Expression of cAMP response element-binding protein in the duct ...

Histochem Cell Biol (2009) 132:647–657 DOI 10.1007/s00418-009-0649-2

ORIGINAL PAPER

Expression of cAMP response element-binding protein in the duct system of the mouse submandibular gland Sunisa Keattikunpairoj · Tomohiko Wakayama · Miyuki Yamamoto · Masa-aki Nakaya · Hiroki Nakata · Wiphawi Hipkaeo · Natthiya Sakulsak · Shoichi Iseki

Accepted: 7 September 2009 / Published online: 17 October 2009 © Springer-Verlag 2009

Abstract The submandibular gland (SMG) of mice shows a marked sexual dimorphism in which a duct portion called the granular convoluted tubule (GCT) is developed preferentially in males during puberty. The administration of testosterone to female mice causes the conversion of striated duct (SD) cells into GCT cells, but the underlying molecular mechanisms are unclear. Cyclic AMP response element-binding protein (CREB) is a transcription factor functioning downstream of a variety of signal transduction pathways. In the present study, we examined the expression, activation and cellular localization of CREB in the mouse SMG using Western blotting and immunohistochemistry. Both total CREB (t-CREB) and phosphorylated CREB (p-CREB) were signiWcantly more abundant in the female than in the male gland and were localized to the nuclei of intercalated duct cells and a subpopulation of SD cells. In contrast, the GCT cells in males were negative for t- and p-CREB. The levels of CREB in the SMG were increased by castration in males and decreased by repeated administration of testosterone to females or castrated males. From 3 h after a single administration of testosterone to females, many SD cells temporarily gained nuclear immunoreactivity for both t- and p-CREB, which was lost as the

S. Keattikunpairoj · T. Wakayama · M. Yamamoto · M. Nakaya · H. Nakata · W. Hipkaeo · N. Sakulsak · S. Iseki (&) Department of Histology and Embryology, Graduate School of Medical Science, Kanazawa University, 13-1 Takara-machi, Kanazawa 920-8640, Japan e-mail: [email protected] Present Address: W. Hipkaeo · N. Sakulsak Department of Anatomy, Faculty of Medical Sciences, Naresuan University, Phitsanulok, Thailand

cells were converted to GCT cells by 24 h. These results suggest the involvement of CREB in the androgen-dependent diVerentiation of the duct system in the mouse SMG. Keywords CREB · Submandibular gland · Sexual dimorphism · Granular convoluted tubule · DiVerentiation · Androgens · Mouse

Introduction The duct system of the submandibular gland (SMG) of rodents is composed of the intercalated duct (ID), striated duct (SD), granular convoluted tubule (GCT) and excretory duct (PinkstaV 1980). Extensive development of the GCT from the SD takes place around puberty in an androgendependent manner, resulting in a marked sexual dimorphism in the morphology and function of the duct system, with the GCT developing preferentially in the male gland (Jacoby and Leeson 1959; Caramia 1966a; Gresik 1980). The epithelial cells of the GCT possess abundant secretory granules that contain a variety of biologically active peptides, including nerve growth factor (NGF), epidermal growth factor, transforming growth factor , renin and kallikrein (Barka 1980; Gresik 1994). Castration of the animals causes involution of the GCT accompanied by the conversion of GCT cells into SD cells; whereas, the administration of androgens to females or castrated males causes the opposite phenomenon (Caramia 1966b; Chretien 1977). Such androgen-induced GCT diVerentiation is accompanied by upregulation of the gene expression of GCT-speciWc products (Gubits et al. 1986). The androgens, similar to other steroid hormones, exert their biological functions by binding to the androgen receptor (AR), a cytoplasmic/nuclear receptor, which is believed

123

648

to act as a transcription factor in itself by binding to the androgen response element (ARE), which is located upstream of androgen-regulated genes (Zhou et al. 1994; Chang et al. 1995). Although the AR is expressed in both acinar and duct cells of the rodent SMG (Morrell et al. 1987; Sar et al. 1990), little is known about the mechanism by which androgens cause the diVerentiation of GCT cells. In our previous study, we demonstrated in the rat SMG that the transcription factor cAMP response element-binding protein (CREB) is expressed abundantly in the nuclei of ID and SD cells at 3–5 W postpartum, but is no longer expressed in mature GCT cells in adults (Amano and Iseki 1998; Kim et al. 2001). The administration of testosterone to immature or hypophysectomized rats caused a temporary rise in the number of CREB-positive SD cells prior to their conversion to GCT cells, suggesting cross-talk between the signaling pathways involving CREB and AR. Evidence has accumulated in recent years that steroid hormones, including androgens, can exert rapid, non-classical or non-genomic eVects by activating a variety of second messenger cascades (Wehling 1997; Schmidt et al. 2000; Heinlein and Chang, 2002; Losel et al. 2003; Foradori et al. 2008; Michels and Hoppe 2008). These cascades include increases in the concentration of free intracellular calcium and activation of protein kinase A (PKA), protein kinase C (PKC), and mitogen-activated (MAP) kinase, all of which are known to activate CREB as a downstream transcription factor. On activation by phosphorylation at serine 133, CREB forms a dimer, translocates to the nucleus, binds to the consensus sequence cAMP response element (CRE), which is located in the promotor regions of a variety of genes, and Wnally transactivates these genes by recruiting coactivators (Shaywitz and Greenberg 1999; Johannessen and Moens 2007; Johannessen et al. 2004). In the present study, using the mouse SMG, which shows more apparent sexual dimorphism than the rat SMG, we examined the expression and localization of total CREB (t-CREB) and phosphorylated CREB (p-CREB) in the duct system, with reference to sexual dimorphism and androgeninduced diVerentiation of SD cells into GCT cells.

Histochem Cell Biol (2009) 132:647–657

from Cell Signaling Technology (Beverly, MA) and SigmaAldrich (St. Louis, MO), respectively. Testosterone, dehydrotestosterone (DHT) and Xutamide were purchased from Wako Pure Chemical Industries (Osaka, Japan). Animals and preparation of tissues Male and female Slc:ddY mice, some of which underwent testectomy at 6 weeks (W) of age, were purchased from Nippon SLC (Hamamatsu, Japan) and reared until 8 W of age (adulthood) under standard 12-h light/12-h dark laboratory conditions with free access to standard food and water. All subsequent experiments were conducted in accordance with the Guidelines for the Care and Use of Laboratory Animals at Kanazawa University. Groups of Wve male and female animals at various postnatal ages between 1 and 8 W were used without hormone treatment. Groups of Wve females and testectomized male adults were subjected to subcutaneous injections of 25 mg/kg of testosterone or DHT dissolved in 0.1 ml of olive oil (Wako Pure Chemical Industries), or the vehicle alone, every 24 h for 5 consecutive days and killed at 6 h after the last injection. In some groups of testectomized males, 100 mg/kg of Xutamide, a competitive inhibitor of androgens, dissolved in olive oil, was administered either alone or 1 h prior to each administration of DHT. In another experiment, groups of Wve female mice were administered a single subcutaneous injection of 50 mg/kg of testosterone or the vehicle alone and were killed at 1, 3, 6 or 24 h after the injection. All animals were anesthetized with an intraperitoneal injection of sodium pentobarbital (50 mg/kg) and killed by bleeding from the right atrium followed by transcardial perfusion with cold physiological saline. To make tissue homogenates for Western blotting, the SMG were dissected out, frozen immediately in liquid nitrogen and stored at ¡80°C until use. To make tissue sections for immunohistochemistry (IHC), the animals were Wxed by perfusion with cold 4% paraformaldehyde in 0.1 M phosphate buVer (pH 7.2), and the SMG were dissected out. They were further Wxed by immersion in the same Wxative for 4 h at 4°C, rinsed in 0.1 M phosphate buVer, dehydrated in a series of ethanol and embedded in paraYn.

Materials and methods Western blotting Antibodies, cell lysates and chemical reagents Rabbit polyclonal antibodies against t- and p-CREB with the corresponding immunizing peptides were purchased from Upstate Biotechnology Inc. (Lake Placid, NY). Rabbit polyclonal anti-NGF antibody, NIH/3T3 cell lysate and NIH/3T3 + forskolin cell lysate were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Mouse monoclonal antibodies against t-CREB and -actin were purchased

123

Frozen SMG tissues were homogenized in a lysis buVer containing 50 mM Tris-HCl (pH 7.5), 150 mM NaCl, 1% NP-40, 0.5% sodium deoxycholate, 0.1%SDS, protease inhibitor cocktail (CompleteTM; Roche Diagnostics, Indianapolis, IN, USA) and phosphatase inhibitor cocktail (PhosStopTM; Roche Diagnostics). The tissue homogenates and NIH/3T3 cell lysates were then separated on 12% SDS–polyacrylamide gels and transferred to PVDF


membranes (BioRad Laboratories, Hercules, CA). After being blocked with 5% non-fat skimmed milk in PBS, the membranes were incubated with rabbit or mouse primary antibody against t-CREB, p-CREB, NGF (1:1000 dilution) or -actin (1:20,000) overnight at 4°C. After being washed, the membranes were incubated with horseradish peroxidase-conjugated secondary antibody against rabbit or mouse IgG (1:2000) (Dako, Glostrup, Denmark) for 1 h. The immunoreaction was detected with a chemiluminescence kit (ECL-plusTM; Amersham Pharmacia Biotech, Uppsala, Sweden) and X-ray Wlms. For re-probing, the membranes were stripped with Re-blot Plus Mild Antibody Stripping SolutionTM (Chemicon International, Temecula, CA). The developed X-ray Wlms were converted to an image Wle by scanning them with an Epson GT-9800F scanner and the optical density of the immunoreactive band was quantiWed with the Image Gage version 3.41 software (Fuji Photo Film, Co., Tokyo, Japan). The relative intensity of the immunoreactive bands for t-CREB, p-CREB or NGF were expressed as the mean § SD of four to Wve samples after being normalized to the corresponding value for -actin. The statistical diVerence between two mean values was examined using Student’s t test, and values of P < 0.05 were considered to be signiWcant. IHC ParaYn sections of the SMG (5-m thick) were made with a microtome and mounted on silanized glass slides (Dako, Glostrup, Denmark). After being deparaYnized in xylene, the sections were pre-treated successively with 0.3% H2O2 in methanol for 10 min and 5% normal goat serum for 30 min. For IHC by the enzyme-detection method, the sections were incubated overnight at 4°C with rabbit antit-CREB antibody (1:1000 dilution), rabbit anti-p-CREB antibody (1:1000) or rabbit anti-NGF antibody (1:500). To conWrm the speciWcity of the immunoreactions, the primary antibodies against t- and p-CREB were absorbed with the corresponding immunizing peptides (10 g for 1 g IgG) for 1 h at room temperature prior to and throughout the incubation with the sections at 4°C. After being washed with PBS, the sites of immunoreaction were visualized by incubating the sections successively with biotinylated antirabbit IgG antibody (1:200) (Vector Laboratories, Burlingame, CA) for 1 h, horseradish peroxidase-conjugated streptavidin (1:300) (Dako) for 1 h, and a peroxidase substrate (ImmPACT DABTM; Vector Laboratories, Burlingame, CA) for about 5 min. The sections were then observed under an Olympus BX50 microscope. For Xuorescent double-immunostaining, the pretreated sections were incubated with mouse anti-t-CREB antibody (1:500) mixed with rabbit anti-p-CREB or anti-NGF antibody (1:500) overnight at 4°C. After being washed, the sections were

649

incubated with a mixture of anti-rabbit IgG antibody conjugated with Alexa Fluor 594 (1:400) and anti-mouse IgG antibody conjugated with Alexa Fluor 488 (1:400) (Molecular Probes; Eugene, OR) for 1 h. They were then mounted in glycerol and observed with an Olympus BX50/BX-FLA Xuorescent microscope.

Results Sexual dimorphism in the expression of CREB in the SMG First, the speciWcity of the present antibodies against t- and p-CREB was examined by Western blot analysis using lysates of mouse NIH/3T3 cells treated and untreated with forskolin, which is known to enrich p-CREB by activating the cAMP signaling pathway (Fig. 1). Both the rabbit and mouse anti-t-CREB antibodies formed single immunoreactive bands (molecular weight: 43 kDa) with similar intensities in the NIH/3T3 and NIH/3T3 + forskolin cell lysates. In contrast, rabbit anti-p-CREB antibody formed a 43-kDa band, which showed a substantially stronger intensity in the NIH/3T3 + forskolin than in the NIH/3T3 cell lysates. These results indicated that the anti-t-CREB antibodies used react with both phosphorylated and unphosphorylated CREB, whereas the anti-p-CREB antibody speciWcally reacts with phosphorylated CREB. Then, the homogenates of the adult male and female SMG were analyzed using rabbit anti-t-CREB and anti-p-CREB antibodies (Fig. 2a), revealing a marked sexual dimorphism in the production of both t- and p-CREB. Both antigens formed stronger

Fig. 1 Western blot analysis of the speciWcity of the antibodies against t- and p-CREB. The cell lysates of NIH/3T3 (3T3) and NIH/ 3T3 + forskolin (3T3 + f) were electrophoresed, blotted and immunostained with rabbit anti-t-CREB, mouse anti-t-CREB and rabbit antip-CREB antibodies. For the control, staining with mouse anti--actin antibody was also performed. A representative result is shown. The molecular weights (kD) are indicated

123

650


Fig. 2 Western blot analysis of the expression of t-CREB, p-CREB, and NGF in the SMG of male and female mice. a The protein samples were electrophoresed, blotted and immunostained with rabbit antibodies against t-CREB, p-CREB and NGF. For the control, staining with mouse anti--actin antibody was also performed. A representative result is shown. The molecular weights (kD) of the immunoreactive

bands are indicated. b The relative levels of t-CREB and p-CREB in the male and female samples in relation to the corresponding -actin levels were determined by densitometric analysis and presented as the mean § SD of Wve samples. For both t-CREB and p-CREB, the male values are arbitrarily set at 1. **SigniWcantly diVerent from the male value (P < 0.01)

Fig. 3 Western blot analysis of the expression of t-CREB, p-CREB and NGF in the SMG of male and female mice after repeated hormonal treatments. a The protein samples from the male mice, testectomized male mice (Tx), testectomized male mice treated for 5 days (Tx + 5d) with testosterone (T) or DHT, female mice and female mice treated for 5 days (F + 5d) with testosterone or DHT were electrophoresed and blotted. b The protein samples from the testectomized male mice treated

for 5 days with the vehicle alone (Oil), Xutamide alone (Flu), Xutamide plus DHT (Flu + DHT) and DHT alone were electrophoresed and blotted. A, B The blots were immunostained with rabbit antibodies against t-CREB, p-CREB and NGF. For the control, staining with mouse anti--actin antibody was also performed. Representative results are shown. The molecular weights (kD) of the immunoreactive bands are indicated

immunoreactive bands of 43 kDa in the female than in the male gland, in contrast with the band for NGF, which was stronger in the male than in the female gland. In quantitative analysis of the immunoreactive bands, the relative levels of t- and p-CREB, after normalization to the level of -actin, were found to be 7.4-fold and 7.9-fold higher, respectively, in the female than in the male gland (P < 0.01) (Fig. 2b). The ratio of the level of p-CREB to that of t-CREB in each sample, which represents the activity of CREB, showed no signiWcant diVerence between the sexes.

regulation, the eVects of testectomy as well as repeated daily administration of androgens for 5 days on the expression of t- and p-CREB in the SMG were examined. As shown by Western blotting, testectomy caused a substantial increase in the levels of both t- and p-CREB, whereas the replacement of testosterone reversed it in the male SMG (Fig. 3a). The administration of testosterone to females also caused a substantial decrease in the levels of t- and p-CREB. The administration of DHT, which is an intracellular metabolite of testosterone and represents the non-aromatizable androgen, to females or testectomized males caused the same eVect as that of testosterone, conWrming that androgens were responsible for reducing the levels of CREB in the SMG. Finally, repeated administration of Xutamide, which binds to the AR and competitively inhibits the action of androgens, at 1 h before that

Dependency of the expression of CREB on the androgen-AR system To clarify whether the observed sexual dimorphism in the level of CREB in the mouse SMG resulted from hormonal

123


651

Fig. 4 Immunohistochemical localization of t-CREB, p-CREB and NGF in the SMG of adult male and female mice. Serial paraYn sections of male (a, c, e) and female (b, d, f) glands were immunostained with rabbit antibodies against t-CREB (a, b), p-CREB (c, d) and NGF (e, f). a, c The duct system of the male gland is composed mostly of the ID and GCT. Strong immunoreactivity for both t-CREB and p-CREB is present in the nuclei of ID cells (ID), but absent in the nuclei of GCT cells (GCT). b, d The duct system of the female gland is composed mostly of the ID and SD. Both t-CREB and p-CREB are present in the nuclei of ID cells and the cells located in the distal portions of the SD (dSD) adjacent to the ID. e, f NGF-immunopositive GCT cells occupy most of the duct portions other than the ID in the male gland, but are scattered amongst immunonegative SD cells in the female gland. Acinar cells (a) are immunonegative for all three factors. Bar = 50 m

of DHT to the testectomized males partially prevented the decreases of the levels of t- and p-CREB; whereas, the administration of Xutamide alone had no eVect, suggesting that the androgen-induced reduction in the level of CREB in the SMG is dependent, at least partly, on the function of the AR (Fig. 3b). The repeated administration of vehicle (oil) alone caused no change in the levels of t- or p-CREB. The eVectiveness of testectomy and hormonal treatment in this experiment was veriWed by the inverse relationship between the changes in the level of NGF, a marker of the male phenotype of SMG and those of CREB.

Localization of CREB in the duct system of the SMG In serial paraYn sections of the adult male and female SMG immunostained for t-CREB, p-CREB and NGF with rabbit antibodies, the duct system of the male gland was largely occupied by the GCT, the epithelial cells of which contained NGF-immunopositive secretory granules (Fig. 4e). In contrast, the SD occupied the largest part of the duct system of the female gland, with only a small number of NGF-positive GCT cells scattered amongst the SD cells (Fig. 4f). In the male gland, intense immunostaining for both t- and p-CREB was localized exclusively to the nuclei

123

652


Fig. 5 Negative controls of the immunostaining for t-CREB and p-CREB. Serial paraYn sections of female SMG representing the same area as in Fig. 4b, d were immunostained with anti-t-CREB antibody absorbed with t-CREB immunizing peptide (a) and anti-p-CREB

antibody absorbed with p-CREB immunizing peptide (b). No reactivity for t-CREB or p-CREB is observed in the nuclei of ID cells (ID) or the cells in the distal portions of the SD (dSD). A acini, pSD proximal portions of the SD. Bar = 50 m

of ID cells and was not detected in GCT cells or acinar cells (Fig. 4a, c). In the female gland, in contrast, intense immunoreactivity for both antigens was localized to the nuclei of ID cells and cells occupying the distal portions of the SD adjacent to the ID (Fig. 4b, d). Most of the cells in the proximal portions of the SD, as well as acinar cells, were immunonegative for t- and p-CREB. In both male and female glands, the nuclei of the excretory duct cells showed a weak immunoreactivity for both antigens (picture not shown). When the primary antibodies against t- and p-CREB were absorbed with the corresponding immunizing peptides, no immunostaining was observed in any cell or structure, conWrming the speciWcity of the immunoreactions (Fig. 5a, b). The result of the immunostaining with mouse anti-t-CREB antibody was exactly the same as that with rabbit antibody (picture not shown).

complete absence of t-CREB immunoreactivity in the diVerentiated GCT cells that had been labeled for NGF (Fig. 6c, e). In the female gland, the cells of the proximal portions of the SD decreased their t-CREB immunoreactivity to some extent; whereas, those of the distal portions continued to show strong reactivity at 5 W (Fig. 6d). No NGF-labeled GCT cells were seen in the female gland at 5 W (Fig. 6f).

Postnatal changes in the expression and localization of CREB We then examined the changes in the expression and localization of t-CREB during postnatal development of the male and female SMG. Early in the postnatal period at 1–2 W, the duct system was composed primarily of the ID and SD. T-CREB immunoreactivity was present in most of the cell nuclei in the duct system in both sexes (Fig. 6a, b). At 4–6 W, the female duct system was still mostly composed of SD cells, whereas extensive diVerentiation of SD cells into GCT cells was seen in the male duct system, a phenomenon known to be androgen dependent (Chretien 1977). The proportion of immunopositive duct cells decreased in both sexes during this period, but the extent of the decrease was substantially larger in the male than in the female gland, because of the

123

EVect of testosterone on the expression and localization of CREB To further clarify their relationship with GCT-cell diVerentiation, the expression and localization of t- and p-CREB in the SMG were examined at diVerent intervals after a single administration of testosterone to adult female mice. At 1 h after a subcutaneous injection of 50 mg/kg of testosterone, there was no signiWcant change in the levels of t- or p-CREB compared with the control levels observed after the injection of vehicle (oil) alone (Fig. 7a–c). Thereafter, the levels of both t- and p-CREB were signiWcantly elevated, reaching 1.33 and 1.56 times those of the controls (P < 0.05), respectively, at 3 h. The levels were further increased to 2.10 and 1.75 times those of the controls (P < 0.01), respectively, at 6 h, but declined to the control levels by 24 h. In IHC, a temporary increase in the number of t- and p-CREB-positive nuclei in the duct system was observed from 3 h through 6 h after the administration of testosterone (Fig. 8a, b, d, e). Many immunopositive cell nuclei newly appeared in the proximal portions of the SD, in addition to those in the ID and the distal portions of the SD. At these time points, the number of NGF-positive GCT cells remained unchanged from that before stimulation (Fig. 8c, f). By 24 h after the


653

Fig. 6 Immunohistochemical localization of t-CREB and NGF in the duct system of male and female submandibular glands during postnatal development. Serial paraYn sections of male (a, c, e) and female (b, d, f) glands at 2 W (a, b) and 5 W (c–f) postpartum were immunostained with rabbit anti-t-CREB (a–d) and anti-NGF (e, f) antibodies. a, b The duct system is composed of ID (ID) and SD (SD) cells, all of which are immunostained for CREB in the nuclei, in both the male and female glands. c, e The duct system is composed of ID cells, a small number of remaining SD cells adjacent to the ID and abundant GCT cells (GCT) that have been positively immunostained for NGF in the cytoplasm. CREB staining is positive in the nuclei of ID and SD cells, but negative in GCT cells. d, f The duct system is composed of ID and SD cells, but lack in NGF-positive GCT cells. CREB staining is strong in the nuclei of ID cells and cells in the distal portions of the SD (dSD), but weaker in those of the cells in the proximal portions of the SD (pSD). Acinar cells (A) are immunonegative in both sexes and both ages. Bar = 50 m

stimulation, the majority of SD cells had been converted into GCT cells with clear secretory granules that were immunopositive for NGF, similar to those of the male gland (Fig. 8i). No t- or p-CREB immunoreactivity was present in the nuclei of the diVerentiated GCT cells (Fig. 8g, h). The double immunoXuorescence indicated that, at all time points before and after the administration of testosterone, the cells positive for nuclear t-CREB reactivity overlapped with those that were positive for nuclear p-CREB reactivity almost completely, but never overlapped with those that were positive for cytoplasmic NGF reactivity (Fig. 8b, c, e, f, h, i).

Discussion In the present study, we have demonstrated that the expression of the transcription factor CREB in the adult mouse SMG has a pronounced sexual dimorphism, in which CREB is much more abundant in the female than in the male gland. No such sexual dimorphism was apparent in the rat SMG and therefore, was not analyzed in our previous study (Amano and Iseki 1998; Kim et al. 2001). No signiWcant diVerence was found between the sexes in the ratio of p-CREB to t-CREB, suggesting that the absolute amount of CREB, but not the extent of the activation of CREB, is

123

654


Fig. 7 Western blot analysis of the time course of the expression of t-CREB and p-CREB in the SMG of female mice stimulated with testosterone. a The protein samples of adult female glands at 1, 3, 6 and 24 h after a single administration of vehicle alone (V) or testosterone (T) were electrophoresed, blotted and immunostained with rabbit antibodies against t-CREB and p-CREB. For the control, staining with mouse anti--actin antibody was also performed. A representative result is shown. The molecular weights (kDa) of the immunoreactive

bands are indicated. b, c The relative levels of t-CREB (b) and p-CREB (c) in the testosterone-administered and vehicle-administered (control) groups in relation to the corresponding -actin levels at the same time points were determined by densitometric analysis and presented as the mean § SD of four to Wve samples. For both t-CREB and p-CREB, the values of the control group were arbitrarily set at 1. * and **SigniWcantly diVerent from the control values at P < 0.05 and P < 0.01, respectively

higher in the female than in the male gland. Based on the results of IHC, this sexual dimorphism can be accounted for by a lack of CREB in the nuclei of GCT cells, which are known to diVerentiate from SD cells preferentially in the male gland around 3–5 W postpartum (Cutler and Chaundhry 1975; Srinivasan and Chang 1975). Furthermore, the results of castration and of the administration of testosterone, DHT or Xutamide indicate that the androgenAR system is responsible for the disappearance of CREB associated with the diVerentiation of SD cells into GCT cells. Unlike GCT cells, the majority of SD cells lack secretory granules, but instead have extensive basal striations. However, the cells in the distal portions of the SD adjacent to the ID have both basal striations and apical granules and are designated as striated granular duct (SGD) cells (Caramia 1966a). SGD cells are much more common in the female gland, representing another sexual dimorphism of the mouse SMG. In the developing SMG, it is generally accepted that ID cells serve as stem cells that proliferate and give rise to both acinar and duct cell populations (Zajicek et al. 1985). On the other hand, there is evidence in the adult male SMG, where most SD cells have been

converted to GCT cells, that ID cells, as well as SGD cells, primarily give rise to the adjacent GCT cells and not to acinar cells, suggesting that SGD cells are the intermediate form between ID and GCT cells (Denny et al. 1993, 1999). In the present study, intense CREB immunoreactivity was found in the nuclei of ID cells in both sexes and those of a subpopulation of SD cells located in the distal portions of the SD in females, which is likely to represent SGD cells. We hypothesize that nuclear accumulation of CREB in ID cells is a prerequisite for these cells to diVerentiate into GCT cells, in which CREB has been lost. In the female gland, SGD cells may represent cells that were interrupted when diVerentiating into GCT cells and so retained CREB in their nuclei. The present study has also demonstrated that after a single administration of testosterone to females, the cells in the proximal portions of the SD become temporarily positive for CREB immunoreactivity in the nuclei at 3–6 h after the administration, but then lose it as they are converted to GCT cells by 24 h. This phenomenon suggests that the successive accumulation and loss of CREB in the nuclei is associated with androgen-induced diVerentiation of SD cells to GCT cells. We hypothesize that SD cells in the adult female gland, unlike ID and SGD cells, normally

123


655

Fig. 8 Immunohistochemical localization of t-CREB, p-CREB and NGF in the SMG of female mice stimulated with testosterone. Serial paraYn sections of adult female glands at 0 h (a–c), 6 h (d–f) and 24 h (g–i) after a single administration of testosterone were immunostained with rabbit anti-t-CREB (a, d, g), mouse anti-t-CREB plus rabbit antip-CREB (b, e, h), and mouse anti-t-CREB plus rabbit anti-NGF (c, f, i) antibodies and visualized with DAB (a, d, g) and Xuorescence (b, c, e, f, h, i). a–c Both t- and p-CREB immuoreactivities are localized to the nuclei of ID cells (ID) and cells in the distal portions of the SD (dSD), but are not detected in cells in the proximal portions of the SD (pSD),

including the NGF-positive GCT cells. d–f Many SD cells in the proximal portions of the SD have turned positive for t- and p-CREB in the nuclei. g–i Most SD cells have been replaced by NGF-positive GCT cells (GCT), which are negative for t- or p-CREB. Only ID cells and yet remaining SD cells in the distal portions of the SD (dSD) are positive for t- or p-CREB. b, c, e, f, h, i The merged double immunoXuorescence pictures indicate that cells that show nuclear immunoreactivity for t-CREB (green) overlap with those for p-CREB (red) almost completely, but never overlap with the cells that show cytoplasmic reactivity for NGF (red). Bar = 50 m

avoid diVerentiating into GCT cells, but can be induced by androgens to enter this process. The nuclear accumulation of CREB seems to be a prerequisite for this process, but is not required in diVerentiated GCT cells. The present study has suggested a relationship between the androgen-AR system and signaling pathways involving CREB in the diVerentiation of the duct system in the mouse SMG. The activity of CREB is regulated at the posttranslational level by protein phosphorylation. A variety of signal-

ing pathways originating in the membrane receptors cause phosphorylation of CREB and thereby increase its transactivating potential (Shaywitz and Greenberg 1999; Johannessen and Moens 2007; Johannessen et al. 2004). On the other hand, there is evidence that androgens can exert their eVects in a non-genomic manner by interacting with signaling molecules other than the AR (Heinlein and Chang 2002; Foradori et al. 2008; Michels and Hope 2008). For instance, in cultured Sertoli cells, the eVects of testosterone leading

123

656

to spermatogenesis involve the association of the classical AR with membrane Src kinase followed by the activation of ERK, the major factor of the MAP kinase pathway, and Wnally the activation of CREB (Fix et al. 2004; Cheng et al. 2007). It is tempting to assume that a similar mechanism of non-genomic androgen action involving CREB operates in the duct cells of the mouse SMG. However, the present study has indicated that changes in the levels of t- and p-CREB always occur in parallel, suggesting that androgens regulate the amount of CREB (e.g., its temporary increase in SD cells followed by its permanent decrease in GCT cells) rather than the activation of CREB. Interestingly, our previous study demonstrated that the amount of another transcription factor, JunD, undergoes almost the same pattern of androgen-dependent changes as that of CREB in the mouse SMG (Hipkaeo et al. 2004, 2008). To date, the regulation of the amount of CREB remains poorly understood, except that the transcription of CREB mRNA is upregulated by several mechanisms, including the positive autoregulation of the CREB gene by binding of activated CREB to the CRE located in the CREB promotor (Walker et al. 1995). The androgen-induced diVerentiation of duct cells in the mouse SMG may provide an attractive system for investigating the regulatory mechanism and biological signiWcance of the accumulation and loss of certain transcription factors including CREB. Acknowledgments This work was supported by a Grant-in-Aid for ScientiWc Research from the Ministry of Education, Science and Culture of Japan to SI. We thank S.Yamazaki for technical assistance.

References Amano O, Iseki S (1998) Occurrence and nuclear localization of cAMP response element-binding protein in the postnatal development of the rat submandibular gland. Histochem J 30:591–601 Barka T (1980) Biologically active polypeptides in submandibular glands. J Histochem Cytochem 28:836–859 Caramia F (1966a) Ultrastructure of the mouse submaxillary gland. I. Sexual diVerences. J Ultrastruct Res 16:333–345 Caramia F (1966b) Ultrastructure of the mouse submaxillary gland. II. EVect of castration in the male. J Ultrastruct Res 16:505–523 Chang C, Salzman A, Yeh S, Young W, Keller E, Lee HJ, Wang C, Mizokami A (1995) Androgen receptor: an overview. Crit Rev Eukar Gene Expr 5:97–125 Cheng J, Watkins SC, Walker WH (2007) Testosterone activates mitogen-activated protein kinase via Src kinase and the epidermal growth factor receptor in Sertoli cells. Endocrinology 148:2066– 2074 Chretien M (1977) Action of testosterone on the diVerentiation and secretory activity of a target organ: the submaxillary gland of the mouse. Int Rev Cytol 50:333–396 Cutler LS, Chaundhry AP (1975) CytodiVerentiation of striated duct cells and secretory cells of the convoluted granular tubules of the rat submandibular gland. Am J Anat 143:201–218 Denny PC, Chai Y, Klauser DK, Denny PA (1993) Parenchymal cell proliferation and mechanisms for maintenance of granular duct

123

Histochem Cell Biol (2009) 132:647–657 and acinar cell populations in adult male mouse submandibular gland. Anat Rec 235:475–485 Denny PC, Liu P, Denny PA (1999) Evidence of a phenotypically determined ductal cell lineage in mouse salivary glands. Anat Rec 256:84–90 Fix C, Jordan C, Cano P, Walker WH (2004) Testosterone activates mitogen-activated protein kinase and the cAMP response element binding protein transcription factor in Sertoli cells. Proc Natl Acad Sci USA 101:10919–10924 Foradori CD, Weiser MJ, Handa RJ (2008) Non-genomic actions of androgens. Front Neuroendocrinol 29:169–181 Gresik EW (1980) Postnatal development changes in submandibular glands of rats and mice. J Histochem Cytochem 28:860–870 Gresik EW (1994) The granular convoluted tubule (GCT) cell of rodent submandibular gland. Microsc Res Technol 27:1–24 Gubits RM, Shaw PA, Gresik EW, Onetti-Muda A, Barka T (1986) Epidermal growth factor gene expression is regulated diVerently in mouse kidney and submandibular gland. Endocrinology 119:1382–1387 Heinlein CA, Chang C (2002) The roles of androgen receptors and androgen-binding proteins in nongenomic androgen actions. Mol Endocrinol 16:2181–2187 Hipkaeo W, Wakayama T, Yamamoto M, Iseki S (2004) Expression and localization of the transcription factor JunD in the duct system of mouse submandibular gland. J Histochem Cytochem 52:479–490 Hipkaeo W, Sakulsak N, Wakayama T, Yamamoto M, Nakaya M, Keattikunpairoj S, Kurobo M, Iseki S (2008) Coexpression of menin and JunD during the duct cell diVerentiation in mouse submandibular gland. Tohoku J Exp Med 214:23–245 Jacoby F, Leeson C (1959) The postnatal development of the rat submaxillary gland. J Anat 93:201–206 Johannessen M, Moens U (2007) Multisite phosphorylation of the cAMP response element-binding protein (CREB) by a diversity of protein kinases. Front Biosci 12:1814–1832 Johannessen M, Delghandi MP, Moens U (2004) What turns CREB on? Cell Signal 16:1211–1227 Kim J-G, Amano O, Wakayama T, Takahagi H, Iseki S (2001) The role of cyclic AMP response element-binding protein in testosteroneinduced diVerentiation of granular convoluted tubule cells in the rat submandibular gland. Arch Oral Biol 46:495–507 Losel RM, Falkenstein E, Feuring M, Schultz A, Tillmann HC, RossolHaseroth K, Wehling M (2003) Nongenomic steroid action: controversies, questions, and answers. Physiol Rev 83:965–1016 Michels G, Hoppe UC (2008) Rapid actions of androgens. Front Neuroendocrinol 29:182–198 Morrell JI, Gresik EW, Barka T (1987) Autoradiographic localization of dihydrotestosterone binding in the major salivary glands and other androgen-responsive organs of the mouse. J Histochem Cytochem 35:1053–1058 PinkstaV CA (1980) The cytology of salivary glands. Int Rev Cytol 63:141–161 Sar M, Lubahn DB, French FS, Wilson EM (1990) Immunohistochemiccal localization of the androgen receptor in rat and human tissues. Endocrinology 127:3180–3186 Schmidt BMW, Gerdes D, Feuring M, Falkenstein E, Christ M, Wehling M (2000) Rapid, nongenomic steroid actions: a new age? Front Neuroendocrinol 21:57–94 Shaywitz AJ, Greenberg ME (1999) CREB: a stimulus-induced transcription factor activated by a diverse array of extracellular signals. Annu Rev Biochem 68:821–861 Srinivasan R, Chang WWL (1975) The development of the granular convoluted duct in the rat submandibular gland. Anat Rec 182:29–40 Walker WH, Fucci L, Habener JF (1995) Expression of the gene encoding transcription factor adenosine 3⬘, 5⬘-monophosphate

Histochem Cell Biol (2009) 132:647–657 (cAMP) response element-binding protein: regulation by folliclestimulating hormone-induced cAMP signaling in primary rat Sertoli cells. Endocrinology 136:3534–3545 Wehling M (1997) SpeciWc, nongenomic actions of steroid hormones. Annu Rev Physiol 59:365–393

657 Zajicek G, Yagil C, Michaeli Y (1985) The streaming submandibular gland. Anat Rec 213:150–158 Zhou Z-X, Wong C-I, Sar M, Wilson EM (1994) The androgen receptor: an overview. Rec Prog Horm Res 49:249–274

123