Impaired electrical signaling disrupts gamma frequency oscillations in connexin 36-deficient mice

Neural processing occurs in parallel in distant cortical areas even for simple perceptual tasks. Associated cognitive binding is believed to occur through the interareal synchronization of rhythmic activity in the gamma (30-80 Hz) range. Such oscillations arise as an emergent property of the neuronal network and require conventional chemical neurotransmission. To test the potential role of gap junction-mediated electrical signaling in this network property, we generated mice lacking connexin 36, the major neuronal connexin. Here we show that the loss of this protein disrupts gamma frequency network oscillations in vitro but leaves high frequency (150 Hz) rhythms, which may involve gap junctions between principal cells (Schmitz et al., 2001), unaffected. Thus, specific connexins differentially deployed throughout cortical networks are likely to regulate different functional aspects of neuronal information processing in the mature brain.     

Mouse horizontal cells do not express connexin26 or connexin36

Gap junctions between neurons function as electrical synapses, and are present in all layers of mammalian and teleost retina. These synapses are largest and most prominent between horizontal cells where they function to increase the receptive field of a single neuron beyond the width of its dendrites. Receptive field size and the extent of gap junctional coupling between horizontal cells is regulated by ambient light levels and may mediate light/dark adaptation. Furthermore, teleost horizontal cell gap junction hemichannels may facilitate a mechanism of feedback inhibition between horizontal cells and cone photoreceptors. As a prelude to using mouse genetic models to study horizontal cell gap junctions and hemichannels, we sought to determine the connexin complement of mouse horizontal cells. Cx36, Cx37, Cx43, Cx45 and Cx57 mRNA could be detected in mouse retina by RT-PCR. Microscopy was used to further examine the distribution of Cx26 and Cx36. Cx26 immunofluorescence and a beta-gal reporter under regulatory control of the Cx36 promoter did not colocalize with a horizontal cell marker, indicating that these genes are not expressed by horizontal cells. The identity of the connexin(s) forming electrical synapses between mouse horizontal cells and the connexin that may form hemichannels in the horizontal cell telodendria remains unknown.     

IKAP/hELP1 down-regulation in neuroblastoma cells causes enhanced cell adhesion mediated by contactin overexpression

A splicing mutation in the IKBKAP gene encoding the IKAP/hELP1 (IKAP) protein was found to be the major cause of Familial Dysautonomia (FD). This mutation affects both the normal development and survival of sensory and sympathetic neurons of the peripheral nervous system (PNS). To understand the FD phenotype it is important to study the specific role played by IKAP in developing and mature PNS neurons. We used the neuroblastoma SHSY5Y cell line, originated from neural crest adrenal tumor, and simulated the FD phenotype by reducing IKAP expression with retroviral constructs. We observed that IKAP – down - regulated cells formed cell clusters compared to control cells under regular culture conditions. We examined the ability of these cells to differentiate into mature neurons in the presence of laminin, an essential extracellular matrix for developing PNS neurons. We found that the cells showed reduced attachment to laminin, morphological changes and increased cell-to-cell adhesion resulting in cell aggregates. We identified Contactin as the adhesion molecule responsible for this phenotype. We show that Contactin expression is related to IKAP expression, suggesting that IKAP regulates Contactin levels for appropriate cell-cell adhesion that could modulate neuronal growth of PNS neurons during development.     

Phorbol ester induces phosphorylation and down-regulation of connexin 43 in WB cells.

Western blot analysis indicated that WB cells contained multiple proteins of molecular mass 42-47 K which reacted with anti-connexin 43 antibody. Incubation of cells with 12-O-tetradecanoylphorbol-13-acetate (TPA) for 5 min caused the inhibition of dye transfer and the appearance of reactive material of 50 K which was abolished by acid phosphatase. Longer term incubation with TPA caused loss of connexin 43 protein.     

Immunocytochemical and RT-PCR analysis of connexin36 in cultures of mammalian glial cells

The expression of connexin36 (Cx36) was studied in primary cultures of rat brain glial cells: mature astrocytes, ameboid and ramified microglia and immature oligodendrocytes (at middle period of myelinogenesis). The data from these cells were compared with those obtained from cultures of neocortical and hypothalamic neurons. mRNA encoding Cx36 was investigated by RT-PCR, the Cx36 protein by immunocytochemistry using a polyclonal antibody against Cx36 in cells characterized by antibodies specific for the single cell types. The Cx36 was found in oligodendrocytes, both ameboid and ramified microglial cells and in neurons. Astrocytes showed no detectable expression of the Cx36. The expression of Cx36 in oligodendrocytes and microglial cells suggests an involvement of the direct cell-cell communication channels formed by Cx36 in myelin formation and in brain development, damage and repair processes.     

Association of connexin36 with zonula occludens-1 in HeLa cells, betaTC-3 cells, pancreas, and adrenal gland

The PDZ domain-containing protein zonula occludens-1 (ZO-1), a well-established component of tight junctions, has recently been shown to interact with various connexin proteins that form gap junctions. We investigated the association of connexin36 (Cx36) with ZO-1 in various cultured cells and tissues. Punctate immunofluorescence labeling for Cx36 was detected in Cx36-transfected HeLa cells, betaTC-3 cells, pancreatic islets, and adrenal medulla. Immunofluorescence for ZO-1 was also punctate in cells and tissues, and was colocalized with Cx36 at points of cell-cell contact. Immunoprecipitation of either Cx36 or ZO-1 from cell lysates and tissue homogenates resulted in immunoblot detection of ZO-1 or Cx36, respectively, in immunoprecipitates. A 14-amino acid peptide corresponding to the carboxy-terminus of Cx36 showed binding capacity to the PDZ1 domain of ZO-1, which was eliminated after removal of the last 4 carboxy-terminus amino acids. Low micromolar concentrations of the 14-amino acid peptide produced up to 85% inhibition of Cx36 interaction with the PDZ1 domain of ZO-1. These results provide evidence for molecular interaction between Cx36 and ZO-1 in vitro, and in vivo, and suggest that the interference with Cx36/ZO-1 interaction by short carboxy-terminus peptides of Cx36 may be of value for functional studies of this interaction.     

Immunohistochemical detection of the neuronal connexin36 in the mouse central nervous system in comparison to connexin36-deficient tissues

Investigating the spatial and temporal expression of connexin36 (Cx36) protein in neuronal tissue is of prime importance to understand the molecular mechanisms underlying extensive electrical coupling. Although Cx36 mRNA was shown to be expressed in neurons of the central nervous system in different studies, only the determination of Cx36 protein expression allows a correlation between localization and its functional role in gap junction-mediated neuronal coupling. After the initial use of antibodies recognizing the skate connexin35 protein, antibodies directed to the mammalian Cx36 sequence allowed the detailed investigation of Cx36 cellular localization. However, results on Cx36 protein distribution still remained controversial in some areas of the central nervous system. In the present study, we have investigated: (a) the distribution of Cx36 protein in various areas of the central nervous system and (b) determined the specificity in the immunohistochemical staining of two polyclonal antibodies comparing wildtype and Cx36-deficient mice. In some areas of the central nervous system, for example in the retina and the inferior nuclear olivary complex, Cx36 antibodies were highly specific, and in the cerebellar cortex, Cx36 protein expression was partly specific. In other regions, particularly in pyramidal cells of the hippocampal formation, non-specific staining was prevalent, indicating that Cx36 antibodies also recognize proteins other than Cx36 in these tissues. The present results argue for a re-evaluation of many documented immunohistochemical protein distribution patterns and require, not only in connexin research, their assessment using null-mutant animals.     

2-Ketoisocaproate transport in insulin-secreting cells

The transport of the nutrient secretagogue 2-ketoisocaproate (KIC) was studied in isolated rat pancreatic islets and in the HIT-T15 insulinoma cell line using an oil-filtration technique. In both islets and HIT-T15 cells, KIC uptake was a slow process, not reaching equilibrium within 10 min KIC transport was not dependent upon Na⁺ in the medium, was not inhibited by -cyano-4-hydroxy-cinnamate nor by 2-amino-2-norborane carboxylic acid (BCH) and did not appear to be electrogenic. Evidence was obtained to suggest that KIC uptake occurred via passive diffusion into the cell of the undissociated acid species. This possibility was supported by the apparent unsaturability of KIC uptake in HIT-T15 cells. Addition of 10–30 mM KIC to dispersed islets cells or HIT-T15 cells produced a rapid intracellular acidification. In islets, the rate of transport of 10 mM KIC was comparable with oxidation rate of the keto-acid suggesting that uptake could be rate-limiting factor for KIC oxidation and thus stimulated insulin release. However, in HIT-T15 cells, the rate of uptake of KIC greatly exceeded the oxidation rate. The low rate of KIC oxidation could explain the poor secretory response of HIT-T15 cells to KIC     

Membrane potential dependent modulations of calcium oscillations in insulin-secreting INS-1 cells

This study was undertaken to examine the role of K(+) channels on cytosolic Ca(2+) ([Ca(2+)](i)) in insulin secreting cells. [Ca(2+)](i) was measured in single glucose-responsive INS-1 cells using the fluorescent Ca(2+) indicator Fura-2. Glucose, tolbutamide and forskolin elevated [Ca(2+)](i) and induced [Ca(2+)] oscillations. Whereas the glucose effect was delayed and observed in 60% and 93% of the cells, in a poorly and a highly glucose-responsive INS-1 cell clone, respectively, tolbutamide and forskolin increased [Ca(2+)](i) in all cells tested. In the latter clone, glucose induced [Ca(2+)](i) oscillations in 77% of the cells. In 16% of the cells a sustained rise of [Ca(2+)](i) was observed. The increase in [Ca(2+)](i) was reversed by verapamil, an L-type Ca(2+) channel inhibitor. Adrenaline decreased [Ca(2+)](i) in oscillating cells in the presence of low glucose and in cells stimulated by glucose alone or in combination with tolbutamide and forskolin. Adrenaline did not lower [Ca(2+)](i) in the presence of 30mM extracellular K(+), indicating that adrenaline does not exert a direct effect on Ca(2+) channels but increases K(+) channel activity. As for primary b-cells, [Ca(2+)](i) oscillations persisted in the presence of closed K(ATP) channels; these also persisted in the presence of thapsigargin, which blocks Ca(2+) uptake into Ca(2+) stores. In contrast, in voltage-clamped cells and in the presence of diazoxide (50mM), which hyperpolarizes the cells by opening K(ATP) channels, [Ca(2+)](i) oscillations were abolished. These results support the hypothesis that [Ca(2+)](i) oscillations depend on functional voltage-dependent Ca(2+) and K(+) channels and are interrupted by a hyperpolarization in insulin-secreting cells.     

Peroxisome proliferator-activated receptor gamma overexpression suppresses proliferation of human colon cancer cells

Peroxisome proliferator-activated receptor gamma (PPARγ) plays an important role in the differentiation of intestinal cells and tissues. Our previous reports indicate that PPARγ is expressed at considerable levels in human colon cancer cells. This suggests that PPARγ expression may be an important factor for cell growth regulation in colon cancer. In this study, we investigated PPARγ expression in 4 human colon cancer cell lines, HT-29, LOVO, DLD-1, and Caco-2. Real-time polymerase chain reaction (PCR) and Western blot analysis revealed that the relative levels of PPARγ mRNA and protein in these cells were in the order HT-29>LOVO>Caco-2>DLD-1. We also found that PPARγ overexpression promoted cell growth inhibition in PPARγ lower-expressing cell lines (Caco-2 and DLD-1), but not in higher-expressing cells (HT-29 and LOVO). We observed a correlation between the level of PPARγ expression and the cells' sensitivity for proliferation.     

Thioredoxin overexpression in HT-1080 cells induced cellular senescence and sensitization to gamma radiation

An increment of thioredoxin-1 (TRX) is observed in many human primary cancers and appears to contribute to an increase of cell growth and a resistance to chemotherapy. On the contrary, when TRX was overexpressed in the HT-1080 fibrosarcoma cells, the cell growth was retarded and chromosomal polyploidy and cellular senescence were induced. TRX-overexpression made HT-1080 cells resistant to an oxidative stress caused by H2O2 or paraquat. But these cells were significantly sensitive to ionizing radiation, showing an abrogation of the G2 checkpoint. Their DNA contents were twice of the controls and they expressed typical senescence markers. Their expression levels of p53 and cyclin-dependent kinase inhibitors (CDKI) were about 2-3-fold higher than the control. Nevertheless, cyclin D1 and D3, which are negatively regulated by CDKIs, were also increased. Overall, in HT-1080 cells the TRX-overexpression created a state of cellular senescence caused by a simultaneous stimulation of the mitogen-activated pathways and an inhibition of the cyclin-dependent kinases, which is known as a hypermitogenic arrest.     

Voltage-activated Ca2+ currents in insulin-secreting cells

Membrane voltage and voltage-clamped membrane currents have been investigated with the whole-cell patch clamp method in the insulin-secreting cell line RINm5F. The mean resting membrane potential of RINm5F cells was found to be -52 mV. Overshooting spike potentials could be evoked by depolarising voltage steps in the absence of a secretagogue. Inward membrane currents evoked by depolarising voltage steps were dependent upon extracellular Ca2+ and blocked by Co2+, nifedipine and verapamil. Outward membrane currents which were evoked by depolarising voltage steps to positive membrane potentials were reduced when Ca2+ entry was prevented. It is concluded that the voltage-activated Ca2+ currents underlie the voltage-activated spike potentials recorded from insulin-secreting cells.     

Effects of overexpression of pancreatic derived factor (FAM3B) in isolated mouse islets and insulin-secreting betaTC3 cells

PANcreatic DERived factor (PANDER, FAM3B) is a recently discovered islet-specific cytokine. We have previously shown that, in vitro, truncated recombinant PANDER isoforms (20 and 21 kDa) are cytotoxic to beta-cell lines but the effects of full-length PANDER on islet biology remain unclear. In this study, we used adenovirus (Ad-PANDER) to overexpress full-length cDNA of PANDER in islets and betaTC3 cells. BetaTC3 cells were infected with Ad-PANDER or control vector. After 48 h, cell viability was significantly decreased as evaluated by MTT assay. The number of dead cells was significantly increased as indicated by the fluorescent intensity of the propidium iodide-stained cells (160 +/- 13 vs. control 100 +/- 7%, P = 0.001). Flow cytometric Tunel assay showed that overexpressing PANDER induced a significant fourfold increase in beta-cell apoptosis (19.4 +/- 6.3 vs. control 4.1 +/- 0.8%, P < 0.05). There was a significant increase in the number of annexin V-positive (apoptotic) cells and propidium iodide-positive (dead) cells in mouse islets infected with Ad-PANDER compared with control cells infected with Ad-LacZ. Addition of 4 nM recombinant PANDER protein to betaTC3 cells or infection of Ad-PANDER did not affect Akt and STAT1 phosphorylation, Bcl-2, Fas, and NF-kappaB protein levels. However, activation of caspase-3 was observed in betaTC3 and islets infected with Ad-PANDER. Overexpression of PANDER in mouse islets or addition of recombinant PANDER decreased insulin secretion induced by carbachol plus glucose or high potassium but not that by glucose alone. Culture with recombinant PANDER did not affect glucose-induced NAD(P)H elevation in mouse islets. In conclusion, Ad-PANDER infection is as effective as truncated recombinant PANDER to induce betaTC3 cell and mouse islet apoptosis.