1. Chromosomes and Gene Expression

UV irradiation alters TFAM binding specificity and compaction of DNA

  1. Dillon E King
  2. Emily E Beard
  3. Matthew J Satusky
  4. Alex George
  5. Ian Ryde
  6. Caitlin Johnson
  7. Emma L Dolan
  8. Yuning Zhang
  9. Wei Zhu
  10. Hunter Wilkins
  11. Evan Corden
  12. Susan K Murphy
  13. Dorothy A Erie  Is a corresponding author
  14. Raluca Gordân  Is a corresponding author
  15. Joel Meyer  Is a corresponding author
  1. Nicholas School of the Environment, Duke University, United States
  2. Department of Chemistry, University of North Carolina at Chapel Hill, United States
  3. Renaissance Computing Institute, University of North Carolina at Chapel Hill, United States
  4. Department of Pharmacology, Duke University, United States
  5. Department of Obstetrics and Gynecology, Duke University Medical Center, United States
  6. Department of Biostatistics and Bioinformatics, Duke University, United States
  7. Department of Molecular Genetics and Microbiology, Duke University, United States
  8. Program in Computational Biology and Bioinformatics, Duke University, United States
  9. Lineberger Comprehensive Cancer Center, University of North Carolina at Chapel Hill, United States
  10. Department of Computer Science, Duke University, United States
  11. Department of Genomics and Computational Biology, University of Massachusetts Chan Medical School, United States
https://doi.org/10.7554/eLife.108862.3
Share this article
Cite this article
  1. Dillon E King
  2. Emily E Beard
  3. Matthew J Satusky
  4. Alex George
  5. Ian Ryde
  6. Caitlin Johnson
  7. Emma L Dolan
  8. Yuning Zhang
  9. Wei Zhu
  10. Hunter Wilkins
  11. Evan Corden
  12. Susan K Murphy
  13. Dorothy A Erie
  14. Raluca Gordân
  15. Joel Meyer
(2026)
UV irradiation alters TFAM binding specificity and compaction of DNA
eLife 14:RP108862.
https://doi.org/10.7554/eLife.108862.3
  2. Download BibTeX
  3. Download .RIS
6 figures, 1 table and 3 additional files

Figures

Figure 1 with 3 supplements
Download asset Open asset
Live-cell imaging 24 hr following ultraviolet-C (UVC) treatment indicates increased mitochondrial DNA (mtDNA) degradation.

(A) Representative images for control (left) and UVC-treated (right) cells. Merged channels include mitotracker (red, stains mitochondria), lysotracker (pink, stains lysosomes), picogreen (green, stains mtDNA), and nuclei (blue). (B) Inset outlined in the UVC-treated cells in panel A. Channels from left to right include mitochondria, mtDNA, lysosome, and the merged image. The yellow arrow indicates a mtDNA spot colocalized with lysotracker. For panels C-F, x-axes represent the treatment, and data was analyzed via two-tailed unpaired t-test. Data includes at least n=30 cells per treatment group per imaging session and three distinct imaging sessions. (C) Quantification of the total mitochondrial area per cell normalized to the size of the cell (p<0.0001). (D) Quantification of the number of lysosomes normalized to the size of the cell (p=0.0054). (E) Quantification of the number of mtDNA spots normalized to the size of the cell (p<0.0001). (F) Quantification of the proportion of mtDNA spots within lysosomes (p<0.0001).

Figure 1—figure supplement 1
Download asset Open asset
Transcription Factor A, Mitochondrial (TFAM) roles in regulating the mitochondrial genome.

(A) TFAM is one of the dominant proteins of the mitochondrial nucleoid and has been shown in vitro to be sufficient to coat and compact the genome. Compaction status of the mitochondrial nucleoid is associated with the activity of the nucleoid wherein loose and open genomes are accessible to polymerases and protein complexes required for transcription and replication, while compacted and closed-off genomes are inaccessible and inactive. (B) In its role as a transcription factor, TFAM serves to initiate both transcription and replication. TFAM binds upstream of one of the transcription start sites and recruits mitochondrial RNA polymerase (POLRMT). Transcription factor B2 mitochondrial (TFB2M) is then recruited to form the initiation complex. Synthesis of RNA by POLRMT then proceeds with transcription elongation factor mitochondrial (TEFM). Transcription of genome-length RNA transcripts can then occur. Production of a short RNA primer by POLRMT allows for the replication of the mitochondrial genome by the replicative polymerase γ, comprising the catalytic subunit (POLG) and its accessory subunit (POLG2). Created with BioRender.com.

Figure 1—figure supplement 2
Download asset Open asset
Cell viability at 0-, 24-, and 48 hr following exposure to ultraviolet-C (UVC).

Cell viability following exposure to UVC, normalized to the control for each replicate (n=3), determined using a resazurin cell viability assay. The x-axis represents the UVC dose delivered to the cells in J/m2. The y-axis represents the fluorescent readout normalized as percent control. Data was analyzed via two-way ANOVA with a Dunnett’s correction for multiple comparisons to the control for each time point (dose: p=0.004, time: p<0.001, interaction: p=0.002, 24 hr 100 J/m2 p<0.0001, 48 hr 100 J/m2 p<0.0001).

Figure 1—figure supplement 3
Download asset Open asset
Cell area and mitochondrial morphology analysis from live-cell imaging experiments.

For all panels, x-axes represent the treatment, and data was analyzed via two-tailed unpaired t-test. Data includes at least n=30 cells per treatment group per imaging session and three distinct imaging sessions. (A) Quantification of the total area per cell (p<0.0001). (B) Quantification of the mean area of individual mitochondria within each cell (p=0.84). (C) Quantification of the mean perimeter of individual mitochondria within each cell (p=0.45). (D) Quantification of the mean eccentricity values of individual mitochondria within each cell (p=0.02).

Figure 2 with 1 supplement
Download asset Open asset
Ultraviolet-C (UVC) causes mitochondrial DNA (mtDNA) damage that decreases over time and is associated with upregulation of mtDNA replication genes, in the absence of apparent mitochondrial dysfunction.

For panels A-F, x-axes represent recovery time, i.e., time following the UVC exposure. Doses of UVC used were 0, 10, and 30 J/m2. (A) Mitochondrial DNA damage levels following UVC exposure. The y-axis represents the level of damage (lesions/10 kb). Data analyzed via two-way ANOVA with a Dunnett’s post-hoc test for multiple comparisons (dose: p<0.0001, time: p=0.0001, interaction: p=0.0024). For panels B-E, y-axes represent fold change normalized to the control (0 J/m2) at each time point. All data was analyzed via two-way ANOVA with Dunnett’s post-hoc test for multiple comparisons. (B) TFAM expression level assessed via qPCR following UVC exposure (dose: p=0.003, time: p<0.0001, interaction: p=0.01). (C) POLG expression level assessed via qPCR following UVC exposure (dose: p=<0.0001, time: p<0.0001, interaction: p=0.001). (D) POLRMT expression level assessed via qPCR following UVC exposure (Dose: p=0.002, time: p=0.008, interaction: p=0.05). (E) ND-1 expression level assessed via qPCR following UVC exposure (Dose: p<0.0001, time: p=0.01, interaction: p=0.002). (F) Mitochondrial membrane potential assessed via flow cytometry following exposure to 0, 10, 30, or 50 J/m2 UVC at 6 and 24 hr after exposure. The x-axis represents the exposure group and time point, and the y-axis represents the change in Median Fluorescent Intensities (MFI) of tetramethylrhodamine, methyl ester (TMRM) normalized to the control for each time point. Cells were also exposed to FCCP, a well-known chemical that causes mitochondrial depolarization, as a positive control. Data was analyzed via a two-way ANOVA (treatment: p=0.0008, time: p<0.0001, interaction: p=0.14). (G) Cellular ATP levels following exposure to 0, 10, 30, or 50 J/m2 UVC at 6 and 24 hr after exposure. The x-axis represents the exposure group and time point, and the y-axis represents the ATP content (μM). Data was analyzed via a two-way ANOVA (treatment: p=0.38, time: p<0.001, interaction: p=0.94).

Figure 2—figure supplement 1
Download asset Open asset
Ultraviolet-C (UVC) exposures alter gene expression of PGC1α and NRF-1.

For all both panels, x-axes represent recovery time, i.e., time following the UVC exposure, and y-axes represent fold change normalized to the control (0 J/m2) at each time point. Doses of UVC used were 0, 10, and 30 J/m2. All data was analyzed via two-way ANOVA with Dunnett’s post-hoc test for multiple comparisons. (A) PGC1α expression level assessed via qPCR following UVC exposure (dose: p<0.0001, time: p<0.0001, interaction: p<0.0001). (B) NRF-1 expression level assessed via qPCR following UVC exposure (dose: p<0.0001, time: p=0.0022, interaction: p<0.0001).

Figure 3 with 11 supplements
Download asset Open asset
Transcription Factor A, Mitochondrial (TFAM) has specific binding across the mitochondrial genome and exhibits a reduction in specificity in the context of ultraviolet-C (UVC)-irradiated DNA.

The median z-score is plotted to the coordinate of the middle nucleotide of the variable mitochondrial region of the sequence for the non-UVC-irradiated chamber containing 30 nM TFAM (A) and the chamber irradiated with 1500 J/m2 UVC containing 30 nM TFAM (B). The gene map of the mitochondrial genome is shown in the center. Z-score variation is color-coded such that positive z-scores associated with high binding are in blue and progressively get lighter as the z-scores get higher. Negative z-scores associated with low binding are in red. Regions highlighted in yellow are the promoter sequences of the mitochondrial genome on the light strand (LSP1 and LSP2), while regions highlighted in pink are the promoter sequences on the heavy strand (HSP1 and HSP2). Plots of these regions can be found in Figure 3—figure supplement 8.

Figure 3—figure supplement 1
Download asset Open asset
Custom DNA library with full coverage of human mitochondrial DNA (mtDNA) genome with a sliding window width of two nucleotides.

(A) Schematic depicting the library generation. Each 60 nt sequence contains a 33 nt variable region from the human mitochondrial genome and a 27 nt primer. To ensure full coverage of the mitochondrial genome, the variable region of the sequence was generated using a sliding window with a width of 2 nt. (B) The custom DNA library was synthesized and double-stranded on a chip. Two of the chambers were subjected to ultraviolet-C (UVC) irradiation to induce UVC-associated lesions. Chambers were incubated with either 30 or 300 nM Transcription Factor A, Mitochondrial (TFAM) and a fluorophore-labeled antibody. The fluorescent signal associated with bound protein for each DNA spot was determined using a microarray scanner.

Figure 3—figure supplement 2
Download asset Open asset
Fluorescence intensity distribution of the bottom 116 non-mitochondrial sequences from a universal DNA-binding array used to calculate z-scores for each chamber.

X-axes represent the normalized median fluorescence intensity values and y-axes represent the count of sequences within each bin for 30 nM Transcription Factor A, Mitochondrial (TFAM) treatment without ultraviolet-C (UVC) irradiation (A), 300 nM TFAM treatment without UVC irradiation (B), 30 nM TFAM treatment with UVC irradiation (C), and 300 nM TFAM treatment with UVC irradiation (D). The red line is the Gaussian fit using the parameters in each plot and the equation below, where mu is the mean and sigma is the standard deviation.

Figure 3—figure supplement 3
Download asset Open asset
Experiments performed at 300 nM Transcription Factor A, Mitochondrial (TFAM) concentrations demonstrate that TFAM has specific binding across the mitochondrial genome and exhibits a reduction in specificity in the context of ultraviolet-C (UVC)-irradiated DNA.

The median z-score is plotted to the coordinate of the middle nucleotide of the variable mitochondrial region of the sequence for the non-UVC-irradiated chamber containing 300 nM TFAM (A) and the UVC-irradiated chamber containing 300 nM TFAM (B). The gene map of the mitochondrial genome is shown in the center. Z-score variation is color-coded such that positive z-scores associated with high binding are in blue and progressively get lighter as the z-scores get higher. Negative z-scores associated with low binding are in red. Regions highlighted in yellow are the promoter sequences of the mitochondrial genome on the light strand (LSP1 and LSP2), while regions highlighted in pink are the promoter sequences on the heavy strand (HSP1 and HSP2).

Figure 3—figure supplement 4
Download asset Open asset
Correlation between z-scores obtained at 30 nM and 300 nM Transcription Factor A, Mitochondrial (TFAM).

The x-axis represents the median z-score for each probe in experiments performed at 30 nM TFAM and the y-axis represents the median z-score for each in experiments performed at 300 nM TFAM. Distributions to the right and above the graphs represent the distribution of z-scores for each concentration. The red line indicates the line of best fit. Data was analyzed via Pearson’s correlation.

Figure 3—figure supplement 5
Download asset Open asset
Across all experiments, GN10G motifs are not enriched in high Transcription Factor A, Mitochondrial (TFAM) occupancy groups.

The y-axis represents the z-score for each sequence, and the x-axis depicts the concentration of TFAM used in each experiment as well as the presence or absence of ultraviolet-C (UVC) exposure. The violin plots are color-coded to represent whether a GN10G motif was present (green, n=7397) or absent (blue, n=939) within each sequence. For all mitochondrial sequences on the array, sequences were classified as having a GN10G motif if the motif was present in the variable region and the data analyzed was the maximum z-score between the two orientations. Data was analyzed using a one-tailed Kolmogorov-Smirnov test where the null hypothesis was sequences containing GN10G motif have higher z-scores than those without.

Figure 3—figure supplement 6
Download asset Open asset
DNAse I footprints across the mitochondrial genome are enriched for sites with high in vitro binding signals for Transcription Factor A, Mitochondrial (TFAM).

(A) For all DNase Genomics Footprinting (DGF) sites reported by Blumberg et al., 2018, that were identified in >90% of analyzed human samples, we selected all 33 bp mitochondrial DNA (mtDNA) probes in our on-chip DNA library that were contained entirely within DGFs. For DGFs shorter than 33 bp, we selected the DNA chip probes that contained the entire DGF. (B) The TFAM DNA-binding signal (i.e. the fluorescence intensity measured on the DNA chip) was significantly higher at probes that overlap footprints (blue) vs. sites outside of footprints (gray); Mann-Whitney test p-value <2.2×10–16. (C) We also performed a randomization test where we shuffled the positions of the DGF sites 1000 times (keeping the number of DGFs and the distribution of their lengths constant). For each randomization, we repeated the analysis and found that in only 2 of 1000 random samples, the mean TFAM binding was at least as large as in the real DGF data (p=0.002).

Figure 3—figure supplement 7
Download asset Open asset
Comparison between fiber-seq high and low Transcription Factor A, Mitochondrial (TFAM) binding regions and our high-density TFAM-DNA binding array data.

For high and low TFAM binding regions reported by Isaac et al., 2024, determined using fiber-seq on linear mitochondrial DNA (mtDNA), we selected all 33 bp mtDNA probes in our on-chip DNA library that were contained entirely within the region. For regions shorter than 33 bp, we selected the DNA chip probes that contained the entire region. (A) For probes within the low binding regions, we observe lower TFAM DNA-binding signal (i.e. the fluorescence intensity measured on the DNA chip) (red), and for probes within the high binding regions, we observe moderate to high levels of TFAM DNA-binding signal (blue) than all other probes that were not within the high and low TFAM regions reported by Isaac et al. Data was analyzed via Mann-Whitney test (all other probes vs. low regions p-value <3.0×10–18; all other probes vs. high regions p-value <8.9×10–05; low regions vs. high regions p-value <2.8×10–13). We also performed a randomization test where we shuffled the positions of both the low (B) and high (C) TFAM binding regions 1000 times (keeping the number of regions and the distribution of their lengths constant). For the low binding regions, only 8 of the 1000 random samples had binding signal less than the mean for the probes within the low regions (p=0.008). However, for the high binding regions, 81 of the 1000 random samples had binding signal higher than the mean for the probes within the high regions (p=0.081).

Figure 3—figure supplement 8
Download asset Open asset
High-resolution view of Transcription Factor A, Mitochondrial (TFAM) binding to promoter sequences and shift in z-scores following ultraviolet-C (UVC)-irradiation.

For panels A-D, the median z-score is plotted to the coordinate of the middle nucleotide of the variable mitochondrial region of the sequence for the non-UVC-irradiated chamber containing 30 nM TFAM (top panels, in blue) and the UVC-irradiated chamber containing 30 nM TFAM (bottom panels, in green). Z-score variation is color-coded such that positive z-scores associated with high binding are in blue and negative z-scores associated with low binding are in red. Panels represent the shaded regions in Figure 3 for LSP1 (A), HSP1 (B), LSP2 (C), and HSP2 (D). Regions highlighted in yellow are LSP TFAM binding sites, regions highlighted pink are HSP TFAM binding sites, and regions highlighted in green are transcription start sites.

Figure 3—figure supplement 9
Download asset Open asset
Ultraviolet-C (UVC)-irradiation is associated with a reduction in Transcription Factor A, Mitochondrial (TFAM) sequence specificity as weakest binders become tighter and tightest binders become weaker.

Panels A and B are kernel density estimates of the probes in the top 5% of z-scores (A) and the bottom 5% of z-scores (B) in the context of non-damaged DNA and UVC-irradiated DNA. X-axes represent the median z-score of the probe and y-axes represent the probability density. Note differences in y-axes between panels A and B. (C) The x-axis represents the median z-score for each probe in experiments performed at 30 nM TFAM and the y-axis represents the change in z-score for each probe between non-UVC and UVC-irradiated experiments. The distribution plot above of the graph indicates the distribution of median z-score values in the non-damaged probes performed at 30 nM TFAM. The distribution plot to the right of the graph indicates the distribution of the difference in z-scores calculated between non-UVC and UVC-irradiated experiments. The red line indicates the line of best fit. Data was analyzed via Pearson’s correlation.

Figure 3—figure supplement 10
Download asset Open asset
Individual anisotropy plots for all sequences tested.

For all graphs, the x-axes represent the Transcription Factor A, Mitochondrial (TFAM) concentration (nM) and the y-axes represent anisotropy. The three lines on each graph represent three replicates performed. Graphs in the top half (white) are the sequences tested without UV exposure. Graphs in the bottom half (gray) represent sequences that were first irradiated with ultraviolet-C (UVC). Sequences can be found in Supplementary file 2. KD values and n values can be found in Table 1.

Figure 3—figure supplement 11
Download asset Open asset
Histogram plots of volumes for Transcription Factor A, Mitochondrial (TFAM) with and without DNA.

(A) Top panel is a histogram plot for the volumes of TFAM protein only at 7.5 nM. The red line is aligned with the peak at 33 nm3 in the TFAM only histogram to help show the shift in peaks for the TFAM with short DNA sequences in the two bottom panels. (B) Middle panel is the volume distribution of a 30 nM TFAM incubated with a final concentration of 10 nM of the low occupancy sequence ND4-473. (C) The bottom panel is the volume distribution of 30 nM TFAM in the presence of 10 nM high occupancy sequence ND1-353.

Figure 4 with 3 supplements
Download asset Open asset
Atomic force microscopy of Transcription Factor A, Mitochondrial (TFAM)-DNA substrates indicates an increase in compaction associated with ultraviolet-C (UVC) exposure.

(A) Atomic force microscopy (AFM) images of the pUC19 DNA only (control) and the 30 nM TFAM-DNA complexes in two different conditions: one with (+UV) or without (-UV) UVC irradiated DNA. Colored arrows on the AFM images represent the different TFAM-DNA complexes categorizations: dispersed (green), intermediate (yellow), punctate (red), and free DNA (blue). The white scale bar represents 1 μm. UV-irradiated plasmids were exposed to 100 J/m2 UVC. (B) Histogram plot of the volumes distribution of plasmid DNA only (control) as well as the 15 nM and 30 nM TFAM-DNA complexes with (+UV) or without (-UV) UVC-damaged DNA. All axes in the histogram are scaled the same. The data for control pUC19 only was replicated for each TFAM concentration for clarity in comparisons. (C) Atomic force microscopy images of three different TFAM-DNA complex categorizations labeled as dispersed (small clusters with no protein tracts on DNA), intermediate (small clusters with protein tracts), and punctate (tightly associated punctate clusters). The white scale bar represents 200 nm. (D) A bar graph representing the percent total number of plasmids in the 15 nM TFAM concentration with (+UV) (N=66) or without (-UV) (N=136) UVC damaged DNA and the 30 nM TFAM concentration with (+UV) (N=65) or without (-UV) (N=91) UVC damaged DNA. Detailed counts of each classification can be found in Supplementary file 1. (E) Schematic of the TFAM-DNA binding and compaction mechanism.

Figure 4—figure supplement 1
Download asset Open asset
Area quantification of undamaged and UV-irradiated pUC191341 plasmids using atomic force microscopy (AFM).

(A) Representative AFM images of undamaged pUC19 (top) and UV-irradiated pUC19 (bottom). (B) Overview of how area (nm2) of each plasmid was analyzed. The area is defined as 1343 the region encapsulated within the DNAs colored in black. (C) Histogram plots of the areas for both 1344 undamaged pUC19 (top) and UV-irradiated pUC19 (bottom). The average area for the undamaged pUC19 is ~42,170 nm2, and the average area for the UV-irradiated pUC19 is ~42,200 nm2. Each AFM1346 image is 2×2 μm and 512×512 pixels.

Figure 4—figure supplement 2
Download asset Open asset
2D and 3D atomic force microscopy (AFM) images of Transcription Factor A, Mitochondrial (TFAM) tracts along DNA.

(A) 2D AFM images with TFAM-DNA complexes, with white arrows pointing at the tracts of TFAM along the DNA (white scale bars are all 100 nm). (B) 3D AFM images of two images in panel A (images labeled 1 and 2) showing the difference in areas along the DNA without protein and regions with DNA and protein (TFAM tracts). In the 3D AFM images, it is clear to see that the areas with DNA only (no protein) are much lower in height (nm), whereas the areas with protein along the DNA (TFAM tracts) are higher in height.

Figure 4—figure supplement 3
Download asset Open asset
Atomic force microscopy (AFM) images of 15 nM Transcription Factor A, Mitochondrial (TFAM) incubated with either damaged or undamaged DNA.

The top panels show examples of 15 nM TFAM in the presence of undamaged (-UV) pUC19 DNA and the bottom panels show examples of 15 nM TFAM in the presence of damaged (+UV) pUC19 DNA. Colored arrows on the AFM images represent the different TFAM-DNA complexes categorizations dispersed (green), intermediate (yellow), punctate (red), and protein-free DNA (blue). Each AFM image is 2×2 μm and 512×512 size pixels.

Figure 5 with 1 supplement
Download asset Open asset
Increased Transcription Factor A, Mitochondrial (TFAM) levels in vitro or in vivo do not protect mitochondrial DNA (mtDNA) from ultraviolet-C (UVC)-induced DNA damage or alter damage levels over time.

(A) Exposure paradigm for TFAM overexpression experiments. Cells were exposed to doxycycline for 48 hr prior to UVC exposure to ensure upregulation of TFAM at the time of exposure. Protein was quantified in control cells only to ensure TFAM upregulation. (B) Representative western blot of TFAM-tetON cell lysates following 48, 72, and 96 hr of doxycycline treatment to confirm TFAM upregulation. TFAM-tetON cells contain an HA tag that when expressed, results in a second band. (C) Pixel quantifications of n=3 western blots shown in panel (B) at each time point. TFAM protein levels were normalized to β-actin and then normalized to the non-doxycycline-treated controls. The x-axis represents the doxycycline treatment and y-axis represents the fold change relative to the non-doxycycline-treated cells. Data was analyzed via two-way ANOVA (doxycycline treatment p=0.0006, time p=0.25, interaction p=0.25). (D) TFAM mRNA quantification following doxycycline treatment. The x-axis represents the doxycycline treatment, and the y-axis represents the fold change relative to the non-doxycycline-treated cells. Data was analyzed via two-way ANOVA (doxycycline treatment p=0.0003, time p=0.52, interaction p=0.52). (E) Lesion frequency following UVC exposure in TFAM-tetON cells immediately after the exposure (0 hr recovery time). The x-axis represents the doxycycline treatment across a range of UVC doses and y-axis represents the lesion frequency (lesions per 10 kb). Data was analyzed via two-way ANOVA (UVC dose: p<0.0001, doxycycline treatment p=0.15, interaction: p=0.77). (F) Lesion frequency following UVC exposure in TFAM-tetON cells 24 hr after the exposure (24 hr recovery time). The x-axis represents the doxycycline treatment across a range of UVC doses and y-axis represents the lesion frequency (lesions per 10 kb). Data was analyzed via two-way ANOVA (UVC dose: p<0.0001, doxycycline treatment p=0.71, interaction: p=0.65). (G) Lesion frequency following UVC exposure in TFAM-tetON cells 48 hr after the exposure (48 hr recovery time). The x-axis represents the doxycycline treatment across a range of UVC doses and y-axis represents the lesion frequency (lesions per 10 kb). Data was analyzed via two-way ANOVA (UVC dose: p<0.0001, doxycycline treatment p=0.01, interaction: p=0.61). (H) Representative atomic force microscopy (AFM) images of in vitro nucleoids generated using purified TFAM and PCR amplified human mtDNA at 0, 100, 250, and 1000 nM TFAM. All scale bars represent 500 nm. (I) Lesion frequency following UVC exposure in in vitro nucleoids. The x-axis represents the UVC dose and y-axis represents the lesion frequency (lesions per 10 kb). Data was analyzed via two-way ANOVA (UVC dose: p<0.0001, TFAM concentration p=0.20, interaction: p=0.86). (J) Lesion frequency following UVC exposure in cells with and without a TFAM knockdown. The x-axis represents the recovery time, i.e., time following exposure to UVC, and the y-axis represents the level of mtDNA damage (lesions per 10 kb). Data was analyzed via a three-way ANOVA (UVC dose: p<0.0001, recovery time: p=0.002, siRNA treatment: p<0.0001, recovery time*UVC dose: p<0.0001, recovery time*siRNA treatment: p=0.001, UVC dose*siRNA treatment: p=0.08, recovery time*UVC dose*siRNA treatment: p=0.66). (K) Ratio of mtDNA copy number to nuclear DNA copy number following UVC exposure in cells with and without a TFAM knockdown. The x-axis represents the recovery time, i.e., time following exposure to UVC, and the y-axis represents the level of mtDNA damage (lesions per 10 kb). Data was analyzed via a three-way ANOVA (UVC dose: p=0.01, recovery time: p<0.01, siRNA treatment: p<0.0001, recovery time*UVC dose: p<0.001, recovery time*siRNA treatment: p<0.0001, UVC dose*siRNA treatment: p=0.04, recovery time*UVC dose*siRNA treatment: p=0.01).

Figure 5—source data 1

Original files for western blot analysis displayed in Figure 5B.

https://cdn.elifesciences.org/articles/108862/elife-108862-fig5-data1-v1.zip
Figure 5—source data 2

PDF file containing original western blot analysis displayed in Figure 5B, indicating relevant bands, treatment groups, and time points.

https://cdn.elifesciences.org/articles/108862/elife-108862-fig5-data2-v1.zip
Figure 5—figure supplement 1
Download asset Open asset
Transcription Factor A, Mitochondrial (TFAM) overexpression does not protect cells from loss of cell viability from ultraviolet-C (UVC) exposures.

Cell viability following exposure to UVC normalized to the control for each replicate (n=3), determined using a resazurin cell viability assay. For all panels, x-axes represent the doxycycline treatment conditions across a range of UVC dose delivered to the cells in J/m2. Y-axes represent the fluorescence readout normalized as percent control. Data was analyzed via two-way ANOVA. (A) Cell viability immediately after exposure to UVC (UVC dose: p=0.8087, doxycycline treatment: 0.0029, interaction: p=0.6965). (B) Cell viability 24 hr after exposure to UVC (UVC dose: p<0.0001, doxycycline treatment: 0.7183, interaction: p=0.9790). (C) Cell viability 48 hr after exposure to UVC (UVC dose: p<0.0001, doxycycline treatment: 0.9968, interaction: p=0.7527).

Figure 6
Download asset Open asset
Hypothesized cellular model of the role of Transcription Factor A, Mitochondrial (TFAM) in mitochondrial DNA (mtDNA) damage sensing.

The in vitro data provided in this study indicate that TFAM more readily compacts DNA harboring UV-induced lesions. While there is currently no in vivo evidence to suggest damaged mtDNA is more compacted, future work should determine whether this phenomenon is occurring in cells. In vivo, this feature may serve to ‘tag’ mitochondrial genomes as damaged, which could lend itself to repression of the replication of damaged genomes, flagging damaged genomes for targeted degradation, or both. Both active removal and repression of replication would allow for removal of damaged genomes during mtDNA turnover processes and provide a mechanism for preventing mtDNA mutagenesis. Created with BioRender.com.

Tables

Table 1
Binding of Transcription Factor A, Mitochondrial (TFAM) to array sequences, measured using fluorescence anisotropy.
Sequence-UV+UV
Array z-score percentileKD ±SEM (nM)n±SEMArray z-score percentileKD ±SEM (nM)n±SEM
ND4_473*0.484.95±0.035.06±0.0928.285.36±0.144.76±0.11
ND2_40125.324.36±0.242.64±0.2223.626.24±0.152.61±0.25
ND3_9225.684.53±0.132.63±0.2210.235.54±0.142.19±0.06
RNR2_61925.714.89±0.332.46±0.2351.198.13±0.392.33±0.10
COX1_2775.326.09±0.432.07±0.0774.455.90±0.102.12±0.28
ND1_28875.163.41±0.033.04±0.1935.796.12±0.221.90±0.12
ND6_8775.745.20±0.322.84±0.1495.507.63±0.212.27±0.26
ND1_45090.254.33±0.172.40±0.4370.406.36±0.512.20±0.13
COX2_22990.783.94±0.032.77±0.1780.814.87±0.352.18±0.14
TRNT_1090.904.24±0.122.15±0.2496.077.79±0.801.81±0.05
ND1_353 40mer99.995.69±0.153.09±0.1194.247.25±0.394.38±0.33
ND1_353†99.997.56±0.343.85±0.4994.247.66±0.882.94±0.40
  1. *

    This sequence was used for AFM oligomerization studies to represent a low occupancy sequence in the array-based TFAM binding data, referred to as ‘low occupancy sequence’ in the text.

  2. †

    This sequence was used for AFM oligomerization studies to represent a high occupancy sequence in the array-based TFAM binding data, referred to as ‘high occupancy sequence’ in the text.

Additional files

Supplementary file 1

Total number of Transcription Factor A, Mitochondrial (TFAM) complexes analyzed in Figure 4D.

https://cdn.elifesciences.org/articles/108862/elife-108862-supp1-v1.docx
Supplementary file 2

Sequences used for Transcription Factor A, Mitochondrial (TFAM) KD measurements.

https://cdn.elifesciences.org/articles/108862/elife-108862-supp2-v1.docx
MDAR checklist
https://cdn.elifesciences.org/articles/108862/elife-108862-mdarchecklist1-v1.docx

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Open citations (links to open the citations from this article in various online reference manager services)

Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)

  1. Dillon E King
  2. Emily E Beard
  3. Matthew J Satusky
  4. Alex George
  5. Ian Ryde
  6. Caitlin Johnson
  7. Emma L Dolan
  8. Yuning Zhang
  9. Wei Zhu
  10. Hunter Wilkins
  11. Evan Corden
  12. Susan K Murphy
  13. Dorothy A Erie
  14. Raluca Gordân
  15. Joel Meyer
(2026)
UV irradiation alters TFAM binding specificity and compaction of DNA
eLife 14:RP108862.
https://doi.org/10.7554/eLife.108862.3