Figure 1.
Unprocessed NOA1 precursor protein is localized in the nucleus.
(A) Schematic representation of functional domains of the NOA1 protein and description of constructs used in the study. (B) Western blot analysis of C2C12 cells transfected with a NOA1 expression construct. The antibody against the C-terminal Flag-epitope detects the unprocessed precursor protein and the shorter processed protein present in mitochondria. Deletion of the N-terminal mitochondrial targeting sequence (MTS) prevents processing. (C) Subcellular fractionation of HEK293 cells expressing NOA1-Flag and NOA1ΔMTS-Flag reveals presence of unprocessed precursor proteins in the nuclear fraction marked by Histone H3 and the processed mature form in the mitochondrial fraction marked by Complex I 39 k. GAPDH served as marker for the cytosolic fraction.
Figure 2.
A fraction of endogenous NOA1 is found in the nucleus of C2C12 cells and primary mouse myofibers.
(A) Immunofluorescence staining of NOA1-transfected C2C12 cells. Mitochondria were visualized by antibody staining with the outer mitochondrial membrane markers Tom20 and Omp-25. Overexpressed NOA1-EGFP mainly localizes in mitochondria as shown by co-localization with the mitochondrial markers Tom20 and Omp-25. Endogenous NOA1 protein detected with a rabbit polyclonal anti-NOA1 antibody also localizes with mitochondria although a fraction of NOA1 is present in the nucleus. (B) Mitochondria in primary myofibers show a striped pattern alternating with F-actin. In primary mouse myofibers endogenous NOA1 presents a striated pattern similar to Tom20 and is also seen in punctae in myonuclei. (C) Diagram of the alternating F-actin/Phalloidin-FITC and Mitochondria/Tom20-Alexa594 fluorescence intensities shown in (B). The arrow highlights the analyzed region. Data points were fitted by a line with moving average of 10. Confocal images, scale bars: 10 µm.
Figure 3.
NOA1 carries a functional nuclear localization signal mediating GTP dependent active nuclear import.
(A) Immunofluorescence staining of NOA1-transfected C2C12 cells. Wild type NOA1 co-localizes with the mitochondrial marker protein Tom20 while deletion of the N-terminal MTS causes accumulation of EGFP-tagged NOA1 protein in the nucleus. (B) Overexpression of NOA1ΔMTS in C2C12 leads to apoptosis, which is partially rescued by addition of 1 µM pan-caspase inhibitor Z-VAD-FMK. A p-value <0.05 was considered significant. (C) Optical section of isolated RAW264.7 nuclei stained with an antibody against recombinant NOA1-His6. NOA1 is imported into the nuclei in a GTP-dependent manner. Import is inhibited by addition of the non-hydrolysable GTP analogue GTPγS to the nuclear import assay. (D) Western Blot analysis of RAW264.7 nuclei after incubation with NOA1-His6 in nuclear import assays demonstrates equal loading of experiment and control nuclei with recombinant protein. Scale bars: 10 µm.
Figure 4.
A fraction of NOA1 localizes in the nucleolus and interacts with UBF1.
(A) Immunofluorescence staining of endogenous NOA1 in primary mouse myofibers reveals that a fraction of NOA1 is localized in the nucleolus. (B) Quantitative evaluation of confocal images of NIH 3T3 fibroblasts stained with antibodies against NOA1, UBF1 and Fibrillarin. NOA1 and UBF1 or NOA1 and Fibrillarin are co-localized in the nucleolus. Moving average smoothing was applied to fit data points into curves. (C) Pull-down assays demonstrating interaction of NOA1 and UBF1. Recombinant NOA1-His6 protein was loaded on Ni-NTA beads and mixed with whole cell or nucleoli lysates from C2C12 cells to pull down interacting proteins. (D) Co-immunoprecipitation of NOA1-His6 with endogenous UBF1 from C2C12 lysates. (E) Addition of RNAse H or RNAse A to pull-down assays increased the efficiency of the interaction between NOA1 and UBF1 while addition of DNAse I had no effect. Scale bars: 10 µm, Zoom scale bars: 1 µM.
Figure 5.
Nuclear export of NOA1 is mediated by a Crm1-dependent Nuclear Export Signal (NES).
A) Immunofluorescence staining of endogenous NOA1 in C2C12 cells. One-hour treatment with 10 ng/ml leptomycin-B leads to accumulation of NOA1 in nuclear foci. (B) Nuclear NOA1 is localizes in distinct foci of UBF1-stained nucleoli and accumulates after leptomycin-B (10 ng/ml) and actinomycin D (1 µg/ml) treatment. (C) Amino acid sequences of the nuclear export sequence (NES) of different proteins including NOA1. (D) Different cellular targeting sequences of NOA1 direct cellular localization of EGFP fusion proteins. Leptomycin-B treatment (10 ng/ml) leads to increased accumulation of NLS-NES-EGFP fusion protein in nucleus. Tom20 staining was used to identify mitochondria. Scale bars: 10 µm, zoom scale bar: 1 µM.
Figure 6.
Nuclear localization and the 25 kDa C-terminus is essential for subsequent mitochondrial import of NOA1.
(A) Immunofluorescence staining of different Flag tagged NOA1 proteins transfected into C2C12 cells reveals requirement of the NLS and the C-terminus of NOA1 for mitochondrial import. Deletion of the MTS abolishes mitochondrial localization leading to accumulation of NOA1 in the nucleus. Mutation of the NLS in the ΔMTS mutant confirms that the NLS is necessary for nuclear localization. Mitochondria were identified by co-staining for Tom20. (B) Quantitative evaluation of the subcellular dynamics of wild type and mutant NOA1 protein. Wild type NOA1 protein shows a predominant “Mitochondrial” localization. ΔMTS-NOA1 shows a predominant “nucleus” accumulation. NOA1-NLS mutants show a mixed localization as indicated by the term “aggregated”. Wild type NOA1 is dynamically distributed between nucleus and mitochondria. Leptomycin-B treatment prevents nuclear export and subsequent mitochondrial import of wild type NOA1. Mutation of the NLS prevents nuclear accumulation of NOA1 after leptomycin-B treatment. (C) N-terminal tagging prevents import of NOA1 into mitochondria of NIH 3T3 cells by masking the MTS. (D) A Western blot analysis of N-terminally (HA tag) and C-terminally (Flag tag) modified versions of NOA2 is shown. (E) Deletion of the RNA-binding domain containing C-terminus prevents mitochondrial import in NIH 3T3 cells. Western blot analysis demonstrated strongly reduced processing of the MTS after truncation of the C-terminus. Scale bars: 10 µm.
Figure 7.
NOA1 is a substrate of the mitochondrial matrix protease complex ClpXP.
(A) The C-terminus of NOA1 contains three motifs resembling known ClpX recognition motifs found in proteins of E.coli. (B) Exemplary Western blot analysis revealing increased concentrations of NOA1 in C2C12 cells after mutation of lysin690 and lysin691 to alanines in the proximal ClpX recognition motif. (C) Quantitative analysis of Western blots demonstrating ca. 60% stabilization of the NOA1-KK690,691AA mutant compared to wild type NOA1. Densitometries from three independent experiments were normalized to Porin. (D) Overexpression of ClpX is sufficient to promote degradation of NOA1 independent of a C-terminal Flag tag in C2C12 cell lysates. (E) Recombinant bacterial ClpXP (E.coli ClpXP) (top left) and mammalian ClpXP (mouse ClpX, human ClpP) (top right) cleaves recombinant NOA1 (2.5 µM) in vitro in a time dependent manner. Mammalian ClpXP but not by E. coli ClpXP cleaves α-Casein (5 µM) in vitro. (F) Quantitation analysis of Western blots shown in (E) demonstrating the degradation of NOA1 by bacterial and mammalian ClpXP in vitro.
Figure 8.
Model of the intracellular routing of NOA1.
NOA1 is a nuclear encoded protein translated in the cytosol. The unprocessed precursor NOA1 protein (1) is imported into the nucleus in a NLS dependent manner mediated by the importin system, which requires GTP. (2). NOA1 localizes to the nucleolus and interacts with the UBF1 protein (3). NOA1 binds G-quadruplex RNA, which destabilizes interaction of NOA1 with the UBF1 protein complex followed by NES dependent Crm1 mediated nuclear export (4). Following nuclear export NOA1 is imported into the mitochondrial matrix where the mitochondrial targeting sequence is removed (5). The matrix protease complex ClpXP most likely mediates degradation of NOA1 in mitochondria (6) although the mammalian ClpXP complex was less efficient to degrade NOA1 compared to bacterial ClpXP.