E XP ER I ME NTAL C E LL RE S E ARCH
331 (2 015 ) 20 0 – 210
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/yexcr
Research article
Brain and muscle Arnt-like 1 promotes skeletal muscle regeneration through satellite cell expansion Somik Chatterjeea, Hongshan Yina,b, Deokhwa Nama, Yong Lic, Ke Maa,n a
Center for Diabetes Research, Department of Medicine, Houston Methodist Research Institute, Houston, TX 77030, USA Department of Cardiovascular Medicine, Third Affiliated Hospital, Hebei Medical University, Shijiazhuang 050051, Hebei, China c Department of Pediatric Surgery, Center for Stem Cell Research and Regenerative Medicine, University of Texas Health Science Center at Houston, Houston, TX 77030, USA b
article information
abstract
Article Chronology:
Circadian clock is an evolutionarily conserved timing mechanism governing diverse biological
Received 29 July 2014
processes and the skeletal muscle possesses intrinsic functional clocks. Interestingly, although the
Received in revised form
essential clock transcription activator, Brain and muscle Arnt-like 1 (Bmal1), participates in
27 August 2014
maintenance of muscle mass, little is known regarding its role in muscle growth and repair. In
Accepted 28 August 2014
this report, we investigate the in vivo function of Bmal1 in skeletal muscle regeneration using two
Available online 9 September 2014
muscle injury models. Bmal1 is highly up-regulated by cardiotoxin injury, and its genetic ablation
Keywords: Muscle regeneration Circadian rhythm Satellite cell Proliferation
significantly impairs regeneration with markedly suppressed new myofiber formation and attenuated myogenic induction. A similarly defective regenerative response is observed in Bmal1-null mice as compared to wild-type controls upon freeze injury. Lack of satellite cell expansion accounts for the regeneration defect, as Bmal1 / mice display significantly lower satellite cell number with nearly abolished induction of the satellite cell marker, Pax7. Furthermore, satellite cell-derived primary myoblasts devoid of Bmal1 display reduced growth and proliferation ex vivo. Collectively, our results demonstrate, for the first time, that Bmal1 is an integral component of the pro-myogenic response that is required for muscle repair. This mechanism may underlie its role in preserving adult muscle mass and could be targeted therapeutically to prevent muscle-wasting diseases. & 2014 Elsevier Inc. All rights reserved.
Introduction The circadian clock is a universal timing mechanism that generates the 24 h rhythm. Nearly all tissues and cell types possess
intrinsic molecular clocks, central and peripheral clocks, that are driven by an interlocked feed-back network of transcriptional regulators coupled with translational and post-translational control [1,2]. This temporal regulatory mechanism governs a wide
Abbreviations: Bmal1, brain and muscle Arnt-like 1; CTX, Cardiotoxin; CME, crushed muscle extract; CLOCK, circadian locomotor output cycles kaput; FGF2, fibroblast growth factor 2; Wnt, wingless-type MMTV integration site family; eMyHC, embryonic myosin heavy chain; TA, tibialis anterior n
Corresponding author. Fax: þ1 713 793 7162. E-mail address: [email protected] (K. Ma).
http://dx.doi.org/10.1016/j.yexcr.2014.08.041 0014-4827/& 2014 Elsevier Inc. All rights reserved.
E XP ER I ME NTAL C E LL RE S E ARCH
variety of cyclical biological processes ranging from rhythmic gene expression to behavioral and physiological cycles. The skeletal muscle possesses cell-autonomous functional clocks and a significant 3.4% of the skeletal muscle transcriptome displays circadian oscillation [3]. Notably, circadian clock-controlled processes in muscle encompass a diverse range from contractile performance, cell cycle progression, sarcomere structural integrity to mitochondrial bioenergetics [3–6]. Brain and muscle Arnt-like 1 (Bmal1), a basis helix-loop-helix transcription factor, is an essential activator of the core molecular clock transcription network that is highly expressed in skeletal muscle [7,8]. It dimerizes with circadian locomotor output cycles kaput (CLOCK) to initiate transcription of target promoters through canonical E-box elements [9]. Recent evidence indicates that Bmal1 and CLOCK are involved in maintaining skeletal muscle sarcomere integrity and proper contractile function [6], while Bmal1 is also implicated in preserving muscle mass during aging [4,5]. Inactivation of Bmal1 in mice leads to severe agingassociated sarcopenia that substantially shortens life span [4], and its muscle-specific overexpression rescues survival [5]. However, although these studies suggest that the circadian clock exerts critical regulatory control on skeletal muscle functions, the mechanisms mediating these effects are not known. Thus, despite the fundamental importance of the molecular clock in skeletal muscle physiology, our understanding of how it maintains muscle mass and function is nearly non-existent, and whether the clock network participates in muscle growth and remodeling has not been studied. Intriguingly, the orderly progression of myogenesis, from commitment of mesenchymal progenitors to myoblast lineage to terminal differentiation and fusion to form multinucleated myotubes, is a highly temporally coordinated event that requires the sequential activation of the myogenic regulatory factor cascade [10]. Importantly, a temporal order of Notch activation preceding Wnt activity has been shown to be required for proper muscle regeneration to ensure stem cell activation, proliferation before terminal differentiation [11], while inappropriate activation of either of these myogenic signals leads to adverse outcomes of muscle repair. We recently demonstrated that Bmal1 has a pro-myogenic function in myogenesis through transcriptional control of the canonical Wnt signaling pathway [12], suggesting that indeed a molecular clock regulation is involved in the progression of myogenic events. Given that Wnt is a major inductive signaling during embryonic and regenerative myogenesis [13,14], our findings imply that Bmal1 may participate in these processes. Based on the role of Bmal1 in preserving muscle mass, we postulate that it may positively regulate muscle regeneration in vivo. Therefore, employing established muscle injury models, our current study interrogates the function of Bmal1 in adult muscle repair, an important aspect of physiological remodeling of skeletal muscle due to activity or exercise-induced injuries [15]. Our findings reveal that Bmal1 is an inherent component of the myogenic response that facilitates regenerative myogenesis.
Materials and methods Animals Mice were maintained in the Methodist Hospital Research Institute vivarium under a constant 12:12 light dark cycle, with lights
331 (2 015 ) 2 0 0 – 210
201
on at 7:00 a.m. (ZT0). All animal procedures were conducted in accordance with the Guidelines for the Care and Use of Laboratory Animals and were approved by the IACUC committee of Houston Methodist Research Institute. Bmal1 / mice [7] (B6.129Arntltm1Bra/J) were obtained originally from the Jackson Laboratory and maintained in the laboratory. Mice 8–12 weeks of age were used throughout the study to exclude effects of aging-related pathologies [4].
Muscle injury models For cardiotoxin (CTX) injection in tibialis anterior (TA) muscle, 30 μl of 10 μM solution was injected at three sites along the longitudinal length to ensure uniform injury of the entire muscle as described [16], with PBS injection as controls. Local freeze injury in TA was induced by applying a metal rod that was dipped in liquid nitrogen to exposed muscle surface for 10 s, and repeated three times to ensure depth of the injury, as described [17]. Shamoperated animals were used as controls. Muscle tissues were collected at 3 p.m., the time of the initial CTX injection, to ensure comparable circadian time sampling.
Primary myoblast isolation and culture Primary myoblasts were isolated from hind limb muscle of 4 week-old mice as described previously [18]. Briefly, muscles were minced into small pieces and subjected to 1% collagenase digestion at 37 1C for 30 min with agitation. Digested cells were seeded on collagen-coated plates, and serial preplating was used to deplete fibroblasts consecutively. Subsequent selective growth enriches myoblasts, and cell amplified after 6 passages were used for the study. Purity of myoblasts obtained by this method was confirmed by their uniform differentiation into myosin heavy chain (MHC)-positive myotubes. The cells were maintained in DMEM with 20% FBS on collagen-coated plates. Cell samples were collected under normal culture conditions without synchronization of cellular clocks.
Crushed muscle extract Crushed muscle extract was obtained following the procedure described by Bischoff [19]. Briefly, leg muscles from 8–10 weeks old C57BL/6 mice were excised and pressed gently seven times with a pair of blunt forceps. Muscles were then incubated at 4 1C at a concentration of 1 g muscle/ml with gentle shaking in Tris-buffered saline for 2 h. The supernatant was collected, filtered, and assayed for protein concentration by a DC Protein Assay Kit (Bio-Rad). Pooled batches of muscle extract were stored at 80 1C until use. For primary myoblast treatment, muscle extract was used at a protein concentration of 300 μg/ml in DMEM containing 10% FBS.
Histology, immunofluorescence staining and quantification Muscles were snap frozen in liquid-nitrogen cooled isopentane and embedded in OCT. 10 μm thick cryosections of middle region of the injured TA muscle were obtained. Histological analysis was performed using hematoxylin and eosin staining. For immunofluorescence, cryosections were fixed using 4% paraformaldehyde and permeabilized by 0.2% Triton X-100. Endogenous IgG was
202
EX PE R IM E NTAL C ELL RE S EA RC H
blocked using non-specific mouse IgG and further blocking using 5% goat serum was carried out prior to primary antibody incubation at 4 1C overnight. The antibodies used were Pax7 (1:20), embryonic myosin heavy chain (BF-G6, 1:100) and dystrophin (MANDRA1 7A10, 1:100), Developmental Studies Hybridoma Bank; laminin (sc-6017, 1:500), Santa Cruz Biotechnology. Secondary antibodies were goat anti-mouse and rabbit anti-goat (Thermo Scientific) used at 1:500 for 1 h incubation at room temperature together with DAPI for nuclei staining. Cross-section area of nascent fibers containing centralized nuclei was measured using H/E staining sections of three representative fields (20X) from each muscle section using Nikon NIS-Elements analysis software. The number of Pax7þ/DAPI double-positive cells was determined from three representative fields (40X) and the percentage to total DAPI-stained nuclei was calculated.
Immunoblot analysis 20–50 mg of total protein was resolved on SDS-PAGE gels followed by immunoblotting after nitrocellulose membrane transfer, and developed by chemiluminescence (Supersignal; Pierce Biotechnology), as described [18]. Myf5 (sc-302, 1:1000), MyoD1 (sc-760, 1:1000), and β-actin (sc-1616, 1:3000) antibodies were obtained from Santa Cruz Biotechnology, myogenin antibody (MAB 3879, 1:1000) was from Millipore, and Bmal1 antibody (PAI-523, 1:500) was from Thermo Scientific.
RNA extraction and quantitative reverse-transcriptase PCR analysis RNeasy miniprep kits (Qiagen) were used to isolate total RNA from snap-frozen muscle tissues and cells. cDNA was generated using a q-Script cDNA Supermix kit (Quanta Biosciences) and quantitative PCR was performed using a Roche 480 Light Cycler with Perfecta SYBR Green Supermix (Quanta Biosciences), as described [18]. Relative expression levels were determined using the comparative Ct method to normalize target genes to 36B4 internal control and compared to controls as indicated.
EdU proliferation assay Primary myoblasts seeded on collagen-coated chamber slides at a density of 80,000 per well were labeled with 10 μM 5-ethynyl-20 deoxyuridine (EdU) for 5 h. Fgf2 (25 ng/ml) was added to cells for 24 h. EdU incorporation was detected by a Click-iT Imaging Kit (Invitrogen) with Alexa 488 Fluorophore. DAPI was used to label for nuclei. Total number of EdUþ cells was counted from 4 representative fields at 20X and the rate of proliferation was calculated as percentage of EdUþ/DAPI.
331 (2015 ) 20 0 –210
Statistical analysis Data are presented as mean7SE. One way ANOVA or two-tailed Student's t-test were performed as appropriate as indicated. Statistical analysis was carried out using SigmaPlot. Po0.05 was considered statistically significant.
Results Robust induction of Bmal1 during muscle regeneration Bmal1 is highly enriched in skeletal muscle [8]. Consistent with this notion, examination of Bmal1 protein level in mesodermal cell types reveals that it is more abundant in C2C12 and primary myoblasts than C3H10T1/2 mesenchymal precursors and 3T3-L1 preadipocytes (Fig. 1A), with the highest level detected in C2C12 myoblasts. Based on our previous finding that Bmal1 promotes myogenic differentiation [18], we postulate that it may respond to myogenic stimuli. Remarkably, cardiotoxin (CTX)-induced muscle injury elicits a robust elevation of Bmal1 protein, with a peak expression detected at day 5 following injury that declines but persists till day 15 (Fig. 1B). This persistent elevation at day 15, a regeneration time point with on-going myogenic activity but absence of inflammation, indicates that Bmal1 induction is likely correlated with myogenesis but not simply a part of the initial CTX-induced inflammatory response. Interestingly, Bmal1 induction kinetics during this regeneration time course mimics that of a known myogenic regulatory factor, myogenin, suggesting that it may respond to injury-induced myogenic signals in a similar manner as a bona fide myogenic regulatory factor. Indeed, treatment of primary myoblasts with crushed muscle extract (CME) [19], obtained from physically damaged muscle fibers, rapidly induces Bmal1 expression after 6 h with 2-fold up-regulation at 12 h (Fig. 1C). In addition, the induction of Bmal1 precedes the ability of CME to up-regulate the myogenic terminal differentiation factor, myogenin, and embryonic myosin heavy chain (Myhc3), both occurring only at 12 h. As various agents released upon muscle injury have been shown to possess myogenic stimulatory effects [20], we next determined whether specific factors, including cAMP-inducer, Forskolin, Wnt3a and Fgf2, can elicit the observed Bmal1 response to muscle injury. Surprisingly, all these known myogenic stimuli tested are capable of activating Bmal1 expression, with Fgf2 leading to the highest 2-fold up-regulation (Fig. 1D). On the other hand, both Forskolin and Fgf2 suppressed the expression of the negative regulator of the molecular clock, Rev-erbα [21], suggesting that these factors may exert a coordinated influence on the clock machinery (data not shown). These findings indicate that Bmal1 responds to signals elicited by muscle injury during initiation of regeneration, implicating its involvement in the early myogenic activity in muscle repair.
Quantitative TUNEL assay Loss of Bmal1 impairs muscle regeneration Primary myoblasts were plated on collagen coated 96 well plate at seeding density of 20,000–120,000 cells per well. Cells treated with endonuclease for 1 h to induce apoptosis were used as a positive control. Terminal deoxynucleotidyl transferase (TdT) in situ labeling and detection of apoptosis was performed 24 h after seeding by a colorimetric TACS apoptosis detection kit (Trevigen).
Based on the above findings, we explored a potential pro-myogenic function of Bmal1 in vivo using two different injury models, cardiotoxin injection to the entire length of the tibialis anteria (TA) muscle and freezing injury of the same site [20]. These injuries inflict extensive damage throughout the muscle to elicit uniform regenerative responses. Following the time course of CTX-induced
331 (2 015 ) 2 0 0 – 210
Arbitrary U Unit /TBP
E XP ER I ME NTAL C E LL RE S E ARCH
Bmal1
203
3 2 1 0
TBP
Cardiotoxin Day
0
5
8
15
Bmal1 Myogenin β-actin
Rel. mRNA Exp./Ctr
2.5
**
2
*
1.5
*
**
Ctr
1
CME-6hr
0.5
CME-12hr
0
Bmal1
Myogenin
Myhc3
Rel. mRNA m Exp. /PBS or DMSO
2.5
**
2
** **
** **
*
1.5
2hr 4hr
1 0.5 0
PBS
DMSO
Forskolin
Wnt3a
Fgf2
Fig. 1 – Induction of Bmal1 expression by myogenic stimuli. Immunoblot analysis of Bmal1 protein abundance in mesodermal cell types with quantitation (A), and its response to cardiotoxin-induced muscle injury in wild-type mice (B). RT-qPCR analysis of Bmal1 mRNA expression in primary myoblasts following crude muscle extract treatment (C), or in response to various soluble signals induced by muscle injury (D, n¼ 3). n,nn: po0.05 and 0.01 vs. respective controls ( Forskolin vs. DMSO, Wnt3a and Fgf2 vs. PBS) by Student's t test.
regeneration in normal wild-type (WT) mice, analysis of total protein lysate of regenerating muscle reveals robust induction of myogenic regulatory factors MyoD1 and Myogenin early on at day 5, which declines but remains elevated at higher levels than day 0 expression till day 15 post-injury (Fig. 2A). In contrast, Bmal1 / mice exhibit a blunted initial elevation and a severe lack of sustained induction of MyoD1 and Myogenin, indicating an attenuated myogenic response with shortened duration as compare to the WT. In addition, protein level of desmin, a marker of nascent myofibers, is significantly lower throughout the regeneration, suggesting impaired new fiber formation. Histological analysis of the injured TA further demonstrates significantly attenuated myofiber formation in Bmal1 / mice throughout the regeneration time course (Fig. 2B). There are less nascent myofibers appearing in Bmal1-null than in WT mice at the beginning of regeneration at
day 5. This defect in regeneration persists and becomes more evident at day 8 and 15, with a higher percentage of smaller size fibers containing single central nucleus as compared to WT. As shown by size distribution of these myofibers with varying diameters in Fig. 2C, the measured cross section area of nascent central-nucleated myofibers in Bmal1-null mice is markedly reduced as compared to that of WT. This reduction of caliber of newly formed myofibers is further confirmed by dystrophin immunoflurescence staining that outlines muscle fiber perimeter, as shown in Fig. 2E. Additionally, quantification of myonuclei per myofiber reveals a similarly shifted distribution toward lower numbers in the Bmal1 / mice (Fig. 2D). As newly-regenerated myofibers transiently express the embryonic isoform of myosin heavy chain (eMyHC), day 3-injured muscles were stained with an eMyHC antibody to definitively assess nascent myofiber synthesis. Compare to the strong
204
EX PE R IM E NTAL C ELL RE S EA RC H
WT 5
WT Bmal1-/-
Bmal1-/8
15
0
5
D8
8 15 % of Total
Day 0
331 (2015 ) 20 0 –210
Bmal1 MyoD1
60
**
40 20
D15
**
**
**
**
**
**
0
** μm2
Myogenin Desmin D8
D15
80
% of Total
β-actin
60
**
**
40 20
**
**
0
1
2
≥3
1
2
≥3
WT
Bmal1-/-
Dystrophin /DAPI-D8 WT
Bmal1-/-
Fig. 2 – Impaired muscle regeneration in Bmal1 / mice. (A) Immunoblot analysis of myogenic markers during regeneration time course as indicated in WT and Bmal1 / mice. Protein samples represent pooled mice of 4–5/group. Representative images of H/E staining (B) with quantification of distribution of myofiber cross section area (C), and myonuclei numbers (D) of cardiotoxininduced TA muscle regeneration (n¼4–5). (E) Representative images of dystrophin immunofluorescence staining of day 8 injured muscles to outline the perimeter of muscle fibers. n,nn: po0.05 and 0.01 Bmal1 / vs. WT by Student's t test.
immunostaining of eMyHC in WT, Bmal1 / mice display a strikingly weaker signal, as demonstrated in Fig. 3A. This finding is further validated by the barely induced total protein level of eMyHC detected in these mice at 3 days post-CTX (Fig. 3B). Thus, these results demonstrate that loss of Bmal1 significantly impairs myogenic response and new myofiber formation during muscle regeneration.
Using an additional freeze injury-induced muscle regeneration model, we further tested the pro-myogenic role of Bmal1 in vivo. At 30 days after freeze injury when the repair phase of regeneration is nearly complete, regenerated TA muscle weight (Fig. 4A) as well as its relative ratio to body weight (Fig. 4B) of Bmal1-null mice is significantly lower than that of sham-operated WT controls. This overall deficient regenerative capacity due to loss
E XP ER I ME NTAL C E LL RE S E ARCH
eMyHC
205
331 (2 015 ) 2 0 0 – 210
Laminin
Merge
WT
Bmal1-/-
WT Day
0
Bmal1-/-
3
5
0
3
5
eMyHC β-actin Fig. 3 – Loss of Bmal1 attenuates induction of embryonic myosin heavy chain during cardiotoxin-induced muscle regeneration. (A) Representative images (20X) of eMyHC immunofluorescence staining with co-staining of laminin to outline myofibers at day 3 of regeneration. (B) Immunoblot analysis of embryonic MyHC protein levels during regeneration time course in WT and Bmal1 / mice (n¼4–5).
0.24
TA/BW (%)
TA Weight (g)
0.06 0.04 0.02
Rel.. Exp./ 36B4/WT-D0
0
0.18 0.12 0.06 0
WT
Bmal1-/-
WT
Bmal1-/-
4
40
200
3
30
150
2
20
100
1
10
50
0
0
D0
D5 Myf5
D15
0
D0
D5
D15
Myogenin
D0
D5 Myhc3
D15
0
D0
D5 D15 Mrf4
Fig. 4 – Loss of Bmal1 impairs muscle regeneration and myogenic induction in freeze-injury model. Regenerated TA muscle weight (A), and ratio to total body weight (B) 30 days after freeze injury in WT and Bmal1-null mice (n¼ 4–6/group). (C) RT-qPCR analysis of myogenic gene expression in regenerating TA at indicated time after freeze injury. n¼ 4–6/group. n,nn: po0.05 and 0.01 Bmal1 / vs. WT; #, ##: po0.05 or 0.01 vs. D0 by Student's t test.
of Bmal1 is also evident in the blunted mRNA expression of myogenic regulatory factors, including Myf5, myogenin, Mrf4 as well as the new myofiber marker, Myhc3 (Fig. 4C). The similarly
defective muscle regeneration induced by freeze injury observed in Bmal1-null suggests that Bmal1 may facilitate muscle repair independent of the specific inciting damage to the muscle.
206
EX PE R IM E NTAL C ELL RE S EA RC H
Bmal1 regulation of Wnt signaling in muscle regeneration We have previously demonstrated that Bmal1 exerts transcriptional control on Wnt pathway genes, and this regulation of Wnt signaling promotes myogenesis [18]. As Wnt is a key signal that drives embryonic and regenerative myogenesis [13], we tested whether altered Wnt signaling may have contributed to the regenerative defect of Bmal1-null mice. First, we examined the mRNA expression of genes of the canonical Wnt signaling pathway. CTX-induced regeneration leads to an evident increase in Wnt signaling activity as indicated by the induction of β-catenin, TCF3 and Axin2 transcript levels in WT controls, with a 3–4 fold up-regulated peak level detected at day 5 and remains high at day 15 (Fig. 5A). In contrast, in mice lacking Bmal1, analysis of β-catenin, the ultimate signaling transducer of the canonical Wnt pathway, reveals its attenuated transcript level at day 15 (Fig. 5A), and induction of TCF3, the transcription factor mediating Wnt signal, is also compromised. Notably, expression of Axin2, a canonical Wnt target gene, is significantly lower at both day 5 and 15 in the regenerating Bmal1null muscle, with marked 70% suppression at day 15 as compared to WT. The lower Axin2 expression indicates impaired Wnt activity in Bmal1 / mice, likely a result of collective contribution of attenuated expression of a few key steps along the Wnt signal transduction cascade. Moreover, in comparison to the substantially elevated protein level of β-catenin in the WT mice (Fig. 5B), Bmal1 / mice display a weaker β-catenin level at day 8 post-CTX and its sustained increase observed in the WT at day 15 is completely absent. Based on our findings of impaired regeneration with Bmal1 ablation and the established role of Wnt in regenerative myogenesis, it is possible that the defect in Wnt signaling observed in vivo due to loss of Bmal1 may have, at least in part, contributed to the phenotype.
Bmal1 promotes satellite cell expansion in muscle regeneration Satellite cells, a well-characterized muscle stem cell population that resides beneath the basal lamina of muscle fibers, are a major 6
Rel. Exp./ 36B4/WT-D0
WT Bmal1-/-/
##
##
4
#
*
##
#
2 0
* D0
D5
**
D15
D0
β-catenin
D5
D15
TCF3
WT Day
0
5
D0
D5
** D15
Axin2 Bmal1 -/-
8
15
0
5
8
15
β-catenin β t i β-actin
Fig. 5 – Attenuated Wnt signaling in muscle regeneration due to absence of Bmal1. (A) RT-qPCR analysis of mRNA expression of genes of the Wnt signaling pathway at indicated time points during CTX-induced muscle regeneration (n ¼4–5/group). (B) Immunoblot analysis of Wnt signaling effector β-catenin protein during the course of muscle regeneration (n¼ 4–5). n nn , : po0.05 and 0.01 Bmal1 / vs. WT; ##po0.01 D14 vs. D0 by Student's t test.
331 (2015 ) 20 0 –210
source of myogenic precursors required for muscle regeneration [20]. Upon muscle injury, satellite cells are activated to proliferate, subsequently terminally differentiate and fuse with each other or existing myofibers to repair the damage. Therefore, proper expansion of satellite cell pool in response to injury-induced myogenic signals plays a critical role in determining the outcomes of regeneration [22,23]. We thus examined this key process in Bmal1-null mice. During CTX-induced regeneration in control mice, protein expression of Pax7, a definitive satellite cell marker [24,25], is highly induced at day 5 with substantially higher levels still detected at day 8 (Fig. 6A), suggesting a marked increase of the satellite cell pool during the peak of regenerative myogenesis, as documented previously [22,23]. In contrast, this Pax7 response reflecting satellite cell expansion is severely blunted in Bmal1null, with only slight induction at day 5 and weakly detectable levels at day 8. Analysis of Pax7 transcripts revealed significantly lower levels in Bmal1 / mice throughout injury as compared to the WT (Fig. 6B). To directly visualize assess satellite cell expansion in regenerating muscle, we performed Pax7 immunostaining at post-CTX day 5 (Fig. 6C), a peak of satellite expansion phase. Quantitative analysis of number of Pax7þ satellite cells at this time point after injury reveals a significant 30% reduction of the percentage of Pax7þ cells (Fig. 6E) in Bmal1 / mice with a tendency toward reduced total number of satellite cells per field (Fig. 6D).
Bmal1 positively regulates primary myoblast growth and proliferation Satellite cell activation and proliferation in response to muscleinjury induced mitogens critically determine the myogenic progenitor pool for regeneration [22,23]. Therefore, the defect in satellite cell expansion observed with loss of Bmal1 function in vivo may reflect an attenuated proliferative capacity of satellite cells. We thus isolated satellite cell-derived primary myoblasts from WT and Bmal1 / mice and examined their growth and proliferative properties ex vivo. After a week in culture, primary myoblasts devoid of Bmal1 exhibit a markedly slower growth rate reaching merely 50% of that of the normal cells, as shown in Fig. 7A. The loss of Bmal1 also leads to a 40% reduction of proliferation rate as assessed by EdU incorporation (Fig. 7B and C). In addition, this reduced proliferation is accompanied by a significantly increased apoptotic rate as demonstrated by quantitative TUNEL assay (Fig. 7D). It is possible that both the defect in proliferation and enhanced apoptosis in Bmal1 / primary myoblasts contributed to their substantially slower growth. Notably, the mitogenic effect of a known mitogen induced by muscle injury, Fgf2, is also diminished in Bmal1 / cells as indicated by significantly lower EdU incorporation rate (Fig. 7E and F). Together, these results from satellite cell-derived primary myoblasts indicate that Bmal1 positively regulates myoblast proliferation and growth, and this mechanism may underlie its effect on promoting satellite cell expansion during muscle regeneration.
Discussion The clock machinery has been implicated in the maintenance of muscle mass, with its disruption in mutant mice resulting in severe aging-associated sarcopenia [4,5], although the underlying
E XP ER I ME NTAL C E LL RE S E ARCH
0
15
0
Bmal1-/5 8 15
Pax7 β-actin
WT Bmal1-/-
8 Rel.Exp./36B4 4 /WT-D0
Day
WT 5 8
207
331 (2 015 ) 2 0 0 – 210
##
##
6
*,##
4
**,##
#
*,##
2 0
D0
Pax-7
Pax-7/DAPI
D5
D8
D15
Phase contrast
WT
30
40
ax-7+/DAPI (%) Pa
Numbe er of Pax-7+ /Field /
Bmal1-/-
30 20 10 0
WT Bmal1-/-
20
**
10 0
WT Bmal1-/-
Fig. 6 – Impaired satellite cell expansion during muscle regeneration in Bmal1-null mice. Immunoblot (A) and RT-qPCR analysis (B) of satellite cell marker Pax7 during CTX-induced muscle regeneration in WT and Bmal1 / mice. Protein analysis was carried out using pooled sample from 4–5 mice/time point and qPCR was performed using individual samples (n¼ 4–5). (C) Representative images of Pax7 immunofluorescence staining with DAPI nuclei stain and phase contrast images (20X), and (D) quantification of total number of Pax7þ satellite cells per field, or (E) percentage of Pax-7þ/DAPI at day 5 of regeneration. Arrows indicate representative satellite cells with Pax7þ staining co-localize with DAPI. n,nn: po0.05 and 0.01 Bmal1 / vs. WT; ##po0.01 vs. D0 control by Student's t test.
mechanisms are not clear. In the current study, we interrogated the in vivo function of the essential clock activator, Bmal1, and reveal its role as an integral component of the myogenic response to promote muscle regeneration. Furthermore, our study indicates that Bmal1 critically impacts muscle stem cell expansion, the essential step of an appropriate muscle repair process. A surprising finding from our study is that Bmal1 protein is strongly stimulated by muscle injury, with an induction kinetic that closely mimics that of the myogenic response factors. This implicates that Bmal1 could be part of a pro-myogenic network that responds to myogenic stimuli in a similar manner. Damage to skeletal muscle has been shown to cause the release of various factors that activates satellite cells and stimulates their proliferation as well as subsequent differentiation [20]. To mimic the muscle damage-induced myogenic process in vitro, soluble
fractions extracted from the crushed muscle, which was shown to contain mitogenic signals [19,26–28], were tested on myoblasts. Although various growth factors [29–31], inflammatory cytokines [32,33] or signaling molecules [11,16,34] have been reported to possess myogenic properties, the precise nature of the myogenic stimuli in vivo is likely to be a combination of varying molecules at specific phase of the regeneration process. Notably, Bmal1 is induced by CME as well as the candidate myogenic signals tested, Fgf2, Forskolin, and Wnt, suggesting that this clock activator is highly responsive to myogenic signals induced by muscle injury. Whether additional mitogenic components of CME, such as the hepatocyte growth factor [35], epidermal growth factor or insulin-like growth factor [30,31], have similar effects on Bmal1 remains to be investigated in future studies. As a molecule that is entrained by light in the
208
EX PE R IM E NTAL C ELL RE S EA RC H
WT Bmal1-/-
Growth Rate
8
WT
6
OD (450nm)
Cell Number (x106)
10
331 (2015 ) 20 0 –210
Bmal1-/-
4 2 0
0
2
4 Days
WT
6
3.5 3 2.5 2 1.5 1 0.5 0
**
0.6
**
**
**
0.4 0.2 0
Nucleasetreated
8
TUNEL Assay
0.8
Unlabeled
20K
40K 80K Cells/well
Bmal1-/-
EdU/D DAPI
EdU+/DAPI (%)
EdU U
30 25 20
10 5 0
WT+Fgf2 g
**
15
WT
Bmal1-/-
Bmal1-/- +Fgf2 g
Edu u+/DAPI (%)
EdU
60
40
**
20
dU/DAPI Ed
0
WT +Fgf2
Bmal1-/+Fgf2
Fig. 7 – Bmal1 / primary myoblast display reduced growth and proliferation with enhanced apoptosis. (A) Growth rate of satellite cell-derived primary myoblasts at indicated time in culture (n¼4). Cells were seeded at 1 106 per 10 cm plate. nnpo0.01 and 0.01 Bmal1 / vs. WT by one way ANOVA. (B, E) Cell proliferation rate as assessed by EdU labeling and staining in the absence (B), and presence of Fgf2 (E) with quantification of percentage of EdUþ cells among total nuclei (C, F). Fgf2 (25 ng/ml) was added to cells for 24 h prior to EdU labeling (E, F). (D) Apoptosis rate as analyzed by quantitative TUNEL assay in primary myoblasts isolated from WT and Bmal1 / mice with endonuclease-treated cells as positive controls (n¼3). n,nn: po0.05 and 0.01 Bmal1 / vs. WT by Student's t test.
central clock or endogenous timing cues in peripheral tissues [1,2], the responsiveness of Bmal1 to environmental signals in order to convey the timing cues has been demonstrated in various systems [7,36]. Interestingly, up-regulation of Bmal1 precedes that of the myogenic regulatory factor and embryonic myosin heavy chain by CME treatment. Given its pro-myogenic action [18], this result implies that Bmal1 could be part of a
myogenic network involved in initiation of regenerative myogenesis. Since Bmal1 is a master regulator of the core molecular clock circuit, these findings also indicate that the molecular clock could be involved in orchestrating the orderly progression of myogenic events during muscle regeneration. Future investigations of involvement of additional clock genes in muscle regeneration may shed light on this intriguing possibility.
E XP ER I ME NTAL C E LL RE S E ARCH
Using two muscle injury models, we definitively establish that Bmal1 plays a positive regulatory role in muscle regeneration. The most striking finding we observed in Bmal1-null animals is that the eMyHC protein level is barely elevated at day 3 after injury, in sharp contrast to its peak induction in the controls. This near absence of protein induction is also evident from the immunostaining result. A similarly deficient Pax7 induction further indicates that the initial phase of the satellite cell expansion and de novo new fiber formation is markedly impaired by the loss of Bmal1 function in vivo. Although we observed moderate phenotype in later stages of the regeneration, the regeneration defect in Bmal1 / mice persists, as indicated by significant reduction of regenerating myofiber caliber at 14 days post-CTX and lower regenerated muscle weight at 30 days postfreeze-injury. These findings likely reflect that Bmal1 and its temporal regulation play a more dominant role in the initial myogenic events that are important at early stages of regeneration. Muscle repair in response to injury follows a defined process, which involves an initial inflammatory infiltration, subsequent myogenic precursors (MPC) activation, proliferation, differentiation and fusion [20,37,38] leading to new myofiber formation. The expansion of MPC, comprised mostly of satellite cells, provides a major source for muscle repair [37,38]. As we previously found that Bmal1 regulates myoblast differentiation through the Wnt signaling pathway [18], we studied whether the regeneration defect observed in Bmal1-null mice is associated with attenuated Wnt signal transduction. Based on the significantly blunted expression levels of Wnt pathway genes, β-catenin, Tcf3 and established Wnt target gene Axin2, Wnt activity is compromised with the absence of Bmal1 in vivo. We demonstrated that Bmal1 exerts direct transcriptional regulation on canonical E-boxes present in β-catenin and Tcf3 promoters [18]. Therefore, loss of this Bmal1 transcriptional activity is a likely mechanism responsible for blunted Wnt signaling observed in Bmal1-null mice regeneration. A critical signal involved in embryonic myogenesis [13], Wnt activation has also been shown to be a key switch regulating satellite cell self-renewal to differentiation during muscle regeneration [11,34]. Although our study does not rule out the potential involvement of additional signaling mechanisms, these data provide evidence that the lower Wnt signaling activity we observed in vivo may contribute to the suppressed myogenic response in Bmal1-null mice regeneration. Interestingly, as Notch pathway is known to regulate satellite cell expansion phase of skeletal muscle regeneration [14] and could be under the circadian modulation of Bmal1[39]. Therefore, although beyond the scope of the current study, a potential contribution of Notch-mediated mechanism warrants future investigations. It has been increasingly recognized that the circadian clock regulates stem cell properties in various tissues, such as the hair follicle [39–41], intestine [42], the hematopoetic system [43] and neuronal organs [44]. Particularly, in epidermal and intestinal stem cells, the circadian regulation is a critical timing mechanism involved in physiological tissue remodeling [39–42]. Our results indicate that loss of Bmal1 hampers satellite cell expansion during muscle regeneration in vivo, a finding that is in line with observed reduced growth and proliferation of these cells ex vivo. Therefore, Bmal1 regulation of muscle stem cell behavior could underlie the defective regeneration seen in Bmal1-null mice. As adult skeletal muscle goes through constant repair and remodeling due to exercise-induced injuries [15], we postulate that this mechanism may mediate, at least in part, the established role of Bmal1 in
331 (2 015 ) 2 0 0 – 210
209
preserving muscle mass in adult mice. Therefore, these findings from our study indicate a novel functional circadian regulatory mechanism in the muscle stem cell population, which may participate in adult skeletal muscle remodeling.
Conclusion We demonstrate, for the first time, that a key clock regulator, Bmal1, plays a critical role in regulating muscle stem cell expansion to impact muscle regeneration. Our study implies that the intrinsic muscle clock may exert a temporal control over a cascade of intricately coordinated myogenic events to ensure the orderly progression of regenerative myogenesis. Comprehensive elucidation of this temporal regulatory mechanism involved in maintaining muscle mass may lead to discovery of targeted strategies to help preserve critical muscle mass in muscle-wasting diseases.
Grants This project is supported by Houston Methodist Research Institute start-up funds, American Heart Association Grant 12SDG12080076 and American Diabetes Association Grant 1-13-BS-118 to K. Ma.
Author contributions S.C., H.Y and D.N. designed, performed experiments and prepared figures; Y.L. carried out cardiotoxin-injury histological analysis and edited manuscript; K.M. conceived research, interpreted results of experiments, prepared figures and manuscript.
Acknowledgments We thank the Core Laboratory of Center for Diabetes Research at Houston Methodist Research Institute for their expert technical assistance, and Dr. Yong Li (UT-HSC) for technical support and proof reading of the manuscript.
references [1] C. Dibner, U. Schibler, U. Albrecht, The mammalian circadian timing system: organization and coordination of central and peripheral clocks, Annu. Rev. Physiol. 72 (2010) 517–549. [2] S.M. Reppert, D.R. Weaver, Coordination of circadian timing in mammals, Nature 418 (2002) 935–941. [3] B.H. Miller, E.L. McDearmon, S. Panda, K.R. Hayes, J. Zhang, J.L. Andrews, M.P. Antoch, J.R. Walker, K.A. Esser, J.B. Hogenesch, J.S. Takahashi, Circadian and CLOCK-controlled regulation of the mouse transcriptome and cell proliferation, Proc. Natl. Acad. Sci. USA 104 (2007) 3342–3347. [4] R.V. Kondratov, A.A. Kondratova, V.Y. Gorbacheva, O.V. Vykhovanets, M.P. Antoch, Early aging and age-related pathologies in mice deficient in BMAL1, the core componentof the circadian clock, Genes Dev. 20 (2006) 1868–1873. [5] E.L. McDearmon, K.N. Patel, C.H. Ko, J.A. Walisser, A.C. Schook, J.L. Chong, L.D. Wilsbacher, E.J. Song, H.K. Hong, C.A. Bradfield, J.S. Takahashi, Dissecting the functions of the mammalian clock protein BMAL1 by tissue-specific rescue in mice, Science 314 (2006) 1304–1308.
210
EX PE R IM E NTAL C ELL RE S EA RC H
[6] J.L. Andrews, X. Zhang, J.J. McCarthy, E.L. McDearmon, T.A. Hornberger, B. Russell, K.S. Campbell, S. Arbogast, M.B. Reid, J.R. Walker, J.B. Hogenesch, J.S. Takahashi, K.A. Esser, CLOCK and BMAL1 regulate MyoD and are necessary for maintenance of skeletal muscle phenotype and function, Proc. Natl. Acad. Sci. USA 107 (2010) 19090–19095. [7] M.K. Bunger, L.D. Wilsbacher, S.M. Moran, C. Clendenin, L.A. Radcliffe, J.B. Hogenesch, M.C. Simon, J.S. Takahashi, C.A. Bradfield, Mop3 is an essential component of the master circadian pacemaker in mammals, Cell 103 (2000) 1009–1017. [8] M. Ikeda, M. Nomura, cDNA cloning and tissue-specific expression of a novel basic helix-loop-helix/PAS protein (BMAL1) and identification of alternatively spliced variants with alternative translation initiation site usage, Biochem. Biophys. Res. Commun. 233 (1997) 258–264. [9] G. Rey, F. Cesbron, J. Rougemont, H. Reinke, M. Brunner, F. Naef, Genome-wide and phase-specific DNA-binding rhythms of BMAL1 control circadian output functions in mouse liver, PLoS Biol. 9 (2011) e1000595. [10] M. Buckingham, Myogenic progenitor cells and skeletal myogenesis in vertebrates, Curr. Opin. Genet. Dev. 16 (2006) 525–532. [11] A.S. Brack, I.M. Conboy, M.J. Conboy, J. Shen, T.A. Rando, A temporal switch from notch to Wnt signaling in muscle stem cells is necessary for normal adult myogenesis, Cell Stem Cell 2 (2008) 50–59. [12] S. Chatterjee, D. Nam, B. Guo, J.M. Kim, G.E. Winnier, J. Lee, R. Berdeaux, V.K. Yechoor, K. Ma, Brain and muscle Arnt-like 1 is a key regulator of myogenesis, J. Cell Sci. 126 (2013) 2213–2224. [13] G. Cossu, U. Borello, Wnt signaling and the activation of myogenesis in mammals, EMBO J. 18 (1999) 6867–6872. [14] S. Tsivitse, Notch and Wnt signaling, physiological stimuli and postnatal myogenesis, Int. J. Biol. Sci. 6 (2010) 268–281. [15] A.J. Wagers, I.M. Conboy, Cellular and molecular signatures of muscle regeneration: current concepts and controversies in adult myogenesis, Cell 122 (2005) 659–667. [16] R. Stewart, L. Flechner, M. Montminy, R. Berdeaux, CREB is activated by muscle injury and promotes muscle regeneration, PloS One 6 (2011) e24714. [17] B. Gayraud-Morel, F. Chretien, P. Flamant, D. Gomes, P.S. Zammit, S. Tajbakhsh, A role for the myogenic determination gene Myf5 in adult regenerative myogenesis, Dev. Biol. 312 (2007) 13–28. [18] S. Chatterjee, D. Nam, B. Guo, J.M. Kim, G.E. Winnier, J. Lee, R. Berdeaux, V.K. Yechoor, K. Ma, Brain and muscle Arnt-like 1 is a key regulator of myogenesis, J. Cell Sci. 126 (2013) 2213–2224. [19] R. Bischoff, A satellite cell mitogen from crushed adult muscle, Dev. Biol. 115 (1986) 140–147. [20] S.B. Charge, M.A. Rudnicki, Cellular and molecular regulation of muscle regeneration, Physiol. Rev. 84 (2004) 209–238. [21] N. Preitner, F. Damiola, L. Lopez-Molina, J. Zakany, D. Duboule, U. Albrecht, U. Schibler, The orphan nuclear receptor REVERBalpha controls circadian transcription within the positive limb of the mammalian circadian oscillator, Cell 110 (2002) 251–260. [22] A. Sacco, R. Doyonnas, P. Kraft, S. Vitorovic, H.M. Blau, Selfrenewal and expansion of single transplanted muscle stem cells, Nature 456 (2008) 502–506. [23] S. Kuang, M.A. Gillespie, M.A. Rudnicki, Niche regulation of muscle satellite cell self-renewal and differentiation, Cell Stem Cell 2 (2008) 22–31. [24] S. Gunther, J. Kim, S. Kostin, C. Lepper, C.M. Fan, T. Braun, Myf5positive satellite cells contribute to Pax7-dependent long-term maintenance of adult muscle stem cells, Cell Stem Cell 13 (2013) 590–601. [25] J. von Maltzahn, A.E. Jones, R.J. Parks, M.A. Rudnicki, Pax7 is critical for the normal function of satellite cells in adult skeletal muscle, Proc. Natl. Acad. Sci. USA 110 (2013) 16474–16479.
331 (2015 ) 20 0 –210
[26] K.L. Haugk, R.A. Roeder, M.J. Garber, G.T. Schelling, Regulation of muscle cell proliferation by extracts from crushed muscle, J. Anim. Sci. 73 (1995) 1972–1981. [27] G. Chen, L.S. Quinn, Partial characterization of skeletal myoblast mitogens in mouse crushed muscle extract, J. Cell. Physiol. 153 (1992) 563–574. [28] R. Bischoff, Cell cycle commitment of rat muscle satellite cells, J. Cell Biol. 111 (1990) 201–207. [29] J.P. Lefaucheur, A. Sebille, Muscle regeneration following injury can be modified in vivo by immune neutralization of basic fibroblast growth factor, transforming growth factor beta 1 or insulin-like growth factor I, J. Neuroimmunol. 57 (1995) 85–91. [30] D.R. Cook, M.E. Doumit, R.A. Merkel, Transforming growth factorbeta, basic fibroblast growth factor, and platelet-derived growth factor-BB interact to affect proliferation of clonally derived porcine satellite cells, J. Cell. Physiol. 157 (1993) 307–312. [31] M.E. Doumit, D.R. Cook, R.A. Merkel, Fibroblast growth factor, epidermal growth factor, insulin-like growth factors, and platelet-derived growth factor-BB stimulate proliferation of clonally derived porcine myogenic satellite cells, J. Cell. Physiol. 157 (1993) 326–332. [32] L. Austin, J. Bower, J. Kurek, N. Vakakis, Effects of leukaemia inhibitory factor and other cytokines on murine and human myoblast proliferation, J. Neurol. Sci. 112 (1992) 185–191. [33] J.B. Kurek, S. Nouri, G. Kannourakis, M. Murphy, L. Austin, Leukemia inhibitory factor and interleukin-6 are produced by diseased and regenerating skeletal muscle, Muscle Nerve 19 (1996) 1291–1301. [34] A.S. Brack, F. Murphy-Seiler, J. Hanifi, J. Deka, S. Eyckerman C. Keller, M. Aguet, T.A. Rando, BCL9 is an essential component of canonical Wnt signaling that mediates the differentiation of myogenic progenitors during muscle regeneration, Dev. Biol. 335 (2009) 93–105. [35] R. Tatsumi, J.E. Anderson, C.J. Nevoret, O. Halevy, R.E. Allen, HGF/SF is present in normal adult skeletal muscle and is capable of activating satellite cells, Dev. Biol. 194 (1998) 114–128. [36] A. Balsalobre, F. Damiola, U. Schibler, A serum shock induces circadian gene expression in mammalian tissue culture cells, Cell 93 (1998) 929–937. [37] F. Le Grand, M.A. Rudnicki, Skeletal muscle satellite cells and adult myogenesis, Curr. Opin. Cell Biol. 19 (2007) 628–633. [38] X. Shi, D.J. Garry, Muscle stem cells in development, regeneration, and disease, Genes Dev. 20 (2006) 1692–1708. [39] P. Janich, G. Pascual, A. Merlos-Suarez, E. Batlle, J. Ripperger U. Albrecht, H.Y. Cheng, K. Obrietan, L. Di Croce, S.A. Benitah, The circadian molecular clock creates epidermal stem cell heterogeneity, Nature 480 (2011) 209–214. [40] P. Janich, K. Toufighi, G. Solanas, N.M. Luis, S. Minkwitz, L. Serrano, B. Lehner, S.A. Benitah, Human epidermal stem cell function is regulated by circadian oscillations, Cell Stem Cell 13 (2013) 745–753. [41] M.V. Plikus, C. Vollmers, D. de la Cruz, A. Chaix, R. Ramos S. Panda, C.M. Chuong, Local circadian clock gates cell cycle progression of transient amplifying cells during regenerative hair cycling, Proc. Natl. Acad. Sci. USA 110 (2013) E2106–E2115. [42] P. Karpowicz, Y. Zhang, J.B. Hogenesch, P. Emery, N. Perrimon, The circadian clock gates the intestinal stem cell regenerative state, Cell Rep. 3 (2013) 996–1004. [43] S. Mendez-Ferrer, A. Chow, M. Merad, P.S. Frenette, Circadian rhythms influence hematopoietic stem cells, Curr. Opin. Hematol. 16 (2009) 235–242. [44] P. Bouchard-Cannon, L. Mendoza-Viveros, A. Yuen, M. Kaern H.Y. Cheng, The circadian molecular clock regulates adult hippocampal neurogenesis by controlling the timing of cell-cycle entry and exit, Cell Rep. 5 (2013) 961–973.
331 (2 015 ) 20 0 – 210
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/yexcr
Research article
Brain and muscle Arnt-like 1 promotes skeletal muscle regeneration through satellite cell expansion Somik Chatterjeea, Hongshan Yina,b, Deokhwa Nama, Yong Lic, Ke Maa,n a
Center for Diabetes Research, Department of Medicine, Houston Methodist Research Institute, Houston, TX 77030, USA Department of Cardiovascular Medicine, Third Affiliated Hospital, Hebei Medical University, Shijiazhuang 050051, Hebei, China c Department of Pediatric Surgery, Center for Stem Cell Research and Regenerative Medicine, University of Texas Health Science Center at Houston, Houston, TX 77030, USA b
article information
abstract
Article Chronology:
Circadian clock is an evolutionarily conserved timing mechanism governing diverse biological
Received 29 July 2014
processes and the skeletal muscle possesses intrinsic functional clocks. Interestingly, although the
Received in revised form
essential clock transcription activator, Brain and muscle Arnt-like 1 (Bmal1), participates in
27 August 2014
maintenance of muscle mass, little is known regarding its role in muscle growth and repair. In
Accepted 28 August 2014
this report, we investigate the in vivo function of Bmal1 in skeletal muscle regeneration using two
Available online 9 September 2014
muscle injury models. Bmal1 is highly up-regulated by cardiotoxin injury, and its genetic ablation
Keywords: Muscle regeneration Circadian rhythm Satellite cell Proliferation
significantly impairs regeneration with markedly suppressed new myofiber formation and attenuated myogenic induction. A similarly defective regenerative response is observed in Bmal1-null mice as compared to wild-type controls upon freeze injury. Lack of satellite cell expansion accounts for the regeneration defect, as Bmal1 / mice display significantly lower satellite cell number with nearly abolished induction of the satellite cell marker, Pax7. Furthermore, satellite cell-derived primary myoblasts devoid of Bmal1 display reduced growth and proliferation ex vivo. Collectively, our results demonstrate, for the first time, that Bmal1 is an integral component of the pro-myogenic response that is required for muscle repair. This mechanism may underlie its role in preserving adult muscle mass and could be targeted therapeutically to prevent muscle-wasting diseases. & 2014 Elsevier Inc. All rights reserved.
Introduction The circadian clock is a universal timing mechanism that generates the 24 h rhythm. Nearly all tissues and cell types possess
intrinsic molecular clocks, central and peripheral clocks, that are driven by an interlocked feed-back network of transcriptional regulators coupled with translational and post-translational control [1,2]. This temporal regulatory mechanism governs a wide
Abbreviations: Bmal1, brain and muscle Arnt-like 1; CTX, Cardiotoxin; CME, crushed muscle extract; CLOCK, circadian locomotor output cycles kaput; FGF2, fibroblast growth factor 2; Wnt, wingless-type MMTV integration site family; eMyHC, embryonic myosin heavy chain; TA, tibialis anterior n
Corresponding author. Fax: þ1 713 793 7162. E-mail address: [email protected] (K. Ma).
http://dx.doi.org/10.1016/j.yexcr.2014.08.041 0014-4827/& 2014 Elsevier Inc. All rights reserved.
E XP ER I ME NTAL C E LL RE S E ARCH
variety of cyclical biological processes ranging from rhythmic gene expression to behavioral and physiological cycles. The skeletal muscle possesses cell-autonomous functional clocks and a significant 3.4% of the skeletal muscle transcriptome displays circadian oscillation [3]. Notably, circadian clock-controlled processes in muscle encompass a diverse range from contractile performance, cell cycle progression, sarcomere structural integrity to mitochondrial bioenergetics [3–6]. Brain and muscle Arnt-like 1 (Bmal1), a basis helix-loop-helix transcription factor, is an essential activator of the core molecular clock transcription network that is highly expressed in skeletal muscle [7,8]. It dimerizes with circadian locomotor output cycles kaput (CLOCK) to initiate transcription of target promoters through canonical E-box elements [9]. Recent evidence indicates that Bmal1 and CLOCK are involved in maintaining skeletal muscle sarcomere integrity and proper contractile function [6], while Bmal1 is also implicated in preserving muscle mass during aging [4,5]. Inactivation of Bmal1 in mice leads to severe agingassociated sarcopenia that substantially shortens life span [4], and its muscle-specific overexpression rescues survival [5]. However, although these studies suggest that the circadian clock exerts critical regulatory control on skeletal muscle functions, the mechanisms mediating these effects are not known. Thus, despite the fundamental importance of the molecular clock in skeletal muscle physiology, our understanding of how it maintains muscle mass and function is nearly non-existent, and whether the clock network participates in muscle growth and remodeling has not been studied. Intriguingly, the orderly progression of myogenesis, from commitment of mesenchymal progenitors to myoblast lineage to terminal differentiation and fusion to form multinucleated myotubes, is a highly temporally coordinated event that requires the sequential activation of the myogenic regulatory factor cascade [10]. Importantly, a temporal order of Notch activation preceding Wnt activity has been shown to be required for proper muscle regeneration to ensure stem cell activation, proliferation before terminal differentiation [11], while inappropriate activation of either of these myogenic signals leads to adverse outcomes of muscle repair. We recently demonstrated that Bmal1 has a pro-myogenic function in myogenesis through transcriptional control of the canonical Wnt signaling pathway [12], suggesting that indeed a molecular clock regulation is involved in the progression of myogenic events. Given that Wnt is a major inductive signaling during embryonic and regenerative myogenesis [13,14], our findings imply that Bmal1 may participate in these processes. Based on the role of Bmal1 in preserving muscle mass, we postulate that it may positively regulate muscle regeneration in vivo. Therefore, employing established muscle injury models, our current study interrogates the function of Bmal1 in adult muscle repair, an important aspect of physiological remodeling of skeletal muscle due to activity or exercise-induced injuries [15]. Our findings reveal that Bmal1 is an inherent component of the myogenic response that facilitates regenerative myogenesis.
Materials and methods Animals Mice were maintained in the Methodist Hospital Research Institute vivarium under a constant 12:12 light dark cycle, with lights
331 (2 015 ) 2 0 0 – 210
201
on at 7:00 a.m. (ZT0). All animal procedures were conducted in accordance with the Guidelines for the Care and Use of Laboratory Animals and were approved by the IACUC committee of Houston Methodist Research Institute. Bmal1 / mice [7] (B6.129Arntltm1Bra/J) were obtained originally from the Jackson Laboratory and maintained in the laboratory. Mice 8–12 weeks of age were used throughout the study to exclude effects of aging-related pathologies [4].
Muscle injury models For cardiotoxin (CTX) injection in tibialis anterior (TA) muscle, 30 μl of 10 μM solution was injected at three sites along the longitudinal length to ensure uniform injury of the entire muscle as described [16], with PBS injection as controls. Local freeze injury in TA was induced by applying a metal rod that was dipped in liquid nitrogen to exposed muscle surface for 10 s, and repeated three times to ensure depth of the injury, as described [17]. Shamoperated animals were used as controls. Muscle tissues were collected at 3 p.m., the time of the initial CTX injection, to ensure comparable circadian time sampling.
Primary myoblast isolation and culture Primary myoblasts were isolated from hind limb muscle of 4 week-old mice as described previously [18]. Briefly, muscles were minced into small pieces and subjected to 1% collagenase digestion at 37 1C for 30 min with agitation. Digested cells were seeded on collagen-coated plates, and serial preplating was used to deplete fibroblasts consecutively. Subsequent selective growth enriches myoblasts, and cell amplified after 6 passages were used for the study. Purity of myoblasts obtained by this method was confirmed by their uniform differentiation into myosin heavy chain (MHC)-positive myotubes. The cells were maintained in DMEM with 20% FBS on collagen-coated plates. Cell samples were collected under normal culture conditions without synchronization of cellular clocks.
Crushed muscle extract Crushed muscle extract was obtained following the procedure described by Bischoff [19]. Briefly, leg muscles from 8–10 weeks old C57BL/6 mice were excised and pressed gently seven times with a pair of blunt forceps. Muscles were then incubated at 4 1C at a concentration of 1 g muscle/ml with gentle shaking in Tris-buffered saline for 2 h. The supernatant was collected, filtered, and assayed for protein concentration by a DC Protein Assay Kit (Bio-Rad). Pooled batches of muscle extract were stored at 80 1C until use. For primary myoblast treatment, muscle extract was used at a protein concentration of 300 μg/ml in DMEM containing 10% FBS.
Histology, immunofluorescence staining and quantification Muscles were snap frozen in liquid-nitrogen cooled isopentane and embedded in OCT. 10 μm thick cryosections of middle region of the injured TA muscle were obtained. Histological analysis was performed using hematoxylin and eosin staining. For immunofluorescence, cryosections were fixed using 4% paraformaldehyde and permeabilized by 0.2% Triton X-100. Endogenous IgG was
202
EX PE R IM E NTAL C ELL RE S EA RC H
blocked using non-specific mouse IgG and further blocking using 5% goat serum was carried out prior to primary antibody incubation at 4 1C overnight. The antibodies used were Pax7 (1:20), embryonic myosin heavy chain (BF-G6, 1:100) and dystrophin (MANDRA1 7A10, 1:100), Developmental Studies Hybridoma Bank; laminin (sc-6017, 1:500), Santa Cruz Biotechnology. Secondary antibodies were goat anti-mouse and rabbit anti-goat (Thermo Scientific) used at 1:500 for 1 h incubation at room temperature together with DAPI for nuclei staining. Cross-section area of nascent fibers containing centralized nuclei was measured using H/E staining sections of three representative fields (20X) from each muscle section using Nikon NIS-Elements analysis software. The number of Pax7þ/DAPI double-positive cells was determined from three representative fields (40X) and the percentage to total DAPI-stained nuclei was calculated.
Immunoblot analysis 20–50 mg of total protein was resolved on SDS-PAGE gels followed by immunoblotting after nitrocellulose membrane transfer, and developed by chemiluminescence (Supersignal; Pierce Biotechnology), as described [18]. Myf5 (sc-302, 1:1000), MyoD1 (sc-760, 1:1000), and β-actin (sc-1616, 1:3000) antibodies were obtained from Santa Cruz Biotechnology, myogenin antibody (MAB 3879, 1:1000) was from Millipore, and Bmal1 antibody (PAI-523, 1:500) was from Thermo Scientific.
RNA extraction and quantitative reverse-transcriptase PCR analysis RNeasy miniprep kits (Qiagen) were used to isolate total RNA from snap-frozen muscle tissues and cells. cDNA was generated using a q-Script cDNA Supermix kit (Quanta Biosciences) and quantitative PCR was performed using a Roche 480 Light Cycler with Perfecta SYBR Green Supermix (Quanta Biosciences), as described [18]. Relative expression levels were determined using the comparative Ct method to normalize target genes to 36B4 internal control and compared to controls as indicated.
EdU proliferation assay Primary myoblasts seeded on collagen-coated chamber slides at a density of 80,000 per well were labeled with 10 μM 5-ethynyl-20 deoxyuridine (EdU) for 5 h. Fgf2 (25 ng/ml) was added to cells for 24 h. EdU incorporation was detected by a Click-iT Imaging Kit (Invitrogen) with Alexa 488 Fluorophore. DAPI was used to label for nuclei. Total number of EdUþ cells was counted from 4 representative fields at 20X and the rate of proliferation was calculated as percentage of EdUþ/DAPI.
331 (2015 ) 20 0 –210
Statistical analysis Data are presented as mean7SE. One way ANOVA or two-tailed Student's t-test were performed as appropriate as indicated. Statistical analysis was carried out using SigmaPlot. Po0.05 was considered statistically significant.
Results Robust induction of Bmal1 during muscle regeneration Bmal1 is highly enriched in skeletal muscle [8]. Consistent with this notion, examination of Bmal1 protein level in mesodermal cell types reveals that it is more abundant in C2C12 and primary myoblasts than C3H10T1/2 mesenchymal precursors and 3T3-L1 preadipocytes (Fig. 1A), with the highest level detected in C2C12 myoblasts. Based on our previous finding that Bmal1 promotes myogenic differentiation [18], we postulate that it may respond to myogenic stimuli. Remarkably, cardiotoxin (CTX)-induced muscle injury elicits a robust elevation of Bmal1 protein, with a peak expression detected at day 5 following injury that declines but persists till day 15 (Fig. 1B). This persistent elevation at day 15, a regeneration time point with on-going myogenic activity but absence of inflammation, indicates that Bmal1 induction is likely correlated with myogenesis but not simply a part of the initial CTX-induced inflammatory response. Interestingly, Bmal1 induction kinetics during this regeneration time course mimics that of a known myogenic regulatory factor, myogenin, suggesting that it may respond to injury-induced myogenic signals in a similar manner as a bona fide myogenic regulatory factor. Indeed, treatment of primary myoblasts with crushed muscle extract (CME) [19], obtained from physically damaged muscle fibers, rapidly induces Bmal1 expression after 6 h with 2-fold up-regulation at 12 h (Fig. 1C). In addition, the induction of Bmal1 precedes the ability of CME to up-regulate the myogenic terminal differentiation factor, myogenin, and embryonic myosin heavy chain (Myhc3), both occurring only at 12 h. As various agents released upon muscle injury have been shown to possess myogenic stimulatory effects [20], we next determined whether specific factors, including cAMP-inducer, Forskolin, Wnt3a and Fgf2, can elicit the observed Bmal1 response to muscle injury. Surprisingly, all these known myogenic stimuli tested are capable of activating Bmal1 expression, with Fgf2 leading to the highest 2-fold up-regulation (Fig. 1D). On the other hand, both Forskolin and Fgf2 suppressed the expression of the negative regulator of the molecular clock, Rev-erbα [21], suggesting that these factors may exert a coordinated influence on the clock machinery (data not shown). These findings indicate that Bmal1 responds to signals elicited by muscle injury during initiation of regeneration, implicating its involvement in the early myogenic activity in muscle repair.
Quantitative TUNEL assay Loss of Bmal1 impairs muscle regeneration Primary myoblasts were plated on collagen coated 96 well plate at seeding density of 20,000–120,000 cells per well. Cells treated with endonuclease for 1 h to induce apoptosis were used as a positive control. Terminal deoxynucleotidyl transferase (TdT) in situ labeling and detection of apoptosis was performed 24 h after seeding by a colorimetric TACS apoptosis detection kit (Trevigen).
Based on the above findings, we explored a potential pro-myogenic function of Bmal1 in vivo using two different injury models, cardiotoxin injection to the entire length of the tibialis anteria (TA) muscle and freezing injury of the same site [20]. These injuries inflict extensive damage throughout the muscle to elicit uniform regenerative responses. Following the time course of CTX-induced
331 (2 015 ) 2 0 0 – 210
Arbitrary U Unit /TBP
E XP ER I ME NTAL C E LL RE S E ARCH
Bmal1
203
3 2 1 0
TBP
Cardiotoxin Day
0
5
8
15
Bmal1 Myogenin β-actin
Rel. mRNA Exp./Ctr
2.5
**
2
*
1.5
*
**
Ctr
1
CME-6hr
0.5
CME-12hr
0
Bmal1
Myogenin
Myhc3
Rel. mRNA m Exp. /PBS or DMSO
2.5
**
2
** **
** **
*
1.5
2hr 4hr
1 0.5 0
PBS
DMSO
Forskolin
Wnt3a
Fgf2
Fig. 1 – Induction of Bmal1 expression by myogenic stimuli. Immunoblot analysis of Bmal1 protein abundance in mesodermal cell types with quantitation (A), and its response to cardiotoxin-induced muscle injury in wild-type mice (B). RT-qPCR analysis of Bmal1 mRNA expression in primary myoblasts following crude muscle extract treatment (C), or in response to various soluble signals induced by muscle injury (D, n¼ 3). n,nn: po0.05 and 0.01 vs. respective controls ( Forskolin vs. DMSO, Wnt3a and Fgf2 vs. PBS) by Student's t test.
regeneration in normal wild-type (WT) mice, analysis of total protein lysate of regenerating muscle reveals robust induction of myogenic regulatory factors MyoD1 and Myogenin early on at day 5, which declines but remains elevated at higher levels than day 0 expression till day 15 post-injury (Fig. 2A). In contrast, Bmal1 / mice exhibit a blunted initial elevation and a severe lack of sustained induction of MyoD1 and Myogenin, indicating an attenuated myogenic response with shortened duration as compare to the WT. In addition, protein level of desmin, a marker of nascent myofibers, is significantly lower throughout the regeneration, suggesting impaired new fiber formation. Histological analysis of the injured TA further demonstrates significantly attenuated myofiber formation in Bmal1 / mice throughout the regeneration time course (Fig. 2B). There are less nascent myofibers appearing in Bmal1-null than in WT mice at the beginning of regeneration at
day 5. This defect in regeneration persists and becomes more evident at day 8 and 15, with a higher percentage of smaller size fibers containing single central nucleus as compared to WT. As shown by size distribution of these myofibers with varying diameters in Fig. 2C, the measured cross section area of nascent central-nucleated myofibers in Bmal1-null mice is markedly reduced as compared to that of WT. This reduction of caliber of newly formed myofibers is further confirmed by dystrophin immunoflurescence staining that outlines muscle fiber perimeter, as shown in Fig. 2E. Additionally, quantification of myonuclei per myofiber reveals a similarly shifted distribution toward lower numbers in the Bmal1 / mice (Fig. 2D). As newly-regenerated myofibers transiently express the embryonic isoform of myosin heavy chain (eMyHC), day 3-injured muscles were stained with an eMyHC antibody to definitively assess nascent myofiber synthesis. Compare to the strong
204
EX PE R IM E NTAL C ELL RE S EA RC H
WT 5
WT Bmal1-/-
Bmal1-/8
15
0
5
D8
8 15 % of Total
Day 0
331 (2015 ) 20 0 –210
Bmal1 MyoD1
60
**
40 20
D15
**
**
**
**
**
**
0
** μm2
Myogenin Desmin D8
D15
80
% of Total
β-actin
60
**
**
40 20
**
**
0
1
2
≥3
1
2
≥3
WT
Bmal1-/-
Dystrophin /DAPI-D8 WT
Bmal1-/-
Fig. 2 – Impaired muscle regeneration in Bmal1 / mice. (A) Immunoblot analysis of myogenic markers during regeneration time course as indicated in WT and Bmal1 / mice. Protein samples represent pooled mice of 4–5/group. Representative images of H/E staining (B) with quantification of distribution of myofiber cross section area (C), and myonuclei numbers (D) of cardiotoxininduced TA muscle regeneration (n¼4–5). (E) Representative images of dystrophin immunofluorescence staining of day 8 injured muscles to outline the perimeter of muscle fibers. n,nn: po0.05 and 0.01 Bmal1 / vs. WT by Student's t test.
immunostaining of eMyHC in WT, Bmal1 / mice display a strikingly weaker signal, as demonstrated in Fig. 3A. This finding is further validated by the barely induced total protein level of eMyHC detected in these mice at 3 days post-CTX (Fig. 3B). Thus, these results demonstrate that loss of Bmal1 significantly impairs myogenic response and new myofiber formation during muscle regeneration.
Using an additional freeze injury-induced muscle regeneration model, we further tested the pro-myogenic role of Bmal1 in vivo. At 30 days after freeze injury when the repair phase of regeneration is nearly complete, regenerated TA muscle weight (Fig. 4A) as well as its relative ratio to body weight (Fig. 4B) of Bmal1-null mice is significantly lower than that of sham-operated WT controls. This overall deficient regenerative capacity due to loss
E XP ER I ME NTAL C E LL RE S E ARCH
eMyHC
205
331 (2 015 ) 2 0 0 – 210
Laminin
Merge
WT
Bmal1-/-
WT Day
0
Bmal1-/-
3
5
0
3
5
eMyHC β-actin Fig. 3 – Loss of Bmal1 attenuates induction of embryonic myosin heavy chain during cardiotoxin-induced muscle regeneration. (A) Representative images (20X) of eMyHC immunofluorescence staining with co-staining of laminin to outline myofibers at day 3 of regeneration. (B) Immunoblot analysis of embryonic MyHC protein levels during regeneration time course in WT and Bmal1 / mice (n¼4–5).
0.24
TA/BW (%)
TA Weight (g)
0.06 0.04 0.02
Rel.. Exp./ 36B4/WT-D0
0
0.18 0.12 0.06 0
WT
Bmal1-/-
WT
Bmal1-/-
4
40
200
3
30
150
2
20
100
1
10
50
0
0
D0
D5 Myf5
D15
0
D0
D5
D15
Myogenin
D0
D5 Myhc3
D15
0
D0
D5 D15 Mrf4
Fig. 4 – Loss of Bmal1 impairs muscle regeneration and myogenic induction in freeze-injury model. Regenerated TA muscle weight (A), and ratio to total body weight (B) 30 days after freeze injury in WT and Bmal1-null mice (n¼ 4–6/group). (C) RT-qPCR analysis of myogenic gene expression in regenerating TA at indicated time after freeze injury. n¼ 4–6/group. n,nn: po0.05 and 0.01 Bmal1 / vs. WT; #, ##: po0.05 or 0.01 vs. D0 by Student's t test.
of Bmal1 is also evident in the blunted mRNA expression of myogenic regulatory factors, including Myf5, myogenin, Mrf4 as well as the new myofiber marker, Myhc3 (Fig. 4C). The similarly
defective muscle regeneration induced by freeze injury observed in Bmal1-null suggests that Bmal1 may facilitate muscle repair independent of the specific inciting damage to the muscle.
206
EX PE R IM E NTAL C ELL RE S EA RC H
Bmal1 regulation of Wnt signaling in muscle regeneration We have previously demonstrated that Bmal1 exerts transcriptional control on Wnt pathway genes, and this regulation of Wnt signaling promotes myogenesis [18]. As Wnt is a key signal that drives embryonic and regenerative myogenesis [13], we tested whether altered Wnt signaling may have contributed to the regenerative defect of Bmal1-null mice. First, we examined the mRNA expression of genes of the canonical Wnt signaling pathway. CTX-induced regeneration leads to an evident increase in Wnt signaling activity as indicated by the induction of β-catenin, TCF3 and Axin2 transcript levels in WT controls, with a 3–4 fold up-regulated peak level detected at day 5 and remains high at day 15 (Fig. 5A). In contrast, in mice lacking Bmal1, analysis of β-catenin, the ultimate signaling transducer of the canonical Wnt pathway, reveals its attenuated transcript level at day 15 (Fig. 5A), and induction of TCF3, the transcription factor mediating Wnt signal, is also compromised. Notably, expression of Axin2, a canonical Wnt target gene, is significantly lower at both day 5 and 15 in the regenerating Bmal1null muscle, with marked 70% suppression at day 15 as compared to WT. The lower Axin2 expression indicates impaired Wnt activity in Bmal1 / mice, likely a result of collective contribution of attenuated expression of a few key steps along the Wnt signal transduction cascade. Moreover, in comparison to the substantially elevated protein level of β-catenin in the WT mice (Fig. 5B), Bmal1 / mice display a weaker β-catenin level at day 8 post-CTX and its sustained increase observed in the WT at day 15 is completely absent. Based on our findings of impaired regeneration with Bmal1 ablation and the established role of Wnt in regenerative myogenesis, it is possible that the defect in Wnt signaling observed in vivo due to loss of Bmal1 may have, at least in part, contributed to the phenotype.
Bmal1 promotes satellite cell expansion in muscle regeneration Satellite cells, a well-characterized muscle stem cell population that resides beneath the basal lamina of muscle fibers, are a major 6
Rel. Exp./ 36B4/WT-D0
WT Bmal1-/-/
##
##
4
#
*
##
#
2 0
* D0
D5
**
D15
D0
β-catenin
D5
D15
TCF3
WT Day
0
5
D0
D5
** D15
Axin2 Bmal1 -/-
8
15
0
5
8
15
β-catenin β t i β-actin
Fig. 5 – Attenuated Wnt signaling in muscle regeneration due to absence of Bmal1. (A) RT-qPCR analysis of mRNA expression of genes of the Wnt signaling pathway at indicated time points during CTX-induced muscle regeneration (n ¼4–5/group). (B) Immunoblot analysis of Wnt signaling effector β-catenin protein during the course of muscle regeneration (n¼ 4–5). n nn , : po0.05 and 0.01 Bmal1 / vs. WT; ##po0.01 D14 vs. D0 by Student's t test.
331 (2015 ) 20 0 –210
source of myogenic precursors required for muscle regeneration [20]. Upon muscle injury, satellite cells are activated to proliferate, subsequently terminally differentiate and fuse with each other or existing myofibers to repair the damage. Therefore, proper expansion of satellite cell pool in response to injury-induced myogenic signals plays a critical role in determining the outcomes of regeneration [22,23]. We thus examined this key process in Bmal1-null mice. During CTX-induced regeneration in control mice, protein expression of Pax7, a definitive satellite cell marker [24,25], is highly induced at day 5 with substantially higher levels still detected at day 8 (Fig. 6A), suggesting a marked increase of the satellite cell pool during the peak of regenerative myogenesis, as documented previously [22,23]. In contrast, this Pax7 response reflecting satellite cell expansion is severely blunted in Bmal1null, with only slight induction at day 5 and weakly detectable levels at day 8. Analysis of Pax7 transcripts revealed significantly lower levels in Bmal1 / mice throughout injury as compared to the WT (Fig. 6B). To directly visualize assess satellite cell expansion in regenerating muscle, we performed Pax7 immunostaining at post-CTX day 5 (Fig. 6C), a peak of satellite expansion phase. Quantitative analysis of number of Pax7þ satellite cells at this time point after injury reveals a significant 30% reduction of the percentage of Pax7þ cells (Fig. 6E) in Bmal1 / mice with a tendency toward reduced total number of satellite cells per field (Fig. 6D).
Bmal1 positively regulates primary myoblast growth and proliferation Satellite cell activation and proliferation in response to muscleinjury induced mitogens critically determine the myogenic progenitor pool for regeneration [22,23]. Therefore, the defect in satellite cell expansion observed with loss of Bmal1 function in vivo may reflect an attenuated proliferative capacity of satellite cells. We thus isolated satellite cell-derived primary myoblasts from WT and Bmal1 / mice and examined their growth and proliferative properties ex vivo. After a week in culture, primary myoblasts devoid of Bmal1 exhibit a markedly slower growth rate reaching merely 50% of that of the normal cells, as shown in Fig. 7A. The loss of Bmal1 also leads to a 40% reduction of proliferation rate as assessed by EdU incorporation (Fig. 7B and C). In addition, this reduced proliferation is accompanied by a significantly increased apoptotic rate as demonstrated by quantitative TUNEL assay (Fig. 7D). It is possible that both the defect in proliferation and enhanced apoptosis in Bmal1 / primary myoblasts contributed to their substantially slower growth. Notably, the mitogenic effect of a known mitogen induced by muscle injury, Fgf2, is also diminished in Bmal1 / cells as indicated by significantly lower EdU incorporation rate (Fig. 7E and F). Together, these results from satellite cell-derived primary myoblasts indicate that Bmal1 positively regulates myoblast proliferation and growth, and this mechanism may underlie its effect on promoting satellite cell expansion during muscle regeneration.
Discussion The clock machinery has been implicated in the maintenance of muscle mass, with its disruption in mutant mice resulting in severe aging-associated sarcopenia [4,5], although the underlying
E XP ER I ME NTAL C E LL RE S E ARCH
0
15
0
Bmal1-/5 8 15
Pax7 β-actin
WT Bmal1-/-
8 Rel.Exp./36B4 4 /WT-D0
Day
WT 5 8
207
331 (2 015 ) 2 0 0 – 210
##
##
6
*,##
4
**,##
#
*,##
2 0
D0
Pax-7
Pax-7/DAPI
D5
D8
D15
Phase contrast
WT
30
40
ax-7+/DAPI (%) Pa
Numbe er of Pax-7+ /Field /
Bmal1-/-
30 20 10 0
WT Bmal1-/-
20
**
10 0
WT Bmal1-/-
Fig. 6 – Impaired satellite cell expansion during muscle regeneration in Bmal1-null mice. Immunoblot (A) and RT-qPCR analysis (B) of satellite cell marker Pax7 during CTX-induced muscle regeneration in WT and Bmal1 / mice. Protein analysis was carried out using pooled sample from 4–5 mice/time point and qPCR was performed using individual samples (n¼ 4–5). (C) Representative images of Pax7 immunofluorescence staining with DAPI nuclei stain and phase contrast images (20X), and (D) quantification of total number of Pax7þ satellite cells per field, or (E) percentage of Pax-7þ/DAPI at day 5 of regeneration. Arrows indicate representative satellite cells with Pax7þ staining co-localize with DAPI. n,nn: po0.05 and 0.01 Bmal1 / vs. WT; ##po0.01 vs. D0 control by Student's t test.
mechanisms are not clear. In the current study, we interrogated the in vivo function of the essential clock activator, Bmal1, and reveal its role as an integral component of the myogenic response to promote muscle regeneration. Furthermore, our study indicates that Bmal1 critically impacts muscle stem cell expansion, the essential step of an appropriate muscle repair process. A surprising finding from our study is that Bmal1 protein is strongly stimulated by muscle injury, with an induction kinetic that closely mimics that of the myogenic response factors. This implicates that Bmal1 could be part of a pro-myogenic network that responds to myogenic stimuli in a similar manner. Damage to skeletal muscle has been shown to cause the release of various factors that activates satellite cells and stimulates their proliferation as well as subsequent differentiation [20]. To mimic the muscle damage-induced myogenic process in vitro, soluble
fractions extracted from the crushed muscle, which was shown to contain mitogenic signals [19,26–28], were tested on myoblasts. Although various growth factors [29–31], inflammatory cytokines [32,33] or signaling molecules [11,16,34] have been reported to possess myogenic properties, the precise nature of the myogenic stimuli in vivo is likely to be a combination of varying molecules at specific phase of the regeneration process. Notably, Bmal1 is induced by CME as well as the candidate myogenic signals tested, Fgf2, Forskolin, and Wnt, suggesting that this clock activator is highly responsive to myogenic signals induced by muscle injury. Whether additional mitogenic components of CME, such as the hepatocyte growth factor [35], epidermal growth factor or insulin-like growth factor [30,31], have similar effects on Bmal1 remains to be investigated in future studies. As a molecule that is entrained by light in the
208
EX PE R IM E NTAL C ELL RE S EA RC H
WT Bmal1-/-
Growth Rate
8
WT
6
OD (450nm)
Cell Number (x106)
10
331 (2015 ) 20 0 –210
Bmal1-/-
4 2 0
0
2
4 Days
WT
6
3.5 3 2.5 2 1.5 1 0.5 0
**
0.6
**
**
**
0.4 0.2 0
Nucleasetreated
8
TUNEL Assay
0.8
Unlabeled
20K
40K 80K Cells/well
Bmal1-/-
EdU/D DAPI
EdU+/DAPI (%)
EdU U
30 25 20
10 5 0
WT+Fgf2 g
**
15
WT
Bmal1-/-
Bmal1-/- +Fgf2 g
Edu u+/DAPI (%)
EdU
60
40
**
20
dU/DAPI Ed
0
WT +Fgf2
Bmal1-/+Fgf2
Fig. 7 – Bmal1 / primary myoblast display reduced growth and proliferation with enhanced apoptosis. (A) Growth rate of satellite cell-derived primary myoblasts at indicated time in culture (n¼4). Cells were seeded at 1 106 per 10 cm plate. nnpo0.01 and 0.01 Bmal1 / vs. WT by one way ANOVA. (B, E) Cell proliferation rate as assessed by EdU labeling and staining in the absence (B), and presence of Fgf2 (E) with quantification of percentage of EdUþ cells among total nuclei (C, F). Fgf2 (25 ng/ml) was added to cells for 24 h prior to EdU labeling (E, F). (D) Apoptosis rate as analyzed by quantitative TUNEL assay in primary myoblasts isolated from WT and Bmal1 / mice with endonuclease-treated cells as positive controls (n¼3). n,nn: po0.05 and 0.01 Bmal1 / vs. WT by Student's t test.
central clock or endogenous timing cues in peripheral tissues [1,2], the responsiveness of Bmal1 to environmental signals in order to convey the timing cues has been demonstrated in various systems [7,36]. Interestingly, up-regulation of Bmal1 precedes that of the myogenic regulatory factor and embryonic myosin heavy chain by CME treatment. Given its pro-myogenic action [18], this result implies that Bmal1 could be part of a
myogenic network involved in initiation of regenerative myogenesis. Since Bmal1 is a master regulator of the core molecular clock circuit, these findings also indicate that the molecular clock could be involved in orchestrating the orderly progression of myogenic events during muscle regeneration. Future investigations of involvement of additional clock genes in muscle regeneration may shed light on this intriguing possibility.
E XP ER I ME NTAL C E LL RE S E ARCH
Using two muscle injury models, we definitively establish that Bmal1 plays a positive regulatory role in muscle regeneration. The most striking finding we observed in Bmal1-null animals is that the eMyHC protein level is barely elevated at day 3 after injury, in sharp contrast to its peak induction in the controls. This near absence of protein induction is also evident from the immunostaining result. A similarly deficient Pax7 induction further indicates that the initial phase of the satellite cell expansion and de novo new fiber formation is markedly impaired by the loss of Bmal1 function in vivo. Although we observed moderate phenotype in later stages of the regeneration, the regeneration defect in Bmal1 / mice persists, as indicated by significant reduction of regenerating myofiber caliber at 14 days post-CTX and lower regenerated muscle weight at 30 days postfreeze-injury. These findings likely reflect that Bmal1 and its temporal regulation play a more dominant role in the initial myogenic events that are important at early stages of regeneration. Muscle repair in response to injury follows a defined process, which involves an initial inflammatory infiltration, subsequent myogenic precursors (MPC) activation, proliferation, differentiation and fusion [20,37,38] leading to new myofiber formation. The expansion of MPC, comprised mostly of satellite cells, provides a major source for muscle repair [37,38]. As we previously found that Bmal1 regulates myoblast differentiation through the Wnt signaling pathway [18], we studied whether the regeneration defect observed in Bmal1-null mice is associated with attenuated Wnt signal transduction. Based on the significantly blunted expression levels of Wnt pathway genes, β-catenin, Tcf3 and established Wnt target gene Axin2, Wnt activity is compromised with the absence of Bmal1 in vivo. We demonstrated that Bmal1 exerts direct transcriptional regulation on canonical E-boxes present in β-catenin and Tcf3 promoters [18]. Therefore, loss of this Bmal1 transcriptional activity is a likely mechanism responsible for blunted Wnt signaling observed in Bmal1-null mice regeneration. A critical signal involved in embryonic myogenesis [13], Wnt activation has also been shown to be a key switch regulating satellite cell self-renewal to differentiation during muscle regeneration [11,34]. Although our study does not rule out the potential involvement of additional signaling mechanisms, these data provide evidence that the lower Wnt signaling activity we observed in vivo may contribute to the suppressed myogenic response in Bmal1-null mice regeneration. Interestingly, as Notch pathway is known to regulate satellite cell expansion phase of skeletal muscle regeneration [14] and could be under the circadian modulation of Bmal1[39]. Therefore, although beyond the scope of the current study, a potential contribution of Notch-mediated mechanism warrants future investigations. It has been increasingly recognized that the circadian clock regulates stem cell properties in various tissues, such as the hair follicle [39–41], intestine [42], the hematopoetic system [43] and neuronal organs [44]. Particularly, in epidermal and intestinal stem cells, the circadian regulation is a critical timing mechanism involved in physiological tissue remodeling [39–42]. Our results indicate that loss of Bmal1 hampers satellite cell expansion during muscle regeneration in vivo, a finding that is in line with observed reduced growth and proliferation of these cells ex vivo. Therefore, Bmal1 regulation of muscle stem cell behavior could underlie the defective regeneration seen in Bmal1-null mice. As adult skeletal muscle goes through constant repair and remodeling due to exercise-induced injuries [15], we postulate that this mechanism may mediate, at least in part, the established role of Bmal1 in
331 (2 015 ) 2 0 0 – 210
209
preserving muscle mass in adult mice. Therefore, these findings from our study indicate a novel functional circadian regulatory mechanism in the muscle stem cell population, which may participate in adult skeletal muscle remodeling.
Conclusion We demonstrate, for the first time, that a key clock regulator, Bmal1, plays a critical role in regulating muscle stem cell expansion to impact muscle regeneration. Our study implies that the intrinsic muscle clock may exert a temporal control over a cascade of intricately coordinated myogenic events to ensure the orderly progression of regenerative myogenesis. Comprehensive elucidation of this temporal regulatory mechanism involved in maintaining muscle mass may lead to discovery of targeted strategies to help preserve critical muscle mass in muscle-wasting diseases.
Grants This project is supported by Houston Methodist Research Institute start-up funds, American Heart Association Grant 12SDG12080076 and American Diabetes Association Grant 1-13-BS-118 to K. Ma.
Author contributions S.C., H.Y and D.N. designed, performed experiments and prepared figures; Y.L. carried out cardiotoxin-injury histological analysis and edited manuscript; K.M. conceived research, interpreted results of experiments, prepared figures and manuscript.
Acknowledgments We thank the Core Laboratory of Center for Diabetes Research at Houston Methodist Research Institute for their expert technical assistance, and Dr. Yong Li (UT-HSC) for technical support and proof reading of the manuscript.
references [1] C. Dibner, U. Schibler, U. Albrecht, The mammalian circadian timing system: organization and coordination of central and peripheral clocks, Annu. Rev. Physiol. 72 (2010) 517–549. [2] S.M. Reppert, D.R. Weaver, Coordination of circadian timing in mammals, Nature 418 (2002) 935–941. [3] B.H. Miller, E.L. McDearmon, S. Panda, K.R. Hayes, J. Zhang, J.L. Andrews, M.P. Antoch, J.R. Walker, K.A. Esser, J.B. Hogenesch, J.S. Takahashi, Circadian and CLOCK-controlled regulation of the mouse transcriptome and cell proliferation, Proc. Natl. Acad. Sci. USA 104 (2007) 3342–3347. [4] R.V. Kondratov, A.A. Kondratova, V.Y. Gorbacheva, O.V. Vykhovanets, M.P. Antoch, Early aging and age-related pathologies in mice deficient in BMAL1, the core componentof the circadian clock, Genes Dev. 20 (2006) 1868–1873. [5] E.L. McDearmon, K.N. Patel, C.H. Ko, J.A. Walisser, A.C. Schook, J.L. Chong, L.D. Wilsbacher, E.J. Song, H.K. Hong, C.A. Bradfield, J.S. Takahashi, Dissecting the functions of the mammalian clock protein BMAL1 by tissue-specific rescue in mice, Science 314 (2006) 1304–1308.
210
EX PE R IM E NTAL C ELL RE S EA RC H
[6] J.L. Andrews, X. Zhang, J.J. McCarthy, E.L. McDearmon, T.A. Hornberger, B. Russell, K.S. Campbell, S. Arbogast, M.B. Reid, J.R. Walker, J.B. Hogenesch, J.S. Takahashi, K.A. Esser, CLOCK and BMAL1 regulate MyoD and are necessary for maintenance of skeletal muscle phenotype and function, Proc. Natl. Acad. Sci. USA 107 (2010) 19090–19095. [7] M.K. Bunger, L.D. Wilsbacher, S.M. Moran, C. Clendenin, L.A. Radcliffe, J.B. Hogenesch, M.C. Simon, J.S. Takahashi, C.A. Bradfield, Mop3 is an essential component of the master circadian pacemaker in mammals, Cell 103 (2000) 1009–1017. [8] M. Ikeda, M. Nomura, cDNA cloning and tissue-specific expression of a novel basic helix-loop-helix/PAS protein (BMAL1) and identification of alternatively spliced variants with alternative translation initiation site usage, Biochem. Biophys. Res. Commun. 233 (1997) 258–264. [9] G. Rey, F. Cesbron, J. Rougemont, H. Reinke, M. Brunner, F. Naef, Genome-wide and phase-specific DNA-binding rhythms of BMAL1 control circadian output functions in mouse liver, PLoS Biol. 9 (2011) e1000595. [10] M. Buckingham, Myogenic progenitor cells and skeletal myogenesis in vertebrates, Curr. Opin. Genet. Dev. 16 (2006) 525–532. [11] A.S. Brack, I.M. Conboy, M.J. Conboy, J. Shen, T.A. Rando, A temporal switch from notch to Wnt signaling in muscle stem cells is necessary for normal adult myogenesis, Cell Stem Cell 2 (2008) 50–59. [12] S. Chatterjee, D. Nam, B. Guo, J.M. Kim, G.E. Winnier, J. Lee, R. Berdeaux, V.K. Yechoor, K. Ma, Brain and muscle Arnt-like 1 is a key regulator of myogenesis, J. Cell Sci. 126 (2013) 2213–2224. [13] G. Cossu, U. Borello, Wnt signaling and the activation of myogenesis in mammals, EMBO J. 18 (1999) 6867–6872. [14] S. Tsivitse, Notch and Wnt signaling, physiological stimuli and postnatal myogenesis, Int. J. Biol. Sci. 6 (2010) 268–281. [15] A.J. Wagers, I.M. Conboy, Cellular and molecular signatures of muscle regeneration: current concepts and controversies in adult myogenesis, Cell 122 (2005) 659–667. [16] R. Stewart, L. Flechner, M. Montminy, R. Berdeaux, CREB is activated by muscle injury and promotes muscle regeneration, PloS One 6 (2011) e24714. [17] B. Gayraud-Morel, F. Chretien, P. Flamant, D. Gomes, P.S. Zammit, S. Tajbakhsh, A role for the myogenic determination gene Myf5 in adult regenerative myogenesis, Dev. Biol. 312 (2007) 13–28. [18] S. Chatterjee, D. Nam, B. Guo, J.M. Kim, G.E. Winnier, J. Lee, R. Berdeaux, V.K. Yechoor, K. Ma, Brain and muscle Arnt-like 1 is a key regulator of myogenesis, J. Cell Sci. 126 (2013) 2213–2224. [19] R. Bischoff, A satellite cell mitogen from crushed adult muscle, Dev. Biol. 115 (1986) 140–147. [20] S.B. Charge, M.A. Rudnicki, Cellular and molecular regulation of muscle regeneration, Physiol. Rev. 84 (2004) 209–238. [21] N. Preitner, F. Damiola, L. Lopez-Molina, J. Zakany, D. Duboule, U. Albrecht, U. Schibler, The orphan nuclear receptor REVERBalpha controls circadian transcription within the positive limb of the mammalian circadian oscillator, Cell 110 (2002) 251–260. [22] A. Sacco, R. Doyonnas, P. Kraft, S. Vitorovic, H.M. Blau, Selfrenewal and expansion of single transplanted muscle stem cells, Nature 456 (2008) 502–506. [23] S. Kuang, M.A. Gillespie, M.A. Rudnicki, Niche regulation of muscle satellite cell self-renewal and differentiation, Cell Stem Cell 2 (2008) 22–31. [24] S. Gunther, J. Kim, S. Kostin, C. Lepper, C.M. Fan, T. Braun, Myf5positive satellite cells contribute to Pax7-dependent long-term maintenance of adult muscle stem cells, Cell Stem Cell 13 (2013) 590–601. [25] J. von Maltzahn, A.E. Jones, R.J. Parks, M.A. Rudnicki, Pax7 is critical for the normal function of satellite cells in adult skeletal muscle, Proc. Natl. Acad. Sci. USA 110 (2013) 16474–16479.
331 (2015 ) 20 0 –210
[26] K.L. Haugk, R.A. Roeder, M.J. Garber, G.T. Schelling, Regulation of muscle cell proliferation by extracts from crushed muscle, J. Anim. Sci. 73 (1995) 1972–1981. [27] G. Chen, L.S. Quinn, Partial characterization of skeletal myoblast mitogens in mouse crushed muscle extract, J. Cell. Physiol. 153 (1992) 563–574. [28] R. Bischoff, Cell cycle commitment of rat muscle satellite cells, J. Cell Biol. 111 (1990) 201–207. [29] J.P. Lefaucheur, A. Sebille, Muscle regeneration following injury can be modified in vivo by immune neutralization of basic fibroblast growth factor, transforming growth factor beta 1 or insulin-like growth factor I, J. Neuroimmunol. 57 (1995) 85–91. [30] D.R. Cook, M.E. Doumit, R.A. Merkel, Transforming growth factorbeta, basic fibroblast growth factor, and platelet-derived growth factor-BB interact to affect proliferation of clonally derived porcine satellite cells, J. Cell. Physiol. 157 (1993) 307–312. [31] M.E. Doumit, D.R. Cook, R.A. Merkel, Fibroblast growth factor, epidermal growth factor, insulin-like growth factors, and platelet-derived growth factor-BB stimulate proliferation of clonally derived porcine myogenic satellite cells, J. Cell. Physiol. 157 (1993) 326–332. [32] L. Austin, J. Bower, J. Kurek, N. Vakakis, Effects of leukaemia inhibitory factor and other cytokines on murine and human myoblast proliferation, J. Neurol. Sci. 112 (1992) 185–191. [33] J.B. Kurek, S. Nouri, G. Kannourakis, M. Murphy, L. Austin, Leukemia inhibitory factor and interleukin-6 are produced by diseased and regenerating skeletal muscle, Muscle Nerve 19 (1996) 1291–1301. [34] A.S. Brack, F. Murphy-Seiler, J. Hanifi, J. Deka, S. Eyckerman C. Keller, M. Aguet, T.A. Rando, BCL9 is an essential component of canonical Wnt signaling that mediates the differentiation of myogenic progenitors during muscle regeneration, Dev. Biol. 335 (2009) 93–105. [35] R. Tatsumi, J.E. Anderson, C.J. Nevoret, O. Halevy, R.E. Allen, HGF/SF is present in normal adult skeletal muscle and is capable of activating satellite cells, Dev. Biol. 194 (1998) 114–128. [36] A. Balsalobre, F. Damiola, U. Schibler, A serum shock induces circadian gene expression in mammalian tissue culture cells, Cell 93 (1998) 929–937. [37] F. Le Grand, M.A. Rudnicki, Skeletal muscle satellite cells and adult myogenesis, Curr. Opin. Cell Biol. 19 (2007) 628–633. [38] X. Shi, D.J. Garry, Muscle stem cells in development, regeneration, and disease, Genes Dev. 20 (2006) 1692–1708. [39] P. Janich, G. Pascual, A. Merlos-Suarez, E. Batlle, J. Ripperger U. Albrecht, H.Y. Cheng, K. Obrietan, L. Di Croce, S.A. Benitah, The circadian molecular clock creates epidermal stem cell heterogeneity, Nature 480 (2011) 209–214. [40] P. Janich, K. Toufighi, G. Solanas, N.M. Luis, S. Minkwitz, L. Serrano, B. Lehner, S.A. Benitah, Human epidermal stem cell function is regulated by circadian oscillations, Cell Stem Cell 13 (2013) 745–753. [41] M.V. Plikus, C. Vollmers, D. de la Cruz, A. Chaix, R. Ramos S. Panda, C.M. Chuong, Local circadian clock gates cell cycle progression of transient amplifying cells during regenerative hair cycling, Proc. Natl. Acad. Sci. USA 110 (2013) E2106–E2115. [42] P. Karpowicz, Y. Zhang, J.B. Hogenesch, P. Emery, N. Perrimon, The circadian clock gates the intestinal stem cell regenerative state, Cell Rep. 3 (2013) 996–1004. [43] S. Mendez-Ferrer, A. Chow, M. Merad, P.S. Frenette, Circadian rhythms influence hematopoietic stem cells, Curr. Opin. Hematol. 16 (2009) 235–242. [44] P. Bouchard-Cannon, L. Mendoza-Viveros, A. Yuen, M. Kaern H.Y. Cheng, The circadian molecular clock regulates adult hippocampal neurogenesis by controlling the timing of cell-cycle entry and exit, Cell Rep. 5 (2013) 961–973.
