- Email: [email protected]
Measurement of Activin B in Human Saliva and Localization of Activin Subunits in Rat Salivary Glands
JOBNAME: BBRC 222#2 PAGE: 1 SESS: 8 OUTPUT: Wed May 29 08:37:10 1996 /xypage/worksmart/tsp000/71401a/4
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS ARTICLE NO.
222, 230–235 (1996)
0727
Measurement of Activin B in Human Saliva and Localization of Activin Subunits in Rat Salivary Glands Merja Bläuer,*,1 Lars Wichmann,† Reijo Punnonen,† and Pentti Tuohimaa* *Department of Anatomy, Medical School, University of Tampere; and †Department of Obstetrics and Gynecology, Tampere University Hospital, FIN-33101 Tampere, Finland Received April 4, 1996 The present paper is the first report to demonstrate that measurable amounts of activin B are secreted into the saliva. The results show wide fluctuations in activin B concentrations during the menstrual cycle with peak values detected at the follicular phase. Estrogen replacement therapy was found to increase salivary activin B levels in post-menopausal subjects. The concentration in males was negligible. These data suggest that activin B concentrations in the human saliva may be under hormonal regulation and propose that salivary activin B measurements may prove useful in investigating the as yet less well defined local and/or physiological roles of activin B. The present paper reports, in addition, the immunohistochemical localization of activin/inhibin subunits in the duct systems of the rat submandibular, sublingual and parotid salivary glands. © 1996 Academic Press, Inc.
Activins and inhibins belong to the TGFb superfamily and have been shown to possess multiple functions in cell growth and differentiation, reproductive processes and early embryonic development (1). Structurally, activins are disulfide-linked dimeric proteins composed of two homologous but distinct b-subunits, bA and bB. Recently, putative new activin BC and bD chains were also discovered (2,3). So far, three forms of activins are known: activin A (bAbA), activin AB (bAbB) and activin B (bBbB) which is the latest dimeric form isolated from natural sources (4). Inhibins are heterodimers of either of the two b-subunits and a related a-subunit (1). There is considerable interest in activin and inhibin measurement in biological fluids. The hormones were initially isolated from ovarian follicular fluid and, later, measurements have been made in serum, seminal plasma, amniotic fluid and various tissue extracts (5). Recently, several assays have been developed and quantitative studies made to screen activin/inhibin concentrations in human sera to investigate the normal physiological role of these regulatory factors and in search for possible clinical applications (6–13). Up till now, little is known about their diagnostic value (14). The present paper introduces the saliva as an interesting new target for quantitative activin research. We have investigated the level of activin B in human saliva using a recently developed immunoenzymometric assay (IEMA) and determined, in addition, the immunohistochemical localization of activin/inhibin subunit proteins in rat salivary glands. MATERIALS AND METHODS Saliva samples. Basal saliva samples were collected from 17 healthy adults from the following groups: females with normal menstrual cycles (n45); postmenopausal females without (n42) or with (n44) estrogen replacement therapy; males <50 y, (n44); males >60 y (n42). Saliva was collected in the morning, and no eating or drinking was allowed for 2h prior to collection. The samples were immediately frozen and stored at −20°C until analyzed. For activin B measurement the samples were thawed at room temperature and centrifuged at 16 000 × g for 10 min. The first group collected samples daily during one menstrual cycle, the other groups for a period of at least one week. LH measurement. To estimate the time of ovulation (group A) the day of LH surge was determined from urine samples collected in the morning and evening at days 8–18 of the menstrual cycle. A two-site IEMA based on two monoclonal
1
To whom correspondence should be addressed. 230
0006-291X/96 $18.00 Copyright © 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
JOBNAME: BBRC 222#2 PAGE: 2 SESS: 7 OUTPUT: Wed May 29 08:37:10 1996 /xypage/worksmart/tsp000/71401a/4
Vol. 222, No. 2, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
anti-human LH antibodies (Medix Biochemica, Kauniainen, Finland) was applied (15). Human LH (0.5–200 I.U./l) was used as standard. Activin B IEMA. Activin B concentrations were measured using an IEMA based on the monoclonal anti-bB(101–115) antibody (16). The IEMA is described in detail in Bläuer et al.(17). In brief, standards and diluted saliva samples were added to the anti-bB(101–115) precoated microtitre plates and incubated for 2 h at 37°C. After thorough washing, biotinylated anti-bB(101–115) was added and incubated for 1h at 37°C. The plates were washed and avidin-horseradish peroxidase was added and allowed to bind for 15 min at room temperature. After a further washing step the OPD-hydrogen peroxide substrate was added and the plates incubated for 5 min at room temperature. The absorbances were read at 492 nm after stopping the enzyme reaction with H2SO4. Data are presented as arbitrary units. A pregnant human serum pool used as standard was given the arbitrary value of 100 U activin B/ml. Animals. Adult male and female Sprague-Dawley rats were killed by cervical dislocation and their submandibular, sublingual and parotid salivary glands were removed. The tissues were fixed in ice-cold Bouin’s fluid for 2.5 h and embedded in paraffin. Immunohistochemistry. Immunohistochemical staining was performed essentially as described in Bläuer et al. (18). Rabbit polyclonal antibodies against synthetic human inhibin a(1–32) and bA(88–102) peptides were used (18). The anti-bB antibody was the same as used for IEMA.
RESULTS Dilution linearity. Serial dilutions of two saliva samples were made and assayed for activin B. Figure 1 shows that saliva diluted in parallel to the serum standard. The dilution curves were linear over the range from 1:5 to 1:20 (r 4 0.98). Activin B concentration in human saliva. The data are summarized in Table 1. IEMA results for four subjects having normal menstrual cycles are shown in Figure 2. In all cases, a peak in activin B concentration was observed before the day of LH surge. Immunohistochemistry. In the submandibular gland, bB-subunit immunoreactivity was localized in the epithelial cells of the striated ducts (SD), excretory ducts (ED) and granular convoluted tubules (GCT) (Fig. 3a,c). Secretory acini were not immunoreactive. Duct cells with varying staining intensities were seen, and in the SD and ED bB-subunit immunoreactivity was predominantly confined to the apical cytoplasm. Only some cells in the GCT were immunoreactive (Fig. 3a). In the parotid and sublingual glands, bB-subunit immunoreactivity was observed in the epithelial cells of the SD (Fig. 3d,e). The staining in the ED was faint (data not shown). In the sublingual gland, the intercalated ducts were strongly immunopositive (Fig. 3d). bA-subunit immunoreactivity was demonstrated in some GCT cells and on or close to the luminal surface of the SD in the submandibular, sublingual and parotid salivary glands (Fig. 4).
FIG. 1. Dose-response curves of the standard serum (v) and two saliva samples (■) and (m). Serial dilutions were made in assay buffer and analyzed for activin B by the IEMA. Each point represents the mean value of duplicate determinations. 231
JOBNAME: BBRC 222#2 PAGE: 3 SESS: 8 OUTPUT: Wed May 29 08:37:10 1996 /xypage/worksmart/tsp000/71401a/4
Vol. 222, No. 2, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS TABLE 1 Activin B Levels in Basal Saliva Group /
Normal menstrual cycle
Postmenopausal; no estrogen replacement Postmenopausal; estrogen replacement ?
<50y
>60y
Subject
Days
Activin B U/ml ±SD
a b c d e a b
30 29 30 30 30 14 14
9.2 ± 16.1 7.8 ± 7.2 5.4 ± 6.0 5.2 ± 13.6 0.2 ± 0.4 0.4 ± 0.2 0.09 ± 0.2
a* b* c d a b c d a b
7 7 7 7 13 10 14 14 14 14
1.4 ± 0.9 1.1 ± 1.2 1.0 ± 0.6 0.07 ± 0.2 0.7 ± 0.4 0.4 ± 0.5 0.1 ± 0.2 0 0.04 ± 0.09 0.01 ± 0.05
* Hysterectomy and bilateral ovariectomy.
In all glands, immunohistochemical staining was negative when absorbed antibodies were used (Fig. 3b, 4b). No differences could be demonstrated in the localization of immunoreactivity between females and males in the samples studied. Immunoreactive a-subunits were not observed in any of the three glands.
FIG. 2. Salivary activin B concentrations during the menstrual cycle. a, b, c, and d show the activin B concentrations in saliva samples collected daily by four subjects during one menstrual cycle. The day of the LH surge is indicated by an arrow. 232
JOBNAME: BBRC 222#2 PAGE: 4 SESS: 7 OUTPUT: Wed May 29 08:37:10 1996 /xypage/worksmart/tsp000/71401a/4
Vol. 222, No. 2, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 3. Expression of activin bB-subunits in rat salivary glands. a, In the submandibular gland, immunoreactive cells are seen in the granular convoluted tubules (arrows) and striated ducts (arrowheads). b, A control section incubated with anti-bB(101–115) preabsorbed with a 50-fold molar excess of peptide bB(101–115). A granular convoluted tubule and a striated duct are indicated by a large and a small asterisk, respectively. c, A submandibular gland excretory duct showing apical localization of bB-subunit immunoreactivity in its epithelial cells (arrowheads). d, In the sublingual gland, striated ducts (arrowheads) and intercalated ducts (arrow) were immunopositive. e, A less intense immunoreaction is seen in parotid striated ducts (arrowheads). Bars, 25 mm.
DISCUSSION Salivary glands have been shown to be a rich source of a multitude of peptide growth factors (19–23). It has been demonstrated, that salivary growth factors significantly affect oral and esophageal mucosal morphology and function (24,25). Furthermore, there is substantial evidence that some of these growth factors, in addition to being components of the saliva, also may have endocrine effects (23,26). The concentration of some of the growth factors in the murine submandibular gland appears to be regulated by hormonal steroids (27,28). Also in humans, the composition of the female saliva has been shown to vary with the menstrual cycle (29) and in pregnancy (30), suggesting an involvement of ovarian hormones in the secretory function of the salivary glands. Many salivary enzymes show a midcycle peak coinciding with the time of ovulation (29). This study represents the first evidence that immunoreactive activin B is present in the human saliva. When assayed with the recently developed IEMA, salivary activin B diluted in parallel to the serum standard, indicating little interference by other salivary components with assay performance. A stimulating effect of estrogens on salivary gland activin B secretion is suggested, based on increased activin B levels during the menstrual cycle and estrogen replacement therapy. In both groups, one subject showed negligible salivary activin B concentrations. The causes and potential significance of the two exceptions are not known but they may reflect disorders, irregularity or individual variation in ovarian function and estrogen effect. 233
JOBNAME: BBRC 222#2 PAGE: 5 SESS: 7 OUTPUT: Wed May 29 08:37:10 1996 /xypage/worksmart/tsp000/71401a/4
Vol. 222, No. 2, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 4. Activin bA-subunit immunoreactivity in rat salivary glands. Figure a shows immunoreactive cells in the granular convoluted tubules of the submandibular gland (arrows). Arrowheads indicate immunoreactivity on or close to the surface of striated ducts. b, A control section incubated with anti-bA (88–102) preabsorbed with a 50-fold molar excess of peptide bA(88–102). A granular convoluted tubule and a striated duct are indicated by a large and a small asterisk, respectively. c and d show a sparse distribution of bA-subunit immunoreactivity (arrowheads) in the striated ducts in the sublingual and parotid salivary glands, respectively. Bars, 25 mm.
The present paper is the first report to describe the localization of immunoreactive activin subunit proteins in salivary glands. Previously, the presence of activin bB-subunit mRNA has been described in the salivary glands of rat embryos (31,32). In mouse embryos, bB-subunit transcripts have been localized in the developing acinar epithelial cells and bA-subunit transcripts in the smooth muscle cells surrounding the salivary duct and some larger vessels (33). We demonstrated here that immunoreactive activin bB was localized in the ductal epithelial cells in adult rat salivary glands. The secretory acini were immunonegative, as previously shown also for other growth factors in salivary glands (19,22). Interestingly, in the present study, activin bA-subunit immunoreactivity was predominantly localized in the epithelial cells of the granular convoluted tubules in the submandibular gland, which in rodents is known to exhibit sexual dimorphism (34). In conclusion, this study shows that immunoreactive activin subunits and activin B are produced by the salivary glands. The data on activin B levels in the human saliva suggest a connection between salivary activin B and the female reproductive physiology. Salivary activin B measurements may prove useful in monitoring ovarian function and the physiological response to estrogen treatment. ACKNOWLEDGMENTS We thank Pekka Vilja, Ph.D., for biotinylation of the antibodies and Ms. Hilkka Mäkinen and Mr. Olli Ström for technical assistance. This work was supported by grants from the Tampere University Hospital Research Foundation and the University of Tampere. 234
JOBNAME: BBRC 222#2 PAGE: 6 SESS: 7 OUTPUT: Wed May 29 08:37:10 1996 /xypage/worksmart/tsp000/71401a/4
Vol. 222, No. 2, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
REFERENCES 1. Vale, W., Hsueh, A., Rivier, C., and Yu, J. (1990) in “Peptide Growth Factors and Their Receptors, Handbook of Experimental Pharmacology (M. B. Sporn and A. B. Robert, Eds.), Vol. 95/II, pp. 211–248, Springer-Verlag, Berlin. 2. Hötten, G., Neidhardt, H., Schneider, C., and Pohl, J. (1995) Biochem. Biophys. Res. Commun. 206, 608–613. 3. Oda, S., Nishimatsu, S., Murakami, K., and Ueno, N. (1995) Biochem. Biophys. Res. Commun. 210, 581–588. 4. Nakamura, T., Asashima, M., Eto, Y., Takio, K., Uchiyama, H., Moriya, N., Ariizumi, T., Yashiro, T., Sugino, K., Titani, K., and Sugino, H. (1992) J. Biol. Chem. 267, 16385–16389. 5. de Jong, F. H. (1988) Physiol. Rev. 68, 555–607. 6. Betteridge, A., and Craven, R. P. (1991) Biol. Reprod. 45, 748–754. 7. Groome, N. (1991) J. Immunol. Meth. 145, 65–69. 8. Shintani, Y., Takada, Y., Yamasaki, R., and Saito, S. (1991) J. Immunol. Meth. 137, 267–274. 9. Shinosich, M. J., Sieg, S., Zakher, A., Ling, N., Saunders, D. M., Rosenwaks, Z., and Hodgen, G. D. (1991) Clin. Chem. 37, 40–46. 10. Baly, D., Allison, D., Krummen, L., Woodruff, T., Mather, J., Fendley, B., Bald, L., and Lucas, C. (1993) Endocrinology 132, 2099–2108. 11. Groome, N., and O’Brien, M. (1993) J. Immunol. Meth. 165, 167–176. 12. Wong, W. L. T., Garg, S. J., Woodruff, T., Bald, L., Fendly, B., and Lofgren, J. A. (1993) J. Immunol. Meth. 165, 1–10. 13. Groome, N. P., Illingworth, P. J., O’Brien, M., Priddle, J., Weaver, K., and McNeilly, A. S. (1995) J. Clin. Endocrinol. Metab. 80, 2926–2932. 14. Burger, H. G. (1993) J. Clin. Endocrinol. Metab. 76, 1391–1396. 15. Vilja, P., Wichmann, L., Isola, J., and Tuohimaa, P. (1988) Clin. Chem. 34, 1585–1590. 16. Bläuer, M., and Tuohimaa, P. (1995) Endocrinology 136, 1482–1487. 17. Bläuer, M., Wichmann, L., Heinonen, P. K., Punnonen, R., and Tuohimaa, P. (1995) Submitted. 18. Bläuer, M., Kohonen, J., Leivonen, I., Vilja, P., and Tuohimaa, P. (1992) Development 115, 1079–1086. 19. Watson, A. Y., Anderson, J. K., Siminoski, K., Mole, J. E., and Murphy, R. A. (1985) Anat. Rec. 213, 365–376. 20. Wu, H. H., Kawamata, H., Wang, D. D., and Oyasu, R. (1993) Histochem. J. 25, 613–618. 21. Amano, O., Tsuji, T., Nakamura, T., and Iseki, S. (1991) J. Histochem. Cytochem. 39, 1707–1711. 22. Amano, O., Yoshitake, Y., Nishikawa, K., and Iseki, S. (1993) Cell Tissue Res. 273, 467–474. 23. Tsutsumi, O., Kurachi, H., and Oka, T. (1986) Science 233, 975–977. 24. Noguchi, S., Ohba, Y., and Oka, T. (1991) 269, E620–E625. 25. Sariosiek, J., Feng, T., and McCallum, R. W. (1991) Am. J. Med. Sci. 302, 359–363. 26. Okamoto, S., and Oka, T. (1984) Proc. Natl. Acad. Sci. 81, 6059–6063. 27. Kurachi, H., and Oka, T. (1986) J. Endocrinol. 109, 221–225. 28. Black, M. A., Lefebvre, F.-A., Pope, L., Lefebvre, Y. A., and Walker, P. (1992) Mol. Cell. Endocrinol. 84, 145–154. 29. Tenovuo, J., Laine, M., Söderling, E., and Irjala, K. (1981) Biochem. Med. 25, 337–345. 30. Laine, M., Tenovuo, J., Lehtonen, O.-P., Ojanotko-Harri, A., Vilja, P., and Tuohimaa, P. (1988) Archs. Oral. Biol. 33, 913–917. 31. Roberts, V. J., Sawchenko, P. E., and Vale, W. (1991) Endocrinology 128, 3122–3129. 32. Roberts, V. J., and Barth, S. L. (1994) Endocrinology 134, 914–923. 33. Ritvos, O., Tuuri, T., Erämaa, M., Sainio, K., Hildén, K., Saxén, L., and Gilbert, S. F. (1995) Mech. Devel. 50, 229–245. 34. Sawada, K., and Noumura, T. (1991) Acta. Anat. 140, 97–103.
235
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS ARTICLE NO.
222, 230–235 (1996)
0727
Measurement of Activin B in Human Saliva and Localization of Activin Subunits in Rat Salivary Glands Merja Bläuer,*,1 Lars Wichmann,† Reijo Punnonen,† and Pentti Tuohimaa* *Department of Anatomy, Medical School, University of Tampere; and †Department of Obstetrics and Gynecology, Tampere University Hospital, FIN-33101 Tampere, Finland Received April 4, 1996 The present paper is the first report to demonstrate that measurable amounts of activin B are secreted into the saliva. The results show wide fluctuations in activin B concentrations during the menstrual cycle with peak values detected at the follicular phase. Estrogen replacement therapy was found to increase salivary activin B levels in post-menopausal subjects. The concentration in males was negligible. These data suggest that activin B concentrations in the human saliva may be under hormonal regulation and propose that salivary activin B measurements may prove useful in investigating the as yet less well defined local and/or physiological roles of activin B. The present paper reports, in addition, the immunohistochemical localization of activin/inhibin subunits in the duct systems of the rat submandibular, sublingual and parotid salivary glands. © 1996 Academic Press, Inc.
Activins and inhibins belong to the TGFb superfamily and have been shown to possess multiple functions in cell growth and differentiation, reproductive processes and early embryonic development (1). Structurally, activins are disulfide-linked dimeric proteins composed of two homologous but distinct b-subunits, bA and bB. Recently, putative new activin BC and bD chains were also discovered (2,3). So far, three forms of activins are known: activin A (bAbA), activin AB (bAbB) and activin B (bBbB) which is the latest dimeric form isolated from natural sources (4). Inhibins are heterodimers of either of the two b-subunits and a related a-subunit (1). There is considerable interest in activin and inhibin measurement in biological fluids. The hormones were initially isolated from ovarian follicular fluid and, later, measurements have been made in serum, seminal plasma, amniotic fluid and various tissue extracts (5). Recently, several assays have been developed and quantitative studies made to screen activin/inhibin concentrations in human sera to investigate the normal physiological role of these regulatory factors and in search for possible clinical applications (6–13). Up till now, little is known about their diagnostic value (14). The present paper introduces the saliva as an interesting new target for quantitative activin research. We have investigated the level of activin B in human saliva using a recently developed immunoenzymometric assay (IEMA) and determined, in addition, the immunohistochemical localization of activin/inhibin subunit proteins in rat salivary glands. MATERIALS AND METHODS Saliva samples. Basal saliva samples were collected from 17 healthy adults from the following groups: females with normal menstrual cycles (n45); postmenopausal females without (n42) or with (n44) estrogen replacement therapy; males <50 y, (n44); males >60 y (n42). Saliva was collected in the morning, and no eating or drinking was allowed for 2h prior to collection. The samples were immediately frozen and stored at −20°C until analyzed. For activin B measurement the samples were thawed at room temperature and centrifuged at 16 000 × g for 10 min. The first group collected samples daily during one menstrual cycle, the other groups for a period of at least one week. LH measurement. To estimate the time of ovulation (group A) the day of LH surge was determined from urine samples collected in the morning and evening at days 8–18 of the menstrual cycle. A two-site IEMA based on two monoclonal
1
To whom correspondence should be addressed. 230
0006-291X/96 $18.00 Copyright © 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
JOBNAME: BBRC 222#2 PAGE: 2 SESS: 7 OUTPUT: Wed May 29 08:37:10 1996 /xypage/worksmart/tsp000/71401a/4
Vol. 222, No. 2, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
anti-human LH antibodies (Medix Biochemica, Kauniainen, Finland) was applied (15). Human LH (0.5–200 I.U./l) was used as standard. Activin B IEMA. Activin B concentrations were measured using an IEMA based on the monoclonal anti-bB(101–115) antibody (16). The IEMA is described in detail in Bläuer et al.(17). In brief, standards and diluted saliva samples were added to the anti-bB(101–115) precoated microtitre plates and incubated for 2 h at 37°C. After thorough washing, biotinylated anti-bB(101–115) was added and incubated for 1h at 37°C. The plates were washed and avidin-horseradish peroxidase was added and allowed to bind for 15 min at room temperature. After a further washing step the OPD-hydrogen peroxide substrate was added and the plates incubated for 5 min at room temperature. The absorbances were read at 492 nm after stopping the enzyme reaction with H2SO4. Data are presented as arbitrary units. A pregnant human serum pool used as standard was given the arbitrary value of 100 U activin B/ml. Animals. Adult male and female Sprague-Dawley rats were killed by cervical dislocation and their submandibular, sublingual and parotid salivary glands were removed. The tissues were fixed in ice-cold Bouin’s fluid for 2.5 h and embedded in paraffin. Immunohistochemistry. Immunohistochemical staining was performed essentially as described in Bläuer et al. (18). Rabbit polyclonal antibodies against synthetic human inhibin a(1–32) and bA(88–102) peptides were used (18). The anti-bB antibody was the same as used for IEMA.
RESULTS Dilution linearity. Serial dilutions of two saliva samples were made and assayed for activin B. Figure 1 shows that saliva diluted in parallel to the serum standard. The dilution curves were linear over the range from 1:5 to 1:20 (r 4 0.98). Activin B concentration in human saliva. The data are summarized in Table 1. IEMA results for four subjects having normal menstrual cycles are shown in Figure 2. In all cases, a peak in activin B concentration was observed before the day of LH surge. Immunohistochemistry. In the submandibular gland, bB-subunit immunoreactivity was localized in the epithelial cells of the striated ducts (SD), excretory ducts (ED) and granular convoluted tubules (GCT) (Fig. 3a,c). Secretory acini were not immunoreactive. Duct cells with varying staining intensities were seen, and in the SD and ED bB-subunit immunoreactivity was predominantly confined to the apical cytoplasm. Only some cells in the GCT were immunoreactive (Fig. 3a). In the parotid and sublingual glands, bB-subunit immunoreactivity was observed in the epithelial cells of the SD (Fig. 3d,e). The staining in the ED was faint (data not shown). In the sublingual gland, the intercalated ducts were strongly immunopositive (Fig. 3d). bA-subunit immunoreactivity was demonstrated in some GCT cells and on or close to the luminal surface of the SD in the submandibular, sublingual and parotid salivary glands (Fig. 4).
FIG. 1. Dose-response curves of the standard serum (v) and two saliva samples (■) and (m). Serial dilutions were made in assay buffer and analyzed for activin B by the IEMA. Each point represents the mean value of duplicate determinations. 231
JOBNAME: BBRC 222#2 PAGE: 3 SESS: 8 OUTPUT: Wed May 29 08:37:10 1996 /xypage/worksmart/tsp000/71401a/4
Vol. 222, No. 2, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS TABLE 1 Activin B Levels in Basal Saliva Group /
Normal menstrual cycle
Postmenopausal; no estrogen replacement Postmenopausal; estrogen replacement ?
<50y
>60y
Subject
Days
Activin B U/ml ±SD
a b c d e a b
30 29 30 30 30 14 14
9.2 ± 16.1 7.8 ± 7.2 5.4 ± 6.0 5.2 ± 13.6 0.2 ± 0.4 0.4 ± 0.2 0.09 ± 0.2
a* b* c d a b c d a b
7 7 7 7 13 10 14 14 14 14
1.4 ± 0.9 1.1 ± 1.2 1.0 ± 0.6 0.07 ± 0.2 0.7 ± 0.4 0.4 ± 0.5 0.1 ± 0.2 0 0.04 ± 0.09 0.01 ± 0.05
* Hysterectomy and bilateral ovariectomy.
In all glands, immunohistochemical staining was negative when absorbed antibodies were used (Fig. 3b, 4b). No differences could be demonstrated in the localization of immunoreactivity between females and males in the samples studied. Immunoreactive a-subunits were not observed in any of the three glands.
FIG. 2. Salivary activin B concentrations during the menstrual cycle. a, b, c, and d show the activin B concentrations in saliva samples collected daily by four subjects during one menstrual cycle. The day of the LH surge is indicated by an arrow. 232
JOBNAME: BBRC 222#2 PAGE: 4 SESS: 7 OUTPUT: Wed May 29 08:37:10 1996 /xypage/worksmart/tsp000/71401a/4
Vol. 222, No. 2, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 3. Expression of activin bB-subunits in rat salivary glands. a, In the submandibular gland, immunoreactive cells are seen in the granular convoluted tubules (arrows) and striated ducts (arrowheads). b, A control section incubated with anti-bB(101–115) preabsorbed with a 50-fold molar excess of peptide bB(101–115). A granular convoluted tubule and a striated duct are indicated by a large and a small asterisk, respectively. c, A submandibular gland excretory duct showing apical localization of bB-subunit immunoreactivity in its epithelial cells (arrowheads). d, In the sublingual gland, striated ducts (arrowheads) and intercalated ducts (arrow) were immunopositive. e, A less intense immunoreaction is seen in parotid striated ducts (arrowheads). Bars, 25 mm.
DISCUSSION Salivary glands have been shown to be a rich source of a multitude of peptide growth factors (19–23). It has been demonstrated, that salivary growth factors significantly affect oral and esophageal mucosal morphology and function (24,25). Furthermore, there is substantial evidence that some of these growth factors, in addition to being components of the saliva, also may have endocrine effects (23,26). The concentration of some of the growth factors in the murine submandibular gland appears to be regulated by hormonal steroids (27,28). Also in humans, the composition of the female saliva has been shown to vary with the menstrual cycle (29) and in pregnancy (30), suggesting an involvement of ovarian hormones in the secretory function of the salivary glands. Many salivary enzymes show a midcycle peak coinciding with the time of ovulation (29). This study represents the first evidence that immunoreactive activin B is present in the human saliva. When assayed with the recently developed IEMA, salivary activin B diluted in parallel to the serum standard, indicating little interference by other salivary components with assay performance. A stimulating effect of estrogens on salivary gland activin B secretion is suggested, based on increased activin B levels during the menstrual cycle and estrogen replacement therapy. In both groups, one subject showed negligible salivary activin B concentrations. The causes and potential significance of the two exceptions are not known but they may reflect disorders, irregularity or individual variation in ovarian function and estrogen effect. 233
JOBNAME: BBRC 222#2 PAGE: 5 SESS: 7 OUTPUT: Wed May 29 08:37:10 1996 /xypage/worksmart/tsp000/71401a/4
Vol. 222, No. 2, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 4. Activin bA-subunit immunoreactivity in rat salivary glands. Figure a shows immunoreactive cells in the granular convoluted tubules of the submandibular gland (arrows). Arrowheads indicate immunoreactivity on or close to the surface of striated ducts. b, A control section incubated with anti-bA (88–102) preabsorbed with a 50-fold molar excess of peptide bA(88–102). A granular convoluted tubule and a striated duct are indicated by a large and a small asterisk, respectively. c and d show a sparse distribution of bA-subunit immunoreactivity (arrowheads) in the striated ducts in the sublingual and parotid salivary glands, respectively. Bars, 25 mm.
The present paper is the first report to describe the localization of immunoreactive activin subunit proteins in salivary glands. Previously, the presence of activin bB-subunit mRNA has been described in the salivary glands of rat embryos (31,32). In mouse embryos, bB-subunit transcripts have been localized in the developing acinar epithelial cells and bA-subunit transcripts in the smooth muscle cells surrounding the salivary duct and some larger vessels (33). We demonstrated here that immunoreactive activin bB was localized in the ductal epithelial cells in adult rat salivary glands. The secretory acini were immunonegative, as previously shown also for other growth factors in salivary glands (19,22). Interestingly, in the present study, activin bA-subunit immunoreactivity was predominantly localized in the epithelial cells of the granular convoluted tubules in the submandibular gland, which in rodents is known to exhibit sexual dimorphism (34). In conclusion, this study shows that immunoreactive activin subunits and activin B are produced by the salivary glands. The data on activin B levels in the human saliva suggest a connection between salivary activin B and the female reproductive physiology. Salivary activin B measurements may prove useful in monitoring ovarian function and the physiological response to estrogen treatment. ACKNOWLEDGMENTS We thank Pekka Vilja, Ph.D., for biotinylation of the antibodies and Ms. Hilkka Mäkinen and Mr. Olli Ström for technical assistance. This work was supported by grants from the Tampere University Hospital Research Foundation and the University of Tampere. 234
JOBNAME: BBRC 222#2 PAGE: 6 SESS: 7 OUTPUT: Wed May 29 08:37:10 1996 /xypage/worksmart/tsp000/71401a/4
Vol. 222, No. 2, 1996
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
REFERENCES 1. Vale, W., Hsueh, A., Rivier, C., and Yu, J. (1990) in “Peptide Growth Factors and Their Receptors, Handbook of Experimental Pharmacology (M. B. Sporn and A. B. Robert, Eds.), Vol. 95/II, pp. 211–248, Springer-Verlag, Berlin. 2. Hötten, G., Neidhardt, H., Schneider, C., and Pohl, J. (1995) Biochem. Biophys. Res. Commun. 206, 608–613. 3. Oda, S., Nishimatsu, S., Murakami, K., and Ueno, N. (1995) Biochem. Biophys. Res. Commun. 210, 581–588. 4. Nakamura, T., Asashima, M., Eto, Y., Takio, K., Uchiyama, H., Moriya, N., Ariizumi, T., Yashiro, T., Sugino, K., Titani, K., and Sugino, H. (1992) J. Biol. Chem. 267, 16385–16389. 5. de Jong, F. H. (1988) Physiol. Rev. 68, 555–607. 6. Betteridge, A., and Craven, R. P. (1991) Biol. Reprod. 45, 748–754. 7. Groome, N. (1991) J. Immunol. Meth. 145, 65–69. 8. Shintani, Y., Takada, Y., Yamasaki, R., and Saito, S. (1991) J. Immunol. Meth. 137, 267–274. 9. Shinosich, M. J., Sieg, S., Zakher, A., Ling, N., Saunders, D. M., Rosenwaks, Z., and Hodgen, G. D. (1991) Clin. Chem. 37, 40–46. 10. Baly, D., Allison, D., Krummen, L., Woodruff, T., Mather, J., Fendley, B., Bald, L., and Lucas, C. (1993) Endocrinology 132, 2099–2108. 11. Groome, N., and O’Brien, M. (1993) J. Immunol. Meth. 165, 167–176. 12. Wong, W. L. T., Garg, S. J., Woodruff, T., Bald, L., Fendly, B., and Lofgren, J. A. (1993) J. Immunol. Meth. 165, 1–10. 13. Groome, N. P., Illingworth, P. J., O’Brien, M., Priddle, J., Weaver, K., and McNeilly, A. S. (1995) J. Clin. Endocrinol. Metab. 80, 2926–2932. 14. Burger, H. G. (1993) J. Clin. Endocrinol. Metab. 76, 1391–1396. 15. Vilja, P., Wichmann, L., Isola, J., and Tuohimaa, P. (1988) Clin. Chem. 34, 1585–1590. 16. Bläuer, M., and Tuohimaa, P. (1995) Endocrinology 136, 1482–1487. 17. Bläuer, M., Wichmann, L., Heinonen, P. K., Punnonen, R., and Tuohimaa, P. (1995) Submitted. 18. Bläuer, M., Kohonen, J., Leivonen, I., Vilja, P., and Tuohimaa, P. (1992) Development 115, 1079–1086. 19. Watson, A. Y., Anderson, J. K., Siminoski, K., Mole, J. E., and Murphy, R. A. (1985) Anat. Rec. 213, 365–376. 20. Wu, H. H., Kawamata, H., Wang, D. D., and Oyasu, R. (1993) Histochem. J. 25, 613–618. 21. Amano, O., Tsuji, T., Nakamura, T., and Iseki, S. (1991) J. Histochem. Cytochem. 39, 1707–1711. 22. Amano, O., Yoshitake, Y., Nishikawa, K., and Iseki, S. (1993) Cell Tissue Res. 273, 467–474. 23. Tsutsumi, O., Kurachi, H., and Oka, T. (1986) Science 233, 975–977. 24. Noguchi, S., Ohba, Y., and Oka, T. (1991) 269, E620–E625. 25. Sariosiek, J., Feng, T., and McCallum, R. W. (1991) Am. J. Med. Sci. 302, 359–363. 26. Okamoto, S., and Oka, T. (1984) Proc. Natl. Acad. Sci. 81, 6059–6063. 27. Kurachi, H., and Oka, T. (1986) J. Endocrinol. 109, 221–225. 28. Black, M. A., Lefebvre, F.-A., Pope, L., Lefebvre, Y. A., and Walker, P. (1992) Mol. Cell. Endocrinol. 84, 145–154. 29. Tenovuo, J., Laine, M., Söderling, E., and Irjala, K. (1981) Biochem. Med. 25, 337–345. 30. Laine, M., Tenovuo, J., Lehtonen, O.-P., Ojanotko-Harri, A., Vilja, P., and Tuohimaa, P. (1988) Archs. Oral. Biol. 33, 913–917. 31. Roberts, V. J., Sawchenko, P. E., and Vale, W. (1991) Endocrinology 128, 3122–3129. 32. Roberts, V. J., and Barth, S. L. (1994) Endocrinology 134, 914–923. 33. Ritvos, O., Tuuri, T., Erämaa, M., Sainio, K., Hildén, K., Saxén, L., and Gilbert, S. F. (1995) Mech. Devel. 50, 229–245. 34. Sawada, K., and Noumura, T. (1991) Acta. Anat. 140, 97–103.
235