incisal section was obviously thinner than that of the control, groups (Figure 5). In the incisal sections, a decrease of dentin thickness occurred at labial area of all animals of 8 and 10 weeks sampling points (Table 4).
Histomorphometric analysis of the middle sections revealed that the values of both odontoblastic cell height and thickness of dentin remained unchanged at 2 weeks after implantation. At 5 weeks after implantation, the odontoblastic cell heights of points l- and m-cej were shorter than observed in their respective controls; however, no significant difference in the point mc was detected. At 8 and 10 weeks after implantation, all points of odontoblastic cell height were shorter than that of their respective controls (Figure 6). The thickness of dentin was significantly reduced at the point mc in the animals of 10 weeks’ sampling point compared with their respective controls. This reduction was not detected at animals of earlier sampling points, nor at points l- and m-cej in all animals (Figure 6).
Discussion
The incisors of rats are known to grow throughout life, are composed of a dentin core, covered by enamel and cementum at the labial and lingual surface, respectively, and enclosing an incisor pulp in its center. Within the pulp, mesenchymal cells around the apical end of the incisor differentiate into odontoblasts, which start secreting dentin, and moving toward the incisal direction with their own differentiation (Kuijpers et al., 1996). Thus, it is important to determine the morphological characteristics along with the regional distribution and time-course progression when histopathologically examining rat incisors.
In our previous report, the HHM model animals of 12 weeks after tumor implantation showed not only dentin and/or odontoblastic lesions but lesions of other incisal compartments as well. Seeing that the localizations of the lesions in other incisal compartments were sporadic and inconsistent, and the characteristics of the lesions were similar to those of known fracture-related lesions (Kuijpers et al., 1996), the lesions were considered to be a secondary response to the incisal fracture. In this study, lesions of other incisal compartments were not observed in any portions of any sampling points, including the cervical loop of apical section where new odontoblasts and ameloblasts are generated. This and aforementioned findings indicate that the lesions of other incisor compartments were induced at late stage (from 10 to 12 weeks after implantation) of the model, confirming the secondary nature of such lesions. Because PTHrP and PTHR1 are known to express in enamel epithelia of fetal and neonatal, and cementoblasts of mature rodents (Beck et al., 1995; Lee et al., 1995; Tenorio and Hughes, 1996; Liu et al., 1998; Calvi et al., 2004), it is striking but considered extrinsic increase of PTHrP does not affect the generation and maturation of those cells in this HHM model using matured animals.
Hypercalcified dentin was observed on the animals of HHM group at 5, 8, and 10 weeks after implantation, and the characteristic of the lesion was identical to that previously reported in animals of 12 weeks after implantation (Kato et al., 2003). In the previous report, the localization of the lesion was distributed to the dentin of entire areas of all sections, without any correlations with the fractured area. Other previous reports, which evaluated hypercalcemia-related incisor changes, the identical histopathological change was recorded without any reference to the incisal fracture (Harris and Innes, 1931; Schour and Massler, 1934; Kuijpers et al., 1996). Through current observations, lesion with same nature as observed in the previous reports, occurred with hypercalcemia during 5, 8, and 10 weeks after implantation, expanding its affected area from inner to outer layers of the dentin along with the prolongation of the hypercalcemic duration. The evidence indicates that the lesions were induced by the systemic hypercalcemia of this model but did not cause the fracture.
Dentin niche with osteodentin, observed in 1 of the animals of 10 weeks sampling point in the HHM group of the present study, also occurred in all 5 animals of previously reported 12 weeks sampling point with identical histopathological characteristic and localization (Kato et al., 2003). Since the lesion was observed only after incisal fractures and long sustained high PTHrP concentration, the lesion is neither the cause of the fracture, nor the early response to the high PTHrP concentration of this model. Additionally, the lesion was identical to those occurring in response to the treatment with several cytotoxic agents, and considered to be the reparative response to the preceding cytotoxic effect of those agents, which restricted to certain susceptible populations of the odontoblasts (Adkins, 1972; Koppang, 1973; Adatia, 1975; Mikkelsen, 1978; Stent and Koppang, 1976, 1980; Stene, 1978, 1979; Nogueira et al., 1981; Dahl, 1984, 1985; Karim and Eddy, 1984; Dahl and Koppang, 1985; Moule et al., 1993). According to the accumulating knowledge of PTHrP, it is not known as a cytotoxic agent but as one of the factors related to the reparative response in several tissues (Ferguson et al., 1998; Vortkamp et al., 1998; Blomme et al., 1999; Kudo et al., 2000; Nakase et al., 2001; Okazaki et al., 2003). Taking into consideration the fact that there were no preceding cytotoxic changes in any animals of earlier sampling time points, it is suggested that the exposure to extremely long and sustained high PTHrP concentrations may induce the odontoblastic reparative mechanisms without preceding cytotoxitiy, in the late stage of this model.
A decrease of odontoblastic cell height was first observed in HHM animals 5 weeks after implantation. At 8 and 10 weeks after implantation, the lesion was accompanied with the thinning of the dentin, with affected areas expanding into the incisal sections at 8 weeks and both incisal and middle sections at 10 weeks after implantation. Since the finding was the only change observed prior to the incisor fracture other than the hypercalcemia induced hypercalcified dentin, it is possible to conclude the decrease of odontoblastic cell height was caused incisor fracture.
Both histopathological and histomorphmetrical analysis revealed that around the points of l- and m-cej of labial area of the middle section, the normal odontoblasts of the control group increased its cell height and initiated the change of the phenotype from columnar to high-columnar. At those points, the decrease of odontoblastic cell height initially observed at 5 weeks after implantation of HHM group. In the previous observation of this model at 12 weeks after tumor implantation, we reported its unique, restricted, and symmetrical localization and because of this localization pattern, it was suggested that the cause of the findings was local factor such as PTHrP. With the lack of reference to similar findings in previous reported incisor changes of hypercalcemic rats, it was suggested that the cause of the findings are most likely to be the increased levels of PTHrP (Kato et al., 2003). From current observations, since the localization of the finding was further restricted in 5 weeks and was expanded later, along with the detection of blood PTHrP, it is considered that the change was to be the earliest response of incisor to the high PTHrP concentration of this model.
Taken together the observations of the current study, the time course progression of the lesions in the model can be interpreted in the following paragraphs (Figure 7). As PTHrP concentration increases in the systemic circulation, it causes a decrease in cell height in the high-columnar odontoblasts that exist around the points of l- and m-cej of the middle region (Figures 7E and 7G), and continues to expand its affected areas toward point mc and incisal direction (Figures 7I and 7K). The affected odontoblasts reduce the production in the course of their migration toward the incisal direction, resulting in thinning of dentin (Figure 7L). Incisor fractures occur once the strength of dentin, reduced by continual thinning, surpasses the bearable limit for the mechanical stress (Figure 7I). Further sustained high concentration of PTHrP induces the formation of dentin niche in labial area of the apical region (Figures 7M and 7N), and along with the aforementioned processes, hypercalcemia induces hypercalcification of the dentin over the entire area of all region.
Recent studies on the relationships of PTHrP and/or its receptor with chondrogenesis have provided evidences of a temporal and spatially specific expression (Vortkamp et al., 1998; St-Jacques et al., 1999; Van Der Eerden et al., 2000; Nakase et al., 2001). The reports regarding PTHrP or its receptor knock out mice have revealed that the mice manifest chondrodysplasia due to the premature hypertrophic differentiation (Amizuka et al., 1996; Chung et al., 1998, 2001; Karp et al., 2000, Kobayashi et al., 2002). On the other hand, PTHrP oversignaling conditions, such as the animal models with PTHrP overexpression, caused to delay hypertrophic differentiation in chondrocytes (Weir et al., 1996; Schipani et al., 1995, 1997).
The accumulated knowledge suggests that PTHrP plays a role in suppressing hypertrophic differentiation in chondrogenesis. In addition, histopathological analysis of the tibias from the current model revealed a decrease or disappear of the calcified- and hypertrophic-chondrocytes, which could be interpreted due to both delay of hypertrophic differentiation and intensive resorption of cartilaginous matrix by the PTHrP-activated osteoclasts. Considering that the current model animals were under the PTHrP oversignaling condition, the interpretation of the tibias changes matched well with the previously proposed concepts of PTHrP role. Recently, Calvi and colleagues (2004) reported fetal and neonatal odontogenesis in collagen promoter-driven constitutively active PTHR1 mice, and described the consequences of the activation as odontoblastic maturation delay and formation of abnormal dentin matrix. Because the PTHR1 constitutively activated fetal or neonatal odontoblasts were under PTHrP oversignaling, it may be possible to expand the concept of chondrogenesis and early odontogenesis to odontogenesis of matured animals. From this point of view, the earliest response of odontoblasts in this PTHrP oversignaling model could be interpreted as the differentiational delay of the odontoblasts from columnar phenotype to high-columnar phenotype, rather than the mere loss of odonotoblastic cell height.
Furthermore, current investigation of this model introduces 2 new aspects of the physiology of odontogensis. First, because the earliest response to PTHrP is decrease of the odontoblastic cell height in labial area of the middle section leading to incisor fractures, the odontoblasts of this area may play an important role in maintaining the strength of incisor under the physiological condition. Second, since PTHrP regulates physiological growth and development in several tissues (Roskams and Desmet, 1997), PTHrP may be partially involved in regulating this maintenance mechanism through the suppression of the odontoblastic differentiation physiologically.
To clarify the possible relation between odontogenesis and PTHrP, continuing studies on the expression of the PTHrP and its receptor in the odontoblasts will be conducted.