- Email: [email protected]
Human mitochondrial transcription factor A is required for the segregation of mitochondrial DNA in cultured cells
E XP ER I ME NT AL CE L L R ES E AR CH 3 1 7 (2 0 11 ) 2 10 – 2 20
available at www.sciencedirect.com
www.elsevier.com/locate/yexcr
Research Article
Human mitochondrial transcription factor A is required for the segregation of mitochondrial DNA in cultured cells Katsumi Kasashima⁎, Megumi Sumitani, Hitoshi Endo Department of Biochemistry, Jichi Medical University, Tochigi 329-0498, Japan
A R T I C L E I N F O R M A T I O N
A B S T R A C T
Article Chronology:
The segregation and transmission of the mitochondrial genome in humans are complicated
Received 2 September 2010
processes but are particularly important for understanding the inheritance and clinical
Revised version received
abnormalities of mitochondrial disorders. However, the molecular mechanism of the
8 October 2010
segregation of mitochondrial DNA (mtDNA) is largely unclear. In this study, we demonstrated
Accepted 9 October 2010
that human mitochondrial transcription factor A (TFAM) is required for the segregation of mtDNA
Available online 15 October 2010
in cultured cells. RNAi-mediated knockdown of TFAM in HeLa cells resulted in the enlarged mtDNA, as indicated by the assembly of fluorescent signals stained with PicoGreen. Fluorescent in
Keywords:
situ hybridization confirmed the enlarged mtDNA and further showed the existence of increased
Mitochondria
numbers of mitochondria lacking mtDNA signals in TFAM knockdown cells. By complementation
TFAM
analysis, the C-terminal tail of TFAM, which enhances its affinity with DNA, was found to be
mtDNA
required for the appropriate distribution of mtDNA. Furthermore, we found that TFAM knockdown
Segregation
induced asymmetric segregation of mtDNA between dividing daughter cells. These results suggest
Nucleoid
an essential role for human TFAM in symmetric segregation of mtDNA. © 2010 Elsevier Inc. All rights reserved.
Introduction Mitochondria are semi-autonomous organelles that contain their own genomic DNA called mitochondrial DNA (mtDNA). Mammalian mtDNA is a closed circular molecule of about 16 kbp in length and exists in multiple copies per cell. Human mtDNA encodes proteins that are essential components of respiratory chain complexes. Mutations in human mtDNA are closely associated with neuromuscular diseases, such as MELAS [1]. mtDNA is maternally inherited, and mammalian mtDNA variants segregate rapidly between generations. Mutated mtDNA molecules some-
times co-exist with wild-type mtDNA, which is called heteroplasmy. How the mutant mtDNA molecules become fixed is controversial issue, and it is likely that the mechanism is not explained only by random genetic drift [2]. Several studies have demonstrated that particular mtDNA variants show non-random segregation in human cells [3,4]. The biased segregation of mtDNA variants is dependent on cell type and is suggested to play a role in human mtDNA diseases and mutant load [2,5]. Thus, the segregation of mtDNA seems to be important for understanding the inheritance and clinical abnormalities of mitochondrial disorders; however, its molecular basis is largely unknown.
⁎ Corresponding author. Department of Biochemistry, Jichi Medical University, 3311-1 Yakushiji, Shimotsuke, Tochigi 329-0498, Japan. Fax: +81 285 44 1827. E-mail address: [email protected] (K. Kasashima). Abbreviations: TFAM, mitochondrial transcription factor A; mtDNA, mitochondrial DNA; FISH, fluorescent in situ hybridization; MELAS, mitochondrial myopathy, encephalopathy, lactic acidosis, stroke-like episodes; OPA1, optic atrophy 1; ddC, dideoxycytidine; PHB2, prohibitin 2; mtSSB, mitochondrial single stranded DNA-binding protein; LRP130, leucine-rich protein 130 kDa; GAPDH, glyceraldehydes 3-phosphate dehydrogenase; RNAi, RNA interference; POLG, mtDNA polymerase gamma; siRNA, small interfering RNA; D-loop, displacement loop; rRNA, ribosomal RNA; rTFAM, recombinant TFAM 0014-4827/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.yexcr.2010.10.008
E XP E RI ME N T AL C ELL R ES E AR C H 3 1 7 ( 2 0 11 ) 2 1 0– 2 20
Recent studies have shown that mtDNA exists not as naked DNA but rather as a highly organized structure associated with many proteins (the nucleoprotein complex) as in the nuclear genome. These complex structures are called mitochondrial nucleoids, and the nucleoid components are believed to regulate the stability, replication, transcription, and segregation of mtDNA [6]. Mitochondrial transcription factor A (TFAM) is known to be a major nucleoid protein that binds directly to mtDNA [7]. In addition to its role as a transcription factor for mtDNA, mammalian TFAM acts to maintain the copy number of mtDNA [8,9] and is involved in its replication [10]. TFAM contains two high mobility group (HMG)-box domains, which are required for DNA binding, and a C-terminal tail abundant in basic amino acid residues. The C-terminal tail was originally identified as a domain required for the activation of transcription [11] that interacts with mitochondrial transcription factor B1 [12]. Recently, it has been shown that the C-terminal tail is important for efficient interaction between TFAM and DNA [13,14]. In molecular and structural studies, it was found that TFAM acts as a homodimer during its interaction with DNA [14–16] and that the Cterminal tail facilitates its formation [14]. Therefore, the C-terminal tail seems to be an indispensable domain for TFAM function. In yeast, Abf2, a yeast homologue of TFAM, maintains the mitochondrial genome and also regulates its segregation within zygotes [17,18]. The segregation of mtDNA has been extensively studied in yeast because several mtDNA deletion variants are available and their transmission can be easily tracked. mtDNA segregation is considered to be a directed process that is driven by a putative segregation apparatus; however, the molecular mechanism underlying it remains largely unknown, even in yeast. In mammals, the factors that regulate the segregation of mtDNA into dividing cells have not been reported. In this study, we identified a novel function of human TFAM: it is required for the segregation of mtDNA in cultured cells. RNAimediated knockdown of TFAM in HeLa cells resulted in the enlarged mtDNA signals produced by staining with the DNAspecific fluorescent dye PicoGreen. Fluorescent in situ hybridization (FISH) confirmed that the PicoGreen-stained signals reflected the enlarged mtDNA. The FISH results also showed that the number of mitochondria lacking mtDNA signals was increased in these cells, suggesting that the segregation of mtDNA is impaired by TFAM knockdown. In a complementation assay, the C-terminal tail of TFAM, which enhances its affinity with DNA [13,14], was found to be required for the proper distribution of mtDNA. Furthermore, it was shown that TFAM knockdown induced asymmetric segregation of mtDNA between dividing daughter cells. Here, we demonstrate that human TFAM is required for symmetric segregation of mtDNA in cultured cells and provide an insight into the segregation apparatus of mtDNA in mammals. These results suggest that TFAM participates in the asymmetric segregation and transmission of heteroplasmic mtDNA mutations in mitochondrial disorders.
Materials and methods Plasmid construction For RNA interference (RNAi), small interfering RNA (siRNA) sequences for TFAM, OPA1, and M19 were introduced into the pSilencer 3.1-H1
211
Puro vector (Ambion), as described previously [19,20]. For targeting of the 3′ UTR of human TFAM, another siRNA sequence (5′- AAGATTGAGATGTGTTCACAA-3′) was designed and introduced into the expression vector. The coding region of TFAM and its C-terminal deletion mutant (amino acids 1–221) were amplified by PCR from a human heart cDNA library, and the PCR products were introduced into the mammalian expression vector pIRESpuro3 (Clontech).
Cell culture and transfection HeLa and ρ0 HeLa cells were cultured as previously described [19]. HeLa cells were also cultured in medium containing 10 μM dideoxycytidine (ddC) for the indicated periods. Transfection was performed using Lipofectamine 2000 (Invitrogen) according to the manufacturer's instructions. HeLa cells were plated on 35mm poly-L-lysine-coated glass-bottomed dishes (Matsunami Glass Ind.). The mitochondria in living HeLa cells were stained with MitoTracker Red CM-H2XRos (250 nM, Molecular Probes) for 30 min at 37 °C, and mtDNA was stained with PicoGreen solution (3 μl/ml, Molecular Probes) for 60 min at 37 °C. Cells expressing pSilencer 3.1-H1 Puro constructs were selected with 2 μg/ml of puromycin (Sigma). Fluorescent images were captured and analyzed with a μRadiance™ Laser Scanning Confocal Microscope System (Bio-Rad). For the synchronization of cells, HeLa cells were treated with 2 mM thymidine for 23 h and released from the G1/S phase by being placed in thymidine-free medium for 11 h in order to restart the cell cycle.
Fluorescent in situ hybridization FISH was performed according to a slightly modified version of a previously described method [21]. Two micrograms of the 1.1 kb displacement loop (D-loop) mtDNA fragment (nt: 16029-599) was labeled with Alexa Fluor 488 fluorescent dye (ARES DNA labeling kit, Molecular Probes) by nick translation (Roche). The probe was ethanol precipitated with herring sperm DNA (Promega) as a carrier and COT-1 DNA (Invitrogen) as a general blockant, and resolved in hybridization buffer (50% deionized formamide, 10% Dextran Sulfate, 2× SSC). For hybridization, we used a probe that was diluted to a final concentration of about 2 ng/μl with hybridization buffer. HeLa cells were stained with MitoTracker Red CM-H2XRos (2.5 μM) for 30 min, fixed for 20 min at room temperature with 4% paraformaldehyde in PBS, and then treated for 15 min with 0.1% Triton X-100 in PBS. The cells were treated with RNase A (100 μg/ml in 2× SSC) for 1.5 h at 37 °C and then dehydrated and rehydrated using 70%, 90%, and 100% ethanol for 2 min at each step. Then, prehybridization was performed using 2× SSC buffer (2× SSC, 0.3 M NaCl, and 0.04 M sodium citrate) for 1 h at 37 °C, and the cells were dehydrated using a graded series of ethanols. The cells were then denatured with 70% deionized formamide in 2× SSC for 5 min at 72 °C, dehydrated again with ice-cold ethanol (70% and 90%) for 1 min at each step, and then stored in 100% ethanol at room temperature. After denaturing the probe for 5 min at 72 °C, it was applied to the cells, and hybridization was performed overnight at 37 °C in a humid chamber in the dark. After washing the cells with 0.4× SSC with 0.3% Tween-20 for 5 min at 72 °C and 2× SSC with 0.1% Tween-20 for 1 min at room temperature, fluorescent images were captured with the Axio Observer D1 system (Carl Zeiss) or the Leica TCS SP5 confocal microscope system (Leica).
212
E XP ER I ME NT AL CE L L R ES E AR CH 3 1 7 (2 0 11 ) 2 10 – 2 20
Combined immunofluorescence and FISH After 11 h of thymidine release, the cells were fixed for 15 min with 4% formaldehyde in PBS and then treated for 15 min with 0.1% Triton X-100 in PBS. The cells were then probed with anti-αtubulin mouse monoclonal antibody (DM1A, Abcam) and labeled with Cy3-conjugated anti-mouse IgG antibody (Molecular Probes). Mitochondria were also immunostained with anti-PHB2 rabbit polyclonal antibody (the product produced by immunization with both GST and His-tagged PHB2 protein), anti-OPA1 (our product), anti-mtSSB (Sigma Aldrich), or anti-LRP130 (GeneTex, Inc.) rabbit polyclonal antibody. The signals were visualized with Alexa Fluor 488, Cy3-conjugated, or Cy5-conjugated anti-rabbit IgG (Molecular Probes, GE Healthcare). After the immunofluorescence, the cells were fixed again for 20 min with 4% paraformaldehyde in PBS, and FISH was performed as described above.
Western blotting Samples were separated by electrophoresis on SDS-polyacrylamide gels (12% acrylamide) and electrophoretically transferred to a nitrocellulose membrane (Hybond ECL, GE Healthcare). The membrane was probed with antibodies and detected with an enhanced chemiluminescence system (GE Healthcare), as previously described [22]. The following primary antibodies were used: anti-TFAM (1:100; SantaCruz), anti-glyceraldehyde 3-phosphate dehydrogenase (GAPDH, 1:3000; Chemicon), and anti-OPA1 (1:100, our product).
Quantitative PCR Genomic DNA from HeLa cells was extracted by the standard Proteinase K digestion method [19]. LightCycler-FastStart DNA Master SYBR Green I (Roche) was used for the quantitative PCR with LightCycler (Roche). To produce a standard curve, 0.5, 1, 2, 4, and 8 ng (for mtDNA amplification) or 1, 2, 4, 8, and 16 ng (for 18S ribosomal RNA (rRNA) gene amplification) of HeLa cell genomic DNA were used. For the amplification of mtDNA (132 bp) and 18S rRNA (104 bp) fragments, the following primer sets were used: mtDNA: forward, 5′-GCCTGCCTGATCCTCCAAAT-3′ (nt: 14,862– 14,881), reverse, 5′-AAGGTAGCGGATGATTCAGCC-3′ (nt: 14,973– 14,993), 18S rRNA: forward, 5′-TAGAGGGACAAGTGGCGTTC-3′, reverse, 5′-CGCTGAGCCAGTCAGTGT-3′.
[23], about half of the knockdown cells showed large areas displaying intense PicoGreen signals but a reduced numbers of signals (Fig. 1B, Supplemental Fig. 2). PicoGreen signals were confirmed to be specific to mtDNA because ρ0 HeLa cells that lack mtDNA completely were not stained by the dye (Supplemental Fig. 3). The brightly stained PicoGreen fluorescent signals, in which their size was the twice or more larger than that in control cells and the number was reduced to less than half of the control, were defined as enlarged mtDNA. To investigate the relationship between knockdown efficiency of TFAM and the enlarged mtDNA phenotype, they were examined at 2 to 4 days after transfection with the RNAi vector. In the analysis, knockdown of TFAM expression became high as days go by (Supplemental Fig. 4). By counting the number of cells containing enlarged mtDNA, its frequency was found to be increased in parallel with the knockdown (Supplemental Fig. 4). Therefore, the decrease of TFAM expression correlates with the formation frequency of enlarged mtDNA, which is similar to the relationship between TFAM protein level and mtDNA copy number [9]. No enlarged fluorescent signals was observed in the non-transfected control or the cells transfected with an RNAi vector targeting another mitochondrial nucleoid protein, M19 [20] (Supplemental Fig. 5), indicating that it is a phenotype specific to TFAM knockdown cells. PicoGreen and MitoTracker staining showed that the enlarged signals were detected in mitochondria and that a large number of mitochondria in the TFAM knockdown cells contained no visible mtDNA signals (Fig. 1C). These results strongly suggest that the mechanism for the correct distribution of mtDNA in newly dividing mitochondria does not work well in TFAM knockdown cells. We then directly detected mtDNA in TFAM knockdown cells by fluorescent in situ hybridization (FISH). In non-transfected control cells, the FISH signals were extensively distributed in mitochondria and detected non-specifically in nucleus (Fig. 1D). The mitochondrial signals were confirmed to be specific because they were hardly detected in HeLa cells containing only 11% of the control level of mtDNA after 4-day ddC treatment (Supplemental Fig. 6). ddC is an inhibitor of mtDNA replication, and its addition reduces the copy number of mtDNA [23]. FISH confirmed the presence of enlarged mtDNA signals by confocal microscopy and also showed the existence of a large number of mitochondria lacking mtDNA in the TFAM knockdown cells (Fig. 1D). These results confirm that the PicoGreen-stained signals seen in the TFAM knockdown cells represented the enlarged mtDNA signals.
No enlarged mtDNA was observed after the inhibition of its replication
Results Enlarged mtDNA signals in TFAM knockdown cells TFAM maintains the copy number of mtDNA in mammals [8,9]. Using RNA interference (RNAi), we successfully and specifically silenced human TFAM in HeLa cells (Fig. 1A). When we performed knockdown of TFAM in HeLa cells, these cells showed a reduction in the amount of mtDNA they contained on quantitative PCR analysis (Supplemental Fig. 1). Since the same result was previously obtained by Southern blot analysis [19], we quantified the amount of mtDNA by this method. In addition to the abovementioned phenotype, when TFAM knockdown cells were stained with PicoGreen, which quantitatively stained mtDNA in living cells
TFAM also regulates the replication of mtDNA [10]. To test whether the enlarged mtDNA is caused by the inhibition of mtDNA replication in TFAM knockdown cells, HeLa cells were treated with the replication inhibitor ddC. Under treatment with ddC, the intensity of the mtDNA signals stained with PicoGreen and the amount of mtDNA, which was measured by quantitative PCR, gradually decreased (Fig. 2 A, B). However, no enlarged signals were observed in these cells. After 24-h treatment with ddC, the amount of mtDNA was decreased to about half of the control level (Fig. 2B), which was almost same as that caused by TFAM knockdown (Supplemental Fig. 1). Even in these conditions, no enlarged mtDNA was observed. We also confirmed that the TFAM protein level was not significantly affected by 24 h ddC treatment
E XP E RI ME N T AL C ELL R ES E AR C H 3 1 7 ( 2 0 11 ) 2 1 0– 2 20
213
Fig. 1 – Enlarged mtDNA signals in TFAM knockdown cells. (A) At 4 days after transfection with the RNAi vector for TFAM, protein expression was determined by Western blotting using the indicated antibodies. (B) The TFAM-knocked down HeLa cells were stained with PicoGreen, and the images were obtained by confocal microscopy. The square boxes represent the area enlarged. Quantification of the percentage of cells that contains enlarged mtDNA signals is shown. (C) The TFAM-knocked down HeLa cells were stained with PicoGreen and MitoTracker, and the images were obtained by confocal microscopy. (D) At 4 days after transfection with the RNAi vector, the cells and non-transfected control cells were probed with mtDNA probes (green signals). The cells were also stained with MitoTracker (red signals), and the merged images were obtained with a fluorescent microscope or a confocal microscope (confocal image). mtDNA probes also stained nucleus non-specifically (Supplemental Figure 6). FISH image of siTFAM was obtained by using shorter exposure time than control image because the intensity of the mtDNA signals was strong. Scale bar: 10 μm.
(Fig. 2C). These results indicate that the enlarged mtDNA signals seen in the TFAM knockdown cells were not caused by the inhibition of replication. We also observed that, at least, 24-h treatment with ddC did not efficiently block the enlarged phenotype induced by TFAM knockdown (Supplemental Fig. 7).
The enlarged mtDNA by TFAM knockdown occurs independently of mitochondrial fusion and reflects assembled nucleoids Recently, Ashley et al. reported that DNA intercalators cause enlarged mtDNA (also called aggregation of mtDNA nucleoids), which they termed remodeled nucleoids [24]. The remodeling of nucleoids
requires the fusion machinery contained in mitochondria because it is suppressed by the knockdown of the pro-fusion proteins mitofusin 1 (MFN1) or optic atrophy 1 (OPA1) [24]. So, we examined the effect of mitochondrial fusion on the aggregation of mtDNA caused by TFAM knockdown. We performed double knockdown of TFAM and OPA1 in HeLa cells and confirmed that they were specifically knocked down by RNAi (Fig. 3A). We also confirmed that OPA1 was efficiently knocked down in cells containing fragmented mitochondria by immunofluoresence (Supplemental Fig. 8). In the double knockdown cells, enlarged mtDNA was still observed in fragmented mitochondria (Fig. 3B), as was observed in tubular mitochondria in TFAM knockdown or control cells (Fig. 3B). No enlarged mtDNA was seen in the fragmented mitochondria, in which only OPA1 was knocked down (Fig. 3B). These
214
E XP ER I ME NT AL CE L L R ES E AR CH 3 1 7 (2 0 11 ) 2 10 – 2 20
Fig. 2 – Absence of the enlarged mtDNA in replication inhibited cells. HeLa cells were treated with the inhibitor of the mtDNA replication ddC for the indicated periods. (A) The cells were stained with PicoGreen, and the images were obtained by confocal microscopy. Scale bar: 10 μm. (B) The mtDNA content of these ddC-treated cells was quantified by quantitative PCR. The amount of mtDNA per nuclear gene (18S rRNA) is shown. (C) The protein expression in these ddC-treated cells was determined by Western blotting using the indicated antibodies.
results indicate that the enlarged mtDNA caused by TFAM knockdown is independent of mitochondrial fusion and is not composed of remodeled nucleoids. Next, to reveal whether the enlarged mtDNA induced by TFAM knockdown reflects the enlarged nucleoids, submitochondrial localization of the nucleoid proteins, such as mitochondrial single-stranded DNA-binding protein (mtSSB) and leucine-rich protein 130 kDa (LRP130), was investigated. mtSSB is a core nucleoid protein that shows similar localization with mtDNA [25,26]. When TFAM knockdown cells were immunostained with mtSSB antibody, the immunofluorescent signals were found to be assembled, which corresponded to the mtDNA signals by FISH (Fig. 3C). In contrast, the immunofluorescent signals against LRP130, a mitochondrial RNAbinding protein included in the nucleoids [25,27], were overlapped with mtDNA signals in TFAM knockdown cells but were not assembled (Fig. 3C). These results strongly suggest that the enlarged mtDNA seen
in the TFAM knockdown cells reflects the enlarged nucleoids, which are enriched with core nucleoid proteins, such as mtSSB.
The C-terminal tail of TFAM is required for proper distribution of mtDNA To clarify the molecular mechanism underlying the distribution of mtDNA induced by TFAM, we performed a complementation assay using RNAi and the expression of recombinant TFAM (rTFAM). RNAi targeting of the 3′ UTR of TFAM was also performed to avoid the suppression of rTFAM expression by the RNAi vector. The novel RNAi vector efficiently and specifically silenced TFAM expression in HeLa cells (Fig. 4A). After PicoGreen staining, the knockdown cells showed enlarged fluorescent signals (Fig. 4B), further confirming that the phenotype is specific to TFAM knockdown. We then transfected the RNAi vector with the TFAM expression
Fig. 3 – The enlarged mtDNA induced by TFAM knockdown occurs independently of mitochondrial fusion. (A) HeLa cells were double knocked down with TFAM and OPA1 RNAi vectors. At 4 days after transfection with the RNAi vectors or the control vector (puro-vec.), protein expression was determined by Western blotting using the indicated antibodies. (B) The double knockdown and single knockdown (TFAM or OPA1) cells were stained with PicoGreen and MitoTracker. Enlarged mtDNA induced by TFAM knockdown was still observed in fragmented mitochondria caused by OPA1 knockdown. (C) At 4 days after transfection with the RNAi vector for TFAM, the knockdown cells were probed with the indicated antibodies against the nucleoid proteins by immunofluorescence and mtDNA probe by FISH. Scale bar: 10 μm.
E XP E RI ME N T AL C ELL R ES E AR C H 3 1 7 ( 2 0 11 ) 2 1 0– 2 20
215
216
E XP ER I ME NT AL CE L L R ES E AR CH 3 1 7 (2 0 11 ) 2 10 – 2 20
Fig. 4 – The C-terminal tail of TFAM is required for proper distribution of mtDNA. At 4 days after transfection with a novel RNAi vector targeting the 3′ UTR of TFAM (siTFAM-3U) or the control vector (puro-vec.), protein expression was quantified by Western blotting using the indicated antibodies (A), and the cells were stained with PicoGreen (B). The images were obtained by confocal microscopy. Scale bar: 10 μm. (C) At 3 days after co-transfection with the RNAi vector for TFAM-3U and the indicated rTFAM expression plasmid, protein expression was quantified by Western blotting using the indicated antibodies. (D) At four days after co-transfection with the RNAi vector and rTFAM expression plasmid, the proportion of cells containing enlarged mtDNA was measured by PicoGreen staining in three independent experiments (in each experiment, the total cell number was more than 100 cells).
plasmid and examined whether the enlarged mtDNA was suppressed by its expression. The expression of rTFAM significantly rescued the enlarged phenotype (Fig. 4 C, D), confirming that the phenotype is caused by the downregulation of TFAM. The C-terminal tail of TFAM, which is required for efficient interaction between TFAM and DNA, is known to be important for TFAM functions, such as transcription [11]. So, we examined its importance for the distribution of mtDNA. When a deletion mutant of TFAM lacking the C-terminal tail (TFAM-delta C) was used in the complementation experiment, it did not rescue the phenotype as well as the wild type (Fig. 4C, D). Because the precursor protein of TFAM-delta C, which has similar molecular weight with endogenous TFAM, was detected by Western blotting (Supplemental Fig. 9), the endogenous TFAM protein in the rescue with the mutant was estimated slightly high (Fig. 4C). We also confirmed that no enlarged mtDNA was observed by the expression of TFAM-delta C only (Supplemental Fig. 10), indicating that the mutant does not work as dominant-negative mutant. As has been reported by other group [9], the C-terminal deletion mutant was sufficient for the rescue of mtDNA copy number
in HeLa cells (Supplemental Fig. 11). Thus, the C-terminal tail is required for the proper distribution of mtDNA by TFAM.
TFAM knockdown induces asymmetric segregation of mtDNA between dividing two daughter cells The enlarged mtDNA caused by TFAM knockdown strongly suggests that the segregation of mtDNA is impaired in these cells. To obtain direct evidence to support our hypothesis that TFAM is required for mtDNA segregation, the distribution of mtDNA in dividing cells was investigated. HeLa cells were synchronized at the G1/S phase using the single thymidine block procedure and subsequently released into thymidine-free medium to restart their cell cycles. Eleven hours after their release, most cells were in the cytokinesis phase, which was confirmed by the characteristic localization of α-tubulin at the midbody-like structure present between dividing cells (Fig. 5A). Hereafter, these cells are referred to as midbody-positive cells. To investigate the segregation of mtDNA between two cells, we chose midbodypositive cells that had been produced by division of the same
E XP E RI ME N T AL C ELL R ES E AR C H 3 1 7 ( 2 0 11 ) 2 1 0– 2 20
217
Fig. 5 – TFAM knockdown induces asymmetric segregation of mtDNA between dividing cells. At 3 days after transfection with the RNAi vector for TFAM, the cells were synchronized with thymidine treatment as described in the Materials and Methods section. After 11 h of the release from the thymidine block, the cells were fixed. (A) The cells were immunostained with anti-α-tubulin (red) and anti-PHB2 (green) antibodies. Cells in the cytokinesis phase showed the localization of α-tubulin at a midbody-like structure (midbody-positive cells). (B) The cells were subjected to combined immunofluorescence and FISH. The cells were immunostained with anti-α-tubulin (red) and probed with mtDNA probes (green) and observed at several z-axis. For TFAM knockdown cells, two representative images (asymmetric (left) and symmetric (right) distribution of mtDNA) are shown. Merged images are also shown. (C) Cell populations containing symmetric or asymmetric segregation of mtDNA is shown. In the control, all ‘midbody-positive’ cells showed symmetric segregation of mtDNA. In the TFAM knockdown cells, which contained enlarged mtDNA and were ‘midbody-positive’, the symmetric segregation was significantly impaired. The criteria in determining whether asymmetric or not are described in text. (D) The TFAM-knocked down HeLa cells were immunostained with anti-α-tubulin (red) and anti-PHB2 (blue) and probed with mtDNA probes (green), and a merged image is shown. The arrowheads represent aggregated mtDNA signals. Scale bar: 10 μm.
parent cell. We first observed the segregation of mitochondria in TFAM knockdown cells by labeling mitochondria with antibody against PHB2, which mainly localizes to mitochondria in HeLa cells
[22]. As was observed in control cells, mitochondria were segregated almost evenly between TFAM knockdown cells (Fig. 5A). Next, using combined immunofluorescence and FISH
218
E XP ER I ME NT AL CE L L R ES E AR CH 3 1 7 (2 0 11 ) 2 10 – 2 20
technology [20,28], the segregation of mtDNA between dividing cells was examined. We estimated the nucleoid number by mtDNA signals stained with FISH and found that each dividing cell contained about 100 to 200 nucleoids. The case, in which the nucleoid number in dividing one cell was the twice or more than that in the other cell, was defined as asymmetric distribution. As was predicted, mtDNA was almost symmetrically segregated between dividing cells in the control experiment (Fig. 5B). However, under TFAM knockdown, about half of the midbodypositive cells that contained enlarged mtDNA signals showed asymmetric segregation of mtDNA between dividing cells (Fig. 5B, C), which was estimated by counting the number of enlarged mtDNA at several z-axis (Supplemental Fig. 12). The knockdown efficiency of TFAM between the dividing cells was considered to be almost equal because the cells were just divided from same parent cell. We further confirmed that mitochondria were equally segregated in the TFAM knockdown cells, in which mtDNA was segregated asymmetrically (Fig. 5D). These results demonstrated that knockdown of TFAM induced asymmetric segregation of mtDNA.
Discussion In this study, we observed that mtDNA signals were enlarged within mitochondria in TFAM knockdown cells. Since mtDNA is depleted in TFAM knockout mice [8], this phenotype is only seen in knockdown studies. To date, similar phenotypes have been reported in ATAD3A knockdown [5], mtDNA polymerase gamma beta subunit (POLGbeta) or Twinkle mutant-overexpressed cells [29,30], and cells treated with DNA intercalators [24]. The AAA+ protein ATAD3A, a nucleoid factor, binds to the D-loop within mtDNA and regulates the organization of mitochondrial nucleoids [31]. In ATAD3A knockdown 143B cells, larger nucleoids were observed by PicoGreen staining at 72 h after treatment with siRNA [5]. In the case of overexpression of POLGbeta or Twinkle mutant, the number of nucleoids, which was identified by immunofluorescence with antiDNA antibodies, was reduced, and the nucleoids were enlarged in the cells [29,30]. These cases indicate the possibility that these nucleoid proteins are involved in the segregation of mtDNA. In this study, we demonstrated directly that symmetric segregation of mtDNA was significantly impaired in dividing cells by TFAM knockdown (Fig. 5). Although it is possible that the asymmetric distribution of mtDNA occurs stochastically by the reduced number of enlarged nucleoids, it is evident that TFAM is required for the proper distribution of mtDNA.
How does TFAM regulate the distribution of mtDNA? How does depletion of TFAM lead to enlarged mtDNA and asymmetric segregation of mtDNA? TFAM is a core nucleoid protein and wraps entire region of mtDNA [7]. Depletion of TFAM might alter the nucleoids organization and impair their segregation within mitochondria, which results in the enlarged mDNA nucleoids. For determining whether TFAM directly regulates the segregation of mtDNA, more experimental evidence would be required. Mitochondria share several similarities with their prototype bacteria. In bacteria, the segregation of plasmid DNA is controlled by ATPase and DNA-binding proteins [32]. Given that
TFAM contains no ATPase domain, an unknown factor that possesses an ATPase domain and binds to TFAM-mtDNA complex might regulate the segregation of nucleoids as a molecular motor. In this respect, ATAD3 is a candidate for this factor and might be involved in the regulation of segregation through its binding to TFAM–mtDNA complexes. Further investigation is required to examine this possibility. As shown in Fig. 4, the C-terminal tail of TFAM is required for appropriate distribution of mtDNA. Considering the importance of the domain in transcription, TFAM might regulate the segregation of mtDNA by a transcription-coupled mechanism. Alternatively, the C-terminal tail enhances the DNA-binding activity of TFAM, which leads to the formation of its dimmer and enables appropriate packaging of mtDNA [13,14]. Therefore, homodimerization of TFAM and proper packaging of mtDNA might be important for mtDNA segregation. As reported by other group [9], the C-terminal deletion mutant was sufficient for the rescue of mtDNA copy number in HeLa cells. Thus, the mechanism for maintaining mtDNA copy number might be different from that for the segregation. Yeast mtDNA shows a high frequency of recombination, and the removal of recombination junctions by resolving enzymes is required for efficient segregation of mtDNA [33,34]. These mutations causing reduced resolving activity induce the aggregation of mtDNA, which results in reduced segregation of mtDNA into daughter cells. Thus, it is possible that active aggregation causes the abnormal segregation of mtDNA. Considering that the recombination of mtDNA in mammalian cultured cells is rare, the possibility that mtDNA actively aggregates through impairment of the resolving activity of recombination junctions is low. Furthermore, interlinked molecules of mtDNA, which were observed in a resolvase yeast mutant, were not seen in TFAM knockdown cells because the electrophoretic mobility of mtDNA showed almost no changes in these cells [10,19]. Therefore, from this point of view, it is likely that the enlarged mtDNA induced by THAM knockdown is caused by impaired segregation. Considering that Abf2 regulates the segregation of mtDNA in yeast, this function seems to be highly conserved among yeast and humans.
The relationship between the mtDNA distribution and other TFAM functions TFAM is also involved in the replication of mtDNA [10]. We observed that the inhibition of mtDNA replication using the inhibitor ddC did not produce enlarged mtDNA (Fig. 2), and no enlarged mtDNA was observed in PicoGreen-stained cells in which Twinkle, a helicase involved in the replication of mtDNA, was knocked down [35]. In addition, ddC treatment did not efficiently block the enlarged mtDNA phenotype induced by TFAM knockdown. Therefore, the enlarged mtDNA is likely to be independent of replication. As discussed above, the aggregation and impaired segregation of mtDNA are interlinked processes. The lack of mtDNA aggregation induced by the replication inhibitor also suggests that replication does not affect the segregation of mtDNA. If this is true, it is possible that the replication and segregation of mtDNA are independent processes in mitochondria and that mtDNA can be segregated without replication. One of the major functions of TFAM is the maintenance of the copy number of mtDNA. The widely accepted mechanism
E XP E RI ME N T AL C ELL R ES E AR C H 3 1 7 ( 2 0 11 ) 2 1 0– 2 20
underlying the regulation of mtDNA copy number is a simple titration model because the level of TFAM protein and amount of mtDNA are closely related [7]. However, the mechanism is not all because the amounts of TFAM protein and mtDNA do not always correlate [10]. In yeast, the mmm1 mutant shows a reduced mtDNA copy number, possibly due to its defective transmission [36]. In the case of mammalian cells, the defective segregation of mtDNA by TFAM knockdown might be one of causes that leads to a reduction of the mtDNA copy number.
Conclusions Considering the above, we conclude that human TFAM is required for the proper distribution of mtDNA in human cultured cells. Recently, it has been suggested that asymmetric mtDNA segregation is involved in human mtDNA diseases [5]. Therefore, downregulation or mutation of TFAM may affect the segregation of particular mtDNA molecules in human tissues. Elucidating the mechanism of this process would provide an insight into the segregation and transmission of heteroplasmic mtDNA mutations in human mtDNA diseases. Supplementary data to this article can be found online at doi:10.1016/j.yexcr.2010.10.008.
Acknowledgments We are grateful to Dr. Kugao Oishi for reading the manuscript and his valuable comments. This publication was subsidized by JKA through its promotion funds from KEIRIN RACE.
REFERENCES
[1] H.A. Tuppen, E.L. Blakely, D.M. Turnbull, R.W. Taylor, Mitochondrial DNA mutations and human disease, Biochem. Biophys. Acta. (Bioenergetics) 1797 (2010) 113–128. [2] A. Malena, E. Loro, M. Di Re, I.J. Holt, L. Vergani, Inhibition of mitochondrial fission favours mutant over wild-type mitochondrial DNA, Hum. Mol. Genet. 18 (2009) 3407–3416. [3] D.R. Dunbar, P.A. Moonie, H.T. Jacobs, I.J. Holt, Different cellular backgrounds confer a marked advantage to either mutant or wild-type mitochondrial genomes, Proc. Natl. Acad. Sci. USA 92 (1995) 6562–6566. [4] I.J. Holt, D.R. Dunbar, H.T. Jacobs, Behaviour of a population of partially duplicated mitochondrial DNA molecules in cell culture: segregation, maintenance and recombination dependent upon nuclear background, Hum. Mol. Genet. 6 (1997) 1251–1260. [5] I.J. Holt, J. He, C.C. Mao, J.D. Boyd-Kirkup, P. Martinsson, H. Sembongi, A. Reyes, J.N. Spelbrink, Mammalian mitochondrial nucleoids: organizing an independently minded genome, Mitochondrion 7 (2007) 311–321. [6] X.J. Chen, R.A. Butow, The organization and inheritance of the mitochondrial genome, Nat. Rev. Genet. 6 (2005) 815–825. [7] D. Kang, S.H. Kim, N. Hamasaki, Mitochondrial transcription factor A (TFAM): roles in maintenance of mtDNA and cellular functions, Mitochondrion 7 (2007) 39–44. [8] N.G. Larsson, J. Wang, H. Wilhelmsson, A. Oldfors, P. Rustin, M. Lewandoski, G.S. Barsh, D.A. Clayton, Mitochondrial transcription factor A is necessary for mtDNA maintenance and embryogenesis in mice, Nat. Genet. 18 (1998) 231–236.
219
[9] T. Kanki, K. Ohgaki, M. Gaspari, C.M. Gustafsson, A. Fukuoh, N. Sasaki, N. Hamasaki, D. Kang, Architectural role of mitochondrial transcription factor A in maintenance of human mitochondrial DNA, Mol. Cell. Biol. 24 (2004) 9823–9834. [10] J.L. Pohjoismaki, S. Wanrooij, A.K. Hyvarinen, S. Goffart, I.J. Holt, J.N. Spelbrink, H.T. Jacobs, Alterations to the expression level of mitochondrial transcription factor A, TFAM, modify the mode of mitochondrial DNA replication in cultured human cells, Nucleic Acids Res. 34 (2006) 5815–5828. [11] D.J. Dairaghi, G.S. Shadel, D.A. Clayton, Addition of a 29 residue carboxyl-terminal tail converts a simple HMG box-containing protein into a transcriptional activator, J. Mol. Biol. 249 (1995) 11–28. [12] V. McCulloch, G.S. Shadel, Human mitochondrial transcription factor B1 interacts with the C-terminal activation region of h-mtTFA and stimulates transcription independently of its RNA methyltransferase activity, Mol. Cell. Biol. 23 (2003) 5816–5824. [13] K. Ohgaki, T. Kanki, A. Fukuoh, H. Kurisaki, Y. Aoki, M. Ikeuchi, S.H. Kim, N. Hamasaki, D. Kang, The C-terminal tail of mitochondrial transcription factor A markedly strengthens its general binding to DNA, J. Biochem. 141 (2007) 201–211. [14] T.S. Wong, S. Rajagopalan, S.M. Freund, T.J. Rutheford, A. Andreeva, F.M. Townsley, M. Petrovich, A.R. Fersht, Biophysical characterizations of human mitochondrial transcription factor A and its binding to tumor suppressor p53, Nucleic Acids Res. 37 (2009) 6765–6783. [15] B.A. Kaufman, N. Durisic, J.M. Mativetsky, S. Costantino, M.A. Hancock, P. Grutter, E.A. Shoubridge, The mitochondrial transcription factor TFAM coordinates the assembly of multiple DNA molecules into nucleoid-like structures, Mol. Biol. Cell 18 (2007) 3225–3236. [16] T.A. Gangelhoff, P.S. Mungalachetty, J.C. Nix, M.E. Churchill, Structural analysis and DNA binding of the HMG domains of the human mitochondrial transcription factor A, Nucleic Acids Res. 37 (2009) 3153–3164. [17] J.F. Diffley, B. Stillman, A close relative of the nuclear, chromosomal high-mobility group protein HMG1 in yeast mitochondria, Proc. Natl. Acad. Sci. USA 88 (1991) 7864–7868. [18] O. Zelenaya-Troitskaya, S.M. Newman, K. Okamoto, P.S. Perlman, R.A. Butow, Functions of the high mobility group protein, Abf2p, in mitochondrial DNA segregation, recombination and copy number in Saccharomyces cerevisiae, Genetics 148 (1998) 1763–1776. [19] K. Kasashima, M. Sumitani, M. Satoh, H. Endo, Human Prohibitin 1 maintains the organization and stability of the mitochondrial nucleoids, Exp. Cell Res. 314 (2008) 988–996. [20] M. Sumitani, K. Kasashima, E. Ohta, D. Kang, H. Endo, Association of a novel mitochondrial protein M19 with mitochondrial nucleoids, J. Biochem. 146 (2009) 725–732. [21] D.H. Margineantu, W. Gregory Cox, L. Sundell, S.W. Sherwood, J. M. Beechem, R.A. Capaldi, Cell cycle dependent morphology changes and associated mitochondrial DNA redistribution in mitochondria of human cell lines, Mitochondrion 1 (2002) 425–435. [22] K. Kasashima, E. Ohta, Y. Kagawa, H. Endo, Mitochondrial functions and estrogen receptor-dependent nuclear translocation of pleiotropic human prohibitin 2, J. Biol. Chem. 281 (2006) 36401–36410. [23] N. Ashley, D. Harris, J. Poulton, Detection of mitochondrial DNA depletion in living human cells using PicoGreen staining, Exp. Cell Res. 303 (2005) 432–446. [24] N. Ashley, J. Poulton, Anticancer DNA intercalators cause p53dependent mitochondrial DNA nucleoid re-modelling, Oncogene 28 (2009) 3880–3891. [25] D.F. Bogenhagen, D. Rousseau, S. Burke, The layered structure of human mitochondrial DNA nucleoids, J. Biol. Chem. 283 (2008) 3665–3675. [26] N. Garrido, L. Griparic, E. Jokitalo, J. Wartiovaara, A.M. van der Bliek, J.N. Spelbrink, Composition and dynamics of human mitochondrial nucleoids, Mol. Biol. Cell 14 (2003) 1583–1596.
220
E XP ER I ME NT AL CE L L R ES E AR CH 3 1 7 (2 0 11 ) 2 10 – 2 20
[27] S. Mili, S. Pinol-Roma, LRP130, a pentatricopeptide motif protein with a noncanonical RNA-binding domain, is bound in vivo to mitochondrial and nuclear RNAs, Mol. Cell. Biol. 23 (2003) 4972–4982. [28] R.W. Gilkerson, E.A. Schon, E. Hernandez, M.M. Davidson, Mitochondrial nucleoids maintain genetic autonomy but allow for functional complementation, J. Cell Biol. 181 (2008) 1117–1128. [29] M. Di Re, H. Sembongi, J. He, A. Reyes, T. Yasukawa, P. Martinsson, L.J. Bailey, S. Goffart, J.D. Boyd-Kirkup, T.S. Wong, A.R. Fersht, J.N. Spelbrink, I.J. Holt, The accessory subunit of mitochondrial DNA polymerase gamma determines the DNA content of mitochondrial nucleoids in human cultured cells, Nucleic Acids Res. 37 (2009) 5701–5713. [30] S. Goffart, H. M. Cooper, H. Tyynismaa, S. Wanrooij, A. Suomalainen, J. N. Spelbrink, Twinkle mutations associated with autosomal dominant progressive external ophthalmoplegia lead to impaired helicase function and in vivo mtDNA replication stalling, Hum. Mol. Genet. 18, 328–340. [31] J. He, C.C. Mao, A. Reyes, H. Sembongi, M. Di Re, C. Granycome, A.B. Clippingdale, I.M. Fearnley, M. Harbour, A.J. Robinson, S. Reichelt, J.N. Spelbrink, J.E. Walker, I.J. Holt, The AAA+protein ATAD3 has
[32]
[33]
[34]
[35]
[36]
displacement loop binding properties and is involved in mitochondrial nucleoid organization, J. Cell Biol. 176 (2007) 141–146. F. Hayes, D. Barilla, The bacterial segrosome: a dynamic nucleoprotein machine for DNA trafficking and segregation, Nat. Rev. Microbiol. 4 (2006) 133–143. D. Lockshon, S.G. Zweifel, L.L. Freeman-Cook, H.E. Lorimer, B.J. Brewer, W.L. Fangman, A role for recombination junctions in the segregation of mitochondrial DNA in yeast, Cell 81 (1995) 947–955. C.L. Doe, F. Osman, J. Dixon, M.C. Whitby, The Holliday junction resolvase SpCCE1 prevents mitochondrial DNA aggregation in Schizosaccharomyces pombe, Mol. Gen. Genet. 263 (2000) 889–897. H. Tyynismaa, H. Sembongi, M. Bokori-Brown, C. Granycome, N. Ashley, J. Poulton, A. Jalanko, J.N. Spelbrink, I.J. Holt, A. Suomalainen, Twinkle helicase is essential for mtDNA maintenance and regulates mtDNA copy number, Hum. Mol. Genet. 13 (2004) 3219–3227. A.E. Hobbs, M. Srinivasan, J.M. McCaffery, R.E. Jensen, Mmm1p, a mitochondrial outer membrane protein, is connected to mitochondrial DNA (mtDNA) nucleoids and required for mtDNA stability, J. Cell Biol. 152 (2001) 401–410.
available at www.sciencedirect.com
www.elsevier.com/locate/yexcr
Research Article
Human mitochondrial transcription factor A is required for the segregation of mitochondrial DNA in cultured cells Katsumi Kasashima⁎, Megumi Sumitani, Hitoshi Endo Department of Biochemistry, Jichi Medical University, Tochigi 329-0498, Japan
A R T I C L E I N F O R M A T I O N
A B S T R A C T
Article Chronology:
The segregation and transmission of the mitochondrial genome in humans are complicated
Received 2 September 2010
processes but are particularly important for understanding the inheritance and clinical
Revised version received
abnormalities of mitochondrial disorders. However, the molecular mechanism of the
8 October 2010
segregation of mitochondrial DNA (mtDNA) is largely unclear. In this study, we demonstrated
Accepted 9 October 2010
that human mitochondrial transcription factor A (TFAM) is required for the segregation of mtDNA
Available online 15 October 2010
in cultured cells. RNAi-mediated knockdown of TFAM in HeLa cells resulted in the enlarged mtDNA, as indicated by the assembly of fluorescent signals stained with PicoGreen. Fluorescent in
Keywords:
situ hybridization confirmed the enlarged mtDNA and further showed the existence of increased
Mitochondria
numbers of mitochondria lacking mtDNA signals in TFAM knockdown cells. By complementation
TFAM
analysis, the C-terminal tail of TFAM, which enhances its affinity with DNA, was found to be
mtDNA
required for the appropriate distribution of mtDNA. Furthermore, we found that TFAM knockdown
Segregation
induced asymmetric segregation of mtDNA between dividing daughter cells. These results suggest
Nucleoid
an essential role for human TFAM in symmetric segregation of mtDNA. © 2010 Elsevier Inc. All rights reserved.
Introduction Mitochondria are semi-autonomous organelles that contain their own genomic DNA called mitochondrial DNA (mtDNA). Mammalian mtDNA is a closed circular molecule of about 16 kbp in length and exists in multiple copies per cell. Human mtDNA encodes proteins that are essential components of respiratory chain complexes. Mutations in human mtDNA are closely associated with neuromuscular diseases, such as MELAS [1]. mtDNA is maternally inherited, and mammalian mtDNA variants segregate rapidly between generations. Mutated mtDNA molecules some-
times co-exist with wild-type mtDNA, which is called heteroplasmy. How the mutant mtDNA molecules become fixed is controversial issue, and it is likely that the mechanism is not explained only by random genetic drift [2]. Several studies have demonstrated that particular mtDNA variants show non-random segregation in human cells [3,4]. The biased segregation of mtDNA variants is dependent on cell type and is suggested to play a role in human mtDNA diseases and mutant load [2,5]. Thus, the segregation of mtDNA seems to be important for understanding the inheritance and clinical abnormalities of mitochondrial disorders; however, its molecular basis is largely unknown.
⁎ Corresponding author. Department of Biochemistry, Jichi Medical University, 3311-1 Yakushiji, Shimotsuke, Tochigi 329-0498, Japan. Fax: +81 285 44 1827. E-mail address: [email protected] (K. Kasashima). Abbreviations: TFAM, mitochondrial transcription factor A; mtDNA, mitochondrial DNA; FISH, fluorescent in situ hybridization; MELAS, mitochondrial myopathy, encephalopathy, lactic acidosis, stroke-like episodes; OPA1, optic atrophy 1; ddC, dideoxycytidine; PHB2, prohibitin 2; mtSSB, mitochondrial single stranded DNA-binding protein; LRP130, leucine-rich protein 130 kDa; GAPDH, glyceraldehydes 3-phosphate dehydrogenase; RNAi, RNA interference; POLG, mtDNA polymerase gamma; siRNA, small interfering RNA; D-loop, displacement loop; rRNA, ribosomal RNA; rTFAM, recombinant TFAM 0014-4827/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.yexcr.2010.10.008
E XP E RI ME N T AL C ELL R ES E AR C H 3 1 7 ( 2 0 11 ) 2 1 0– 2 20
Recent studies have shown that mtDNA exists not as naked DNA but rather as a highly organized structure associated with many proteins (the nucleoprotein complex) as in the nuclear genome. These complex structures are called mitochondrial nucleoids, and the nucleoid components are believed to regulate the stability, replication, transcription, and segregation of mtDNA [6]. Mitochondrial transcription factor A (TFAM) is known to be a major nucleoid protein that binds directly to mtDNA [7]. In addition to its role as a transcription factor for mtDNA, mammalian TFAM acts to maintain the copy number of mtDNA [8,9] and is involved in its replication [10]. TFAM contains two high mobility group (HMG)-box domains, which are required for DNA binding, and a C-terminal tail abundant in basic amino acid residues. The C-terminal tail was originally identified as a domain required for the activation of transcription [11] that interacts with mitochondrial transcription factor B1 [12]. Recently, it has been shown that the C-terminal tail is important for efficient interaction between TFAM and DNA [13,14]. In molecular and structural studies, it was found that TFAM acts as a homodimer during its interaction with DNA [14–16] and that the Cterminal tail facilitates its formation [14]. Therefore, the C-terminal tail seems to be an indispensable domain for TFAM function. In yeast, Abf2, a yeast homologue of TFAM, maintains the mitochondrial genome and also regulates its segregation within zygotes [17,18]. The segregation of mtDNA has been extensively studied in yeast because several mtDNA deletion variants are available and their transmission can be easily tracked. mtDNA segregation is considered to be a directed process that is driven by a putative segregation apparatus; however, the molecular mechanism underlying it remains largely unknown, even in yeast. In mammals, the factors that regulate the segregation of mtDNA into dividing cells have not been reported. In this study, we identified a novel function of human TFAM: it is required for the segregation of mtDNA in cultured cells. RNAimediated knockdown of TFAM in HeLa cells resulted in the enlarged mtDNA signals produced by staining with the DNAspecific fluorescent dye PicoGreen. Fluorescent in situ hybridization (FISH) confirmed that the PicoGreen-stained signals reflected the enlarged mtDNA. The FISH results also showed that the number of mitochondria lacking mtDNA signals was increased in these cells, suggesting that the segregation of mtDNA is impaired by TFAM knockdown. In a complementation assay, the C-terminal tail of TFAM, which enhances its affinity with DNA [13,14], was found to be required for the proper distribution of mtDNA. Furthermore, it was shown that TFAM knockdown induced asymmetric segregation of mtDNA between dividing daughter cells. Here, we demonstrate that human TFAM is required for symmetric segregation of mtDNA in cultured cells and provide an insight into the segregation apparatus of mtDNA in mammals. These results suggest that TFAM participates in the asymmetric segregation and transmission of heteroplasmic mtDNA mutations in mitochondrial disorders.
Materials and methods Plasmid construction For RNA interference (RNAi), small interfering RNA (siRNA) sequences for TFAM, OPA1, and M19 were introduced into the pSilencer 3.1-H1
211
Puro vector (Ambion), as described previously [19,20]. For targeting of the 3′ UTR of human TFAM, another siRNA sequence (5′- AAGATTGAGATGTGTTCACAA-3′) was designed and introduced into the expression vector. The coding region of TFAM and its C-terminal deletion mutant (amino acids 1–221) were amplified by PCR from a human heart cDNA library, and the PCR products were introduced into the mammalian expression vector pIRESpuro3 (Clontech).
Cell culture and transfection HeLa and ρ0 HeLa cells were cultured as previously described [19]. HeLa cells were also cultured in medium containing 10 μM dideoxycytidine (ddC) for the indicated periods. Transfection was performed using Lipofectamine 2000 (Invitrogen) according to the manufacturer's instructions. HeLa cells were plated on 35mm poly-L-lysine-coated glass-bottomed dishes (Matsunami Glass Ind.). The mitochondria in living HeLa cells were stained with MitoTracker Red CM-H2XRos (250 nM, Molecular Probes) for 30 min at 37 °C, and mtDNA was stained with PicoGreen solution (3 μl/ml, Molecular Probes) for 60 min at 37 °C. Cells expressing pSilencer 3.1-H1 Puro constructs were selected with 2 μg/ml of puromycin (Sigma). Fluorescent images were captured and analyzed with a μRadiance™ Laser Scanning Confocal Microscope System (Bio-Rad). For the synchronization of cells, HeLa cells were treated with 2 mM thymidine for 23 h and released from the G1/S phase by being placed in thymidine-free medium for 11 h in order to restart the cell cycle.
Fluorescent in situ hybridization FISH was performed according to a slightly modified version of a previously described method [21]. Two micrograms of the 1.1 kb displacement loop (D-loop) mtDNA fragment (nt: 16029-599) was labeled with Alexa Fluor 488 fluorescent dye (ARES DNA labeling kit, Molecular Probes) by nick translation (Roche). The probe was ethanol precipitated with herring sperm DNA (Promega) as a carrier and COT-1 DNA (Invitrogen) as a general blockant, and resolved in hybridization buffer (50% deionized formamide, 10% Dextran Sulfate, 2× SSC). For hybridization, we used a probe that was diluted to a final concentration of about 2 ng/μl with hybridization buffer. HeLa cells were stained with MitoTracker Red CM-H2XRos (2.5 μM) for 30 min, fixed for 20 min at room temperature with 4% paraformaldehyde in PBS, and then treated for 15 min with 0.1% Triton X-100 in PBS. The cells were treated with RNase A (100 μg/ml in 2× SSC) for 1.5 h at 37 °C and then dehydrated and rehydrated using 70%, 90%, and 100% ethanol for 2 min at each step. Then, prehybridization was performed using 2× SSC buffer (2× SSC, 0.3 M NaCl, and 0.04 M sodium citrate) for 1 h at 37 °C, and the cells were dehydrated using a graded series of ethanols. The cells were then denatured with 70% deionized formamide in 2× SSC for 5 min at 72 °C, dehydrated again with ice-cold ethanol (70% and 90%) for 1 min at each step, and then stored in 100% ethanol at room temperature. After denaturing the probe for 5 min at 72 °C, it was applied to the cells, and hybridization was performed overnight at 37 °C in a humid chamber in the dark. After washing the cells with 0.4× SSC with 0.3% Tween-20 for 5 min at 72 °C and 2× SSC with 0.1% Tween-20 for 1 min at room temperature, fluorescent images were captured with the Axio Observer D1 system (Carl Zeiss) or the Leica TCS SP5 confocal microscope system (Leica).
212
E XP ER I ME NT AL CE L L R ES E AR CH 3 1 7 (2 0 11 ) 2 10 – 2 20
Combined immunofluorescence and FISH After 11 h of thymidine release, the cells were fixed for 15 min with 4% formaldehyde in PBS and then treated for 15 min with 0.1% Triton X-100 in PBS. The cells were then probed with anti-αtubulin mouse monoclonal antibody (DM1A, Abcam) and labeled with Cy3-conjugated anti-mouse IgG antibody (Molecular Probes). Mitochondria were also immunostained with anti-PHB2 rabbit polyclonal antibody (the product produced by immunization with both GST and His-tagged PHB2 protein), anti-OPA1 (our product), anti-mtSSB (Sigma Aldrich), or anti-LRP130 (GeneTex, Inc.) rabbit polyclonal antibody. The signals were visualized with Alexa Fluor 488, Cy3-conjugated, or Cy5-conjugated anti-rabbit IgG (Molecular Probes, GE Healthcare). After the immunofluorescence, the cells were fixed again for 20 min with 4% paraformaldehyde in PBS, and FISH was performed as described above.
Western blotting Samples were separated by electrophoresis on SDS-polyacrylamide gels (12% acrylamide) and electrophoretically transferred to a nitrocellulose membrane (Hybond ECL, GE Healthcare). The membrane was probed with antibodies and detected with an enhanced chemiluminescence system (GE Healthcare), as previously described [22]. The following primary antibodies were used: anti-TFAM (1:100; SantaCruz), anti-glyceraldehyde 3-phosphate dehydrogenase (GAPDH, 1:3000; Chemicon), and anti-OPA1 (1:100, our product).
Quantitative PCR Genomic DNA from HeLa cells was extracted by the standard Proteinase K digestion method [19]. LightCycler-FastStart DNA Master SYBR Green I (Roche) was used for the quantitative PCR with LightCycler (Roche). To produce a standard curve, 0.5, 1, 2, 4, and 8 ng (for mtDNA amplification) or 1, 2, 4, 8, and 16 ng (for 18S ribosomal RNA (rRNA) gene amplification) of HeLa cell genomic DNA were used. For the amplification of mtDNA (132 bp) and 18S rRNA (104 bp) fragments, the following primer sets were used: mtDNA: forward, 5′-GCCTGCCTGATCCTCCAAAT-3′ (nt: 14,862– 14,881), reverse, 5′-AAGGTAGCGGATGATTCAGCC-3′ (nt: 14,973– 14,993), 18S rRNA: forward, 5′-TAGAGGGACAAGTGGCGTTC-3′, reverse, 5′-CGCTGAGCCAGTCAGTGT-3′.
[23], about half of the knockdown cells showed large areas displaying intense PicoGreen signals but a reduced numbers of signals (Fig. 1B, Supplemental Fig. 2). PicoGreen signals were confirmed to be specific to mtDNA because ρ0 HeLa cells that lack mtDNA completely were not stained by the dye (Supplemental Fig. 3). The brightly stained PicoGreen fluorescent signals, in which their size was the twice or more larger than that in control cells and the number was reduced to less than half of the control, were defined as enlarged mtDNA. To investigate the relationship between knockdown efficiency of TFAM and the enlarged mtDNA phenotype, they were examined at 2 to 4 days after transfection with the RNAi vector. In the analysis, knockdown of TFAM expression became high as days go by (Supplemental Fig. 4). By counting the number of cells containing enlarged mtDNA, its frequency was found to be increased in parallel with the knockdown (Supplemental Fig. 4). Therefore, the decrease of TFAM expression correlates with the formation frequency of enlarged mtDNA, which is similar to the relationship between TFAM protein level and mtDNA copy number [9]. No enlarged fluorescent signals was observed in the non-transfected control or the cells transfected with an RNAi vector targeting another mitochondrial nucleoid protein, M19 [20] (Supplemental Fig. 5), indicating that it is a phenotype specific to TFAM knockdown cells. PicoGreen and MitoTracker staining showed that the enlarged signals were detected in mitochondria and that a large number of mitochondria in the TFAM knockdown cells contained no visible mtDNA signals (Fig. 1C). These results strongly suggest that the mechanism for the correct distribution of mtDNA in newly dividing mitochondria does not work well in TFAM knockdown cells. We then directly detected mtDNA in TFAM knockdown cells by fluorescent in situ hybridization (FISH). In non-transfected control cells, the FISH signals were extensively distributed in mitochondria and detected non-specifically in nucleus (Fig. 1D). The mitochondrial signals were confirmed to be specific because they were hardly detected in HeLa cells containing only 11% of the control level of mtDNA after 4-day ddC treatment (Supplemental Fig. 6). ddC is an inhibitor of mtDNA replication, and its addition reduces the copy number of mtDNA [23]. FISH confirmed the presence of enlarged mtDNA signals by confocal microscopy and also showed the existence of a large number of mitochondria lacking mtDNA in the TFAM knockdown cells (Fig. 1D). These results confirm that the PicoGreen-stained signals seen in the TFAM knockdown cells represented the enlarged mtDNA signals.
No enlarged mtDNA was observed after the inhibition of its replication
Results Enlarged mtDNA signals in TFAM knockdown cells TFAM maintains the copy number of mtDNA in mammals [8,9]. Using RNA interference (RNAi), we successfully and specifically silenced human TFAM in HeLa cells (Fig. 1A). When we performed knockdown of TFAM in HeLa cells, these cells showed a reduction in the amount of mtDNA they contained on quantitative PCR analysis (Supplemental Fig. 1). Since the same result was previously obtained by Southern blot analysis [19], we quantified the amount of mtDNA by this method. In addition to the abovementioned phenotype, when TFAM knockdown cells were stained with PicoGreen, which quantitatively stained mtDNA in living cells
TFAM also regulates the replication of mtDNA [10]. To test whether the enlarged mtDNA is caused by the inhibition of mtDNA replication in TFAM knockdown cells, HeLa cells were treated with the replication inhibitor ddC. Under treatment with ddC, the intensity of the mtDNA signals stained with PicoGreen and the amount of mtDNA, which was measured by quantitative PCR, gradually decreased (Fig. 2 A, B). However, no enlarged signals were observed in these cells. After 24-h treatment with ddC, the amount of mtDNA was decreased to about half of the control level (Fig. 2B), which was almost same as that caused by TFAM knockdown (Supplemental Fig. 1). Even in these conditions, no enlarged mtDNA was observed. We also confirmed that the TFAM protein level was not significantly affected by 24 h ddC treatment
E XP E RI ME N T AL C ELL R ES E AR C H 3 1 7 ( 2 0 11 ) 2 1 0– 2 20
213
Fig. 1 – Enlarged mtDNA signals in TFAM knockdown cells. (A) At 4 days after transfection with the RNAi vector for TFAM, protein expression was determined by Western blotting using the indicated antibodies. (B) The TFAM-knocked down HeLa cells were stained with PicoGreen, and the images were obtained by confocal microscopy. The square boxes represent the area enlarged. Quantification of the percentage of cells that contains enlarged mtDNA signals is shown. (C) The TFAM-knocked down HeLa cells were stained with PicoGreen and MitoTracker, and the images were obtained by confocal microscopy. (D) At 4 days after transfection with the RNAi vector, the cells and non-transfected control cells were probed with mtDNA probes (green signals). The cells were also stained with MitoTracker (red signals), and the merged images were obtained with a fluorescent microscope or a confocal microscope (confocal image). mtDNA probes also stained nucleus non-specifically (Supplemental Figure 6). FISH image of siTFAM was obtained by using shorter exposure time than control image because the intensity of the mtDNA signals was strong. Scale bar: 10 μm.
(Fig. 2C). These results indicate that the enlarged mtDNA signals seen in the TFAM knockdown cells were not caused by the inhibition of replication. We also observed that, at least, 24-h treatment with ddC did not efficiently block the enlarged phenotype induced by TFAM knockdown (Supplemental Fig. 7).
The enlarged mtDNA by TFAM knockdown occurs independently of mitochondrial fusion and reflects assembled nucleoids Recently, Ashley et al. reported that DNA intercalators cause enlarged mtDNA (also called aggregation of mtDNA nucleoids), which they termed remodeled nucleoids [24]. The remodeling of nucleoids
requires the fusion machinery contained in mitochondria because it is suppressed by the knockdown of the pro-fusion proteins mitofusin 1 (MFN1) or optic atrophy 1 (OPA1) [24]. So, we examined the effect of mitochondrial fusion on the aggregation of mtDNA caused by TFAM knockdown. We performed double knockdown of TFAM and OPA1 in HeLa cells and confirmed that they were specifically knocked down by RNAi (Fig. 3A). We also confirmed that OPA1 was efficiently knocked down in cells containing fragmented mitochondria by immunofluoresence (Supplemental Fig. 8). In the double knockdown cells, enlarged mtDNA was still observed in fragmented mitochondria (Fig. 3B), as was observed in tubular mitochondria in TFAM knockdown or control cells (Fig. 3B). No enlarged mtDNA was seen in the fragmented mitochondria, in which only OPA1 was knocked down (Fig. 3B). These
214
E XP ER I ME NT AL CE L L R ES E AR CH 3 1 7 (2 0 11 ) 2 10 – 2 20
Fig. 2 – Absence of the enlarged mtDNA in replication inhibited cells. HeLa cells were treated with the inhibitor of the mtDNA replication ddC for the indicated periods. (A) The cells were stained with PicoGreen, and the images were obtained by confocal microscopy. Scale bar: 10 μm. (B) The mtDNA content of these ddC-treated cells was quantified by quantitative PCR. The amount of mtDNA per nuclear gene (18S rRNA) is shown. (C) The protein expression in these ddC-treated cells was determined by Western blotting using the indicated antibodies.
results indicate that the enlarged mtDNA caused by TFAM knockdown is independent of mitochondrial fusion and is not composed of remodeled nucleoids. Next, to reveal whether the enlarged mtDNA induced by TFAM knockdown reflects the enlarged nucleoids, submitochondrial localization of the nucleoid proteins, such as mitochondrial single-stranded DNA-binding protein (mtSSB) and leucine-rich protein 130 kDa (LRP130), was investigated. mtSSB is a core nucleoid protein that shows similar localization with mtDNA [25,26]. When TFAM knockdown cells were immunostained with mtSSB antibody, the immunofluorescent signals were found to be assembled, which corresponded to the mtDNA signals by FISH (Fig. 3C). In contrast, the immunofluorescent signals against LRP130, a mitochondrial RNAbinding protein included in the nucleoids [25,27], were overlapped with mtDNA signals in TFAM knockdown cells but were not assembled (Fig. 3C). These results strongly suggest that the enlarged mtDNA seen
in the TFAM knockdown cells reflects the enlarged nucleoids, which are enriched with core nucleoid proteins, such as mtSSB.
The C-terminal tail of TFAM is required for proper distribution of mtDNA To clarify the molecular mechanism underlying the distribution of mtDNA induced by TFAM, we performed a complementation assay using RNAi and the expression of recombinant TFAM (rTFAM). RNAi targeting of the 3′ UTR of TFAM was also performed to avoid the suppression of rTFAM expression by the RNAi vector. The novel RNAi vector efficiently and specifically silenced TFAM expression in HeLa cells (Fig. 4A). After PicoGreen staining, the knockdown cells showed enlarged fluorescent signals (Fig. 4B), further confirming that the phenotype is specific to TFAM knockdown. We then transfected the RNAi vector with the TFAM expression
Fig. 3 – The enlarged mtDNA induced by TFAM knockdown occurs independently of mitochondrial fusion. (A) HeLa cells were double knocked down with TFAM and OPA1 RNAi vectors. At 4 days after transfection with the RNAi vectors or the control vector (puro-vec.), protein expression was determined by Western blotting using the indicated antibodies. (B) The double knockdown and single knockdown (TFAM or OPA1) cells were stained with PicoGreen and MitoTracker. Enlarged mtDNA induced by TFAM knockdown was still observed in fragmented mitochondria caused by OPA1 knockdown. (C) At 4 days after transfection with the RNAi vector for TFAM, the knockdown cells were probed with the indicated antibodies against the nucleoid proteins by immunofluorescence and mtDNA probe by FISH. Scale bar: 10 μm.
E XP E RI ME N T AL C ELL R ES E AR C H 3 1 7 ( 2 0 11 ) 2 1 0– 2 20
215
216
E XP ER I ME NT AL CE L L R ES E AR CH 3 1 7 (2 0 11 ) 2 10 – 2 20
Fig. 4 – The C-terminal tail of TFAM is required for proper distribution of mtDNA. At 4 days after transfection with a novel RNAi vector targeting the 3′ UTR of TFAM (siTFAM-3U) or the control vector (puro-vec.), protein expression was quantified by Western blotting using the indicated antibodies (A), and the cells were stained with PicoGreen (B). The images were obtained by confocal microscopy. Scale bar: 10 μm. (C) At 3 days after co-transfection with the RNAi vector for TFAM-3U and the indicated rTFAM expression plasmid, protein expression was quantified by Western blotting using the indicated antibodies. (D) At four days after co-transfection with the RNAi vector and rTFAM expression plasmid, the proportion of cells containing enlarged mtDNA was measured by PicoGreen staining in three independent experiments (in each experiment, the total cell number was more than 100 cells).
plasmid and examined whether the enlarged mtDNA was suppressed by its expression. The expression of rTFAM significantly rescued the enlarged phenotype (Fig. 4 C, D), confirming that the phenotype is caused by the downregulation of TFAM. The C-terminal tail of TFAM, which is required for efficient interaction between TFAM and DNA, is known to be important for TFAM functions, such as transcription [11]. So, we examined its importance for the distribution of mtDNA. When a deletion mutant of TFAM lacking the C-terminal tail (TFAM-delta C) was used in the complementation experiment, it did not rescue the phenotype as well as the wild type (Fig. 4C, D). Because the precursor protein of TFAM-delta C, which has similar molecular weight with endogenous TFAM, was detected by Western blotting (Supplemental Fig. 9), the endogenous TFAM protein in the rescue with the mutant was estimated slightly high (Fig. 4C). We also confirmed that no enlarged mtDNA was observed by the expression of TFAM-delta C only (Supplemental Fig. 10), indicating that the mutant does not work as dominant-negative mutant. As has been reported by other group [9], the C-terminal deletion mutant was sufficient for the rescue of mtDNA copy number
in HeLa cells (Supplemental Fig. 11). Thus, the C-terminal tail is required for the proper distribution of mtDNA by TFAM.
TFAM knockdown induces asymmetric segregation of mtDNA between dividing two daughter cells The enlarged mtDNA caused by TFAM knockdown strongly suggests that the segregation of mtDNA is impaired in these cells. To obtain direct evidence to support our hypothesis that TFAM is required for mtDNA segregation, the distribution of mtDNA in dividing cells was investigated. HeLa cells were synchronized at the G1/S phase using the single thymidine block procedure and subsequently released into thymidine-free medium to restart their cell cycles. Eleven hours after their release, most cells were in the cytokinesis phase, which was confirmed by the characteristic localization of α-tubulin at the midbody-like structure present between dividing cells (Fig. 5A). Hereafter, these cells are referred to as midbody-positive cells. To investigate the segregation of mtDNA between two cells, we chose midbodypositive cells that had been produced by division of the same
E XP E RI ME N T AL C ELL R ES E AR C H 3 1 7 ( 2 0 11 ) 2 1 0– 2 20
217
Fig. 5 – TFAM knockdown induces asymmetric segregation of mtDNA between dividing cells. At 3 days after transfection with the RNAi vector for TFAM, the cells were synchronized with thymidine treatment as described in the Materials and Methods section. After 11 h of the release from the thymidine block, the cells were fixed. (A) The cells were immunostained with anti-α-tubulin (red) and anti-PHB2 (green) antibodies. Cells in the cytokinesis phase showed the localization of α-tubulin at a midbody-like structure (midbody-positive cells). (B) The cells were subjected to combined immunofluorescence and FISH. The cells were immunostained with anti-α-tubulin (red) and probed with mtDNA probes (green) and observed at several z-axis. For TFAM knockdown cells, two representative images (asymmetric (left) and symmetric (right) distribution of mtDNA) are shown. Merged images are also shown. (C) Cell populations containing symmetric or asymmetric segregation of mtDNA is shown. In the control, all ‘midbody-positive’ cells showed symmetric segregation of mtDNA. In the TFAM knockdown cells, which contained enlarged mtDNA and were ‘midbody-positive’, the symmetric segregation was significantly impaired. The criteria in determining whether asymmetric or not are described in text. (D) The TFAM-knocked down HeLa cells were immunostained with anti-α-tubulin (red) and anti-PHB2 (blue) and probed with mtDNA probes (green), and a merged image is shown. The arrowheads represent aggregated mtDNA signals. Scale bar: 10 μm.
parent cell. We first observed the segregation of mitochondria in TFAM knockdown cells by labeling mitochondria with antibody against PHB2, which mainly localizes to mitochondria in HeLa cells
[22]. As was observed in control cells, mitochondria were segregated almost evenly between TFAM knockdown cells (Fig. 5A). Next, using combined immunofluorescence and FISH
218
E XP ER I ME NT AL CE L L R ES E AR CH 3 1 7 (2 0 11 ) 2 10 – 2 20
technology [20,28], the segregation of mtDNA between dividing cells was examined. We estimated the nucleoid number by mtDNA signals stained with FISH and found that each dividing cell contained about 100 to 200 nucleoids. The case, in which the nucleoid number in dividing one cell was the twice or more than that in the other cell, was defined as asymmetric distribution. As was predicted, mtDNA was almost symmetrically segregated between dividing cells in the control experiment (Fig. 5B). However, under TFAM knockdown, about half of the midbodypositive cells that contained enlarged mtDNA signals showed asymmetric segregation of mtDNA between dividing cells (Fig. 5B, C), which was estimated by counting the number of enlarged mtDNA at several z-axis (Supplemental Fig. 12). The knockdown efficiency of TFAM between the dividing cells was considered to be almost equal because the cells were just divided from same parent cell. We further confirmed that mitochondria were equally segregated in the TFAM knockdown cells, in which mtDNA was segregated asymmetrically (Fig. 5D). These results demonstrated that knockdown of TFAM induced asymmetric segregation of mtDNA.
Discussion In this study, we observed that mtDNA signals were enlarged within mitochondria in TFAM knockdown cells. Since mtDNA is depleted in TFAM knockout mice [8], this phenotype is only seen in knockdown studies. To date, similar phenotypes have been reported in ATAD3A knockdown [5], mtDNA polymerase gamma beta subunit (POLGbeta) or Twinkle mutant-overexpressed cells [29,30], and cells treated with DNA intercalators [24]. The AAA+ protein ATAD3A, a nucleoid factor, binds to the D-loop within mtDNA and regulates the organization of mitochondrial nucleoids [31]. In ATAD3A knockdown 143B cells, larger nucleoids were observed by PicoGreen staining at 72 h after treatment with siRNA [5]. In the case of overexpression of POLGbeta or Twinkle mutant, the number of nucleoids, which was identified by immunofluorescence with antiDNA antibodies, was reduced, and the nucleoids were enlarged in the cells [29,30]. These cases indicate the possibility that these nucleoid proteins are involved in the segregation of mtDNA. In this study, we demonstrated directly that symmetric segregation of mtDNA was significantly impaired in dividing cells by TFAM knockdown (Fig. 5). Although it is possible that the asymmetric distribution of mtDNA occurs stochastically by the reduced number of enlarged nucleoids, it is evident that TFAM is required for the proper distribution of mtDNA.
How does TFAM regulate the distribution of mtDNA? How does depletion of TFAM lead to enlarged mtDNA and asymmetric segregation of mtDNA? TFAM is a core nucleoid protein and wraps entire region of mtDNA [7]. Depletion of TFAM might alter the nucleoids organization and impair their segregation within mitochondria, which results in the enlarged mDNA nucleoids. For determining whether TFAM directly regulates the segregation of mtDNA, more experimental evidence would be required. Mitochondria share several similarities with their prototype bacteria. In bacteria, the segregation of plasmid DNA is controlled by ATPase and DNA-binding proteins [32]. Given that
TFAM contains no ATPase domain, an unknown factor that possesses an ATPase domain and binds to TFAM-mtDNA complex might regulate the segregation of nucleoids as a molecular motor. In this respect, ATAD3 is a candidate for this factor and might be involved in the regulation of segregation through its binding to TFAM–mtDNA complexes. Further investigation is required to examine this possibility. As shown in Fig. 4, the C-terminal tail of TFAM is required for appropriate distribution of mtDNA. Considering the importance of the domain in transcription, TFAM might regulate the segregation of mtDNA by a transcription-coupled mechanism. Alternatively, the C-terminal tail enhances the DNA-binding activity of TFAM, which leads to the formation of its dimmer and enables appropriate packaging of mtDNA [13,14]. Therefore, homodimerization of TFAM and proper packaging of mtDNA might be important for mtDNA segregation. As reported by other group [9], the C-terminal deletion mutant was sufficient for the rescue of mtDNA copy number in HeLa cells. Thus, the mechanism for maintaining mtDNA copy number might be different from that for the segregation. Yeast mtDNA shows a high frequency of recombination, and the removal of recombination junctions by resolving enzymes is required for efficient segregation of mtDNA [33,34]. These mutations causing reduced resolving activity induce the aggregation of mtDNA, which results in reduced segregation of mtDNA into daughter cells. Thus, it is possible that active aggregation causes the abnormal segregation of mtDNA. Considering that the recombination of mtDNA in mammalian cultured cells is rare, the possibility that mtDNA actively aggregates through impairment of the resolving activity of recombination junctions is low. Furthermore, interlinked molecules of mtDNA, which were observed in a resolvase yeast mutant, were not seen in TFAM knockdown cells because the electrophoretic mobility of mtDNA showed almost no changes in these cells [10,19]. Therefore, from this point of view, it is likely that the enlarged mtDNA induced by THAM knockdown is caused by impaired segregation. Considering that Abf2 regulates the segregation of mtDNA in yeast, this function seems to be highly conserved among yeast and humans.
The relationship between the mtDNA distribution and other TFAM functions TFAM is also involved in the replication of mtDNA [10]. We observed that the inhibition of mtDNA replication using the inhibitor ddC did not produce enlarged mtDNA (Fig. 2), and no enlarged mtDNA was observed in PicoGreen-stained cells in which Twinkle, a helicase involved in the replication of mtDNA, was knocked down [35]. In addition, ddC treatment did not efficiently block the enlarged mtDNA phenotype induced by TFAM knockdown. Therefore, the enlarged mtDNA is likely to be independent of replication. As discussed above, the aggregation and impaired segregation of mtDNA are interlinked processes. The lack of mtDNA aggregation induced by the replication inhibitor also suggests that replication does not affect the segregation of mtDNA. If this is true, it is possible that the replication and segregation of mtDNA are independent processes in mitochondria and that mtDNA can be segregated without replication. One of the major functions of TFAM is the maintenance of the copy number of mtDNA. The widely accepted mechanism
E XP E RI ME N T AL C ELL R ES E AR C H 3 1 7 ( 2 0 11 ) 2 1 0– 2 20
underlying the regulation of mtDNA copy number is a simple titration model because the level of TFAM protein and amount of mtDNA are closely related [7]. However, the mechanism is not all because the amounts of TFAM protein and mtDNA do not always correlate [10]. In yeast, the mmm1 mutant shows a reduced mtDNA copy number, possibly due to its defective transmission [36]. In the case of mammalian cells, the defective segregation of mtDNA by TFAM knockdown might be one of causes that leads to a reduction of the mtDNA copy number.
Conclusions Considering the above, we conclude that human TFAM is required for the proper distribution of mtDNA in human cultured cells. Recently, it has been suggested that asymmetric mtDNA segregation is involved in human mtDNA diseases [5]. Therefore, downregulation or mutation of TFAM may affect the segregation of particular mtDNA molecules in human tissues. Elucidating the mechanism of this process would provide an insight into the segregation and transmission of heteroplasmic mtDNA mutations in human mtDNA diseases. Supplementary data to this article can be found online at doi:10.1016/j.yexcr.2010.10.008.
Acknowledgments We are grateful to Dr. Kugao Oishi for reading the manuscript and his valuable comments. This publication was subsidized by JKA through its promotion funds from KEIRIN RACE.
REFERENCES
[1] H.A. Tuppen, E.L. Blakely, D.M. Turnbull, R.W. Taylor, Mitochondrial DNA mutations and human disease, Biochem. Biophys. Acta. (Bioenergetics) 1797 (2010) 113–128. [2] A. Malena, E. Loro, M. Di Re, I.J. Holt, L. Vergani, Inhibition of mitochondrial fission favours mutant over wild-type mitochondrial DNA, Hum. Mol. Genet. 18 (2009) 3407–3416. [3] D.R. Dunbar, P.A. Moonie, H.T. Jacobs, I.J. Holt, Different cellular backgrounds confer a marked advantage to either mutant or wild-type mitochondrial genomes, Proc. Natl. Acad. Sci. USA 92 (1995) 6562–6566. [4] I.J. Holt, D.R. Dunbar, H.T. Jacobs, Behaviour of a population of partially duplicated mitochondrial DNA molecules in cell culture: segregation, maintenance and recombination dependent upon nuclear background, Hum. Mol. Genet. 6 (1997) 1251–1260. [5] I.J. Holt, J. He, C.C. Mao, J.D. Boyd-Kirkup, P. Martinsson, H. Sembongi, A. Reyes, J.N. Spelbrink, Mammalian mitochondrial nucleoids: organizing an independently minded genome, Mitochondrion 7 (2007) 311–321. [6] X.J. Chen, R.A. Butow, The organization and inheritance of the mitochondrial genome, Nat. Rev. Genet. 6 (2005) 815–825. [7] D. Kang, S.H. Kim, N. Hamasaki, Mitochondrial transcription factor A (TFAM): roles in maintenance of mtDNA and cellular functions, Mitochondrion 7 (2007) 39–44. [8] N.G. Larsson, J. Wang, H. Wilhelmsson, A. Oldfors, P. Rustin, M. Lewandoski, G.S. Barsh, D.A. Clayton, Mitochondrial transcription factor A is necessary for mtDNA maintenance and embryogenesis in mice, Nat. Genet. 18 (1998) 231–236.
219
[9] T. Kanki, K. Ohgaki, M. Gaspari, C.M. Gustafsson, A. Fukuoh, N. Sasaki, N. Hamasaki, D. Kang, Architectural role of mitochondrial transcription factor A in maintenance of human mitochondrial DNA, Mol. Cell. Biol. 24 (2004) 9823–9834. [10] J.L. Pohjoismaki, S. Wanrooij, A.K. Hyvarinen, S. Goffart, I.J. Holt, J.N. Spelbrink, H.T. Jacobs, Alterations to the expression level of mitochondrial transcription factor A, TFAM, modify the mode of mitochondrial DNA replication in cultured human cells, Nucleic Acids Res. 34 (2006) 5815–5828. [11] D.J. Dairaghi, G.S. Shadel, D.A. Clayton, Addition of a 29 residue carboxyl-terminal tail converts a simple HMG box-containing protein into a transcriptional activator, J. Mol. Biol. 249 (1995) 11–28. [12] V. McCulloch, G.S. Shadel, Human mitochondrial transcription factor B1 interacts with the C-terminal activation region of h-mtTFA and stimulates transcription independently of its RNA methyltransferase activity, Mol. Cell. Biol. 23 (2003) 5816–5824. [13] K. Ohgaki, T. Kanki, A. Fukuoh, H. Kurisaki, Y. Aoki, M. Ikeuchi, S.H. Kim, N. Hamasaki, D. Kang, The C-terminal tail of mitochondrial transcription factor A markedly strengthens its general binding to DNA, J. Biochem. 141 (2007) 201–211. [14] T.S. Wong, S. Rajagopalan, S.M. Freund, T.J. Rutheford, A. Andreeva, F.M. Townsley, M. Petrovich, A.R. Fersht, Biophysical characterizations of human mitochondrial transcription factor A and its binding to tumor suppressor p53, Nucleic Acids Res. 37 (2009) 6765–6783. [15] B.A. Kaufman, N. Durisic, J.M. Mativetsky, S. Costantino, M.A. Hancock, P. Grutter, E.A. Shoubridge, The mitochondrial transcription factor TFAM coordinates the assembly of multiple DNA molecules into nucleoid-like structures, Mol. Biol. Cell 18 (2007) 3225–3236. [16] T.A. Gangelhoff, P.S. Mungalachetty, J.C. Nix, M.E. Churchill, Structural analysis and DNA binding of the HMG domains of the human mitochondrial transcription factor A, Nucleic Acids Res. 37 (2009) 3153–3164. [17] J.F. Diffley, B. Stillman, A close relative of the nuclear, chromosomal high-mobility group protein HMG1 in yeast mitochondria, Proc. Natl. Acad. Sci. USA 88 (1991) 7864–7868. [18] O. Zelenaya-Troitskaya, S.M. Newman, K. Okamoto, P.S. Perlman, R.A. Butow, Functions of the high mobility group protein, Abf2p, in mitochondrial DNA segregation, recombination and copy number in Saccharomyces cerevisiae, Genetics 148 (1998) 1763–1776. [19] K. Kasashima, M. Sumitani, M. Satoh, H. Endo, Human Prohibitin 1 maintains the organization and stability of the mitochondrial nucleoids, Exp. Cell Res. 314 (2008) 988–996. [20] M. Sumitani, K. Kasashima, E. Ohta, D. Kang, H. Endo, Association of a novel mitochondrial protein M19 with mitochondrial nucleoids, J. Biochem. 146 (2009) 725–732. [21] D.H. Margineantu, W. Gregory Cox, L. Sundell, S.W. Sherwood, J. M. Beechem, R.A. Capaldi, Cell cycle dependent morphology changes and associated mitochondrial DNA redistribution in mitochondria of human cell lines, Mitochondrion 1 (2002) 425–435. [22] K. Kasashima, E. Ohta, Y. Kagawa, H. Endo, Mitochondrial functions and estrogen receptor-dependent nuclear translocation of pleiotropic human prohibitin 2, J. Biol. Chem. 281 (2006) 36401–36410. [23] N. Ashley, D. Harris, J. Poulton, Detection of mitochondrial DNA depletion in living human cells using PicoGreen staining, Exp. Cell Res. 303 (2005) 432–446. [24] N. Ashley, J. Poulton, Anticancer DNA intercalators cause p53dependent mitochondrial DNA nucleoid re-modelling, Oncogene 28 (2009) 3880–3891. [25] D.F. Bogenhagen, D. Rousseau, S. Burke, The layered structure of human mitochondrial DNA nucleoids, J. Biol. Chem. 283 (2008) 3665–3675. [26] N. Garrido, L. Griparic, E. Jokitalo, J. Wartiovaara, A.M. van der Bliek, J.N. Spelbrink, Composition and dynamics of human mitochondrial nucleoids, Mol. Biol. Cell 14 (2003) 1583–1596.
220
E XP ER I ME NT AL CE L L R ES E AR CH 3 1 7 (2 0 11 ) 2 10 – 2 20
[27] S. Mili, S. Pinol-Roma, LRP130, a pentatricopeptide motif protein with a noncanonical RNA-binding domain, is bound in vivo to mitochondrial and nuclear RNAs, Mol. Cell. Biol. 23 (2003) 4972–4982. [28] R.W. Gilkerson, E.A. Schon, E. Hernandez, M.M. Davidson, Mitochondrial nucleoids maintain genetic autonomy but allow for functional complementation, J. Cell Biol. 181 (2008) 1117–1128. [29] M. Di Re, H. Sembongi, J. He, A. Reyes, T. Yasukawa, P. Martinsson, L.J. Bailey, S. Goffart, J.D. Boyd-Kirkup, T.S. Wong, A.R. Fersht, J.N. Spelbrink, I.J. Holt, The accessory subunit of mitochondrial DNA polymerase gamma determines the DNA content of mitochondrial nucleoids in human cultured cells, Nucleic Acids Res. 37 (2009) 5701–5713. [30] S. Goffart, H. M. Cooper, H. Tyynismaa, S. Wanrooij, A. Suomalainen, J. N. Spelbrink, Twinkle mutations associated with autosomal dominant progressive external ophthalmoplegia lead to impaired helicase function and in vivo mtDNA replication stalling, Hum. Mol. Genet. 18, 328–340. [31] J. He, C.C. Mao, A. Reyes, H. Sembongi, M. Di Re, C. Granycome, A.B. Clippingdale, I.M. Fearnley, M. Harbour, A.J. Robinson, S. Reichelt, J.N. Spelbrink, J.E. Walker, I.J. Holt, The AAA+protein ATAD3 has
[32]
[33]
[34]
[35]
[36]
displacement loop binding properties and is involved in mitochondrial nucleoid organization, J. Cell Biol. 176 (2007) 141–146. F. Hayes, D. Barilla, The bacterial segrosome: a dynamic nucleoprotein machine for DNA trafficking and segregation, Nat. Rev. Microbiol. 4 (2006) 133–143. D. Lockshon, S.G. Zweifel, L.L. Freeman-Cook, H.E. Lorimer, B.J. Brewer, W.L. Fangman, A role for recombination junctions in the segregation of mitochondrial DNA in yeast, Cell 81 (1995) 947–955. C.L. Doe, F. Osman, J. Dixon, M.C. Whitby, The Holliday junction resolvase SpCCE1 prevents mitochondrial DNA aggregation in Schizosaccharomyces pombe, Mol. Gen. Genet. 263 (2000) 889–897. H. Tyynismaa, H. Sembongi, M. Bokori-Brown, C. Granycome, N. Ashley, J. Poulton, A. Jalanko, J.N. Spelbrink, I.J. Holt, A. Suomalainen, Twinkle helicase is essential for mtDNA maintenance and regulates mtDNA copy number, Hum. Mol. Genet. 13 (2004) 3219–3227. A.E. Hobbs, M. Srinivasan, J.M. McCaffery, R.E. Jensen, Mmm1p, a mitochondrial outer membrane protein, is connected to mitochondrial DNA (mtDNA) nucleoids and required for mtDNA stability, J. Cell Biol. 152 (2001) 401–410.