- Email: [email protected]
Different expressions of connexin 43 and 32 in the fibroblasts of human dental pulp
Tissue & Cell, 2002 34 (3) 170–176 © 2002 Elsevier Science Ltd. All rights reserved. doi:10.1016/S0040-8166(02)00028-9, available online at http://www.idealibrary.com
Tissue&Cell
Different expressions of connexin 43 and 32 in the fibroblasts of human dental pulp N. Ibuki, Y. Yamaoka, Y. Sawa, T. Kawasaki, S. Yoshida
Abstract. The expression and localization of gap junctional proteins connexin (Cx) 26, 32, and 43 was examined in human dental pulp. Dental pulp tissues were obtained from human third molars immediately after extraction. Some pulp tissues were used for cell culture, and the rest for histological observations. Immunostaining for cultured dental pulp fibroblasts (DPFs) showed that Cx32 and 43 were expressed in human DPFs, and proteins corresponding to 27 (Cx32) and 43 kDa (Cx43) were identified by Western blot analysis. Immunostaining for tissue sections showed that the expression of Cx32 and 43 was observed in the entire region of the pulp and further strong expression of Cx32 was established beneath the cell-rich zone. Considering the close relationship between Cx types and cell functions, the results indicate that DPFs beneath the cell-rich zone may have specific, Cx32-related functions. The cell rich zone is thought to contain progenitor odontoblasts that can be induced to differentiate into mature odontoblasts in response to wounding. Therefore, it may be hypothesized that DPFs just beneath the cell-rich zone produce proteins and induce odontoblast differentiation from the cells in the cell-rich zone. © 2002 Elsevier Science Ltd. All rights reserved.
Keywords: dental pulp, fibroblast, gap junction, connexin 43, connexin 32
Introduction Gap junctions are specialized plasma membrane regions containing collections of transmembrane channels that provide a low resistance pathway directly linking the adjacent cells. Each channel is formed by two sets of hexametric structures (connexon) composed of six homologous transmembrane proteins (connexin, Cx) that surround a central pore. The channels permit intercytoplasmic exchange of molecules of less than 1000 Da including second messengers (Kumar & Department of Oral Functional Science, Graduate School of Dental Medicine, Hokkaido University, N13 W7, Kita-ku, Sapporo 060-8586, Japan Received 7 September 2001 Revised 9 April 2002 Accepted 11 April 2002 Correspondence to: Naoko Ibuki, Department of Oral Functional Science, Graduate School of Dental Medicine, Hokkaido University, N13 W7, Kita-ku, Sapporo 060-8586, Japan. Tel.: +81 11 706 4270; Fax: +81 11 706 4928; E-mail: [email protected]
170
Gilula, 1996). Different types of Cxs with various degrees of sequence homology have been reported in different tissues and organs. They are classified by the molecular weight as predicted from the corresponding cDNA (Beyer, 1993; Beyer et al., 1987). At present, 15 types of Cxs are known. Elfgang et al. (1995) transfected different types of CxcDNAs into HeLa cells, and found that there are Cx-specific differences in channel permeability. This indicates close relationships between Cx types and cell functions. The expression of Cx43 has been reported in fibroblasts (Beyer et al., 1989) and recent studies have shown other Cx expressions in fibroblasts. Abdullah et al. (1999) reported double expressions of Cx43 and 26 in human skin fibroblasts, and Yamaoka et al. (2000) reported Cx43 and 32 in human periodontal ligament fibroblasts. They suggested that the plural Cx expressions on the fibroblasts could reflect different functions of the cells. Dental pulp is a fibrous connective tissue containing fibroblasts, odontoblasts, and undifferentiated mesenchymal
connexin 43 and 32 in the fibroblasts of human dental pulp 171
cells, and it is surrounded by dentin. Among these cells, fibroblasts are major components of the dental pulp cells and produce collagen, collagenase, proteoglycans (Bartold et al., 1995), and glycosaminoglycans (Mangkornkarn & Steiner, 1992). Besides these functions, dental pulp fibroblasts (DPFs) in the cell-rich zone have been known to differentiate into odontoblasts when primary odontoblasts have been damaged (Chiego, 1994). To perform these different functions simultaneously, DPFs possibly express different types of Cxs. However, there have been no reports on the Cx expressions of DPFs except by Fried et al. (1996). Therefore, this study aims to examine Cx expression on human DPFs both in vivo and in vitro.
Materials and methods Human dental pulp tissues were obtained from third molars extracted for pericoronitis from ten 18–30 years old individuals. Immediately after extraction, the pulp tissues were carefully removed from the dentin, and seven were used for tissue sections and three for cell culture. Immunostaining for tissue sections Immediately after removal of the pulp, frozen 10 m serial sections were cut with a cryostat, air-dried, and fixed in 30% acetone diluted with phosphate-buffered saline (PBS) for 10 min at 4 ◦ C. The sections were immersed in PBS for 1 h at 4 ◦ C, and treated with 1% normal rabbit serum (NRS) or goat serum (GS) diluted with PBS for 1 h at room temperature (RT). They were reacted with anti-Cx43, anti-Cx32, and anti-Cx26 (mouse IgGs to rat Cx43, Zymed Laboratories, USA, mouse IgGs to rat Cx32, Chemicon International, USA, and goat IgGs to human Cx26, Santa Cruz Biotechnology Inc., USA) diluted with PBS for 8 h at 4 ◦ C. Then the sections were treated with AlexaTM 488 goat anti-mouse IgG (H + L) conjugate (Molecular Probes Inc., USA) or with AlexaTM 568 rabbit anti-goat IgG (H + L) conjugate (Molecular Probes Inc., USA). The sections were mounted in a VECTASHIELDTM (Vector Laboratories Inc., USA) and observed under a fluorescence microscope. Since myelinated peripheral nerve fibers have been known to express Cx32 (Bergoffen et al., 1993), adjacent sections were immunostained with anti-CD56 (mouse IgGs to human CD56, Antigenix America, USA) to discriminate Cx32 expression on the DPFs from that on nerve fibers (Roche et al., 1997). Immunostaining with secondary antibodies only was performed as a negative control, and no cross-reaction products were observed in the section (data is not shown). Hematoxilin–eosin (H–E) stained sections were also made in this study. Cell culture The dental pulp tissues were cultured in Dulbecco’s Modified Eagle Medium (DMEM, GIBCO Laboratories, USA) containing 10% fetal bovine serum (FBS), penicillin G
(100 U/ml), streptomycin (100 mg/ml), and amphotericin B (0.25 mg/ml) in type-I collagen-coated plastic culture dishes at 37 ◦ C in an atmosphere containing 5% CO2 for 7 days. After a confluent monolayer of the migrating cells had formed (5 × 105 –7 × 105 cells per dish), the cells were passaged by trypsinization and the DPFs in passage 3 were obtained. Immunostaining for cultured cells The DPFs in passage 3 (2 × 105 –2.5 × 105 cells per well) were cultured on cover slips in 6-well plates for 3 days until they were approximately 70% confluent. After the cultured cells were washed three times with PBS, they were air-dried and fixed in 30% acetone in PBS for 10 min. These cells were immunostained with anti-Cx43, anti-Cx32, and anti-Cx26 as described above, and observed under a fluorescence microscope. The cells were also immunostained with anti-vimentin (DAKO Corporation, USA) and observed under a phase-contrast and a fluorescence microscope before examination. Western blot The cultured DPFs in passage 3 (3 × 106 –4 × 106 ) were harvested without trypsinization and homogenized in cell lysis buffer (1% cholic acid, 0.1% sodium dodecyl sulfate [SDS] in PBS). The samples were suspended in equal volumes of sample buffer (10% glycerol, 2% SDS, 2 mM EDTA, 1% -mercaptoethanol, 0.05% bromophenol blue, 63 mM Tris–HCl) and boiled for 2 min, applied to 15% polyacrylamide gel (separating gel) with 4% stacking gel. Electrophoresis was conducted in a running buffer (50 mM Tris–HCl, 383 mM glycine, 0.1% SDS, 0.4 mM EDTA) for 4 h at constant current (30 mA). The separated proteins were electroblotted onto PVDF membranes (NOVEX, USA) in a transfer buffer (20% methanol, 20 mM Tris, 150 mM glycine) for 1 h at constant voltage (100 V). After the transfer, membranes were blocked with washing buffer (10 mM Tris–HCl: pH 7.4, 0.15 M NaCl, 0.1% Tween 20) containing 5% skim milk (Difco, USA) for 8 h at 4 ◦ C, and then incubated with anti-Cx43, anti-Cx32, and anti-Cx26 (mouse IgGs to rat Cx43, Zymed Laboratories, USA, mouse IgGs to rat Cx32, Chemicon International, USA, and rabbit IgGs to rat Cx26, Zymed Laboratories, USA) diluted 1:1000 in washing buffer for 2 h at RT. Membranes were washed three times for 20 min each in washing buffer and incubated with a second antibody (biotinylated horse anti-mouse IgGs or biotinylated goat anti-rabbit IgGs in a VECTOR-ABCTM kit; Vector Laboratories Inc.) for 0.5 h at RT. After washing for 1 h in washing buffer with changes of buffer every 20 min, the reaction products on the membranes were visualized in purple with a VECTOR-VIPTM kit (Vector Laboratories Inc.). The following proteins were used as molecular weight markers (Bio-Rad laboratories, USA): myosin (MW 199 000); -galactosidase (133 000); bovine serum albumin (BSA, 87 000); carbonic anhydrase (40 100); soybean trypsin inhibitor (31 600); lysozyme (18 500); aprotinin (7100); phosphorylase b (97 400); BSA
172 ibuki et al.
Fig. 1a–e (Continued )
connexin 43 and 32 in the fibroblasts of human dental pulp 173
Fig. 1a–e Serial sections of dental pulp. (a) Stained with H–E. The cell-rich zone of the subodontoblastic layer is clearly observed (arrows). ×120. (b) Immunostained with anti-Cx43. Reaction products of anti-Cx43 are observed in the entire region of the tissue section. It appeared weak in the cell-rich zone of the subodontoblastic region, and stronger in the rest of the section. ×120. (c) Immunostained with anti-Cx32. Reaction products of anti-Cx32 are observed in the entire region of the tissue section, with especially strong expression beneath the cell-rich zone (arrows). ×120. (d & e) Immunostained with anti-Cx26 (d) and CD56 (e). Reaction products of anti-Cx26 and anti-CD56 are prominent in the subodontoblastic region (arrows) and in the deeper pulp region with bundle shaped expression (arrowheads). The region expressing CD56 coincides with that of Cx26, but not with that of Cx32 (c). ×120.
(66 200); ovalbumin (45 000); carbonic anhydrase (31 000); soybean trypsin inhibitor (21 500); and lysozyme (14 400). For negative controls, Western blot of Cx43, 32, and 26 in the HeLa cells were also carried out.
Results Immunostained tissue sections Figure 1a–e are serial sections of the dental pulp stained with H–E (Fig. 1a), immunostained with anti-Cx43 (Fig. 1b),
anti-Cx32 (Fig. 1c), anti-Cx26 (Fig. 1d), and anti-CD56 (Fig. 1e), respectively. The expression of Cx43 was observed in the entire region of the pulp. It appeared weak in the cell-rich zone of the subodontoblastic region, and stronger in the rest of the section (Fig. 1b). The expression of Cx32 was also found in the entire region of the pulp, however, it was not uniform and there was strong expression only beneath the cell-rich zone (Fig. 1c). Different to these two Cxs, the expression of Cx26 and CD56 was prominent in the subodontoblastic region and in the deeper pulp region with bundle shaped expression (Fig. 1d & e), and the region
174 ibuki et al.
Fig. 2 Cultured dental pulp fibroblasts. (a) Phase contrast image of cultured dental pulp fibroblasts. The cells are homogeneous and spindle shaped. ×90. (b) The same cells as in (a) immunostained with anti-vimentin. The cells react strongly with anti-vimentin, indicating that there has been no epithelial cell contamination in the cultured dental pulp fibroblasts. ×95. Fig. 3 Cultured dental pulp fibroblasts immunostained with anti-Cx43 (a), and anti-Cx32 (b). Reaction products to anti-Cx43 and anti-Cx32 are observed around nuclei and at the periphery of the cells. ×190.
Fig. 4 Western blot of Cx43, Cx32, and Cx26. Lane 1: the reaction to anti-Cx43 in DPFs. There is a single band at approximately 43 kDa. Lane 2: the reaction to anti-Cx32 in DPFs. There is a single band at approximately 27 kDa. Lane 3: the reaction to anti-Cx26 in DPFs. No band is detected at 26.5 kDa. Lane 1 through 3 : the reactions to anti-Cx43 (Lane 1 ), anti-Cx32 (Lane 2 ), and Cx26 (Lane 3 ) in the HeLa cells. No bands are detected. M, molecular weight marker.
connexin 43 and 32 in the fibroblasts of human dental pulp 175
expressing Cx26 coincided with that of CD56. Comparing the four Figure 1a (H–E), 1c (Cx32), 1d (Cx26), and 1e (CD56) shows that the region strongly expressing Cx32 was located beneath the subodontoblastic region expressing Cx26 and CD56 where the nerve plexus of Raschkow exists. Cultured cells In the cultured human DPFs derived from extracted teeth, it was clear that the cells were homogeneous with a spindle shape (Fig. 2a) and that they reacted strongly with anti-vimentin (Fig. 2b), indicating that the cultured cells were of mesenchymal origin and that there was no contamination of the epithelial cells. Immunostained cultured cells In the cultured human DPFs, reaction products for anti-Cx43 and anti-Cx32 were found around the nuclei and at the periphery of the cells (Fig. 3a & b). Reaction products to anti-Cx26 were not observed (data is not shown). Western blot The results of the Western blot analysis of the cultured human DPFs and HeLa cells homogenates with the antibodies specific to Cx43, Cx32, and Cx26 are shown in Figure 4. In cultured DPFs, proteins were immunodetected as single bands at approximately 43 (Cx43) and 27 kDa (Cx32), but not detected at 26.5 kDa (Cx26). No proteins were immmunoreacted in HeLa cells.
Discussion Double expression of Cx43 and 32 in DPFs It has been reported that there are Cx-specific differences in the permeabilities of channels (Brissette et al., 1994; Steinberg et al., 1994; Elfgang et al., 1995; Veenstra et al., 1995). In these reports, Brissette et al. (1994) showed that during keratinocyte differentiation, the Cx expression had changed from Cx43 and 26 to Cx31 and 31.1. Further, parallel to the changes in the Cx expression, junctional permeability also changed selectively even though the cells remain electrically coupled. Elfgang et al. (1995) transfected different types of Cx-cDNAs into HeLa cells, and reported that the gap junctions composed of different types of Cxs showed different permeabilities with the different types of tracers, apparently dependent on the molecular structure of each tracer, i.e. the size, charge, and possibly rigidity. These reports strongly suggest that Cx types and functions are closely related to each other. The study here clearly demonstrated that human DPFs express both Cx43 and Cx32 both in vivo and in vitro. The results suggest the existence of gap junctions composed of Cx43 and Cx32 in human DPFs. Plural Cx expressions in fibroblasts have been reported in human skin (Abdullah et al., 1999; Cx43 and Cx26) and human periodontal ligament (Yamaoka et al., 2000; Cx43 and Cx32). Muramatsu et al. (1996) examined the expression and localization of Cxs in rat salivary
glands, and found that Cx43 was expressed between the processes of myoepithelial cells and Cx32 between acinar cells. It was suggested that Cx43 is associated with regulation of the contraction of the myoepithelial cells, and Cx32 with secretion of acinar cells. Considering these reports, the double expression of Cx43 and Cx32 in human DPFs may relate to the regulation of different functions of DPFs. In the tissue section, reaction products to anti-Cx26 were found in the subodontoblastic region and the deeper pulp region with bundle shaped expression. Further, Western blot analysis for cultured DPFs did not detect Cx26. The Cx26 has been known to express on the perineurium (Nagaoka et al., 1999), and the region expressing Cx26 coincided with that of CD56, which has been known to express in mature human normal peripheral nerves (Roche et al., 1997). Therefore, Cx26 expression in the subodontoblastic regions of tissue sections was thought to appear on the perineurium of the nerve plexus of Raschkow. Strong expression of Cx32 beneath the cell-rich zone The study clearly demonstrated strong expression of Cx32 just beneath the cell-rich zone, as well as the entire region of the pulp. In contrast to Cx26, this region was obviously different from that expressing CD56. Therefore, it is suggested that strong expression of Cx32 is not the result of nerve fibers but of DPFs. Assuming that DPFs just beneath the cell-rich zone strongly express Cx32, it is possible to speculate about the specific function of the DPFs there. The DPFs are known to produce bone-associated proteins, such as osteocalcin (Ranly et al., 1997), osteonectin (Salonen et al., 1990), and osteopontin (Nagata et al., 1995) in vitro under particular stimuli, e.g. bacterial toxins, growth factors, cytokines etc. This means that DPFs have the ability to differentiate into hard tissue forming cells. It is well known that DPFs in the cell-rich zone differentiate into odontoblasts when primary odontoblasts had been damaged (Chiego, 1994), and Fried et al. (1996) reported that differentiating odontoblasts in rat embryos express Cx43 and 32 while fully differentiated odontoblasts do not express Cx32. Considering these reports, it may be hypothesized that DPFs just beneath the cell-rich zone produce some proteins and induce odontoblast differentiation from the cells in the cell-rich zone. Another explanation for the Cx32 expression just beneath the cell-rich zone could be that this layer acts as a barrier to protect the pulp from various stimuli through the dentinal tubules in case of dental caries. It is still unclear, however, what functions are related to Cx43 and 32. Further study concerning the function of DPFs in the production of proteins which stimulate odontoblast differentiation will be needed.
REFERENCES Abdullah, K.M., Luthra, G., Bilski, J.J., Abdullah, S.A., Reynolds, L.P., Redmer, D.A. and Grazul-Bilska, A.T. 1999. Cell-to-cell communication and expression of gap junctional proteins in human diabetic and nondiabetic skin fibroblasts. Endocrine, 10, 35–41.
176 ibuki et al.
Bartold, P.M., Moule, A.J., Li, H. and Rigby, P. 1995. Isolation and characterization of the proteoglycans synthesized by adult human pulp fibroblasts in vitro. Int. Endocr. J., 28, 163–171. Bergoffen, J., Scherer, S.S., Wang, S., Scott, M.O., Bone, L.J., Paul, D.L., Chen, K., Lensch, M.W., Chance, P.F. and Fischbeck, K.H. 1993. Connexin mutations in X-linked Charcot–Marie-Tooth disease. Science, 262, 2039–2042. Beyer, E.C. 1993. Gap junctions. Int. Rev. Cytol., 137C, 1–37. Beyer, E.C., Paul, D.L. and Goodenough, D.A. 1987. Connexin43: a protein from rat heart homologous to a gap junction protein from liver. J. Cell Biol., 105, 2621–2629. Beyer, E.C., Kistler, J., Paul, D.L. and Goodenough, D.A. 1989. Antisera directed against connexin43 peptides react with a 43-kDa protein localized to gap junctions in myocardium and other tissues. J. Cell Biol., 108, 595–605. Brissette, J.L., Kumar, N.M., Gilula, N.B., Hall, J.E. and Dotto, G.P. 1994. Switch in gap junction protein expression is associated with selective changes in junctional permeability during keratinocyte differentiation. Proc. Natl. Acad. Sci. USA, 91, 6453–6457. Chiego Jr., D.J., 1994. Histology of pulp. In: Avery, J.K. (ed) Oral Development and Histology, 2nd edn, Thieme Medical Publishers Inc., New York, p 267. Elfgang, C., Eckert, R., Lichtenberg-Fraté, H., Butterweck, A., Traub, O., Klein, R.A., Hüser, D.F. and Willecke, K. 1995. Specific permeability and selective formation of gap junction channels in connexin-transfected HeLa cells. J. Cell Biol., 129, 805–817. Fried, K., Mitsiadis, T.A., Guerrier, A., Haegerstrand, A. and Meister, B. 1996. Combinatorial expression patterns of the connexins 26, 32, and 43 during development, homeostasis, and regeneration of rat teeth. Int. J. Dev. Biol., 40, 985–995. Kumar, N.M. and Gilula, N.B. 1996. The gap junction communication channel. Cell, 84, 381–388. Mangkornkarn, C. and Steiner, J.C. 1992. In vivo and in vitro glycosaminoglycans from human dental pulp. J. Endocrinol., 18, 327–331.
Muramatsu, T., Hashimoto, S. and Shimono, M. 1996. Differential expression of gap junction proteins connexin 32 and 43 in rat submandibular and sublingual glands. J. Histochem. Cytochem., 44, 49–56. Nagaoka, T., Oyamada, M., Okajima, S. and Takamatsu, T. 1999. Differential expression of gap junction proteins connexin 26, 32, and 43 in normal and crush-injured rat sciatic nerves: close relationship between connexin 43 and occludin in the perineurium. J. Histochem. Cytochem., 47, 937–948. Nagata, T., Yokota, M., Nishikawa, S., Ishida, H. and Wakano, Y. 1995. Osteopontin expression in clonal dental pulp cells. Ann. NY Acad. Sci., 760, 342–345. Ranly, D.M., Thomas, H.F., Chen, J. and MacDougall, M. 1997. Osteocalcin expression in young and aged dental pulps as determined by RT-PCR. J. Endocrinol., 23, 374–377. Roche, P.H., Figarella-Branger, D., Daniel, L., Bianco, N., Pellet, W. and Pellissier, J.F. 1997. Expression of cell adhesion molecules in normal nerves, choronic axonal neuropathies and Schwann cell tumors. J. Neurol. Sci., 151, 127–133. Steinberg, T.H., Civitelli, R., Geist, S.T., Robertson, A.J., Hick, E., Veenstra, R.D., Wang, H.-Z., Warlow, P.M., Westphale, E.M., Laing, J.G. and Beyer, E.C. 1994. Connexin43 and connexin45 form gap junctions with different molecular permeabilities in osteoblastic cells. EMBO J., 13, 744–750. Salonen, J., Domenicucci, C., Goldberg, H.A. and Sodek, J. 1990. Immunohistochemical localization of SPARC (osteonectin) and denatured collagen and their relationship to remodeling in rat dental tissues. Arch. Oral Biol., 35, 337–346. Veenstra, R.D., Wang, H.-Z., Beblo, D.A., Chilton, M.G., Harris, A.L., Beyer, E.C. and Brink, P.R. 1995. Selectivity of connexin-specific gap junctions does not correlate with channel conductance. Circ. Res., 77, 1156–1165. Yamaoka, Y., Sawa, Y., Ebata, N., Ibuki, N., Yoshida, S. and Kawasaki, T. 2000. Double expressions of connexin 43 and 32 in human periodontal ligament fibroblasts. Tissue Cell, 32, 328–335.
Tissue&Cell
Different expressions of connexin 43 and 32 in the fibroblasts of human dental pulp N. Ibuki, Y. Yamaoka, Y. Sawa, T. Kawasaki, S. Yoshida
Abstract. The expression and localization of gap junctional proteins connexin (Cx) 26, 32, and 43 was examined in human dental pulp. Dental pulp tissues were obtained from human third molars immediately after extraction. Some pulp tissues were used for cell culture, and the rest for histological observations. Immunostaining for cultured dental pulp fibroblasts (DPFs) showed that Cx32 and 43 were expressed in human DPFs, and proteins corresponding to 27 (Cx32) and 43 kDa (Cx43) were identified by Western blot analysis. Immunostaining for tissue sections showed that the expression of Cx32 and 43 was observed in the entire region of the pulp and further strong expression of Cx32 was established beneath the cell-rich zone. Considering the close relationship between Cx types and cell functions, the results indicate that DPFs beneath the cell-rich zone may have specific, Cx32-related functions. The cell rich zone is thought to contain progenitor odontoblasts that can be induced to differentiate into mature odontoblasts in response to wounding. Therefore, it may be hypothesized that DPFs just beneath the cell-rich zone produce proteins and induce odontoblast differentiation from the cells in the cell-rich zone. © 2002 Elsevier Science Ltd. All rights reserved.
Keywords: dental pulp, fibroblast, gap junction, connexin 43, connexin 32
Introduction Gap junctions are specialized plasma membrane regions containing collections of transmembrane channels that provide a low resistance pathway directly linking the adjacent cells. Each channel is formed by two sets of hexametric structures (connexon) composed of six homologous transmembrane proteins (connexin, Cx) that surround a central pore. The channels permit intercytoplasmic exchange of molecules of less than 1000 Da including second messengers (Kumar & Department of Oral Functional Science, Graduate School of Dental Medicine, Hokkaido University, N13 W7, Kita-ku, Sapporo 060-8586, Japan Received 7 September 2001 Revised 9 April 2002 Accepted 11 April 2002 Correspondence to: Naoko Ibuki, Department of Oral Functional Science, Graduate School of Dental Medicine, Hokkaido University, N13 W7, Kita-ku, Sapporo 060-8586, Japan. Tel.: +81 11 706 4270; Fax: +81 11 706 4928; E-mail: [email protected]
170
Gilula, 1996). Different types of Cxs with various degrees of sequence homology have been reported in different tissues and organs. They are classified by the molecular weight as predicted from the corresponding cDNA (Beyer, 1993; Beyer et al., 1987). At present, 15 types of Cxs are known. Elfgang et al. (1995) transfected different types of CxcDNAs into HeLa cells, and found that there are Cx-specific differences in channel permeability. This indicates close relationships between Cx types and cell functions. The expression of Cx43 has been reported in fibroblasts (Beyer et al., 1989) and recent studies have shown other Cx expressions in fibroblasts. Abdullah et al. (1999) reported double expressions of Cx43 and 26 in human skin fibroblasts, and Yamaoka et al. (2000) reported Cx43 and 32 in human periodontal ligament fibroblasts. They suggested that the plural Cx expressions on the fibroblasts could reflect different functions of the cells. Dental pulp is a fibrous connective tissue containing fibroblasts, odontoblasts, and undifferentiated mesenchymal
connexin 43 and 32 in the fibroblasts of human dental pulp 171
cells, and it is surrounded by dentin. Among these cells, fibroblasts are major components of the dental pulp cells and produce collagen, collagenase, proteoglycans (Bartold et al., 1995), and glycosaminoglycans (Mangkornkarn & Steiner, 1992). Besides these functions, dental pulp fibroblasts (DPFs) in the cell-rich zone have been known to differentiate into odontoblasts when primary odontoblasts have been damaged (Chiego, 1994). To perform these different functions simultaneously, DPFs possibly express different types of Cxs. However, there have been no reports on the Cx expressions of DPFs except by Fried et al. (1996). Therefore, this study aims to examine Cx expression on human DPFs both in vivo and in vitro.
Materials and methods Human dental pulp tissues were obtained from third molars extracted for pericoronitis from ten 18–30 years old individuals. Immediately after extraction, the pulp tissues were carefully removed from the dentin, and seven were used for tissue sections and three for cell culture. Immunostaining for tissue sections Immediately after removal of the pulp, frozen 10 m serial sections were cut with a cryostat, air-dried, and fixed in 30% acetone diluted with phosphate-buffered saline (PBS) for 10 min at 4 ◦ C. The sections were immersed in PBS for 1 h at 4 ◦ C, and treated with 1% normal rabbit serum (NRS) or goat serum (GS) diluted with PBS for 1 h at room temperature (RT). They were reacted with anti-Cx43, anti-Cx32, and anti-Cx26 (mouse IgGs to rat Cx43, Zymed Laboratories, USA, mouse IgGs to rat Cx32, Chemicon International, USA, and goat IgGs to human Cx26, Santa Cruz Biotechnology Inc., USA) diluted with PBS for 8 h at 4 ◦ C. Then the sections were treated with AlexaTM 488 goat anti-mouse IgG (H + L) conjugate (Molecular Probes Inc., USA) or with AlexaTM 568 rabbit anti-goat IgG (H + L) conjugate (Molecular Probes Inc., USA). The sections were mounted in a VECTASHIELDTM (Vector Laboratories Inc., USA) and observed under a fluorescence microscope. Since myelinated peripheral nerve fibers have been known to express Cx32 (Bergoffen et al., 1993), adjacent sections were immunostained with anti-CD56 (mouse IgGs to human CD56, Antigenix America, USA) to discriminate Cx32 expression on the DPFs from that on nerve fibers (Roche et al., 1997). Immunostaining with secondary antibodies only was performed as a negative control, and no cross-reaction products were observed in the section (data is not shown). Hematoxilin–eosin (H–E) stained sections were also made in this study. Cell culture The dental pulp tissues were cultured in Dulbecco’s Modified Eagle Medium (DMEM, GIBCO Laboratories, USA) containing 10% fetal bovine serum (FBS), penicillin G
(100 U/ml), streptomycin (100 mg/ml), and amphotericin B (0.25 mg/ml) in type-I collagen-coated plastic culture dishes at 37 ◦ C in an atmosphere containing 5% CO2 for 7 days. After a confluent monolayer of the migrating cells had formed (5 × 105 –7 × 105 cells per dish), the cells were passaged by trypsinization and the DPFs in passage 3 were obtained. Immunostaining for cultured cells The DPFs in passage 3 (2 × 105 –2.5 × 105 cells per well) were cultured on cover slips in 6-well plates for 3 days until they were approximately 70% confluent. After the cultured cells were washed three times with PBS, they were air-dried and fixed in 30% acetone in PBS for 10 min. These cells were immunostained with anti-Cx43, anti-Cx32, and anti-Cx26 as described above, and observed under a fluorescence microscope. The cells were also immunostained with anti-vimentin (DAKO Corporation, USA) and observed under a phase-contrast and a fluorescence microscope before examination. Western blot The cultured DPFs in passage 3 (3 × 106 –4 × 106 ) were harvested without trypsinization and homogenized in cell lysis buffer (1% cholic acid, 0.1% sodium dodecyl sulfate [SDS] in PBS). The samples were suspended in equal volumes of sample buffer (10% glycerol, 2% SDS, 2 mM EDTA, 1% -mercaptoethanol, 0.05% bromophenol blue, 63 mM Tris–HCl) and boiled for 2 min, applied to 15% polyacrylamide gel (separating gel) with 4% stacking gel. Electrophoresis was conducted in a running buffer (50 mM Tris–HCl, 383 mM glycine, 0.1% SDS, 0.4 mM EDTA) for 4 h at constant current (30 mA). The separated proteins were electroblotted onto PVDF membranes (NOVEX, USA) in a transfer buffer (20% methanol, 20 mM Tris, 150 mM glycine) for 1 h at constant voltage (100 V). After the transfer, membranes were blocked with washing buffer (10 mM Tris–HCl: pH 7.4, 0.15 M NaCl, 0.1% Tween 20) containing 5% skim milk (Difco, USA) for 8 h at 4 ◦ C, and then incubated with anti-Cx43, anti-Cx32, and anti-Cx26 (mouse IgGs to rat Cx43, Zymed Laboratories, USA, mouse IgGs to rat Cx32, Chemicon International, USA, and rabbit IgGs to rat Cx26, Zymed Laboratories, USA) diluted 1:1000 in washing buffer for 2 h at RT. Membranes were washed three times for 20 min each in washing buffer and incubated with a second antibody (biotinylated horse anti-mouse IgGs or biotinylated goat anti-rabbit IgGs in a VECTOR-ABCTM kit; Vector Laboratories Inc.) for 0.5 h at RT. After washing for 1 h in washing buffer with changes of buffer every 20 min, the reaction products on the membranes were visualized in purple with a VECTOR-VIPTM kit (Vector Laboratories Inc.). The following proteins were used as molecular weight markers (Bio-Rad laboratories, USA): myosin (MW 199 000); -galactosidase (133 000); bovine serum albumin (BSA, 87 000); carbonic anhydrase (40 100); soybean trypsin inhibitor (31 600); lysozyme (18 500); aprotinin (7100); phosphorylase b (97 400); BSA
172 ibuki et al.
Fig. 1a–e (Continued )
connexin 43 and 32 in the fibroblasts of human dental pulp 173
Fig. 1a–e Serial sections of dental pulp. (a) Stained with H–E. The cell-rich zone of the subodontoblastic layer is clearly observed (arrows). ×120. (b) Immunostained with anti-Cx43. Reaction products of anti-Cx43 are observed in the entire region of the tissue section. It appeared weak in the cell-rich zone of the subodontoblastic region, and stronger in the rest of the section. ×120. (c) Immunostained with anti-Cx32. Reaction products of anti-Cx32 are observed in the entire region of the tissue section, with especially strong expression beneath the cell-rich zone (arrows). ×120. (d & e) Immunostained with anti-Cx26 (d) and CD56 (e). Reaction products of anti-Cx26 and anti-CD56 are prominent in the subodontoblastic region (arrows) and in the deeper pulp region with bundle shaped expression (arrowheads). The region expressing CD56 coincides with that of Cx26, but not with that of Cx32 (c). ×120.
(66 200); ovalbumin (45 000); carbonic anhydrase (31 000); soybean trypsin inhibitor (21 500); and lysozyme (14 400). For negative controls, Western blot of Cx43, 32, and 26 in the HeLa cells were also carried out.
Results Immunostained tissue sections Figure 1a–e are serial sections of the dental pulp stained with H–E (Fig. 1a), immunostained with anti-Cx43 (Fig. 1b),
anti-Cx32 (Fig. 1c), anti-Cx26 (Fig. 1d), and anti-CD56 (Fig. 1e), respectively. The expression of Cx43 was observed in the entire region of the pulp. It appeared weak in the cell-rich zone of the subodontoblastic region, and stronger in the rest of the section (Fig. 1b). The expression of Cx32 was also found in the entire region of the pulp, however, it was not uniform and there was strong expression only beneath the cell-rich zone (Fig. 1c). Different to these two Cxs, the expression of Cx26 and CD56 was prominent in the subodontoblastic region and in the deeper pulp region with bundle shaped expression (Fig. 1d & e), and the region
174 ibuki et al.
Fig. 2 Cultured dental pulp fibroblasts. (a) Phase contrast image of cultured dental pulp fibroblasts. The cells are homogeneous and spindle shaped. ×90. (b) The same cells as in (a) immunostained with anti-vimentin. The cells react strongly with anti-vimentin, indicating that there has been no epithelial cell contamination in the cultured dental pulp fibroblasts. ×95. Fig. 3 Cultured dental pulp fibroblasts immunostained with anti-Cx43 (a), and anti-Cx32 (b). Reaction products to anti-Cx43 and anti-Cx32 are observed around nuclei and at the periphery of the cells. ×190.
Fig. 4 Western blot of Cx43, Cx32, and Cx26. Lane 1: the reaction to anti-Cx43 in DPFs. There is a single band at approximately 43 kDa. Lane 2: the reaction to anti-Cx32 in DPFs. There is a single band at approximately 27 kDa. Lane 3: the reaction to anti-Cx26 in DPFs. No band is detected at 26.5 kDa. Lane 1 through 3 : the reactions to anti-Cx43 (Lane 1 ), anti-Cx32 (Lane 2 ), and Cx26 (Lane 3 ) in the HeLa cells. No bands are detected. M, molecular weight marker.
connexin 43 and 32 in the fibroblasts of human dental pulp 175
expressing Cx26 coincided with that of CD56. Comparing the four Figure 1a (H–E), 1c (Cx32), 1d (Cx26), and 1e (CD56) shows that the region strongly expressing Cx32 was located beneath the subodontoblastic region expressing Cx26 and CD56 where the nerve plexus of Raschkow exists. Cultured cells In the cultured human DPFs derived from extracted teeth, it was clear that the cells were homogeneous with a spindle shape (Fig. 2a) and that they reacted strongly with anti-vimentin (Fig. 2b), indicating that the cultured cells were of mesenchymal origin and that there was no contamination of the epithelial cells. Immunostained cultured cells In the cultured human DPFs, reaction products for anti-Cx43 and anti-Cx32 were found around the nuclei and at the periphery of the cells (Fig. 3a & b). Reaction products to anti-Cx26 were not observed (data is not shown). Western blot The results of the Western blot analysis of the cultured human DPFs and HeLa cells homogenates with the antibodies specific to Cx43, Cx32, and Cx26 are shown in Figure 4. In cultured DPFs, proteins were immunodetected as single bands at approximately 43 (Cx43) and 27 kDa (Cx32), but not detected at 26.5 kDa (Cx26). No proteins were immmunoreacted in HeLa cells.
Discussion Double expression of Cx43 and 32 in DPFs It has been reported that there are Cx-specific differences in the permeabilities of channels (Brissette et al., 1994; Steinberg et al., 1994; Elfgang et al., 1995; Veenstra et al., 1995). In these reports, Brissette et al. (1994) showed that during keratinocyte differentiation, the Cx expression had changed from Cx43 and 26 to Cx31 and 31.1. Further, parallel to the changes in the Cx expression, junctional permeability also changed selectively even though the cells remain electrically coupled. Elfgang et al. (1995) transfected different types of Cx-cDNAs into HeLa cells, and reported that the gap junctions composed of different types of Cxs showed different permeabilities with the different types of tracers, apparently dependent on the molecular structure of each tracer, i.e. the size, charge, and possibly rigidity. These reports strongly suggest that Cx types and functions are closely related to each other. The study here clearly demonstrated that human DPFs express both Cx43 and Cx32 both in vivo and in vitro. The results suggest the existence of gap junctions composed of Cx43 and Cx32 in human DPFs. Plural Cx expressions in fibroblasts have been reported in human skin (Abdullah et al., 1999; Cx43 and Cx26) and human periodontal ligament (Yamaoka et al., 2000; Cx43 and Cx32). Muramatsu et al. (1996) examined the expression and localization of Cxs in rat salivary
glands, and found that Cx43 was expressed between the processes of myoepithelial cells and Cx32 between acinar cells. It was suggested that Cx43 is associated with regulation of the contraction of the myoepithelial cells, and Cx32 with secretion of acinar cells. Considering these reports, the double expression of Cx43 and Cx32 in human DPFs may relate to the regulation of different functions of DPFs. In the tissue section, reaction products to anti-Cx26 were found in the subodontoblastic region and the deeper pulp region with bundle shaped expression. Further, Western blot analysis for cultured DPFs did not detect Cx26. The Cx26 has been known to express on the perineurium (Nagaoka et al., 1999), and the region expressing Cx26 coincided with that of CD56, which has been known to express in mature human normal peripheral nerves (Roche et al., 1997). Therefore, Cx26 expression in the subodontoblastic regions of tissue sections was thought to appear on the perineurium of the nerve plexus of Raschkow. Strong expression of Cx32 beneath the cell-rich zone The study clearly demonstrated strong expression of Cx32 just beneath the cell-rich zone, as well as the entire region of the pulp. In contrast to Cx26, this region was obviously different from that expressing CD56. Therefore, it is suggested that strong expression of Cx32 is not the result of nerve fibers but of DPFs. Assuming that DPFs just beneath the cell-rich zone strongly express Cx32, it is possible to speculate about the specific function of the DPFs there. The DPFs are known to produce bone-associated proteins, such as osteocalcin (Ranly et al., 1997), osteonectin (Salonen et al., 1990), and osteopontin (Nagata et al., 1995) in vitro under particular stimuli, e.g. bacterial toxins, growth factors, cytokines etc. This means that DPFs have the ability to differentiate into hard tissue forming cells. It is well known that DPFs in the cell-rich zone differentiate into odontoblasts when primary odontoblasts had been damaged (Chiego, 1994), and Fried et al. (1996) reported that differentiating odontoblasts in rat embryos express Cx43 and 32 while fully differentiated odontoblasts do not express Cx32. Considering these reports, it may be hypothesized that DPFs just beneath the cell-rich zone produce some proteins and induce odontoblast differentiation from the cells in the cell-rich zone. Another explanation for the Cx32 expression just beneath the cell-rich zone could be that this layer acts as a barrier to protect the pulp from various stimuli through the dentinal tubules in case of dental caries. It is still unclear, however, what functions are related to Cx43 and 32. Further study concerning the function of DPFs in the production of proteins which stimulate odontoblast differentiation will be needed.
REFERENCES Abdullah, K.M., Luthra, G., Bilski, J.J., Abdullah, S.A., Reynolds, L.P., Redmer, D.A. and Grazul-Bilska, A.T. 1999. Cell-to-cell communication and expression of gap junctional proteins in human diabetic and nondiabetic skin fibroblasts. Endocrine, 10, 35–41.
176 ibuki et al.
Bartold, P.M., Moule, A.J., Li, H. and Rigby, P. 1995. Isolation and characterization of the proteoglycans synthesized by adult human pulp fibroblasts in vitro. Int. Endocr. J., 28, 163–171. Bergoffen, J., Scherer, S.S., Wang, S., Scott, M.O., Bone, L.J., Paul, D.L., Chen, K., Lensch, M.W., Chance, P.F. and Fischbeck, K.H. 1993. Connexin mutations in X-linked Charcot–Marie-Tooth disease. Science, 262, 2039–2042. Beyer, E.C. 1993. Gap junctions. Int. Rev. Cytol., 137C, 1–37. Beyer, E.C., Paul, D.L. and Goodenough, D.A. 1987. Connexin43: a protein from rat heart homologous to a gap junction protein from liver. J. Cell Biol., 105, 2621–2629. Beyer, E.C., Kistler, J., Paul, D.L. and Goodenough, D.A. 1989. Antisera directed against connexin43 peptides react with a 43-kDa protein localized to gap junctions in myocardium and other tissues. J. Cell Biol., 108, 595–605. Brissette, J.L., Kumar, N.M., Gilula, N.B., Hall, J.E. and Dotto, G.P. 1994. Switch in gap junction protein expression is associated with selective changes in junctional permeability during keratinocyte differentiation. Proc. Natl. Acad. Sci. USA, 91, 6453–6457. Chiego Jr., D.J., 1994. Histology of pulp. In: Avery, J.K. (ed) Oral Development and Histology, 2nd edn, Thieme Medical Publishers Inc., New York, p 267. Elfgang, C., Eckert, R., Lichtenberg-Fraté, H., Butterweck, A., Traub, O., Klein, R.A., Hüser, D.F. and Willecke, K. 1995. Specific permeability and selective formation of gap junction channels in connexin-transfected HeLa cells. J. Cell Biol., 129, 805–817. Fried, K., Mitsiadis, T.A., Guerrier, A., Haegerstrand, A. and Meister, B. 1996. Combinatorial expression patterns of the connexins 26, 32, and 43 during development, homeostasis, and regeneration of rat teeth. Int. J. Dev. Biol., 40, 985–995. Kumar, N.M. and Gilula, N.B. 1996. The gap junction communication channel. Cell, 84, 381–388. Mangkornkarn, C. and Steiner, J.C. 1992. In vivo and in vitro glycosaminoglycans from human dental pulp. J. Endocrinol., 18, 327–331.
Muramatsu, T., Hashimoto, S. and Shimono, M. 1996. Differential expression of gap junction proteins connexin 32 and 43 in rat submandibular and sublingual glands. J. Histochem. Cytochem., 44, 49–56. Nagaoka, T., Oyamada, M., Okajima, S. and Takamatsu, T. 1999. Differential expression of gap junction proteins connexin 26, 32, and 43 in normal and crush-injured rat sciatic nerves: close relationship between connexin 43 and occludin in the perineurium. J. Histochem. Cytochem., 47, 937–948. Nagata, T., Yokota, M., Nishikawa, S., Ishida, H. and Wakano, Y. 1995. Osteopontin expression in clonal dental pulp cells. Ann. NY Acad. Sci., 760, 342–345. Ranly, D.M., Thomas, H.F., Chen, J. and MacDougall, M. 1997. Osteocalcin expression in young and aged dental pulps as determined by RT-PCR. J. Endocrinol., 23, 374–377. Roche, P.H., Figarella-Branger, D., Daniel, L., Bianco, N., Pellet, W. and Pellissier, J.F. 1997. Expression of cell adhesion molecules in normal nerves, choronic axonal neuropathies and Schwann cell tumors. J. Neurol. Sci., 151, 127–133. Steinberg, T.H., Civitelli, R., Geist, S.T., Robertson, A.J., Hick, E., Veenstra, R.D., Wang, H.-Z., Warlow, P.M., Westphale, E.M., Laing, J.G. and Beyer, E.C. 1994. Connexin43 and connexin45 form gap junctions with different molecular permeabilities in osteoblastic cells. EMBO J., 13, 744–750. Salonen, J., Domenicucci, C., Goldberg, H.A. and Sodek, J. 1990. Immunohistochemical localization of SPARC (osteonectin) and denatured collagen and their relationship to remodeling in rat dental tissues. Arch. Oral Biol., 35, 337–346. Veenstra, R.D., Wang, H.-Z., Beblo, D.A., Chilton, M.G., Harris, A.L., Beyer, E.C. and Brink, P.R. 1995. Selectivity of connexin-specific gap junctions does not correlate with channel conductance. Circ. Res., 77, 1156–1165. Yamaoka, Y., Sawa, Y., Ebata, N., Ibuki, N., Yoshida, S. and Kawasaki, T. 2000. Double expressions of connexin 43 and 32 in human periodontal ligament fibroblasts. Tissue Cell, 32, 328–335.