UMD Undergraduate Research Journal

Fluorescent Polarization Assays Quantify Human Mitochondrial Transcription Factor A DNA Binding

Deryck Pearson, Fernanda Lodeiro, Akira Uchida, and Craig E. Cameron


Human mitochondrial transcription factor A (TFAM) is a DNA binding protein re- sponsible for transcription activation in mitochondria. It is also required for DNA compaction into nucleoids and DNA maintenance. TFAM belongs to the high mobil- ity group (HMG) superfamily, containing two HMG boxes and an additional carboxy- terminal domain. To date, quantitative studies addressing DNA binding specificity have not been performed. A detailed understanding of basic events underlying the specific DNA binding of TFAM as a transcription factor and its role in nucleoid formation has yet to be achieved. The purpose of this study is to develop an assay to further understand what these determinants are, in regards to specific DNA binding. We have used, in the past, fluorescence polarization to study DNA binding. In these assays, increasing concentrations of TFAM are incubated with fluorescently labeled DNA, and the change in polarization is utilized to infer DNA binding affinity. Here, we discuss results obtained with mutant TFAM carrying deletions in the carboxy terminal domain. Our results suggest a key role of this domain in DNA binding. We also describe the development of an alternative approach in which the fluorescent label is located on the protein. This approach may permit the characterization of TFAM binding to various DNA sequences and gain insight into the specificity of DNA binding.


Mitochondria are the powerhouses of the cell. Their main role is the production of energy for cellular function through oxidative phosphorylation. Mitochondria also play a role in fatty acid oxidation, metabolism of calcium and iron, biosynthesis of amino acids and heme, and apoptosis [1]. Mitochondria have their own genome that encodes information essential for the expression of components of oxidative phosphorylation including, mRNA, tRNA and rRNA, which are necessary for protein synthesis in organello [2]-[4]. It is very well established that mutations in the genes that encode components of the oxidative phosphorylative system can lead to mitochondrial diseases, typically with manifestations that affect the main energy-consuming organs, such as the heart, 47 brain and muscles. In recent years, it has been recognized that several human diseases, such as cancer, diabetes, Parkinson's disease, dementia, and diabetes mellitus, all have a correlation with defective mitochondrial function [5]-[10]. One hypothesis explaining this phenomenon is that mitochondrial DNA (mtDNA) can accumulate mutations over the life span of the mitochondria. An underlying reason why mtDNA is highly susceptible to mutation is because it lacks efficient DNA repair mechanisms and it functions in a highly oxidative environment.

Figure 1: Model of TFAM-DNA binding and mitochondrial transcription initiation. First, TFAM binds the LSP promoter, forming a complex. The complex changes the DNA conformation, allowing the recruitment of POLRMT and TFB2M to initiate transcription [11].

Transcription is the process of converting information stored in the form of DNA into RNA. This then allows that information to be converted into proteins for a wide array of functions. Only a single regulatory region, called a displacement loop (D-loop), is contained in mtDNA. Transcription in mitochondria starts from three promoters: the light strand promoter (LSP) and heavy strand promoters 1 and 2 (HSP1 and HSP2). Transcription from the most studied promoter, LSP, requires three protein components: mitochondrial RNA polymerase (POLRMT), mitochondrial transcription factor A (TFAM) and mitochondrial transcription factor B (TFB2M) (Figure 1). We hypothesize that mutations within the mtDNA regulatory region, particularly within the promoters, may affect TFAM binding and/or POLRMT/TFB2M recruitment to the promoter. Thus transcription initiation, and therefore gene expression, may ultimately lead to impairment of mitochondrial function.

TFAM is a member of the high mobility group (HMGB) superfamily of DNA binding proteins defined by the HMG DNA binding domain (HMG box). Proteins from this fold are classified based on DNA binding: sequence-specific through promoter regions and nonsequence-specific based on the physical structure of DNA.

TFAM has multiple functions. First, it can act as a transcription factor by binding to a specific sequence in the LSP promoter upstream of the initiation site [11]. In addition, it has been shown that TFAM can activate transcription from the HSP1 promoter. The activation is enhanced by the presence of the LSP promoter and takes place at concentrations in which LSP transcription is inhibited [12]. Moreover, TFAM participates in mtDNA packaging into nucleoids and mtDNA maintenance [13], [14]. The protein determinants for each of these functions have not been elucidated.

TFAM has two HMG-box domains, separated by a short link. In addition, it has a carboxy-terminal domain of 26 Figure 1: Model of TFAM-DNA binding and mitochondrial transcription initiation. First, TFAM binds the LSP promoter, forming a complex. The complex changes the DNA conformation, allowing the recruitment of POLRMT and TFB2M to initiate transcription [11].amino acids [15]. This domain has been implicated in transcription initiation since its deletion abolishes transcription [16]. However, the mechanism underlying this observation has not been described. While it is known where TFAM binds in LSP, it has not been studied in great detail which base pairs are responsible for this interaction and whether mutations in this region in human mtDNA can affect transcription initiation. Here we report the use of fluorescence polarization to study TFAM and DNA interaction. By tagging the DNA we can quantitatively probe the necessary TFAM structures for binding. Then, by tagging TFAM, we should be able to find the target DNA sequences for TFAM binding.

Materials and Methods

Materials: Purified TFAM, TFAMdCT, TFAMd10CT, and TC-TFAM were previously prepared in the Cameron laboratory. DNA oligonucleotides were obtained from Integrated DNA Technologies Inc. and purified by PAGE gel electrophoresis. BSA was obtained from New England Biolabs. The Biarsenical Fluorescein derivative FlAsH was obtained from Invitrogen. All other reagents were received from VWR and Fisher.

Annealing: The non-template strand, 5'-ATGTGTTAGTTGGGGGGT-GACTGTTAAA-Fl-3', where Fl means fluorescein, and template strand (reverse complimentary strand) were annealed at 25 µM in 10 mM Tris-HCl, pH 8.0, 1 mM EDTA, and 50 mM NaCl by using a Progene Thermocycler (Techne). The annealing solution was then heated at 90°C for 1 minute, then cooled at a rate of 5°C per minute until it reached 10°C.

TFAM Titration for TFAM-DNA Binding: The reaction mixture for anisotropy for measurement was prepared as follows. TFAM was serially diluted in Enzyme Dilution Buffer (10 mM HEPES, pH 7.5, 20% Glycerol, 100 mM NaCl and 1mM DTT) to acquire the appropriate concentrations. Fluorescein-labeled DNA was diluted in annealing buffer. The DNA binding reaction contained 10 mM HEPES, pH 7.5, 10 mM MgCl2, 1 mM DTT, 0.1 µg/µL BSA, 100 mM NaCl, 0.1 nM DNA, and various concentrations of protein typically in 100 µL. The samples were then transferred to glass tubes and incubated at 25°C for approximately 30 sec before obtaining the mini polarization (mP) values.

Plots of the change in mP as a function of TFAM concentration were used to determine the equilibrium dissociation constant (Kd) for the interaction between TFAM and LSP binding site. The data was fit to a hyperbola (Eq. 1) using the program KaleidaGraph (Synergy Software, Reading, PA). This procedure was used for wild-type TFAM, TFAMdCT and TFAMd10CT, two other TFAM variants.


Figure 2: Experimental design to study protein-DNA interaction by FP. a- A fluorescent labeled DNA is incubated with increasing concentrations of TFAM. This approach permits the study of determinants for specific binding in the protein. b- A fluorescently labeled TFAM derivative is incubated with increasing concentrations of DNA. This approach permits the study of different DNA sequences and dissection of determinants for specific binding within the DNA.

TFAM-DNA Binding Characterization using fluorescence polarization (FP): To study TFAM target site we used fluorescence polarization (FP). The advantages of fluorescence polarization include that it is fast, reproducible, reports on the equilibrium in solution and does not require radioactivity materials. FP permits the study of molecular interactions by tracking size changes of fluorescent molecules. The polarization depends on the apparent size of the fluorescent molecule. As the fluorescent molecule is incubated with its binding partner, binding occurs. The complex formed has a higher apparent size, which polarizes the emitted light. For protein-DNA interactions, the fluorophor can be incorporated in the protein or in the DNA (Figure 2).

To begin our studies we performed FP assays using a fluorescently labeled TFAM binding site and titrating increasing concentrations of TFAM. The data was plotted to a hyperbola and the Kd obtained was 1 nM (Figure 3). This indicates tight binding here.

Figure 3: TFAM-DNA Binding Fluoresce- in-labeled DNA was used to titrate TFAM concentrations ranging from 0.03125 nM to 500 nM. The data points were plotted using Eq 1 to determine the Kd.

TFAM carboxy terminal domain is important for DNA binding: As mentioned above TFAM contains two HMG boxes and a carboxy-terminal domain (CTD) that have been implicated in transcription initiation. In order to determine the function of the carboxy-terminal domain we used a protein variant in which this domain was deleted (TFAMdCT). We performed the FP assay as described above, titrating TFAMdCT. We found that the deletion of the CT domain greatly decreased the binding affinity (30 fold). The Kd value is 31 +/- 4 nM (Figure 4). This result explains the observation that the CT domain fails to activate transcription and suggests that this domain is essential for high-affinity DNA binding [16].

We observed that within the CTD (amino acids 220 to 246) there are 16 amino acids that are highly conserved between mammalian species, while the last 10 amino acids differ (Figure 5). We wanted to know whether these last 10 amino acids are as critical as the entire CTD for DNA binding. We used FP to study the binding of TFAMd10CT to the fluorescently labeled DNA as described previously. We found that the binding affinity was moderately affected, with a Kd value of 9 nM between the full length (TFAM) and the CTD deletion (TFAM-dCT). TFAMd10CT was able to support transcription in our in vitro assays. The Last 10 amino acids contribute to DNA binding by 3-fold, while the whole CTD contributes 30-fold.

Figure 4: The carboxy terminal domain of TFAM is essential for specific DNA binding. a - Schematic of TFAM domains. b - TFAMdCT-DNA Binding. Fluorescein-labeled DNA was incubated with increasing concentrations of TFAMdCT, ranging from 1 nM to 1000 nM. The data points were plotted using Eq 1 to measure the Kd.

Figure 5: The carboxy terminal domain of TFAM is essential for specific DNA binding. a - Sequence alignment of TFAM from human, mouse and bovine species. b - TFAMd10CT-DNA Binding. Fluorescein-labeled DNA was incubated with increasing concentrations of TFAMdCT, ranging from 1 nM to 1000 nM. The data points were plotted using Eq 1 to measure the Kd.

TC-TFAM-DNA Binding Characterization: As depicted in Figure 2b, to study the determinants of binding specificity within the DNA, it is useful to incorporate the fluorescent label into the protein. To do so, we introduced a tetra cystein (TC) motif in the N-terminal region of TFAM. The TC motif can then conjugate the fluorescent dye, FlAsH. It was important to confirm that TFAM function has not been affected by this modification. This was also determined using FP in which TC-TFAM was titrated.

TC-TFAM binds with relatively tight affinity with a Kd of 3 nM (Figure 6). This result is comparable to the one obtained with wild-type TFAM and indicates that the addition of the TC motif does not impact DNA binding. This finding is in agreement with our results, showing that TC-TFAM functions as TFAM in transcription assays in vitro.


Figure 6: TC-TFAM-DNA Binding. Fluorescein-labeled DNA was used to titrate TC-TFAM concentrations ranging from 0.0625 nM to 200 nM. The data points were plotted using Eq 1 to measure the Kd. The Kd was found to be 3 +/- 0.3 nM.

The present work illustrates the use of FP to study TFAM binding to its DNA cognate. Here we showed that the carboxy-terminal domain has a critical role in DNA binding, explaining previous observations that TFAMdCT does not support transcription.

This approach can be useful to address the impact of each amino acid within TFAM on specific binding. In relation to human disease, this assay represents a fast and accurate measurement to address whether mutations within TFAM that can be observed in human populations will have an effect on DNA binding that could impair transcription and cause disease.

On the other hand, the finding that the derivative TC-TFAM maintains its DNA binding properties is very promising. An approach based on fluorescently labeled TC-TFAM should permit a detailed study of determinants for specific binding within the DNA, as well as testing mutations that may be present in human populations within the TFAM binding site. Utilization of this technique can hopefully provide a new tool to better investigate human mitochondrial diseases.


[1] N. D. Bonawitz et al., "Initiation and beyond: multiple functions of the human mitochondrial transcription machinery," Mol Cell, vol. 24, no. 6, pp. 813-825, 2006.

[2] D. A. Clayton, "Transcription and replication of mitochondrial DNA," Hum Reprod, vol. 15, supp. 2, pp. 11-17, 2000.

[3] G. S. Shadel, "Coupling the mitochondrial transcription machinery to human disease," Trends Genet, vol. 20, no. 10, pp. 513-519, 2004.

[4] M. Falkenberg et al., "DNA replication and transcription in mammalian mitochondria," Annu Rev Biochem, vol. 76, pp. 679-699, 2007.

[5] D. C. Wallace, "Mitochondrial DNA mutations in diseases of energy metabolism," J Bioenerg Biomembr, vol. 26, no. 3, pp. 241-250, 1994.

[6] D. C. Wallace, Mitochondrial defects in neurodegenerative disease," Ment Retard Dev Disabil Res Rev, vol. 7, no. 3, pp. 158-166, 2001.

[7] D. C. Wallace, A mitochondrial paradigm for degenerative diseases and ageing," Novartis Found Symp, vol. 235, pp. 247-266, 2001.

[8] D. C. Wallace, A mitochondrial paradigm of metabolic and degenerative diseases, aging, and cancer: a dawn for evolutionary medicine," Annu Rev Genet, vol. 39, pp. 359-407, 2005.

[9] D. C. Wallace, "Mitochondria and cancer: Warburg addressed," Cold Spring Harb Symp Quant Biol, vol. 70, pp. 363-374, 2005.

[10] D. C. Wallace et al., "Mitochondrial DNA mutations in human degenerative diseases and aging," Biochim Biophys Acta, vol. 1271, no. 1, pp. 141-151, 1995.

[11] M. F., Lodeiro et al., "Identification of multiple rate-limiting steps during the human mitochondrial transcription cycle in vitro," J Biol Chem, vol. 285, no. 21, pp. 16387-16402, 2010.

[12] T. E. Shutt et al., "Core human mitochondrial transcription apparatus is a regulated two-component system in vitro," Proc Natl Acad Sci USA, vol. 107, no. 27, pp. 12133-12138, 2010.

[13] R. P. Fisher et al., "DNA wrapping and bending by a mitochondrial high mobility group-like transcriptional activator protein," J Biol Chem, vol. 267, no. 5, pp. 3358-3367, 1992.

[14] B. A. Kaufman et al., "The mitochondrial transcription factor TFAM coordinates the assembly of multiple DNA molecules into nucleoid-like structures," Mol Biol Cell, vol. 18, no. 9, pp. 3225-3236, 2007.

[15] T. A. Gangelhoff et al., "Structural analysis and DNA binding of the HMG domains of the human mitochondrial transcription factor A," Nucleic Acids Res, vol. 37, no. 10, pp. 3153-3164, 2009.

[16] D. J. Dairaghi et al., "Addition of a 29 residue carboxyl-terminal tail converts a simple HMG box-containing protein into a transcriptional activator," J Mol Biol, vol. 249, no. 1, pp. 11-28, 1995.