Phosphorylation of mitochondrial transcription factor B2 controls mitochondrial DNA binding and transcription

Abstract

Mammalian cells contain genetic information in two compartments, the nucleus and the mitochondria. Mitochondrial gene expression must be coordinated with nuclear gene expression to respond to cellular energetic needs. To gain insight into the coordination between the nucleus and mitochondria, there is a need to understand the regulation of transcription of mitochondrial DNA (mtDNA). Reversible protein post-translational modifications of the mtDNA transcriptional machinery may be one way to control mtDNA transcription. Here we focus on a member of the mtDNA transcription initiation complex, mitochondrial transcription factor B2 (TFB2M). TFB2M melts mtDNA at the promoter to allow the RNA polymerase (POLRMT) to access the DNA template and initiate transcription. Three phosphorylation sites have been previously identified on TFB2M by mass spectrometry: threonine 184, serine 197, and threonine 313. Phosphomimetics were established at these positions. Proteins were purified and analyzed for their ability to bind mtDNA and initiate transcription in vitro. Our results indicate phosphorylation at threonine 184 and threonine 313 impairs promoter binding and prevents transcription. These findings provide a potential regulatory mechanism of mtDNA transcription and help clarify the importance of protein post-translational modifications in mitochondrial function.

Keywords: Mitochondrial DNA; Mitochondrial transcription factor B2 (TFB2M); Phosphorylation; Transcription regulation.

Figure 1.. TFB2M phosphomimetics display altered binding to mtDNA promoter regions.

A. Representative western blot of LSP binding. WT or phosphomimetic TFB2M (400 nM) was incubated with 1.5 μM biotinylated-LSP dsDNA (−17 to +19) for 15 minutes. Streptavidin conjugated agarose beads were added to capture DNA-protein complexes. After extensive washing, bound protein was eluted by heating with 1x Laemmli sample buffer. Proteins were subject to western blotting using a TFB2M primary antibody and IRDye 800CW labeled secondary antibodies for quantitation. B. Quantitation of western blots. Results are shown relative to WT and were normalized to total protein controls. Data presented are averages and standard deviations of experiments performed with 2 to 4 independent protein preparations, with 2 technical replicates performed in each experiment. C, D. The experiment in A was repeated in the presence of 1.5 μM biotinylated-HSP1 duplex DNA (−25 to +20), and western blots were quantified as in B. *p<0.1, **p<0.05, Student’s t-test.

Figure 2.. Binding dissociation constants for the LSP determined by fluorescence anisotropy confirm altered binding affinity of TFB2M phosphomimetics.

A. 5’-fluorescein labeled LSP (−17 to +19, 3 nM) was titrated with increasing concentrations of TFB2M WT or phosphomimetic. Anisotropy was measured and background corrected. Anisotropy values from two independent experiments were averaged and plotted. The average binding data were fit to determine binding dissociation constants (K_d), tabulated in (B). The error is the standard error of the fit.

Figure 3.. Transcription initiation activity of TFB2M phosphomimetics.

A. The sequence of LSP DNA used in the experiment is shown. B, C. The denaturing gel image identifies abortive products (2- to 6-mer) and run-off products (18- and 19-mer) of transcription reactions performed for 30 minutes at 25°C. Reactions were ca rried out in the presence of NTPs (250 μM each of ATP, GTP, UTP, dCTP, and spiked with γ-³²P-ATP) and 1 μM each of POLRMT, TFB2M, LSP DNA, and in the absence (B) or presence (C) of TFAM. D, E. Quantitation of run-off products (D) and abortive products (E). Error bars represent the percent error between two technical gel replicates of the same reaction samples. A comparison of the ratio of run-off versus abortive products indicates transcription initiation efficiency is unchanged across samples.

Figure 4.. Structural model to explain the role of phosphorylation of TFB2M in transcription initiation.

A. Crystal structure (PDB 6ERP [10]) of human TFB2M (blue) in complex with open LSP mtDNA (blue/teal) and POLRMT (wheat). Phosphorylation sites are labeled. TFAM was removed for clarity. B. Close up view of TFB2M helix α8 (shown in gray) and key amino acids required for mtDNA binding and transcription initiation. Threonine 313 lies at the N-terminal end of this helix and may cause structural changes leading to movements of R330/331 and H326 that are unfavorable for mtDNA binding and transcription. C. T184 and T313 lie within 10–12Å of the thumb domain of POLRMT (shown in green), suggesting a negative impact of modification of these amino acids on POLRMT-TFB2M interactions (PDB 4BOC [22]).