HOpic

Phospho-tyrosine phosphatase inhibitor Bpv(Hopic) enhances C2C12 myoblast migration in vitro. Requirement of PI3K/AKT and MAPK/ERK pathways

Georgi A. Dimchev • Nasser Al-Shanti •
Claire E. Stewart

Received: 20 January 2013 / Accepted: 26 March 2013 / Published online: 4 April 2013
© Springer Science+Business Media Dordrecht 2013

Abstract Muscle progenitor cell migration is an impor- tant step in skeletal muscle myogenesis and regeneration. Migration is required for muscle precursors to reach the site of damage and for the alignment of myoblasts prior to their fusion, which ultimately contributes to muscle regeneration. Limited spreading and migration of donor myoblasts are reported problems of myoblast transfer therapy, a proposed therapeutic strategy for Duchenne Muscular Dystrophy, warranting further investigation into different approaches for improving the motility and homing of these cells. In this article, the effect of protein phospho-tyrosine phosphatase and PTEN inhibitor BpV(Hopic) on C2C12 myoblast migration and differentiation was investigated. Applying a
wound healing migration model, it is reported that 1 lM
BpV(Hopic) is capable of enhancing the migration of C2C12 myoblasts by approximately 40 % in the presence of

Electronic supplementary material The online version of this article (doi:10.1007/s10974-013-9340-2) contains supplementary material, which is available to authorized users.

· ·
G. A. Dimchev N. Al-Shanti C. E. Stewart
Institute for Biomedical Research into Human Movement and Health (IRM), Faculty of Science and Engineering, Manchester Metropolitan University, John Dalton Building, Oxford Road, Manchester, UK

Present Address:
G. A. Dimchev (&)
Institute of Genetics, University of Bonn, Karlrobert-Kreiten Street 13, 53115 Bonn, Germany e-mail: [email protected]

Present Address:
C. E. Stewart
Research Institute for Sport and Exercise Sciences, School of Sport and Exercise Sciences, Liverpool John Moores University, Tom Reilly Building, Byrom Street Campus, Liverpool L3 3AF, UK
myotube conditioned media, without significantly affecting their capacity to differentiate and fuse into multinucleated myotubes. Improved migration of myoblasts treated with 1 lM BpV(Hopic) was associated with activation of PI3K/ AKT and MAPK/ERK pathways, while their inhibition with
either LY294002 or UO126, respectively, resulted in a reduction of C2C12 migration back to control levels. These results propose that bisperoxovanadium compounds may be considered as potential tools for enhancing the migration of myoblasts, while not reducing their differentiation capacity and underpin the importance of PI3K/AKT and MAPK/ ERK signalling for the process of myogenic progenitor migration.

Keywords Skeletal muscle · Myoblast · Migration ·
Bisperoxovanadium · Differentiation
Abbreviations
Bpv Bisperoxovanadium CM Conditioned media
DMD Duchenne muscular dystrophy

Introduction

Migration of skeletal muscle stem and progenitor cells is an important step during both developmental and postnatal regenerative myogenesis. Myoblasts are capable of migrat- ing towards areas of muscle injury and regeneration (Watt et al. 1994) and the process of migration is also a key step in the alignment of myoblasts for their subsequent fusion into multinucleated myotubes (Leloup et al. 2006; Vaz et al. 2012). Importantly, myoblast transfer therapy, a proposed strategy for restoring normal dystrophin expression by

injection of donor myoblasts into dystrophic muscles, has failed to achieve satisfactory results in clinical trials, partly because of limited spreading/migration of injected myo- blasts (Fan et al. 1996; Gussoni et al. 1997; Moens et al. 1996; Skuk et al. 2004; Smythe et al. 2001). Although dif- ferent stem cells have been demonstrated to be able to con- tribute to skeletal muscle regeneration and recent advances in gene therapies have emerged as promising strategies to treat muscle myopathies (Cirak et al. 2011; Meng et al. 2011; Peault et al. 2007), myoblast transplantation could also provide a useful strategy to improve and restore muscle function following injury or with ageing and disuse. This encourages the development of strategies to enhance the migration of donor myoblasts into muscle and to reveal in more detail the regulatory mechanisms driving this process. Different aspects of myogenic cell migration have been investigated. A number of articles have reported the ben- eficial role of various growth factors for inducing chemo- taxis or migration of either satellite cells or myoblasts (Amano et al. 2002; Bischoff, 1997; Corti et al. 2001; Lafreniere et al. 2004; Ranzato et al. 2009; Torrente et al. 2003). The importance of matrix degrading enzymes for the migration of myogenic cells has also been revealed (Allen et al. 2003; El Fahime et al. 2000; Lewis et al. 2000), as have signalling pathways regulating the migra- tion of myogenic cells, such as PI3K and MAPK, although results have been conflicting (Al-Shanti et al. 2011; Kawamura et al. 2004; Kim et al. 2011; Leloup et al. 2007; Ranzato et al. 2009; Suzuki et al. 2000). In order to improve our understanding of and therefore ultimately the migration of myogenic cells in vivo, a combination of different strategies might be required, such as the presen- tation of different soluble factors, genetic modifications of the donor cells, or the modulation of intracellular signalling pathways, warranting further investigation into the intra- and extracellular regulators of myogenic cell migration. Developing tools to enhance the migration of muscle pro- genitors could have important implications in proposed therapies for muscle myopathies and could also contribute to a better understanding of the process of skeletal muscle
myogenesis and regeneration throughout the lifespan.
Bisperoxovanadium (BpV) compounds are well estab- lished phospho-tyrosine phosphatase (PTPase) inhibitors, acting as PTEN inhibitors at low concentrations (Lai et al. 2007; Schmid et al. 2004). Lower concentrations of BpV compounds have been shown to induce cell migration of non-myogenic cells (Lai et al. 2007; Mihai et al. 2012). Castaldi et al. demonstrated that C2C12 myoblasts treated
with 10 lM BpV displayed reversible inhibition in their
myogenic differentiation capacity, as shown by reduced myotube formation and downregulation of myogenic reg- ulatory factors (Castaldi et al. 2007). Additionally, 10 lM BpV-induced markers of circulatory progenitors in C2C12
myoblasts and improved their recruitment, via the circu- lation, in a murine model of dystrophic muscle. However, the effects of BpV compounds on myogenic cell motility have not yet been directly investigated. It was hypothesised that PTP and PTEN inhibitor BpV at lower concentrations might enhance the migration of myoblasts, while not negatively influencing the myogenic differentiation
capacity of these cells. In this article, it was tested whether 1 lM BpV(Hopic) influenced the differentiation capacity of C2C12 myoblasts and whether it was capable of enhancing their migration. The effect of 1 lM BpV(Hopic) on PI3K/AKT and MAPK/ERK pathways and their requirement for the effective migration of myogenic cells in response to the inhibitor was also assessed.

Materials and methods

Cell culture reagents

Sterile Lonza BioWhittaker Dulbecco’s Modified Eagle’s Media (DMEM) w/4.5 g Glucose per Liter, w/L-Glutamine (584 mg/L) was purchased from BioWhittaker (Wokingham, UK). Heat-inactivated (hi) foetal bovine serum (FBS) was purchased from Gibco (Paisley, UK) and hi horse serum— from Southern Group Laboratory (Corby, UK). Sterile peni- cillin–streptomycin solution and 109 trypsin solution were purchased from BioWhittaker (Wokingham, UK). L-gluta- mine was purchased from BDH (Poole, Dorset, UK) and Gelatin type A from porcine skin was purchased from Sigma– Aldrich Chemie (Steinheim, Germany). Phosphate buffered saline (PBS) tablets were from Oxoid Ltd. (Basingstoke, UK).

Inhibitors, chemicals and solvents

All general chemicals and solvents (analytical (AnalaR) or molecular biology/tissue culture grade) were purchased from Sigma (Dorset, UK) and BDH (Poole, Dorset, UK), unless otherwise specified. Inhibitors were purchased from Merck–Calbiochem (Darmstadt, Germany).

Cell culture treatments

Cells were induced to proliferate in growth media, con- taining 10 % foetal bovine serum (FBS) DMEM media, 1 % Pen–strep and 200 mM L-glutamine. Myoblasts were induced to differentiate and to fuse into myotubes, once high confluency was achieved, upon withdrawal of growth media and addition of differentiation media (2 % horse serum, DMEM media, 1 % Pen–strep and 200 mM L-glu- tamine). For all experiments, cells were seeded on flasks or multi-well dishes (Nunc Life Sciences, Thermo Fisher Scientific; Rockslide, Denmark), pre-coated for 15 min at

room temperature with 0.2 % Gelatin type A from porcine skin.

Creatine kinase assay

Samples for CK assays were prepared by lysing cells with TMT buffer (0.05 M Tris/Mes and 1 % Triton-X100). CK assay kits were purchased from Catachem Inc. (Connecti- cut, NE, USA). Reagents of the kit were reconstituted according to manufacturer’ s instructions and loaded in a 96 well UV plate from BD Biosciences (San Jose, CA, USA). The plate was incubated for 3 min on a shaker at RT before absorbance activity of CK was measured in 2 min intervals on a Bio-Tek ELISA plate reader (Winooski, VT, USA) at 3–9 min post reaction initiation at a wavelength of 340 nm. Changes in absorbance/min were calculated prior to normalisation against total protein content, as deter- mined by the BCATM assay (Pierce, Rockford, IL, USA).

Conditioned media generation

C2C12 myoblasts were grown to high confluency in the presence of growth media and were induced to differentiate and fuse into myotubes by transfer to differentiation media. Differentiation media was replenished on day 3 and cells cultured for two more days, thus allowing C2C12 myo- blasts 5 days in total to differentiate and form large myo- tubes. On day 5, differentiation media was discarded, cell monolayers were washed 3 times with PBS, incubated for 3 h in serum-free DMEM media (1 % Pen–strep and
200 mM L-glutamine) at 37 °C/5 % CO2 and washed three
more times in PBS, in order to reduce the possibility of residual serum contaminants. Fresh serum-free DMEM media (1 % Pen–strep and 200 mM L-glutamine) was added to the myotube cultures, which were incubated at
37 °C/5 % CO2 for 3 days and the resultant conditioned
medium was collected, filtered through 0.22 lm syringe filter (Corning Life Sciences; Lowell, MA, USA) to
remove cell debris and incubated at 4 °C until further use.

Wound healing assay

Cells were seeded in pre-gelatinised six well plates at 80,000 cells/ml in growth media. Following 24 h incuba- tion, once *80 % cell confluency was reached, growth media was removed, cell monolayers were washed with PBS and 0.1 % FBS DMEM media (quiescent media), containing 1 % Pen–Strep and 200 mM L-glutamine, was added. Cells were incubated for 20 h in quiescent media
before a wound was inflicted (of *600 lm), using a
sterilised pipette tip. Following three washes with PBS, conditioned media with or without inhibitors were added to each well and live imaging examination of the wound
closure was initiated. To study migration in the absence of conflicting proliferation, cell monolayers were pre-incu- bated for 3 h in 10 lg/ml of Mitomycin-C (Sigma–Aldrich (Dorset, England)) in quiescent media before wound
healing and live imaging was initiated. Photomicroscopy and live imaging
Cell microscopy and imaging were performed using Leica Microsystems (Wetzlar, Germany) CMS GMbH light microscope (Leica DMI6000B) with digital photography, recording capabilities (Leica CTR 6000), using Leica Application Suite software (Wetzlar, Germany). Live imag- ing microscopy was performed using the Leica DMI6000B microscope equipped with an autoflow incubator, CO2
controller, heating unit and temperature controller, thus allowing maintenance of 37 °C/5 % CO2 atmosphere. Videos were generated and exported in .AVI format for further analysis using the Leica Application Suite software (Wetzlar, Germany).

Analysis of cell migration

Migration analyses of the total number of cells migrating into a wound were performed using Keynote’09 software version 5.1.1. (Apple Inc.), which was employed to draw lines into the first frame of the exported wound healing videos, outlining the initial wound edges. Preserving these lines, images were taken of the last frame of each video and ImageJ v1.45i software was used to count the number of cells present within the outlined wound. For each experi- ment, at least six representative videos were analysed per condition, taken from at least two duplicate wells. Each experiment was repeated a minimum three times.
Cell movement trajectories were created with Key- note’09 software version 5.1.1. (Apple Inc.) by manually marking frame by frame the migration trajectories of cells into exported.AVI videos. Trajectories of 16 randomly selected leading cells of the wound edges (eight from each edge of the wound) were created, distance of migration
trajectories was established by ImageJ v1.45i software (normalised against a lm marker) and average cell migration distance per movie was calculated. For each experiment, at least six representative videos were analysed
per condition, taken from at least 2 duplicate wells. Each experiment was repeated a minimum of three times.
Cell movement velocity and directionality were ana- lysed by ImageJ and Ibidi Chemotaxis and Migration Tool Version 1.01 (authors: Gerhard Trapp and Elias Horn) plugin for ImageJ. Directionality/directness was calculated by comparing the Euclidian distance to the Accumulated distance (values were represented from 0 to 1, as a value of 1 would be equivalent to linear movement).

RNA extraction and real-time pcr

Extraction of RNA was performed using Trizol® reagent (Invitrogen, Life Technologies, Paisley, UK). RNA con- centrations and purity were assessed using a Biotech Photometer (WPA UV1101, Biochrom, Cambridge, UK).
TaqMan® RNA-to-CTTM 1-Step Kit was utilised
(Applied Biosystems, San Jose, CA, USA). Total RNA amount of 85 ng was loaded per reaction. Applied Biosys- tems StepOnePlus Real-time PCR system with StepOne Software v2.2.2 (Applied Biosystems, Life Technologies)
was used with the following PCR program: 1) 15 min at 48 °C; 2) 10 min at 95 °C; 3) 40x(15 s at 95 °C, 1 min at 60 °C). The following predesigned primer sets were ordered from Applied Biosystems and were used with the TaqMan® detection method: Polr2b (Mm00464214_m1), MyoD
(Mm00440387_m1), Myogenin (Mm00446194_m1). Rela- tive quantification of gene expression was determined by the delta–delta Ct method. All samples were normalised to a housekeeping gene (Polr2b). Values were represented as fold change relative to a common calibrator sample.

SDS-PAGE

Cells, as cultured for wound healing assays (described above), were stimulated by conditioned media with or without inhibitor treatments. 30 min post treatment, cells were lysed with lysis buffer (10 mM TrisCl, 5 mM EDTA,
50 mM NaCl, 30 mM Na4P207, 50 mM NaF, 100 lM
Na3VO4, 1 mM PMSF and 1 % Triton X-100, pH 8.3).
Protein concentrations were determined using BCATM Protein Assay (Pierce (Rockford, IL, USA)). Sample con- centration was adjusted to 30 lg, samples loaded on an
8.5 % polyacrylamide gel and transferred to Hybond ECL Nitrocellulose membranes, purchased from GE Healthcare Life Sciences, (Buckinghamshire, UK). HRP substrate used was West Dura Supersignal kit (Pierce, Rockford, IL, USA). Chemiluminescence was detected on Biorad Imag- ing System (Hercules, CA, USA) and densitometry performed on Quantity One Software 4.6.2 (Bio-Rad, Hercules, CA, USA). Values were represented as percent- age change to a common calibrator sample loaded on each gel.
Primary AKTpan, pAKT (Ser473), beta-actin and ERKpan rabbit IGg antibodies were purchased from New England Biolabs (Hertfordshire, UK) and were incubated with membranes overnight (RT) at 1:1,000, diluted in 5 % BSA/TBST. Primary pERK1/2 (V803A, Promega) rabbit antibody was incubated with membranes overnight at 1:5,000 concentration, diluted in 0.1 % BSA/TBST. Sec- ondary Goat Anti-rabbit HRP-conjugated antibodies were purchased from MP Biomedicals (United Kingdom) and were incubated with membranes for 1 h at RT at 1:5,000 in
blocking solution. Membranes were blocked for 1 h at RT in 5 % non-fat dry milk/TBST (for antibodies purchased from New England Biolabs) or in 1 % BSA/TBST (for pERK1/2 antibody).

Statistics

Statistical analyses and significance of data were deter- mined using GraphPad Prism version 5.00 for Mac OS X (GraphPad Software, San Diego California USA, www. graphpad.com). Statistical significance for interactions between two paired groups was determined with a Paired t test. Statistical significance for interactions between more than two groups was determined by one-way repeated measures ANOVA with Bonferroni post hoc analysis. All results are presented as mean ± standard error of the mean (SEM). All values below p \ 0.05 were considered as significant.

Results

Myogenic differentiation is suppressed by 10 lM, but not by 1 lM Bpv(Hopic)

Previous studies by Castaldi et al. demonstrated the inhi- bition of myogenic differentiation of C2C12 myoblasts, treated with high concentrations of Bisperoxovanadium (10 lM) (Castaldi et al. 2007). The ability of myogenic
cells to initiate the myogenic differentiation program and fuse, following migration to the injury site, would be essential for successful muscle repair. Therefore, we investigated whether the presence of lower concentrations of BpV, would affect the capacity of C2C12 myoblasts to differentiate and fuse into multinucleated myotubes. C2C12
myoblasts were grown to confluence and treated with either differentiation media (DM), DM ? 1 lM BpV(Hopic) or DM ? 10 lM BpV(Hopic). As shown in Fig. 1A, 72 h post treatment, 10 lM Bpv(Hopic) significantly reduced myo- tube formation, as compared to control DM, while 1 lM Bpv(Hopic) did not visually cause inhibition of myotube formation, as compared to control DM. These results were further substantiated by measuring the enzymatic activity of Creatine Kinase (CK), demonstrating no significant differ- ence in the CK activity of cells treated for 72 h with DM versus DM ? 1 lM Bpv(Hopic) (Fig. 1B). On the other hand, significantly lower levels of CK were indicated in cells treated with DM ? 10 lM Bpv(Hopic) versus DM (352.6 ± 15.2 vs 791.8 ± 37.1 ng.ml-1, p \ 0.001, equiv-
alent to 2.25 fold decrease) and versus DM ? 1 lM
Bpv(Hopic) (352.6 ± 15.2 vs 768.4 ± 22 ng.ml-1, p \ 0.001, equivalent to 2.18 fold decrease). MyoD expression levels remained unchanged between cells treated for 24 h

Fig. 1 C2C12 myoblast differentiation is significantly reduced by 10 lM, but not by 1 lM BpV(Hopic) treatment. C2C12 myoblasts were grown to high confluency and triggered to differentiate in the presence of DM, supplemented with 1 lM or 10 lM BpV(Hopic). Differentiation capacity of C2C12 myoblasts treated with 10 lM Bpv(Hopic) was significantly reduced, while treatment of myoblasts with 1 lM Bpv(Hopic) did not significantly affect the differentiation capacity of myoblasts, as compared to control DM, 72 h post treatment. Phase-contrast images demonstrate visually unchanged myotube formation between C2C12 cells treated for 72 h with DM versus DM ? 1 lM Bpv(Hopic), while cells treated with DM ? 10 lM Bpv(Hopic) had clearly impaired myotube formation
(A). Creatine kinase assays indicated significantly lower levels of CK activity in C2C12 cells treated with DM ? 10 lM Bpv(Hopic) versus cells treated with either DM or DM ? 1 lM Bpv(Hopic) (p \ 0.001, represented by *) (B). MyoD gene expression levels, 24 h post treatment, were significantly lower in C2C12 cells treated with DM ? 10 lM Bpv(Hopic) versus cells treated with either DM or DM ? 1 lM Bpv(Hopic) (p \ 0.01, represented by *) (C). Myogenin gene expression levels, 24 h post treatment, are significantly lower in C2C12 cells treated with DM ? 10 lM Bpv(Hopic) versus cells treated with either DM or DM ? 1 lM Bpv(Hopic) (p \ 0.001, represented by *) (D). n = 5; DM Differentiation media

with DM or DM ? 1 lM BpV(Hopic), but were signifi- cantly lower when cells were treated with DM ? 10 lM Bpv(Hopic) versus either DM (2.4 fold decrease, p \ 0.01) or versus DM ? 1 lM BpV(Hopic) (2.2 fold decrease, p \ 0.01) (Fig. 1C). Myogenin expression levels were also significantly lower when C2C12 cells were treated with DM ? 10 lM BpV(Hopic) versus either DM only (7.74 fold decease, p \ 0.001) or DM ? 1 lM Bpv(Hopic) (6.3 fold decrease, p \ 0.001) (Fig. 1D).

1 lM Bpv(Hopic) enhances migration of C2C12 myoblasts and is associated with activation of PI3K/ AKT and MAPK/ERK signalling pathways

Having established that myoblasts are capable of fusing and initiating the myogenic differentiation program in the
presence of 1 lM BpV(Hopic), we tested whether their migration would be enhanced, as hypothesised, following treatment with the same inhibitor concentration. Migration
of C2C12 myoblasts was assessed in the presence of condi- tioned medium (CM) with or without 1 lM BpV(Hopic) treatment. In preliminary experiments in our group (data not shown), myotube conditioned media was demonstrated to
enhance migration of myoblasts in wound healing models and chemotaxis in transwell insert assays, as compared to control serum-free media and was thus adopted as a positive stimulus for migration. Migration in the presence of CM or
1 lM BpV(Hopic) was analysed in wound healing assays
following 20 h stimulation and utilising two approaches: one) counting the total number of cells infiltrating the wound and two) calculating the average migration distance of the leading cells at the wound edge. As shown in Fig. 2A, while

Fig. 2 1 lM Bpv(Hopic) enhances C2C12 myoblast migration in a wound healing assay model. Wound healing was visually enhanced when
C2C12 myoblasts were treated with 1 lM Bpv(Hopic), in the presence of myotube conditioned media, with greater
numbers of cells infiltrating the wound (top panels, a) and also enhanced migration distance (bottom panels, a) with inhibitor versus control CM (A). These observations were confirmed by analyses of the total number of cells infiltrating the wound, indicating significantly higher
number with 1 lM Bpv(Hopic)
versus control CM (p \ 0.0001, represented by *) (B) and also by analyses of the migration distance, indicating higher capacity of myoblasts to migrate
when treated with 1 lM
Bpv(Hopic) versus control CM (p \ 0.0001, represented by *) (C). Furthermore, myoblasts treated with 1 lM Bpv(Hopic) migrated with higher velocity,
as compared to control CM treated myoblasts during both 0–10 h (p \ 0.0001,
represented by *) and 10–20 h (p \ 0.0001, represented by *). Velocity was slightly higher with BpV treatment in the period of 0–10 h of migration, as compared to 10–20 h
(p \ 0.0001, represented by #),
unlike CM treatment, where migration velocity remained unchanged between these two time periods analysed (D).
Analysis of the directionality of cell movement, indicated that BpV treated cells in a wound healing assay model was slightly reduced, as compared to control CM treated cells
(p \ 0.05) (E). n = 14
experiments, performed in duplicate (A, B, C); n = 416 cells tracked throughout 13 experiments performed in duplicate (D, E).
CM Conditioned media

efficient migration was evident in the presence of CM, this migration was visually improved following treatment with 1 lM Bpv(Hopic) treatment. Analyses of the total number of
cells infiltrating the wound (Fig. 2B) indicated a 38 % increase in infiltrating cell numbers with CM ? 1 lM Bpv (Hopic) versus CM alone (91 ± 4 vs 66 ± 3, >p \ 0.0001).

Analysis of the migration distance indicated 41 % increase with CM ? 1 lM Bpv(Hopic) versus CM alone (598 ± 47.4 vs 422 ± 33.2 lm, p \ 0.0001) (Fig. 2C). Analysis of migration velocity supported these observations, with sig- nificantly higher velocities of migration observed with CM ? 1 lM Bpv(Hopic) versus CM alone (0.4 ± 0.005 vs
0.29 ± 0.004 lm/min, p \ 0.0001; Fig. 2D). Directionality of cell movement between the two treatments over 20 h remained relatively unchanged, however due to the high number of cell trajectories analysed (n = 416), a very small, but statistically significant reduction in directionality was indicated with BpV treatment, as compared to CM treatment
(0.73 ± 0.005 vs 0.755 ± 0.006, p \ 0.05; Fig. 2E). Thus,
enhanced wound closure by 1 lM BpV(Hopic) treatment was caused by enhanced cell velocity, but not by BpV- induced changes in cell directionality.
Following investigation of the migration capacity of C2C12 myoblasts in response to 1 lM Bpv(Hopic), acti- vation of PI3K/AKT and MAPK/ERK signalling pathways was assessed. SDS-PAGE was performed, investigating
the phosphorylation of pAKT (Ser 473) and pERK1/2 at 30 min following treatment of C2C12 cells with either CM or CM ? 1 lM BpV(Hopic). Results revealed greater activation of both PI3K/AKT and MAPK/ERK pathways
(in the face of unchanged total Akt and ERK) following 1 lM Bpv(Hopic) treatment, as compared to CM alone (Fig. 3).

Inhibition of PI3K/AKT and MAPK/ERK pathways with LY294002 and UO126 reduced the migration induced by Bpv(Hopic)

In order to investigate whether PI3K/AKT and MAPK/ ERK activation, following 1 lM Bpv(Hopic) treatment, are required for the enhanced myoblast migration induced by the inhibitor, cells were treated with PI3K inhibitor LY294002 (10 lM) or with MEK inhibitor UO126 (5 lM), in the presence of 1 lM BpV(Hopic).
C2C12 myoblasts were incubated with CM ? vehicle (DMSO), CM ? 1 lM BpV(Hopic) ? vehicle or CM ? 1 lM BpV(Hopic) ? 10 lM LY294002 and lysates for SDS-PAGE were collected at 30 min post stimulation. As anticipated, results demonstrated inhibition of the PI3K/ AKT pathway to CM levels, following treatment with 10 lM LY294002, without significant effect on the MAPK/ERK pathway, which remained activated in the presence of 1 lM BpV(Hopic) (Fig. 4).
The effect of PI3K/AKT inhibition in the presence of 1 lM BpV(Hopic) on myoblast migration was investigated (Fig. 5). Analyses of the total number of cells infiltrating the wound 20 h post stimulation indicated significant increases with CM ? 1 lM BpV(Hopic) versus CM alone (90 ± 5 vs 65 ± 3, p \ 0.001, equivalent to 38 % increase), similar to
previous results. However, when 10 lM LY294002 was co- incubated with 1 lM BpV(Hopic), the total number of cells infiltrating the wound was significantly reduced (90 ± 5 vs 66 ± 3, p \ 0.01, equivalent to 27 % decrease; Fig. 5B) and approached control CM levels. Similarly, examination of the migration distance of cells at the leading edge of the wound, indicated an increase when myoblasts were incubated with CM ? 1 lM BpV(Hopic) versus CM (678 ± 63.1 vs 463 ± 49.4 lm, p \ 0.0001, equivalent to 46 % increase). The increase in migration distance caused by 1 lM BpV(Hopic) was reduced to baseline levels after co-incubation with 10 lM LY294002 (678 ± 63.1 vs. 521 ± 54.3 lm, p \ 0.001, equivalent to 23 % decrease; Fig. 5C). These results suggest that PI3K/ AKT signalling is required for the enhanced, rather than basal myoblast migration following 1 lM BpV(Hopic) treatment.
To examine whether MAPK/ERK signalling is also required for the enhanced migration of myoblasts induced by BpV(Hopic), cells were treated with 5 lM UO126, co- incubated with 1 lM BpV(Hopic). Lysates for SDS-PAGE were collected at 30 min following treatment with either CM ? vehicle (DMSO), CM ? 1 lM BpV(Hopic) ? vehicle or CM ? 1 lM BpV(Hopic) ? 5 lM UO126. SDS-PAGE results revealed that 5 lM UO126 blocked the increase in pERK1/2 levels, previously shown to be caused by 1 lM BpV(Hopic) treatment. Furthermore, 5 lM UO126 treatment did not significantly affect the levels of pAKT (Ser 473), which remained activated in the presence of 1 lM BpV(Hopic) (Fig. 6) again suggesting specificity of inhibition.
Analyses of the total number of cells infiltrating the wound and migration distance were performed 20 h post stimulation with control CM ? vehicle, CM ? 1 lM BpV(Hopic) ? vehicle and CM ? 1 lM BpV(Hopic) ?
5 lM UO126 (Fig. 7). Increases in total cell numbers were indicated with CM ? 1 lM BpV(Hopic) versus CM (97 ± 7 vs 68 ± 3, p \ 0.0001, equivalent to 43 %
increase), with reduction back to control CM levels with CM ? 1 lM Bpv(Hopic) ? 5 lM UO126 versus CM ? 1 lM BpV(Hopic) (68 ± 3 vs 97 ± 7, p \ 0.0001, equiv- alent to 30 % decrease; Fig. 7B). Similarly, analyses of the migration distances of the cells at the leading edge indi- cated significant increase with CM ? 1 lM BpV(Hopic) versus CM (573 ± 72.6 vs 413 ± 44.4 lm, p \ 0.01, equivalent to 39 % increase), which was decreased with CM ? 1 lM Bpv(Hopic) ? 5 lM UO126 (573 ± 72.6 vs
346 ± 42.1 lm, p \ 0.0001, equivalent to 40 % decrease;
Fig. 7C). Thus, results suggest that MAPK/ERK signalling is also required for the enhanced myoblast migration following 1 lM BpV(Hopic) treatment and that neither ERK1/2, nor PI3K activation can compensate for the other in enhancing basal migration in the presence of BpV(Hopic).

Fig. 3 1 lM BpV(Hopic) activates PI3K/AKT and MAPK/ERK signalling. SDS-PAGE revealed increased activation of both PI3K/ AKT and MAPK/ERK signalling pathways 30 min post treatment with 1 lM BpV(Hopic) (A). Densitometric and statistical analyses indicated significant increase in the levels of pAKT (Ser 473) in cells
treated with 1 lM BpV(Hopic) versus control CM (p \ 0.001, represented by *) (B) and also higher levels of pERK1/2 with 1 lM BpV(Hopic) versus control CM (p \ 0.01, represented by *) (C). n = 4experiments. CM Conditioned media; Bpv = 1 lM Bpv(Hopic)

Fig. 4 LY294002 was applied to block PI3K/AKT activation induced by 1 lM Bpv(Hopic). SDS-PAGE revealed that 10 lM LY294002 is capable of inhibiting the activation of PI3K/AKT signalling caused by 1 lM BpV(Hopic) treatment, without signifi- cantly affecting the MAPK/ERK signalling (A). Densitometric and statistical analyses indicated significant increases in the levels of pAKT (Ser 473) in cells treated with 1 lM BpV(Hopic) versus control CM (p \ 0.001, represented by *), which was significantly inhibited when 1 lM BpV(Hopic) was co-incubated with 10 lM
LY294002 (p \ 0.001, represented by #). (B) Analyses of the levels of pERK1/2 indicated significant increase in cells treated with 1 lM BpV(Hopic) versus CM (p \ 0.001, represented by *) and with 1 lM BpV(Hopic) ? 10 lM LY294002 versus CM (p \ 0.001, represented by #). pERK1/2 levels were not significantly affected by treatment with LY294002 in the presence of 1 lM BpV(Hopic) (C). n = 3 experiments. CM Conditioned media; Bpv = 1 lM Bpv(Hopic); Bpv ? LY = 1 lM Bpv(Hopic) ? 10 lM LY294002

Discussion

In this study, it was demonstrated that C2C12 myoblasts treated with 1 lM BpV(Hopic) displayed enhanced migra- tion in the presence of myotube conditioned media, without
this treatment exerting a negative impact on myogenic differentiation. Strategies for enhancing the migration of donor myoblasts, while not compromising their myogenic differentiation and fusion, might be particularly relevant to proposed therapies for DMD, where one of the limiting factors is inefficient myoblast migration. Results obtained in recent articles, however, question the significance of
migration for the success of myoblast transfer therapy (Lafreniere et al. 2009; Skuk et al. 2011). Coinjection of growth factors improved migration of donor myoblasts in nonhuman primates, but myoblasts failed to fuse with undamaged fibres and thus the enhanced migration did not improve the transplantation success (Lafreniere et al. 2009). Similarly, Skuk et al. (2011) demonstrated, using monkey models, that transplanted myoblasts were capable of migrating several millimetres, but failed to fuse with undamaged myofibers (Skuk et al. 2011). Due to ethical reasons, the generation of primates with dystrophin defi- ciency is not possible and therefore the question still remains

Fig. 5 LY294002 blocks migration enhanced by 1 lM BpV(Hopic). Wound healing assay, examining total cell number infiltrating the wound (top panels, a) and migration distance (bottom panels, a), demonstrated that C2C12 cell migration, enhanced by 1 lM BpV(Ho- pic) in the presence of CM, was reduced when PI3K/AKT signalling was inhibited by 10 lM LY294002 (A). Analyses of the total number of cells infiltrating the wound site indicated significant increases following 1 lM BpV(Hopic) treatment, as compared to control CM (p \ 0.001, represented by *). The increase in cell numbers was abolished when
BpV(Hopic) was co-incubated with 10 lM LY294002 (p \ 0.01, represented by #) (B). Similarly, migration distance of cells at the leading edge was significantly increased following 1 lM BpV(Hopic) treatment, as compared to control CM (p \ 0.0001, represented by *) and this increase was abolished when BpV(Hopic) was co-incubated with 10 lM LY294002 (p \ 0.001, represented by #) (C). n = 6 experiments performed in duplicate. CM Conditioned media; Bpv = 1 lM Bpv(Hopic); Bpv(Hopic) ? LY = 1 lM Bpv(Hopic) ? 10 lM LY294002

of whether enhanced migration of donor myoblasts in dys- trophic muscle, where extensive muscle fibre damage occurs, would lead to improved transplantation success. Importantly, however, one should consider that only enhancing migration would not be sufficient for improving the transplantation success. Enhanced capacity of myoblasts to migrate should also be coupled with enhanced ability of these cells to fuse with resident myofibres. Tools for enhancing donor myoblast migration may indeed lead to improvement in myoblast transfer therapy, but it must be considered that this should be
combined with other strategies for improving the engraftment of these cells. Importantly, the effect of exogenous com- pounds on other cell types within the muscle would also require evaluation. Thus, for example, BpV could have a stimulatory effect on myogenic cell migration, but being a PTEN inhibitor, it might also affect the process of myofibro- blast differentiation to promote fibrosis (White et al. 2006), an unwanted impact in already damaged muscle in DMD patients. This warrants further investigation in vivo, if BpV(Hopic), or other bisperoxovanadium compounds, are to

Fig. 6 UO126 was applied to block MAPK/ERK activation induced by 1 lM Bpv(Hopic). SDS-PAGE revealed that 5 lM UO126 was capable of inhibiting the activation of MAPK/ERK signalling caused by 1 lM BpV(Hopic) treatment, without significantly affecting the PI3K/AKT signalling (A). Densitometric and statistical analyses of pERK1/2 indicated significant increases in cells treated with 1 lM BpV(Hopic) versus control CM (p \ 0.001, represented by *), which was significantly inhibited when 1 lM BpV(Hopic) was co-incubated
with 5 lM UO126 (p \ 0.001, represented by #) (B). Analyses of the levels of pAKT (Ser473) indicated significant increase in cells treated with 1 lM BpV(Hopic) versus CM (p \ 0.05, represented by *) and with 1 lM BpV(Hopic) ? 5 lM UO126 versus CM (p \ 0.05, represented by #). pAKT levels were not significantly affected by treatment with 5 lM UO126 in the presence of 1 lM BpV(Hopic) (C). n = 4 experiments. CM Conditioned media; Bpv = 1 lM Bpv(Hopic); Bpv ? UO = 1 lM Bpv(Hopic) ? 5 lM UO126

be considered as a potential strategy for enhancing donor myoblast migration during muscle repair. Furthermore, clear understanding is emerging in the cell migration field that migration on flat two-dimensional (2D) surfaces may differ from cell motility in three-dimensional (3D) environments (Petrie and Yamada, 2012). Therefore, further analysis is required to evaluate the effects of bisperoxovanadium com- pounds on myoblast migration in physiologically relevant, in vivo models.
In addition to the enhanced migration elicited by 1 lM
BpV(Hopic), an increase in the levels of pAKT and pERK1/2 phosphorylation was also observed. Considering the inhibi- tory effect of BpV on PTEN, these results are in agreement with the established role of PTEN as an upstream negative effector of PI3K and MAPK pathways (Georgescu, 2010; Gu et al. 1998). PI3K/AKT and MAPK pathways have been investigated in the context of myoblast migration by a number of groups, but results have been divergent. Some articles have reported requirement of PI3K signalling on motility or chemotaxis of myogenic precursors (Al-Shanti et al. 2011; Kawamura et al. 2004; Kim et al. 2011; Ranzato et al. 2009), while others reported no effects of PI3K inhi- bition on myoblast migration (Leloup et al. 2007; Suzuki et al. 2000). Similarly, some have reported requirement for MAPK signalling for migration of myogenic cells (Al-Shanti et al. 2011; Leloup et al. 2007), while others reported no effect of MAPK inhibition on myoblast migration (Ranzato et al. 2009; Suzuki et al. 2000). This inconsistency in the results obtained is possibly due to different culture and experimental conditions, suggesting the importance of the balance in factors regulating migration and possibly the
existence of some redundancy between signalling pathways regulating myogenic cell migration. As 1 lM BpV(Hopic) induced activation of both PI3K and MAPK pathways, and
possibly other signalling pathways not examined here, it was investigated whether individual inhibition of either PI3K/ AKT or MAPK/ERK would affect the migration of C2C12 myoblasts. It was found that in the presence of 1 lM BpV(Hopic), inhibiting either PI3K/AKT or MAPK/ERK
pathways alone, resulted in a reduction in the migration of myoblasts to control levels, regardless of the elevated activity of the other. These results suggest that enhanced activation of both pathways is required for supplementary migration induced by BpV(Hopic) and that activation of one pathway alone is not sufficient to maintain enhanced migration. It is, however, of interest to note that when either PI3K/AKT or MAPK/ERK activation is suppressed, basal migration is not inhibited, suggesting that for baseline myoblast migration, compensation is probably sufficient to enable migration. Therefore, both PI3K/AKT and MAPK/ ERK activation appear essential for BpV(Hopic)-induced supplementation of myoblast migration, but basal migration can still occur when one of them is pharmacologically inhibited. A recent article by Mihai et al. published in the course of this study, has demonstrated that PTEN inhibition and BpV(Phen) treatment enhances migration of epithelial cells and also demonstrated that BpV(Phen) altered the mechanical properties of epithelial cells, such as cell stiff- ness, which was mediated by both Akt and ERK pathways, ultimately leading to an enhanced migratory phenotype (Mihai et al. 2012). It is entirely possible that similar mechanisms act on skeletal muscle myoblasts as well. This would explain the effect of BpV(Hopic) and the requirement for PI3K/Akt and MAPK/ERK pathways on their enhanced migration in the face of BpV(Hopic).
In conclusion, it was revealed that 1 lM BpV(Hopic) treatment was capable of increasing the in vitro migration of C2C12 myoblasts, without significantly reducing their

Fig. 7 UO126 blocks migration enhanced by 1 lM BpV(Hopic). Wound healing assay, examining total cell number infiltrating the wound (top panels, a) and migration distance (bottom panels, a), demonstrated that C2C12 cell migration, enhanced by 1 lM BpV(Hopic) in the presence of CM, was reduced when MAPK/ ERK signalling was inhibited by 5 lM UO126 (A). Analyses of the total number of cells infiltrating the wound site indicated significant increases following 1 lM BpV(Hopic) treatment, as compared to control CM (p \ 0.0001, represented by *). This increase in cell
numbers was abolished when BpV(Hopic) was co-incubated with 5 lM UO126 (p \ 0.0001, represented by #) (B). Similarly, migra- tion distance of cells at the leading edge was significantly increased following 1 lM BpV(Hopic) treatment, as compared to control CM (p \ 0.01, represented by *) and this increase was abolished when BpV(Hopic) inhibitor was co-incubated with 5 lM UO126 (p \ 0.0001, represented by #) (C). n = 4 experiments performed in duplicate. CM Conditioned media; Bpv = 1 lM Bpv(Hopic); Bpv(Hopic) ? UO = 1 lM Bpv(Hopic) ? 5 lM UO126

ability to differentiate and fuse into multinucleated myotu- bes. The effect of 1 lM BpV(Hopic) on myoblast migration was in conjuction with enhanced AKT and ERK1/2 sig- nalling and the requirement for activation of both of these
signalling pathways for increased myoblast migration was revealed by individually inhibiting both and demonstrating reductions in migration back to baseline levels. BpV(Hopic), or other bisperoxovanadium compounds at proper concen- trations, might prove to be a useful tool for enhancing the migration of myoblasts, while not reducing their
differentiation capabilities—both parameters potentially important in the successful transplantation and function of myoblasts in damaged or diseased muscle tissue.

Acknowledgments Work was sponsored by Institute for Biomedi- cal Research into Human Movement and Health (IRM), Manchester Metropolitan University, Faculty of Science and Engineering, John Dalton Building, Oxford Road, Manchester, UK.

Conflict of interest The authors declare that they have no conflict of interest.

References

Allen DL, Teitelbaum DH, Kurachi K (2003) Growth factor stimulation of matrix metalloproteinase expression and myoblast migration and invasion in vitro. Am J Physiol Cell Physiol 284(4):C805–C815
Al-Shanti N, Faulkner SH, Saini A, Loram I, Stewart CE (2011) A semi-automated programme for tracking myoblast migration following mechanical damage: manipulation by chemical inhib- itors. Cell Physiol Biochem 27(6):625–636
Amano O, Yamane A, Shimada M, Koshimizu U, Nakamura T, Iseki S (2002) Hepatocyte growth factor is essential for migration of myogenic cells and promotes their proliferation during the early periods of tongue morphogenesis in mouse embryos. Dev Dyn 223(2):169–179
Bischoff R (1997) Chemotaxis of skeletal muscle satellite cells. Dev Dyn 208(4):505–515
Castaldi L, Serra C, Moretti F, Messina G, Paoletti R, Sampaolesi M, Torgovnick A, Baiocchi M, Padula F, Pisaniello A, Molinaro M, Cossu G, Levrero M, Bouche M (2007) Bisperoxovanadium, a phospho-tyrosine phosphatase inhibitor, reprograms myogenic cells to acquire a pluripotent, circulating phenotype. FASEB J 21(13):3573–3583
Cirak S, Arechavala-Gomeza V, Guglieri M, Feng L, Torelli S, Anthony K, Abbs S, Garralda ME, Bourke J, Wells DJ, Dickson G, Wood MJ, Wilton SD, Straub V, Kole R, Shrewsbury SB, Sewry C, Morgan JE, Bushby K, Muntoni F (2011) Exon skipping and dystrophin restoration in patients with Duchenne muscular dystrophy after systemic phosphorodiamidate morpho- lino oligomer treatment: an open-label, phase 2, dose-escalation study. Lancet 378(9791):595–605
Corti S, Salani S, Del Bo R, Sironi M, Strazzer S, D’Angelo MG, Comi GP, Bresolin N, Scarlato G (2001) Chemotactic factors enhance myogenic cell migration across an endothelial mono- layer. Exp Cell Res 268(1):36–44
El Fahime E, Torrente Y, Caron NJ, Bresolin MD, Tremblay JP (2000) In vivo migration of transplanted myoblasts requires matrix metalloproteinase activity. Exp Cell Res 258(2):279–287
Fan Y, Maley M, Beilharz M, Grounds M (1996) Rapid death of injected myoblasts in myoblast transfer therapy. Muscle Nerve 19(7):853–860
Georgescu MM (2010) PTEN Tumor Suppressor Network in PI3 K- Akt Pathway Control. Genes Cancer 1(12):1170–1177
Gu J, Tamura M, Yamada KM (1998) Tumor suppressor PTEN inhibits integrin- and growth factor-mediated mitogen-activated protein (MAP) kinase signaling pathways. J Cell Biol 143(5):1375–1383 Gussoni E, Blau HM, Kunkel LM (1997) The fate of individual myoblasts after transplantation into muscles of DMD patients.
Nat Med 3(9):970–977
Kawamura K, Takano K, Suetsugu S, Kurisu S, Yamazaki D, Miki H, Takenawa T, Endo T (2004) N-WASP and WAVE2 acting downstream of phosphatidylinositol 3-kinase are required for myogenic cell migration induced by hepatocyte growth factor. J Biol Chem 279(52):54862–54871
Kim MJ, Froehner SC, Adams ME, Kim HS (2011) Alpha-syntrophin is required for the hepatocyte growth factor-induced migration of cultured myoblasts. Exp Cell Res 317(20):2914–2924
Lafreniere JF, Mills P, Tremblay JP, El Fahime E (2004) Growth factors improve the in vivo migration of human skeletal myoblasts by modulating their endogenous proteolytic activity. Transplantation 77(11):1741–1747
Lafreniere JF, Caron MC, Skuk D, Goulet M, Cheikh AR, Tremblay JP (2009) Growth factor coinjection improves the migration potential of monkey myogenic precursors without affecting cell transplantation success. Cell Transplant 18(7):719–730
Lai JP, Dalton JT, Knoell DL (2007) Phosphatase and tensin homologue deleted on chromosome ten (PTEN) as a molecular target in lung epithelial wound repair. Br J Pharmacol 152(8): 1172–1184
Leloup L, Mazeres G, Daury L, Cottin P, Brustis JJ (2006) Involvement of calpains in growth factor-mediated migration. Int J Biochem Cell Biol 38(12):2049–2063
Leloup L, Daury L, Mazeres G, Cottin P, Brustis JJ (2007) Involvement of the ERK/MAP kinase signalling pathway in milli-calpain activation and myogenic cell migration. Int J Biochem Cell Biol 39(6):1177–1189
Lewis MP, Tippett HL, Sinanan AC, Morgan MJ, Hunt NP (2000) Gelatinase-B (matrix metalloproteinase-9; MMP-9) secretion is involved in the migratory phase of human and murine muscle cell cultures. J Muscle Res Cell Motil 21(3):223–233
Meng J, Muntoni F, Morgan JE (2011) Stem cells to treat muscular dystrophies—where are we? Neuromuscul Disord 21(1):4–12
Mihai C, Bao S, Lai JP, Ghadiali SN, Knoell DL (2012) PTEN inhibition improves wound healing in lung epithelia through changes in cellular mechanics that enhance migration. Am J Physiol Lung Cell Mol Physiol 302(3):L287–L299
Moens PD, Van-Schoor MC, Marechal G (1996) Lack of myoblasts migration between transplanted and host muscles of mdx and normal mice. J Muscle Res Cell Motil 17(1):37–43
Peault B, Rudnicki M, Torrente Y, Cossu G, Tremblay JP, Partridge T, Gussoni E, Kunkel LM, Huard J (2007) Stem and progenitor cells in skeletal muscle development, maintenance, and therapy. Mol Ther 15(5):867–877
Petrie RJ, Yamada KM (2012) At the leading edge of three- dimensional cell migration. J Cell Sci Dec 15; 125(Pt 24): 5917–5926
Ranzato E, Balbo V, Boccafoschi F, Mazzucco L, Burlando B (2009) Scratch wound closure of C2C12 mouse myoblasts is enhanced by human platelet lysate. Cell Biol Int 33(9):911–917
Schmid AC, Byrne RD, Vilar R, Woscholski R (2004) Bisperoxova- nadium compounds are potent PTEN inhibitors. FEBS Lett 566(1–3):35–38
Skuk D, Roy B, Goulet M, Chapdelaine P, Bouchard JP, Roy R, Dugre FJ, Lachance JG, Deschenes L, Helene S, Sylvain M, Tremblay JP (2004) Dystrophin expression in myofibers of Duchenne muscular dystrophy patients following intramuscular injections of normal myogenic cells. Mol Ther 9(3):475–482
Skuk D, Goulet M, Tremblay JP (2011) Transplanted myoblasts can migrate several millimeters to fuse with damaged myofibers in nonhuman primate skeletal muscle. J Neuropathol Exp Neurol 70(9):770–778
Smythe GM, Hodgetts SI, Grounds MD (2001) Problems and solutions in myoblast transfer therapy. J Cell Mol Med 5(1): 33–47
Suzuki J, Yamazaki Y, Li G, Kaziro Y, Koide H (2000) Involvement of Ras and Ral in chemotactic migration of skeletal myoblasts. Mol Cell Biol 20(13):4658–4665
Torrente Y, El Fahime E, Caron NJ, Del Bo R, Belicchi M, Pisati F, Tremblay JP, Bresolin N (2003) Tumor necrosis factor-alpha (TNF-alpha) stimulates chemotactic response in mouse myo- genic cells. Cell Transplant 12(1):91–100
Vaz R, Martins GG, Thorsteinsdottir S, Rodrigues G (2012) Fibronectin promotes migration, alignment and fusion in an in vitro myoblast cell model. Cell Tissue Res 348(3):569–578
Watt DJ, Karasinski J, Moss J, England MA (1994) Migration of muscle cells. Nature 368(6470):406–407
White ES, Atrasz RG, Hu B, Phan SH, Stambolic V, Mak TW, Hogaboam CM, Flaherty KR, Martinez FJ, Kontos CD, Toews GB (2006) Negative regulation of myofibroblast differentiation by PTEN (Phosphatase and Tensin Homolog Deleted on chromosome 10). Am J Respir Crit Care Med 173(1):112–121