Skip to main content
NCBI home page
iScience logoLink to iScience
. 2021 Apr 17;24(5):102432. doi: 10.1016/j.isci.2021.102432

MyoD induces ARTD1 and nucleoplasmic poly-ADP-ribosylation during fibroblast to myoblast transdifferentiation

Lavinia Bisceglie 1,2, Ann-Katrin Hopp 1, Kapila Gunasekera 1, Roni H Wright 3,4, François Le Dily 3, Enrique Vidal 3, Alessandra Dall’Agnese 5, Luca Caputo 5, Chiara Nicoletti 5, Pier Lorenzo Puri 5, Miguel Beato 3,6, Michael O Hottiger 1,7,∗
PMCID: PMC8102911  PMID: 33997706

Summary

While protein ADP-ribosylation was reported to regulate differentiation and dedifferentiation, it has so far not been studied during transdifferentiation. Here, we found that MyoD-induced transdifferentiation of fibroblasts to myoblasts promotes the expression of the ADP-ribosyltransferase ARTD1. Comprehensive analysis of the genome architecture by Hi-C and RNA-seq analysis during transdifferentiation indicated that ARTD1 locally contributed to A/B compartmentalization and coregulated a subset of MyoD target genes that were however not sufficient to alter transdifferentiation. Surprisingly, the expression of ARTD1 was accompanied by the continuous synthesis of nuclear ADP ribosylation that was neither dependent on the cell cycle nor induced by DNA damage. Conversely to the H2O2-induced ADP-ribosylation, the MyoD-dependent ADP-ribosylation was not associated to chromatin but rather localized to the nucleoplasm. Together, these data describe a MyoD-induced nucleoplasmic ADP-ribosylation that is observed particularly during transdifferentiation and thus potentially expands the plethora of cellular processes associated with ADP-ribosylation.

Subject areas: Biological Sciences, Molecular Biology, Cell Biology, Developmental Biology

Graphical abstract

graphic file with name fx1.jpg

Open in a new tab

Highlights

  • •

    MyoD-dependent transdifferentiation of IMR90 to myoblasts induces ARTD1 expression

  • •

    Transdifferentiation induces nuclear ARTD1-dependent ADP-ribosylation in myoblasts

  • •

    This ADP-ribosylation is induced independent of cell cycle and of DNA damage

  • •

    ARTD1-mediated poly-ADP-ribosylation localizes to the nucleoplasm in myoblasts

Biological Sciences; Molecular Biology; Cell Biology; Developmental Biology

Introduction

Transdifferentiation is the conversion of a fully differentiated cell type into another type without dedifferentiation and reacquisition of progenitor-like features (Mills et al., 2019). Unlike other types of transdifferentiation that rely on the co-operation of multiple transcription or chemical factors (Ieda et al., 2010; Mollinari et al., 2018), MyoD is sufficient to convert fibroblasts and other cell types to myoblasts (Davis et al., 1987; Sartorelli and Puri, 2018; Choi et al., 1990; Weintraub et al., 1989). However, MyoD alone fails to transdifferentiate HeLa (Weintraub et al., 1989), HepG2 (Weintraub et al., 1989), or P19 cells (Skerjanc et al., 1994), suggesting that the chromatin landscape of the initial cell type affects the accessibility of the DNA binding sites of MyoD (i.e. E boxes) and thus its ability to induce the myogenic program (Sartorelli and Puri, 2018). Moreover, trimethylation of histone 3 lysine 27 (H3K27me3) reduced the binding ability of MyoD to chromatin (Caretti et al., 2004; Sartorelli and Puri, 2018) and depletion of polycomb repressor complex 2 (PRC2) increased MyoD-dependent cell-type conversion (Patel et al., 2012) highlighting that histone post-translational modifications (PTMs) plays a crucial role in tuning transdifferentiation. This is supported by the observation that in some epigenetic contexts, the interaction of MyoD with the pioneer factor Pbx overcomes the limitations associated with the accessibility of the E boxes (Berkes et al., 2004; Maves et al., 2007). Moreover, components of the SWItch/Sucrose Non-Fermentable (SWI/SNF) complex, for example, Brg1 and ATPase Brahma, are essential for successful fibroblasts to myoblasts conversion (de la Serna et al., 2001), and human embryonic stem cells are unable to differentiate to muscle cells owing to the lack of Baf60c (Albini et al., 2013).

ADP-ribosylation (ADPR) is a dynamic PTM, involving the transfer of ADP-ribose units from NAD+ to specific amino acids or ADP-ribose itself with concomitant release of nicotinamide (Kraus, 2020). While the transfer of one ADP-ribose unit is called mono-ADPR or MARylation, the formation of a linear or branched chain is defined as poly-ADPR or PARylation (PAR) (Gibson and Kraus, 2012). ADPR is a fully reversible PTM, catalyzed by ADP-ribosyltransferases (Hottiger et al., 2010) and removed by ADPR hydrolases with different subcellular localizations (Luscher et al., 2018). Among the nuclear enzymes responsible for ADPR catalysis, ARTD1 (also PARP1) is the best studied, mainly for its function in DNA damage response pathways and cancer progression (Azarm and Smith, 2020). ARTD1 activity is regulated by its flexible helical subdomain (HD), that in the properly folded status covers the NAD+ pocket, while binding to DNA changes the protein conformation, thus unfolding the HD domain and allowing access of NAD+ (Alemasova and Lavrik, 2019). In addition, activation of ARTD1 can be mediated by RNA (Kim et al., 2019) and regulated by other PTMs (Piao et al., 2014). Although the function of ARTD1 was intensively studied in the last decade, the mechanisms regulating its expression are poorly understood. While highly expressed in cancer cells, ARTD1 protein levels are rather low in noncancer cells as IMR90, skin fibroblasts, and hepatocytes (Chen et al., 2014). Its expression is regulated during cell cycle progression and cell-cycle arrest in G1 leads to the repression of ARTD1 by binding of the HDAC-PRC2-SWI/SNF complex to its promoter and deposition of the repressive H3K27me3 histone mark (Wisnik et al., 2017; Pietrzak et al., 2018).

ADPR has been observed in several models of cell differentiation, either promoting or preventing the progression of differentiation depending on the context, timing, and NAD+ availability (Abplanalp and Hottiger, 2017; Ryu et al., 2018). For instance, ADPR levels are modulated during adipogenesis to either repress or promote gene expression in different phases of the differentiation (Szanto and Bai, 2020) and application of the ADP-ribose chromatin-affinity purification method confirmed the enrichment of chromatin ADPR at promoters of PPARγ target genes during the later stage of adipogenesis (Bartolomei et al., 2016). Interestingly, the expression of ARTD1 and its activity increased during reprogramming and depletion or inhibition of ARTD1 strongly decreased the efficiency of reprogramming (Chiou et al., 2013; Weber et al., 2013). Although ARTD1-dependent ADPR has been reported to play important roles in several differentiation and reprogramming pathways (Abplanalp and Hottiger, 2017), the existence and the contribution of ADPR during transdifferentiation was so far not investigated. In this study, we observed that MyoD-dependent transdifferentiation of fibroblasts to myoblasts induced nuclear ARTD1 levels and consequently increased nuclear ADPR that was catalyzed in a DNA damage-independent manner and was not associated to chromatin but rather localized to the nucleoplasm.

Results

MyoD transdifferentiates IMR90 fibroblasts to myotubes

While ADPR was studied in several types of differentiation and reprogramming, its presence in transdifferentiation is so far still unknown. Thus, the MyoD-induced system was established, which has the unique property to allow distinguishing between the transdifferentiation step (i.e. cell type conversion) and the terminal differentiation (Dall'Agnese et al., 2019), allowing us to investigate ADPR specifically during transdifferentiation independently of its function in differentiation. The murine MyoD cDNA controlled by a doxycycline (dox)-inducible promoter was stably integrated into human IMR90 fibroblasts using the PiggyBac system (Dall'Agnese et al., 2019). IMR90 fibroblasts selected for the integrated MyoD construct (MyoD+ fibroblasts) were subsequently treated with dox and the expression of MyoD was controlled by immunofluorescence (IF) microscopy. While MyoD was not detectable in untreated MyoD+ fibroblasts, dox-treatment strongly induced its expression (Figure S1B). After MyoD-dependent transdifferentiation of human fibroblasts to myoblasts, these can be further differentiated to myotubes via serum deprivation (Figure S1A). The ability of the MyoD+ myoblasts to further differentiate to myotubes was investigated by qPCR analysis of known myogenic markers such as Myogenin, MyoD, and muscular creatine kinase. Gene expression of untreated or dox-treated MyoD+ cells after transdifferentiation (GM1) and either 1, 2, 3, or 5 days after induction of differentiation (DM1, DM2, DM3, DM5, respectively; Figures S1A and S1C) revealed that dox treatment and subsequent serum deprivation increased the expression of all three tested genes (Figure S1C). Because myotubes are multinucleated and characterized by high levels of myosin heavy chain (MHC), MHC expression was analyzed by IF at DM5. Differentiated MyoD+ cells were multinucleated and expressed MHC in a dox-dependent manner (Figure S1D), confirming that the induction of MyoD successfully transdifferentiated IMR90 fibroblasts to committed myoblasts. To investigate ADPR exclusively during MyoD-dependent transdifferentiation, excluding its contribution to muscle differentiation, hereafter we exclusively focused on the first 24 h after dox treatment (GM1, Figure S1A).

MyoD induces ARTD1 expression in transdifferentiated IMR90

Because ARTD1 participates in several differentiation and reprogramming systems (Abplanalp and Hottiger, 2017) but its contribution to transdifferentiation is unknown, we took advantage of the MyoD-induced transdifferentiation system to investigated its expression exclusively in transdifferentiated cells, thus excluding its contribution to differentiation. MyoD+ cells were treated with dox, and protein levels of both ARTD1 and MyoD were analyzed at the single-cell level by quantitative image-based cytometry (QIBC) (Toledo et al., 2013) (Figure 1A). In contrast to other reported systems, where the level of ARTD1 was easily detectable (Erener et al., 2012; Luo et al., 2017), it was weakly expressed in IMR90 fibroblasts and the dox-induced expression of MyoD significantly increased ARTD1 protein levels, as observed by IF but also by WB (Figures 1A and 1B), suggesting that MyoD might regulate the expression of ARTD1. Analysis of the ARTD1 mRNA by qPCR indeed confirmed its increase in MyoD+ cells upon dox-treatment. Conversely, control fibroblasts treated with dox (i.e. MyoD− cells) did not show any change in ARTD1 levels, confirming that the observed increase of ARTD1 is not a consequence of the dox treatment itself but is indeed MyoD-dependent (Figure 1C). To test whether MyoD can directly bind the promoter of ARTD1, recently published ChIP-seq data of MyoD were analyzed (Dall'Agnese et al., 2019). MyoD binding was indeed observed at the proximal promoter of ARTD1 in GM1 MyoD+ cells (Figure 1D), suggesting that MyoD directly promotes ARTD1 expression. Furthermore, the observed increased acetylation of histone 3 lysine 27 (H3K27ac) at the transcriptional start site of ARTD1 in GM1 MyoD+ cells confirmed the upregulation of ARTD1 transcription in transdifferentiated cells.

Figure 1.

Figure 1

Open in a new tab

MyoD induces ARTD1 expression in IMR90 cells

(A) IF of untreated (untr) or dox-treated (dox) MyoD+ fibroblasts using anti-MyoD and anti-ARTD1 antibodies (magnification 20×). Scale bars indicate 20μm. Quantification on the right: the nuclear mean fluorescence intensity (arbitrary unit) of each event was normalized over the mean of the control/untreated, arbitrarily set to 30. The Y axes of violin plots are shown as log10 scales. For statistical analysis, a ratio-paired student t-test was used (n = 3–5; ∗, p < 0.05; ∗∗, p < 0.005; ∗∗∗, p < 0.0005).

(B) WB analysis of untreated (dox -) or dox-treated (dox +) MyoD+ fibroblasts upon siMock or siARTD1.

(C) qPCR for ARTD1 in untreated (untr) or dox-treated (dox) MyoD+ and MyoD− IMR90. Data are shown as mean ± SD. For statistical analysis, 3 to 5 independent experiments were compared using a ratio-paired student t-test with ∗, p < 0.05; ∗∗, p < 0.005; ∗∗∗, p < 0.0005.

(D) UCSC browser showing MyoD binding (pink track) on human ARTD1 gene of MyoD+ cells in GM1 and H3K27ac of MyoD− and MyoD+ cells in GM1 (black tracks).

To analyze whether the overexpression of MyoD is sufficient to increase ARTD1 levels in any cell line, we generated MyoD+ HEK 293 cells and repeated the experiments. Despite the strong induction of MyoD upon dox treatment, ARTD1 level remained unchanged in MyoD+ HEK 293 (Figures S2A and S2B), suggesting that the observed MyoD-dependent increase of ARTD1 is not ubiquitous. Furthermore, comparison of the basal (i.e. untreated) ARTD1 levels in IMR90 and HEK 293 cells revealed that ARTD1 levels were much lower in IMR90 cells (Figures S2B and 2C) (Chen et al., 2014), suggesting that the promoter of ARTD1 is differently regulated in the two tested cell lines. qPCR analysis for myogenic markers in MyoD+ HEK 293 cells treated with dox revealed that the overexpression of the murine MyoD did not activate the human myogenic program in this cell line (Figure S2D), confirming that not every cell line responds to MyoD-dependent cell conversion, most likely owing to the lack of MyoD cofactors (Choi et al., 1990; Skerjanc et al., 1994; Weintraub et al., 1989). Indeed, these results confirmed that the observed MyoD-dependent increase of ARTD1 is not ubiquitous in every cell line but is rather specific of cells able to transdifferentiate.

ARTD1 does not contribute to MyoD-dependent regulation of chromatin compartmentalization

MyoD-induced transdifferentiation was described to be associated with an extensive three-dimensional (3D) reorganization of the genome via binding of MyoD to CTCF boundaries (Dall'Agnese et al., 2019). Because ARTD1 interacts with CTCF to regulate chromatin architecture (Zhao et al., 2015), we tested whether and to which extent ARTD1 would affect the chromatin compartmentalization during transdifferentiation. Hi-C analysis of untreated or dox-treated MyoD+ cells confirmed that MyoD expression strongly altered A/B compartmentalization (Figure 2A, Table S1, S2). However, the eigenvector PCA analysis revealed that knockdown of ARTD1 only mildly affected A/B compartmentalization compared with the corresponding siMock control in both untreated and dox-treated MyoD+ cells (Figure 2A), suggesting that the presence of ARTD1 has only a mild, local effect on A/B compartmentalization, however, in a transdifferentiation-independent manner.

Figure 2.

Figure 2

Open in a new tab

MyoD-dependent ARTD1 regulates gene expression of a subset of MyoD target genes

(A) PCA eigenvector analysis of changes in A/B compartmentalization of untreated and dox-treated MyoD+ cells transfected with siMock or siARTD1 (shape indicates different batches; size indicates sequencing depth).

(B) Volcano plot showing the comparison of gene expression between untreated (untr) or dox-treated (dox) MyoD+ fibroblasts. (Red dots: downregulated genes; Blue dots: upregulated genes; Green dots: unchanged genes; cutoff: fold change ≥1.5; p value on the y axes).

(C) PCA analysis showing the variation of gene expression between untreated and dox-treated MyoD+ cells transfected with siMock or siARTD1 (each dot represents a biological replicate).

(D) Volcano plot showing the comparison of gene expression between dox-treated MyoD+ fibroblasts transfected with either siARTD1 or siMock. (Red dots: downregulated genes; Blue dots: upregulated genes; Green dots: unchanged genes; cutoff: fold change ≥1.5; p value on the y axes).

(E) Heat map of differential gene expression between MyoD+ cells untreated (untr), dox-treated (dox) transfected with either siMock or siARTD1 (cut-off: ≥ 2-fold difference between samples).

Because knock down of ARTD1 does not allow distinguishing between the function of the protein and its enzymatic activity, the Hi-C analysis was repeated in presence of the PARP inhibitor Niraparib (PARPi) (Figure S3A). The samples treated with niraparib clustered slightly differently from the ones treated with siARTD1 (Figure S3A), suggesting that ARTD1 enzymatic activity might locally affect A/B compartmentalization but in a transdifferentiation-independent manner. Interestingly, cotreatment with niraparib and siARTD1 did not induce any additive changes to the knockdown of ARTD1 alone (Figure S3A), confirming that ARTD1 is the main ARTD family member contributing to A/B compartmentalization under the tested conditions. However, because both niraparib treatment and ARTD1 knockdown induced comparable changes in untreated as well as dox-treated cells, we concluded that the MyoD-induced ARTD1 and its enzymatic activity do not contribute to transdifferentiation by affecting the chromatin compartmentalization. This results also suggest that the newly transcribed, MyoD-dependent ARTD1 pool exerts a distinct function than the already present ARTD1 pool, indicating that functionally different ARTD1 pools can coexist at the same time.

ARTD1 mildly regulates the expression of a subset of MyoD-dependent genes

Because ARTD1 was described as transcription cofactor in several differentiation processes (Abplanalp and Hottiger, 2017), we tested whether it would coregulate MyoD-induced target genes. Therefore, RNA sequencing was performed on siMock MyoD+ cells with or without dox treatment (siMock untreated versus siMock dox) and on dox-treated siARTD1 MyoD+ cells (siARTD1 dox). The comparison of untreated versus dox-treated siMock MyoD+ cells confirmed that MyoD strongly induces gene expression changes (Figure 2B) and gene ontology (GO) analysis confirmed that MyoD activated the myogenic program (Dall'Agnese et al., 2019) (Figure S3B). However, PCA analysis of the three samples revealed that while untreated and dox-treated MyoD+ cells significantly differed from each other, the transcription profiles of cells transdifferentiated in presence or absence of ARTD1 were similar (Figure 2C, 2% variance). A more detailed analysis of the differentially expressed genes of siMock dox and siARTD1 dox samples indicated that ARTD1 regulates the gene expression of a subset of upregulated as well as downregulated genes (Figure 2D). A heat map comparing the expression of genes differentially regulated in all the three samples allowed the identification of four clusters of genes (Figure 2E, Table S3: Deferentially regulated genes detected by RNAseq, Related to Figure 2). GO analysis of genes in cluster 1, whose expression was enhanced in absence of ARTD1, revealed that these genes were associated with general biological processes (e.g. cytoskeleton regulation and cell motility) (Figure S3C). GO analysis of genes in cluster 2 (the largest cluster, whose genes were induced by MyoD and showed reduced expression in absence of ARTD1) correlated with muscle differentiation and function (Figure S3D), genes of cluster 3 (that were repressed by MyoD but activated in absence of ARTD1) associated with different types of differentiation (e.g. chondrogenesis, osteogenesis, neurogenesis) (Figure S3E). Finally, GO analysis of genes in cluster 4 (that were repressed by MyoD and further reduced in absence of ARTD1) included inflammatory response genes, whose regulation by MyoD was reported before (Dall'Agnese et al., 2019) (Figure S3F). These results suggested that ARTD1 coregulates the expression of a small subset of MyoD target genes (304 genes, 8% of MyoD target genes, Figure S3G), some of which favor transdifferentiation to myoblasts and prevent the induction of transdifferentiation to other cell types (i.e. clusters 2 and 3).

MyoD induces ARTD1-dependent nuclear ADP-ribosylation

To investigate whether MyoD-induced ARTD1 catalyzed PAR in transdifferentiated cells, the presence of PAR was assessed by QIBC in untreated and dox-treated MyoD+ cells. Intriguingly, 24 h after dox treatment, we observed a significant increase of nuclear PAR (Figure 3A) that could also be detected with another anti-PAR antibody (Figure S4A). In dox-treated MyoD− cells, the basal nuclear PAR levels were barely detectable and did not change after dox treatment (Figure S4B), confirming that the increase of nuclear PAR was dependent on MyoD and was not a consequence of the dox treatment.

Figure 3.

Figure 3

Open in a new tab

MyoD induces ARTD1-dependent nuclear ADP-ribosylation

(A) IF of untreated (untr) or dox-treated (dox) MyoD+ fibroblasts using anti-MyoD and anti-PAR antibodies (magnification 20×). Scale bars indicate 20μm. Quantification on the right.

(B) IF of PARPi treatment of untreated or dox-treated MyoD+ fibroblasts using anti MyoD and anti-PAR antibodies (Nir: niraparib, magnification 20×) Scale bars indicate 20μm. Quantification of PAR on the right.

(C) IF of untreated or dox-treated MyoD+ fibroblasts upon siMock or siARTD1 transfection using anti-MyoD and anti-PAR antibodies (magnification 20×). Scale bars indicate 20μm. Quantification of PAR on the right.

(D) Quantification of the IF of MyoD+ fibroblasts either untreated (untr) or dox-treated (dox) using anti-PAR and anti-MyoD antibodies. After 24h (GM1), dox was either maintained for additional 24 or 48 h (GM1 and GM2) or removed for the same time.

(E) Quantification of IF of MyoD+ fibroblasts either untreated or dox-treated using anti-PAR, anti-MyoD and anti-ARTD1 antibodies. After 24 h (GM1), dox was either maintained for additional 24 h (GM2) or removed for the same time. Niraparib (Nir) was added at GM1 together with dox withdrawal (GM2 dox-, nir+). For every quantification, the IF signal was normalized as described in Figure 1. The Y axes of all violin plots are shown as log10 scales. For statistical analysis, a ratio-paired student t-test was used (n = 3-5; ∗, p < 0.05; ∗∗, p < 0.005; ∗∗∗, p < 0.0005).

To strengthen that the MyoD-dependent nuclear PAR signal was enzymatically catalyzed, MyoD+ cells were cotreated with dox and either Niraparib or Rucaparib, two different PARPi. Both inhibitors completely abolished the MyoD-dependent nuclear PAR (Figure 3B and S4C). To finally confirm that ARTD1 is the enzyme responsible for the observed PAR signal, ARTD1 was knocked down by siRNA in MyoD+ cells, and these cells were subsequently treated with dox to induce transdifferentiation. As expected, knockdown of ARTD1 significantly reduced ARTD1 protein (Figure S4D) and, similar to the PARPi treatment, prevented the nuclear PAR accumulation (Figure 3C), confirming that ARTD1 is responsible for the PAR formation in transdifferentiation.

MyoD-induced nuclear ADP-ribosylation is dynamic and maintained in a MyoD-independent manner

Because continuous expression of MyoD is required to sustain transdifferentiation (Dall'Agnese et al., 2019), we tested whether the expression of ARTD1 and the accumulation of PAR would be reversible when the expression of MyoD is terminated by removing dox from the cells. Therefore, cells were transdifferentiated by dox treatment (GM1) and then dox was either maintained or removed for additional 24 or 48 h (GM2 and GM3, respectively). When dox was removed, the expression of MyoD drastically declined already within 24 h (GM2). Interestingly, neither ARTD1 nor nuclear PAR levels followed this drastic reduction but remained significantly higher as compared with untreated cells even 48 h after dox removal (Figure 3D), suggesting that after the initial induction by MyoD, the expression of ARTD1 is regulated in a MyoD-independent manner. We therefore compared the expression of ARTD1 and the MyoD target gene Myogenin after dox removal. Although MyoD binds the promoters of both genes after dox induction, the expression of Myogenin was immediately reduced after removal of dox, while ARTD1 levels remained high even in absence of MyoD (Figures S4E and 4F), suggesting that once the expression of ARTD1 is induced, MyoD is not further required. Moreover, this result suggested that MyoD-dependent ARTD1 might play a function in transdifferentiation that is not directly linked to the expression of the myogenic program but rather points toward a different function of ARTD1 in transdifferentiation.

Figure 4.

Figure 4

Open in a new tab

MyoD-dependent ADP-ribosylation is independent of dsDNA breaks

(A) γ-H2AX levels depicted as a function of cell cycle progression of untreated or dox-treated MyoD+ cells, in presence of etoposide. Each dot represents a single cell. (Gray: low γ-H2AX; blue: high γ-H2AX).

(B) γ-H2AX levels depicted as a function of cell cycle progression and color-coded for nuclear PAR levels of untreated or dox-treated MyoD+ cells, in presence of etoposide. Each dot represents a single cell (Gray: low PAR; red: high PAR).

(C) Nuclear 53BP1 foci depicted as a function of cell cycle progression of untreated or dox-treated MyoD+ cells, in presence of etoposide. Each dot represents a single cell (Gray: low 53BP1; yellow: high 53BP1).

(D) Nuclear PAR levels depicted as a function of cell cycle progression of untreated or dox-treated MyoD+ cells, in presence of etoposide. Each dot represents a single cell (Gray: low PAR; red: high PAR).

ARTD1-induced PAR formation is very dynamic, and the half-life of the PAR is estimated to be within the minute range (Andersson et al., 2016). We thus investigated whether the PAR signal observed in absence of MyoD was continuously synthesized and degraded or rather stable. Therefore, MyoD+ cells were treated with dox for 24 h (GM1) to induce ARTD1 and PAR and niraparib was added when dox was removed for additional 24 hr (GM2) (Figure 3E). While PAR levels were comparable in GM2 both in presence and absence of dox, niraparib treatment inhibited the nuclear PAR formation at GM2 (Figure 3E), suggesting that the nuclear PAR indeed undergoes continuous turnover, being synthesized by ARTD1 and likely degraded by an active nuclear PAR-eraser.

MyoD-dependent PAR synthesis is cell cycle independent and not induced by DNA damage

Because the activation of ARTD1 was linked to replication stress (Hanzlikova et al., 2018; Maya-Mendoza et al., 2018), we took advantage of the QIBC technology to analyze the distribution of PAR at the single-cell level and in a cell-cycle-dependent manner. The analysis confirmed that most of the IMR90 fibroblasts were accumulated in G1 phase before and after transdifferentiation (Figure S5A) (Dall'Agnese et al., 2019), however, cells with high or low PAR signals did not accumulate in any particular cell cycle phase (Figure S5B), thus excluding a functional contribution of MyoD-dependent PAR in cell cycle progression.

Alternatively, the activity of ARTD1 can be induced by damaged DNA (Azarm and Smith, 2020). We thus tested whether double-strand (ds) DNA breaks would activate ARTD1 during transdifferentiation. To test whether dsDNA breaks would induce PAR formation, we analyzed the accumulation of γ-H2AX and 53BP1 foci upon transdifferentiation by QIBC including treatment with the topoisomerase II inhibitor etoposide as a positive control. As expected, in untreated IMR90, etoposide induced γ-H2AX mainly in cells allocated to the S-phase (Figure 4A). Interestingly, although a similar induction of γ-H2AX was observed in dox-treated cells without etoposide treatment, the high signal of γ-H2AX was again S-phase specific (Figure 4A), while cells with high MyoD-dependent PAR levels were equally distributed in all phases of the cell cycle (Figures 4B and S5B), suggesting that dsDNA breaks are unlikely responsible for the observed PAR formation. Moreover, dox treatment alone did not induce 53BP1 foci formation (Figure 4C), which was instead increased after the etoposide treatment (Figure 4C), providing further evidence that MyoD-dependent PAR is not induced by DNA damage but rather points to another yet to be defined mechanism that is peculiar of IMR90 cells.

Interestingly, although the levels of PAR were unchanged in transdifferentiated cells treated with etoposide (Figure 4D), they showed higher γ-H2AX and 53BP1 signals than untreated cells after etoposide treatment (Figures 4A and 4C), suggesting that the nuclear PAR accumulated during transdifferentiation might enhance the formation of γ-H2AX foci and thus the DNA damage response, for example by sustaining Nudix-dependent ATP production (Wright et al., 2016). To test whether the synthesized PAR would indeed be responsible for the sensitization of the transdifferentiated cells to genotoxic stress, we repeated the previously described experiments upon ARTD1 depletion. However, knock down of ARTD1 during transdifferentiation did not affect the enhanced etoposide-induced γ-H2AX and 53BP1 foci formation after dox treatment (Figures S5C and 5D), suggesting that the MyoD-dependent PAR itself is not responsible for the increased sensitivity to DNA damage in transdifferentiated cells.

Figure 5.

Figure 5

Open in a new tab

MyoD-dependent nuclear ADP-ribosylation localizes in foci that are not chromatin-bound

(A) Confocal pictures of dox-treated (dox) or H2O2-treated (H2O2) MyoD+ fibroblasts stained with anti-PAR (zoom in of magnification 63×). Scale bars indicate 10 μm.

(B) IF for chromatin associated MyoD and PAR on MyoD+ fibroblasts untreated or dox-treated or treated with H2O2 after pre-extraction with 0.2% Triton (magnification 20×). Scale bars indicate 20 μm.

(C) IF of dox-treated MyoD+ fibroblasts using anti-H3K4me3 (upper panel), anti-H3K9me2 (lower panel) and anti-PAR antibodies (zoom in of magnification 20×) Scale bars indicate 10 μm.

(D) IF for chromatin associated MyoD and ARTD1 on MyoD+ fibroblasts either untreated or dox-treated or co-treated with Methyl-methanesulfonate and Olaparib (MMS + Ola), after pre-extraction with 0.2% Triton (magnification 20×). Scale bars indicate 20 μm.

MyoD-dependent ADP-ribosylation localizes to the nucleoplasm

To date, ARTD1-mediated PAR was associated to chromatin either contributing to DNA repair or regulating gene expression (Azarm and Smith, 2020; Luo et al., 2017; Hottiger, 2015). Because the MyoD-induced PAR was not involved in the regulation of the myogenic program and was not induced by DNA damage, we hypothesized that it might differ from the canonical one. When compared with H2O2-induced PAR, the signal intensity of the MyoD-dependent one was lower but detectable (Figure S6A). Interestingly, confocal microscopy revealed that the MyoD-dependent PAR signal displayed a different pattern profile in comparison with the H2O2-induced one, with big and round foci dispersed throughout the nucleus (Figure 5A).

Because histones are the main targets of ARTD1-mediated PAR upon H2O2 treatment (Gibbs-Seymour et al., 2016), we tested whether MyoD-induced PAR is associated to chromatin comparable with the described H2O2-induced one. Therefore, the chromatin association of PAR in untreated, dox-treated, or H2O2-treated MyoD+ cells was examined by IF in combination with a pre-extraction step (Figures 5B and S6B). While H2O2-induced PAR was as expected strongly chromatin-associated after a mild (i.e. 0.1% Triton) or a more stringent pre-extraction treatment (i.e. 0.2% Triton), MyoD-dependent nuclear PAR was undetectable in both pre-extraction conditions (Figures 5B, S6B), suggesting that it is not associated to chromatin and probably does not serve the same function than chromatin-bound PAR. Costaining of PAR with two different histone marks, histone 3 lysine 4 methylation (H3K4me3) and histone 3 lysine 9 di-methylation (H3K9me2) revealed that MyoD-dependent nuclear PAR did not colocalize with either mark (Figure 5C), confirming that MyoD-induced PAR is rather localized to the nucleoplasm.

PARylation of ARTD1 was suggested to evict it from the chromatin (Muthurajan et al., 2014). To analyze the chromatin retention of ARTD1 during transdifferentiation, dox-treated cells were pre-extracted and analyzed by IF, using methanesulfonate (MMS) and olaparib cotreated cells as the positive control (Michelena et al., 2018). While MMS/olaparib induced a strong chromatin retention of ARTD1, transdifferentiation did not increase the affinity of ARTD1 for chromatin (Figures 5D, S6C). Moreover, as additional control, endogenous topoisomerase II was successfully detected on chromatin even after stringent pre-extraction (Figure S6D). Furthermore, chromatin fractionation followed by WB confirmed that ARTD1 is accumulated in the soluble fraction and does not associate to chromatin during transdifferentiation (Figure S6E). Altogether, these results revealed that the MyoD-induced ARTD1 indeed was not retained on chromatin and that MyoD-dependent PAR localized to the nucleoplasm, suggesting that automodification of ARTD1 might lead to its nucleoplasmic localization.

Discussion

Although ARTD1 was described to coregulate several differentiation systems and cell reprogramming (Abplanalp and Hottiger, 2017), the contribution of ADPR to transdifferentiation is still unknown. Our results revealed that MyoD-dependent transdifferentiation induced ARTD1 expression, leading to increased nuclear PAR that was neither cell-cycle-dependent nor induced by DNA damage. MyoD-dependent nuclear PAR was highly dynamic being continuously synthesized and removed and unlike H2O2-induced PAR was not chromatin-bound but localized to the nucleoplasm.

Because we wanted to exclude the effect of the contribution of ARTD1 to further differentiation, the MyoD-dependent transdifferentiation of fibroblasts to myoblasts allowed us to focus exclusively on the transdifferentiation process (i.e. cell type conversion). In IMR90, MyoD bound to the promoter of ARTD1 and increased its expression, leading to increased nuclear ARTD1 levels. This observation differed from what was shown for most of the studied differentiation systems where ARTD1 levels were rather stable or even decreased (Erener et al., 2012; Luo et al., 2017; Wisnik et al., 2017), pointing toward a new mechanism of regulating ARTD1 expression in cellular plasticity. The MyoD-dependent increase of ARTD1 was not observed when MyoD was overexpressed in HEK 293 cells, suggesting that cells unable to transdifferentiate (e.g. HEK293) do not show any increase of ARTD1 expression. This result might be explained by different chromatin landscapes in different cell lines. Interestingly, after transcriptional activation of the ARTD1 gene, removal of MyoD only slightly reduced the ARTD1 mRNA levels, while the expression of the canonical MyoD target gene Myogenin completely ceased, suggesting that the promoter region of ARTD1 once activated, remains accessible and the gene is continuously expressed. The reason behind this different behavior of the promoters of ARTD1 and Myogenin in response to MyoD is so far still unknown, but this observed difference suggests that while Myogenin is essential for the activation of the myogenic program (Andres and Walsh, 1996), MyoD-induced ARTD1 has rather a different function in transdifferentiation.

The increased ARTD1 protein levels even 48 h after MyoD withdrawal cannot be explained with the stability of the protein because siRNA experiments revealed that ARTD1 protein is undetectable already 48 h after knockdown, indicating that ARTD1 has quite a short life span. Thus, we speculate that the ARTD1 expression is regulated by a positive feedback loop. We cannot exclude that a pool of ARTD1 might be in complex with other transcription factors than MyoD, regulating its own expression. Although the regulation of ARTD1 transcription is not completely understood, several chromatin remodelers and transcription factors, such as Sp1, were described to modulate its expression (Zaniolo et al., 2005). MyoD recruits HATs and SWI/SNF and interacts with Sp1 (Tapscott, 2005). Thus, the expression of ARTD1 might be regulated and maintained by a transcription factor that binds to the ARTD1 promoter in a MyoD-dependent manner but then remains associated independently of MyoD.

Hi-C analysis during transdifferentiation revealed that MyoD is involved in chromatin rewiring at CTCF binding sites and MyoD-dependent changes in genome architecture are needed to activate the myogenic transcriptional program (Dall'Agnese et al., 2019). Moreover, CTCF was reported to be ADP-ribosylated by ARTD1 (Farrar et al., 2010) and interaction of ARTD1 and CTCF regulated the formation of Lamin-associated domains in a circadian-rhythm-dependent fashion (Zhao et al., 2015). Our Hi-C analysis confirmed that MyoD-dependent transdifferentiation led to strong changes in A/B compartmentalization, thus regulating the 3D genome architecture (Dall'Agnese et al., 2019). However, ARTD1 or its enzymatic activity had only a mild effect on A/B compartmentalization that was comparable both in fibroblasts and myoblasts, suggesting that the MyoD-dependent ARTD1 does not contribute to these changes.

The efficiency of transdifferentiation is very difficult to evaluate (Mills et al., 2019). An option to overcome this limitation is to compare transdifferentiated cells with normally differentiated ones. However, our current, dynamic envision of the epigenetic landscape rather suggests that cells changing their cellular identity by transdifferentiation use pathways that differ from the ones used for differentiation (Rajagopal and Stanger, 2016). Moreover, both transdifferentiation and differentiation are characterized by high levels of heterogeneity of chromatin landscapes and gene expression profiles, even between cells of the same population, thus making a comparison very difficult (Gulati et al., 2020). In this regard, although C2C12 myotubes and myotubes obtained from transdifferentiated IMR90 are globally comparable in terms of morphology and expression of myogenic markers, they are not identical because, for instance, they differ in the extent of ARTD1 expression and PAR formation (our own unpublished data). Our data rather point toward a different and distinct function of ARTD1 during differentiation and transdifferentiation. In this perspective, to investigate ARTD1's function in transdifferentiation by comparing transdifferentiated IMR90 with C2C12 would be very simplistic because the molecular pathway characterizing the two different types of myoblasts are very different. In addition, to dissect the role of ARTD1 in transdifferentiation by further differentiating MyoD-induced IMR90 to myotubes after ARTD1 knock down or inhibition would be inconclusive because ARTD1 is also involved in myogenesis (Olah et al., 2015) and to distinguish the function of ARTD1 for the initial cell type conversion versus its function during the subsequent differentiation would thus be very difficult. The best way to investigate ARTD1's contribution to MyoD-induced cell type conversion is to study fibroblasts to myoblasts transdifferentiation to focus exclusively on the initial step of the process. Our RNA-seq analysis during transdifferentiation confirmed the induction of the myogenic program. The number of MyoD target genes, whose expression was affected by ARTD1, was however quite small, and the loss of ARTD1 did not prevent overall transdifferentiation, suggesting that ARTD1 is dispensable for MyoD-dependent gene expression during transdifferentiation under the tested conditions. This is in line with the notion that MyoD alone is sufficient to activate the myogenic program (Sartorelli and Puri, 2018).

Although the enzymatic activity of ARTD1 is usually induced by replication stress or damaged DNA (Alemasova and Lavrik, 2019), the MyoD-dependent PAR formation was observed throughout the cell cycle and was independent of dsDNA breaks. In line with this finding, basal ARTD1 activity was reported in different cancer cell lines in absence of DNA damage (Krukenberg et al., 2014), suggesting the existence of a different mechanism for ARTD1 activation that could be used also in transdifferentiated IMR90. Further experiments are required to elucidate the molecular mechanism responsible for the induction of the enzymatic activity of ARTD1 in MyoD-dependent transdifferentiation.

The current manuscript provides evidence that MyoD-induced PAR is not involved in regulation of gene expression, does not contribute to MyoD-dependent 3D chromatin architecture, is not induced by DNA damage, and does not contribute to cell cycle regulation. Unfortunately, we could not identify yet the function of MyoD-dependent PAR induction. However, our data revealed that MyoD-dependent ARTD1 and PAR were not associated to chromatin but localized to the nucleoplasm. In contrast to oxidative stress conditions, our pre-extraction experiments suggest that histones are very unlikely to be modified by MyoD-dependent ARTD1. Auto-ADPR of ARTD1 changes its physicochemical properties and chromatin affinity (Muthurajan et al., 2014). Thus, the present study suggests that the MyoD-dependent ARTD1 pool that is modified and localizes to the nucleoplasm is unlikely regulating any of the canonical, chromatin-linked functions (e.g. DNA repair or transcriptional regulation). This evidence rather suggests a so far undescribed function of PAR in the nucleoplasm. As HPF1 regulates ARTD1 specificity for histones as target proteins (Gibbs-Seymour et al., 2016), the lack of HPF1 or the expression of other cofactors might regulate the shift toward ARTD1 automodification and/or modification of other nucleoplasmic proteins. To investigate the ADP-ribosylome of IMR90 cells upon transdifferentiation, a special MS protocol is applied (Martello et al., 2016), which requires a large amount of starting material. IMR90 cells grow very slowly and do undergo senescence very quickly, thus not allowing to further investigate the ADP-ribosylome of transdifferentiated IMR90. In the nucleoplasm, PAR might recruit and regulate the local diffusion of nuclear proteins during transdifferentiation. Comparably, PAR mediates the shuttling of TARG between the nucleoli and the nucleoplasm after DNA damage (Butepage et al., 2018). Furthermore, owing to its negative charge, PAR behaves as a seeder of liquid-liquid demixing (Altmeyer et al., 2015). Thus, MyoD-dependent PAR foci might locally increase the concentration of PAR-binding proteins, allowing the establishment of membraneless nucleoplasmic environments with liquid-droplet-like behavior.

Transplantation of transdifferentiated cells, starting either with healthy donor cells or with genetically engineered patient fibroblasts, represents a potential therapeutic approach for several genetic diseases (Mollinari et al., 2018). Although, ARTD1-dependent PAR does not seem to contribute to transdifferentiation under the tested conditions, we speculate that it might increase its efficiency. However, single-cell RNA seq would be required to better investigate this hypothesis. Thus, further investigation of the function of ADPR in transdifferentiation might lead to the use of targeting ADPR (e.g. by inhibitors of the potential eraser) as possible therapeutic approach for genetic diseases.

Limitations of the study

The enzymatic activity of ARTD1 was induced during transdifferentiation in a cell cycle and DNA damage independent manner; however, the molecular mechanism behind ARTD1's activation remains elusive.

Moreover, although we observed that ARTD1-dependent PAR was not chromatin associated but localized to the nucleoplasm of transdifferentiated myoblasts, we could not identify the function of nuclear ADPR in the used system. Thus, the role of MyoD-dependent PAR is so far still unknown. Identifying which target proteins are modified by ARTD1 could help dissecting the molecular mechanism behind the increase of PAR after MyoD expression. Unfortunately, the used IMR90 cells proliferate slowly and quickly undergo senescence, thus not allowing to reach the required protein amounts to further investigate the ADP-ribosylome of transdifferentiated IMR90 cells.

It remains to be investigated to which extent the observed ARTD1-induced PAR can be also observed in other transdifferentiation systems.

Resource availability

Lead contact

michael.hottiger@dmmd.uzh.ch.

Materials availability

Plasmids and materials related to transdifferentiation should be requested to P. L. Puri (email address: lpuri@sbpdiscovery.org). All other materials are available from Michael O. Hottiger. This study did not generate any new unique reagent.

Data and code availability

All data is available in the main text or the supplementary information.

The Hi-C data and the RNA-seq data have been deposited in the NCBI database und ID BioProject: PRJNA610985.

Methods

All methods can be found in the accompanying transparent methods supplemental file.

Acknowledgments

We thank Deena M. Leslie Pedrioli and Tobias Suter (University of Zurich) for providing editorial assistance. We thank the Center for Microscopy and Image Analysis (ZMB) and the Functional Genomics Center of the University of Zurich (FGCZ) for services and assistance. We especially thank Catharine Aquino from the FGCZ for helpful suggestions and discussions as well as technical support. PLP lab support from NIH/NIGMS (R01 GM134712-01), entitled “MYOD Regulation of 3D Chromatin Structure”. Research in Beato's lab is supported by funds from the European Research Council under the European Union's Seventh Framework Program (FP7/2007-2013)/ERC Synergy grant agreement 609989 (4DGenome), from the Spanish Ministry of Economy and Competitiveness, ‘Centro de Excelencia Severo Ochoa 2013-2017’ and Plan Nacional (SAF2016-75006-P), as well as support of the CERCA Program/Generalitat de Catalunya. L.B was supported by the Forschungskredit 2019 of the University of Zurich and K.G by a grant of the Oncosuisse (Nr. KFS-3740-08-2015-R). ADP-ribosylation research in the laboratory of MOH is funded by the Kanton of Zurich and the Swiss National Science Foundation, Switzerland (grant 31003A_176177).

Author contributions

Project conceptualization and administration: M.O.H. and LB (lead).

Investigation: L.B. (lead), A-K.H., R.H.W., F.L.D., E.V., A.D., L.C., C.N. (supporting).

Methodology: L.B. (lead for the QIBC) and A-K.H. (supporting for the QIBC); R.H.W. and F.L.D (lead for the Hi-C) and L.B. (supporting); L.B. (lead for transdifferentiation establishment) and A.D. and P.L.P. (supporting).

Data curation and formal analysis: L.B. (lead), A-K.H. (supporting for the QIBC analysis); K.G. (lead for RNA-seq), L.B. (supporting); E.V.(lead for Hi-C), R.H.W. and F.L.D. and (supporting); L.C., C.N. and P.L.P. (lead for the ChIP-seq analysis) and L.B. (supporting).

Writing, review, and editing of MS: L.B. and M.O.H (lead), M.B., P.L.P. (supporting), A-K.H., R.H.W., F.L.D., K.G., A.D., L.C., C.N. (editing).

Declaration of Interests

The authors declare no financial interest.

Published: May 21, 2021

Footnotes

Supplemental information can be found online at https://doi.org/10.1016/j.isci.2021.102432.

Supplemental information

Document S1. Transparent methods, Figures S1–S6, and Tables S1 and S2
mmc1.pdf (2.2MB, pdf)
Table S3. Deferentially regulated genes detected by RNAseq, related to figure 2
mmc2.xlsx (42KB, xlsx)

References

  1. Abplanalp J., Hottiger M.O. Cell fate regulation by chromatin ADP-ribosylation. Semin. Cell Dev. Biol. 2017;63:114–122. doi: 10.1016/j.semcdb.2016.09.010. [DOI] [PubMed] [Google Scholar]
  2. Albini S., Coutinho P., Malecova B., Giordani L., Savchenko A., Forcales S.V., Puri P.L. Epigenetic reprogramming of human embryonic stem cells into skeletal muscle cells and generation of contractile myospheres. Cell Rep. 2013;3:661–670. doi: 10.1016/j.celrep.2013.02.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Alemasova E.E., Lavrik O.I. Poly(ADP-ribosyl)ation by PARP1: reaction mechanism and regulatory proteins. Nucleic Acids Res. 2019;47:3811–3827. doi: 10.1093/nar/gkz120. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Altmeyer M., Neelsen K.J., Teloni F., Pozdnyakova I., Pellegrino S., Grofte M., Rask M.D., Streicher W., Jungmichel S., Nielsen M.L., Lukas J. Liquid demixing of intrinsically disordered proteins is seeded by poly(ADP-ribose) Nat. Commun. 2015;6:8088. doi: 10.1038/ncomms9088. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Andersson A., Bluwstein A., Kumar N., Teloni F., Traenkle J., Baudis M., Altmeyer M., Hottiger M.O. PKCalpha and HMGB1 antagonistically control hydrogen peroxide-induced poly-ADP-ribose formation. Nucleic Acids Res. 2016;44:7630–7645. doi: 10.1093/nar/gkw442. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Andres V., Walsh K. Myogenin expression, cell cycle withdrawal, and phenotypic differentiation are temporally separable events that precede cell fusion upon myogenesis. J. Cell Biol. 1996;132:657–666. doi: 10.1083/jcb.132.4.657. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Azarm K., Smith S. Nuclear PARPs and genome integrity. Genes Dev. 2020;34:285–301. doi: 10.1101/gad.334730.119. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Bartolomei G., Leutert M., Manzo M., Baubec T., Hottiger M.O. Analysis of chromatin ADP-ribosylation at the genome-wide level and at specific loci by ADPr-ChAP. Mol. Cell. 2016;61:474–485. doi: 10.1016/j.molcel.2015.12.025. [DOI] [PubMed] [Google Scholar]
  9. Berkes C.A., Bergstrom D.A., Penn B.H., Seaver K.J., Knoepfler P.S., Tapscott S.J. Pbx marks genes for activation by MyoD indicating a role for a homeodomain protein in establishing myogenic potential. Mol. Cell. 2004;14:465–477. doi: 10.1016/s1097-2765(04)00260-6. [DOI] [PubMed] [Google Scholar]
  10. Butepage M., Preisinger C., von Kriegsheim A., Scheufen A., Lausberg E., Li J., Kappes F., Feederle R., Ernst S., Eckei L. Nucleolar-nucleoplasmic shuttling of TARG1 and its control by DNA damage-induced poly-ADP-ribosylation and by nucleolar transcription. Sci. Rep. 2018;8:6748. doi: 10.1038/s41598-018-25137-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Caretti G., Di Padova M., Micales B., Lyons G.E., Sartorelli V. The Polycomb Ezh2 methyltransferase regulates muscle gene expression and skeletal muscle differentiation. Genes Dev. 2004;18:2627–2638. doi: 10.1101/gad.1241904. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Chen H., Ruiz P.D., Novikov L., Casill A.D., Park J.W., Gamble M.J. MacroH2A1.1 and PARP-1 cooperate to regulate transcription by promoting CBP-mediated H2B acetylation. Nat. Struct. Mol. Biol. 2014;21:981–989. doi: 10.1038/nsmb.2903. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Chiou S.H., Jiang B.H., Yu Y.L., Chou S.J., Tsai P.H., Chang W.C., Chen L.K., Chen L.H., Chien Y., Chiou G.Y. Poly(ADP-ribose) polymerase 1 regulates nuclear reprogramming and promotes iPSC generation without c-Myc. J. Exp. Med. 2013;210:85–98. doi: 10.1084/jem.20121044. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Choi J., Costa M.L., Mermelstein C.S., Chagas C., Holtzer S., Holtzer H. MyoD converts primary dermal fibroblasts, chondroblasts, smooth muscle, and retinal pigmented epithelial cells into striated mononucleated myoblasts and multinucleated myotubes. Proc. Natl. Acad. Sci. U S A. 1990;87:7988–7992. doi: 10.1073/pnas.87.20.7988. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Dall'Agnese A., Caputo L., Nicoletti C., di Iulio J., Schmitt A., Gatto S., Diao Y., Ye Z., Forcato M., Perera R. Transcription factor-directed Re-wiring of chromatin architecture for somatic cell nuclear reprogramming toward trans-differentiation. Mol. Cell. 2019;76:453–472 e8. doi: 10.1016/j.molcel.2019.07.036. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Davis R.L., Weintraub H., Lassar A.B. Expression of a single transfected cDNA converts fibroblasts to myoblasts. Cell. 1987;51:987–1000. doi: 10.1016/0092-8674(87)90585-x. [DOI] [PubMed] [Google Scholar]
  17. de la Serna I.L., Carlson K.A., Imbalzano A.N. Mammalian SWI/SNF complexes promote MyoD-mediated muscle differentiation. Nat. Genet. 2001;27:187–190. doi: 10.1038/84826. [DOI] [PubMed] [Google Scholar]
  18. Erener S., Hesse M., Kostadinova R., Hottiger M.O. Poly(ADP-ribose)polymerase-1 (PARP1) controls adipogenic gene expression and adipocyte function. Mol. Endocrinol. 2012;26:79–86. doi: 10.1210/me.2011-1163. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Farrar D., Rai S., Chernukhin I., Jagodic M., Ito Y., Yammine S., Ohlsson R., Murrell A., Klenova E. Mutational analysis of the poly(ADP-ribosyl)ation sites of the transcription factor CTCF provides an insight into the mechanism of its regulation by poly(ADP-ribosyl)ation. Mol. Cell Biol. 2010;30:1199–1216. doi: 10.1128/MCB.00827-09. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Gibbs-Seymour I., Fontana P., Rack J.G.M., Ahel I. HPF1/C4orf27 is a PARP-1-interacting protein that regulates PARP-1 ADP-ribosylation activity. Mol. Cell. 2016;62:432–442. doi: 10.1016/j.molcel.2016.03.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Gibson B.A., Kraus W.L. New insights into the molecular and cellular functions of poly(ADP-ribose) and PARPs. Nat. Rev. Mol. Cell Biol. 2012;13:411–424. doi: 10.1038/nrm3376. [DOI] [PubMed] [Google Scholar]
  22. Gulati G.S., Sikandar S.S., Wesche D.J., Manjunath A., Bharadwaj A., Berger M.J., Ilagan F., Kuo A.H., Hsieh R.W., Cai S. Single-cell transcriptional diversity is a hallmark of developmental potential. Science. 2020;367:405–411. doi: 10.1126/science.aax0249. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Hanzlikova H., Kalasova I., Demin A.A., Pennicott L.E., Cihlarova Z., Caldecott K.W. The importance of poly(ADP-ribose) polymerase as a sensor of unligated okazaki fragments during DNA replication. Mol. Cell. 2018;71:319–331 e3. doi: 10.1016/j.molcel.2018.06.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Hottiger M.O. Nuclear ADP-ribosylation and its role in chromatin plasticity, cell differentiation, and epigenetics. Annu. Rev. Biochem. 2015;84:227–263. doi: 10.1146/annurev-biochem-060614-034506. [DOI] [PubMed] [Google Scholar]
  25. Hottiger M.O., Hassa P.O., Luscher B., Schuler H., Koch-Nolte F. Toward a unified nomenclature for mammalian ADP-ribosyltransferases. Trends Biochem. Sci. 2010;35:208–219. doi: 10.1016/j.tibs.2009.12.003. [DOI] [PubMed] [Google Scholar]
  26. Ieda M., Fu J.D., Delgado-Olguin P., Vedantham V., Hayashi Y., Bruneau B.G., Srivastava D. Direct reprogramming of fibroblasts into functional cardiomyocytes by defined factors. Cell. 2010;142:375–386. doi: 10.1016/j.cell.2010.07.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Kim D.S., Camacho C.V., Nagari A., Malladi V.S., Challa S., Kraus W.L. Activation of PARP-1 by snoRNAs controls ribosome biogenesis and cell growth via the RNA helicase DDX21. Mol. Cell. 2019;75:1270–1285 e14. doi: 10.1016/j.molcel.2019.06.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Kraus W.L. PARPs and ADP-ribosylation: 60 years on. Genes Dev. 2020;34:251–253. doi: 10.1101/gad.336420.120. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Krukenberg K.A., Jiang R., Steen J.A., Mitchison T.J. Basal activity of a PARP1-NuA4 complex varies dramatically across cancer cell lines. Cell Rep. 2014;8:1808–1818. doi: 10.1016/j.celrep.2014.08.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Luo X., Ryu K.W., Kim D.S., Nandu T., Medina C.J., Gupte R., Gibson B.A., Soccio R.E., Yu Y., Gupta R.K., Kraus W.L. PARP-1 controls the adipogenic transcriptional program by PARylating C/EBPbeta and modulating its transcriptional activity. Mol. Cell. 2017;65:260–271. doi: 10.1016/j.molcel.2016.11.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Luscher B., Butepage M., Eckei L., Krieg S., Verheugd P., Shilton B.H. ADP-ribosylation, a multifaceted posttranslational modification involved in the control of cell physiology in health and disease. Chem. Rev. 2018;118:1092–1136. doi: 10.1021/acs.chemrev.7b00122. [DOI] [PubMed] [Google Scholar]
  32. Martello R., Leutert M., Jungmichel S., Bilan V., Larsen S.C., Young C., Hottiger M.O., Nielsen M.L. Proteome-wide identification of the endogenous ADP-ribosylome of mammalian cells and tissue. Nat. Commun. 2016;7:12917. doi: 10.1038/ncomms12917. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Maves L., Waskiewicz A.J., Paul B., Cao Y., Tyler A., Moens C.B., Tapscott S.J. Pbx homeodomain proteins direct Myod activity to promote fast-muscle differentiation. Development. 2007;134:3371–3382. doi: 10.1242/dev.003905. [DOI] [PubMed] [Google Scholar]
  34. Maya-Mendoza A., Moudry P., Merchut-Maya J.M., Lee M., Strauss R., Bartek J. High speed of fork progression induces DNA replication stress and genomic instability. Nature. 2018;559:279–284. doi: 10.1038/s41586-018-0261-5. [DOI] [PubMed] [Google Scholar]
  35. Michelena J., Lezaja A., Teloni F., Schmid T., Imhof R., Altmeyer M. Analysis of PARP inhibitor toxicity by multidimensional fluorescence microscopy reveals mechanisms of sensitivity and resistance. Nat. Commun. 2018;9:2678. doi: 10.1038/s41467-018-05031-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Mills J.C., Stanger B.Z., Sander M. Nomenclature for cellular plasticity: are the terms as plastic as the cells themselves? EMBO J. 2019;38:e103148. doi: 10.15252/embj.2019103148. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Mollinari C., Zhao J., Lupacchini L., Garaci E., Merlo D., Pei G. Transdifferentiation: a new promise for neurodegenerative diseases. Cell Death Dis. 2018;9:830. doi: 10.1038/s41419-018-0891-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Muthurajan U.M., Hepler M.R., Hieb A.R., Clark N.J., Kramer M., Yao T., Luger K. Automodification switches PARP-1 function from chromatin architectural protein to histone chaperone. Proc. Natl. Acad. Sci. U S A. 2014;111:12752–12757. doi: 10.1073/pnas.1405005111. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Olah G., Szczesny B., Brunyanszki A., Lopez-Garcia I.A., Gero D., Radak Z., Szabo C. Differentiation-associated downregulation of poly(ADP-ribose) polymerase-1 expression in myoblasts serves to increase their resistance to oxidative stress. PLoS One. 2015;10:e0134227. doi: 10.1371/journal.pone.0134227. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Patel T., Tursun B., Rahe D.P., Hobert O. Removal of Polycomb repressive complex 2 makes C. elegans germ cells susceptible to direct conversion into specific somatic cell types. Cell Rep. 2012;2:1178–1186. doi: 10.1016/j.celrep.2012.09.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Piao L., Kang D., Suzuki T., Masuda A., Dohmae N., Nakamura Y., Hamamoto R. The histone methyltransferase SMYD2 methylates PARP1 and promotes poly(ADP-ribosyl)ation activity in cancer cells. Neoplasia. 2014;16:257–264, 64 e2. doi: 10.1016/j.neo.2014.03.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Pietrzak J., Spickett C.M., Ploszaj T., Virag L., Robaszkiewicz A. PARP1 promoter links cell cycle progression with adaptation to oxidative environment. Redox Biol. 2018;18:1–5. doi: 10.1016/j.redox.2018.05.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Rajagopal J., Stanger B.Z. Plasticity in the adult: how should the waddington diagram Be applied to regenerating tissues? Dev. Cell. 2016;36:133–137. doi: 10.1016/j.devcel.2015.12.021. [DOI] [PubMed] [Google Scholar]
  44. Ryu K.W., Nandu T., Kim J., Challa S., DeBerardinis R.J., Kraus W.L. Metabolic Regulation of Transcription through Compartmentalized NAD(+) Biosynthesis. Science. 2018;360:eaan5780. doi: 10.1126/science.aan5780. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Sartorelli V., Puri P.L. Shaping gene expression by landscaping chromatin architecture: lessons from a master. Mol. Cell. 2018;71:375–388. doi: 10.1016/j.molcel.2018.04.025. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Skerjanc I.S., Slack R.S., McBurney M.W. Cellular aggregation enhances MyoD-directed skeletal myogenesis in embryonal carcinoma cells. Mol. Cell Biol. 1994;14:8451–8459. doi: 10.1128/mcb.14.12.8451. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Szanto M., Bai P. The role of ADP-ribose metabolism in metabolic regulation, adipose tissue differentiation, and metabolism. Genes Dev. 2020;34:321–340. doi: 10.1101/gad.334284.119. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Tapscott S.J. The circuitry of a master switch: myod and the regulation of skeletal muscle gene transcription. Development. 2005;132:2685–2695. doi: 10.1242/dev.01874. [DOI] [PubMed] [Google Scholar]
  49. Toledo L.I., Altmeyer M., Rask M.B., Lukas C., Larsen D.H., Povlsen L.K., Bekker-Jensen S., Mailand N., Bartek J., Lukas J. ATR prohibits replication catastrophe by preventing global exhaustion of RPA. Cell. 2013;155:1088–1103. doi: 10.1016/j.cell.2013.10.043. [DOI] [PubMed] [Google Scholar]
  50. Weber F.A., Bartolomei G., Hottiger M.O., Cinelli P. Artd1/Parp1 regulates reprogramming by transcriptional regulation of Fgf4 via Sox2 ADP-ribosylation. Stem Cells. 2013;31:2364–2373. doi: 10.1002/stem.1507. [DOI] [PubMed] [Google Scholar]
  51. Weintraub H., Tapscott S.J., Davis R.L., Thayer M.J., Adam M.A., Lassar A.B., Miller A.D. Activation of muscle-specific genes in pigment, nerve, fat, liver, and fibroblast cell lines by forced expression of MyoD. Proc. Natl. Acad. Sci. U S A. 1989;86:5434–5438. doi: 10.1073/pnas.86.14.5434. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Wisnik E., Ploszaj T., Robaszkiewicz A. Downregulation of PARP1 transcription by promoter-associated E2F4-RBL2-HDAC1-BRM complex contributes to repression of pluripotency stem cell factors in human monocytes. Sci. Rep. 2017;7:9483. doi: 10.1038/s41598-017-10307-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Wright R.H., Lioutas A., Le Dily F., Soronellas D., Pohl A., Bonet J., Nacht A.S., Samino S., Font-Mateu J., Vicent G.P. ADP-ribose-derived nuclear ATP synthesis by NUDIX5 is required for chromatin remodeling. Science. 2016;352:1221–1225. doi: 10.1126/science.aad9335. [DOI] [PubMed] [Google Scholar]
  54. Zaniolo Karine, Rufiange Anne, Leclerc Steeve, Desnoyers Serge, Guérin Sylvain. Regulation of the poly(ADP-ribose) polymerase-1 gene expression by the transcription factors Sp1 and Sp3 is under the influence of cell density in primary cultured cells. Biochem. J. 2005;389:423–433. doi: 10.1042/BJ20041718. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Zhao H., Sifakis E.G., Sumida N., Millan-Arino L., Scholz B.A., Svensson J.P., Chen X., Ronnegren A.L., Mallet de Lima C.D., Varnoosfaderani F.S. PARP1- and CTCF-mediated interactions between active and repressed chromatin at the lamina promote oscillating transcription. Mol. Cell. 2015;59:984–997. doi: 10.1016/j.molcel.2015.07.019. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Document S1. Transparent methods, Figures S1–S6, and Tables S1 and S2
mmc1.pdf (2.2MB, pdf)
Table S3. Deferentially regulated genes detected by RNAseq, related to figure 2
mmc2.xlsx (42KB, xlsx)

Data Availability Statement

All data is available in the main text or the supplementary information.

The Hi-C data and the RNA-seq data have been deposited in the NCBI database und ID BioProject: PRJNA610985.

Articles from iScience are provided here courtesy of Elsevier

RESOURCES