Immunohistochemical analysis of dentin matrix protein 1 (Dmp1) phosphorylation by Fam20C in bone: implications for the induction of biomineralization.

Dmp1 is an acidic phosphoprotein that is specifically expressed in osteocytes. During the secretory process, the full-length, precursor Dmp1 is cleaved into N- and C-terminal fragments. C-terminal Dmp1 is phosphorylated, becoming a highly negatively charged domain that may assist in bone mineralization by recruiting calcium ions and influencing subsequent mineral deposition. It has been recently reported that the Golgi-localized protein kinase Fam20C phosphorylates Dmp1 in vitro. To investigate this phosphorylation in situ, we determined the locations of phosphorylated Dmp1 and Fam20C in rat bones using immunohistochemistry. During osteocytogenesis, osteoblastic, osteoid, and young osteocytes (but not old osteocytes) express Dmp1 mRNA and contain Dmp1 protein in the Golgi apparatus. These Dmp1-producing cells were distributed across the surface layer of cortical bone. Using immunofluorescence, we found that N- and C-terminal Dmp1 fragments were predominantly distributed along the lacunar walls and canaliculi of mineralized bone, respectively, but were not present in the osteoid matrix. We also found that Fam20C and its substrate, C-terminal Dmp1, colocalized in the Golgi of osteoblastic, osteoid, and young osteocytes. Furthermore, phosphorylated C-terminal Dmp1 was present in the Golgi of young osteocytes. Double-labeling immunoelectron microscopy revealed that phosphorylated C-terminal Dmp1 localized to the canalicular wall in mineralized bone. These findings suggest that C-terminal Dmp1 is phosphorylated within osteocytes and then secreted into the pericanalicular matrix of mineralized bone. Phosphorylated, negatively charged C-terminal Dmp1 in the pericanalicular matrix may play an important role in bone mineralization by recruiting calcium ions.

Transcription factors related to chondrogenesis in pleomorphic adenoma of the salivary gland: a mechanism of mesenchymal tissue formation.

In salivary gland pleomorphic adenoma, expression of extracellular matrix (ECM) substances indicates that tumor epithelial cells are becoming chondrogenic and will produce cartilage-like mesenchymal tissues. Sox9, the master transcription factor of chondrogenesis, is expressed in mouse salivary gland cells. To clarify the mechanism behind chondrogenesis in tumor epithelial cells, we examined the expression of transcription factors related to chondrogenesis in tumors and salivary glands. Reverse transcriptase-polymerase chain reaction (RT-PCR), quantitative real-time RT-PCR, and immunostaining were performed on pleomorphic adenoma tissues, salivary gland tissues, and human submandibular gland (HSG) cells. The mRNAs of essential transcription factors for chondrogenesis-Sox9, Sox6, and Sox5-were detected in both tumor and salivary gland tissues. The mRNAs of aggrecan and type II collagen-cartilage-specific ECM substances-were detected only in tumors. Sox9 and Sox6 proteins were colocalized in many epithelial cells in tumors and salivary glands. Tumor epithelial cells also possessed aggrecan protein and occasionally type II collagen protein. Moreover, mRNAs for transcription repressors of chondrogenesis δEF1 and AP-2α were detected in both tumors and salivary glands, whereas Twist1 mRNA was detected only in salivary glands and was at significantly low-to-undetectable levels in tumors. Twist1 protein was localized in the Sox9-expressing salivary gland cells. HSG cells expressed Sox9, Sox6, and Twist1, but not aggrecan or type II collagen, and thus were similar to salivary gland cells. Twist1 depletion by Twist1 siRNA led to the upregulation of aggrecan and type II collagen mRNA expression in HSG cells. In contrast, forced expression of Twist1, using Twist1 cDNA, resulted in the downregulation of both these genes. Taken together, these results indicate that salivary gland cells have a potential for chondrogenesis, and Twist1 depletion concomitant with neoplastic transformation, which would permit tumor epithelial cells to produce cartilage-like mesenchymal tissues in salivary gland pleomorphic adenoma.

Intercellular adhesion molecule-1 (ICAM-1) expression correlates with oral cancer progression and induces macrophage/cancer cell adhesion.

Intercellular adhesion molecule-1 (ICAM-1) is a transmembrane glycoprotein in the immunoglobulin superfamily, which plays an important role in cell adhesion and signal transduction. Although ICAM-1 is believed to play a role in several malignancies, it is still uncertain whether or not ICAM-1 expression contributes to cancer progression. In this study, we performed clinicopathological and cell biological analyses of ICAM-1 expression in oral squamous cell carcinoma (SCC). First, we examined the ICAM-1 expression in tongue SCC immunohistochemically, and revealed that ICAM-1 was expressed predominantly at the invasive front area of tongue SCC. ICAM-1 expression at the invasive front area was correlated with invasion, lymph node metastasis and increased blood and lymphatic vessel density of the tongue SCC. The relationship between ICAM-1 expression and clinicopathological factors were consistent with the increased proliferation, invasion and cytokine-production activities of ICAM-1-transfected SCC cells. Second, we analyzed the relationship between macrophages and ICAM-1-expressing tongue SCC cells because ICAM-1 is known to act as a ligand for adhesion of immune cells. Increased ICAM-1 expression in tongue SCC was correlated with increased macrophage infiltration within SCC nests. Moreover, macrophage/SCC-cell adhesion through ICAM-1 molecule was revealed using an in vitro cell adhesion and blockade assay. These findings indicate that ICAM-1 plays an important role in tongue SCC progression, which may result from the SCC-cell activity, angiogenic activity, lymphangiogenic activity and macrophage/SCC-cell adhesion.

Chronological histological changes during bone regeneration on a non-crosslinked atelocollagen matrix.

Cleavage of the antigenic telopeptide region from type I collagen yields atelocollagen, and this is widely used as a scaffold for bone regeneration combined with cells, growth factors, etc. However, neither the biological effect of atelocollagen alone or its contribution to bone regeneration has been well studied. We evaluated the chronological histological changes during bone regeneration following implantation of non-crosslinked atelocollagen (Koken Co., Ltd.) in rat calvarial defects. One week after implantation, osteogenic cells positive for runt-related transcription factor 2 (Runx2) and osteoclasts positive for tartrate-resistant acid phosphatase (TRAP) were present in the atelocollagen implant in the absence of bone formation. The number of Runx2-positive osteogenic cells and Osterix-positive osteoblasts increased 2 weeks after implantation, and bone matrix proteins (osteopontin, OPN; osteocalcin, OC; dentin matrix protein 1, DMP1) were distributed in newly formed bone in a way comparable to normal bone. Some resorption cavities containing osteoclasts were also present. By 3 weeks after implantation, most of the implanted atelocollagen was replaced by new bone containing many resorption cavities, and OPN, OC, and DMP1 were deposited in the residual collagenous matrix. After 4 weeks, nearly all of the atelocollagen implant was replaced with new bone including hematopoietic marrow. Immunohistochemistry for the telopeptide region of type I collagen (TeloCOL1) during these processes demonstrated that the TeloCOL1-negative atelocollagen implant was replaced by TeloCOL1-positive collagenous matrix and new bone, indicating that new bone was mostly composed of endogenous type I collagen. These findings suggest that the atelocollagen itself can support bone regeneration by promoting osteoblast differentiation and type I collagen production.

Multifocal nodular oncocytic hyperplasia of bilateral parotid glands: A case report with a histological variant of clear cells.

A case of multifocal nodular oncocytic hyperplasia (MNOH) of the bilateral parotid gland is presented. An 80-year-old woman was admitted to hospital because of painless swellings in bilateral parotid regions. Histologically, the nodular lesion had incomplete capsules and engulfed the surrounding parotid gland at the periphery. The lesions were mostly composed of clear cells, while the peripheries of the lesions had typical oncocytic cells with abundant fine granules. The histological existence of the clear cell component in the lesions led to misdiagnoses of other clear cell neoplasms. However, this case had multiple nodules in bilateral glands. No evidence of malignant histological findings was found. Moreover, the clear cells, as well as the oncocytic cells, were demonstrated to have mitochondria and glycogen in their cytoplasm using special staining. Based on these findings, the diagnosis of this case was MNOH in the parotid gland. We also discuss the differential diagnosis for clear cell lesions.

The prevalence of human papillomavirus in oral premalignant lesions and squamous cell carcinoma in comparison to cervical lesions used as a positive control.

BACKGROUND: Previous reports concerning the prevalence of human papillomavirus (HPV) in oral squamous cell carcinoma (OSCC) have observed varied results. The aim of this study was to evaluate the prevalence of HPV in oral premalignant lesions (OPL) and OSCC. For accurate HPV detection in oral lesions, comparative analysis was performed on cervical lesions as positive controls. METHODS: Fifty-seven cases with OPL and 50 with OSCC were selected. Twenty-nine control cases were selected from cervical lesions. The HPV infection rate was analysed by consensus polymerase chain reaction (PCR) using the My09/My11 and Gp5+/Gp6+ primers, and genotyping detection was employed using a PCR-based micro-array. Immunohistochemical staining for p16(INK4a) was performed. RESULTS: Twenty-eight (96.6%) cases of cervical lesions were positive for HPV by consensus PCR and 24 cases (82.8%) were positive by genotyping. The total HPV-positive rate in cervical lesions was 96.6%. HPV-DNA was detected in nine cases (15.8%) of OPL and six cases (12.0%) of OSCC by consensus PCR. Six cases (10.5%) of OPL and three cases (6.0%) of OSCC were positive by genotyping. The total HPV-positive rate in oral lesions was 22.4% (26.3% of OPL and 18.0% of OSCC). In cervical lesions, immunohistochemistry of p16(INK4a) identified 27 cases (93.1%) as positive. Fifteen cases (26.3%) of OPL and eight cases (16.0%) of OSCC were positive for p16(INK4a). CONCLUSIONS: The HPV infection and p16(INK4a)-positive rates in oral lesions are lower than previously reported. This suggests that HPV may not play a major role in oral lesions although its involvement cannot completely be ruled out.

Lymphatic vessels and related factors in adenoid cystic carcinoma of the salivary gland.

Adenoid cystic carcinoma of the salivary gland preferentially metastasizes to distant organs. It rarely metastasizes to lymph nodes. Recently, lymphangiogenesis has been associated with lymph node metastasis. Therefore, lymphangiogenesis in adenoid cystic carcinoma was evaluated from the number of lymphatic vessels and the expression of lymphangiogenic factors. Immunohistochemistry and molecular analysis were performed on clinical materials (29 cases for immunohistochemistry and 9 cases for molecular analysis). Normal submandibular gland was used as a negative control of lymphangiogenesis (10 cases for immunohistochemistry and 5 cases for molecular analysis). In adenoid cystic carcinoma, podoplanin-positive lymphatic vessels were small and often constricted, and localized to the tumor periphery. They did not have Ki67-positive endothelial cells. The lymphatic vessel density of the tumor did not exceed that of the salivary gland. By reverse transcriptase-polymerase chain reaction, adenoid cystic carcinoma and the salivary gland expressed vascular endothelial growth factor receptor-3 (VEGFR-3) similarly but VEGF-C and VEGF-D differently. Adenoid cystic carcinoma expressed VEGF-C, whereas the salivary gland expressed both VEGF-C and VEGF-D. VEGF-C was weak in adenoid cystic carcinoma and strong in the salivary gland. Real-time reverse transcriptase-polymerase chain reaction of VEGF-C showed that the ratio of the tumor to the salivary gland was 1 to 30 (P<0.01). Immunohistochemistry barely detected VEGF-C in adenoid cystic carcinoma. VEGF-C was expressed faintly by the tumor cells. VEGF-C and VEGF-D were detected in the serous acinar and duct cells and in the duct contents in the salivary gland. VEGFR-3 appeared to be expressed by lymphatic vessels in both adenoid cystic carcinoma and the salivary gland. These results indicate that lymphangiogenesis does not occur in adenoid cystic carcinoma. This condition would lead to the uncommon lymphatic metastasis.

Label-retaining cells in the rat submandibular gland.

To identify stem cells in salivary glands, label-retaining cells (LRCs) were established in rat submandibular glands. Developing and regenerating glands were labeled with bromodeoxyuridine (BrdU). To cause gland regeneration, the glands were injured by duct obstruction. BrdU LRCs were observed in all the parenchymal structures except for the acinus of the glands labeled during regeneration. Among these LRCs, a few, but not many, expressed neither keratin18 (K18; an acinar/duct cell marker) nor alpha-smooth muscle actin (alphaSMA; a myoepithelial cell marker), and thus were putative stem cells. These (K18 and alphaSMA)(neg) LRCs were invariably observed in the intercalated duct and the excretory duct. In the intercalated duct, they were at the proximal end bordering the acinus (the neck of the intercalated duct). Next, to test the above identification, gland extirpation experiments were performed. LRCs were established by labeling developing glands with iododeoxyuridine (IdU) in place of BrdU. Removal of one submandibular gland forced the IdU-LRCs in the remaining gland to divide. They were labeled with chlorodeoxyuridine (CldU). The (K18 and alphaSMA)(neg) LRCs in the neck of the intercalated duct and in the excretory duct did not change in number or in IdU label. The CldU label appeared in these cells and then disappeared. These results indicate that the (K18 and alphaSMA)(neg) LRCs have divided asymmetrically and are thus considered salivary gland stem cells.

Ossifying fibroma vs fibrous dysplasia of the jaw: molecular and immunological characterization.

Ossifying fibroma and fibrous dysplasia of the jaw are maxillofacial fibro-osseous lesions that should be distinguished each other by a pathologist because they show distinct patterns of disease progression. However, both lesions often show similar histological and radiological features, making distinction between the two a diagnostic dilemma. In this study, we performed immunological and molecular analyses of five ossifying fibromas, four cases of extragnathic fibrous dysplasia, and five cases of gnathic fibrous dysplasia with typical histological and radiographic features. First, we examined the difference between fibrous dysplasia and ossifying fibroma in the expression of Runx2 (which determined osteogenic differentiation from mesenchymal stem cells) and other osteogenic markers. Fibroblastic cells in fibrous dysplasia and ossifying fibroma showed strong Runx2 expression in the nucleus. The bone matrices of both lesions showed similar expression patterns for all markers tested except for osteocalcin. Immunoreactivity for osteocalcin was strong throughout calcified regions in fibrous dysplasia, but weak in ossifying fibroma lesions. Second, we performed PCR analysis with peptide nucleic acid (PNA) for mutations at the Arg(201) codon of the alpha subunit of the stimulatory G protein gene (GNAS), which has reported to be a marker for extragnathic fibrous dysplasia. All nine cases of extragnathic or gnathic fibrous dysplasia were positive for this mutation. On the other hand, none of the five cases of ossifying fibroma showed the mutation. These findings indicate that although fibrous dysplasia and ossifying fibroma are similar disease entities, especially in the demonstration of the osteogenic lineage in stromal fibroblast-like cells, they show distinct differences that can be revealed by immunohistochemical detection of osteocalcin expression. Furthermore, PCR analysis with PNA for GNAS mutations at the Arg(201) codon is a useful method to differentiate between fibrous dysplasia and ossifying fibroma.

Involvement of intracellular free Ca2+ in enhanced release of herpes simplex virus by hydrogen peroxide.

BACKGROUND: It was reported that elevation of the intracellular concentration of free Ca2+ ([Ca2+]i) by a calcium ionophore increased the release of herpes simplex virus type 1 (HSV-1). Freely diffusible hydrogen peroxide (H2O2) is implied to alter Ca2+ homeostasis, which further enhances abnormal cellular activity, causing changes in signal transduction, and cellular dysfunction. Whether H2O2 could affect [Ca2+]i in HSV-1-infected cells had not been investigated. RESULTS: H2O2 treatment increased the amount of cell-free virus and decreased the proportion of viable cells. After the treatment, an elevation in [Ca2+]i was observed and the increase in [Ca2+]i was suppressed when intracellular and cytosolic Ca2+ were buffered by Ca2+ chelators. In the presence of Ca2+ chelators, H2O2-mediated increases of cell-free virus and cell death were also diminished. Electron microscopic analysis revealed enlarged cell junctions and a focal disintegration of the plasma membrane in H2O2-treated cells. CONCLUSION: These results indicate that H2O2 can elevate [Ca2+]i and induces non-apoptotic cell death with membrane lesions, which is responsible for the increased release of HSV-1 from epithelial cells.

A role of salivary carbonic anhydrase VI in dental plaque.

OBJECTIVE: Carbonic anhydrase (CA) VI is a unique secreted isozyme of CA, which catalyzes the reversible reaction CO2 +H2O<-->H+ +HCO3-. CA VI has been thought to provide a greater buffering capacity to fluids into which it is secreted. This study was performed to confirm this in saliva. DESIGN: Nine healthy subjects participated in the study. The pH of the dental plaque from each subject was monitored after a mouth rinse with 10% sucrose with or without 10(-5)M acetazolamide, a specific inhibitor of CA. Also CA was examined in plaque by enzyme histochemistry, immunohistochemistry and Western blot analysis. RESULTS: Though sucrose and sucrose plus inhibitor yielded Stephan curves with a similar temporal pattern, the pH values of the latter were significantly lower than those of the former. Plaque exhibited CA activity by enzyme histochemistry. Immunohistochemistry and Western analysis demonstrated that the activity was due to CA VI but not to CA I or CA II. CONCLUSIONS: The results indicate that CA VI in saliva penetrates plaque and facilitates acid neutralization by salivary bicarbonate. Therefore, CA VI may be considered an anti-caries protein in saliva.

Benign fibrous histiocytoma of the mandible.

A very rare case of benign fibrous histiocytoma of the mandible is presented. A 49-year-old woman was admitted because of left buccal swelling and pain. Panoramic radiograph showed well-demarcated soap-bubble appearance without sclerotic rim in the left mandibular bone. A yellowish-white and partly brown solid tumor was noted in the excised mandibular bone specimen. The tumor histologically consisted of spindle cells, in which areas showing a storiform pattern and other areas composed of histiocytic cells with erythrophagocytosis and foam cells were mixed. Immunohistochemically, the tumor cells were positive for vimentin, CD68, alpha-1-antichymotrypsin and alpha-1-antitrypsin. From these findings the tumor was diagnosed as a primary BFH of the mandible. No recurrence has been noted 2 years and 11 months after surgery.

Carbonic anhydrase VI in the mouse nasal gland.

Western blotting analysis of mouse nasal tissue using a specific anti-mouse secreted carbonic anhydrase (CA VI) antibody has shown that CA VI is present in this tissue. A single immunoreactive band of 42 kD was observed, as has been found previously for salivary tissues. RT-PCR analysis has shown that nasal mucosa expressed CA VI mRNA. By immunohistochemistry (IHC), CA VI was observed in acinar cells, in duct contents of the anterior gland of the nasal septum, and in the lateral nasal gland. The Bowman's gland, the posterior gland of the nasal septum, and the maxillary sinus gland were negative. Immunoreactivity was also observed in the mucus covering the respiratory and olfactory mucosa and in the lumen of the nasolacrimal duct. In contrast, an anti-rat CA II antibody (that crossreacts with the mouse enzyme) stained only known CA II-positive cells and an occasional olfactory receptor neuron. These results indicate that CA VI is produced by the nasal gland and is secreted over the nasal mucosa. By reversible hydration of CO(2), CA VI is presumed to play a role in mucosal functions such as CO(2) sensation and acid-base balance. It may also play a role in olfactory function as a growth factor in maturation of the olfactory epithelial cells.

Adenoid cystic carcinoma associated with salivary duct cyst in the sublingual gland.

We described an extremely rare case of adenoid cystic carcinoma associated with salivary duct cyst in the sublingual gland of a 40-year-old Japanese woman. The tumor was growing from the cyst wall and almost occluded the cyst lumen. The epithelium lining the cyst lumen contained both keratin 19-positive cells and alpha-smooth muscle actin-positive cells, indicating the cyst being derived from the acinus/intercalated duct of the sublingual gland. Therefore, our case has presented for the first time a direct evidence that adenoid cystic carcinoma arises from acinus/intercalated duct.

Plasmacytoid cells in salivary-gland pleomorphic adenomas: evidence of luminal cell differentiation.

To determine the cellular origin of plasmacytoid cells in salivary gland adenomas, immunohistochemistry was performed on sections from 12 pleomorphic adenomas rich in these cells. In normal salivary glands included in these sections, the myoepithelial cells (MECs) expressed alpha-smooth muscle actin (alphaSMA) and smooth muscle myosin heavy chain (SMMHC), whereas the duct luminal cells expressed keratins 19, 18 and 8. Some of the salivary duct basal cells expressed these keratins, and the acinar cells expressed keratins 18 and 8. The expression profile was similar in rat salivary glands not only after but also during development. The immature MECs never expressed the keratins nor did the immature duct cells express alphaSMA. In seven cases, up to 60% of the plasmacytoid cells expressed keratin 19. In three of these cases, about 10% of the plasmacytoid cells expressed keratin 18. No plasmacytoid cells expressed alphaSMA, SMMHC or keratin 8. These results indicate that plasmacytoid cells originate from luminal cells and not from MECs. Furthermore, in addition to the luminal tumor cells, the non-luminal cells could express keratins 19, 18 and 8. Therefore, it is necessary to re-evaluate the prevailing notion that non-luminal cells are modified MECs. Keratin 14, basic calponin, vimentin and p63 were bi-specific for the MECs and the duct cells. Therefore, expression of these proteins by significant numbers of the non-luminal tumor cells and the plasmacytoid cells never denied the above notion.

Immunocytochemistry of myoepithelial cells in the salivary glands.

MECs are distributed on the basal aspect of the intercalated duct and acinus of human and rat salivary glands. However, they do not occur in the acinus of rat parotid glands, and sometimes occur in the striated duct of human salivary glands. MECs, as the name implies, have structural features of both epithelial and smooth muscle cells. They contract by autonomic nervous stimulation, and are thought to assist the secretion by compressing and/or reinforcing the underlying parenchyma. MECs can be best observed by immunocytochemistry. There are three types of immunocytochemical markers of MECs in salivary glands. The first type includes smooth muscle protein markers such as alpha-SMA, SMMHC, h-caldesmon and basic calponin, and these are expressed by MECs and the mesenchymal vasculature. The second type is expressed by MECs and the duct cells and includes keratins 14, 5 and 17, alpha 1 beta 1 integrin, and metallothionein. Vimentin is the third type and, in addition to MECs, is expressed by the mesenchymal cells and some duct cells. The same three types of markers are used for studying the developing gland. Development of MECs starts after the establishment of an extensively branched system of cellular cords each of which terminates as a spherical cell mass, a terminal bud. The pluripotent stem cell generates the acinar progenitor in the terminal bud and the ductal progenitor in the cellular cord. The acinar progenitor differentiates into MECs, acinar cells and intercalated duct cells, whereas the ductal progenitor differentiates into the striated and excretory duct cells. Both in the terminal bud and in the cellular cord, the immediate precursors of all types of the epithelial cells appear to express vimentin. The first identifiable MECs are seen at the periphery of the terminal bud or the immature acinus (the direct progeny of the terminal bud) as somewhat flattened cells with a single cilium projecting toward them. They express vimentin and later alpha-SMA and basic calponin. At the next developmental stage, MECs acquire cytoplasmic microfilaments and plasmalemmal caveolae but not as much as in the mature cell. They express SMMHC and, inconsistently, K14. This protein is consistently expressed in the mature cell. K14 is expressed by duct cells, and vimentin is expressed by both mesenchymal and epithelial cells. After development, the acinar progenitor and the ductal progenitor appear to reside in the acinus/intercalated duct and the larger ducts, respectively, and to contribute to the tissue homeostasis. Under unusual conditions such as massive parenchymal destruction, the acinar progenitor contributes to the maintenance of the larger ducts that result in the occurrence of striated ducts with MECs. The acinar progenitor is the origin of salivary gland tumors containing MECs. MECs in salivary gland tumors are best identified by immunocytochemistry for alpha-SMA. There are significant numbers of cells related to luminal tumor cells in the non-luminal tumor cells that have been believed to be neoplastic MECs.

Characterization of lacrimal gland carbonic anhydrase VI.

We have previously demonstrated by immunohistochemistry the presence of secreted carbonic anhydrase (CA VI) in the acinar cells of the rat lacrimal glands. In this study we purified the sheep lacrimal gland CA VI to homogeneity and demonstrated by Western analysis that it has the same apparent subunit molecular weight (45 kD) as the enzyme isolated from saliva. RT-PCR analysis showed that CA VI mRNA from the lacrimal gland was identical to that of the parotid gland CA VI mRNA. An RIA specific for sheep CA VI showed the lacrimal gland tissue concentration of the enzyme to be 4.20 +/- 2.60 ng/mg protein, or about 1/7000 of the level found in the parotid gland. Immunohistochemistry (IHC) and in situ hybridization (ISH) showed that lacrimal acinar cells expressed both immunoreactivity and mRNA for CA VI. Moreover, CA VI immunoreactivity was occasionally observed in the lumen of the ducts. Unlike the parotid gland, in which all acinar cells expressed CA VI immunoreactivity and mRNA, only some of the acinar cells of the lacrimal gland showed expression. These results indicate that the lacrimal gland synthesizes and secretes a very small amount of salivary CA VI. In tear fluid, CA VI is presumed to have a role in the maintenance of acid/base balance on the surface of the eye, akin to its role in the oral cavity.