Nuclear depletion of apurinic/apyrimidinic endonuclease 1 (Ape1/Ref-1) is an indicator of energy disruption in neurons

Abstract

Apurinic/apyrimidinic endonuclease 1 (Ape1/Ref-1) is a multifunctional protein critical for cellular survival. Its involvement in adaptive survival responses includes key roles in redox sensing, transcriptional regulation, and repair of DNA damage via the base excision repair (BER) pathway. Ape1 is abundant in most cell types and central in integrating the first BER step catalyzed by different DNA glycosylases. BER is the main process for removal of oxidative DNA lesions in postmitotic brain cells, and after ischemic brain injury preservation of Ape1 coincides with neuronal survival, while its loss has been associated with neuronal death. Here, we report that in cultured primary neurons, diminution of cellular ATP by either oligomycin or H(2)O(2) is accompanied by depletion of nuclear Ape1, while other BER proteins are unaffected and retain their nuclear localization under these conditions. Importantly, while H(2)O(2) induces γH2AX phosphorylation, indicative of chromatin rearrangements in response to DNA damage, oligomycin does not. Furthermore, despite comparable diminution of ATP content, H(2)O(2) and oligomycin differentially affect critical parameters of mitochondrial respiration that ultimately determine cellular ATP content. Taken together, our findings demonstrate that in neurons, nuclear compartmentalization of Ape1 depends on ATP and loss of nuclear Ape1 reflects disruption of neuronal energy homeostasis. Energy crisis is a hallmark of stroke and other ischemic/hypoxic brain injuries. In vivo studies have shown that Ape1 deficit precedes neuronal loss in injured brain regions. Thus, our findings bring to light the possibility that energy failure-induced Ape1 depletion triggers neuronal death in ischemic brain injuries.

Figure 1. Reduction of neuronal ATP content by H₂O₂ and oligomycin exposures

Cellular ATP content was measured in DIV7 neurons incubated with indicated concentrations of H₂O₂ and oligomycin. Cultures were lysed and ATP content determined using ATP Bioluminescence Assay and normalized to protein amount. ATP content was significantly reduced: 3-hour exposure to 25 μM H₂O₂ resulted in ~30% reduction, whereas exposure to 50 μM, reduced ATP content by ~55%. Exposures to 1 nM and 2.5 nM oligomycin resulted in 30% and nearly 50% reduction, respectively. Values are presented as means ± SEM relative to non-treated controls for 6 independent experiments; *P<0.05 versus control.

Figure 2. Induction of γH2AX phosphorylation by exposure to H₂O₂ but not to oligomycin

Representative Immunofluorescent images of DIV7 neurons reacted with anti Map2 (neuronal marker, red) and anti-γH2AX (phosphorylated at serine 139, green; nuclei stain blue with DAPI) are shown. γH2AX positive foci emerge and intensify in the course of exposure to 50 μM H₂O₂ (green), indicative of chromatin modifications associated with formation and processing of DNA damage. In contrast, γH2AX immunoreactivity was undetectable in the course of exposure to 2.5 nM oligomycin, indicating that significant DNA damage is not induced under these conditions (right), bar=6 μm.

Figure 3. Depletion of nuclear Ape1 by exposure of neurons to H₂O₂ and oligomycin

Immunofluorescent images of primary neuronal cultures are shown. Neuronal soma and extensions are visualized by Map2 immunoreactivity (red), whereas Ape1 is observed in green (nuclei stain blue with DAPI; bar=12 μm). On DIV7, cultures were exposed to H₂O₂ or oligomycin at indicated concentrations. Under control conditions (left panel), neuronal extensions are robust (Map2, red) and nuclei show uniform Ape1 immunoreactivity (green), which overlaps with DAPI (blue). In addition, few condensed apoptotic nuclei, which are inherent to primary neuronal cultures and stain with DAPI alone, are also observed. Following incubations with the higher concentration of either H₂O₂ (50 μM) or oligomycin (2.5 nM), neuronal extensions are shortened, nuclei become rounded and Ape1 immunoreactivity (green) increases in cytoplasm and decreases in nuclear compartment. Following exposure to 25 μM H₂O₂ ~30% of neurons undergo changes, while in the case of 50 μM H₂O₂ or 2.5 nM oligomycin, the proportion of affected cells increases to ~70% and 60%, respectively. Concomitantly, nuclear Ape1 immunoreactivity (green) decreases and shifts to cytoplasm. In merged images, yellow color (bottom) reflects an overlap of cytoplasmic Ape1 (green) and Map2 (red).

Figure 4. Nuclear localization of BER proteins, ligase 3 and Xrcc1 is maintained, while nuclear Ape1 is depleted following H₂O₂ and oligomycin exposures

(A) Immunofluorescent images of DIV7 neurons reacted with anti Map2 antibody (red) and anti-ligase3 (green), show nuclear immunoreactivity of ligase 3, which coincides with DAPI (blue; scale bar=6 μm). Nuclear localization of ligase 3 (green, left panel) was not altered by either 50 μM H₂O₂ or 2.5 nM oligomycin exposures, and fully retained in the nuclear compartment (middle and right panels). (B) Xrcc1 is retained while Ape1 is depleted from nuclei and accumulates in cytoplasm following H₂O₂ exposure. Following exposure to H₂O₂, Xrcc1 immunoreactivity is retained in nucleus (red), while Ape1 immunoreactivity (green) shifts to cytoplasm (top, right); consequently, no overlap of Xrcc1 and Ape1 immunoreactivity is detected in merged images after treatment (bottom, right).

Figure 5. Abasic site incision and nick ligation activities in nuclear extracts from DIV7 neurons are differentially affected by exposures to H₂O₂ and oligomycin

Reactions were assembled with radioactively labeled substrates. Products resolved in denaturing polyacrylamide gels were visualized by autoradiography; representative autoradiograms are shown. Time points are indicated and a substrate-only lane is included (lane 1). (A) Incision assay of abasic site (THF) carrying substrate. Reaction with recombinant human Ape1 serves as positive control (lane 14). 5′-end labeled 20-mer substrate (S) and the 9-mer incision product (P) are indicated. (B) Nick ligation assay. 5′-end labeled 21-mer substrate (S) and ligated 45-mer product (P) are indicated. Labeled oligonucleotides resolved in parallel provide size markers (M). Bar graphs represent product yields calculated from phosphorimager values and expressed as pmoles/min/mg protein. Values for three or four independent assays for each substrate in the linear range of reaction were used to obtain mean ± SEM. *P<0.05 versus control.

Figure 6. Differential effects of H₂O₂ and oligomycin on mitochondrial respiratory profiles in primary neurons

On DIV7 neurons were exposed on to H₂O₂ or oligomycin and along with non-treated control neurons, processed for measurements of oxygen consumption rates (OCR) with XF24 analyzer. Respiratory profile established for non-treated control neurons (A & C, circles) showed baseline OCR of approximately 200 pmoles O₂/min/10⁵ neurons. Sequential, in port additions of modifiers of mitochondrial activities (downward arrows), revealed that under normal conditions ~65% of consumed oxygen feeds into ATP synthesis, proton leak is limited to ~15% of oxygen consumption, spare respiratory capacity is at ~60%, while non mitochondrial respiration accounts for ~20% of total oxygen consumption. Respiratory profile was changed following a 3-hour exposure to 25 μM H₂O₂ (A, solid triangles) and to a greater extent following exposure to 50 μM H₂O₂ (A, solid squares). Changes in respiratory profile emerged also after exposure to 1 nM oligomycin (C, open triangles) and to a much greater extent after a 2.5 nM treatment (C, open squares). The extent of changes induced by each treatment versus respective control, was calculated for the different segments of respiratory profiles (baseline, ATP synthesis, proton leak and spare respiratory capacity) and presented as means ± SEM of 4–5 independent experiments for each treatment: H₂O₂ (B, solid bars) and oligomycin (D, open bars). * P<0.05 versus respective control.