Production and Localization of Muc4/Sialomucin Complex and Its Receptor Tyrosine Kinase ErbB2 in the Rat Lacrimal Gland

Abstract

purpose. To show the presence and forms of sialomucin complex (rat Muc4) and receptor tyrosine kinase ErbBs in the rat lacrimal gland and analyze for complexes of ErbB2 and its ligand Muc4.

methods. Northern blot analyses were used to identify sialomucin complex/Muc4 (SMC/Muc4) mRNA in rat lacrimal gland. Immunoblot analyses were performed to detect SMC/Muc4 and ErbBs. Sequential immunoprecipitation and immunoblot analyses were used to differentiate membrane and soluble forms of the SMC/Muc4 transmembrane subunit ASGP-2. Methacarn-fixed, paraffin-embedded sections of lacrimal glands from female adult rats were immunocytochemically stained using antisera to SMC/Muc4 and ErbBs to determine their relative locations in the gland. Colocalization of SMC/Muc4 and ErbB2 was confirmed by confocal immunofluorescence. Sequential immunoprecipitation and immunoblot were performed to analyze complexes of the SMC/Muc4 and ErbB2 in the lacrimal tissue.

results. Northern blot analyses of rat lacrimal glands, with a probe for SMC/Muc4, demonstrated the presence of a ∼9-kb transcript, consistent with observations in other tissues. Similarly, immunoblot analyses with antibodies against both the transmembrane (ASGP-2) and mucin (ASGP-1) subunits showed the presence of the two SMC/Muc4 subunits in lysates from rat lacrimal gland. Significantly, two different forms of ASGP-2 were observed, a high-molecular-weight (∼200-kDa) form and the more common 120- to 140-kDa form. Sequential immunoprecipitation and immunoblot analyses to differentiate membrane and soluble forms of SMC/Muc4 indicated that the high-molecular-weight form of ASGP-2 was predominantly associated with membranes, whereas the 120- to 140-kDa form was both membrane-associated and soluble. The lacrimal gland consists of acini connected by intercalated and interlobular ducts. Both acini and some intercalated ducts were stained by anti-ASGP-2 monoclonal antisera. Two patterns of acinar staining were observed: membrane staining at the borders of the epithelial cells and a granular staining within the cells. Staining of ductal surfaces with antibody to the cytoplasmic domain of ASGP-2 suggests that membrane SMC/Muc4 is being produced by the ductal cells and is not simply an adsorbed soluble product from the acinar cells. Immunoblot and immunocytochemical analyses demonstrated the presence of all four ErbBs, with ErbB2 showing the most widespread distribution, similar to that of SMC/Muc4. Immunofluorescence colocalization of membrane SMC/Muc4 and ErbB2 and coimmunoprecipitation of a complex of the two provided evidence of their association in membranes of lacrimal gland acinar cells.

conclusions. SMC/Muc4 is produced by the rat lacrimal gland as both membrane and soluble forms, specifically associated with both acinar and ductal cells. Because sialomucin complex is also present in the ocular tear film, the rat lacrimal gland represents a second source of this mucin for the tear film, in addition to the corneal and conjunctival epithelia. Moreover, the presence of a complex of SMC/Muc4 and the receptor tyrosine kinase ErbB2 in lacrimal tissue suggests that SMC/Muc4 acts as a ligand for the receptor and has functions in the lacrimal gland other than that of a mucin.

Covering the ocular surface is a tear film with the primary function of maintaining a smooth and clear refractive optical surface for vision in a hostile external environment. The tear film is biochemically complex, composed of an outer layer of lipid and an inner mucous gel layer separated by a midlayer of aqueous fluid. ¹ This aqueous layer contains electrolytes, water, and proteins. These components are primarily secreted by the lacrimal gland, which consists of lobules of acini. Each acinus possesses a luminal lining of columnar epithelial cells surrounded by a basal layer of myoepithelial cells and is enclosed by a basement membrane. Intercalated and interlobular ducts drain the fluid produced by the lacrimal gland into the conjunctival space beneath the upper eyelid. 

The major components of the inner gel layer of the tear film are high-molecular-weight glycoproteins called mucins, which are characterized by extensively O-glycosylated tandem repeats of serine- and threonine-rich domains. ² mRNA analyses have shown the expression of MUC4 and MUC5AC human mucin genes in the conjunctiva. ^{3 4} MUC1 has also been immunolocalized in cornea and conjunctiva. ⁵ Recently, sialomucin complex (SMC) and its human homologue MUC4 ⁶ have been found on the cornea and conjunctiva and in the ocular tear film of rat ⁷ and human ⁸ eyes. This mucin was first identified on the surface of metastatic strain 13762 rat ascites mammary adenocarcinoma cells. ^{9 10} SMC/Muc4 consists of an O-glycosylated mucin subunit ASGP-1, which is noncovalently bound to an N-glycosylated membrane glycoprotein ASGP-2. ^{11 12 13} Mature glycosylated ASGP-1 has a molecular weight of more than 500 kDa ⁹ ; its polypeptide is ∼220 kDa. ¹⁴ This subunit comprises three domains: an N-terminal unique sequence, a large tandem repeat region rich in serine and threonine residues similar to that of other mucins, and a C-terminal unique sequence. ¹⁴ ASGP-2 is a ∼120- to 140-kDa protein consisting of seven domains: two hydrophilic N-glycosylated regions, two epidermal growth factor (EGF)-like domains, a cysteine-rich domain, a transmembrane domain, and a small cytoplasmic domain. ¹⁵  

A number of functions have been attributed to SMC/Muc4. In tumor cells, ASGP-1 confers antirecognition and antiadhesive properties ¹⁶ that have been demonstrated by the transfection of tetracycline-regulated SMC/Muc4 DNA constructs into A375 melanoma cells. SMC/Muc4 overexpression was found to abolish cell–matrix adhesion and cell–cell interactions. ¹⁷ In addition, the overexpression of SMC/Muc4 by these cells reduces their killing by natural killer cells, ¹⁸ which may be important to tumor progression of mammary tumor cells. In normal epithelia SMC/Muc4 is proposed to act as a membrane barrier to protect the epithelial surface from effects of noxious agents, including microbes. ^{19 20} SMC/Muc4 is also proposed to modulate cellular signaling through the EGF family of receptors through its interaction with ErbB-2. ²¹ This function resides in one of its EGF-like domains, which contains all the consensus residues present in active members of the EGF family. ¹⁵ Recent studies have shown that SMC/Muc4 acts as an antiapoptotic agent in transfected tumor cells ²² and thus may also provide a protective mechanism to retard loss of cells in damaged epithelia. 

Because SMC/Muc4 is present in the rat ocular tear film, we investigated whether the lacrimal gland synthesizes and secretes this mucin complex. Northern and Western blot analyses indicated that SMC/Muc4 was present in the rat lacrimal gland. Sequential immunoprecipitation and immunoblot analyses showed that both soluble and membrane-bound forms of SMC/Muc4 are produced by the lacrimal gland. SMC/Muc4 was immunolocalized to both membranes and cytoplasmic granular structures in acinar cells. These data provide direct evidence for the production of mucins by the lacrimal gland, indicating that this tissue represents another source of SMC/Muc4 for the ocular tear film. The presence of a membrane form of SMC/Muc4 in the lacrimal gland raised questions about its functions. Of particular interest was the possibility of a complex with ErbB2 that may be involved in cellular signaling. Combined immunocytochemistry, immunofluorescence, and immunoprecipitation experiments demonstrated the colocalization of SMC/Muc4 and ErbB2 in lacrimal cells and provided evidence for a putative signaling complex. 

Materials and Methods

Tissues

All procedures used in these studies followed the tenets of the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Ten virgin adult female Fischer 344 rats (8–10 weeks old) were used for the immunologic analyses, and the data shown are representative of those animals. Animals were killed by CO₂ asphyxiation, and exorbital lacrimal glands were dissected from and snap frozen in liquid N₂. For immunocytochemical staining, glands were fixed in modified methacarn (60% [vol/vol] methanol, 30% [vol/vol] 1,1,1-trichloroethane, 10% [vol/vol] acetic acid) for 24 hours, processed, and embedded in paraffin wax, as previously described. ²³  

Northern Blot Analyses

Total RNA was isolated from lacrimal gland tissue or ascites cells using a commercial procedure (Totally RNA; Ambion, Austin, TX). RNA (25 μg) was separated on a 1% formaldehyde/agarose gel and transferred to a positively charged nylon membrane (BrightStar-Plus; Ambion), followed by cross-linking (Stratalinker; Stratagene, La Jolla, CA). The membranes were prehybridized for at least 2 hours at 42°C in hybridizing solution (DIG Easy Hyb; Roche Molecular Biochemicals, Indianapolis, IN). The probe A2G2-9, which spans the 5′ unique region and four tandem repeats of SMC/Muc4 cDNA, was random prime labeled with digoxigenin-11-dUTP, with a kit (Random Prime Labeling; Roche). Hybridization was performed overnight at 42°C in the hybridizing solution, and the membranes were washed once at room temperature in 2× SSC with 0.2% SDS for 15 minutes, twice at 50°C in 2× SSC with 0.2% SDS for 20 minutes each, and once at 50°C in 0.1× SSC with 0.2% SDS for 15 minutes. Signals were detected using anti-DIG-AP conjugate and a 3-(4-methoxyspiro{1,2-dioxetane-3,2′-(5′-chloro)tricyclo[3.3.1.1^3,7]decan}-4-yl)phenyl phosphate chemiluminescence substrate for alkaline phosphatase detection (Roche). 

Preparation of Rat Lacrimal Cell Membranes

According to a published procedure, ²⁴ rat lacrimal tissue was cleaned and trimmed of any fatty tissue and homogenized in 10 volumes of ice-cold buffered sucrose (250 mM sucrose, 50 mM sodium phosphate [pH 7.4]) supplemented with protease inhibitors. The homogenate was centrifuged at 700g for 5 minutes at 4°C, resuspended in buffered sucrose, and centrifuged at 40,000g for 15 minutes at 4°C to yield the membranes. 

Preparation of Rat Lacrimal Acini

According to a published method, ²⁵ cleaned, trimmed rat lacrimal tissue was washed three times in Krebs-Ringer bicarbonate buffer (119 mM NaCl, 4.8 mM KCl, 1.0 mM CaCl₂, 1.2 mM MgSO₄, 1.2 mM KH₂PO₄, and 25 mM NaHCO₃), 10 mM HEPES buffer, 0.1% bovine serum albumin, and protease inhibitors at 37°C and then incubated for 45 minutes in 20 ml of the buffer supplemented with 200 U/ml collagenase type III. Tissue was homogenized by pipetting through pipette tips of decreasing diameters (10 times per 15 minutes), filtered through 120-μm nylon mesh and centrifuged 5 minutes at 50g to collect the acini. 

Antibodies

Monoclonal antibodies (mAb) 4F12, 13C4, and 1B1 to the ASGP-2 subunit of rat SMC/Muc4 have been described, ²⁶ as have rabbit polyclonal antisera raised against whole ASGP-2 and a synthetic peptide from the COOH-terminal cytoplasmic domain. ²⁶ Polyclonal antibody HA-1, specific to a peptide in the C-terminal region of ASGP-1, has been previously described by Price-Schiavi et al. ⁷ Polyclonal antibody against ErbB2 was obtained from Dako (Carpinteria, CA); both monoclonal and polyclonal antibodies against ErbB2 from Neomarkers (Fremont, CA); polyclonal antibodies against ErbB3 and ErbB4 from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA); antibody against ErbB1 from Upstate Biotechnology (Lake Placid, NY); and biotinylated goat anti-mouse and anti-rabbit immunoglobulin (Igs) and horseradish peroxidase (HRP)-conjugated avidin from Dako. Rabbit antiserum to rodent laminin was a gift from Michael J. Warburton (St. George’s Hospital Medical School, London, UK). Smooth-muscle actin mAb 1A4, mAb V9 reactive to vimentin, and mAbs CK5 and CY-90 to keratin 18 were obtained from Sigma (St. Louis, MO), and mAbs MNF116 to keratins 5, 6, 8, 17, and 19, and pankeratin mAb AE1/3 were obtained from Dako. Nonimmune serum (IgG) fraction was used as a control. Fluorescein isothiocyanate (FITC) and Texas-red–labeled secondary antibodies for immunofluorescence were obtained from Molecular Probes (Eugene, OR). In addition, the specificities of the antibodies were checked by preabsorption with the appropriate antigen, and the antigen–antibody complex was then used for immunocytochemistry. ²⁷ Antisera to laminin was absorbed with 1 mg/ml reconstituted matrix (EHS derived from the Engelbreth-Holm-Swarm murine sarcoma; Promega, Madison, WI). mAbs V9 and 1A4 were absorbed with purified vimentin and actin (Sigma), respectively. Keratin mAbs were absorbed with human callus keratin (Sigma). mAbs 4F12 and 1B1 were absorbed with ASGP-2 purified from the MAT-C1 subline of 13762 ascites cells. ¹³  

SDS-PAGE and Immunoblotting

Lysates of lacrimal tissue and immunoprecipitates were subjected to SDS-PAGE under reducing conditions in 6% (vol/vol) polyacrylamide gels using a commercial system (Mini-Protean II; Bio-Rad, Hercules, CA). Resolved proteins were transferred onto nitrocellulose membranes and blocked for 1 hour with 5% nonfat dry milk in TTBS (10 mM Tris, 150 mM NaCl, 0.05% Tween 20, pH 7.5), rinsed in TTBS and incubated with primary antibody for 1 hour. Membranes were rinsed, then incubated with HRP-conjugated secondary anti-mouse or anti-rabbit Ig (Promega). After further washing, signals were detected using a chemiluminescence kit (Renaissance Enhanced Chemiluminescence; NEN Life Science Products Inc., Boston, MA), according to the manufacturer’s instructions. MAT-C1 ascites subline of the 13762 mammary tumor was used as a positive control for immunoblotting. 

Immunocytochemical Staining

For immunocytochemical staining, tissues were fixed in modified methacarn (60% [vol/vol] methanol, 30% [vol/vol] inhibsol, and 10% [vol/vol] acetic acid) for 24 hours, processed and embedded in paraffin wax, as described previously. ²⁸ Immunocytochemical staining of paraffin-embedded sections of lacrimal glands with mAbs and polyclonal antisera was performed using a standard avidin-biotin peroxidase complex assay. ²⁷ Briefly, 5-μm-thick sections were dewaxed in xylene and hydrated in absolute ethanol followed by successive immersions in a graded series of ethanol and water. Antigenic sites were retrieved from fixed samples using retrieval solution (Dako) for 20 minutes as recommended. Sections were rinsed with phosphate-buffered saline (PBS), then treated with 3% (vol/vol) H₂O₂ to inhibit endogenous peroxidases. The sections were incubated with primary antibody in antibody diluent (Dako), rinsed with PBS, and treated with reagent (LINK; Dako) for 30 minutes. After a PBS rinse, samples were incubated for 30 minutes in streptavidin peroxidase, rinsed with PBS, and incubated with a substrate–chromogen solution (Dako). Sections were rinsed in tap water, counterstained with hematoxylin for 15 seconds, washed in tap water, dipped in a wash bath of 1% LiCl, treated with absolute ethanol, cleared in xylene, and mounted (Permount; Fisher Scientific, Pittsburgh, PA). ²⁷ For staining with antisera to keratins and laminin, sections were pretreated with pronase. ²⁸ Acinar structure was examined by standard hematoxylin and eosin staining of tissue sections. 

Immunofluorescence Staining

Briefly, 5-μm-thick sections were dewaxed in xylene and hydrated in absolute ethanol followed by successive immersions in a graded series of ethanol-water. Sections were washed for 5 minutes in 0.85% NaCl followed by a wash in PBS. Next, tissue was fixed in 4% formalin in PBS for 15 minutes at room temperature, washed twice in PBS, and permeabilized in 0.02% Triton for 5 minutes. Tissue sections were fixed again in 4% formalin in PBS for 5 minutes followed by two washes with PBS. Nonspecific sites were blocked for 1 hour with 10% normal goat serum (Dako) and 0.2% BSA (Sigma) in PBS. Tissue sections were incubated with the first antibody for 1 hour, washed twice with PBS, and incubated in second antibody for 1 hour. Slides were mounted with a kit (Slow Fade; Molecular Probes). 

Preparation of Lysates and Immunoprecipitation

Frozen lacrimal tissues were pulverized with a mortar and pestle ²⁴ and solubilized with RIPA buffer (150 mM NaCl, 1% [vol/vol] Nonidet P-40, 0.5% [wt/vol] sodium deoxycholate, 0.1% [wt/vol] SDS, and 50 mM Tris-HCl [pH 8.0]) containing a cocktail of protease inhibitors (Sigma, St. Louis, MO). The lysates were clarified by centrifugation for 20 minutes at 12,000g and the pellet discarded. Lysates were immunoprecipitated with rotation at 4°C, as described previously. ⁷  

Results

Detection of Lacrimal Gland SMC/Muc4 mRNA by Northern Blot Analysis

A 1.7-kb probe from the 5′ end of SMC/Muc4 (ASGP-1) was used to investigate the presence of its transcript in rat lacrimal tissue. A predominant broad band at ∼9 kb was found for both lacrimal tissue and strain 13762 ascites cells (Fig. 1B) , as previously reported for the ascites cells. ¹⁵ Comparisons of the staining of ribosomal RNA (Fig. 1A) and SMC/Muc4 transcript (Fig. 1B) in lacrimal tissue and ascites cells suggest that SMC/Muc4 is an abundant gene product in the lacrimal gland. 

Immunoblot Analysis of Lacrimal Gland SMC/Muc4 Glycoproteins

Immunoblot analyses of lacrimal tissue with anti-SMC/Muc4 antibodies demonstrated the presence of both the transmembrane subunit ASGP-2 (Fig. 2) and the mucin subunit ASGP-1 (data not shown). Two forms of ASGP-2 were unexpectedly prominently displayed, by using the mAb anti-ASGP-2 4F12. A similar higher molecular weight form of ASGP-2 has been seen in other tissues in minor amounts and has recently been observed as a significant component in salivary glands. ²⁹ Both the higher molecular weight form and the more common 120- to 140-kDa form were present in isolated lacrimal gland acini and membranes as well as in the intact tissue (Fig. 2) . Because lacrimal gland is a secretory tissue and SMC/Muc4 is found in many tissues in both soluble and membrane forms, we used sequential immunoprecipitation and immunoblot analysis to investigate the SMC/Muc4 forms. Anti-C-pep antibody, which recognizes the cytoplasmic domain of ASGP-2, ²⁶ immunoprecipitated both the higher molecular weight and the 120- to 140-kDa forms (Fig. 3) . However, subsequent immunoprecipitation of the supernatant with polyclonal anti-ASGP-2, which recognizes both the soluble and membrane forms of ASGP-2, yielded only the 120- to 140-kDa form, suggesting that the higher molecular weight form is predominantly a membrane species. Moreover, the higher molecular weight form is readily observed in anti-C-pep immunoprecipitations from lacrimal gland membrane preparations (Fig. 3) . These results are consistent with our observations in other tissues, in which a complex of ASGP-1 and part of ASGP-2 are secreted from the epithelial cells. ^{7 19 26}  

Histologic Appearance and Immunocytochemical Staining of Cells in the Rat Lacrimal Gland

Adult female rat lacrimal tissue consists of acini containing tubular columnar secretory cells, surrounded by myoepithelial cells (data not shown). Scattered among acini are intercalated and interlobular ducts that are lined by luminal epithelial cells. A sparse lining of stromal tissue separates individual acini. Acinar cells were specifically stained by the cytokeratin-18 mAbs CY-90 and CK5, whereas ductal epithelial cells were stained by mAbs to MNF116 and AE1/3, which recognize multiple cytokeratins. mAb 1A4, specific for smooth-muscle actin, stained myoepithelial cells that appeared distended around acini but not ducts. Polyclonal antiserum to laminin delineated a basement membrane separating each acinus from the surrounding stroma. Vimentin mAb V9 stained fibroblasts and blood vessels in the stroma. The latter were also stained by mAb 1A4 to smooth-muscle actin. 

Immunologic Analysis of SMC/Muc4 in the Rat Lacrimal Gland

The expression of SMC/Muc4 (Muc4) in the rat lacrimal gland was investigated by immunofluorescence and immunocytochemical analysis with monoclonal anti-ASGP-2. Because both subunits of SMC/Muc4 have been found as a complex in all tissues studied so far, ²⁰ antisera raised to ASGP-2 can be used for immunocytochemical analyses of SMC/Muc4. Immunofluorescence staining with anti-ASGP-2 showed staining in cell clusters expected for localization in lacrimal gland acini (data not shown), compared with the linear organization of some other epithelia. Similarly, immunocytochemical staining of SMC/Muc4 with mAbs 4F12 (Fig. 4) and 1B1 (data not shown) was observed in acinar cells in two patterns. Membrane staining was abundant in some acini (Fig. 4C) and delineated the cell boundaries. However, a greater percentage of the acini (approximately 80%) exhibited primarily granular staining within the cells (Fig. 4D) , possibly in cellular secretory granules. Surfaces and luminal contents of some ducts were observed to be prominently stained (Fig. 4) , suggesting that SMC/Muc4 was produced by these cells as well as acinar cells. That the staining of the ducts was not due to adsorbed secreted material from the acini was suggested by staining of ductal cells with antibody against the cytoplasmic domain of ASGP-2 (Fig. 4F) , which is not present in the secreted form of SMC/Muc4. This antibody (anti-C-pep) also stained membranes of lacrimal acinar cells, but not cytoplasmic granules (Table 1) . 

Detection and Localization of ErbBs in the Rat Lacrimal Gland

The presence of soluble SMC/Muc4 in the rat lacrimal gland suggests that it is a source of mucin for the tear fluid. However, the reason for the presence of the membrane form is uncertain. One possibility is that SMC/Muc4 acts as a ligand for ErbB2, which is present in the gland to modulate cellular regulatory functions. Therefore, the complement of ErbBs in the lacrimal gland was analyzed by immunoblot analysis and immunocytochemistry. Immunoblot data in Figure 5 demonstrate that all four ErbB family members were present in the rat lacrimal gland. However, immunolocalization studies (Fig. 6) suggest that the different ErbBs have distinctly different locations and functions in the gland. ErbB2 appeared to be the most widely distributed, exhibiting a membrane localization in most of the acini of the gland, when analyzed by multiple anti-ErbB2 antibodies (Table 1) . Comparison of the ErbB2 distribution in Figure 6 with that of the membrane form of SMC/Muc4 in Figure 4C suggests that they are colocalized in cells where they are coexpressed. Colocalization was verified by two-color confocal analysis with immunofluorescence using anti-ASGP-2 and anti-ErbB2. Acini exhibiting the membrane form of SMC/Muc4 exhibited a high degree of colocalization of the receptor and mucin (Fig. 7) . In contrast, acini exhibiting the cytoplasmic granule form of SMC/Muc4 did not exhibit colocalization (Fig. 7) . 

SMC/Muc4 Complex with ErbB2 in Rat Lacrimal Gland

The colocalization of SMC/Muc4 and ErbB2 in cells in the lacrimal gland suggests that they may be present in a complex. To test for complex, lacrimal gland lysates were immunoprecipitated with either polyclonal anti-ASGP-2 or a monoclonal anti-ASGP-2 (13C4) or a nonimmune control antibody (Fig. 8) . The immunoprecipitates were then immunoblotted with anti-ErbB2 (Fig. 8) . Coimmunoprecipitation was observed with each of the anti-ASGP-2 antibodies, but not the control. 

Discussion

Previous studies of the expression of mucins in ocular tissues have mainly involved mRNA analyses. Northern blot analysis has shown expression of MUC1 in stratified epithelia of cornea and conjunctiva ⁵ and has also shown the expression of MUC4 and MUC5AC genes by human conjunctival epithelia. ^{3 4} MUC4 and MUC5 transcripts have been observed in stratified conjunctival epithelium and conjunctival goblet cells, respectively, by in situ hybridization. There are conflicting reports whether MUC2 mRNA is present in conjunctival epithelia. ^{4 30 31} Such studies of mRNA expression determine the cellular source of mucins but not their final location. Moreover, the presence of MUC transcripts does not mean that they are actually translated in cells, because it has been shown that mucin gene transcription can occur without translation. ³² Using antisera to glycosylated epitopes, MUC1 has been immunolocalized to the basal layers of the stratified conjunctival epithelium and the apical membranes of the cornea and conjunctiva. ⁵ MUC2 and MUC5AC have been detected by antibody analyses in extracts of conjunctiva. ³³  

In addition to the above tissues and the goblet cells of the conjunctiva, ² the lacrimal gland has been implicated as a potential source of mucins for the tear film. A high sialic acid content is present in lacrimal tissue and lacrimal fluid. ³⁴ The gland is also positive for acidic and neutral glycoprotein stains. ^{35 36} Our study now provides direct evidence that the rat lacrimal gland synthesizes mucins, at least in the case of SMC/Muc4. These results indicate that it is produced by the cytokeratin-18–positive acinar cells and is transported through the interlobular ducts, where it presumably reaches the ocular surface and contributes to the mucin composition of the tear film. This premise is supported by the presence of SMC/Muc4 in the tear fluid. ⁷ SMC/Muc4 is produced in both soluble and membrane-bound forms in the lacrimal gland. Thus, in addition to the contribution as a soluble mucin to the tear film, ⁷ the membrane form of SMC/Muc4 may provide a protective role at the cell surface of the acinar and ductal cells of the gland. Thus, the mucin can extend from the cell surface and block access to harmful agents, such as microbes. ¹⁹  

The presence of ErbB2 in most lacrimal acini and its colocalization with SMC/Muc4 suggests that the protective role may assume a second form in some of the acini. Activation of the ErbB2 by its association with SMC/Muc4 ²¹ may induce signaling events important to lacrimal cell function. The function involved in such signaling is still unknown, but recent studies have implicated SMC/Muc4 as an antiapoptotic agent in cells expressing ErbB2. ³⁷ Thus, SMC/Muc4 may serve as an intrinsic survival factor in cells in which it is coexpressed with ErbB2. Additional work is under way to test that hypothesis. 

Expression of SMC/Muc4 in membrane and soluble forms in different cell types raises the question of how production of these forms differs. Biosynthesis studies in strain 13762 ascites tumor cells indicate that the membrane form follows a typical processing pathway of a cell surface protein. ¹⁶ Experiments with transfected cells suggest that the soluble form is derived from the membrane form by a proteolytic processing step inside the cell (Komsatsu M, unpublished observations, 1998). This scenario is consistent with the observation of SMC/Muc4 in secretory granules and its release from the colon by secretagogue. ²⁶ These observations demonstrate that one function of SMC/Muc4 is to serve as a secreted mucin in fluids such as milk, ²⁶ saliva, ¹⁶ and tears. In each of these cases the secretory cells of the gland secreting the mucin are organized in acini connected to ducts through which the mucin must pass. In all these glands SMC/Muc4 is secreted as a soluble form and is present on the membranes of the acinar and duct cells. The function(s) of the membrane form are still unclear, but the colocalization and complex formation with ErbB2 provides evidence to support a role in cellular signaling. Additional studies are needed to address the pathways initiated by the receptor activation in these glands. 

In conclusion, the lacrimal gland was thought to be responsible only for the secretion of electrolytes, water, and tear fluid proteins. This present study of SMC/Muc4 in the lacrimal gland now shows that it is able to synthesize mucins, which may play a role in the tear fluid as protective agents. Moreover, the membrane form of the SMC/Muc4 may have a second role as a ligand for the tyrosine kinase receptor, which is expressed in most of the acini of the gland. Signaling events induced by this complex may contribute further to protection by providing an antiapoptotic mechanism, as demonstrated by our recent studies on the effect of SMC/Muc4 on xenotransplanted tumors and cultured tumor cells. ³⁷  

 Supported by Grants EY-12343 and CA-74072 from the National Institutes of Health.

 Submitted for publication April 16, 2001; revised July 20, 2001; accepted August 1, 2001.

 Commercial relationships policy: N.

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked“ advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Northern blot analysis of SMC/Muc4 mRNA in rat lacrimal gland. Total RNA was isolated from rat lacrimal gland and 13762 MAT-C1 ascites tumor cells. Samples were run on a 1% agarose gel. (A) Gel stained with ethidium bromide; (B) gel transferred and probed with an SMC/Muc4-specific RNA probe, A2G2-9. (A, arrows) Bands for 18S and 28S ribosomes used as loading controls; (B, arrow) position of the SMC/Muc4 transcript. The broad bands and heterogeneity of the SMC/Muc4 mRNA are typical of mucin transcripts. ³⁸

Figure 2.

 

Immunoblot analyses of SMC/Muc4 expression in rat lacrimal gland. Lysates were prepared from whole lacrimal gland, isolated acini (top), or a membrane preparation (bottom). The expression of SMC/Muc4 was analyzed by immunoblot with anti-ASGP-2 mAb 4F12. The MAT-C1 subline of the 13762 ascites cells was used as a positive control. Two bands are observed for ASGP-2, the transmembrane subunit of SMC/Muc4. The lower band at approximately 120 to 140 kDa is commonly observed in rat epithelial tissues and the ascites cells. The upper band at approximately 200 to 250 kDa is novel and has been observed primarily in the lacrimal gland and salivary glands. ²⁹ The broad bands are typical of highly glycosylated proteins.

Figure 3.

 

Analysis of soluble and membrane-associated SMC/Muc4 forms in rat lacrimal gland. Serial immunoprecipitations with anti-C-pep and anti-ASGP-2 antisera were performed on rat lacrimal gland acini (top) or acinar membranes (bottom) solubilized in RIPA buffer. Tissue lysate was immunoprecipitated with anti-C-pep antisera three times to deplete the membrane-bound forms. The supernatant was then immunoprecipitated with polyclonal anti-ASGP-2 antiserum to precipitate soluble forms. All immunoprecipitates were analyzed by immunoblot analysis with anti-ASGP-2 mAb 4F12. The negative control for immunoprecipitation was nonimmune serum. The positive control for the immunoblot was the 13762 mammary ascites tumor cells from which SMC/Muc4 was originally isolated. The heavy bands at the bottom of the gels were from Ig used in the immunoprecipitations.

Figure 4.

 

Immunocytochemical staining of SMC/Muc4 in the rat lacrimal gland. Paraffin sections of the rat lacrimal gland were stained with secondary antibody only (A), anti-ASGP-2 mAb 4F12 (B–E), or polyclonal anti-C-pep against the cytoplasmic domain of ASGP-2 (F). Low magnification shows extensive staining of acinar cells (B). Higher magnifications of acini demonstrate staining of cell plasma membranes (C) and intracellular granular structures (D) in different acini. Higher magnifications also show staining of intercalated ducts with both anti-ASGP-2 mAb 4F12 (E) and with anti-C-pep (F). The latter indicates that the ductal staining was not due to adsorbed soluble form. Bars, (A, B) 63 μm; (C–E) 16 μm; (F) 31 μm.

Table 1.

 

Immunocytochemical Staining with ErbB2 and ASGP-2 Antibodies

Antibody	Type	Epitope	Sites
Dako anti-ErbB2*	pAb	Cytoplasmic domain peptide	Membrane
NM1 anti-ErbB2, †	pAb	Cytoplasmic domain peptide	Membrane
NM8 anti-ErbB2, †	mAb	Tyrosine kinase domain	Membrane, duct
NM17 anti-ErbB2, †	mAb	Cytoplasmic domain peptide	Membrane, duct
4F12 anti-ASGP-2	mAb	N-terminal domain	Membrane, granules
c-Pep anti-ASGP-2	pAb	Cytoplasmic domain peptide	Membrane

 pAb, polyclonal antibody.

*  Dako, Carpinteria, CA.

^†  Neomarkers, Fremont, CA.

Figure 5.

 

Immunoblot analyses of ErbBs in rat lacrimal gland. Ascites cell lysates were used as positive controls for ErbB2, -3, and -4. A431 tumor cells were used as a positive control for the EGF receptor (ErbB1). Irrelevant antibodies were used as negative controls for each receptor and showed no staining.

Figure 6.

 

Differential localization of the different ErbBs in rat lacrimal gland. Negative controls were run in which the first or second antibody was absent and showed no staining. Note that ErbB2 was the most widely expressed ErbB in the lacrimal gland and that it showed a staining pattern similar to membrane SMC/Muc4 shown in Figure 4C . Bars, (A, ErbB1, -2, -3) 126 μm; (A, ErbB4) 63 μm; (B, ErbB1, -2) 32 μm; (B, ErbB3) 45 μm; (B, ErbB4) 21 μm.

Figure 7.

 

Confocal immunofluorescence analysis of ErbB2 (top; Dako primary antibody; Texas red-labeled second antibody) and ASGP-2 (bottom; 4F12 primary antibody; fluorescein-labeled second antibody) colocalization in rat lacrimal gland tissue sections. Right: merged panels. There was significant colocalization of ErbB2 and SMC/Muc4 in sections containing membrane SMC/Muc4 in acini, but not in those containing cytoplasmic granule SMC/Muc4. Bars, 5 μm.

Figure 8.

 

Coimmunoprecipitation of ErbB2 with ASGP-2. Lacrimal gland tissue lysate was immunoprecipitated with anti-ASGP-2 pAb antisera, anti-ASGP-2 mAb 13C4, or nonimmune sera (negative control). All immunoprecipitates were analyzed by immunoblot analysis with anti-ErbB2 pAb antisera (Dako). Strain 13762 ascites cell lysates were used as a positive control for the position of the ErbB2 band.

The authors thank Shari Price-Schiavi and Steve Pflugfelder for advice and Pedro Salas for assistance with microscopy.