Regulatory cis- and trans-elements of mitochondrial D-loop-driven reporter genes in budding tunicates

Mitochondrion xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Mitochondrion journal homepage: www.elsevier.com/locate/mito

Regulatory cis- and trans-elements of mitochondrial D-loop-driven reporter genes in budding tunicates Kaz Kawamuraa,⁎, Yuhya Saitoha, Loriano Ballarinb, Takeshi Sunanagaa a b

Laboratory of Cellular and Molecular Biotechnology, Faculty of Science, Kochi University, Kochi, Japan Department of Biology, Padova University, Padova, Italy

A R T I C L E I N F O

A B S T R A C T

Keywords: Ascidian Green fluorescent protein Histone methylation Non-coding region 16S ribosomal RNA TFAM

To unveil the underlying mechanism of mitochondrial gene regulation associated with ageing and budding in the tunicate Polyandrocarpa misakiensis, mitochondrial non-coding-region (NCR)-containing reporter genes were constructed. PmNCR2.3K/GFP was expressed spatiotemporally in a pattern quite similar to mitochondrial 16S rRNA. The reporter gene expression was sensitive to high dose of rifampicin similar to mitochondrial genes, suggesting that the transcription indeed occurs in mitochondria. However, the gene expression also occurred in vivo in the cell nucleus and in vitro in the nuclear extracts. Mitochondrial transcription factor A (PmTFAM) enhanced reporter gene expression, depending on the NCR length. A budding-specific polypeptide TC14-3 is an epigenetic histone methylation inducer. It heavily enhanced reporter gene expression that was interfered by histone methylation inhibitors and PmTFAM RNAi. Our results indicate for the first time that the nuclear histone methylation is involved in mitochondrial gene activity via TFAM gene regulation.

1. Introduction Mitochondria are multifunctional organelles that not only participate in ATP and fatty acid biosynthesis (Albert et al., 2014), but also influence embryonic development and cell death by their components such as large ribosomal RNA (16S rRNA) and cytochrome c (Gilbert, 2013). Recently, mitochondria have drawn increased attention regarding organism senescence and mortality, since their activity declines with age (Byrne et al., 1991), and the age-related cumulative mutation of mitochondrial DNA (mtDNA) is implicated in the inevitable and irreversible mitochondrial dysfunction (Hebert et al., 2010). Human mtDNA has a transcription-controlling unit in the intergenic, non-coding region (NCR) (Tracy and Stern, 1995; Shadel and Clayton, 1997). The mitochondrial NCR, commonly referred as “Dloop,” is approximately 1 kb long and contains three transcriptional promoters (Scarpulla, 2008). In the budding tunicate, Polyandrocarpa misakiensis, mtDNA is approximately 15 kb long, and the NCR locates between NADH dehydrogenase subunit 2 (ND2) and 12S ribosomal RNA (12S rRNA) genes (Fig. 1A). In Polyandrocarpa and other tunicates, however, little is known about the role of NCR in mitochondrial gene regulation. In P. misakiensis, buds grow out from the parental body wall,

develop into functional animals, and live for 4–5 months (Fig. 2A1) (Kawamura et al., 2012a). During lifetime, cytochrome c oxidase subunit 1 (PmCOX1) and PmND1 gene activities gradually attenuate as known in mammals, but this decrease in tunicate mitochondrial activities is reversibly restored during budding, although the neighboring parental somatic tissues continue to age (Kawamura et al., 2012a). Mitochondrial ageing and rejuvenation have been repeated for more than four decades in Polyandrocarpa asexual strains that collected in 1970 (Kawamura et al., 2012a). Several endogenous factors have been identified to regulate PmCOX1. TC14-3 is a humoral factor that triggers histone methylation (Kawamura et al., 2012b). It can activate PmCOX1, and its effect is blocked by a histone H3 methyltransferase inhibitor GSK343 (Kawamura et al., 2015). The second factor is Embryonic ectodermal development (PmEED), a component of polycomb group genes. It is strongly expressed during budding, and RNA interference (RNAi) of PmEED downregulates PmCOX1 expression (Kawamura et al., 2012a). The third factor is Mitochondrial transcription factor A (PmTFAM). Human TFAM specifically binds to sites immediately upstream of the promoters (Rubio-Cosials et al., 2011; Ngo et al., 2011) to initiate in vitro transcription (Fisher and Clayton, 1988; Falkenberg et al., 2002; Shi et al., 2012). TFAM also binds non-specifically to mtDNA (Kanki

Abbreviations: COX1, Cytochrome c oxidase subunit 1; ISH, In situ hybridization; NCR, Non-coding region; RT-PCR, Reverse transcription polymerase chain reaction; TFAM, Mitochondrial transcription factor A ⁎ Corresponding author. E-mail address: [email protected] (K. Kawamura). http://dx.doi.org/10.1016/j.mito.2017.05.006 Received 26 July 2016; Received in revised form 13 January 2017; Accepted 12 May 2017 1567-7249/ © 2017 Elsevier B.V. and Mitochondria Research Society. All rights reserved.

Please cite this article as: Kawamura, K., Mitochondrion (2017), http://dx.doi.org/10.1016/j.mito.2017.05.006

Mitochondrion xxx (xxxx) xxx–xxx

K. Kawamura et al.

Fig. 1. Mitochondrial genome and NCR in P. misakiensis. (A) Gene map of 15 kb mtDNA. (B) A 2.3 kb region intervening between ND2 and 16S rRNA genes. (C) Tandem repeat units (1.1, 1.2, 2.1, and 2.2) in PmNCR. (D) Multiple alignment of repeats 1.1 and 1.2 having 231 nucleotides. (E) Multiple alignment of repeats 2.1 and 2.2 having 191 nucleotides.. (For interpretation of the references to colour in this figure, the reader is referred to the web version of this article.)

expression of reporter gene, intracellular localization of gene expression, and cis- and trans-elements essential for gene transcription. To get insights into the way how repeatable changes of mitochondrial gene activities are possible during asexual life span, we examined the relationship among histone modification, TFAM regulation, and reporter gene expression. As far as we know, this is the first report on the mechanism of in vivo mitochondrial gene regulation in invertebrates including tunicates.

et al., 2004; Campbell et al., 2012). In P. misakiensis, TFAM is expressed most abundantly during budding and declines with zooid age. PmTFAM mRNA, when introduced in vivo into adult animals by electroporation, enhances PmCOX1 gene expression (Kawamura et al., 2015). It is uncertain at present how PmEED–related histone methylation is involved in the activation of mitochondrial genes having no histone proteins and what kinds of relationship, if any, exist among TC14-3, PmTFAM, and PmCOX1. In this study, we constructed reporter genes with wild and aberrant NCRs to characterize them with reference to the spatiotemporal 2

Mitochondrion xxx (xxxx) xxx–xxx

K. Kawamura et al.

Fig. 2. Comparison of 16S rRNA and reporter gene expression by ISH during sexual and asexual reproduction. (A1) Schematic illustration of zooid and buds in P. misakiensis. (A2) RT-PCR of 12S (lane 1) and 16S rRNA (lane 2) extracted from buds. (B, C, E, F, G) 16S rRNA expression. (D, H) PmNCR2.3K/GFP expression. (B1) Whole mount zooid, ventral view. Bar, 1 mm. (B2) Mature ovary and testis. Bar, 100 μm. (C1, D1) Primordial gonads. Arrowheads show germline precursor cells. Bars, 50 μm. (C2, D2) Developing ovary and testis. Bars, 50 μm. (E1) Bud primordium and parent. Bar, 0.5 mm. (E2) Distal tip of bud primordium. Bar, 50 μm. (F1) Growing bud. Arrows show the parental atrial epithelium. Bar, 0.5 mm. (F2) Bud section. Arrowheads show possible mitochondria. Bar, 10 μm. (G1) Truncated 2-day-developing bud. Broken circle shows the transdifferentiation area. Bar, 0.5 mm. (G2) Transdifferentiating tissues. Bar, 50 μm. (H1) Growing bud. Bar, 50 μm. (H2) Bud epidermis. Arrowhead shows possible mitochondria. Bar, 10 μm. (H3) Transdifferentiating atrial epithelium. Bar, 20 μm. ae, atrial epithelium; b, bud; c, coelomic cell; e, epidermis; g, gonad; ge, germinal epithelium; i, intestine; o, ovary; p, pharynx; po, primordial ovary; pt., primordial testis; s, stomach; t, testis.

2. Material and methods

2.2. Genes and RNAs

2.1. Animals

Pm16S rDNA [DDBJ: LC055486], PmNCR [DDBJ: LC055488], and PmTFAM [DDBJ: AB920559] were cloned into the pGEM-T vector (Promega, Madison, WI, USA). Digoxigenin (Dig)-labeled antisense RNA probes for in situ hybridization (ISH) were synthesized using T7 RNA polymerase (Roche, Mannheim, Germany). PmTFAM cDNA was subcloned into pCMV-Tag 5 (Stratagene, Santa Clara, CA, USA), and capped mRNA was synthesized using the mMessage mMachine T3 Kit (Ambion, Austin, TX, USA). Some reporter genes were incubated with HeLaScribe Nuclear Extract in vitro Transcription System (Promega) for 60 min at 30 °C, and RNA was purified by MEGAclear Kit (Ambion) for reverse transcription polymerase chain reaction (RT-PCR).

Zooids of P. misakiensis were cultured on glass plates and placed in culture boxes at the Uranouchi inlet near the Usa Marine Biological Institute, Kochi University. Developing buds were prepared by extirpating buds at the growth stage from parental zooids. Juvenile zooids were defined as 3–4-week-old growing zooids prior to budding. Adult zooids were aged for more than 1 month (Kawamura et al., 2012a).

3

Mitochondrion xxx (xxxx) xxx–xxx

K. Kawamura et al.

2.3. Inhibitors and RNAi

2.8. Chromatin immunoprecipitation

An RNA polymerase inhibitor in prokaryotes, Rifampicin (Wako Pure Chemical Industries, Osaka, Japan), was dissolved in distilled water at the concentration of 2 mg/mL. It was diluted with sterile seawater immediately before use. A histone methyltransferase inhibitor, GSK343 (SML0766, Sigma-Aldrich, St. Louis, MO, USA), was dissolved in DMSO at 10 mM and used at a final concentration of 10 μM, as described previously (Kawamura et al., 2015). Three different siRNAs directed to PmTFAM (si1, CCUAUUAAU GCUUUCGUAAUU; si2, GGAAGAUAAAUUAGAACAAUU; and si3, GUUAAAGAUUCAGCAAAUUCA) were mixed with one another at the concentration of 50 μM in sterile seawater. Immediately before use, the solution was coupled with seawater, containing Lipofectamine 3000 (Invitrogen, Carlsbad, CA, USA). RNAi of PmEED has already been described elsewhere (Kawamura et al., 2012a).

The cross-linking method for CHIP (Kimura et al., 2008) was used. The detail and modification were described elsewhere (Kawamura et al., 2015). 2.9. Microscopy Bright field sections were observed with a differential interference microscope (ECLIPSE 80i; Nikon, Tokyo, Japan) and analyzed by NISElements (Nikon). Dark field sections stained with fluorescent dyes were observed with ECLIPSE 80i or a confocal microscope (ECLIPSE C1si system; Nikon). 3. Results 3.1. Comparison between 16S rRNA and PmNCR expression

2.4. Reporter genes

Pm12S rRNA and Pm16S rRNA were arranged tandem in the mitochondrial genome (Fig. 1A, B). Pm16S rRNA expression was much stronger than that of Pm12S rRNA, irrespective of the developmental and senescence stages (Fig. 2A2), suggesting that Pm16S rRNA could be expressed by a specific and strong promoter, except for that as a polycistronic RNA together with 12S rRNA. In P. misakiensis, mature gonads were buried in the ventral body wall of zooid (Fig. 2A1) and strongly expressed Pm16S rRNA, similar to other animals (Fig. 2B1, B2). The expression was also observed in germline precursor cells and the primordial ovary and testis of immature gonads (Fig. 2C1, C2). The original reporter gene contained a 2.3-kb mtDNA insertion, extending from the end of ND2 to the beginning of 16S rRNA (Fig. 1B). The 2.3-kb region had two sequence repeats (green and blue boxes), one of which was located between 12S rRNA and 16S rRNA (Fig. 1C). The repeat 1 (green) was 230 nucleotides long, and the repeat 1.1 was 99.6% identical with repeat 1.2 (Fig. 1D). The repeat 2 (blue) was 191 nucleotides long, and the repeat 2.1 was 98.4% identical with repeat 2.2 (Fig. 1E). The reporter gene, PmNCR2.3K/GFP, was expressed in germline precursor cells and juvenile gonads (Fig. 2D1, D2), similar to Pm16S rRNA. Besides sexual reproduction, Pm16S rRNA also appeared during asexual reproduction, in which buds arise as the outgrowth of the parent body wall (Fig. 2A1). The bud primordium showed signals in the epidermis and atrial epithelium (Fig. 2E1, E2), similar to the expression pattern of PmCOX1 and PmND1 (Kawamura et al., 2012a). The signals became prominent in growing buds (Fig. 2F1), especially around the epidermal nucleus (Fig. 2F2 arrowheads), consistently with our ultrastructural study, which showed that mitochondria crowd around the nucleus at the bud growth stage (K.K., in preparation); however, the neighboring parent tissues showed low signals of 16S rRNA (Fig. 2F1 arrows). In 2-day-developing buds, 16S rRNA was heavily expressed at the proximal area, where transdifferentiation occurs (Fig. 2G1 broken circle). Both the epidermis and atrial epithelium were stained heavily (Fig. 2G2). PmNCR2.3K/GFP was expressed in the epidermis and atrial epithelium of growing buds (Fig. 2H1). Punctate (dot-like) signals were prominent around the nucleus (Fig. 2H2 arrowhead). In 2-day-developing buds, signals were found in transdifferentiating atrial epithelium and coelomic cells (Fig. 2H3).

A 2310-bp DNA fragment (PmNCR2.3K) that contained two repeated sequences between ND2 and 16S rRNA genes (Fig. 1A–C) was ligated to GFP with SalI and PstI sites into the pGEM-T vector (Promega). PmNCR2.3KΔ0.6 lacked 12S rRNA from NCR2.3K, whereas PmNCR2.3KΔ1.4 lacked the repeat 1.2, repeat 2.1, and 12S rRNA (Fig. 5A). A 5′ UTR DNA fragment (1386 bp) of PmTFAM was ligated to GFP and used as a control nuclear reporter gene. Other reporter genes used in this study are listed in Figs. 5A and 6C1. 2.5. Transfection Lipofection was used for DNA transfection and electroporation for mRNA transfection and DNA/RNA co-transfection. Lipofectamine 3000 (Invitrogen) was diluted 25-fold with sterile seawater, and the liposome solution was mixed with an equal volume of P3000 solution diluted 25fold with sterile seawater, containing 0.02–0.04 μg/μL DNA. Buds and zooid pieces were immersed in the liposome-DNA complex solution at 20–25 °C for 45–60 min. For electroporation, the reporter gene (final concentration of 0.01 μg/μL) and/or mRNA (final concentration of 0.01 μg/μL) were added to an electro-cuvette (0.4 mm), containing 400 μL of HEPES buffered salt solution (HBS) 2 min before the electroporation (625 V/cm, 100 μF; GENE Pulser Xcell System; BioRad, Hercules, CA, USA). 2.6. ISH and FISH Whole-mount ISH was performed using Dig-labeled antisense probe. Specimens were stained with anti-Dig monoclonal antibody labeled with alkaline phosphatase (Roche), embedded in Technovit 8100 resin, and sectioned for microscopy as previously described (Kawamura et al., 2012a). For fluorescent ISH (FISH), GFP antisense probe and 16S rRNA antisense probe were labeled with biotin and digoxigenin, respectively. After the hybridization, specimens were sectioned and stained with fluorescein isothiocyanate (FITC)-labeled streptavidin and rhodaminelabeled anti-Dig antibody. 2.7. RT-PCR

3.2. Unusual localization of reporter gene expression signals Poly(A)+ RNA was extracted and purified from zooids and buds using an mRNA isolation kit (Roche). Single-stranded DNA complementary to poly(A)+ RNA was synthesized for 30 min at 55 °C using the Transcriptor First Strand cDNA Synthesis Kit (Roche). PCR was performed as described previously (Kawamura et al., 2015), and PCR products were quantified by agarose gel electrophoresis using ImageJ (https://imagej.nih.gov/ij/).

Punctate or plain signals were usually located in the cytoplasm after ISH of reporter genes (Fig. 2H2, H3), but in some cases nuclear signals were observed (Fig. 3A black arrowheads) other than no nuclear signals (white arrowheads). Results of FISH also showed that the reporter gene signals appeared in the nuclei of the epidermis (Fig. 3B1, B2 black arrowheads) and mesenchymal cells (Fig. 3C1–C3). Green fluorescence 4

Mitochondrion xxx (xxxx) xxx–xxx

K. Kawamura et al.

Fig. 3. In vivo and in vitro reporter gene expression in the nucleus. (A–C) ISH of growing buds. (A) GFP signals merged with DAPI staining. White and black arrowheads show negative and positive nuclear signals, respectively. Bar, 20 μm. (B, C) FISH. (B) Epidermis. Bars, 20 μm. (B1) GFP signals (green). Black arrowheads show positive signals. (B2) Merge of GFP and DAPI signals. (C) Coelomic cells. Black arrowheads show signals in the cytoplasm. Bars, 10 μm. (C1) GFP signals. (C2) 16S rRNA signals (red) merged with DAPI image. (C3) Merge of (C1) and (C2). (D, E) In vitro transcription by HeLa cell nuclear extracts. (D) Gel image after RT-PCR. (E) Signal intensity of respective bands estimated by ImageJ. Longitudinal bars show the mean ± S.D. ae, atrial epithelium; e, epidermis. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

B). On the other hand, the PmCOX1 expression was affected by high concentrations of rifampicin more than 2 μg/mL (Fig. 4C, D). PmNCR2.3K/GFP exhibited similarly mitochondria-type rifampicin sensitivity (Fig. 4C, D). Ethidium bromide (EtBr) has inhibitory effects on mitochondrial transcription (Noda et al., 2002); however, EtBr (1 μg/mL) did not have such inhibitory effects on Polyandrocarpa mitochondria (data not shown).

protein (GFP) signals in the cytoplasm were consistent with 16S rRNA signals as a mitochondrial marker (Fig. 3C black arrowheads). To examine whether nuclear RNA polymerase could recognize possible promoter(s) in PmNCR to begin transcription, PmNCR2.3K/ GFP was incubated with HeLa cell nuclear extracts, and RNA products were analyzed by RT-PCR (Fig. 3D). In the absence of reverse transcriptase (RT), GFP signals were negligible, whereas in the presence of RT, GFP signals increased with increasing amounts of template plasmid (Fig. 3E). A positive control plasmid (PmTFAM1.4K/GFP) contained the 5′ untranslated region (UTR) of a nuclear gene, PmTFAM. It expressed GFP as strongly as PmNCR2.3K/GFP (Fig. 3E). In the negative control that only had a short 5′ UTR (100 bp) of PmNCR, signals became very weak.

3.4. Reporter gene expression from deletion constructs Growing buds (n = 20) were transfected with PmNCRs of variable length (Fig. 5A), and 2 days later RNA was extracted for RT-PCR. PmNCR2.3K/GFP showed a band only in the presence of RT (Fig. 5C lanes 4, 5), the position of band consistent with that of a PCR product from GFP plasmid (Fig. 5C lane 1). In the negative control, NCR was cloned from a related species, Botryllus primigenus. BpNCR/GFP produced no signals when introduced into P. misakiensis buds (Fig. 5C lanes 2, 3, 5D1). PmNCR2.3KΔ0.6 was the reporter gene lacking 12S rRNA from PmNCR2.3K (Fig. 5A). It had GFP signals with intensity similar to that of PmNCR2.3K (Fig. 5C lanes 6, 7, 5D2). PmNCR2.3KΔ1.4 lacked repeat 1.2, repeat 2.1, and 12S rRNA (Fig. 5A). In four independent transfection experiments, strong signals were observed in two cases

3.3. Rifampicin sensitivity of reporter genes Mitochondrial RNA polymerases purified from some species and organs are sensitive to rifampicin, whereas the antibiotic has no effects on nuclear genes. In P. misakiensis, rifampicin did not have any inhibitory effects on the expression of nuclear gene PmTFAM (Fig. 4A, B). Similarly, the reporter gene containing 5′ upstream region of PmTFAM (PmTFAM1.4K/GFP) was insensitive to rifampicin (Fig. 4A, 5

Mitochondrion xxx (xxxx) xxx–xxx

K. Kawamura et al.

Fig. 4. Rifampicin sensitivity of nuclear, mitochondrial, and reporter gene expression. (A) Gel images after RT-PCR of nuclear gene, PmTFAM, and nuclear gene-derived reporter gene, PmTFAM1.4K/GFP. Arrowhead shows the GFP band. PmActin was used as internal standard. (B) Signal intensity of respective bands in (A) estimated by ImageJ. (C) Gel image after RTPCR of mitochondrial gene, PmCOX1, and NCR reporter gene, PmNCR2.3K/GFP. Arrowhead shows the GFP band. PmActin was used as internal standard. (D) Signal intensity of respective bands in (C) estimated by ImageJ. Longitudinal bars show the mean ± S.D.

3.6. Relationship between NCR length and reporter gene expression

(Fig. 5C lanes 8, 9) and weak signals in the remaining two cases (Fig. 5C lane 10). After ISH, epidermal signals became highly weak (Fig. 5D3), indicating that the reporter gene expression became progressively unstable. The repeat 2.2 contained a TATA box-like sequence of 32 nucleotides (Fig. 5B), approximately 100 nucleotides upstream from the transcription start of Pm16S rRNA (Fig. 5B arrowheads). This nucleotide position is considerably different from positions − 15 to −40 between the mitochondrial promoter and the transcription initiation site in mammals (Ringer et al., 2011; Rubio-Cosials and Solà, 2013). When 32 nucleotides were deleted from PmNCR2.3Δ1.4 (Fig. 5A PmNCR2.3Δ1.4 + P), almost no signals appeared after RT-PCR (Fig. 5C lanes 11, 12). To examine whether the complete repeat 2 can solely induce the reporter gene expression, we constructed plasmids containing only one or two copies of repeat 2, named PmNCR(repeat2)1 and PmNCR (repeat2)2 (Fig. 5A). These reporter genes, however, did not support GFP expression (Fig. 5C lanes 13–16). Our results indicated that, although the repeat 2 contained an essential sequence (possible promoter) for GFP expression, some upstream regions were also necessary.

As already shown in Fig. 5, either a single copy or tandem copies of repeat 2 (2.1 plus 2.2) were insufficient for reporter gene expression. Likewise, PmNCR0.3K with the repeat 2.2 (Fig. 6C1) did not provoke gene induction in response to PmTFAM mRNA (Fig. 6C2 lanes 1, 2, C3), suggesting that some upstream sequences from the repeat 2.2 are needed for PmTFAM to enhance the reporter gene expression. Adult animal pieces (n = 12) were co-transfected with PmTFAM mRNA and reporter genes of variable length (Fig. 6C1). PmNCR0.6K consisted of the repeat 2.2 and a truncated 12S rRNA, and PmNCR0.9K had a full length 12S rRNA. As shown in Fig. 5, 12S rRNA was dispensable for reporter gene expression; however, both PmNCR0.6K and PmNCR0.9K dramatically improved GFP expression in the presence of PmTFAM mRNA (Fig. 6C2 lanes 3–6, C3). This experiment also showed that the repeat 1 was dispensable, although the repeat 1.1 could induce the reporter gene expression when combined with the repeat 2.2 (Fig. 5C PmNCR2.3KΔ1.4). PmNCR1.2K, PmNCR1.5K, and PmNCR1.8K had gradually increasing 5′ upstream NCR sequences (Fig. 6C1). The results showed the gradual increase of reporter gene expression (Fig. 6C2 lanes 7–12, C3), suggesting that, in P. misakiensis, the responsiveness to TFAM depends on the nucleotide length upstream from the repeat 2.2, in addition to specific TATA-like box.

3.5. Role of TFAM in reporter gene expression 3.7. Involvement of histone methylation in reporter gene expression In P. misakiensis, PmTFAM expression was the most abundant during budding and declined during senescence to approximately 40% (Fig. 6A1–A4). The bud epidermis and coelomic cells were stained moderately, and the atrial epithelium was stained heavily (Fig. 6A5). In adult zooids, signals from the atrial epithelium and coelomic cells became weak, and those from the epidermis were almost undetectable (Fig. 6A6). Capped PmTFAM mRNA was introduced by electroporation into senescent adult animals (Fig. 6B). It enhanced GFP expression when co-transfected with the reporter gene (Fig. 6B1, B2). Semiquantitative RT-PCR showed that GFP transcription was at least 5-fold higher than that in the control without PmTFAM mRNA (Fig. 6B3, B4).

A large amount of TC14-3 mRNA was transcribed during budding stage as compared with adult stage (Fig. 7A1, A2). TC14-3 recombinant protein accelerated H3K27me3 of PmSirtuin6 and several other genes and downregulated their expression (Fig. 7B1) (Kawamura et al., 2015). In contrast, TC14-3 enhanced the gene expression of PmTFAM and several other genes without affecting H3K27 trimethylation (Fig. 7B2). Sirtuin6 encodes NAD-dependent H3K9 deacetylase that acts as a transcriptional co-repressor (Michishita et al., 2008). PmSirtuin6 mRNA indeed caused histone deacetylation around the promoter region of PmTFAM (Fig. 7C1). It suppressed PmTFAM gene expression (Fig. 7C2). 6

Mitochondrion xxx (xxxx) xxx–xxx

K. Kawamura et al.

Fig. 5. Reporter assay of full and deletion constructs transfected into buds. (A) Reporter genes having variable length of 5′ UTR and constant size of 3′ UTR and poly(A). (B) TATA box-like sequence in the repeat 2.2 (blue) and possible transcription start sites (arrowheads) of 16S rRNA gene (red). (C) Gel images after RT-PCR in the absence or presence of reverse transcriptase (RT). (D) ISH of GFP expression. Bars, 50 μm. (D1) BpNCR/GFP as a negative control. (D2) PmNCR2.3KΔ0.6/GFP. (D3) PmNCR2.3KΔ1.4/GFP. ae, atrial epithelium; e, epidermis. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

4. Discussion

TC14-3 increased PmNCR/GFP transcription approximately 7-fold, as compared to the untreated control (Fig. 7D1 lanes 1, 2, D2, E1, E2). This effect of TC14-3 on reporter gene expression was suppressed by siRNAs of either PmEED (Fig. 7D1 lane 3, D2, E3) or PmTFAM (Fig. 7D1 lane4, D2, E4) and by a H3K27 methyltransferase inhibitor, GSK343 (Fig. 7D1 lane 5, D2, E5). RNAi of PmTFAM showed the most severe inhibition, which was reasonable provided that PmTFAM was located immediately upstream of the reporter gene in the gene cascade. To confirm the effectiveness of siRNA(PmTFAM), we introduced siRNA into buds and found that endogenous PmTFAM mRNA markedly attenuated after RNAi (Fig. 7F1–F3). The result of siRNA(PmEED) has already been shown elsewhere (Kawamura et al., 2012a).

4.1. PmNCR reporter gene reproduces Pm16S rRNA expression As first revealed in fruit fly (Kobayashi and Okada, 1989) and later in other species from flatworms to mammals (Kobayashi et al., 1998; Sato et al., 2001; Ninomiya and Ichinose, 2007), 16S rRNA gene is abundantly expressed in germline cells and a portion of RNA accumulates in cytoplasmic bodies outside of the mitochondria (Kobayashi et al., 1993). In P. misakiensis, a large amount of 16S rRNA similarly appeared in germline precursor cells and primordial gonads. In addition, the gene

7

Mitochondrion xxx (xxxx) xxx–xxx

K. Kawamura et al.

Fig. 6. PmTFAM gene expression from budding to ageing with reference to reporter gene induction. (A) ISH and RT-PCR of PmTFAM. (A1) Growing bud. Bar, 200 μm. (A2) Juvenile zooid. Bar, 100 μm. (A3) Adult zooid. Bar, 100 μm. (A4) Gel images after RT-PCR (upper). The gradual decrease in signal intensity was estimated by Image J (lower). (A5) Bud tissues. Bar, 20 μm. (A6) Adult tissues. Bar, 20 μm. (B) Reporter gene expression in adult tissues after PmTFAM mRNA transfection. Bars, 20 μm. (B1) Control without mRNA. (B2) Experiment treated with mRNA. (B3) Cumulative analysis of RT-PCR. (B4) Signal intensity of respective PCR bands estimated by ImageJ. (C) Effects of 5′ upstream length of NCR on reporter gene expression induced by PmTFAM mRNA. (C1) Reporter genes having increasing upstream regions from the repeat 2.2 (blue box). (C2) Gel images after RT-PCR. (C3) Signal intensity of PCR bands estimated by ImageJ. Longitudinal bars show the mean ± S.D. ae, atrial epithelium; c, coelomic cell; e, epidermis; p, pharynx. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

4.2. PmNCR reporter genes can be expressed in both mitochondria and nuclei

expression was conspicuously enhanced during bud formation and development. This expression pattern of Pm16S rRNA was quite similar to that of PmCOX1 (Kawamura et al., 2012a), indicating that the mitochondrial gene activity is enhanced during both sexual and asexual reproduction. The reporter gene, PmNCR2.3K/GFP, could completely reproduce these expression patterns of 16S rRNA. This result indicates that PmNCR reporter gene affords a reliable probe for searching mitochondrial gene activity.

At first, we expected that the reporter gene was only expressed in mitochondria. However, confocal imaging studies indicated undoubtedly that GFP signals appeared in both the nucleus and cytoplasm. Mitochondrial RNA polymerase (mtRNAP) has similar sequence and structure to bacteriophage T7 RNA polymerase (Ringer et al., 2011). There is no reported information on the nuclear localization of 8

Mitochondrion xxx (xxxx) xxx–xxx

K. Kawamura et al.

Fig. 7. Relationship among TC14-3, PmSirtin6, PmTFAM, and reporter gene expression. (A) Comparison of TC14-3 transcription between buds and adults. (A1) Gel image of cumulative PCR cycles. (A2) Signal intensity of PCR bands estimated by ImageJ. (B) Effects of TC14-3 on histone methylation and gene expression. (B1) Genes in which H3K27me3 occurred and the expression was downregulated. (B2) Genes in which H3K27me3 did not occur and the expression was upregulated. (C) Effects of PmSirtuin6 mRNA on histone deacetylation and gene expression of PmTFAM. (C1) ChIP of PmTFAM using anti-acetylated histone H3K9 antibody. (C2) RT-PCR of PmTFAM. PmActin was used as an internal standard. (D) Effects of siRNAs and GSK343 on TC14-3-induced reporter gene expression. (D1) Gel image. (D2) Relative strength of gene expression estimated by ImageJ. (E) ISH in the absence or presence of TC14-3 and other compounds. Bars, 20 μm. (E1) Control. (E2) TC14-3. (E3) TC14-3 plus siRNA(PmEED). (E4) TC14-3 plus siRNA(PmTFAM). (E5) TC14-3 plus GSK343. (F) Effectiveness of siRNA(PmTFAM) introduced into buds. (F1) RT-PCR of PmTFAM 2 days after siRNA. (F2) ISH of PmTFAM without siRNA. Bar, 50 μm. (F3) ISH of PmTFAM after siRNA. Bar, 50 μm. ae, atrial epithelium; e, epidermis.

cytoplasmic signals of the reporter gene overlapped with those of 16S rRNA. These results strongly suggested that cytoplasmic punctate signals indicated the mitochondrial position, even though 16S rRNA leaks out from mitochondria (Kobayashi et al., 1993). Rifampicin is an antibiotic that inhibits bacterial RNA polymerase (RNAP) by 50% at the concentration of 0.1 μM (0.08 μg/mL) (Campbell et al., 2001). PmNCR/ GFP was sensitive to rifampicin at relatively higher concentration (2 μg/mL), although the nuclear reporter gene (PmTFAM/GFP) was insensitive to it even at this concentration, affording further evidence for the mitochondrial expression of PmNCR reporter gene. Taken altogether, our results indicate that in P. misakiensis, reporter

mtRNAP, whereas the current in vitro transcription assay indicated that PmNCR reporter gene could be transcribed by virtue of nuclear transcription machinery. In mammals, TFAM is a core component of the mitochondrial transcription machinery (Shi et al., 2012). It not only moves toward mitochondria but also acts in the nucleus (Han et al., 2011; Lee et al., 2014). In growing buds of P. misakiensis, reporter gene often showed punctate signals, similar to those of Pm16S rRNA, around epidermal nuclei having no signals after ISH. Epidermal mitochondria were ultrastructurally found to assemble around the nucleus during bud stages (K.K., in preparation). The present FISH study also showed that 9

Mitochondrion xxx (xxxx) xxx–xxx

K. Kawamura et al.

absence of TFAM, human mitochondrial extracts abolish transcription from promoters, whereas the in vitro transcription system restarts successfully after the addition of functional TFAM to the extracts (Shi et al., 2012). In P. misakiensis, PmTFAM encodes 2 HMG boxes and a mitochondrial transport signal. In the previous and current study, the PmTFAM expression was the highest during budding stages, and it declined during zooid ageing (Kawamura et al., 2015), consistent with the expression pattern of mitochondrial genes (Kawamura et al., 2012a). PmTFAM mRNA facilitated PmCOX1 and reporter gene expression. These results indicate that in P. misakiensis, TFAM indeed acts as transelement of mitochondrial genes. It should also be noted that the region immediately upstream from the possible promoter was dispensable for PmTFAM function, suggesting that some mechanisms different from those of mammals may regulate Polyandrocarpa reporter gene. In mammals, TFAM binds non-specific HMG recognition sites on mtDNA, which has a direct effect on the mtDNA copy number without affecting the transcriptional level (Ekstrand et al., 2004; Kanki et al., 2004). The DNase I protection assay indicates that a single TFAM protein protects approximately 20 bp of mtDNA in length (Fisher et al., 1992). In HeLa cells, the TFAM: mtDNA ratio is approximately 1000: 1, indicating that TFAM covers almost the whole mtDNA without any vacant space (Takamatsu et al., 2002). Considering these references, the total amount of PmTFAM protein binding to NCR may play a key role in regulating reporter gene expression.

Fig. 8. Gene cascade connecting TC14-3 with mitochondrial activity. (A) TC14-3 accelerates histone H3K27me3 of downstream genes via PmEED. H3K27me3 suppresses PmSirtuin6 gene expression. H3K9 deacetylation attenuates PmTFAM expression. PmTFAM is a trans-element for PmNCR reporter gene and mitochondrial gene expression. (B) Senescence accompanies the decrease in H3K27me3, resulting in the upregulation of PmSirtuin6 and, inversely, the downregulation of PmTFAM. (C) Budding accompanies the increase in H3K27me3 of PmSirtuin6, making PmTFAM release from the suppressive effect of PmSirtuin6. It is uncertain whether histone acetyltransferase PmGCN5 is involved in H3K9 acetylation of PmTFAM at the beginning of budding.

genes and probably mitochondrial DNA having NCR can be transcribed both in nucleus and mitochondria, which means that if the horizontal DNA transfer (Albert et al., 2014) occurs toward the nucleus, some mitochondrial genes would be expressed immediately without the integration into the nuclear genome.

4.5. Transcriptional co-repressor intervenes between TC14-3 and mitochondrial activity TC14s belong biochemically to a C-type lectin family (Suzuki et al., 1990). TC14-3 acts cytologically as a cytostatic factor (Matsumoto et al., 2001). Recently, TC14-3 was found to be an epigenetic inducer of a polycomb group gene, PmEed, that facilitates trimethylation of histone H3 lysine 27 (H3K27me3) (Fig. 8A) (Kawamura et al., 2012b). It also induced PmCOX1 gene expression by unknown mechanisms (Kawamura et al., 2012a). In the current study, TC14-3 facilitated PmNCR reporter gene expression, and this effect was suppressed by siEED and GSK343, indicating that TC14-3 exerts the effect via histone methylation. TC14-3 induced H3K27me3 in the promoter region of PmSirtuin6 and several other genes and consequently blocked their expression (Fig. 8A) (Kawamura et al., 2015 and present study). In contrast, histone H3K9 acetylation occurs during bud development in transdifferentiation-related genes such as retinoid X receptor (RXR) and ERK, and it conspicuously enhances their expression (Shibuya et al., 2015). Sirtuin6 is a NAD-dependent histone H3K9 deacetylase that acts as a transcriptional co-repressor (Michishita et al., 2008). The present study has shown that PmTFAM is one of direct targets of PmSirtuin6 (Fig. 8A), as PmSirtuin6 mRNA caused the deacetylation of PmTFAM histone H3K9 and attenuated PmTFAM gene expression. It is, therefore, possible to assume that PmSirtuin6 may serve as a negative regulator intervening between TC14-3 and mitochondrial gene activity (Fig. 8A). According to our hypothesis, senescence accompanies the decrease in trimethylated histone of PmSirtuin6 owing to the low level of TC14-3, which would cause the upregulation of PmSirtuin6 gene and then the downregulation of PmTFAM (Fig. 8B). During budding, on the other hand, TC14-3 suppresses PmSirtuin6 via histone methylation, by which PmTFAM would be released from the suppressive effect of PmSirtuin6 (Fig. 8C). Both PmTFAM and mitochondrial genes are rapidly activated at the beginning of bud formation (Kawamura et al., 2012a). We supposed that PmTFAM histone acetylation rather than PmSirtuin6 histone methylation might be involved in the so rapid gene activation (Fig. 8C). However, PmGCN5 encoding H3K9 acetyltransferase is expressed at the transdifferentiation stage later than bud formation stages (Shibuya et al., 2015). Histone H3K9 acetylation of PmTFAM also occurs at later bud stages (unpublished data of KK). Therefore, although

4.3. Reporter gene expression depends on the length of cis-element Polyandrocarpa NCR contained 2 sequence repeats, one of which (repeat 2) had a TATA-like box. These repeats of sequence are absent from related species such as B. schlosseri (Griggio et al., 2014) and Polycarpa mytiligera (Rubinstein et al., 2013). As the TATA-like box in PmNCR was essential for reporter gene expression, it might represent the possible promoter of mitochondrial genes. If this is the case, two promoters from repeats 2.1 and 2.2 would facilitate Pm16S rRNA gene expression, similar to that of human 12S rRNA that is regulated by the heavy strand promoters 1 and 2 (Scarpulla, 2008), explaining in part the reason that Pm16S rRNA expression was much higher than that of Pm12S rRNA. In mammalian mitochondria genome, high mobility group (HMG) recognition sequences are located immediately upstream from the promoter. They are essential for facilitating mitochondrial transcriptional initiation (Litonin et al., 2010; Rubio-Cosials and Solà, 2013; Yakubovskaya et al., 2014). In contrast, in P. misakiensis, repeat 1.1, 12S rDNA, and other upstream regions similarly facilitated reporter gene expression. The strength of expression depended on the length of cis-element upstream from the TATA-like box rather than some specific regions. This relationship between the TATA-like box and its upstream regions in PmNCR is similar to the relationship between promoter and enhancer in eukaryote genomes rather than to prokaryote-type transcription systems. 4.4. PmTFAM is a budding- and ageing-dependent mitochondrial transelement TFAM has 2 tandem HMG boxes to bind mtDNA (Shadel and Clayton, 1997; Scarpulla, 2008; Kanki et al., 2004; Campbell et al., 2012). Human TFAM facilitates the transcriptional initiation by bending DNA immediately upstream from the promoter region (RubioCosials et al., 2011; Ngo et al., 2011; Rubio-Cosials and Solà, 2013) in collaboration with mitochondrial transcription factor B (Falkenberg et al., 2002; Litonin et al., 2010; Yakubovskaya et al., 2014). In the 10

Mitochondrion xxx (xxxx) xxx–xxx

K. Kawamura et al.

Campbell, C.T., Kolesar, J.E., Kaufman, B.A., 2012. Mitochondrial transcription factor a regulates mitochondrial transcription initiation, DNA packaging, and genome copy number. Biochim. Biophys. Acta 1819, 921–929. Ekstrand, M.I., et al., 2004. Mitochondrial transcription factor a regulates mtDNA copy number in mammals. Hum. Mol. Genet. 13, 935–944. Falkenberg, M., et al., 2002. Mitochondrial transcription factors B1 and B2 activate transcription of human mtDNA. Nat. Genet. 31, 289–294. Fisher, R.P., Clayton, D.A., 1988. Purification and characterization of human mitochondrial transcription factor 1. Mol. Cell. Biol. 8, 3496–3509. Fisher, R.P., et al., 1992. DNA wrapping and bending by a mitochondrial high mobility group-like transcriptional activator protein. J. Biol. Chem. 267, 3358–3367. Gilbert, S., 2013. Developmental Biology, 10th ed. Sinauer Associates Inc, Massachusetts, pp. 719P. Griggio, F., et al., 2014. Ascidian mitogenomics: comparison of evolutionary rates in closely related taxa provides evidence of ongoing speciation events. Genome Biol. Evol. 6, 591–605. Han, B., et al., 2011. Human mitochondrial transcription factor a functions in both nuclei and mitochondria and regulates cancer cell growth. Biochem. Biophys. Res. Commun. 408, 45–51. Hebert, S.L., Lanza, I.R., Nair, S., 2010. Mitochondrial DNA alterations and reduced mitochondrial function in aging. Mech. Aging Dev. 131, 451–462. Kanki, T., et al., 2004. Architectural role of mitochondrial transcription factor A in maintenance of human mitochondrial DNA. Mol. Cell. Biol. 24, 9823–9834. Kawamura, K., Kitamura, S., Sekida, S., Tsuda, M., Sunanaga, T., 2012a. Molecular anatomy of tunicate senescence: reversible function of mitochondrial and nuclear genes associated with budding cycles. Development 139, 4083–4093. Kawamura, K., Takakura, K., Mori, D., Ikeda, K., Nakamura, A., Suzuki, T., 2012b. Tunicate cytostatic factor TC14-3 induces a polycomb group gene and histone modification through Ca2 + binding and protein dimerization. BMC Cell Biol. 13, 3. Kawamura, K., Kinoshita, M., Sekida, S., Sunanaga, T., 2015. Histone methylation codes involved in stemness, multipotency, and senescence in budding tunicates. Mech. Ageing Dev. 145, 1–12. Kimura, H., Hayashi-Takanaka, Y., Goto, Y., Takizawa, N., Nozaki, N., 2008. The organization of histone H3 modifications as revealed by a panel of specific monoclonal antibodies. Cell Struct. Funct. 33, 61–73. Kobayashi, S., Okada, M., 1989. Restoration of pole-cell-forming ability to u.v.-irradiated Drosophila embryos by injection of mitochondrial lrRNA. Development 107, 733–742. Kobayashi, S., Amikura, R., Okada, M., 1993. Presence of mitochondrial large ribosomal RNA outside mitochondria in germ plasm of Drosophila melanogaster. Science 260, 1521–1524. Kobayashi, S., Amikura, R., Mukai, M., 1998. Localization of mitochondrial large ribosomal RNA in germ plasm of Xenopus embryos. Curr. Biol. 8, 1117–1120. Lee, E.J., Kang, Y.C., Park, W.-H., Jeong, J.H., Pak, Y.K., 2014. Negative transcriptional regulation of mitochondrial transcription factor A (TFAM) by nuclear TFAM. Biochem. Biophys. Res. Commun. 450, 166–171. Litonin, D., et al., 2010. Human mitochondrial transcription revisited: only TFAM and TFB2M are required for transcription of the mitochondrial genes in vitro. J. Biol. Chem. 285, 18129–18133. Matsumoto, J., et al., 2001. A novel C-type lectin regulating cell growth, cell adhesion and cell differentiation in budding tunicates. Development 128, 3339–3347. Michishita, E., et al., 2008. SIRT6 is a histone H3 lysine 9 deacetylase that modulates telomeric chromatin. Nature 452, 492–496. Ngo, H.B., Kaiser, J.T., Chan, D.C., 2011. Tfam, a mitochondrial transcription and packaging factor, imposes a U-turn on mitochondrial DNA. Nat. Struct. Mol. Biol. 18, 1290–1296. Ninomiya, Y., Ichinose, S., 2007. Subcellular distribution of mitochondrial ribosomal RNA in the mouse oocyte and zygote. PLoS One 2, 1241. Noda, M., et al., 2002. Switch to anaerobic glucose metabolism with NADH accumulation in the β-cell model of mitochondrial diabetes. J. Biol. Chem. 277, 41817–41826. Ringer, R., et al., 2011. Structure of human mitochondrial RNA polymerase. Nature 478, 269–273. Rubinstein, N.D., et al., 2013. Deep sequencing of mixed total DNA without barcodes allows efficient assembly of highly plastic ascidian mitochondrial genomes. Genome Biol. Evol. 5, 1185–1199. Rubio-Cosials, A., Solà, M., 2013. U-turn DNA bending by human mitochondrial transcription factor a. Curr. Opin. Struct. Biol. 23, 116–124. Rubio-Cosials, A., et al., 2011. Human mitochondrial transcription factor A induces a uturn structure in the light strand promoter. Nat. Struct. Mol. Biol. 18, 1281–1289. Sato, K., et al., 2001. Localization of mitochondrial ribosomal RNA on the chromatoid bodies of marine planarian polyclad embryos. Develop. Growth Differ. 43, 107–114. Scarpulla, R.C., 2008. Transcriptional paradigms in mammalian mitochondrial biogenesis and function. Physiol. Rev. 88, 611–638. Shadel, G.S., Clayton, D.A., 1997. Mitochondrial DNA maintenance in vertebrates. Annu. Rev. Biochem. 66, 409–435. Shi, Y., et al., 2012. Mammalian transcription factor A is a core component of the mitochondrial transcription machinery. Proc. Natl. Acad. Sci. U. S. A. 109, 16510–16515. Shibuya, M., Hatano, M., Kawamura, K., 2015. Interactive histone acetylation and methylation in regulating transdifferentiation-related genes during tunicate budding and regeneration. Dev. Dyn. 244, 10–20. Suzuki, T., et al., 1990. A calcium-dependent galactose-binding lectin from the tunicate Polyandrocarpa misakiensis. Isolation, characterization and amino acid sequence. J. Biol. Chem. 265, 1274–1281. Takamatsu, C., et al., 2002. Regulation of mitochondrial D-loops by transcription factor A and single-stranded DNA-binding protein. EMBO Rep. 3, 451–456. Tracy, R.L., Stern, D.B., 1995. Mitochondrial transcription initiation: promoter structures and RNA polymerases. Curr. Genet. 28, 205–216. Yakubovskaya, E., et al., 2014. Organization of the human mitochondrial transcription initiation complex. Nucleic Acids Res. 42, 4100–4112.

our hypothesis well explains how mitochondrial activity attenuates during senescence, the mitochondrial activation mechanism is still obscure. In the forth-coming paper, we will focus the missing parts of mitochondrial activation/inactivation cascade. 5. Conclusion In the tunicate Polyandrocarpa misakiensis, somatic cells can escape from ageing by asexual reproduction in association with mitochondrial gene activation. TFAM is a budding- and ageing-related trans-element that facilitates the reporter gene expression from the non-coding region (NCR) of mitochondrial DNA. Unlike mammals, tunicate TFAM depends on the length of 5′ upstream cis-element from the TATA-like box rather than some specific sequences within NCR. This study affords, for the first time, an insight into the relationship between epigenetic histone methylation and mitochondrial gene regulation. Additionally, we found that the NCR reporter gene can be expressed in the nuclear transcription system, which indicates that after the horizontal transfer toward the nucleus, NCR-containing mitochondrial genome or gene fragments are capable of transcription without integration into the nuclear genome. Significance statement In the tunicate Polyandrocarpa misakiensis, somatic cells escape from senescence by budding in association with mitochondrial activation. Mitochondrial non-coding-region (NCR)-containing reporter genes were expressed spatiotemporally in a pattern similar to mitochondrial 16S rRNA. Unexpectedly, the gene expression was possible in vivo in the cell nucleus and in vitro in the nuclear extracts. Mitochondrial transcription factor A (TFAM) was an ageing-related trans-element that facilitated the reporter gene expression, and unlike mammals, it required adequate length of 5′ upstream region from the TATA-like box rather than specific sequences within NCR. The TFAM regulation depended on budding-related nuclear histone methylation at H3K27. Competing interests The authors declare no conflict of interest. Author contributions KK designed experiments, constructed reporter genes, carried out ISH, and drafted the manuscript. YS prepared capped mRNA and carried out in vivo reporter assays. LB participated in mitochondrial EST. TS carried out ISH and helped YS to conduct reporter gene assays. Acknowledgments We thank Dr. Shigeki Fujiwara for the valuable discussion concerning the expression and function of NCR-reporter genes. We also thank the staff of the Usa Marine Biological Institute, Kochi University, for their assistance with tunicate culturing throughout the experimental period. This study was supported by KAKENHI (Grant-in-Aid for Scientific Research (C); No. 21570227 and No. 15K07078; Grant-inAid for Challenging Exploratory Research, No. 25650081) from the Japan Society for the Promotion of Science. References Albert, B., et al., 2014. Molecular Biology of the Cell, sixth ed. Garland Science Taylor and Francis Group, New York, pp. 1464P. Byrne, E., Trounce, I., Dennett, X., 1991. Mitochondrial theory of senescence: respiratory chain protein studies in human skeletal muscle. Mech. Dev. 60, 295–302. Campbell, E.A., et al., 2001. Structural mechanism for rifampicin inhibition of bacterial RNA polymerase. Cell 104, 901–912.

11