Presence of Nerves and Their Receptors in Mouse and Human Conjunctival Goblet Cells

Abstract

purpose. To determine whether neural pathways for controlling goblet cell secretion are present in mouse and human conjunctiva.

methods. Mouse conjunctiva was homogenized and subjected to electrophoresis and Western blotting to detect PGP 9.5 (indicates nerves), muscarinic receptor subtypes (indicates parasympathetic pathway), and adrenergic receptors (indicates sympathetic pathway). Mouse eyes and human conjunctival tissue were analyzed by immunofluorescence microscopy. Antibodies to vasoactive intestinal peptide (VIP), tyrosine hydroxylase (TH), dopamine β-hydroxylase (DBH), and muscarinic andα ₁- and β-adrenergic receptor subtypes were used.

results. Western blot demonstrated PGP 9.5, M₁, M₂, and M₃ muscarinic receptors and α_1A-,β ₁-, β₂-, and β₃-adrenergic receptors in mouse conjunctiva. Immunoreactivity for VIP, TH, and DBH was found adjacent to mouse and human goblet cells. M₁ and M₂ muscarinic receptors were identified throughout mouse conjunctiva, but M₃ receptor was predominantly on goblet cells. All three muscarinic receptor subtypes were detected on goblet cells in human conjunctiva. α_1A-Adrenergic receptors were found on epithelial cells and on goblet cells in mouse and human conjunctiva. β₁- and β₂-Adrenergic receptors were found on both epithelial and goblet cells in mouse conjunctiva, but not on human conjunctival cells.β ₃-Adrenergic receptors were found on both epithelial and goblet cells in human conjunctiva but not on mouse conjunctival cells.

conclusions. The following conclusions were drawn: parasympathetic nerves and M₁, M₂, and M₃ muscarinic receptors, as well as sympathetic nerves are present on mouse and human goblet cells. The adrenergic receptors β₁ andβ _2, but not α_1A and β₃ are present on mouse conjunctival goblet cells, whereas α_1A and β_3, but not β₁ and β₂ are present on human conjunctival goblet cells, suggesting that these nerves and receptors could activate goblet cell secretion in mouse and humans.

Human conjunctival stroma has a plexiform network of sensory and sympathetic nerves. ¹ The sensory nerves are derived from the nerves of the lids (naso-ciliary, lacrimal, frontal, and infra-orbital nerves) and from the ciliary nerves, which arise from the ophthalmic division of the trigeminal nerve. The sympathetic nerves are derived from the sympathetic plexus associated with the branches of the ophthalmic artery to the conjunctiva. Ruskell ² also found a parasympathetic pathway from the facial nerve to the conjunctiva in human and cynomolgus monkey. However, direct innervation of conjunctival epithelium and even less the goblet cells has traditionally been considered unlikely. ^{2 3}  

The innervation of the conjunctiva has also been studied in other species such as rabbit, ⁴ rat, ^{5 6 7} and monkey. ^{2 8 9} In these species the conjunctiva receives sensory, parasympathetic, and sympathetic innervation of trigeminal, pterygopalatine, and superior cervical ganglia origin respectively. ⁸ The innervation of the murine conjunctiva has not yet been investigated. 

The extraordinary importance of the mucus layer of the tear film in maintaining the health of the ocular surface is well established. Mucins are secreted mainly by the conjunctival goblet cells, although the corneal epithelium and some of the stratified squamous cells from the conjunctiva (the so-called second secretory system) are known to be another source of mucin. Variation in the amount of mucins secreted is associated with several ocular surface diseases with both mucus overproduction and underproduction harmful to the ocular surface. Sjögren’s syndrome, ocular cicatricial pemphigoid, Stevens-Johnson syndrome, keratoconjunctivitis sicca, rosacea, xerophthalmia, trachoma, or alkali burns are some examples of ocular surface conditions associated with altered mucin secretion. This suggests that the mucin secretion is under tight control. 

As previously shown, ¹⁰ the mode of conjunctival goblet cell secretion is different from the one used by most other secretory cell types. Goblet cells appear to secrete mucus in an apocrine manner, ¹¹ that is, all or most of the secretory granules are discharged upon stimulation. In contrast, merocrine secretion is when only a small percentage of secretory granules are released. ¹² If all goblet cells secrete in response to a stimulus, the mucous supply of the ocular surface would be rapidly depleted leaving no reserve. Thus, to produce a graded secretory response of mucus secretion from conjunctival goblet cells, the number of cells responding to a stimulus must be controlled. 

One level at which mucin secretion can be controlled is at the level of neural regulation. Other systems in the body contain goblet cells: the respiratory tract, which includes the nasal mucosa, the trachea, bronchi, and bronchioles; the gastrointestinal tract, which includes the small intestine and colon; and the pancreatic ducts. In these tissues, goblet cells respond to neural stimuli, ^{13 14 15} although these epithelia are not directly innervated. Pancreatic ducts contain nerves, but direct innervation of the goblet cells has not yet been demonstrated. ¹⁶ We previously demonstrated the presence of parasympathetic and sympathetic, but not sensory, nerves adjacent to rat conjunctival goblet cells ¹⁷ and that goblet cell mucous secretion can be under the control of nerves in that species. ^{18 19} Neurotransmitters released by these nerves in the subepithelial plexus or in the epithelium can diffuse to the goblet cells and induce secretion. Exogenous addition of parasympathomimetic or sympathomimetic agonists in vivo and parasympathomimetic agonists in vitro induced conjunctival goblet cell mucous secretion. ^{10 18 20} Thus, neurally mediated regulation of mucin secretion from rat conjunctival goblet cells is possible. 

Direct innervation of conjunctival goblet cells in other species has not yet been demonstrated. Therefore, the purpose of the present study was to determine whether conjunctival goblet cells from mouse and human are innervated and if so, whether those nerves are parasympathetic and/or sympathetic. We also determined whether conjunctival goblet cells express muscarinic and adrenergic receptors. 

Materials and Methods

Materials

All chemicals used were purchased from Sigma (St. Louis, MO) unless otherwise indicated. Formaldehyde was obtained from Polysciences, Inc. (Warrington, PA), the microscope slides were Colorfrost/Plus from Fisher Scientific (Pittsburgh, PA), and the optimal cutting compound (OCT) from Sakura (Tokyo, Japan). Helix pomatia (HPA) and Ulex europaeus-I (UEA-I) agglutinins conjugated to Texas red and tetramethylrhodamine isothiocyanate (TRITC), respectively, were obtained from Sigma. Rabbit polyclonal antibodies to the following nerve markers and receptors were purchased: protein gene product (PGP) 9.5 (Accurate Chemical and Scientific Corporation, Westbury, NY), vasoactive intestinal peptide (VIP; Dia Sorin, Stillwater, MN), dopamine β-hydroxylase (DBH; Eugene Tech International, Inc., Ridgefield Park, NJ), tyrosine hydroxylase (TH; Calbiochem-Novabiochem Corp., San Diego, CA), M₁-, M₂-, and M₃-muscarinic receptor subtypes (R & D Antibodies, Berkeley, CA), and α_1A-,α _1B-, α_1C-,β ₁-, β₂-, andβ ₃-adrenergic receptor subtypes (Santa Cruz Biotechnology, Inc., Santa Cruz, CA). Two different antibodies (raised either in mouse or human) against α_1A orβ ₃-adrenergic receptors were used for both immunohistochemistry and Western blotting. The source, receptor subtype specificity, and species specificity for the primary antibodies to adrenergic receptors are summarized in Table 1 . The secondary antibodies for immunofluorescence experiments were fluorescein isothiocyanate (FITC)-conjugated and were purchased from Jackson Laboratories (West Grove, PA). The secondary antibodies for Western blotting were horseradish peroxidase (HRP)-conjugated anti-IgG and purchased from Santa Cruz Biotechnology, Inc. Vectashield mounting media, with or without DAPI, were from Vector Laboratories (Burlingame, CA). All the reagents for Western blotting were purchased from Bio-Rad Laboratories (Hercules, CA), and the chemiluminescence reagents for visualization from Pierce (Rockford, IL). 

Animals and Tissues

Adult Sprague–Dawley rats and BALB/c mice, from Taconic Farms, Inc. (Germantown, NY) and The Jackson Laboratories (Bar Harbor, ME), respectively, were used for all experiments. Experiments conformed to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and were approved by the Schepens Eye Research Institute Animal Care and Use Committee. Animals were anesthetized for 1 minute in carbon dioxide and decapitated, and the eyes with the lids removed. The whole eye and lids were fixed in 4% formaldehyde for at least 4 hours, rinsed in 5% sucrose dissolved in PBS, placed overnight in 30% sucrose dissolved in PBS at 4°C, embedded in OCT, and frozen. Cryostat sections (6 μm) were placed on Colorfrost/Plus slides and keep at −20°C until use. 

Human Tissues

Human bulbar conjunctival tissues from two cadaveric donors (age 66 and 77 years, male) were kindly provided by Marcia Jumblatt, University of Louisville, Louisville, KY. The research followed the tenets of the Declaration of Helsinki. The tissues were received from the Kentucky Lions Eye Bank (Louisville, KY) in culture medium. Tissues were rinsed in PBS, placed in 4% formaldehyde, and processed as described for rat and mouse tissues. Biopsies of human conjunctival tissue were obtained from Drs. Peter Rubin and Gabriel Garza of the Massachusetts Eye and Ear Infirmary. Informed consent was obtained from each patient, and the procedure was approved by the Institutional Review Board of the Massachusetts Eye and Ear Infirmary. The tissue was rinsed with PBS, embedded in OCT, and frozen. 

Immunohistochemistry

Fixed cryosections were thawed at room temperature for 1 hour, washed in PBS, and blocked in PBS containing 1% bovine serum albumin, 4% goat serum, and 0.2% to 0.3% Triton X-100. The following antibodies were used: antibody to the pan-neuronal marker PGP 9.5 (1:400) was applied for 4 hours at room temperature; antibodies to VIP (1:400) and muscarinic (1:1000) and adrenergic receptor subtypes (1:500) were applied for 20 hours at 4°C; and antibodies to TH (1:400; the enzyme responsible for converting tyrosine to DOPA, a precursor of dopamine) or to DBH (1:400; the enzyme responsible for converting dopamine in norepinephrine) were applied for 36 hours at 4°C. The secondary antibodies (1:100) were FITC-conjugated donkey anti-rabbit or anti-goat (for α₁ andβ ₃ receptor antibodies, respectively) and were applied for 1 hour at room temperature. The sections were double-labeled with TRITC-conjugated UEA-I (for rat sections; 1:1000) or Texas red–conjugated HPA (for mouse and human sections; 1:1000) for goblet cell identification. After incubation, sections were washed in PBS and mounted on coverslips. Sections were viewed and photographed using a Zeiss Axiophot microscope (Thornwood, NY). Negative controls included the omission of the primary antibody. Rat sections were used as positive controls. ^{17 18}  

Electrophoresis and Immunoblotting

Rat and mouse conjunctiva were homogenized in RIPA buffer plus proteinase inhibitors (10 mM Tris-HCL, pH 7.4, containing 150 mM NaCl, 1% deoxycholic acid, 1% Triton X-100, 0.1% SDS, 1 mM EDTA, 10 mg/ml phenylmethylsulfonyl fluoride, 5 U/ml aprotinin, and 100 nM sodium orthovanadate). After homogenization, the samples were centrifuged at 2000g for 30 minutes at 4°C to remove unbroken cells and nuclei. Proteins in the homogenate were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) on 15% acrylamide gels, according to the method of Laemmli. ²¹ Proteins were then transferred to nitrocellulose membranes, blocked in 5% dried milk in TBST (10 mM Tris-HCl, pH 8.0, 150 mM NaCl, and 0.05% Tween-20), and incubated with indicated antibody for 1 hour at room temperature. Membranes were washed three times in TBST and incubated with HRP-conjugated anti-rabbit IgG (1:1000) for 1 hour at room temperature. Immunoreactive bands were visualized using the enhanced chemiluminescence method. Rat corneal homogenate was used as positive control. 

To detect muscarinic and adrenergic receptors, a membrane fraction was prepared from rat and mouse conjunctival homogenates. After homogenization in homogenization buffer plus proteinase inhibitors (30 mM Tris-HCL, pH 7.5, 10 mM EGTA, 5 mM EDTA, 1 mM dithiothreitol, 250 mM sucrose, 10 mg/ml phenylmethylsulfonyl fluoride, 5 U/ml aprotinin, and 100 nM sodium orthovanadate), the samples were centrifuged at 2000g for 15 minutes at 4°C. The pellet was then resuspended in the homogenization buffer and centrifuged at 100,000g for 1 hour at 4°C. The pellet (membrane fraction) was resuspended in homogenization buffer. Proteins in the membrane fraction were separated by SDS-PAGE on 10% acrylamide gels and then transferred to nitrocellulose membranes, as described above. The primary antibodies were incubated for 1 hour at room temperature (1:500). The nitrocellulose membranes were processed as described above. Cellular membranes from rat cornea and kidney were used as positive controls. 

Results

Identification of Goblet Cells using Lectins in Rat, Mouse, and Human Conjunctiva

Plant proteins called lectins have the ability to bind to specific carbohydrate residues of glycoconjugates ²² and are therefore useful as molecular probes to detect glycoprotein content in conjunctival goblet cells. We previously demonstrated that Ulex europaeus agglutinin I (UEA-I) binds almost exclusively to goblet cells in rat conjunctiva. ¹⁸ However, little is known about mouse conjunctival goblet cells and the lectins that recognize its glycoconjugates. ²³ Thus, we tried a panel of lectins with different oligosaccharide residue specificities. Because we previously showed that UEA-I and HPA (Helix pomatia agglutinin) specifically recognized rat conjunctival goblet cells, ¹⁰ we tested these lectins in mouse conjunctiva. We also tested Lotus tetragonolobus (LTL), Euonymus europaeus (EEL), and Bandeiraea simplifolia (BSL-1) agglutinins. HPA recognized mouse goblet cells more specifically than UEA-I, with no binding in the stratified squamous cells (Fig. 1B ). The other lectins were negative (data not shown). Thus, we used HPA (conjugated to Texas red) to identify goblet cells in mouse conjunctiva. 

Because Kawano et al. ²⁴ have reported that lectins binding to GalNAc residues detect conjunctival goblet cell with minimal binding to the nongoblet, epithelial cells, we tried HPA in human cryosections, and we also found that HPA was specific for goblet cells (Fig. 1C) . Therefore, we used HPA for both mouse and human goblet cell identification and UEA-I for rat goblet cells as described previously and shown in (Fig. 1D) . ¹⁸ In mouse and human, but not rat, the lectin used also bound to the mucin layer on the surface of the conjunctiva. 

We observed that mouse goblet cells are clustered and their shape is elongated, being embedded in the whole stratified, squamous epithelium, as they are in the rat conjunctiva. We also found that mouse goblet cells follow a distribution pattern along the conjunctiva, similar to that in the rat. They are absent from the palpebral conjunctiva, but present in clusters in forniceal and bulbar areas. These results were confirmed by Alcian blue/periodic acid-Schiff reagent staining of methacrylate mouse sections (data not shown). No author has previously reported information about mouse conjunctival goblet cells. 

Localization of Nerves in Mouse, Human, and Rat Conjunctival Goblet Cells

In a first set of experiments, we determined the presence of nerves in the conjunctiva by Western blot analysis. The presence of PGP 9.5, a cytosolic ubiquitin carboxyl-terminal hydrolase protein present in almost all neuronal types, ²⁵ was analyzed in conjunctival homogenates from rat and mouse by immunoblotting and from rat corneal homogenate, the positive control. As expected, a single 27-kDa band appeared in the two species (Fig. 1A) , indicating that nerves are present in the conjunctiva of rats and mice. 

We next studied the type of nerves and their distribution in the mouse, human, and rat conjunctiva by using antibodies against PGP 9.5, VIP (a neurotransmitter for the parasympathetic nerves), TH (indicates sympathetic nerves ²⁶ ), and DBH (indicates sympathetic nerves ²⁶ ). PGP 9.5-immunoreactive fibers were detected throughout the entire mouse conjunctival epithelium between stratified squamous cells and surrounding the goblet cell clusters (Fig. 1B) . In human conjunctiva, immunoreactivity to PGP 9.5 appeared to surround the goblet cells (Fig. 1C) . In addition, in the human conjunctiva PGP 9.5 immunoreactivity was found on the goblet cell body. In rat conjunctiva PGP 9.5 immunoreactivity was found along the junction between the epithelium and the stroma where the base of the goblet cells is located (Fig. 1D) . Immunoreactivity was also detected in the stroma in particular surrounding blood vessels. 

Immunoreactivity to VIP was observed in both squamous epithelial and goblet cells in mouse conjunctiva (Fig. 2A ). The goblet cell staining was mainly localized to the lateral walls and subjacent to the secretory granules in the goblet cells (Fig. 2A) . Some nonspecific staining also appeared apically as in the rat (Fig. 2C) . For human conjunctiva, immunoreactivity was seen in the basal epithelial cell layer as well as surrounding the body of the goblet cells (Fig. 2B) . Diffuse immunoreactivity was detected around other epithelial cells in the conjunctiva. Immunoreactivity to VIP was mainly localized subjacent to the secretory granules in rat conjunctival goblet cells (Fig. 2C) . 

TH-containing nerve fibers were immunodetected in the conjunctival epithelium of the three species (Fig. 3) . These fibers were localized subjacent to the secretory granules in goblet cells of mouse and along the epithelial-stromal junction (Fig. 3A) . In human conjunctiva, immunoreactivity for TH was detected subjacent to the goblet cell secretory granules and adjacent to stratified squamous cells (Fig. 3B) . The extensive fluorescence in the stroma was nonspecific binding. In the rat conjunctiva nerves were localized surrounding goblet cell secretory product and in blood vessels of the stroma (Fig. 3C) . Similarly, DBH-containing nerve fibers were immunodetected in the conjunctival epithelium of mouse and human (data not shown). 

These results, summarized in Table 2 , show that mouse, human, and rat conjunctival goblet cells are innervated and that those nerves may be parasympathetic and sympathetic, because VIP, TH, and DBH were immunodetected. 

Localization of Muscarinic Receptors in Mouse, Human, and Rat Conjunctival Goblet Cells

The presence of muscarinic neurotransmitter receptors in the membrane fraction from rat and mouse conjunctival homogenates was determined by Western blot analysis and immunoblotting. Muscarinic and adrenergic receptors were investigated using polyclonal antibodies specific to M₁, M₂, and M₃ muscarinic receptor subtypes. The specificity of the antibodies used has been described previously (Ref. ¹⁸ and unpublished data). One major band was detected for the M₁, M₂, and M₃ muscarinic (approximately 55 kDa) receptor subtypes in rat and mouse conjunctiva as well as rat corneas, the positive control (Fig. 4) . 

The cellular distribution of muscarinic receptors was determined by immunofluorescence microscopy. Polyclonal antibodies to M₁, M₂, and M₃ muscarinic receptor subtypes were used in mouse, human, and rat conjunctival cryosections. In Figures 5 and 6 , each section was double-labeled with HPA to indicate the position of goblet cells. Immunoreactivity to M₁ and M₂ receptors was detected in both stratified squamous epithelial cells and goblet cells in mouse (Figs. 5A 5C) and rat (data not shown) conjunctiva. M₃ receptors were observed predominantly in goblet cells of mice (Fig. 5E) and rats. ¹⁸ The location of goblet cells in each section is shown in Figures 5B 5D and 5F . The immunoreactivity of M₁ receptors in stratified squamous cells was less intense and more diffuse than for M₂ receptors (Fig. 5A) . Immunoreactivity of M₂ receptor was seen mainly in the apical epithelial cells (Fig. 5C) . Immunoreactivity to M₃ receptors was more intense in the goblet cells compared with the stratified squamous cells and located subjacent to the goblet cell secretory granules as in the rat (Fig. 5E) . Intense immunoreactivity to the three receptor subtypes was localized in goblet cells subjacent to the secretory granules and above the nuclei in mouse, with M₃ > M₂ > M₁ receptors. In the human conjunctiva, immunoreactivity to all three muscarinic receptor subtypes was observed in occasional epithelial cells throughout the conjunctiva and surrounding and on the body of the goblet cells (Figs. 6A 6C 6E) . The location of goblet cells is indicated in Figures 6B 6D and 6F . Intense immunoreactivity to M₂ and M₃ muscarinic receptors was also detected in the basal epithelial cell layer (Figs. 6C 6E) . 

Localization of Adrenergic Receptors in Mouse, Human, and Rat Goblet Cells

The presence of adrenergic receptors in mouse, human, and rat conjunctiva was determined by Western blot analysis. A band of approximately 66 kDa was detected forα _1A-adrenergic receptor subtypes in mouse, but not rat, conjunctival membrane fraction, using an antibody raised in human (Santa Cruz antibody sc-1475; Fig. 7 ). A band of 64 kDa was detected forβ ₁-adrenergic receptor subtypes (Santa Cruz antibody sc-9041) present in the rat and mouse conjunctival membrane fractions (Fig. 7) . Similar results were obtained forβ ₂-adrenergic receptor subtypes (Santa Cruz antibody sc-570; Fig. 7 ). Two bands (66 and 67 kDa) were detected forβ ₃-adrenergic receptor subtypes in mouse but not rat conjunctival membrane fraction using antibodies that either recognize rat and mouse (Santa Cruz antibody sc-1473; Fig. 7 ) or human epitopes (Santa Cruz antibody sc-1472). 

The cellular distribution of adrenergic receptors was determined by immunofluorescence microscopy. In Figures 8 9 and 10 each section was double-labeled with HPA to indicate the position of goblet cells. Immunoreactivity to theα _1A-adrenergic receptor subtypes was detected diffusely in stratified squamous epithelial cells but not in goblet cells in mouse conjunctival epithelium (Fig. 8A) . Immunoreactivity to the α_1A-adrenergic receptor subtypes was detected in stratified squamous epithelial cells and also was detected in goblet cells in human conjunctival epithelium (Fig. 8C) . The location of goblet cells was indicated in Figures 8B and 8D . Theα _1A-adrenergic receptor subtypes were not detected in rat conjunctiva by immunofluorescence microscopy (data not shown). 

Immunoreactivity to β₁- andβ ₂-adrenergic receptors was seen in both stratified squamous epithelial cells and goblet cells in mouse (Figs. 9A 9C ) and rat (data not shown). Immunoreactivity toβ ₁-adrenergic receptor was distributed in plasma membranes of stratified squamous cells and goblet cells in the mouse conjunctiva (Fig. 9A) . β₁-Adrenergic receptor immunoreactivity was preferentially localized subjacent to the secretory granules in rat conjunctiva (data not shown). The immunoreactivity of β₂-adrenergic receptor was distributed in the basal portion of goblet cells as well as in the lateral membranes in mouse (Fig. 9C) and rat (data not shown) conjunctiva. Immunoreactivity to theβ ₃-adrenergic receptor was not detected in either mouse or rat conjunctiva (data not shown). 

In a pattern similar to the localization of M₂ and M₃ receptors in conjunctival stratified squamous cells, immunoreactivity toβ ₁-adrenergic receptor was seen only in the basal epithelial cell layer of human conjunctiva sections (Fig. 10A ). Immunoreactivity to the β₂-adrenergic receptor subtype was not detected in human conjunctiva. However, immunoreactivity to the β₃-adrenergic receptor subtypes was detected in both stratified squamous epithelial cells and goblet cells in human conjunctival epithelium (Fig. 10B) . 

The results summarized in Table 2 indicate that mouse, human, and rat conjunctival goblet cells express muscarinic receptors. Mouse and human goblet cells express α _1A-adrenergic receptors. Mouse and rat goblet cells express β₁- andβ ₂-adrenergic receptors, whereas only human goblet cells express β₃-adrenergic receptors. 

Discussion

Using SDS-PAGE and Western blot analysis along with immunofluorescence microscopy, we have identified parasympathetic and sympathetic nerves adjacent to mouse and human conjunctival goblet cells. In addition, muscarinic α_1A- andβ -adrenergic receptors were present on mouse and human conjunctival goblet cells. 

PGP 9.5 is a well-established marker for nerve fibers in other tissues. ^{27 28} Using an antibody against this protein we were able to identify immunoreactive nervelike fibers close to rodent and human conjunctival goblet cells. Recently, Seifert and Spitznas ³ reported, using this marker, that no nerve fibers were seen by electron microscopy in association with human goblet cells of the lid. Goblet cells are absent in the palpebral conjunctiva of rat, ²⁹ and they are less abundant in that particular area than in the fornices in humans. ³⁰ It is possible that those goblet cells in the most critical regions of the conjunctiva, where they are more abundant, are more likely to be tightly regulated, either neurally or by other ways. This may explain the differences in PGP 9.5 immunoreactivity in this study and that of the aforementioned authors, as they studied only the goblet cells in the conjunctiva associated with the lid. Moreover, these authors obtained their lid conjunctiva samples from ectropium surgery. It is known that ectropium patients frequently have chronic conjunctivitis, epiphora, and ocular desiccation problems associated with the exposure of the ocular surface. Therefore, that conjunctiva cannot be considered as a normal tissue to be studied. 

Immunoreactivity to VIP was seen around mouse and human goblet cells as well as in the stratified squamous epithelia. This parasympathetic neurotransmitter has been reported to stimulate mucin secretion from cultured intestinal goblet cells and cat tracheal cells. ^{31 32} Moreover, this 28-amino acid peptide stimulated pancreatic goblet cell degranulation in isolated guinea pig ducts. ^{15 33} Our group has previously demonstrated immunoreactivity for the VIP receptor type 2 in rat conjunctival goblet cells and that VIP stimulated rat conjunctival goblet cell mucin secretion. ¹⁸ The presence of VIP binding sites has also been reported in rat and rabbit conjunctiva. ³⁴ However, Seifert and Spitznas ³ were not able to see VIP-immunopositive nervelike structures adjacent to human goblet cells. Again, regional differences in the goblet cells studied may be responsible for the difference in VIP immunoreactivity. 

Muscarinic receptors have been identified by ³H-quinuclidinyl benzilate binding and in situ hybridization in isolated villus and crypt cells of the rat small intestine and colon and to human airway mucosal glands, respectively. ^{35 36 37 38} In the present work, we identified M₁, M₂, and M₃ muscarinic receptor subtypes on conjunctival goblet cells from mouse and human. The M₃ subtype was present predominantly on goblet cells in the mouse, whereas all three subtypes were present on human goblet cells. M₁ and M₂ receptors were also localized to the basal cell layer of stratified, squamous epithelial cells in human conjunctiva. 

Immunoreactivity to TH was seen in close apposition to goblet cell in the three species studied. The enzymatic synthesis of norepinephrine takes place at the sympathetic neuroeffector junction where tyrosine is actively transported into the axoplasm to be converted to dopa by TH and then to dopamine by DBH. ²⁶ Then, dopamine is transported to the cytoplasmic vesicles where norepinephrine is synthesized and stored. We also detected immunoreactivity to DBH in minute swellings that were located around goblet cells and stratified squamous epithelial cells. The presence of these two cytoplasmic enzymes shows that rodent and human conjunctiva both contain sympathetic nerves that may potentially release dopamine or norepinephrine to mediate the regulation of goblet cell mucin secretion. Norepinephrine activates both α₁- and β-adrenergic receptors. We found that bothα _1A- and β-adrenergic receptors are present on conjunctival goblet cells. Unlike muscarinic receptors, the adrenergic receptors displayed species specificity.α _1A-Adrenergic receptors were present in mouse and human but not in rat conjunctiva. Both goblet cells and stratified squamous epithelial cells from mouse and rat showed immunoreactivity for β₁- and β₂- but notβ ₃-adrenergic receptor subtypes. Onlyβ ₃-, but not β₁- orβ ₂-, adrenergic receptor subtypes were present in both goblet cells and stratified squamous epithelial cells from human conjunctiva. 

Mouse and rat conjunctival goblet cells showed a different pattern of immunoreactivity for nerves and receptors than that shown by human conjunctival goblet cells. The distinct patterns may be explained by the observed differences in the shape and arrangement of the goblet cells in the conjunctiva between these species. Although goblet cells in human conjunctiva are individually present in the apical portion of the epithelium and display a more rounded morphology, rodent goblet cells are clustered in groups of variable number of quite elongated cells embedded within the epithelium. The cell bodies of rodent, but not human, goblet cells reach the basement membrane. 

In conclusion, we demonstrated that parasympathetic and sympathetic nerves as well as muscarinic and adrenergic receptors are present in conjunctival goblet cells from mouse and human, similar to the rat. All three muscarinic receptors were present in mouse, human, and rat conjunctiva but there was species specificity in the presence of α_1A- and β-adrenergic receptor subtypes. This suggests that nerves regulate goblet cell secretion in all three species studied, including human. 

 Supported by National Institutes of Health Grant EY09054 and a grant from the NATO Scientific Program.

 Submitted for publication October 25, 2000; revised May 11, 2001; accepted May 31, 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.

 

Table 1.

 

Antibodies Against Adrenergic Receptor Subtypes Used for Western Blot Analysis or Immunofluorescence Microscopy

Antibodies	Species Specificity	Western Blot	Immunofluorescence
α1A-AR (c-19) sc-9352	Human	−	+
α1A-AR (R-20) sc-1475	Rat	+	+
	Mouse	+	+
β2-AR (M-20) sc-570	Rat	+	+
	Mouse	+	+
β2-AR (H-73) sc-9042	Rat	+	+
	Mouse	+	+
	Human	−	+
β3-AR (M-20) sc-1473	Rat	+	−
	Mouse	+	−
β3-AR (C-20) sc-1472	Human	−	+

 All antibodies were purchased from Santa Cruz Biotechnology. −, not used; +, used.

Figure 1.

 

Immunolocalization of PGP9.5, a general marker for nerves, in mouse, human, and rat goblet cells. Western blot analysis (A) of homogenates from mouse and rat conjunctiva showed a band of 27 kDa. Positive control (PC) was rat cornea. Cryosections from mouse eyes (B), human conjunctival biopsy (C), and rat eyes (D) were triple-labeled with FITC-conjugated antibody anti-PGP 9.5 (green color) to indicate nerves, Texas red–conjugated HPA (for mouse and human) or TRITC-UEA (for rat; red color) to indicate location of goblet cells, and 4,6, diamidino-2-phenylindole (DAPI) to identify nuclei (blue color). Nerves were detected coursing through the conjunctival epithelium and adjacent to or surrounding goblet cells (arrows) in all three species. These pictures are representative of at least three different experiments. Magnification,× 380.

Figure 2.

 

Immunolocalization of VIP,a neurotransmitter for parasympathetic nerves, in mouse, human, and rat goblet cells. Cryosections from mouse eyes (A), human conjunctival biopsy (B), and rat eyes (C) were triple-labeled with FITC-conjugated antibody anti-VIP (green color), Texas red–conjugated HPA (for mouse and human) or TRITC-UEA (for rat; red color) to indicate location of goblet cells, and 4,6, diamidino-2-phenylindole (DAPI) to identify nuclei (blue color). Parasympathetic nerves were detected coursing through the conjunctival epithelium and adjacent to or surrounding goblet cells (arrows) in all three species. VIP bound nonspecifically to the apical surface of mouse and rat conjunctiva. These pictures are representative of at least three different experiments. Magnification, ×380.

Figure 3.

 

Immunolocalization of TH, an enzyme involved in the neurotransmitter synthesis in sympathetic nerves, in mouse, and rat goblet cells. Cryosections from mouse eyes (A), human conjunctival biopsy (B), and rat eyes (C) were triple-labeled with FITC-conjugated antibody anti-TH (green color), Texas red–conjugated HPA (for mouse and human) or TRITC-UEA (for rat; red color) to indicate location of goblet cells, and 4,6, diamidino-2-phenylindole (DAPI) to identify nuclei (blue color). Sympathetic nerves were detected traversing the conjunctival epithelium and adjacent to goblet cells (arrows) in mouse, human, and rat. Antibody to TH bound nonspecifically to the stroma of the human conjunctiva. These pictures are representative of at least three different experiments. Magnification, ×380.

Table 2.

 

Presence of Parasympathetic and Sympathetic Nerves and Receptors in Mouse, Human, and Rat Conjunctival Goblet Cells

			Mouse	Human	Rat
Nerves		PGP 9.5	+	+	+
Parasympathetic	Neurotransmitters	VIP	+	+	+
	Receptors	M1	+	+	+
		M2	+	+	+
		M3	+	+	+
Sympathetic	Neurotransmitters	DBH	+	+	+
		TH	+	+	+
	Receptors	α1A	+	+	−
		β1	+	−	+
		β2	+	−	+
		β3	−	+	−

Figure 4.

 

Western blot of M₁, M₂, and M₃ muscarinic receptor subtypes in membrane fractions from mouse and rat conjunctival homogenate. A single major band of 55 kDa was present for each receptor subtype in mouse and rat conjunctiva as well as in the positive control (PC), which was rat cornea.

Figure 5.

 

Immunolocalization of M₁, M₂, and M₃ muscarinic receptor subtypes on mouse conjunctival goblet cells. M₁ receptors were localized in both stratified squamous cells and goblet cells (A). M₂ receptors were predominantly on stratified squamous epithelial cells (C). M₃ receptor was localized on stratified squamous epithelial cells as well as on goblet cells, preferentially distributed below the secretory granules (arrows, E). The position of goblet cells within the epithelium was identified by using HPA, which preferentially stains goblet cell secretory granules (B, D, and F). These pictures are representative of at least three different experiments. Magnification, ×320.

Figure 6.

 

Immunolocalization of M₁, M₂, and M₃ muscarinic receptor subtypes on human conjunctival goblet cells. Diffuse immunoreactivity for M₁ receptor was localized on goblet cells (arrowheads; A). M₂ and M₃ receptors were localized on goblet cells (arrowheads) as well as in the basal epithelial cell layers (arrows; C and E). The position of goblet cells within the epithelium was identified by using HPA, which preferentially stains goblet cell secretory granules (B, D, and F). These pictures are representative of at least three different experiments. Magnification, ×320.

Figure 7.

 

Western blot analysis of adrenergic receptors in membrane fractions from mouse and rat conjunctiva. A single band of 66 kDa was detected in mouse and rat conjunctiva using an antibody againstα _1A-adrenergic receptors. A single major band of 64 kDa was detected in mouse and rat conjunctiva using an antibody toβ ₁-adrenergic receptors, whereas a single major band of 64 kDa was detected in mouse and rat conjunctiva using an antibody against β₂-adrenergic receptors. Two bands of 67 and 66 kDa were detected in mouse conjunctiva using an antibody againstβ ₃-adrenergic receptors.

Figure 8.

 

Immunolocalization of α_1A-adrenergic receptors on mouse and human conjunctival goblet cells. Immunoreactivity forα _1A-adrenergic receptor was only detected diffusely in the epithelial cell layer and not on the goblet cells in mouse conjunctiva (A). Immunoreactivity forα _1A-adrenergic receptors was detected on stratified squamous epithelial cells and in goblet cells in human conjunctiva (C). The position of goblet cells within the epithelium (B and D) was identified by using HPA, which preferentially stains goblet cell secretory granules in mouse and human conjunctiva. These pictures are representative of at least three different experiments. Magnification, ×270.

Figure 9.

 

Immunolocalization of β₁- and β₂-adrenergic receptors on mouse conjunctival goblet cells. Immunoreactivity forβ ₁-adrenergic receptors was detected on stratified squamous epithelial cells as well as on goblet cells in mouse conjunctiva (A). Immunoreactivity toβ ₂-adrenergic receptors was detected predominantly on the basal and lateral membranes of goblet cells in the mouse (C), but also in the stratified squamous cells. The position of goblet cells within the epithelium (B and D) was identified by using HPA, which preferentially stains goblet cell secretory granules in mouse conjunctiva. These pictures are representative of at least three different experiments. Magnification,× 270.

Figure 10.

 

Immunolocalization of β₁- and β₃-adrenergic receptors on human conjunctival goblet cells. Immunoreactivity forβ ₁-adrenergic receptor was only detected in the basal epithelial cell layer and not on the goblet cells in human conjunctiva (A). Immunoreactivity forβ ₃-adrenergic receptors was detected on stratified squamous epithelial cells as well as on goblet cells in human conjunctiva (B). The position of goblet cells within the epithelium (C) was identified by using HPA, which preferentially stains goblet cell secretory granules in human conjunctiva. These pictures are representative of at least three different experiments. Magnification, ×270.

The authors thank Patricia Pearson for her excellent technical assistance and Marcia Jumblatt, Peter Rubin, and Gabriel Garza for kindly providing the human conjunctival tissue.