Effect of Overexpression of Constitutively Active PKCα on Rat Lacrimal Gland Protein Secretion

Abstract

purpose. The lacrimal gland secretes water, electrolytes, and protein into the tear film. Decreased secretion from the lacrimal gland can lead to dry eye syndromes with deleterious effects on vision. Protein kinase C (PKC)-α plays a major role in cholinergic- and α₁-adrenergic–induced protein secretion from the lacrimal gland. This study was undertaken to determine whether activation of PKCα alone would induce lacrimal gland protein secretion by examining the effects of overexpression of constitutively active PKCα.

methods. Rat lacrimal gland acini were transduced with an adenovirus containing a gene for constitutively active PKCα. Protein secretion was measured in response to cholinergic and α₁-adrenergic agonist stimulation.

results. More than 84% of acinar cells were transduced, and PKCα expression was increased 176-fold. Western blot analysis using an antibody to phosphorylated (activated) PKCα indicated that the overexpressed PKCα was active, and basal secretion was increased. Cholinergic agonist–stimulated protein secretion was not stimulated above basal secretion, whereas α₁-adrenergic-agonist–stimulated protein secretion was increased in transduced acini.

conclusions. Basal lacrimal gland protein secretion can be stimulated by bypassing the release of neurotransmitters and activating PKCα, possibly leading to the development of new treatments for dry eye syndromes.

The primary function of tears is to protect the cornea and conjunctiva, which together compose the ocular surface. Tears are secreted in response to environmental stresses and protect the ocular surface by providing nourishment to the avascular cornea. ¹ The lacrimal gland is a tubuloacinar exocrine gland that functions to synthesize, store, and secrete proteins and electrolytes into the tear film, leading to the formation of the aqueous layer of tears. ¹ If this secretion is altered in either amount or composition, a spectrum of diseases called dry eye syndromes results. ¹ In these types of diseases, symptoms of itchiness, irritation, and feelings of dryness occur. In severe dry eye, vision-threatening conditions such as corneal ulcers and loss of corneal transparency can occur. ¹ Dry eye syndromes are a leading cause of ocular difficulties, with more than 3.2 million women in the United States alone diagnosed with these disorders. ²  

The lacrimal gland is composed of three main cell types: the acinar cells which are the main secretory cells; the ductal epithelial cells which line the ducts and modify the fluid by secreting water and electrolytes; and myoepithelial cells, which surround the acini with long processes. ¹ The lacrimal gland is highly innervated with parasympathetic and sympathetic nerves, and the neurotransmitters released from these nerves are potent stimuli of protein secretion. ^{3 4 5 6} Cholinergic agonists released from parasympathetic nerves bind to M₃ muscarinic receptors. ⁷ These receptors are G-protein–coupled receptors coupled to phospholipase Cβ (PLCβ). ^{7 8} Activated PLC hydrolyzes phosphatidylinositol 4,5-bisphosphate to generate inositol 1,4,5-trisphosphate (1,4,5-IP₃) and diacylglycerol (DAG). ^{9 10} 1,4,5-IP₃ causes the release of Ca²⁺ from intracellular Ca²⁺ [Ca²⁺]_i stores that can stimulate secretion, either on its own or through enzymes such as Ca²⁺/calmodulin-dependent protein kinases and protein kinase C (PKC). ^{11 12 13 14} DAG is necessary for the activation of PKC. α₁-Adrenergic agonists released from sympathetic nerves also stimulate lacrimal gland protein secretion, though the signaling pathway is not as well characterized as the pathway used by cholinergic agonists. ^{15 16} The specific types of α₁-adrenergic receptors present in the lacrimal gland are unknown, as is the primary G protein activated used by these agonists. The phospholipase activated by the G-proteins is also unknown. It is known that α₁-adrenergic agonists cause release of [Ca²⁺]_i, though to a much lesser extent than cholinergic agonists. ¹⁶ α₁-Adrenergic agonists are also known to activate PKC. ^{15 16}  

Our laboratory has shown that PKC plays a major role in both cholinergic- and α₁-adrenergic–stimulated lacrimal gland protein secretion. ¹⁵ However, as PKC is a family of at least 10 different isozymes, its role is somewhat complicated. The family of PKCs are divided into three groups. ^{17 18} The classical PKC isoforms, PKC-α, -βI, -βII, and -γ, are calcium and phospholipid dependent. The novel PKC isoforms, PKC-δ, -ε, -η, and -θ, are calcium independent and phospholipid dependent. The final group is the atypical PKC isoforms consisting of PKC-λ and -ζ. These PKC isoforms are calcium and phospholipid independent. Translocation of PKC isoforms to a membrane structure is an indication of enzyme activation. Anchoring proteins called receptors for activated C-kinase (RACKS) have been shown to provide isoform specificity by binding the activated isoforms and recruiting them to the appropriate substrates. ¹⁹  

The lacrimal gland contains PKC-α, -δ, -ε, and -λ. ²⁰ Using selective isoform inhibitors for PKCα, -δ, and -ε, we found that these three isoforms play roles in cholinergic-agonist–stimulated secretion, with PKCα playing the major role. ¹⁵ In contrast, activation of PKCα and δ inhibits α₁-adrenergic-agonist–stimulated secretion, whereas activation of PKCε stimulates secretion. ¹⁵  

In addition to having a direct role in protein secretion, PKC has been shown to alter cellular calcium handling in the lacrimal gland. Activation of PKC with phorbol esters decreases the production of 1,4,5-IP₃ and the Ca²⁺ response induced by cholinergic agonists. ^{21 22 23 24} Using specific PKC isoform inhibitors, we found that the Ca²⁺-independent PKC isoform PKCδ, and to a lesser extent PKCε, reverses the effect of phorbol esters on the Ca²⁺ response. ²² In contrast the Ca²⁺-dependent PKC isoform PKCα had no effect. ²² We hypothesized that activation of PKCδ, and to a lesser extent PKCε but not PKCα, blocks the Ca²⁺ entry process stimulated by cholinergic agonists. ²²  

With the development of adenoviral vectors and more efficient transfection procedures in combination with molecular biology techniques, the role of a particular signaling molecule can be more thoroughly studied. Constitutively active and dominant negative mutants of many signaling molecules have been generated and overexpressed in target cells. ^{25 26 27 28 29} The effects of these proteins were then measured on various cellular functions. In particular, constitutively active or dominant negative PKC isoforms have been used to provide additional information regarding its downstream effects. ^{30 31} As PKC inhibitors can be nonspecific or difficult to use, these techniques can provide a valuable tool for studying the role of specific isoforms. As the lacrimal gland contains multiple isoforms of PKC, we wanted to examine further the role of a specific isoform, namely PKCα, in Ca²⁺ handling and protein secretion. Thus, we overexpressed a constitutively active form of PKCα, by using an adenoviral vector, and investigated its effects on protein secretion and [Ca²⁺]_i handling. 

Materials and Methods

Collagenase type CLS III was obtained from Worthington Biochemical (Freehold, NJ), polyclonal antibody to PKCα from Santa Cruz Biotechnology (Santa Cruz, CA), and a polyclonal antibody to phosphorylated PKCα from Cell Signaling Technology (Beverly, MA). A peroxidase detection agent (Amplex Red) was purchased from Molecular Probes (Eugene, OR). RPMI-1640 culture medium, l-glutamine, and penicillin/streptomycin were obtained from BioWhittaker (Walkersville, MD). Phorbol 12-myristate 13-acetate (PMA) was from L. C. Services (Waltham, MA). Other chemicals were obtained from Sigma-Aldrich (St. Louis, MO). All reagents were of the highest purity available. 

Generation of Myristoylated PKCα Construct Adenoviruses

A myristoylated PKCα construct was synthesized as previously described. ³¹ In brief, an α-Flag epitope (DYKDDDDK) was added to the carboxyl terminus of the PKCα cDNA, and an Src myristoylation site (MYPYDVPDYA) was added to the amino terminus. The myristoylation moiety targets PKCα to cell membranes, resulting in kinase activity in vitro that has been shown to be higher than in wild-type PKCα. ³¹ Recombinant-deficient adenoviruses containing myr-PKCα (AdV-myrPKCα) and green fluorescent protein (AdV-GFP) were generated and purified at the Harvard Gene Therapy Center (Boston, MA). 

Preparation of Rat Lacrimal Gland Acini and Infection with Adenovirus

All 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. Both exorbital lacrimal glands were removed from male Sprague-Dawley rats that had been anesthetized with CO₂ and then decapitated. Lacrimal glands were trimmed of fatty and connective tissue and fragmented into small pieces. The pieces were washed at 37°C in Krebs-Ringer bicarbonate buffer (119 mM NaCl, 4.8 mM KCl, 1 mM CaCl₂, 1.2 mM KH₂PO₄, and 25 mM NaHCO₃) supplemented with 10 mM HEPES, 5.5 mM glucose, and 0.5% BSA (pH 7.4; KRB-BSA). Lacrimal gland acini were prepared by incubating tissue pieces with collagenase CLS III, (100 U/mL) in KRB-BSA for 30 minutes at 37°C under a stream of 95% O₂-5% CO₂. The preparation was then filtered through nylon mesh (150 μm) and the acini collected with 3-minute centrifugation at 50g. The dispersed acini were allowed to recover for 45 minutes in fresh KRB-BSA. Lacrimal gland acini were pelleted; resuspended in RPMI 1640 medium supplemented with 0.5% BSA, 2 mM l-glutamine and 100 μg/mL penicillin-streptomycin; and incubated (150 μg protein/well) at 37°C in 95% O₂-5% CO₂. The viruses AdV-GFP and AdV-myrPKCα were added at the concentrations indicated to the acini for 18 hours. It is important to note that acini are clumps in the figure legends composed of various numbers of cells. Thus, it is impossible to determine the exact number of cells in a particular preparation. Therefore, the multiplicities of infection (MOIs) used in this study were described as plaque-forming units per micrograms of protein. 

Measurement of Protein Secretion

Lacrimal gland acini were incubated in the presence or absence of adenoviruses for 18 hours and removed from cell culture plates. Acini were washed with KRB-BSA, allowed to recover for 45 minutes in fresh KRB-BSA, and incubated for 20 minutes with either the cholinergic agonist carbachol or the α₁-adrenergic agonist phenylephrine in KRB-BSA. To terminate incubation, acini were centrifuged at 50g and placed on ice. The supernatant was removed and the acini (pellet fraction) were disrupted by sonication in 50 mM Tris-HCl (pH 8.0). Peroxidase secretion, an index of protein secretion in lacrimal gland was measured. Peroxidase activity was measured in both the supernatant and the pellet fraction using a peroxidase detection agent (Amplex Red; Molecular Probes). Peroxidase oxidizes the agent in the presence of hydrogen peroxide to produce resorufin, a highly fluorescent molecule. For the measurement of peroxidase, supernatant and acini homogenate were spotted in duplicate onto 96-well microplates. Assay buffer (50 mM Tris-HCl, pH 7.5) containing 0.2 M peroxidase detector and 0.2 M hydrogen peroxide was added to each well. The fluorescence was determined in a fluorescence microplate reader (model FL600; Bio-Tek, Winooski, VT) with 530-nm excitation wavelength and 590-nm emission wavelength. The amount of secreted peroxidase was expressed as a percentage of the total: (peroxidase in medium/peroxidase in medium + peroxidase in tissue) × 100. 

Western Blot Analysis

Lacrimal gland acini were incubated in the presence or absence of adenoviruses for 18 hours and removed from cell culture plates. The acini were homogenized in ice-cold RIPA buffer containing proteinase inhibitors (10 mM Tris-HCl [pH 7.4], 150 mM NaCl, 1% Triton X-100, 1% sodium deoxycholic acid, 0.1% SDS, 1 mM EDTA, 10 mg/mL phenylmethylsulfonyl fluoride, 5 U/mL aprotinin, and 100 nM sodium orthovanadate) and proteins were separated by SDS-PAGE (10% acrylamide gel). The proteins were then transferred to nitrocellulose membranes. For the anti-PKCα antibody, nonspecific sites were blocked by incubating membranes overnight at 4°C in 5% dried milk in buffer containing 10 mM Tris-HCl (pH 8.0), 150 mM NaCl, and 0.05% Tween-20 (TBST), followed by incubation for 1 hour at room temperature with the primary antibody to PKCα (1:2000). For anti-phosphorylated PKCα antibody, the membranes were blocked for 1 hour at room temperature in 5% dried milk in TBST followed by overnight incubation with primary antibody (1:1000) in 5% BSA in TBST. After primary antibody incubation, all membranes were washed in TBST and incubated with the secondary antibody conjugated to horseradish peroxidase (1:2000). Immunoreactive bands were visualized by the enhanced chemiluminescence method. The films were scanned and analyzed with National Institutes of Health Image software (available by ftp at zippy.nimh.nih.gov/ or at http://rsb.info.nih.gov/nih-image; developed by Wayne Rasband, National Institutes of Health, Bethesda, MD). 

Flow Cytometry and Confocal Immunofluorescence Microscopy

The percentage of lacrimal gland acinar cells expressing GFP was determined using the FITC channel. In some experiments, lacrimal gland acinar cells were fixed with PBS containing 0.5% BSA and 4% paraformaldehyde for 20 minutes and incubated with the primary antibodies (anti-flagM2 or anti-PKCα; 1:50) in PBS containing 0.5% BSA and 0.1% saponin for 20 minutes. Cells were rinsed three times with PBS and 0.5% BSA and incubated with FITC-conjugated secondary antibodies (1:50) for 20 minutes. Cells were rinsed three times and analyzed by flow cytometry with a cell counter (Epics XL Analyzer; Beckman Coulter Inc., Hialeah, FL). 

For confocal microscopy, lacrimal gland acini were incubated with and without the cholinergic agonist carbachol (10⁻⁴ M) for 5 minutes after transduction with 3 × 10⁷ pfu/150 μg protein. The cells were cytospun onto gelatin-coated slides by a cytospin centrifuge (Shandon, Pittsburgh, PA) running at 1000 rpm for 4 minutes. The slides were air dried and fixed in 4% formaldehyde in 0.1 M sodium phosphate buffer (pH 7.4) with 15 mM CaCl₂ for 20 minutes and preserved in 30% sucrose in sodium phosphate buffer with 15 mM CaCl₂ for 20 minutes. The slides were air dried and stored at −20°C. Before addition of antibodies, acini were permeabilized with 1% Triton X-100 diluted in PBS (containing 145 mM NaCl, 7.3 mM Na₂HPO₄, and 2.7 mM NaH₂PO₄) for 10 minutes at room temperature. The antibody against PKCα was used at a 1:1000 dilution for 1 hour at room temperature. The secondary antibody, conjugated to either FITC or rhodamine was used at 1:200 for 1 hour at room temperature. The slides were viewed with a confocal scanning microscope (TCS4D; Leica Microsystems, Bannockburn, IL) equipped with a krypton-argon laser. 

Measurement of [Ca²⁺]_i

After overnight incubation with 3 × 10⁷ pfu/mL of either AdV-GFP or AdV-myrPKCα, acini were allowed to recover for 30 minutes at 37°C in KRB-BSA buffer. Acini were then incubated in KRB-BSA buffer containing 0.5% BSA, 0.5 μM fura-2 tetraacetoxymethyl ester, 10% pluronic F127, and 250 μM sulfinpyrazone for 60 minutes at 22°C. The cells were then washed with KRB-BSA buffer containing 250 μM sulfinpyrazone, and fluorescence was measured at 22°C. Fluorescence was measured at excitation wavelengths of 340 and 380 nm and an emission wavelength of 505 nm using a luminescence spectrofluorometer (model LS-5B; Perkin Elmer, Wellesley, MA). To calculate [Ca²⁺]_i, 5.6 mM EGTA, 7.5 mM Tris-HCl (pH 7.5), and 1% Triton X-100 were added at the end of the reaction to obtain the minimum fluorescence level. Maximum fluorescence was determined by the addition of 14.5 mM CaCl₂. The dissociation constant of 135 nM for fura-2 at 22°C was used to calculate [Ca²⁺]_i by the ratio method. ³²  

Data Presentation and Statistical Analysis

Data are expressed as the mean ± SEM. When appropriate, data were statistically analyzed with Student’s t-test for unpaired values. P < 0.05 was considered to be significant. 

Results

Transduction of Rat Lacrimal Gland Acini by Adenovirus AdV-GFP

To determine the conditions necessary for optimal adenoviral transduction, lacrimal gland acini were incubated with various concentrations of AdV-GFP for various times. The GFP expressed within the acinar cells was viewed by fluorescence microscopy, using the FITC wavelength. Fluorescence microscopy showed that lacrimal acini were maximally transduced after 18 hours by AdV-GFP at 3 × 10⁷ pfu/150 μg protein and that GFP was present in almost every cell throughout the acinus (Fig. 1A) . Because it was impossible to determine the number of cells in an acini preparation, MOIs are expressed as plaque-forming units (pfu) per 150 micrograms of protein. When analyzed by flow cytometry, 88% ± 5% of lacrimal acinar cells expressed GFP when transduced by AdV-GFP under these conditions (Fig. 1B) . 

These results show that freshly isolated lacrimal gland acini can be transduced using an adenovirus containing the gene for GFP, resulting in expression of the GFP protein. 

Overexpression of PKCα in AdV-myrPKCα–Transduced Lacrimal Gland Acini

The myrPKCα construct contains an α-FLAG epitope on the carboxyl terminus of the PKCα gene. Thus, using the same conditions of transfection for AdV-GFP, we used confocal microscopy to view α-FLAG in transduced acini and flow cytometric analysis to determine the number of cells expressing α-FLAG and thus PKCα. After incubation for 18 hours with AdV-myrPKCα (3 × 10⁷ pfu/150 μg cell protein), acini were cytospun onto slides and incubated with an anti-α-FLAG antibody. Confocal microscopy showed that the α-FLAG epitope was located in the plasma membranes and outlined the individual cells within the acinus (Fig. 2B) . In addition, flow cytometric analysis determined that 84.0% ± 0.2% of acinar cells expressed the α-Flag epitope (Fig. 2B) . This epitope was absent in noninfected acini (Fig. 2A) . 

To confirm these results, rat lacrimal gland acini were transduced with AdV-myrPKCα for 18 hours. Expression of myrPKCα was then determined in cell homogenates by Western blot techniques with an anti-PKCα antibody. Figure 3A shows the expression of PKCα in nontransduced acini and in acini transduced by increasing concentrations of AdV-myrPKCα (3 × 10⁵ to 10⁸ pfu/150 μg cell protein). When the blots were analyzed by densitometry (Fig. 3B) , PKCα expression was 176 times higher in cells transduced with 3 × 10⁷ pfu/150 μg cell protein than in nontransduced acini (n = 3). Overexpression of PKCα was dependent on the MOI, with the highest expression corresponding to the highest concentration of AdV-myrPKCα. 

These results show that freshly isolated lacrimal gland acini can be efficiently transduced using an adenovirus containing the gene for constitutively active PKCα, resulting in expression of the PKCα protein. 

Localization of myrPKCα

The myristoylation moiety on the amino terminus of the PKCα construct should target PKCα to cell membranes, thereby activating it. Confocal microscopy showed that the α-FLAG epitope appeared to be located in the plasma membrane. To verify the location of the myrPKCα expressed by transduced lacrimal gland acini, acini were cytospun onto slides and incubated with an anti-PKCα antibody. The localization of PKCα was then determined by confocal microscopy. In unstimulated, nontransduced acini, PKCα had diffuse, cytosolic staining (Fig. 4A) . As expected, in nontransduced acini, PKCα localization changed on stimulation with the cholinergic agonist carbachol (10⁻⁴ M) and appeared to localize to the basolateral membranes (Fig. 4B) . In cells transduced with AdV-myrPKCα, but not stimulated, and viewed by confocal microscopy, PKCα was located in the plasma membranes (Fig. 4C) . This location is consistent with the localization seen using the antibody against α-FLAG epitope. 

These results indicate that endogenous PKCα, in nontransduced acini, translocates to the plasma membrane on stimulation with cholinergic agonists. However, constitutively active myrPKCα in transduced acini is already located in the plasma membrane of freshly isolated lacrimal gland acini. 

PKC Activity in AdV-myrPKCα–Transduced Lacrimal Gland Acini

It has been reported that myristoylated PKCα has increased kinase activity in vitro relative to the wild-type enzyme. ¹⁹ To ensure that the PKCα expressed in myrPKCα-transduced lacrimal gland acini was constitutively active, PKC activity was compared in nontransduced and transduced acini. Acini were transduced with 3 × 10⁷ pfu/150 μg protein overnight before stimulation with the phorbol ester, PMA, a known activator of PKC, at 10⁻⁶ M for 10 minutes. The reaction was terminated by centrifugation, acini were homogenized in RIPA buffer, and proteins were separated by SDS-PAGE. Because PKC activation is regulated by phosphorylation at three sites, Thr 500 in the activation loop, Thr 638 in the autophosphorylation site, and Ser 660 in the hydrophobic site of the carboxyl terminus, ³³ we used an antibody against PKCα phosphorylated at Thr 638 in Western blot analysis. As shown in Figure 5A , activated PKCα was not detected in nontransduced acini until stimulated with PMA. In contrast, there was a substantial amount of activated PKCα in transduced acini. There was no further increase in the amount of activated PKCα in transduced acini after stimulation with PMA. The same samples were also analyzed using an antibody to total PKCα (Fig. 5B) to ensure that transduction had occurred. 

These results indicate that PKCα in transduced lacrimal gland acini is constitutively active. 

Effect of Overexpression of myrPKCα on Stimulated Protein Secretion

To determine whether overexpression of activated PKCα has an effect on protein secretion stimulated by neurotransmitters, lacrimal gland acini were transduced with either AdV-GFP or AdV-myrPKCα (3 × 10⁷ pfu/150 μg protein) before stimulation with the cholinergic agonist carbachol (10⁻³ M) or the α₁-adrenergic agonist phenylephrine (10⁻³ M). Peroxidase, our index of protein secretion, was measured. As shown in Figure 6 , basal secretion in nontransduced acini was 0.28% ± 0.08% of total peroxidase (n = 6) and was significantly increased by stimulation with carbachol (0.75% ± 0.22% of total peroxidase) and phenylephrine (1.20% ± 0.28% of total peroxidase). Basal secretion in acini transduced with AdV-GFP decreased slightly, though not significantly, from basal secretion in nontransduced acini (0.17% ± 0.01% of total peroxidase, n = 6). Carbachol and phenylephrine again significantly increased peroxidase secretion over their basal in acini transduced with AdV-GFP (0.36% ± 0.02% and to 1.22% ± 0.04% of total peroxidase, respectively, Fig. 6 ). Basal secretion was significantly increased in acini transduced with AdV-myrPKCα to 0.89% ± 0.04% of total peroxidase compared with basal secretion in nontransduced acini (n = 6). In contrast to nontransduced and AdV-GFP–transduced acini, carbachol stimulation did not further increase peroxidase secretion in AdV-myrPKCα transduced acini (0.69% ± 0.02% of total peroxidase, Fig. 6 ) above its basal. There was no significant difference in carbachol-stimulated peroxidase secretion between nontransduced acini and acini transduced with either AdV-GFP or AdV-myrPKCα. Phenylephrine, however, significantly increased peroxidase secretion to 1.94% ± 0.11% of total peroxidase in acini transduced with AdV-myrPKCα over its basal secretion. Phenylephrine-induced peroxidase secretion was not significantly different between nontransduced acini and acini transduced with either AdV-myrPKCα or AdV-GFP. 

These results imply that PKCα plays a role in basal and carbachol-induced, but not phenylephrine-induced, peroxidase secretion from freshly isolated lacrimal gland acini. 

Effect of Overexpression of myrPKCα on Basal Protein Secretion

As indicated in Figure 6 , overexpession of myr-PKCα caused a statistically significant increase in basal secretion (i.e., secretion in the absence of exogenous stimuli). To investigate this response further, acini were transduced with increasing concentrations of AdV-myrPKCα overnight, and basal peroxidase secretion was measured. Basal secretion rose with the increasing concentrations of AdV-myrPKCα used for transfection (Fig. 7) . Statistically significant increases in peroxidase secretion over nontransduced acini were obtained with concentrations of 10⁶, 3 × 10⁶, 3 × 10⁷, and 10⁸ pfu/μg protein. 

These results indicate that increasing the expression PKCα increases basal peroxidase secretion and are consistent with the presence of active PKCα. 

Effect of Overexpression of myrPKCα on [Ca²⁺]_i Concentration

To determine whether the increase in basal peroxidase secretion was a result of altered [Ca²⁺]_i handling, the [Ca²⁺]_i was measured with the calcium indicator fura-2 in acini transduced with 3 × 10⁷ pfu/μg protein of either AdV-GFP or AdV-myrPKCα. Basal [Ca²⁺]_i in nontransduced acini was 20.9 ± 2.3 nM (n = 4). This level was not altered in acini tranduced with AdV-GFP (19.0 ± 0.4 nM, n = 4) or AdV-myrPKCα (16.8 ± 1.9 nM, n = 4). To ensure that changes in [Ca²⁺]_i could be detected in these cells, the α₁-adrenergic agonist phenylephrine, which is known to increase [Ca²⁺]_i in lacrimal gland acini, was added to nontransduced and transduced acini. ¹⁶ Phenylephrine (10⁻⁴ M) elicited an increase in [Ca²⁺]_i by 12.0 ± 7.1, 18.6 ± 7.5, and 14.9 ± 4.3 nM over basal levels in nontransduced acini and acini transduced with either 3 × 10⁷ pfu/μg protein AdV-GFP or AdV-myrPKCα, respectively. 

Thus, these results indicate that altered Ca²⁺ handling does not play a role in increased basal protein secretion. 

Discussion

When the lacrimal gland ceases to secrete proteins and electrolytes in the appropriate quantity, dry eye syndromes result. The causes of lacrimal gland dysfunction vary and can range from normal aging to destruction of the gland. Indeed, the autoimmune disease Sjögren’s syndrome is characterized by lymphocytic infiltration of the lacrimal gland and a subsequent decrease in protein secretion. We have shown in a mouse model of Sjögren’s syndrome that the innervation in lacrimal glands is unchanged with the onset and progression of the disease. ³⁴ However, the release of neurotransmitters by nerves surrounding lacrimal gland acini was inhibited, resulting in a denervation-like supersensitivity. ³⁵ By identifying the location of the defect in Sjögren’s syndrome or other dry eye syndromes, it should then be possible to design therapies that bypass the obstruction and restore lacrimal gland secretion. Thus, expression of a constitutively active form of PKCα could circumvent the need for released neurotransmitters and activation of their receptors. 

In this study, we show that a constitutively active form of PKCα can be overexpressed through an adenoviral vector in lacrimal gland acini. This transduction is very efficient, as PKCα can be overexpressed in more than 85% of lacrimal gland acini. This results in a 176-fold increase in expression of constitutively active PKCα over nontransduced cells. This expression increases basal secretion while not affecting [Ca²⁺]_i. 

In general, PKC isoforms have cell- and tissue-specific localizations. The cellular location of PKC has been shown to be critically important to the regulation and function of the enzyme in a variety of cell types. ^{19 36 37} The lacrimal gland is no exception. Of the four PKC isoforms present in the lacrimal gland, each has a specific cellular location. ²⁰ Of interest to this study, PKCα was located in lacrimal gland sections on the basolateral and apical membranes of the acini, although Western blot analysis indicated that a substantial amount of PKCα was located in the cytosolic fraction of unstimulated acini. PKCα was translocated to the membrane fraction on stimulation, but the location was not determined. ²⁰ In this study, confocal microscopy confirmed that PKCα was located in the cytosol of acini and was translocated to the basolateral membranes on stimulation with the cholinergic agonist carbachol. This location is similar to that seen with the overexpressed constitutively active PKCα. Thus, constitutively active PKCα appears to be located in the basolateral membranes of acinar cells, the proper location for exerting its functional effects. 

We have shown that PKCα plays a major role, both positive and negative, in cholinergic and α₁-adrenergic agonist-induced protein secretion. ¹⁵ Thus, it is surprising that the effect of overexpression of PKCα was to increase basal secretion (i.e., absence of stimuli). The increase was dependent on the concentration of adenovirus used and therefore the amount of PKCα expressed, implying that PKCα plays a major role in producing basal protein secretion. Overexpression of wild-type PKCε has also been shown to increase basal secretion of prolactin from rat pituitary cells. ³⁸ Akita et al. ³⁸ determined that the amount of PKCε associated with membrane fractions was higher in cells overexpressing PKCε than in control cells, resulting in increased enzyme activity and an increase in basal secretion. They suggest that the amount of PKCε in these cells is rate limiting in the basal secretory pathway, and thus more PKCε leads to more secretion. In the lacrimal gland, the amount of PKCα could also be rate limiting in basal secretion, with overexpression of PKCα resulting in increased basal secretion. This could be advantageous when designing a treatment to stimulate tear secretion in patients with dry eye syndromes, as the number of compounds needed to be introduced for effective treatment would be low. 

Although cholinergic and α₁-adrenergic agonists activate the same PKC isoforms, namely PKCα, -δ, and -ε, the effects of activation are quite different. Cholinergic agonists translocate these isoforms from cytosolic to membrane fractions; α₁-adrenergic agonists do not. ¹⁶ Activation of PKCα, -δ, and -ε, as measured using pseudosubstrate inhibitor peptides, stimulate cholinergic agonist-induced protein secretion, whereas activation of PKCα and -δ inhibits α₁-adrenergic agonist–stimulated protein secretion, and PKCε stimulates it. PKC isoforms are known to interact with several binding proteins that are responsible for targeting PKC to the appropriate location, bringing the enzyme in close proximity to its substrates and integrating the signal with other signaling pathways. ³⁶ We hypothesized that different agonists are distinctly coupled to the PKC isoforms to account for the differing effects. This is further supported by the present study, in which stimulation of protein secretion by carbachol in acini overexpressing PKCα was not increased, whereas phenylephrine stimulated secretion in these cells. 

The differences seen between the effects of cholinergic and α₁-adrenergic agonists are striking. Cholinergic agonists do not stimulate secretion over basal levels in acini overexpressing PKCα, whereas secretion is stimulated by α₁-adrenergic agonists. This suggests that cholinergic agonists activate PKCα to stimulate secretion and, because PKCα has already been activated, they cannot further activate PKCα. Thus, cholinergic agonists cannot increase secretion over the already elevated levels. It is possible that cholinergic agonists use a different signaling pathway than that used to maintain basal secretion. In this case, instead of the amount of PKCα being rate limiting, the amount of another protein downstream of PKCα would be rate limiting. α₁-Adrenergic agonists can increase secretion over basal levels, suggesting that they do not use the same signaling pathway as activated by cholinergic agonists or the pathway that maintains basal secretion. α₁-Adrenergic agonist use PKCε to stimulate protein secretion and thus can activate this PKC isoform in cells in which PKCα is constitutively active. 

We have shown previously, with isoform-specific inhibitors, that PKCα is not involved in the Ca²⁺ response induced by cholinergic agonists, but the effects on the Ca²⁺ response stimulated by α₁-adrenergic agonists was not determined. ²² It is interesting that activation of PKCα is sufficient to increase basal protein secretion in the absence of an increase in [Ca²⁺]_i, as we have shown that cholinergic and α₁-adrenergic agonists increase [Ca²⁺]_i and this increase is necessary for secretion to occur. ^{11 16} It is possible that there was a localized, but undetected, increase of [Ca²⁺]_i in a microdomain that was sufficient to increase protein secretion. The results in this study confirm the hypothesis that additional PKC isoforms, other than PKCα, are involved in Ca²⁺ handling in the lacrimal gland. 

It is not clear whether the effects on basal peroxidase secretion in this study are a result of the overexpression of PKCα or of the fact that PKCα is constitutively active. Experiments with an adenovirus containing a wild-type PKCα did not increase basal protein secretion. However, the total amount of PKCα expressed in acini transduced with wild-type PKCα was significantly less than in acini transduced with myrPKCα. Indeed, we were unable to achieve a level of expression of wild-type PKCα that was comparable to the level achieved with myrPKCα. Transduction of acinar cells with 1 × 10⁸ pfu/150 μg protein resulted in only a threefold increase in PKCα expression, as measured by Western blot techniques (data not shown). It is unlikely, however, that the effects are a result only of overexpression of PKCα, as we showed that myrPKCα was found in the plasma membrane, the same location as endogenous PKCα when translocated by cholinergic agonists. Wild-type PKCα would not be in the appropriate location under basal conditions to exert any effects. 

It has been suggested that certain proteins in the adenoviral coat uncouple the secretory pathway stimulated by cholinergic agonists in lacrimal gland acini. ³⁹ This study shows that there is a reduced amount of secretory vesicles containing rab3D in cells treated with the penton protein, resulting in an increased basal protein secretion and decreased stimulated protein secretion. In contrast, Chen et al. ⁴⁰ did not detect changes in location of Rab3D and either basal or stimulated amylase secretion in pancreatic acini transduced with either control adenovirus or an adenovirus encoding Rab3D. In the present study, we also saw an increase in basal secretion. However, this effect occurred only when we used AdV-myrPKCα and not with AdV-GFP, indicating that the effect on secretion was not due to the adenovirus itself. The differences might be explained by the fact that the lacrimal gland acini used in the study by Wang et al. ⁴¹ were reconstituted by culturing acini from rabbit lacrimal glands in medium for several days, allowing them to re-form into a structure resembling an acinus. In addition, Wang et al. were measuring total protein secretion compared with regulated protein secretion, as measured in the present study. Finally, the duration of exposure to the adenovirus and the MOI used have been shown to alter the effects of the adenovirus itself. We are unable to determine cell number to calculate MOI in our preparations, making comparisons between these studies difficult. 

We observed that basal and cholinergic agonist-stimulated peroxidase secretion is decreased in acini transduced with AdV-GFP. Although this decrease is not statistically significant, it is notable. The decrease could be due to the adenoviral vector itself, although this effect was not seen when acini were stimulated with α₁-adrenergic agonists. It could also be due to the expression of the GFP protein. This also seems unlikely as again, the effect did not occur when acini were stimulated with α₁-adrenergic agonists. 

In conclusion, overexpression of a constitutively active form of PKCα increases basal protein secretion, but does not alter Ca²⁺ handling, in the lacrimal gland. Thus, in the lacrimal gland, secretion can be stimulated by circumventing the release of neurotransmitters and activation of their receptors, possibly leading to new therapies for the treatment of dry eye syndromes. 

 Supported by National Eye Institute Grant EY06177.

 Submitted for publication May 6, 2004; revised June 9, 2004; accepted June 17, 2004.

 Disclosure: R.R. Hodges, None; I. Raddassi, None; D. Zoukhri, None; A. Toker, None; A. Kazlauskas, None; D.A. Dartt, None

 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.

 

Expression of GFP in lacrimal gland acini transduced with adenovirus containing GFP. Lacrimal gland acini were incubated with AdV-GFP (3 × 10⁷ pfu/150 μg protein) for 18 hours. Acini were either placed on slides and viewed by fluorescence microscopy (A) or analyzed by flow cytometry (B) to determine the number of cells expressing GFP, using the FITC channel. Results are representative of two independent experiments. Magnification, ×200.

Figure 2.

 

Localization and expression of α-FLAG in lacrimal gland acini transduced with adenovirus containing constitutively active myrPKCα. Lacrimal gland acini were incubated either without (A) or with (B) 3 × 10⁷ pfu/150 μg protein AdV-myrPKCα for 18 hours and viewed by confocal microscopy or analyzed by flow cytometry, using an anti-α-FLAG antibody. Results are representative of two independent experiments.

Figure 3.

 

Expression of PKCα in lacrimal gland acini transduced with adenovirus containing constitutively active myrPKCα. Lacrimal gland acini were incubated with increasing concentrations of AdV-myrPKCα for 18 hours. Expression of PKCα was detected by Western blot techniques, using an anti-PKCα antibody. (A) A representative blot showing increasing expression of PKCα (arrow) after incubation with increasing concentrations of AdV-myrPKCα. *Lane was loaded with 40 μg protein instead of 10 μg as in the other lanes. (B) Films were analyzed by densitometry. Data are the mean ± SEM of results in seven independent experiments.

Figure 4.

 

Localization of PKCα in AdV-myrPKCα lacrimal gland acini. Localization of PKCα was determined by confocal microscopy in (A) nontransduced, nonstimulated acini (basal); (B) nontransduced acini stimulated with the cholinergic agonist carbachol (Cch, 10⁻⁴ M) for 5 minutes; and (C) unstimulated acini (basal) transduced with 3 × 10⁷ pfu/150 μg protein AdV-myrPKCα for 18 hours.

Figure 5.

 

PKCα activity in AdV-myrPKCα–transduced lacrimal gland acini. Lacrimal gland acini were transduced with AdV-myrPKCα and then stimulated with the phorbol ester PMA (10⁻⁶ M) for 5 minutes. Acini were homogenized and analyzed by Western blot techniques using an antibody to phosphoPKCα (indicating activity, top) or total PKCα (bottom). C, control, no stimulation; P, PMA; NT, nontransduced; Trans, transduced.

Figure 6.

 

Effect of overexpression of constitutively active myrPKCα on lacrimal gland protein secretion. Acini were incubated with 3 × 10⁷ pfu/150 μg protein AdV-myrPKCα for 18 hours before stimulation with the cholinergic agonist carbachol (Cch, 10⁻³ M) or the α₁-adrenergic agonist phenylephrine (Ph, 10⁻³ M) for 20 minutes. Peroxidase secretion was then measured. Data are expressed as mean ± SEM from six independent experiments. *Significant difference from nontransduced basal acini; **significant difference from AdV-GFP transduced basal acini; #significant difference from AdV-PKCα basal acini.

Figure 7.

 

Effect of increasing amounts of overexpression of constitutively active myrPKCα on lacrimal gland basal protein secretion. Lacrimal gland acini were incubated with increasing concentrations of AdV-myrPKCα for 18 hours before measurement of basal secretion for 20 minutes. Data are expressed as the mean ± SEM from three independent experiments. *Significant difference from nontransduced acini.