Hippocampal poly(ADP-Ribose) polymerase 1 and caspase 3 activation in neonatal bornavirus infection

Abstract

Infection of neonatal rats with Borna disease virus results in a characteristic behavioral syndrome and apoptosis of subsets of neurons in the hippocampus, cerebellum, and cortex (neonatal Borna disease [NBD]). In the NBD rat hippocampus, dentate gyrus granule cells progressively degenerate. Apoptotic loss of granule cells in NBD is associated with accumulation of zinc in degenerating neurons and reduced zinc in granule cell mossy fibers. Excess zinc can trigger poly(ADP-ribose) polymerase 1 (PARP-1) activation, and PARP-1 activation can mediate neuronal death. Here, we evaluate hippocampal PARP-1 mRNA and protein expression levels, activation, and cleavage, as well as apoptosis-inducing factor (AIF) nuclear translocation and executioner caspase 3 activation, in NBD rats. PARP-1 mRNA and protein levels were increased in NBD hippocampi. PARP-1 expression and activity were increased in granule cell neurons and glia with enhanced ribosylation of proteins, including PARP-1 itself. In contrast, levels of poly(ADP-ribose) glycohydrolase mRNA were decreased in NBD hippocampi. PARP-1 cleavage and AIF expression were also increased in astrocytes in NBD hippocampi. Levels of activated caspase 3 protein were increased in NBD hippocampi and localized to nuclei, mossy fibers, and dendrites of granule cell neurons. These results implicate aberrant zinc homeostasis, PARP-1, and caspase 3 activation as contributing factors in hippocampal neurodegeneration in NBD.

FIG. 1.

Granule cell degeneration, zinc distribution, and apoptosis in the NBD HC. (A) Timm's staining (dark-brown stain) in PND28 control rat dentate gyrus. gcl, granule cell layer; iml, inner molecular layer; mf, mossy fibers. (B) Timm's staining in PND28 NBD rat dentate gyrus; note the accumulation of zinc in DGNs (arrows), decreased intensity of zinc stain in mossy fibers, and increased intensity of zinc stain in the iml. (C) Representative TUNEL of PND28 control rat DGNs. (D) Representative TUNEL of PND28 NBD rat DGNs; note the numerous TUNEL-positive neurons (brown stain). (E) TUNEL-positive DGNs with pyknotic nuclei (arrow) and signs of nuclear fragmentation (arrowhead). (F) Quantitation of TUNEL-positive DGNs in control (n = 4; 0.75 ± 0.479 cells) and PND28 NBD (n = 4; 80 ± 17.213 cells) rats (ANOVA, P = 0.0037). The data are presented as the mean number of DGNs per dentate gyrus ± standard error of the mean. The asterisk indicates a P value of <0.05. (G and H) IF staining for BDV nucleoprotein in control (G) and NBD (H) rat DGNs. (I and J) IF staining for BDV nucleoprotein in control (I) and NBD (J) rat hippocampal CA1 region pyramidal neurons. Bars = 100 μm.

FIG. 2.

PARP-1 mRNA and protein expression in NBD and control rat brains. (A) Real-time PCR analysis of PARP-1 mRNA in HC of PND28 control (n = 5) and NBD (n = 7) rats. PARP-1 mRNA in HC was significantly increased for NBD relative to control rats (3.18-fold increase in NBD; ANOVA, P = 0.044). The asterisk indicates a P value of <0.05. The error bars indicate standard errors of the means (SEM). (B) Representative Western immunoblots for PARP-1 protein (top) in extracts from control or NBD rat HC. Corresponding immunoblot signals for GAPDH are shown (bottom). The bar graph shows the determination of the PARP-1 protein band density relative to the GAPDH signal in extracts from control (n = 4) or NBD (n = 4) rat HC. HC PARP-1 protein levels were significantly increased in extracts from NBD rats relative to controls (1.68-fold increase in NBD; ANOVA, P = 0.012). PARP-1 IF in control (C, E, and G) and NBD (D, F, and H) rat HC. Note the faint staining of DGNs in control rats (C and E) and increased staining of DGNs in NBD rats (D and F). Few PARP-1-positive cells were apparent in the molecular layers of control rats (C and G) compared to numerous PARP-1-positive cells scattered throughout the molecular layers of NBD rats (D and H). gcl, granule cell layer; ml, molecular layer; Hi, hilus. Bars = 100 μm.

FIG. 3.

Enhanced poly(ADP-ribosyl)ation of proteins in NBD rats. (A) Protein lysates from HC of PND28 control (n = 4) and NBD (n = 4) rats were evaluated by Western analysis using Abs specific for PAR. (B) Determination of protein band densities relative to that for GAPDH revealed increased ribosylation of PARP-1 (116 kDa; 1.74-fold increase; ANOVA, P = 0.05), an unknown 50-kDa protein (8.25-fold increase; ANOVA, P = 0.02), and an unknown 25-kDa protein (9.15-fold increase; ANOVA, P = 0.03). The asterisks indicate P values of <0.05. The error bars indicate SEM. (C) Real-time PCR analysis of PARG mRNA in HC of PND28 control (n = 5) or NBD (n = 7) rats. PARG mRNA was significantly decreased for NBD rats relative to the results for control rats (1.4-fold decrease; ANOVA, P = 0.041). (D to G) IF analysis for PAR in DGNs (D and E) and the dentate molecular layer (F and G) of PND28 control (D and F) and NBD (E and G) rats. Note the enhanced IF in DGNs and cells in the molecular layer of NBD relative to control rats. gcl, granule cell layer; ml, molecular layer. Bars = 100 μm.

FIG. 4.

Cleavage of PARP-1 and AIF expression in NBD rat glia. (A and B) IF analysis with p85PARP-1 Abs in PND28 control (A) and NBD (B) rat HC. Note the increased IF for p85PARP-1 in astrocytes (colocalization with GFAP is not shown) throughout the molecular layer of the dentate gyrus of NBD rats. (B, inset) Higher magnification of a p85PARP-1 immunoreactive astrocyte. (C and D) IF analysis with AIF Abs in PND28 control (C) and NBD (D) rat HC. Note the increased expression of AIF in astrocytes in the NBD rat HC (D, inset). gcl, granule cell layer; ml, molecular layer; Hi, hilus; lac, stratum lacunosum-moleculare; rad, stratum radiatum. Bars = 100 μm.

FIG. 5.

Caspase 3 activation in the NBD rat HC. (A) Protein lysates from the HC of PND28 control and NBD rats were evaluated by Western analysis using Abs specific for activated (cleaved) caspase 3. Note that activated caspase 3 was not detected in any lysates from control rats but was present as strong bands in all NBD lysates. The bar graph shows the determination of protein band densities from control (n = 4) and NBD (n = 4) protein extracts, which revealed significantly increased levels for both the 17-kDa (14.4-fold increase; ANOVA, P = 0.01) and 19-kDa (6.83-fold increase; ANOVA, P = 0.008) species of activated caspase 3 in NBD HC relative to corresponding blot regions of control rat lysates. The asterisks indicate P values of <0.05. The error bars indicate SEM. (B to J) IF analysis of activated caspase 3 in the HC of control (B, D, F, and H) and NBD (C, E, G, I, and J) rats. (D and E) Higher-magnification views of DGNs in control (D) and NBD (E) rats. (F and G) Activated caspase 3 immunopositive astrocytes in the dentate molecular layer of control (F) and NBD (G) rats. (H to J) Immunolocalization of activated caspase 3 in axons and dendrites of NBD DGNs. Shown is activated caspase 3 IF in control (H) and NBD (I) rat DGN mossy fibers and NBD DGN dendrites extending into the molecular layer of the dentate gyrus (J, arrows). gcl, granule cell layer; ml, molecular layer; CA3, stratum pyramidale of the CA3 subregion. Bars = 100 μm.

FIG. 6.

Colocalization of molecular markers. (A) Double-label IF showing colocalization of PARP-1 and the neuronal marker NeuN in PND28 NBD rat DGNs. (B) Double-label IF showing colocalization of PAR and PARP in PND28 NBD rat DGNs. (C) Double-label IF showing colocalization of PAR and BDV N in PND28 NBD rat DGNs (boxed cells). Note that many DGNs have strong PAR signal in the absence of viral antigen (arrows). (D) Double-label IF showing colocalization of PARP-1 and the microglial marker OX-42 in the molecular layer of the HC of a PND28 NBD rat. (E) Double-label IF showing colocalization of PARP-1 and the astrocytic marker GFAP (arrow indicates double-labeled astrocyte) in PND28 NBD rat HC; note the stronger PARP-1 staining in microglia (boxed cell). (F) Double-label IF showing colocalization of activated caspase 3 and NeuN in DGN of a PND28 NBD rat (arrow indicates double-labeled DGN). (G) Double-label IF showing colocalization of activated caspase 3 and NeuN in a cortical neuron (arrow indicates double-labeled cortical neuron) of a PND28 NBD rat; note the caspase 3-positive astrocyte in the cortex (boxed cell). (H) Double-label IF showing colocalization of activated caspase 3 and GFAP in the molecular layer of the dentate gyrus of a PND28 NBD rat.

FIG. 7.

PARP-1, PAR, and activated caspase 3 in cortex and CBLM. (A) IF analysis of PARP-1 (top row; arrows indicate PARP-1 immunoreactive microglia), PAR (middle row), and activated caspase 3 (bottom row; arrows indicate activated caspase 3-positive cortical neurons) in the cortices of PND28 control (left) and NBD (right) rats. (B) IF analysis of PARP-1 (top row), PAR (middle row), and activated caspase 3 (bottom row) in the CBLM of PND28 control (left) and NBD (right) rats.