Secretion polarity of interferon-beta in epithelial cell lines

Effective intraperitoneal gene transfection system using nanobubbles and ultrasound irradiation

In this study, we demonstrate the low toxicity and highly efficient and spatially improved transfection of plasmid DNA (pDNA) with liposomal nanobubbles (bubble liposomes [BLs]) using ultrasound (US) irradiation in mice. Naked pDNA with BLs was intraperitoneally injected, followed by US irradiation. The injection volume, the duration of US irradiation, and the dose of BLs were optimized. Both BLs and US irradiation were essential to achieve high transgene expression from naked pDNA. We observed transgene expression in the entire peritoneal tissues, including the peritoneal wall, liver, spleen, stomach and small and large intestines. The area of transfection could be controlled with focused US irradiation. There were few changes in the morphology of the peritoneum, the peritoneal function or serum alanine aminotransferase levels, suggesting the safety of BLs with US irradiation. Using a tissue-clearing method, the spatial distribution of transgene expression was evaluated. BLs with US irradiation delivered pDNA to the submesothelial layer in the peritoneal wall, whereas transgene expression was restricted to the surface layer in the liver and stomach. Therefore, BLs with US irradiation could be an effective and safe method of gene transfection to the peritoneum.

Diapause is associated with a change in the polarity of secretion of insulin-like peptides

The insulin/IGF-1 signalling (IIS) pathway plays an important role in the regulation of larval diapause, the long-lived growth arrest state called dauer arrest, in Caenorhabditis elegans. In this nematode, 40 insulin-like peptides (ILPs) have been identified as putative ligands of the IIS pathway; however, it remains unknown how ILPs modulate larval diapause. Here we show that the secretory polarity of INS-35 and INS-7, which suppress larval diapause, is changed in the intestinal epithelial cells at larval diapause. These ILPs are secreted from the intestine into the body cavity during larval stages. In contrast, they are secreted into the intestinal lumen and degraded during dauer arrest, only to be secreted into the body cavity again when the worms return to developmental growth. The process that determines the secretory polarity of INS-35 and INS-7, thus, has an important role in the modulation of larval diapause.

Insulin-like peptides INS-7 and INS-35 suppress larval diapause in Caenorhabditis elegans via unknown mechanism. Here, Matsunaga et al. show that the secretory polarity of both peptides changes in diapause, when these peptides are secreted into the intestinal lumen instead of the body cavity like in other larval stages.

Efficient in vivo gene transfer by intraperitoneal injection of plasmid DNA and calcium carbonate microflowers in mice

Gene transfer to intraperitoneal organs is thought to be a promising approach to treat such conditions as peritoneal fibrosis and peritoneal dissemination of cancers. We previously discovered that simple instillation of naked plasmid DNA (pDNA) onto intraperitoneal organs such as the liver and stomach could effectively transfer foreign genes in mice. In this study, we developed a novel nonviral method to enhance transfection efficiency of naked pDNA to intraperitoneal organs using a calcium carbonate suspension containing pDNA. Using commercially available calcium carbonate, we successfully transfected pDNA to the stomach. Handling of commercially available calcium carbonate, however, was troublesome owing to rapid precipitation and caking. To obtain slowly settling particles of calcium carbonate, we tried to synthesize novel versions of such particles and succeeded in creating flower-shaped particles, named calcium carbonate microflowers. Sedimentation of calcium carbonate microflowers was sufficiently slow for in vivo experiments. Moreover, the transfection efficiency of the suspension of calcium carbonate microflowers to the stomach was more effective than that of commercially available calcium carbonate, especially at low concentrations. Intraperitoneal injection of the suspension of calcium carbonate microflowers containing pDNA greatly enhanced naked pDNA transfer to whole intraperitoneal organs in mice. Furthermore, lactate dehydrogenase activities in intraperitoneal fluid and plasma were not raised by the suspension of calcium carbonate microflowers.

Effects of methyl-beta-cyclodextrin treatment on secretion profile of interferon-beta and zonula occuludin-1 architecture in Madin-Darby canine kidney cell monolayers

The interferon (IFN) is a paradigm of secretory protein. However, it has been poorly understood how its secretion is regulated in polarized epithelial cells. Recently, we had shown that exogenous IFNs transiently expressed in polarized monolayers were predominantly secreted to the side on which gene transfection had been performed, while stably expressed IFNs were secreted almost equally to the both cell sides. Since those modes of secretion did not affect each other, epithelial cell layers seemed to have at least two protein sorting/secretion pathways, one for transient expression and the other for stable expression, for identical secretory proteins. Furthermore, this dual secretion profile seemed to be mediated by distinct post-trans Golgi network vesicles, suggesting the involvement of lipid rafts in the sorting multiplicity. To address this issue, here we studied the effects of cholesterol depletion with methyl-beta-cyclodextrin (MbetaCD) on the secretion profile of IFN-beta exogenously expressed in Madin-Darby canine kidney (MDCK) cells. The MbetaCD-treatment, however, did not affect the profile in either transient or stable expression, although the architecture of zonula occuludin-1, which links to the tight junction, was substantially disrupted by the treatment. Further analysis of Triton X-100-insoluble cell extracts by sucrose density centrifugation demonstrated that IFN-beta was not apparently associated with lipid rafts in either transient or stable expression. These results suggest that lipid rafts may not be crucially involved in the regulation of secretion polarity of IFN-beta in the epithelial cells.

RCE1 Corneal Epithelial Cell Line: Its Variability on Phenotype Expression and Differential Response to Growth Factors

BACKGROUND

By serial transfer of rabbit corneal epithelial cells, the spontaneous RCE1 cell line was previously established. These cells mimic the stage-dependent differentiation of the corresponding cell type.

METHODS

RCE1 cells were cultured either on plastic culture dishes or on collagen rafts to compare the epithelial ultrastructure after growth on these substrata. Phenotypic variability was studied after subcloning of cells. The differentiation ability of each subclone was determined by Western blot with antibodies against the differentiation-linked keratin pair K3/K12 and by measuring LDH activity and LDH isozymes in cytosolic extracts. The proliferative response of RCE1 cells to EGF, TGFalpha, amphiregulin, bFGF or IL-6 was determined under serum-free culture conditions.

RESULTS

Cells grown on collagen rafts formed 5- to 7-layered epithelia with characteristics closer to those found in normal corneal epithelium than cells cultivated on plastic substrata, which formed 3- to 5-layered epithelia. Subcloning experiments demonstrated that every proliferative cell is able to grow and constitute stratified epithelia expressing K3/K12 keratins. LDH levels in RCE1 epithelia were similar to those of cultured or freshly harvested corneal epithelia; however, they showed a slightly altered LDH isozyme set, with prevalence of LDH-3 isoform. Whereas EGF and TGF-alpha were equipotent, amphiregulin elicited a 4-fold lower proliferative response. Also, bFGF was 10-fold less mitogenic than EGF, and IL-6 had the lowest effect with an ED(50) 20-fold lower than EGF.

CONCLUSIONS

The results demonstrate that every RCE1 proliferative cell has the ability to generate epithelial sheets. We conclude that EGF and TGF-alpha are the major effectors of RCE1 cell proliferation.

Cholesterol is required for the polarized secretion of erythropoietin in Madin-Darby canine kidney cells

It has already been reported that stably expressed exogenous human wild-type EPO (wtEPO) is preferentially secreted to the apical side and one of the three N-linked carbohydrate chains critically acts as an apical sorting determinant in Madin-Darby canine kidney (MDCK) cells. It has been suggested that lipid rafts are involved in the apical sorting of membrane and secretory proteins. To investigate the involvement of lipid rafts in the apical sorting of wtEPO, we examined the effect of cholesterol depletion with methyl-beta-cyclodextrin on the secretion polarity of EPO and analyzed Triton X-100 insoluble cell extracts by sucrose density gradients centrifugation in MDCK cells. We found that wtEPO was shifted in non-polarized direction by cholesterol depletion. Most of the wtEPO was not detectable in the raft fractions by sucrose density gradients centrifugation analysis. These results indicate that apical secretion of EPO involves a cholesterol-dependent mechanism probably not involving lipid rafts.

Exogenous expression of interferon-beta in cultured brain microvessel endothelial cells

Brain microvessel endothelial cells (BMECs) make up the blood-brain barrier (BBB) and regulate the passage of therapeutic proteins as well as drugs from the cerebrovasucular circulation to the brain. In the present study, we transferred mouse or human interferon-beta (IFN-beta) gene via cationic liposomes into primary cultures of bovine BMECs developed as an in vitro model of the BBB. The gene-transferred BMECs secreted transiently a substantial amount of IFN activity more efficiently during the growth phase than at confluence. This was suggested to be due to a difference in the potential for plasmid incorporation between growing and confluent BMECs in a series of cell association experiments with (32)P-labelled plasmid DNA. Furthermore, when BMEC monolayers in Transwell plates were transfected with the IFN-beta-expression vectors from the upper side, IFN-beta was predominantly detected in the upper compartments, suggesting polarized secretion of the transgene products in BMEC monolayers. These findings provide important basic information about therapeutic secretory protein gene delivery to BMECs.

Expression and visualization of a human interferon-beta-enhanced green fluorescent protein chimeric molecule in cultured cells

We have constructed a recombinant cDNA encoding the chimeric protein between human IFN-beta (HuIFN-beta) and enhanced green fluorescent protein (EGFP) to elucidate the intracellular localization of IFN-beta. Transient expression of the chimeric molecule, HuIFN-beta-EGFP, in L cells demonstrated that the chimeric molecule secreted from the cells had an intact biological activity as far as antiviral effect was concerned. Immunostaining of the transfected cells using anti-HuIFN-beta antibody demonstrated that green-fluorescence was co-localized with the IFN signal and its profile was similar to IFN signals in the cells transfected with HuIFN-beta expressing plasmid DNA. These results indicate that the HuIFN-beta-EGFP chimeric gene was expressed as a chimeric protein and the chimera was transported via the regular secretory pathway in the cells. In other cell types, the fluorescence derived from the chimeric protein was also seen on cytoplasmic vesicular structures. These results suggest that HuIFN-beta-EGFP will be a useful tool to investigate the intracellular trafficking processes of HuIFN-beta in a variety of cell types.

Permeability modulation of human intestinal Caco-2 cell monolayers by interferons

We investigated the effects of interferon-beta (IFN-beta) and IFN-gamma on the drug efflux activity of the human intestinal Caco-2 cell line, expressing the P-glycoprotein (P-gp) on the apical membrane. The cells grown on Transwell plates were pretreated with 1000U/ml IFN-beta, IFN-gamma or a combination of both for 3 days, and then the transepithelial electrical resistance (TEER) and the vectorial transport of rhodamine-123 (Rho-123) across the cell monolayers were evaluated. Exposure to IFN-gamma reduced substantially the TEER, but the effect of IFN-beta was minimal? The apparent permeability of Rho-123 in both the basolateral-to-apical and apical-to-basolateral directions was significantly increased by IFN-gamma but scarcely by IFN-beta. The combination of IFN-gamma and IFN-beta showed similar effects to IFN-gamma alone. Meanwhile, the cellular uptake of Rho-123 from the apical side was not affected by any IFN treatment. The uptake level was increased approximately three times in the presence of verapamil, a P-gp inhibitor, and the increased level was not affected by any IFN treatment, indicating that the efflux activity mediated by P-gp in the monolayers is not altered by these cytokines. Taken together, these results suggest that IFNs modulate the permeability of Caco-2 monolayer through effect on paracellular transport rather than effect on P-gp activity.

Simultaneous detection of DsRed2-tagged and EGFP-tagged human beta-interferons in the same single cells

The red fluorescent protein DsRed2 is a useful fusion tag for various proteins, together with the enhanced green fluorescent protein (EGFP). These chromoproteins have spectral properties that allow simultaneous distinctive detection of tagged proteins in the same single cells by dual color imaging. We used them for tagging a secretory protein, human interferon-beta (IFN-beta). Expression plasmids for human IFN-beta tagged with DsRed2 or with EGFP at the carboxyl terminal were constructed and their coexpression was examined in Mardin-Darby canine kidney epithelial cells. Although maturation of DsRed2 for coloration was slow and the color intensity was weak compared with EGFP, low temperature treatment (20 degrees C) allowed DsRed2-tagged human IFN-beta to be detected in the cells using color imaging. Consequently, the two chimeric proteins were shown to be colocalized in the same single cells by dual color confocal microscopy. This approach will be useful for investigating subcellular localization of not only cell resident proteins but also secretory proteins.

Secretion mode and subcellular localization of human interferon-beta exogenously expressed in porcine renal epithelial LLC-PK1 cells

We have investigated the secretion polarity of human interferon-beta (HuIFN-beta) exogenously expressed in the porcine renal proximal tubule cell line, LLC-PK1. In these cells, stably expressed HuIFN-beta was secreted to the apical and basolateral sides. However, when transiently expressed by apical lipofection, HuIFN-beta was secreted to both cell sides, while basolateral-preferential secretion was seen for basal transfection. Confocal imaging using enhanced green fluorescent protein (EGFP)-tagged HuIFN-beta revealed no difference in the subcellular distribution in either of the chimeric protein-expressing cells examined. These results suggest that the secretion polarity of HuIFN-beta is regulated by a post trans-Golgi network in a cell type-dependent manner.

Induction of cytotoxic T lymphocytes following immunization with cationized soluble antigen

Antigen presentation on major histocompatibility complex (MHC) class I and subsequent priming of antigen-specific cytotoxic T lymphocytes (CTLs) are essential steps for vaccination but exogenous soluble proteins are conventionally taken up by endosomes and presented on MHC class II rather than class I. In this study, we demonstrated, for the first time, that ovalbumin (OVA) chemically cationized with hexamethylenediamine (HMD) can induce OVA-specific CTLs without any adjuvants. Cationization of OVA greatly enhances cellular uptake by antigen-presenting cells (APCs) through adsorptive endocytosis. Two kinds of Cat-OVAs with different cationic charges were evaluated to elicit a CTL response through enhanced uptake by APCs and concomitant participation in the class I pathway. Cat(20)-OVA, a cationized OVA derivative with more cationic charges, showed pronounced induction of the OVA-specific CTL response after subcutaneous immunization. The CTL response was comparable with that induced by OVA with CFA. In contrast to the CFA formulation that actually produced local tissue damage in this study, local damage at the injection sites was not observed with Cat-OVAs. Cat(20)-OVA also showed a significant protective effect on the growth of OVA-expressing E.G7 tumor cells. In conclusion, cationization of soluble antigen is a useful and safe vaccination strategy.

Subcellular trafficking of exogenously expressed interferon-? in Madin-Darby canine kidney cells

We have recently demonstrated that when IFN-beta was exogenously expressed in epithelial cells, transiently expressed IFN-beta was predominantly secreted from the cell side to which the transfection was performed, while stably expressed one was almost equally secreted to the apical and basolateral sides. In the present study, we analyzed the subcellular transport of IFN-beta using confocal imaging with green fluorescent protein (GFP)-tagged IFN-beta in Madin-Darby canine kidney (MDCK) cells. Stably expressed and transiently expressed human IFN-beta (HuIFN-beta)-GFPs were seen in upper regions of the nucleus. In stable HuIFN beta-GFP-producing transformants, transiently expressed mouse IFN-beta (MuIFN-beta) was apparently co-localized with the bulk of the constitutive HuIFN beta-GFP proteins at TGN, and a significant quantity of them then appeared to pass into distinct post-TGN vesicles, accepting either type of IFN. Meanwhile, when cells were co-transfected with both expression vectors, transiently expressed both IFNs tended to co-localize not only at TGN but in post-TGN vesicles. These results suggest that stably and transiently expressed IFN-betas, albeit co-localized at TGN, were transported through apparently discriminated post-TGN routes.

Effect of interferon-gamma on the pharmacokinetics of digoxin, a P-glycoprotein substrate, intravenously injected into the mouse

P-glycoprotein (P-gp) is an efflux transporter with a wide substrate specificity that plays an important role in the disposition of drugs in the epithelial cells of various tissues, such as the gastrointestinal tract, liver, and kidney. One characteristic feature of this efflux transporter is that its expression and activity are modulated by various factors, including cytokines. Here, we investigated the effect of interferon-gamma (IFN-gamma) on the transport activity of P-gp and its expression in mice, since the cytokine is induced by various stimuli and capable of provoking a variety of cellular responses. Twenty-four hours after a single intraperitoneal injection of IFN-gamma (1 x 10(5) U), mice were intravenously injected with [3H]digoxin, a P-gp substrate, and its pharmacokinetics was examined. IFN-gamma pretreatment resulted in retardation of plasma elimination of the drug with a concomitant increase of its tissue levels in liver, kidney, and intestine. Furthermore, the excretion of [3H]digoxin into the urine and bile, but not into the intestinal lumen, was significantly reduced: the urinary and biliary excretion clearances in IFN-gamma-treated mice were 65 and 55%, respectively, of those clearances in untreated mice. However, the P-gp expression levels were only slightly reduced (20-30% reduction) by IFN-gamma treatment in the liver, kidney, or intestine on Western blot analysis. IFN-gamma also caused a slight down-regulation (20-30% reduction) in the expression of cytochrome P450 3A (CYP3A) on Western blot analysis. Thus, a more pronounced effect may be elicited by IFN-gamma for common substrates of P-gp and CYP3A.