|Home | About | Journals | Submit | Contact Us | Français|
In the mature rat parotid gland, myoepithelial cells (MEC) invest intercalated ducts, but not acini. During postnatal development, however, these cells differentiate around both intercalated ducts and acini, then translocate to only intercalated ducts during weaning. Previously, we found that thyroxine (T4) accelerates translocation of cells with small secretory granules from acini into intercalated ducts and the number of apoptotic cells increased tremendously with high doses. We present here additional analysis of the effects of T4 on developing rat parotid gland, namely, the distribution of MEC and the proliferation of parenchymal cells. Beginning at age 4 days, pups were given daily subcutaneous injections of low, medium and high doses of T4 or vehicle or no injection. At ages 4, 7, 10 and 15 days, glands were excised and processed for light microscopy. Sections were double-immunostained with antibodies against proliferating cell nuclear antigen (PCNA) and actin, and counterstained with hematoxylin. Proliferative activity was assessed via PCNA histochemistry and MEC were identified using actin histochemistry. MEC in the T4 groups invested mostly acini at 15 days in vehicle/normal glands and mostly intercalated ducts after 10 days in the T4 groups. The proliferative activity of acinar cells and MEC in vehicle/normal glands declined progressively with age and T4 increased the rate of this decline in the MEC in a dose-dependent manner. We conclude that T4 accelerates the translocation of MEC from acini to intercalated ducts, with an important mechanism being the more rapid decline in the proliferative activity of MEC than in acinar cells in the T4 groups. Some of the decline in the proliferative activity of all cells in the high and medium dose T4 groups after 7 days may have been due to dose-related thyroxine toxicity.
Most of the growth and differentiation of the rat parotid gland occurs postnatally (Lawson 1970, Redman and Sreebny 1970a, 1971). This permits the study of factors that influence gland development via interventions applied directly to the pups, thus avoiding modifications of the effects of application via the dams. Alterations of diet and hormones have been shown to influence strongly postnatal development of rodent salivary glands (Sasaki et al. 1976, Takeuchi et al. 1977a,b 1978, Takuma et al. 1978, Kumegawa et al. 1980, Johnson 1987, Ikeda et al. 2008, Inukai et al. 2008). Serum thyroid hormone concentration is low in the perinatal rat, increases to twice the adult level by 16 days of age, and returns to the adult level by 30 days (Clos et al. 1974, Porterfield and Hendrick 1981, Dussault et al. 1982, Porterfield 1985, Farwell and Dubord-Tomasetti 1999).
In the mature rat parotid gland, myoepithelial cells (MEC) invest intercalated ducts, but not acini (Garrett and Parsons 1973). During postnatal development, however, these cells differentiate around both intercalated ducts and acini, then translocate to only intercalated ducts during weaning (Redman et al. 1980, Tsujimura et al. 2006). Previously, we found that exogenous thyroxine (T4) accelerates the translocation of cells with small secretory granules from acini into intercalated ducts, and apoptotic cells increased tremendously with high doses (Ikeda et al. 2008). We present here further analysis of parotid glands to determine the effects of T4 on the distribution of MEC and proliferation of parenchymal cells in the developing rat parotid gland during the two weeks after birth.
All experimental procedures were approved by the appropriate review boards of The Nippon Dental University and the Washington, DC, Veterans Affairs Medical Center.
Details of many of the methods have been published earlier (Ikeda et al. 2008). Briefly, certified pathogen-free Sprague-Dawley female rats at 14 days postconception were obtained from Harlan (Indianapolis, IN). Pups were considered 0 days old on the day of birth. Beginning at 4 days, two pups in each of four litters of 10 pups per age were given daily subcutaneous injections of 0.1, 0.5 or 5 μg/g body weight of T4 (low, medium and high dose groups, respectively), or vehicle injection or no injection (minimum n per group = 6). At ages 4, 7, 10 and 15 days (12 for highest dose of T4), at 3 h postinjection, pups were deeply anesthetized with 2-bromo-2-chloro-1,1,1-trifluoraethane (Halothane; Sigma-Aldrich, St. Louis, MO), heart blood was collected for assessing serum T4, and the parotid glands were excised.
The serum T4 assessment was done as described earlier (Ikeda et al. 2008). The parotid glands were fixed in 4% formaldehyde prepared from paraformaldehyde in 0.05 M phosphate buffer, pH 7.4, for 6 h, then processed for embedding in paraffin. Five-micrometer-thick paraffin sections were immersed in 0.3% H2O2/ in methanol for 30 min to inhibit endogenous peroxidase activity, rinsed with phosphate-buffered saline (PBS) and immersed in 10% normal rabbit serum (Immuno-Biological Lab. Co., Ltd., Takasaki, Japan) for 30 min to prevent nonspecific binding. After rinsing with PBS, the sections were incubated for 1 h in a mouse monoclonal antibody against muscle actin (Nichirei, Tokyo, Japan), rinsed again with PBS, then incubated for 10 min with biotinylated rabbit anti-mouse IgG, IgA, and IgM (Nichirei). The streptavidin-horseradish peroxidase (HRP) complex (SABC) method was applied using diaminobenzidine (DAB; DOJINDO, Kumamoto, Japan, lot no. WE052; brown stain). After three rinses in 0.1 M glycine-HCl buffer at pH 2.2 for 2 h, the sections were rinsed with PBS and immunostained with a mouse monoclonal antibody against proliferating cell nuclear antigen (PCNA; DAKO, Glostrup, Denmark) overnight at room temperature using the SABC method, visualized with aminoethylcarbazol (Nichirei, product number H0709; red stain), and counterstained lightly with hematoxylin (Sigma; ScH-2). Proliferative activity was assessed using PCNA histochemistry and MEC were identified via actin histochemistry.
The total cells, the total of each cell type (acinar, myoepithelial, and intercalated and large ductal), and the cells that were PCNA-positive (total and by cell type) were counted in each sample. The PCNA-positive cells divided by the total cells × 100 = percent positive cells, or labeling index (LI). We also counted the number of acini and intercalated ducts that were fully, partly or not invested with MEC, and converted these data into scores representing the predominant distribution of MEC in each section. Scores ranged from 1 (only acini invested) to 5 (all MEC invested intercalated ducts). Only whole numbers were entered for analysis for each sample. Thus, if only a few MEC processes surrounded or nearly surrounded most acini while almost all intercalated ducts were invested with MEC, the score for that sample would be 4.0. The scoring system is described also in Table 1. The data are presented as means ± standard error (SE) and were subjected to one-way analysis of variance using the SPSS ver.14 software package (SPSS Japan, Tokyo).
Because there were no differences between the vehicle-injected and non-injected pups for any measurement at any age, these values were combined into a single vehicle/normal group (minimum n = 14) for comparison with the groups that were injected with T4. The serum T4 values and body weights of the pups were presented in our earlier paper (Ikeda et al. 2008). The former data are reprinted in Fig. 1 to facilitate comparison with the new data. In general, the serum value of T4 was proportional to the dose administered and the gain in body weight with age was slowed in proportion to T4 dose beginning at 7 days. Pups in the high dose group failed to thrive and one died at age 11 days. To prevent an unacceptable discomfort level and loss of data, the remaining pups in this group were euthanized at age 12 days.
Photomicrographs illustrating the changes in proliferative activity and the distribution and differentiation of the myoepithelium by age and treatment are presented in Figs. 2 (vehicle/normal and low dose T4) and and33 (medium and high dose T4). The proportions of MEC investing acini and intercalated ducts by age and treatment are compiled in Table 1. The proliferative activities (percent that reacted with anti-PCNA) by age and treatment of all parenchymal cells and by cell type are presented in Tables 2 and and3,3, respectively.
Only scattered MEC were sufficiently differentiated for reaction with anti-muscle actin in the vehicle/normal pups at age 4 days. Of these, almost all were around acini. At 7 days and older, many more MEC were seen, because most reacted with anti-muscle actin, but these still invested mostly acini at 15 days in vehicle/normal glands. The proportion of MEC investing intercalated ducts in the T4 treatment groups increased in proportion to the increase in T4 at 10 days and older, investing mostly intercalated ducts by 12–15 days.
It is important to note that mitosis of well differentiated acinar (Redman and Sreebny 1970b, Redman 1995) and myoepithelial (Redman 1994) cells in the developing rat parotid gland has been documented by electron microscopy. The proliferative activity of the acinar and MEC in vehicle/normal glands of pups at 4 days was high, then progressively declined with age. Administration of T4 accelerated their rate of decline in a roughly dose-dependent manner; that of MEC declined more rapidly than that of acinar cells in the T4 treatment groups, but not in the vehicle/normal group. There were no overt differences in the proportion of PCNA-positive MEC between those investing acini or intercalated ducts (data not shown).
The translocation of MEC from investing mostly acini to almost entirely intercalated ducts in the rat parotid gland occurs between 14 and 25 days after birth (Redman et al. 1980). Tsujimura et al. (2006) showed that the more rapid slowing of proliferative activity of MEC than of acinar cells during this period is a major factor in this shift. The observed pace of MEC translocation and the chronology of changes in proliferative activity with age in the vehicle/normal group generally are consistent with previous findings (Redman et al. 1980, Redman 1995, Tsujimura et al. 2006).
The acinar and intercalated duct LI, however, tended to be lower than those obtained by Redman (1995) using 3H-thymidine, especially at age 15 days. This is not unexpected given conflicting reports on the validity of PCNA expression as a marker of proliferating cells. Some authors have noted poor correlation of PCNA expression and good correlation of Ki-67 expression with LI assessed via uptake of bromodeoxyuridine or 3H-thymidine (Sullivan et al. 1993, Jones et al. 1995, Mushelishvilli et al 2003). Others have reported PCNA and Ki-67 to have essentially equal value in this regard (Cher et al. 1995, Hicks and Flaitz 2000). A low level of PCNA expression tends to linger long after the mitotic cycle has ended (van Dierendonck et al. 1991). Elias (1997) noted that there are several reasons for this. In addition to a longer half-life than that of Ki-67, some forms of PCNA are associated with DNA repair and other nonreplicative functions. Further, fixation for more than 30 h can affect performance of PCNA antibodies. He noted also, however, that the antibody to Ki-67 that can be used with paraffin sections is less effective than the one that is useful with fresh frozen sections.
The fixation period for our samples was uniform and much less than 30 h, and our criteria for cells being “PCNA-positive” excluded those with faintly brown nuclei. In doing this we may have over-corrected, because the foregoing suggests that PCNA LI should be higher than LI provided by uptake of DNA precursors. Nonetheless, on balance, it seems unlikely that the problems with PCNA cited above were a significant confounder of our within-study LI comparisons.
Apoptosis was rare in the vehicle/normal group in the same parotid glands that were used for our earlier report (Ikeda et al. 2008) and no apoptotic MEC cells were observed by Tsujimura et al. (2006). Thus, apoptosis was not a factor in the translocation of MEC from acini to intercalated ducts. Our results show that T4 clearly accelerates the translocation of MEC from acini to intercalated ducts. An important mechanism for this is an even more rapid decline in proliferative activity of the MEC than of the acinar cells in the pups injected with T4.
We concluded previously that the rapid decline in the rate of body weight gain and a huge increase in vacuoles and apoptosis among the parotid cells indicated that thyroid toxicity had developed in the pups of the high and medium dose groups by 10 days. This makes it difficult to determine how much of the precipitous decline in the proliferative activity of all parotid parenchymal cells observed in the high and medium dose groups is another manifestation of thyroxine-induced acceleration of cytodifferentiation and how much is due to thyroid toxicity. The decline in proliferative activity in the low dose pups can be attributed more clearly to accelerated cytodifferentiation, because signs of thyroid toxicity were minimal or absent in this group. Indeed, as we noted earlier, the serum T4 levels in the low dose pups came close to meeting our goal of causing a premature rise of thyroid hormone to the level that normally occurs 6–10 days later.
The effects of exogenous thyroid hormone on the developing rat parotid gland reported here and in our earlier paper must be considered in the context of the hormone’s other known effects on rat development. These include advancing by 1–2 days the ages at which the eyelids open (14–15 days) and the incisors and first molars erupt enough to become functional (normal 8–10 and15–17 days, respectively (Donaldson 1924, Huxley 1971, Ehrmann et al. 1984, Gamborino et al. 2001), the development of independent thermoregulation (Henning et al. 1979) and the ability to obtain fluids independent of milk from the dam (normal approximately 18 days; Krecek and Kreckova 1957). Indeed, eyelid opening, hair growth, and molar tooth eruption occurred about a day earlier in the T4 low dose pups and even earlier in the medium and high dose pups than in the vehicle/normal pups. Together, these effects could advance the weaning process that normally occurs between 15 and 28 days, during which a significant dietary switch from mostly milk to mostly solid food occurs (Redman and Sweney 1976, Henning et al. 1979). The much greater stimulation of parotid secretion by solid food promotes parotid acinar maturation between 21 and 25 days (Redman and Sreebny 1976). Although no evidence such as stomach contents was collected, the watery vacuoles in the acini of the medium and high dose pups at 10 days and older was taken as evidence of hypersecretion (Ikeda et al. 2008) and thus indirectly of precocious ingestion of solid food. In addition, exogenous thyroid hormone has been shown to induce precocious rises in serum corticosterone and corticosteroid-binding globulin in rat pups (D’Agostino and Henning 1982). Exogenous corticosteroids have been shown to increase the rate of other aspects of rat and mouse salivary gland differentiation (Johannessen 1965, Sasaki et al. 1976, Takeuchi et al. 1977a,b, Kumegawa et al. 1980, Johnson 1987, Ikeda et al. 2008, Inukai et al. 2008). Thus, it is possible that a T4-induced rise in serum corticosteroids also may have been involved in the acceleration of MEC translocation and decline in parenchymal cell proliferative activity observed here.
This research was supported in part by The Nippon Dental University, Tokyo, Japan (RI), Grant DE 14995 from The National Institute of Dental and Craniofacial Research, The National Institutes of Health, Bethesda, MD (RSR), and the Research Service of the Department of Veterans Affairs Medical Center, Washington, DC. We thank Ms Sumie Satoh for her technical assistance.