Despite having its own DNA, the mitochondrion imports the majority of its proteins from the nucleus which, thus, controls the organellar activities, and in a reciprocal manner mitochondria send signals to the nucleus resulting in a coordination of both genomes. Although mitochondrial function depends on the nuclear genome, the organelle still plays a pivotal role in energy production, cell proliferation, maintenance and death. The functional coexistence of proteins encoded by the two separate genomes requires an efficient coordination of gene expression ensuring correct mitochondrial function upon cell specific needs. The mitochondrianucleus networking includes interconnection between anterograde (nucleus to mitochondria) and retrograde (mitochondria to nucleus) signals. The anterograde system consists of signals generated from the nuclear genome, in response to endogenous and exogenous stimuli, and transmitted to mitochondria. The retrograde system links mitochondrially originated stimuli to coregulation of nuclear gene expression.
Nevertheless, anterograde and retrograde signalling are still poorly understood in mammals, if better characterized in yeast. The expression of most of the eukaryotic nuclear genes is controlled at the level of transcription through transacting regulatory proteins (transcription factors) which can recognize and bind to short specific nucleotide sequences (cisacting binding sites) within a promoter, and consequently can modulate (activate or suppress) the expression of genes. Despite bidirectional promoters controlling the transcription of around 11% of mammalian genes, so far only few of them have been studied and properly characterized. I addressed my research study to mapping the bidirectional promoter for the Mrps12/Sarsm genes. Mrps12 (mitochondrial ribosomal protein s12) and Sarsm (seryl tRNA synthetase) are nuclear genes encoding components of the mitochondrial translational apparatus. In my study, I identified an array of four CCAAT boxes, all with the same orientation which, interacting with NFY, regulate the human and mouse Mrps12/Sarsm bidirectional promoter activities. The NFY involvement in Mrps12/Sarsm promoter transcriptional regulation was confirmed using a dual luciferase reporter vector in combination with EMSA and Chip. The NFY yeast counterpart, Hap2/3/5, was previously shown to be the main activator of transcription of nuclear genes involved in mitochondrial biogenesis and OXPHOS. Also in humans, NF Y was demonstrated to be important in regulating mitochondrial function, as shown elsewhere by the rescue of the pathological cytochrome c oxidase assembly mutant in the SURF1 gene in human fibroblasts, obtained by overexpressing NFY subunits. Due to the particular bidirectional property of NFY to recognize its core binding site sequence in either orientation, i.e. as CCAAT or as ATTGG on the coding strand, I hypothesize that it could be a suitable factor to govern bidirectional promoters of this class. Inverting the core binding site from CCAAT to ATTGG, I demonstrated that the CCAAT box orientation of the Mrps12/Sarsm promoter is not important since, the overall promoter activity changed only minimally and transcriptional directionality was maintained. The CCAAT core binding site can be recognized by several multiprotein complex factors; including NFY and members of the C/EBP transcription factor family (CCAAT/Enhancer Binding Protein). I demonstrated with EMSA, that C/EBP? could also interact with the CCAAT boxes of the Mrps12/Sarsm promoter. Bioinformatic analyses of the CCAAT box, within the mouse or human genomes, revealed a preferential distribution in favour of bidirectional promoters over unidirectional for both NFY type CCAAT boxes or in combination with C/EBPtype CCAAT boxes, especially in those promoters presenting multiple CCAAT box sites. In mammals, the nuclear respiratory factors, NRF1 and NRF2, were previously shown to have a key role in the regulation of nuclear genes encoding components of the mitochondrial oxidative phosphorylation (OXPHOS) system, thus linking mitochondrial function to the cell’s energy demand. However, mutations affecting the putative binding site for NRF2 in the Mrps12/Sarsm promoter produced only a small change in the transcriptional activity in both mouse and human. To study how the bidirectional promoter for Mrps12/Sarsm genes can be affected by mitochondrial dysfunction, I established a cell culture model of mitochondrial stress.
Several types of human and mouse cells were treated with toxins which differently affect mitochondrial function, such as inhibitors of mitochondrial protein synthesis, and agents that bring about uncoupling or respiratory chain inhibition. Mitochondrial stress produced multiple effects on promoter activity in the different cell lines I tested and at different times of exposure. High drug doses and/or a prolonged drug exposure generally suppressed transcriptional activity in the tested celltypes: mouse NIH 3T3 or C2C12 myoblasts, human HEK293 cells or U2OS or Hela. However, a shorter drug treatment stimulated the promoter activity in mouse 3T3 or human Hela cells. ROS seemed to be part of the signaling, since the potency of different drugs in producing the transcriptional response was well correlated with the amount of ROS production. HEK 293 and 3T3 cells gave a similar increase in ROS for the same mitochondrial stress induction although exhibiting opposite promoter modulation, which was stimulated in 3T3 cells or suppressed in HEK 293. The array of the four CCAAT boxes was not directly involved on Mrps12/Sarsm promoter stimulation under OXPHOS stress. However, transcriptional downregulation under prolonged mitochondrial stress was CCAAT boxdependent.
In conclusion: the complex mechanism of gene expression modulationcoregulation in the nuclei and the reciprocal tight communication mitochondria and the nucleus are vital for cell biogenesis, proliferation, death and adaptation to endogenous and exogenous stimuli. A clear understanding of the mechanisms by which nucleusmitochondrial signalling occurs, and the pathways by which perturbations are signaled, may allow knowledge at the molecular level of human diseases and thus permit a specific therapeutic intervention.
