Previous Article |
Table of Contents
| Next Article 
BIOLOGICAL SCIENCES / GENETICS
Terminally differentiated muscle cells are defective in base excision DNA repair and hypersensitive to oxygen injury
*Department of Environment and Primary Prevention, Istituto Superiore di Sanità, Viale Regina Elena 299, 00161 Rome, Italy; Department of Genetics and Complex Diseases, Harvard School of Public Health, Boston, MA 02115; and Department of Translational Oncology, Istituto Nazionale per la Ricerca sul Cancro, Largo Rosanna Benzi 10, 16132 Genoa, Italy
Edited by James E. Cleaver, University of California, San Francisco, CA, and approved July 16, 2007 (received for review February 27, 2007)
 |
Abstract |
|---|
 Top Abstract ResultsDiscussionMaterials and MethodsAcknowledgementsReferences |
|---|
. The attenuation of BER in myotubes was associated with significant accumulation of DNA damage as detected by increased DNA single-strand breaks and phosphorylated H2AX nuclear foci upon exposure to hydrogen peroxide. We propose that in skeletal muscle exacerbated by free radical injury, the accumulation of DNA repair intermediates, due to attenuated BER, might contribute to myofiber degeneration as seen in sarcopenia and many muscle disorders.
oxidative stress | XRCC1 | DNA ligases | DNA single-strand breaks | 8-oxoguanine
In skeletal muscle, tissue maintenance and repair are ensured by satellite cells that reside under the basal lamina (1). These cells can be readily isolated, cultured, and induced to differentiate into mononucleated myocytes that precede the fusion in multinucleated myotubes (7, 8).
Here we use mouse satellite cells differentiated in vitro as a model to study BER regulation during skeletal muscle differentiation. This cell system recapitulates the process in vivo as shown by irreversible cell cycle withdrawal, repression of cell proliferation-associated genes, and expression of muscle-specific genes during terminal differentiation in vitro as determined by genome-wide analysis (9). We show that long-patch BER, which shares several partners with DNA replication, is impaired in myotubes, and, to a minor extent, short-patch BER also is decreased. In either case, terminally differentiated cells accumulate unligated repair intermediates in vitro. This finding is in agreement with the lack of DNA ligase I in myotubes and the down-regulation of XRCC1 during differentiation with subsequent destabilization of DNA ligase III. The attenuation in myotubes of BER, and likely of XRCC1-dependent SSBR, is associated with the accumulation of DNA single-strand breaks (SSB) and 
H2AX foci upon cell exposure to hydrogen peroxide. This mechanism might account for the exquisite sensitivity of myotubes to oxidative injury.
|
Results |
|---|
|
TopAbstractResultsDiscussionMaterials and MethodsAcknowledgementsReferences |
|---|
|
The changes in oxidative stress gene expression and the increased level of DNA oxidation in terminally differentiated muscle cells likely reflect an extra burden for DNA repair associated with free radical generation during the differentiation process.
Base Excision Repair Is Less Efficient in Myotubes than in Myoblasts. BER reactions were conducted in vitro by using as substrate a 32P-labeled circular plasmid containing a single AP site (pGEM-AP). Whole-cell extracts were prepared from proliferating myoblasts and terminally differentiated myotubes and incubated in the presence of dNTPs for increasing periods of time. After 10 min postincubation, a delay in the repair rate of AP sites by myotube extracts was observed, compared with extracts from proliferating cells (Fig. 2). Nicked forms (Form II), which consist of both incised and unligated polymerization products, persisted for longer times in the case of terminally differentiated cells (Fig. 2B) than in myoblasts (Fig. 2A), although completion of repair (reconstitution of fully repaired Form I plasmids) was eventually observed in both cases at 1-h repair time (Fig. 2C). To better characterize the steps responsible for the delayed repair kinetics, the 32P-labeled repaired plasmid molecules were digested with EcoRI and PstI to produce a 43-nt fragment that originally contained the lesion (Fig. 3). In this assay, repair intermediates are identified as 25-nt fragments that correspond to the incision step or >25 nt depending on the patch size of the polymerization step. The slower rate of repair of myotubes, compared with myoblasts, was confirmed. In addition, a clear accumulation of repair intermediates was detected in these cell extracts. In particular, the intensity of the band corresponding to complete repair was lower as a function of time in terminally differentiated (Fig. 3B), compared with proliferating (Fig. 3A), cell extracts. This finding was associated with the persistence for longer times of the 25-nt incised and 26-nt gap-filling products. The repair of AP sites in myotubes is affected by inefficient processing of repair intermediates.
|
|
-32P-TTP. Repaired DNA was digested with SmaI and HindIII restriction endonuclease to free a fragment that originally contained the lesion. As shown in Fig. 4A, the repair of the AP site by terminally differentiated cell extracts was less efficient than by proliferating cell extracts, and 1-nt gap-filling products accumulated at early repair times. These repair intermediates are compatible with the processing of AP sites in myotubes mainly by short-patch BER, but their persistence indicates that the end-processing machinery is not fully functional. The repair of the dL residue (Fig. 4B) was even more affected in terminally differentiated cell extracts than that of the AP site. In this case, the accumulation of 2-nt gap-filling products confirmed the occurrence of long-patch BER at dL residues (12), suggesting that the processing of the long-patch repair intermediates also were impaired in terminally differentiated cells. The quantitative analysis of the repair products detected in three independent experiments is provided in SI Materials and Methods and SI Fig. 7.
|
The Ligation Step Is Defective in Terminally Differentiated Muscle Cells.
The incision of the abasic site by the major 5' AP endonuclease APE1 and the resynthesis step by DNA polymerase β (Pol β) are unaffected by the differentiation state, as determined by functional assays, gene expression, and protein levels (SI Fig. 8). In contrast, when the ligation capacity of the two cell extracts was assessed, a significant difference between proliferating and terminally differentiated cells was detected by using a DNA ligase assay (Fig. 5A). This assay mainly reflects the activity of DNA ligase I and is in line with the dramatic decrease in transcription (Fig. 5B) and protein levels (Fig. 5C) of this enzyme observed in myotubes. DNA ligase I is required to seal Okazaki fragments during DNA replication and participates in BER. This last function is shared with the DNA ligase III/XRCC1 heterodimer. RT-PCR analysis of DNA ligase III and XRCC1 (Fig. 5B) revealed that the transcription of XRCC1, but not of DNA ligase III, was heavily down-regulated in myotubes. Western blotting analysis showed that both XRCC1 and DNA ligase III protein levels dropped during differentiation (Fig. 5D) in a coordinated manner, as expected from the requirement of XRCC1 for the stability of DNA ligase III (13).
|
Myotubes Accumulate Unrepaired DNA Damage After Oxidative Stress.
To gain insights into the biological relevance of the decrease in BER activity associated with muscle terminal differentiation, we explored the response to oxidative stress of proliferating versus terminally differentiated cells. Cells were exposed to H2O2, and the formation and repair of DNA SSB and alkali-labile sites (mainly abasic sites) were measured by the comet assay. The tail moment (TM) that reflects the frequency of breaks was used to quantify DNA damage. In myotubes, the level of damage induced by 30-min exposure to 100 µM H2O2 was significantly higher than in myoblasts and persisted for longer times (Fig. 6 A and B). The distribution of proliferating and terminally differentiated cell population by DNA damage classes (according to TM values) is provided in SI Fig. 9. Similar results were obtained after 30-min exposure to 50 µM H2O2 (data not shown). Unrepaired SSB are expected to be converted into double-strand breaks (DSB) at replication. Histone H2AX is rapidly phosphorylated in the chromatin microenvironment surrounding a DSB (14). The formation and repair of H2AX foci were evaluated. The number of foci-positive nuclei immediately after 30-min treatment was significantly higher in myotubes than in myoblasts exposed to 100 µM H2O2 (Fig. 6C). The percentage of damaged nuclei gradually declined during the posttreatment repair time in both cell types, but with slower kinetics in terminally differentiated, compared with proliferating, cells. Similar results were obtained after 30-min exposure to 1 mM H2O2 (data not shown). Interestingly, H2AX foci were present in proliferating cells exclusively in S-phase cells (BrdU-stained), which suggests that they identify DSB arising at replication from unrepaired SSB (Fig. 6D).
|
|
Discussion |
|---|
|
TopAbstractResultsDiscussionMaterials and MethodsAcknowledgementsReferences |
|---|
, provide a mechanistic rationale for the reduced levels of both short- and long-patch BER in myotubes. Down-regulation of DNA ligase I, PCNA, Flap endonuclease-1 (FEN-1), and replicative DNA polymerases 
/
are expected during the genome reprogramming toward differentiation known to involve repression of cell proliferation-associated genes (8). Long-patch repair synthesis is PCNA-dependent and involves FEN-1. The replicative DNA polymerases (2) and DNA ligase I have been shown to be indispensable for the resealing step (15, 16). Therefore, changes in the gene expression of these key players of long-patch BER are likely to account for the impairment of this pathway in myotubes. XRCC1 is critical for the in vivo stabilization of the DNA ligase III protein (13). Accordingly, a decrease in DNA ligase III levels was detected during the muscle differentiation process. XRCC1 plays a crucial role in the coordination of BER and SSBR because of its multiple interactions with other repair proteins (17). In addition to DNA ligase III, XRCC1 interacts with several enzymes involved in BER/SSBR pathways, including APE1, Pol β, PNK, PCNA, and poly (ADP-ribose) polymerase 1 and 2. Therefore, XRCC1 is not only involved in the ligation step of BER/SSBR, but also in earlier repair steps. This finding would explain the accumulation of repair intermediates (incision and repair synthesis products) observed in terminally differentiated myotubes that are likely to represent echoes of defective loading of the BER machinery due to low levels of XRCC1 or defective ligation. XRCC1 down-regulation in terminally differentiated cells might underscore the coregulation of cell cycle genes and XRCC1 expression. It has been demonstrated recently (18) that the transcription factor Forkhead Box M1 (FoxM1), which regulates the expression of cell cycle genes, has a role in XRCC1 expression. FoxM1-deficient cells display increased DNA breaks and reduced expression of XRCC1.
Among repair pathways, NER has been the most extensively analyzed during differentiation. NER was found to be generally repressed in terminally differentiated cells while remaining proficient in active genes (3, 4). Regulation at the level of the ubiquitin-activating enzyme E1 was recently proposed as a strategy to coordinate the NER and other changes occurring during differentiation (19). Our study provides the first mechanistic analysis of the decreased BER activity of differentiated cells. The analysis of the regulation of BER during differentiation should be the challenge for future studies.
What is the biological significance of reduced BER in myotubes? XRCC1 mutant mouse embryos, which die at embryonic day 6.5 (20), were reported to accumulate unrepaired SSB in the egg cylinder and presented a high frequency of spontaneous sister chromatid exchanges. It has been hypothesized that the stage-specificity for embryo lethality in the absence of XRCC1 is caused by the high level of ROS due to increased metabolic activity during gastrulation. We provide evidence that in muscle cells the antioxidant cell response is altered during differentiation and may potentiate damage by the ROS produced during this process. In myotubes, this result increases endogenous oxidative DNA damage, indicated by the increase in 8-OH-Gua content. The endogenous ROS effects are limited, as shown by the relatively low levels of spontaneous DNA SSB and H2AX foci, which are comparable to those of proliferating myoblasts. However, under increased free radical injury, DNA repair capacity becomes the limiting factor in the protection of genome integrity. Upon exposure to hydrogen peroxide, myotubes that are characterized by decreased XRCC1 levels and impaired DNA ligase activity accumulate DNA lesions as shown by the significantly higher frequency and persistence of both DNA SSB and H2AX-positive nuclei, compared with proliferating cells. What is the identity of these lesions? After oxidative DNA damage, SSB are generated either directly by ROS or as repair intermediates during BER/SSBR. The comet assay reveals the presence of SSB and abasic sites induced by H2O2. As expected, their repair is rapid in proliferating cells, but delayed in myotubes characterized by inefficient BER/SSBR. The interpretation of H2AX foci formation is more complex. Phosphorylation of H2AX is an early event in the ATM-dependent DNA damage-signaling pathway, which has been shown to be active in both myoblasts and myotubes after induction of DSB by ionizing radiation exposure (21). Evidence is accumulating that H2AX foci also identify DNA damage-containing structures other than DSB (22). In the case of myoblasts, H2AX foci were exclusively present in S-phase cells, indicating that they mark the formation of DSB produced by the replication of DNA templates containing SSB. In myotubes, which are unable to replicate, H2AX foci are likely to signal the persistence of SSB-containing DNA structures due to attenuated BER/SSBR activity.
Recently, defects in SSBR involving accumulation of abortive DNA repair intermediates have been described in association with neurodegenerative diseases (e.g., several ataxias) (5, 6). Brain and muscle have a high rate of oxygen metabolism that might exacerbate a defect in DNA break repair in nonproliferating, terminally differentiated systems that are deprived of long-patch BER (this study), as well as of alternative replication-associated repair mechanisms such as homologous recombination.
In muscle cells after injury, satellite cells are stimulated to replace dead cells by replicating, but depletion of this reservoir has been reported during aging (23) and pathological conditions (24). During the aging process, ROS production may drastically increase because of altered function of the respiratory chain and insufficient levels of the cellular antioxidant defenses. Such an oxidative insult, combined with a less efficient BER system and the lack of postmitotic repair processes, might play a key role in the age-related decrease of muscle performance and mass (sarcopenia) and in muscle disorders associated with free radical overproduction.
|
Materials and Methods |
|---|
|
TopAbstractResultsDiscussionMaterials and MethodsAcknowledgementsReferences |
|---|
Cells and Growth Conditions. Murine skeletal muscle satellite cells were isolated, cultured, and differentiated as described previously (14).
Cell Extracts. Whole-cell extracts for BER assays (Tanaka type) were prepared as described (27). Nuclear cell extracts for Western blot analysis were prepared by resuspending cells in ipotonic buffer. To compensate for the higher protein content of myotubes than myoblasts, samples in all assays were normalized so that lysates of cells possessed the same total number of nuclei.
Plasmid Assays. pGEM-3Zf (+) plasmid DNA molecules containing a site-specific AP site were constructed as previously described (25). The plasmids containing the lactone precursor residue (1'-t-butylcarbonyluridylate) were constructed as described (12). Repair of the plasmid DNA containing a regular abasic site or a lactone residue (pGEM-AP/dL) was conducted as described (25). All experiments were repeated at least three times, and representative experiments are shown.
Ligation Assay.
Two oligonucleotides, a [-32P]ATP-5'-end labeled 30 nt and a 31 nt, were annealed to a 61-nt complementary strand to obtain duplex DNA molecules bearing a single nick. The incubation buffer was the same described for the BER plasmid assays.
Western Blot Analysis. Proteins were separated on 4–12% polyacrylamide gels and analyzed by Western blotting with the following antibodies: monoclonal antibodies to DNA Ligase I (a gift of A. Montecucco, Institute of Molecular Genetics, Consiglio Nazionale delle Ricerche, Pavia, Italy), DNA Ligase III (BD Biosciences PharMingen, San Diego, CA), XRCC1 (a gift of K. Caldecott, Genome Damage and Stability Center, University of Sussex, Falmer, Brighton, U.K.), polyclonal antibodies to Pol β (a gift of S. H. Wilson, Laboratory of Structural Biology, National Institutes of Health, Research Triangle Park, NC), and Ref1 (APE1) (sc-334; Santa Cruz Biotechnology, Santa Cruz, CA). Western blots were developed by using the West Dura kit (Pierce Chemical, Rockford, IL).
Measurement of Gene Expression. Total RNA was isolated from myoblasts and myotubes by using an RNeasy mini kit (Qiagen, Valencia, CA). The expression levels of selected genes were measured by quantitative RT-PCR using primers pairs designed by the manufacturer (Applied Biosystems, Foster City, CA).
Measurement of 8-Hydroxyguanosine by HPLC/ED. DNA levels of 8-hydroxyguanosine were measured by HPLC with electrochemical detection as described previously (26).
Cell Treatment and Analysis of DNA Damage.
Myoblasts and myotubes were treated with various doses of hydrogen peroxide for 30 min at 37°C (50 and 100 µM in the case of the comet assay, 100 µM and 1 mM in the case of the H2AX foci assay) and harvested at different periods of time. DNA breaks were measured by the comet assay as previously described (27) with minor modifications. The analysis of H2AX foci was performed by incubating cells with mouse monoclonal anti-H2AX antibody (Upstate Biotechnology, Lake Placid, NY) alone or in combination with Alexa488-conjugated anti-BrdU (Invitrogen, Carlsbad, CA). Cells were counterstained with Hoechst 33258 dye. Only nuclei showing 10 bright foci were considered positive.
|
Acknowledgements |
|---|
|
TopAbstractResultsDiscussionMaterials and MethodsAcknowledgementsReferences |
|---|
|
Footnotes |
|---|
Author contributions: L.N. and P.F. contributed equally to this work; B.D., M.C., and E.D. designed research; L.N., P.F., D.P., A.F., P.L., P.D., and M.F. performed research; and E.D. wrote the paper.
Present address: Exsymol, Avenue Albert II 4, 98000 Monaco, Princedom of Monaco. 
The authors declare no conflict of interest.
This article is a PNAS Direct Submission.
This article contains supporting information online at www.pnas.org/cgi/content/full/0701743104/DC1.
© 2007 by The National Academy of Sciences of the USA
|
References |
|---|
|
TopAbstractResultsDiscussionMaterials and MethodsAcknowledgementsReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
