TAp63gamma is required for the late stages of myogenesis

Abstract

p53 family members, p63 and p73, play a role in controlling early stage of myogenic differentiation. We demonstrated that TAp63gamma, unlike the other p53 family members, is markedly up-regulated during myogenic differentiation in murine C2C7 cell line. We also found that myotubes formation was inhibited upon TAp63gamma knock-down, as also indicated by atrophyic myotubes and reduction of myoblasts fusion index. Analysis of TAp63gamma-dependend transcripts identified several target genes involved in skeletal muscle contractility energy metabolism, myogenesis and skeletal muscle autocrine signaling. These results indicate that TAp63gamma is a late marker of myogenic differentiation and, by controlling different sub-sets of target genes, it possibly contributes to muscle growth, remodeling, functional differentiation and tissue homeostasis.

Abbreviations

MRFs

myogenic regulatory factors

Rb

retinoblastoma protein

MHC

myosin heavy chain

Introduction

Myogenesis is a highly ordered process involving a cascade of muscle-specific genes expression that is timely coordinated to cell cycle withdrawal and synthesis of muscle contractile proteins. The commitment of cells to the myogenic lineage depends on the expression of myogenic regulatory factors (MRFs), including: MyoD, Myf5, MyoG, Myogenin and MRF4,^1-8 and microRNAs.^9-14 While early myogenesis markers can be also expressed in proliferating cells, late differentiation markers, like myosin heavy chain (MHC), are induced only after cell cycle arrest.¹⁵ The retinoblastoma protein (Rb) and cyclin D, plays a critical role in myoblasts cycle arrest; indeed myocytes and muscle cells lacking Rb fail to exit the cell cycle,^16-26 altering the late phases of skeletal myogenesis. In addition, the cyclin-dependent-kinase inhibitors p21^Cip/Waf1 (p21) and p57^Kip2 (p57) are highly expressed during myogenesis, where redundantly control cell cycle arrest during differentiation as shown in vitro and in vivo in knock-out mice models.^27-30 In vitro, myogenic differentiation can be recapitulated by myoblasts serum deprivation. Indeed, at low levels of growth factors, mononucleate myoblasts undergo irreversible cell cycle arrest by transcriptional activation of p57, p21 and pRb dephosphorilation followed by myogenin expression.^31-32 Myogenic differentiation then proceeds by induction of muscle-specific genes expression resulting in fusion of myoblasts in myotubes.^27,28,32

There are evidences showing that p53 and its family members, p63 and p73, play a role in controlling myogenic differentiation and rhabdomyosarcoma development.^18,33-36 Both TP63 and TP73 genes are expressed as multiple protein isoforms due to the presence of alternative promoters and splicing sites.³⁷ The usage of different promoters generates trans-activating (TA) isoforms that contain an N-terminal p53-like transactivation domain, and ΔN isoforms that lack this domain being N-terminal truncated.^38,39 The N-terminal-truncated isoforms, ΔNp63 and ΔNp73, bind DNA but are unable to transactivate the canonical p53 target promoters, acting as dominant negative inhibitors of the transcriptionally active family members.³⁸ Note that the ΔNp63 proteins possess intrinsic transactivation activity due to their ability to use a second, yet not well characterized, C-terminus transactivation domain.⁴⁰ Alternative splicing events occurring at the 3’-end generates proteins displaying different C-termini.^37,41,42 p63 and p73 proteins share more than 60% of amino acids identity within the DNA-binding domain and possibly regulate overlapping set of target genes.^37,43-51 In muscle differentiation, the functions of the individual members are distinct but complementary: while p53 transactivates RB gene, p63 and p73 induce the transcription of p57, which maintains Rb in an active, hypo-phosphorylated state,¹⁸ also allowing execution of the later steps in skeletal myogenesis.¹⁶ So far, the engagement of p53 family members in myogenesis has been demonstrated only in the early stage of in vitro muscle differentiation linked to permanent cell cycle exit.

Here, we report that, among the p53 family members, TAp63gamma is specifically upregulated during in vitro C2C7 myogenesis. The data obtained indicate that TAp63gamma has an important role in late differentiation stages. Silencing of TAp63gamma does not significantly alter the expression of early differentiation markers including MRFs, but it causes formation of atrophyc myotubes and reduced myoblasts fusion index. Analysis of TAp63gamma target genes by RT² Profiler PCR Arrays in stable-transfected C2C7 si-TAp63gamma clones, indicated that this transcription factor controls the expression of sub-sets of target genes involved in growth, myoblasts fusion, cell metabolism, muscle remodeling, muscle contractility, therefore having potentially an important role in functional skeleton muscle differentiation.

Results and Discussion

TAp63gamma is expressed in C2C7 mouse myoblast cells induced to differentiate in vitro

To investigate the expression levels of p53 and p53 family members during myogenic differentiation, we used as model the C2C7 myoblastic cell line induced to differentiate by lowering the serum in the culture medium to 2% (differentiation medium). Total RNAs were extracted from these cells at different time-points and RT-PCRs to detect p53, TAp73, ΔNp73, TAp63 and ΔNp63 were performed. The data obtained indicated that, while p53, TAp73, ΔNp73 and ΔNp63 expression level were down-regulated or unchanged during C2C7 differentiation (Fig. 1A), TAp63 isoform expression was markedly up-regulated already after 24h (16 fold) and reach 1535 fold increase at 72 h as compared to undifferentiated myoblasts (Fig. 1B).

Figure 1.

C2C7 mouse myoblast cells induced to differentiate express TAp63gamma. (A) RT-qPCR quantification of p53 family members RNA extracts from C2C7 myoblast grown in differentiation condition for 24, 48 and 72 hours. (B) mRNA expression of TAp63 isoform highlights its up regulation during skeletal muscle differentiation. (C) Western blot analysis of proteins extracts from C2C7 cells. A TAp63 isoform specific antibody was used to detect TAp63 isoform. Protein extracts from SAOS transfected cells with TAp63alpha and TAp63gamma plasmids were used as positive controls. Beta-actin was used as loading control. (D) Immunostaining for p63 is in green color while in blue is the DAPI staining. All the images are presented as merge of both the channels. One representative experiment of 3 is shown. (E) RT-qPCR and (F) protein gel blot analysis of myogenic regulator factor. Error bars in RT-qPCR indicate the SD of triplicate experiments.

Western blot and immunofluorescence analysis further support the evidence that TAp63gamma isoform (Nekulova et al., 2013) is activated during myogenic C2C7 differentiation (Fig. 1C), it is expressed in the nucleus of differentiating myoblasts (Fig. 1D, panel 48h), and accumulated in multinucleated myotubes (Fig. 1D, panel 72h) (Fig. 1D). Myogenic markers, MyoD, Myf5 and MyoG are shown as controls both at mRNA (Fig. 1E) and protein (Fig. 1F) levels. The data shown indicate that the isoform TAp63gamma is expressed in myoblasts nuclei and its expression increases and accumulate in the cells during in vitro muscle differentiation, as late differentiation marker, following myogenin expression and Myf5 down-regulation.

TAp63gamma expression is important for myotube formation

To study the physiological role of TAp63 in muscle differentiation, we performed transient knocking-down experiments of TAp63 transcript by siRNA, followed by 72 h induction of differentiation. TAp63 mRNA was efficiently reduced by siRNA as shown by real-time PCR analysis and by western blot performed at 72 h in differentiation medium (Fig. 2A-B). To examine the effects of TAp63 deficiency on the late stage of myogenesis, control and si-TAp63 myoblasts were plated and allowed to differentiate for 72 h. Co-staining for MHC (red) and DAPI (blue) revealed that the fusion index was significantly (p < 0.005) reduced from 55% (control) to 28% (si-TAp63) (Fig. 2C and D) and the size of the MHC-positive myotubes was significantly reduced in si-TAp63 as compared to control (scramble transfected) cells (Fig. 2C and D).

Figure 2.

Knock-down of TAp63gamma affects late stages of myogenic differentiation in C2C7. (A) RT-qPCR and (B) western blot analysis of C2C7 cells confirm p63 transient silencing after 72 h of differentiation. (C) C2C7 cells scramble transfected (Ctrl) or si-TAp63 (si-p63) were seeded and induced to differentiate for 72 h in differentiate medium. Cells were then fixed and immunostained for MHC (red) to visualize myotubes and counterstained with DAPI (blue) to visualize nuclei. One representative experiment of 3 is shown. (D) Quantification of fusion index. The fusion index (percent differentiation) was determined by dividing the number of nuclei in myotubes-positive by the total number of nuclei in a given microscopic field. Error bars represent the SEM of three independent experiments. Asterisks denote significance (*, P < 0.005). (E) Actin filaments evidentiated by phalloidin-stained Ctrl C2C7 cells versus si-p63. (F) Western blot analysis of proliferation and differentiation markers in Ctrl C2C7 cells (Ctrl-scramble) and si-p63 induced to differentiate for 24, 48 and 72 hours.

In addition, immunostaining with anti-actin antibody during myogenic differentiation showed that the si-p63 myoblasts remain spread in the cell culture plate while the control myoblasts became aligned and organized their cytoskeleton to form myotubes (Fig. 2E). Western blot analysis of extracts obtained by control myoblasts (Ctrl-scramble) and transient transfected si-TAp63 myoblasts induced to differentiate for 24, 48 and 72 h, indicated that reduction of TAp63 does not alter myoblasts cell cycle exit significantly, as evidentiated by unchanged p21 and CyclinA expression and by the weak reduction of p57 expression in si-TAp63 cells. MyoD and myogenin expression levels resulted also unchanged (Fig. 2F), while silencing of TAp63 negatively affects the expression of MHC, indicating a delay of late muscle differentiation. Interestingly, TAp73 and p53 mRNAs are markedly and significantly upregulated upon si-TAp63. In particular, TAp73 mRNA reaches 11 fold over control at 24 h, and goes down to 5.8 and 4.7 fold at 48 and 72 h, respectively, while p53 transcription increases weakly (Figure S1A-E), suggesting that TAp73 and p53, in absence of TAp63, could maintain muscle cells in a early phase of differentiation. All together these data demonstrate that lack of TAp63 does not affect cell cycle exit and early stages of myogenesis but late differentiation stages.

Identification of muscle-specific TAp63gamma down-stream genes

To investigate the genes and the underlying molecular mechanisms that might be involved in the alteration of myoblasts differentiation following TAp63 knock-down, we performed a mouse “Myogenesis and Myopathy” PCR-Array. This array includes genes important for basic skeletal muscle function, development and growth, as well as genes related to the disease processes of metabolic syndrome and muscle wasting. To perform this experiment we generated, by shRNA lentiviral infection, stable si-TAp63 C2C7 clones. We obtained several clones in which TAp63 expression was efficiently inhibited (see clone A as example, Figure S2A and B). Clone A was also characterized, and, as described for transient transfected cells (Fig. 2), expression of p21 and p57 was reduced (Figure S2B and C) at mRNA and protein level. Myoblasts (control and clone A) were induced to differentiate for 48 h and mRNAs were extracted for further analysis. By RT-qPCR based microarray we simultaneously investigated the expression of 84 genes related to myogenesis and myopathy: a scatter plot of the results obtained is shown in Figure S2D. Among the genes differentially expressed, we selected 32 genes that changed by at least two folds in si-TAp63 myoblasts as compared to control cells (Figure S2D, Table 1). In particular, 6 out of 84 genes (7%) showed higher levels of mRNA in si-TAp63 cells, whereas 26 out of 84 genes (31%) showed lower levels of mRNA upon TAp63gamma silencing. The latter are genes whose expression is positively regulated, directly or indirectly, by TAp63. Among the positively regulated ones, 11 genes (42%) are modulated from 2.0 to 4.0 fold, as compared to control samples and 15 genes (58%) are modulated at higher levels (>4 fold). Of these 15 genes, 11 ones (73%) are mainly in skeletal muscle contractility category, including: dystrophin-glycoprotein complex (Cav3); titin complex (Myh1, Myh2, Neb, Tnnt2, Ttn); energy metabolism (Glut4/Slc2a4, HK2); fast-twitch fibers (Atp2a1, Myh1, Myh2); slow-twitch fibers (Mb, Myh1, Tnnc1) (Table 1). Exception is for Igf2 Igfbp5 and Prkag3 that are in other categories (Table 1). Among the 6 genes negatively regulated by TAp63, 2 genes are highy modulated: Mmp9 and Pparg, both in the skeletal muscle wasting and atrophy category (Table 1).

Table 1.

Genes expression modulated upon 48 h si-TAp63

Symbol	Gene name	Fold regulation	Functional gene groups
Genes down-regulated
Actn3	Actinin alpha 3	−2.6528	SMContractility
Adrb2	Adrenergic recept, beta 2	−2.8023	Hypertrophy
Atp2a1	ATPase, Ca++ transporting, cardiac muscle, fast twitch 1	−6.6995	SMContractility
Cav1	Caveolin isoform 1	−2.1306	Myogenesis
Cav3	Caveolin isoform 3	−12.3456	SMContractility
Cryab	Crystallin, alpha B	−3.2976	SMContractility
Cs	Citrate synthase	−2.1938	SMContractility
Dag1	Dystroglycan 1	−2.0579	SMContractility
Dmpk	Dystrophia myotonica-protein kinase	−2.2612	SMContractility
HK2	Hexokinase II	−4,0588	Energy metabolism
Igf2	Insulin-like growth factor 2	−4.1584	Autocrine signaling
Igfbp5	Insulin-like growth factor binding protein 5	−4.3204	Myogenesis Hypertrophy
Il6	Interleukin 6	−2.8418	SMContractility
Mb	Myoglobin	−6.2332	SMContractility
Myh1	Myosin, heavy polypeptide 1	−21.6256	SMContractility
Myh2	Myosin, heavy polypeptide 2	−12.0711	SMContractility
Neb	Nebulin	−8.5486	SMContractility
Ppargc1b	Peroxisome proliferative activated receptor, gamma, coactivator 1 beta	−2.6188	Wasting Atrophy
Prkag3	Protein kinase, AMP-activated, gamma 3 non-catatlytic subunit	−7.271	Metabolic syndrome
Sgca	Sarcoglycan, alpha	−2.0521	SMContractility
Slc2a4	Solute carrier family 2	−5.2538	Metabolic syndrome
Tnnc1	Troponin C, cardiac/slow skeletal	−6.5209	SMContractility
Tnni2	Troponin I, skeletal, fast 2	−4.7124	SMContractility
Tnnt3	Troponin T3, skeletal, fast	−2.6733	SMContractility
Ttn	Titin	−8.0881	SMContractility
Pax7	Pax7	−11	Skeletal myogenesis
Genes up-regulated
Capn3	Calpain 3	2.4538	SMContractility
Des	Desmin	4.8628	SMContractility
Mmp9	Matrix metallopeptidase 9	12.1877	Wasting Atrophy
Nos2	Nitric oxide synthase 2, inducible	2.2585	Wasting Atrophy
Pparg	Peroxisome proliferator activated receptor gamma	2.1007	Metabolic syndrome
Ppargc1a	Peroxisome proliferative activated receptor, gamma, coactivator 1 alpha	2.0689	Metabolic syndrome

To note that the genes down-regulatated upon siTAp63 are the gene POSITIVELY regulated by TAp63 (directly or indirectly); while the genes up-regulated upon so-TAp63 are the one NEGATIVELY regulated by TAp63 (directly or indirectly.

Functional Gene grouping, as identified in RT² Profiler PCR Array:

1- SMContractility: Skeletal Muscle Contractility.

2- Myogenesis: Skeletal Myogenesis.

3- Hypertrophy: Skeletal Muscle Hypertrophy.

4- Autocrine signaling: Skeletal Muscle Autocrine signaling.

5- Metabolic syndrome: Diabetes/Metabolic syndrome.

6- Wasting, Atrophy: Skeletal Muscle Wasting/Atrophy.

The RT-qPCR based microarray results were validated performing qRT-PCR on total mRNAs extracted by C2C7-control and C2C7-cloneA cells induced to differentiate for 48 h. We have screened 10 genes of the 26 genes downregulated upon si-TAp63 and 2 of the 6 genes up-modulated upon siTAp63. The data obtained confirmed the results of the RT-qPCR based micro array (Fig. 3).

Figure 3.

Validation of TAp63gamma down-stream genes. mRNA extracted by control (Ctrl-scamble stable clone, Fig. S2A) and stable silenced-p63 (Clone A, Fig. S2A) cells, grown for 48 h in differentiation medium, were analyzed by RT-qPCR. (A) RT-qPCR showing p63 knockdown. (B) RT-qPCR to detect Pax7, “skeletal myogenesis” category. (C-H) RT-qPCR of genes involved in “skeletal muscle contractility” category. (I) RT-qPCR for Igf2, “autocrine signaling” category. (J) RT-qPCR for HK2, “energy metabolism” category. (K-M) RT-qPCR for Prkag3 and Pparg, “metabolic syndrome” category. (L) RT-qPCR for Mmp9, “wasting atrophy” category. The categories indicated are the one described in Table 1. Data are shown as the mean of 3 experiments +/− standard deviation.

Our results indicated that during myogenic differentiation TAp63gamma regulates genes, directly or indirectly, involved in the skeleton muscle contractility and energy metabolism, is conceivable, therefore, to include TAp63gamma has an important transcription factor for functional skeleton muscle differentiation.

Conclusion

Previous studies have shown the involvement of p53 and its family members, p63 and p73, in activation of RB allowing the permanent cell cycle exit and activation of muscle-specific genes. Whereas p53 controls Rb expression levels, p63 and p73 regulates RB phosphorylation state by inducing p57^Kip2 transcription.¹⁸ Here, using C2C7 as experimental model, we define the role of p53 family member isoforms involved in skeletal muscle differentiation, focusing on the role of TAp63gamma. In contrast to the other isoforms, TAp63gamma is strongly up-regulated during myoblasts differentiation (Fig. 1). Knock-down of TAp63 does not significantly affect skeletal muscle differentiation in terms of cell cycle exit and myogenic markers expression, (MyoD, Myf5 and MyoG; Fig. 2), but alters myotubes fusion and myotubes size (Fig. 2). Interestingly, we observed that, upon si-TAp63, TAp73 mRNA increses significantly, suggesting a compensatory mechanisms toward induction of p57^Kip2 allowing cell cycle exit. Analysis of the putative target genes by RT-qPCR array suggests that TAp63gamma controls, directly or indirectly, genes involved in skeletal muscle contractility, including: dystrophin-glycoprotein complex, titin complex, energy metabolism, fast-twitch fibers, slow-twitch fibers (Table 1, Fig. 3). Our data strongly indicate that TAp63gamma is a late differentiation marker, and by controlling different sub-sets of target genes, it contributes to muscle growth, remodeling, functional differentiation and tissue homeostasis.

Material and Methods

Cell lines, trasfections and lentiviral infection

Mouse C2C7 cell line, a subclone of C2 myoblasts (Yaffe and Saxel, 1977), were grown to confluency under 5% CO₂ at 37°C in D-MEM medium supplemented with 20% (v/v) fetal bovine serum (FBS) and penicillin-streptomycin (100 U/ml). Cells were then switched to differentiation medium (DMEM containing 2% horse serum). Maximal formation of myotubes was usually observed 72 h later. For p63 silencing by siRNA, C2C7 were transfected with FlexiTube GeneSolution (GS22061, QIAGEN). Allstars Negative Control siRNA (1027280, QIAGEN) was used as negative control. Trasfection was performed using Lipofectamine RNAiMax (Invitrogen) according to the manufacturer's protocol. Stable p63 silencing was achieved infecting C2C7 cells with Thermo Scientific SMART Choice Lentiviral shRNA SH01–040654–01 and mCMV-TurboGFP-NTC particles (Dharmacon) at m.o.i. 50. Lentiviral particles were diluted in D-MEM 2% FBS, infection medium was removed after 16 h and replaced with proliferation medium. 48 h after infection, stable transduced clones were selected adding puromycin 1 ug/ml in growth medium and 2 weeks later single clones were picked and grown up for further screening.

p63 shRNA pools sequences used in C2C7 trasfection:

oligo 1; 5′-CCCAGTATGTAGAAGATCCTA-3′,

oligo 2; 5′-CAGCTGCGTCGGAGGAATGAA-3′,

oligo 3: 5′-GACGTACGAGATGTTGCTGAA-3′,

oligo 4: 5′-CCCAGTCATCTGATTCGAGTA-3′,

p63 shRNA sequence used in c2c7 infection: 5′-TTGGACGTCTGTTCATTCC-3′.

RNA extraction, quantitative real-time RT-PCR and array

RNA was extracted from cells by using the RNAeasy MiniKit (Qiagen). A total of 1 μg of RNA was used for reverse transcription using the GoScript Kit (Promega), and 2 μl of the reaction were used for real-time PCR. Normalization was performed amplifying murine GAPDH housekeeping mRNA. All the primers used are listed in Table S1.

Real-time PCR was perfomed using an Applied Biosystems 7500 real-time PCR system. The RT² Profiler PCR Arrays was obtained from SA-Biosciences (PAMM-099Z). These 96-well array plates contained primers for 84 key genes involved in skeletal muscle differentiation, function and disease-related processes, as well as the appropriate control genes. 48 h after differentiation induction, RNA from stable si-p63 clone A and scramble clone (control) was isolated using the GoScript Kit (Promega) according to the manufacturer's protocol and then converted into first strand cDNA with the RT² First Stand Kit. cDNAs of both the clones were then analyzed on their respective RT²-PCR plates. RT-PCR experiments were performed on a ABI 7500 instrument (Applied Biosystems) and analyzed according to manufacturer instructions.

Western blotting

Cells were suspended in RIPA buffer (0.5% Nonidet P-40, 0.5% sodium de oxycolate, 0.1% SDS) including the protease inhibitor cocktail (Sigma-Aldrich). After homogenization through Qiashredder columns (Qiagen) and protein determination, 40 μg of total proteins were loaded onto 10%SDS-PAGE and transferred to PVDF membrane branes (GE Healthcare). The blots were blocked with TBST (Tris-buffered saline and Tween 20) and 10% nonfat dry milk (Biorad), and then incubated with the specific primary antibody: anti-p63⁵² anti-p57 (sc-8298, Santa Cruz), anti MyoD (sc-304, Santa Cruz), anti-Myogenin (sc-576, Santa Cruz), anti-Myf5 (sc-302, Santa Cruz), anti Myosin Fast ( M4276, Sigma), anti-p21 (sc-6246, Santa Cruz). Actin (A5441 mouse monoclonal antibody, Sigma-Aldrich) was used as a loading control). After washings and incubation with the appropriate horseradish peroxidase conjugated secondary antibodies (Biorad), signal detection was performed with the Western Lightning Plus ECL (Perk in Elmer).

Immunofluorescence

Briefly, cells were fixed in 4% paraformaldehyde in PBS for 15 min, washed once with PBS and permeabilized with 0.25% Triton-X-100 in PBS for 5 min. Cells were blocked with 5% goat serum in PBS for 1 h and then exposed to primary antibodies. Cells were tested for MHC (1/100; MY32, Sigma), Alexa fluor Phalloidin 568 (A12380, Invitrogen) and p63 (1/100; sc8343, Santa Cruz) for 2 hours at room temperature. Cells were washed 3 times with PBS and then treated with a 488- or 568-Alexa Fluor secondary antibodies (1/1000 dilution; Invitrogen) and DAPI. After three washes in 1X PBS, the slides were mounted using the Prolong Antifade kit (Invitrogen). Slides were analyzed with a confocal laser microscope (NIKON Eclipse Ti). Detection of the signal was performed using EZ C.1 software (Nikon). Fusion Index (FI) was calculated as the number of nuclei present in myotubes over the total number of nuclei present in the observed field. Data was selected from 4 different and randomly chosen microscopic fields and from 3 independent experiments.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Acknowledgments

We thank Dr Marco Ranalli for technical assistance in confocal microscopy.

Funding

This work has been fully supported by AIRC grant (IG13387) to ED, “Bando Giovani Ricercatori 2008” (GR-2008-1140286, c53), and “Ricerca Corrente 1.1” IDI-IRCCS, Health Ministry. Partially supported by Telethon Grant GGP13013 and ASI to AM and GACR P206/12/G151 to BV.

Supplemental Material

Supplemental data for this article can be accessed on the publisher's website.