Tetramethylpyrazine blocks TFAM degradation and up-regulates mitochondrial DNA copy number by interacting with TFAM

Abstract

The natural small molecule compound: 2,3,5,6-tetramethylpyrazine (TMP), is a major component of the Chinese medicine Chuanxiong, which has wide clinical applications in dilating blood vessels, inhibiting platelet aggregation and treating thrombosis. Recent work suggests that TMP is also an antitumour agent. Despite its chemotherapeutic potential, the mechanism(s) underlying TMP action are unknown. Herein, we demonstrate that TMP binds to mitochondrial transcription factor A (TFAM) and blocks its degradation by the mitochondrial Lon protease. TFAM is a key regulator of mtDNA replication, transcription and transmission. Our previous work showed that when TFAM is not bound to DNA, it is rapidly degraded by the ATP-dependent Lon protease, which is essential for mitochondrial proteostasis. In cultured cells, TMP specifically blocks Lon-mediated degradation of TFAM, leading to TFAM accumulation and subsequent up-regulation of mtDNA content in cells with substantially low levels of mtDNA. In vitro protease assays show that TMP does not directly inhibit mitochondrial Lon, rather interacts with TFAM and blocks degradation. Pull-down assays show that biotinylated TMP interacts with TFAM. These findings suggest a novel mechanism whereby TMP stabilizes TFAM and confers resistance to Lon-mediated degradation, thereby promoting mtDNA up-regulation in cells with low mtDNA content.

Keywords: Lon protease; Tetramethylpyrazine; mitochondria; mitochondrial DNA; mitochondrial transcription factor A.

Figure 1. TMP blocks Lon-mediated TFAM degradation

(A) Chemical structure of TMP. (B) Western blot showing levels of Lon and TFAM. Purified Lon (300 nM) was reacted with purified TFAM (150 nM) in the presence of ATP (5 mM) for 1 h. Velcade was used as the positive control. (C) Western blot showing levels of Lon, actin and TFAM. HeLa ρ⁺ cells were transfected with plasmids for expressing TFAM^HMG1/2 or TFAM^wt transiently. After transfection, cells were treated with TMP (10 μM) while chased by CHX (100 μg/ml) for the next 4 h. (D) Western blot showing levels of Lon, Actin and TFAM. EC-1 cells were transfected with plasmids for expressing TFAM^HMG1/2 or TFAM transiently. After transfection, cells were treated with TMP (10 μM), while chased by CHX (100 μg/ml) for the next 4 h (endoTFAM, the endogenous TFAM).

Figure 2. TMP up-regulates TFAM and mtDNA copy number in HeLa cells

(A) Shown was Western blot analysis of Lon, TFAM and actin from HeLa cells treated with EB for 8 days, cells subsequently grown in the absence of EB (mtDNA recovery) and in the presence of TMP (10 μM) for 1, 3 and 4 days, DMSO (10 μM) was a negative control. (B) Quantificative analysis of data from (A) (see ‘Materials and methods’ section). Plotted on the ordinate was the relative expression of TFAM to actin at each time point during the depletion–repletion experiment. Specific time points in the depletion–repletion experiment were labelled as in (A) above. Data represent the mean value ± S.D. from six separate experiments, *P<0.05, **P<0.01. (C) Shown was the analysis of mtDNA from the samples described in (A) above. (D) Shown was Western blot analysis of Lon, actin and TFAM from HeLa cells with severe mtDNA deficits treated with nothing (lane 1), DMSO (lane 2) or different concentrations of TMP, including 2.5 μM (lane 3), 5 μM (lane 4), 10 μM (lane 5).

Figure 3. TMP does not inhibit Lon protease activity in vitro

(A) Purified Lon (300 nM) was pre-incubated in the presence of TMP at the indicated concentrations for 1 h at 25°C, after which ATP (5 mM) were added and incubated for 1 h at 25°C before fluorescence values were measured. (B) Purified Lon (300 nM) was pre-incubated in the presence of TMP at the indicated concentrations, after which ATP (4 mM) and the AA2-Rh110 (6 μM) were added and incubated for 3 h at 37°C before fluorescence values were measured. (C) Coomassie Brilliant Blue staining showing the effect of TMP on Lon-mediated degradation of casein. DMSO worked as a negative control and Velcade worked as a positive control respectively. (D) The intensities of reaction bands were analysed with the ImageJ software.

Figure 4. TMP interacts directly with TFAM

(A) Chemical structure of biotinylated TMP. (B) HeLa ρ⁺ cell extracts were reacted with or without TMP or biotinylated-TMP followed by incubation with SA-sepharose, pull-down reactions were immunoblotted with Lon and TFAM. Untreated mock samples contained input but without biotinylated-TMP and SA-sepharose. (C) HCT116 cell extracts were reacted with or without TMP or biotinylated-TMP followed by incubating with SA-sepharose, pull-down reactions were immunoblotted with Lon and TFAM. Untreated mock samples contained input but without biotinylated-TMP and SA-sepharose. (D) Biotinylated-TMP was reacted with purified Lon and biotinylated-TMP was pulled down with SA-sepharose.

Figure 5. Schematic diagram of TMP blocks Lon-mediated TFAM degradation