Article Figures & Data
Figures
- Fig. 1.
Cx43 expression confers resistance to injury.A–C, Confocal microscopic images showing double-immunofluorescence labeling for Bcl2 (fluorescein-tagged secondary antibody) and Cx43 (Texas Red-tagged secondary antibody). The nuclei were counterstained with Hoechst (blue).A, C6–mock 1 cells express undetectable levels of either Cx43 or Bcl2. B, C6 cells stably expressing Bcl2 (C6–Bcl2 cells) are strongly immunoreactive. C, Cx43 immunoreactive plaques in C6 cells stably expressing Cx43 (C6–Cx43 cells). Scale bar, 10 μm. D, TUNEL-positive C6–mock 1 cells fixed 24 hr after a 10 min exposure to the calcium ionophore lasalocid (40 μm). E, Percentage of cell death as a function of exposure time to lasalocid (40 μm) in C6–mock 1, C6–Cx43, and C6–Bcl2 cells. Cx43 expression and Bcl2 expression are both associated with a substantial increase in cellular resistance to the calcium ionophore. Error bars indicate SEM.
- Fig. 2.
Anti-death activity in Cx43-, Cx40-, and Cx32-expressing clones. A, LD50 (dosage causing half-maximal death) for the calcium ionophore lasalocid (40 μm) is significantly higher for Cx43, Cx40, and Cx32 cells than for four mock-transfected sister clones, C6–mock 1, C6–mock 2, C6–mock 3, and C6–mock 4. *p < 0.01. Error bars indicate SEM. B, Coupling index in the same clones as in A. Inset, LD50 is a direct function of the coupling index (r = 0.97). *p < 0.01; ANOVA and post hoc Bonferroni t test. Error bars indicate SEM. C, An example of the dye-transfer assay. C6–Cx43 cells were preloaded with DiIC18(red) and the gap junction-permeable tracer CDCF (green). Labeled C6–Cx43 cells were cocultured with unlabeled C6–Cx43 cells for 1 hr; gap junctional coupling was quantified by the transfer of CDCF from DiIC18-labeled cells to unlabeled cells. Donor cells appear yellowbecause of the merge of red and greenlabeling. Clusters of green receiving cells surroundyellow donors when the donor cell establishes a gap junction with adjacent cells (white arrowheads), whereas single yellow–red cells represent donor cells that fail to establish gap junctions (red arrowheads). The coupling index is defined as the mean number of receiving (green) cells per donor (yellow) cell.
- Fig. 3.
Cx43 and Cx32 confer resistance to several but not all injury paradigms. The LD50 of C6–mock 1, C6–Cx43, and C6–Cx32 cells exposed to menadione (200 μm), KCN (1 mm) and iodoacetate (IA; 0.02 mm), tamoxifen (10–25 μm), UV irradiation (0–60 min), dexamethasone (0–6 mm), and staurosporine (0–3 μm) is shown. The expression of Cx43 and Cx32 protects C6 cells against menadione, KCN and IA, tamoxifen, and UV irradiation but not against dexamethasone and staurosporine. *p < 0.01; ANOVA and post hoc Bonferroni t test. Error bars indicate SEM.
- Fig. 4.
Anti-death activity is highest in cells with high Cx expression and is not affected by multiple replating. Comparison of injury resistance of C6–mock 1 cells (coupling index, 0.2), a low Cx32-expressing clone (coupling index, 2.3 ± 0.5), and a high Cx32-expressing clone (coupling index, 11 ± 1). Resistance to various insults is expressed as LD50 for lasalocid (40 μm) (A), menadione (200 μm) (C), and KCN (1 mm) and iodoacetate (0.02 mm) (B). The high-expression clone is consistently more resistant than the low-expression clone. *p < 0.01; ANOVA and post hoc Bonferroni ttest compared with C6–mock 1 cells. D, Multiple replating of C6–Cx43 cells does not decrease the LD50 of lasalocid. Error bars indicate SEM.
- Fig. 5.
Dominant-negative Cx43 mutants reduce the resistance of wild-type C6 cells. A, The expression of the mutant Cx43 L160M as plaques at cell-to-cell contact was visualized via the fluorescence of EGFP (white arrows), whereas C6 wild type transfected with Ad(EGFP) displays diffuse fluorescence. Scale bar, 10 μm. B, C6 wild-type cells endogenously express a low level of Cx43, and their resistance to tamoxifen is significantly reduced after the expression of the Cx43 dominant-negative mutants L160M and Δ130–137 compared with cultures of C6 wild type expressing EGFP only. *p < 0.01; ANOVA and post hoc Bonferroni t test. Error bars indicate SEM.
- Fig. 6.
Resistance of N2A neuroblastoma cells to tamoxifen is increased with Cx43 expression. Cellular resistance to tamoxifen is increased in N2A cells transfected with an adenoviral construct, Ad(Cx43–EGFP), relative to mock controls that have been transfected with Ad(EGFP). *p < 0.01; ANOVA and post hoc Bonferroni t test. Error bars indicate SEM.Inset, Expression of Cx43 was visualized via immunocytochemistry with Cy3-tetramethylrhodamine isothiocyanate-tagged secondary antibodies. Because cDNAs for Cx43 and EGFP are located on separate cassettes, EGFP fluorescence is diffuse, whereas Cx43 immunoreactivity is restricted to a large plaque (red) at a region of cell-to-cell contact (white arrows).
- Fig. 7.
Plaque formation, but not functional gap junction channels, is required for Cx-mediated injury resistance.A, Diffuse Cx43 immunoreactivity in the cytosol of C6 cells transfected with mutant C61S–Cx43, which harbors a cysteine-to-serine point mutation at position 61. The C61S–Cx43 proteins do not form gap junction plaques, and correspondingly, do not increase the injury resistance. B, C6 cells transfected with a chimeric construct, Cx40*43C3, display immunoreactive plaques in cell membrane (arrows) and an increase in injury resistance. The expression of the Cx40*43 chimeras was visualized by an anti-Cx40 antibody, because the Cx43 antibodies target the C-terminal tail, which is not present in the chimeric constructs. Scale bar: (inB) A, B, 35 μm.C, Histograms summarizing the LD50 for ionophore (40 μm) exposure. *p < 0.01; ANOVA and post hoc Bonferroni ttest compared with C6–mock 1 cells. Error bars indicate SEM.
- Fig. 8.
C6–Cx43 cells remain resistant to injury in the absence of gap junctions. Representative fields of C6–Cx43 (left) and C6–mock 1 (middle) cell cultures in 24 well plates at 7000 (A), 15,000 (B), 23,000 (C), and 115,000 (C6–Cx43) or 200,000 (C6–mock 1) (both were confluent cultures) (D) cells per well. Note that at the lowest plating densities, cells rarely contact each other, thereby physically preventing the formation of gap junction channels. Scale bar, 50 μm. Right, Comparison of viability of C6–Cx43 (○) and C6–mock 1 (●) cells after exposure to increasing concentrations of tamoxifen. The sensitivity to tamoxifen increases inversely with the plating density for both C6–Cx43 and C6–mock cells. C6–Cx43 cells maintain their high resistance at plating densities at which gap junction coupling is prevented by the physical separation of the cells. Data are from a representative set of experiments. Similar results have been obtained from two other independent studies.
- Fig. 9.
Cx-mediated resistance is not affected by gap junction blockage. The gap junction inhibitor α-GA does not significantly (p = 0.09 for LD50of C6–Cx32 ± α-GA) reduce the extent of injury after ionophore exposure. Viability (as percentage of vehicle control) is plotted as a function of increasing exposure time to lasalocid (40 μm) of C6–mock 1, C6–Cx43, and C6–Cx32 cells. Error bars indicate SEM.
- Fig. 10.
Cx-mediated resistance persists in suspended cells. A, A sphere of C6–Cx43 cells that grew on a low-attachment plate stained with Texas Red–phalloidin. Actin is organized in a cortical mantle below the plasma membrane and is not in parallel arrays of stress fibers as in attached sister cells (inset). B, Spheres of C6–Cx43 cells exposed for 24 hr to either vehicle (left) or 20 μm tamoxifen (right). The culture was stained with phalloidin (green) and propidium (red). Exposure to tamoxifen does not result in killing C6–Cx43 cells. C, A sphere of C6–mock 1 cells stained with phalloidin. Actin in these cells is organized in a cortical mantle below the plasma membrane both in the sphere and in attached sister cells (inset). Note that phalloidin staining is weaker compared with C6–Cx43 cells (A, B). D, Spheres of C6–mock 1 cells exposed for 24 hr to either vehicle (left) or 20 μm tamoxifen (right). Condensed apoptotic nuclei in the tamoxifen-exposed sphere reveal that C6–mock 1 cells (D, right) remain more sensitive to tamoxifen than C6–Cx43 cells (B, right). Scale bar, 20 μm.
- Fig. 11.
C6–Cx43 cells pre-exposed to purinergic receptor P2Y antagonists remain coupled by gap junctions despite compaction and retraction, but their resistance to tamoxifen is compromised. A, Untreated C6–Cx43 culture loaded with the gap junction-permeable fluorescence indicator CDCF.Top, A field of cells before photobleach.Bottom, A field of cells collected immediately after photobleach, or 1 and 2 min later. The rapid recovery of CDCF fluorescence indicates that the cells are well coupled by gap junctions to neighboring cells. Arrowheads indicate the cell that is subjected to photobleach. Dashed boxes indicate areas of photobleach. B, C6–Cx43 cells exposed to the purinergic receptor antagonist reactive blue (50 μm) for 24 hr. Exposure to reactive blue does not decrease gap junction coupling, despite the reduction in cellular contact. C, Fluorescence recovery after photobleach in vehicle-, reactive blue-, and suramin-treated (50 μm each, 24 hr) C6–Cx43 cultures. D, Comparison of tamoxifen LD50. Reactive blue (RB) and suramin (50 μmeach, 24 hr) significantly reduce LD50 for C6–Cx43 but not the LD50 for C6–mock 1 cells. *p < 0.01; Student's t test.
- Fig. 12.
No difference in ionophore sensitivity between wild-type and Cx43–KO astrocytes. Confluent cultures were treated with 40 μm lasalocid for the times indicated. Viability was evaluated by alamar blue assay and expressed as a percentage of vehicle control. The LD50 for wild-type versus Cx43–KO astrocytes is not significantly different. Error bars indicate SEM.
- Fig. 13.
Ionophore-induced increases in cytosolic [Ca2+]i are suppressed by Cx expression. A, Resting [Ca2+]i levels in a mixed culture of C6–Cx43 (white arrows) and C6–mock 1 cells loaded with the calcium indicator fura-2. B, Peak [Ca2+]i increments during exposure to the calcium ionophore lasalocid (40 μm). Note the lower amplitude of [Ca2+]i increments in C6–Cx43 cells (white arrows) compared with C6–mock 1 cells. C, After 40 min of ionophore exposure, [Ca2+]i in all four C6–Cx43 cells had normalized to a level somewhat higher than resting [Ca2+]i, whereas [Ca2+]i remained elevated in the majority of C6–mock 1 cells. D, Phase contrast micrograph of the same field. The C6–Cx43 cells were prelabeled with the cell tracker CMTMR. The CMTMR labeling was digitally superimposed to visualize C6–Cx43 cells. Note the flat morphology of C6–Cx43 cells compared with the elongated C6–mock 1 cells that typically exhibit less cellular contact. E, [Ca2+]i as a function of time in the cultures shown in A–C. Peak [Ca2+]i increments in C6–Cx43 cells during ionophore exposure are lower than those in surrounding C6–mock cells (left; p < 0.001; Student'st test), normalize faster, and do not show a delayed secondary increase in calcium. Scale bar: (in D)A–D, 50 μm.
In this issue
