QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

Vol. 19, Issue 3, 994-1006, March 2008

This Article Abstract Full Text (PDF) All Versions of this Article: E07-09-0856v1 19/3/994    most recent Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Google Scholar Articles by Bryan, B. A. Articles by D'Amore, P. A. PubMed PubMed Citation Articles by Bryan, B. A. Articles by D'Amore, P. A.

Coordinated Vascular Endothelial Growth Factor Expression and Signaling During Skeletal Myogenic Differentiation

Brad A. Bryan^*, Tony E. Walshe^*, Dianne C. Mitchell, Josh S. Havumaki, Magali Saint-Geniez^*, Arindel S. Maharaj^*, Angel E. Maldonado^*, and Patricia A. D'Amore^*

*Schepens Eye Research Institute, Boston, MA 02114; Acceleron Pharma, Cambridge, MA 02139; University of Massachusetts, Amherst, MA 01003

Submitted September 4, 2007; Revised November 28, 2007; Accepted December 11, 2007
Monitoring Editor: Richard Hynes

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Angiogenesis is largely controlled by hypoxia-driven transcriptional up-regulation and secretion of vascular endothelial growth factor (VEGF) and its binding to the endothelial cell tyrosine receptor kinases, VEGFR1 and VEGFR2. Recent expression analysis suggests that VEGF is expressed in a cell-specific manner in normoxic adult tissue; however, the transcriptional regulation and role of VEGF in these tissues remains fundamentally unknown. In this report we demonstrate that VEGF is coordinately up-regulated during terminal skeletal muscle differentiation. We reveal that this regulation is mediated in part by MyoD homo- and hetero-dimeric transcriptional mechanisms. Serial deletions of the VEGF promoter elucidated a region containing three tandem CANNTG consensus MyoD sites serving as essential sites of direct interaction for MyoD-mediated up-regulation of VEGF transcription. VEGF-null embryonic stem (ES) cells exhibited reduced myogenic differentiation compared with wild-type ES cells, suggesting that VEGF may serve a role in skeletal muscle differentiation. We demonstrate that VEGFR1 and VEGFR2 are expressed at low levels in myogenic precursor cells and are robustly activated upon VEGF stimulation and that their expression is coordinately regulated during skeletal muscle differentiation. VEGF stimulation of differentiating C2C12 cells promoted myotube hypertrophy and increased myogenic differentiation, whereas addition of sFlt1, a VEGF inhibitor, resulted in myotube hypotrophy and inhibited myogenic differentiation. We further provide evidence indicating VEGF-mediated myogenic marker expression, mitogenic activity, migration, and prosurvival functions may contribute to increased myogenesis. These data suggest a novel mechanism whereby VEGF is coordinately regulated as part of the myogenic differentiation program and serves an autocrine function regulating skeletal myogenesis.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Angiogenesis often coincides with increased vascular permeability, allowing extravasation of plasma proteins that lay down a provisional scaffold for migrating endothelial cells (ECs). Subsequent degradation of the extracellular matrix by matrix-metalloproteases relieves pericyte-EC contacts and liberates extracellular matrix-sequestered growth factors. ECs then proliferate and migrate to their final destination to assemble as lumen-bearing cords. These processes are largely controlled by transcriptional up-regulation and subsequent secretion of three splice variant isoforms of vascular endothelial growth factor (VEGF) and their binding to the endothelial cell tyrosine receptor kinases: VEGFR1 and VEGFR2, leading to activation of a number of downstream signaling cascades (Ferrara et al., 2003).

The function and regulation of VEGF has primarily been studied in pathological angiogenesis, such as in tumors of critical mass where hypoxia-driven HIF1 activation and subsequent binding to hypoxia-induced regulatory elements in the VEGF promoter modulates VEGF mRNA expression (Liu et al., 1995; Forsythe et al., 1996; Mazure et al., 1996). Although HIF1 is known to play a role in the regulation of VEGF gene expression in tissues of the embryo and adult, deletion of the hypoxia-responsive element in the VEGF promoter surprisingly leads to only a subtle defect—adult-onset progressive motor neuron degeneration—with other tissues largely unaffected (Oosthuyse et al., 2001). This observation suggests that the importance of hypoxia-driven HIF1 regulation of VEGF may be historically overestimated. Moreover, a recent study by Maharaj et al. (2006) has shown that VEGF is expressed at appreciable levels in a cell-specific manner in nondiseased normoxic adult tissue. This finding, together with the absence of generalized vascular defects in HIF1 null mice, suggests that VEGF gene expression in normal adult tissues is controlled by yet to be identified novel mechanisms.

In addition to its induction of new blood vessel growth, VEGF serves an important role in the maintenance and development of endothelial fenestrations (Esser et al., 1998; Lammert et al., 2003; Yokomori et al., 2003) and in endothelial survival in vitro (Darland et al., 2003) and in vivo (Kamba et al., 2006). A growing number of reports suggest that VEGF has autocrine effects on nonvascular cells such as neuronal cells (Ogunshola et al., 2002; Jin et al., 2006; Nishijima et al., 2007), muscle (Germani et al., 2003), and bone (Fons et al., 2004; Byun et al., 2007). These findings challenge the dogma that VEGF serves solely as a paracrine growth factor specifically targeting ECs (D'Amore, 2007) and may help to explain why VEGF is expressed in normoxic adult tissues, where there is not active ongoing angiogenesis.

In this report, we utilize skeletal muscle as a model system to examine the tissue-specific regulation and function of VEGF under physiological conditions. Although a plethora of data implicates exercise-induced hypoxia as the key signal modulating VEGF expression in the skeletal muscle (Gustafsson and Kraus, 2001; Olfert et al., 2001; Hudlicka et al., 2002), the level of VEGF expression remains high in the skeletal muscle of sedentary mice (Maharaj et al., 2006), suggesting a regulatory mechanism not involving hypoxia. Our data, using multipotent mesenchymal cell lines that are capable of terminally differentiating into skeletal myocytes, demonstrate that skeletal differentiation is associated with increased VEGF expression, suggesting that VEGF is coordinately regulated during myogenesis. We further demonstrate that the myogenic transcription factor, MyoD, and its heterodimeric binding proteins, E12 and E47, up-regulate the expression of endogenous VEGF through direct interaction with the VEGF promoter. Moreover, VEGF-null embryonic stem (ES) cells exhibit a significant reduction in skeletal myogenesis compared with wild-type ES cells, suggesting that VEGF is necessary for skeletal myogenesis. Consistent with these observations, we report that VEGFR1 and VEGFR2 are expressed at low levels in C2C12 cells and are robustly activated upon VEGF stimulation, indicating that VEGF can potentially signal in an autocrine manner. Indeed, VEGF stimulation of C2C12 cells results in increased terminal skeletal muscle differentiation, as evidenced by increased myotube formation, myogenic marker expression, and myotube hypertrophy. Conversely, treatment with the soluble VEGF inhibitor sFlt1 effectively blocks myogenesis and reduces myotube size. These data suggest a novel mechanism by which VEGF expression is coordinately regulated during skeletal myogenesis and demonstrate an autocrine function whereby VEGF serves as an essential regulator in this process.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
LacZ Staining
Adult C57BL/6 mice expressing the β-galactosidase (lacZ) reporter gene cDNA with a nuclear localization signal and an internal ribosome entry site inserted into the 3' untranslated region (3'UTR) of the VEGF gene were used in these studies (Miquerol et al., 1999). This gene yields a bicistronic mRNA that produces both functional VEGF and a reporter β-galactosidase (β-gal) protein. VEGF expression was visualized in adult skeletal muscle and in cryosections of VEGF-LacZ mouse embryos. Embryos were fixed overnight at 4°C in 4% paraformaldehyde in phosphate-buffered saline (PBS). For cryosections, embryos were embedded in OCT compound (Sakura Finetechnical, Torrance, CA). Whole-mounted adult skeletal muscle or cryosections were stained for LacZ using the in situ β-galactosidase staining kit, according to the manufacturer's protocol (Stratagene, La Jolla, CA).

Semiquantitative RT-PCR
Skeletal muscle from the hindlimbs of 2-mo-old C57BL/6 mice and lysates from C2C12 and 10T1/2 cultures were collected under RNase-free conditions. Total RNA was extracted using Trizol reagent (Invitrogen, Carlsbad, CA), according to the manufacturer's protocol. Residual DNA was removed by treatment with 1 U DNase I (Ambion, Austin, TX) at 37°C for 20 min. One microgram of RNA was reverse-transcribed in the presence of 500 ng of oligo dT12–18 with Superscript II reverse transcriptase (Invitrogen) in a 20-µl reaction at 42°C for 50 min and digested with 2 U of RNase H. One microliter of cDNA was used as a template in a 25-µl amplification mixture containing 200 mM dNTPs, 1 U Taq DNA polymerase (Roche Diagnostics, Indianapolis, IN), and 0.2 µM of the appropriate primer pairs capable of amplifying all three isoforms of VEGF (forward: 5' CCT CCG AAA CCA TGA ACT TTC TGC TC 3'; reverse: 5' CAG CCT GGC TCA CCG CCT TGG CTT 3'). Products were amplified for 35 cycles and separated by agarose gel electrophoresis, stained with ethidium bromide, and visualized by UV light. Quantitation of band intensity was performed using Image J software (http://rsb.info.nih.gov/ij/; NIH).

Real-Time PCR
mRNA was purified as described above using the Trizol method. One microgram of RNA was reverse-transcribed as described above, except 500 ng of random hexamers were used. One twentieth of the total cDNA (50 ng of equivalent RNA) was used in each amplification reaction. VEGF isoforms were quantified using the Prism 9700 Sequence Detection System (Applied Biosystems, Foster City, CA) according to the manufacturer's instructions. Reactions were performed in 25 µl with 0.3 µM primers specific for VEGF isoforms (Zhang et al., 2002) and SYBR Green master mix (ABI, Columbia, MD). PCR cycles consisted of an initial denaturation step at 95°C for 10 min, followed by 40 cycles at 95°C for 15 s and at 60°C for 60 s. To confirm amplification specificity, PCR products from each primer pair were subjected to a melting curve analysis. A standard curve was constructed for each PCR reaction and was derived from the serial dilution (10^–3 to 10^–9 ng DNA per reaction) of a plasmid coding for each isoform—VEGF₁₈₈, VEGF₁₆₄, and VEGF₁₂₀—and amplified using the SYBR Green system. The level of isoform expression in each sample was calculated relative to the standard curve. Each sample was run in triplicate, and each experiment included three nontemplate control wells. Results were expressed as the mean ± SD.

Cell Culture and Treatment
C2C12 (generous gift from Dr. M. T. Chin at Harvard Medical School) and 10T1/2 cells (purchased from ATCC, Manassas, VA; CCL-226) were maintained at subconfluent levels in growth medium (GM) consisting of DMEM (GIBCO, Rockville, MD) supplemented with 10% fetal bovine serum (FBS; Hyclone, Logan, UT). 10T1/2 cells, which were used for differentiation experiments, were stably transfected with a MyoD expression plasmid (generous gift from Dr. Mingyao Liu, Texas A&M Health Science Center). To induce differentiation, cells were grown to 100% confluence and GM was replaced with differentiation medium (DM) consisting of DMEM supplemented with 2% horse serum (Invitrogen) and 80 U/ml penicillin/streptomycin C. DM was replaced every 24 h.

Bovine retinal endothelial cells (BRECs) were maintained in EBM supplemented with 10% horse serum, 80 U/ml penicillin/streptomycin C, and 12 µg/ml bovine brain extract. The cells were plated on plastic coated with 50 µg/ml fibronectin.

Wild-type and VEGF-null mouse ES cells (generous gift from Andras Nagy, Mount Sinai Hospital, Toronto, Canada) were cultured on gelatin-coated dishes in high-glucose DMEM (GIBCO BRL) with 15% fetal bovine serum (Lot FRB25667, Hyclone), sodium pyruvate (GIBCO, stock solution diluted 1:100), nonessential amino acids (GIBCO, stock solution diluted1:100), β-mercaptoethanol (GIBCO, final concentration 30 µM), 190 µg/ml L-glutamine, 60 U/ml penicillin G, 60 µg/ml streptomycin (glutamine pen–strep mix, Irvine Scientific, Santa Ana, CA), supplemented with media (1:300 dilution) conditioned by Chinese hamster ovary cells overexpressing LIF (provided by Genetics Institute, Cambridge, MA) as a source of LIF to maintain the ES cells in an undifferentiated state. ES cells were cultured in a humidified tissue culture incubator at 10% CO₂ and 37°C and passaged every 2 to 3 d. ES cells were differentiated into cystic embryoid bodies (CEB) as previously described (Ng et al., 2004). Briefly, trypsinized ES cells were suspended in the same culture medium as described above, but without LIF. A total of 60 aliquots (30 µl) of ES cell suspension containing 2.5 x10³ cells were plated as individual drops onto 100-mm² bacteriological dishes (Valmark, Brampton, Canada). The plates were then inverted and the cells were incubated in hanging drops; this was defined as day 0 of differentiation. The CEB were cultivated via hanging drop for 40–45 h, and then the dishes were turned right side up and 10 ml of ES culture media without LIF was added so that the CEB were then in suspension. Every 3 d, half of the culture media was removed and replaced with fresh media. For attached cultures, day 4 or day 5 CEB were transferred to gelatin-coated glass coverslips, onto which the CEB attached, flattened, and spread.

Recombinant human VEGF₁₆₅ (obtained from National Cancer Institute, www.cancer.gov) was added to the cultures at a final concentration of 25 ng/ml. Recombinant mouse VEGF-R1/Fc chimera (sFlt1; R&D Systems, Minneapolis, MN; 471-F1, Lot BSL1107011) was added to the cultures at a final concentration of 100 ng/ml. Recombinant adenovirus (AV) vectors containing either a null cassette or sFlt1 were obtained from Qbiogene (Montreal, Canada). Undifferentiated C2C12 cells were incubated with AV vectors at multiplicities of infection (MOIs) of 10³ viral genomes per cell.

Western Blotting
Tissues were collected in lysis buffer (10 mM Tris-HCl, pH 7.4, 5 mM EDTA, 50 mM NaCl, 1% Triton X-100, 50 mM NaF, 1 mM phenylmethylsulfonyl fluoride [PMSF], 2 mM Na₃VO₄, and 20 mg/ml aprotinin). Proteins were quantified using the D_c protein assay kit (Bio-Rad, Richmond, CA). Cell extracts were analyzed by SDS-PAGE and probed with anti-MyoD (sc-304, Santa Cruz Biotechnology, Santa Cruz, CA), anti-VEGFR1 (sc-316, Santa Cruz Biotechnology), anti-VEGFR2 (a generous gift from A. Kazlauskas, Schepens Eye Research Institute), anti-myosin heavy chain (MHC; MF20, Iowa Hybridoma Bank), anti-myogenin (F5D, Iowa Hybridoma Bank), PY20 (05-321, Upstate Biotechnology, Lake Placid, NY), and anti-tubulin (CP06-100UG, Oncogene Research Products, Boston, MA). Binding was detected with the appropriate HRP-conjugated secondary antibody (mouse: NA931V, rabbit: NA934V, Amersham Biosciences, Piscataway, NJ) and ECL-Plus Western Blotting Detection System (RPN2132, GE Healthcare, Waukesha, WI).

Immunofluorescence
Cells were fixed in ice-cold methanol for 10 min. Reactions were blocked for 60 min with PBS containing 0.2% bovine serum albumin, followed by 60 min of incubation with anti-MHC antibody (1:20 dilution; Iowa Hybridoma Bank). Fluorescein-conjugated goat anti-mouse antibody was added for 30 min (1:1000; 81-6511, Molecular Probes, Eugene, OR). Nuclear staining was observed after 10 min of 4',6-diamidino-2-phenylindole (DAPI) treatment (1:100 dilution; Molecular Probes). Fluorescence images were captured on a CCD camera mounted on an inverted research microscope using Spot imaging software (Diagnostic Instruments, Sterling Heights, MI).

Luciferase Reporter Assays
VEGF-promoter luciferase constructs have been previously described (Loureiro et al., 2005). C2C12 and 10T1/2 cells were cotransfected at 60–80% confluence using Lipofectamine 2000 with equal amounts of VEGF firefly luciferase (Fluc)-reporter plasmid, Renilla control luciferase (Rluc; Promega, Madison, WI), and the indicated plasmid to a total of 1.5 µg DNA per well of a 24-well plate. Fresh medium was added 6 h after transfection and left undisturbed until cell lysates were collected after 48-h incubation. Luciferase expression was detected using the Dual Luciferase Reporter Assay System (Promega) following the manufacturer's instructions with a Turner Luminometer (Turner Designs, Sunnyvale, CA). VEGF promoter activity, as detected by Fluc activity, was normalized against Rluc readings and this ratio was divided by the ratio of readings from promoterless luciferase to determine fold-change. Assays were performed three times in triplicate with similar results; one representative experiment is shown. Statistical analysis of the data were performed to determine significance using analysis of variance.

Chromatin Immunoprecipitation Assay
Chromatin immunoprecipitation assay (ChIP) was performed following the protocol outlined by the ChIP assay kit (Upstate Biotechnology). Briefly, C2C12 cells were fixed with 1% formaldehyde, scraped into conical tubes, pelleted, and lysed in SDS lysis buffer containing 1 mM phenylmethylsulfonyl fluoride, 1 µg/ml aprotinin, and 1 µg/ml pepstatin A. DNA was sheared to fragments of 200–500 base pairs by eight 10-s sonications. The chromatin was precleared with salmon sperm DNA/protein A-agarose slurry (Upstate Biotechnology) for 1 h at 4°C with gentle agitation. The agarose beads were pelleted, and the precleared supernatant was incubated with antibodies to IgG or MyoD overnight at 4°C. The region between –256 to –122 bases upstream of the VEGF 5'UTR was PCR amplified from the immunoprecipitated chromatin using the appropriate primers (Forward: 5'-CACTCTCCTGTCTCCCCTGA-3'; Reverse: 5'-CACTACCGCGAAATGGAAAG-3'). After PCR, the PCR product was resolved on a 2.5% agarose gel and stained with ethidium bromide. Samples were visualized under UV light.

Invasion/Migration Assay
C2C12 cells were seeded on six-well plates and grown to 100% confluence in DMEM + 10% FBS (serum starvation was not used in this experiment because low serum forced the cells to differentiate and inhibited migration almost completely) and wounded with a sterile pipette tip to remove cells by two perpendicular linear scratches. After washing, the cells were cultured with DMEM + 10% FBS under the following conditions: mock treatment, 25 ng/ml VEGF, or 100 ng/ml sFlt1. The progress of migration was photographed immediately after injury and at 12 h after wounding, near the crossing point, with an inverted microscope equipped with a digital camera (SPOT; Diagnostic Imaging, Sterling Heights, MI).

Proliferation Assay
Proliferation assays were performed according to the method of Lyons (2000). Briefly, C2C12 cells were resuspended at 10⁶ cells/ml in PBS, and 5-(and-6)-carboxyfluorescein diacetate succinimidyl ester (CFDASE) was added to a final concentration of 5 µM and incubated at 37°C for 10 min. The cells were then washed two times with DMEM media supplemented with 10% FBS, plated in six-well plates, and allowed to proliferate for 48 h in DMEM + 10% FBS in the presence of control or 100 ng/ml sFlt1. Cells were collected by trypsinization and washed two times in PBS, and fluorescence was read using a Becton Dickinson Facsort flow cytometer (Mountain View, CA).

Apoptosis Assay
Apoptosis was measured using the annexin V binding method (Koopman et al., 1994). Briefly, C2C12 cells at day 2 of differentiation conditions were treated with control or 100 ng/ml sFlt1, and cells were washed once with PBS, trypsinized, and resuspended briefly in serum containing medium to quench the trypsin. Cells were washed once in PBS, and 1 x 10⁵ cells were added to annexin V binding reagent for 15 min in the dark. Propidium iodide was included in the reaction to allow the identification of cells in late apoptosis. Fluorescence was read using a Becton Dickinson Facsort flow cytometer.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Expression of VEGF in the Developing and Adult Skeletal Muscle
To investigate VEGF expression in the mouse embryo and adult skeletal muscle, representative sections from VEGF-lacZ E18.5 mouse limbs and adult skeletal muscle whole mounts were stained with X-gal. Robust LacZ staining was observed in both the embryonic and adult time points (Figure 1A). Although a high level of VEGF expression is not surprising in embryonic samples, with the exception of exercise-induced hypoxia, little published data exists regarding the physiological expression and control of VEGF in the adult skeletal muscle. Our data are consistent with a report published by Maharaj et al., (2006), indicating strong expression of VEGF in adult mouse skeletal muscle.

View larger version (31K): [in this window] [in a new window]   Figure 1. VEGF is expressed in embryonic and adult skeletal muscle. (A) Cryosections of embryonic mouse limbs (E18.5) and whole mounts of adult mouse skeletal muscle were collected from VEGF-LacZ mice and stained for LacZ using in situ β-galactosidase. Scale bar (E18.5) 50 µm; scale bar (adult) 100 µm. Support of mRNA was collected from the hindlimb skeletal muscle of 2-mo-old C57BL/6 mice, and semiquantitative RT-PCR detection of VEGF188, VEGF164, and VEGF120 was performed. As a reaction control, a PCR reaction composed of mixed VEGF plasmids each encoding single VEGF isoforms was utilized. (C) mRNA was collected from the hindlimb skeletal muscle of 2-mo-old C57BL/6 mice and real-time RT-PCR detection of VEGF188, VEGF164, and VEGF120 was performed. The relative amount of each VEGF isoform is represented as a percentage of the total VEGF expression (VEGF188 + VEGF164 + VEGF120 = 100%). Error bars, SDs.

VEGF Is Up-Regulated during Myogenic Differentiation
We next sought to elucidate the molecular mechanism controlling VEGF expression in skeletal muscle. We utilized in vitro cell culture models that closely recapitulate the formation and maintenance of skeletal muscle. Two cell lines serve as excellent model systems to examine the molecular mechanisms controlling skeletal muscle differentiation: the multipotent mesenchymal progenitor cell line C2C12 (Bains et al., 1984) and the fibroblast cell line 10T1/2 (Davis et al., 1987). To determine if VEGF is endogenously expressed in these cell lines, we performed semiquantitative RT-PCR to detect the VEGF isoform expression. As demonstrated in Figure 2A, VEGF₁₂₀, VEGF₁₆₄, and VEGF₁₈₈ were all expressed in both cell lines, indicating that these lines likely serve as good model systems to examine the mechanisms controlling VEGF gene expression. It is worth noting that VEGF was expressed at higher levels in C2C12 cells compared with 10T1/2 cells.

View larger version (60K): [in this window] [in a new window]   Figure 2. VEGF expression is regulated during skeletal myogenic differentiation. (A) Semiquantitative RT-PCR for VEGF isoform and MyoD expression was performed on cell lysates of C2C12 and 10T1/2 cells. Steady-state mRNA levels were detected and GAPDH expression was utilized to indicate equal loading. (B) Phase-contrast images of C2C12 cells were taken in the proliferative phase (GM) and after 3 d of differentiation (DM). (Arrows indicate the presence of myotubes after 3-d differentiation, 40x magnification) (C) Western blot analysis of the muscle-lineage–specific marker MyoD was performed on lysates collected from proliferating (GM) and 3-d differentiated (DM) C2C12 cells. Tubulin expression was utilized to indicate equal loading. (D–G) RNA was collected from C2C12 and 10T1/2 cells in the proliferative phase of growth (GM) and over 4 d of differentiation (DM1-4). Semiquantitative RT-PCR for VEGF isoform expression was performed for each time point for C2C12 (D) and 10T1/2 (F). Analysis of GAPDH expression was utilized as a loading control. Quantification of VEGF isoform expression during C2C12 (E) and 10T1/2 (G) differentiation was performed using NIH ImageJ software.

The Myogenic Transcription Factor MyoD Controls VEGF Transcriptional Expression
Given the unique expression patterns of the VEGF isoforms during skeletal myogenesis, we sought to determine the molecular mechanisms controlling this process. Using TESS analysis (http://www.cbil.upenn.edu/cgi-bin/tess/tess) of the human VEGF promoter, we identified six putative CANNTG sites within the promoter region, 1600 bases upstream of the coding region (Figure 3A). MyoD homodimers or heterodimers of MyoD plus E12 or E47 serve as transcription factor complexes that bind to CANNTG consensus sites in the promoter regions of genes, performing major functions in specification and differentiation of skeletal muscle precursor cells (Murre et al., 1989).

View larger version (31K): [in this window] [in a new window]   Figure 3. The myogenic transcription factor MyoD regulates the expression of VEGF. (A) Illustration depicting consensus CANNTG MyoD binding sites located within the first 1600 bases upstream of the VEGF 5' UTR. (B and C) Semiquantitative RT-PCR for VEGF isoform expression was performed on cDNA collected from C2C12 cells stably overexpressing an empty plasmid (control), wild-type MyoD (MyoD), or dominant negative MyoD (dnMyoD). GAPDH detection was utilized as a loading control. Quantification of VEGF isoform expression was performed using NIH ImageJ software. (D) 10T1/2 cells were transiently transfected with the 9-kb full-length VEGF promoter region driving expression of the firefly luciferase gene in combination with empty vector, wild-type MyoD, E12, and/or E47. A Renilla luciferase vector was included in all conditions as a transfection control. Firefly luciferase activity was measured and equalized to the control Renilla luciferase activity. Luciferase activity is expressed as arbitrary luciferase units. Each experiment was performed in triplicate and repeated at least three times. Data are reported as the mean ± SD for each representative experiment. (E) 10T1/2 cells were transiently transfected with the indicated truncation constructs of the VEGF promoter region driving expression of the firefly luciferase gene and either a control vector or wild-type MyoD and normalized with Renilla luciferase as above. Data are reported as the percentage of increase in MyoD-transfected cells over control-transfected cells with mean ± SD for each representative experiment. (F) Chromatin immunoprecipitation analysis of C2C12 lysates detecting direct MyoD binding to the VEGF promoter region between –256 to –122 bases upstream of the VEGF 5'UTR. IgG incubation was utilized as a control.

To confirm our finding that MyoD modulates VEGF expression in skeletal muscle precursor cells, we utilized luciferase reporter assays. C2C12 cells were transfected with the 9-kb (full length) VEGF promoter luciferase reporter plasmid and combinations of 1) control vector, 2) wild-type MyoD, 3) E12, or 4) E47, and luciferase assays were performed on the cell lysates. As observed in Figure 3D, coexpression of the VEGF luciferase plasmid with MyoD led to an approximately fourfold increase in VEGF-reporter expression over the control. Transfection with either E12 or E47 alone resulted in no increase in VEGF-reporter expression; however, cotransfection of either MyoD and E12 or MyoD and E47 led to eightfold increase in luciferase activity over the control. These data suggest that MyoD and the E-box proteins act synergistically to up-regulate VEGF-reporter expression.

We next set out to determine the specific transcriptional binding sites that mediate the MyoD-dependent increase of VEGF expression. C2C12 cells were transfected with truncation mutants of the VEGF promoter luciferase reporter plasmid and either a control plasmid or wild-type MyoD. Strong MyoD-driven VEGF-reporter expression was observed in full-length promoter constructs and truncations down to –160 bases upstream of the transcriptional start site (Figure 3E). However, truncations less than this region led to complete loss of MyoD-mediated transcriptional expression of the reporter gene, suggesting that a region between –247 bases and –160 bases upstream of the transcription start site is essential for MyoD activity on this promoter. Indeed, within this region lies three closely spaced CANNTG sites at –165, –187, and –198 bases upstream of the VEGF 5' UTR.

MyoD Physically Associates with the VEGF-Promoter
To demonstrate that MyoD directly interacts and forms a complex with the MyoD sites in this region, we performed ChIP using sheared DNA isolated from C2C12 cells. Immunoprecipitation of chromatin-bound DNA using an antibody against endogenous MyoD was followed by PCR using primers that amplified the –256- to –122-base pair region of the VEGF promoter spanning the length of the three essential consensus MyoD sites. As demonstrated in Figure 3F, the MyoD-specific antibody is capable of immunoprecipitating the VEGF promoter fragment containing the three MyoD sites, whereas control immunoprecipitation using anti-IgG failed to produce a PCR product.

VEGF-Null Embryonic Stem Cells Exhibit Decreased Skeletal Myogenesis
The essential role of VEGF during vasculogenesis and angiogenesis is well established (Ferrara, 1999); however, very little is known regarding nonvascular roles for VEGF. To determine if VEGF influences skeletal myogenic differentiation, we utilized a CEB model in which ECs are cultivated to form aggregates, which efficiently differentiate into a number of cell lineages, including skeletal muscle. This system closely recapitulates the early steps of muscle development in vivo (Rohwedel et al., 1994) and serves as an excellent in vitro system to study this process. ES cells isolated from wild-type or VEGF-null mice were differentiated into CEB and then transferred to gelatin-coated glass coverslips, where they attached, flattened, spread, and differentiated. Over a period of 11 d, we performed immunofluorescent staining specific for the skeletal muscle specific marker, MHC, and calculated the percentage of each field that was positive for MHC. Myosin heavy chain was first detected at day 7 in both the wild-type and VEGF-null ES cultures in sparse, isolated patches throughout the CEB. At this time point staining in wild-type cultures represented 3.5% of the total area of the representative fields, whereas MHC staining in VEGF-null cells represented 0.3% of the total area of the representative fields (Figure 4A and B). At days 9 and 11 of differentiation, VEGF-null ES cultures consistently demonstrated a reduction in MHC expression compared with CEB derived from wild-type ES cells. Because the MHC antibody (Iowa Hybridoma Bank, MF20) we used in these experiments is known to react with any striated muscle actin, including cardiac tissue, we performed Western blots on wild-type and VEGF-null ES cell lysates at day 9 of differentiation to detect the expression of the skeletal muscle specific protein, myogenin. As expected, a significant reduction in myogenin protein expression is observed in VEGF-null ES cell lysates compared with wild-type ES cell lysates (Figure 4C). These data demonstrate that while the loss of VEGF does not appear to affect the onset of myogenic differentiation, VEGF is necessary for this process to proceed in a robust manner. However, these findings do not demonstrate whether myogenic cells are responding to their own VEGF production or from VEGF secreted in a paracrine manner from other cell types in the culture.

View larger version (26K): [in this window] [in a new window]   Figure 4. VEGF regulates skeletal myogenesis during embryonic stem cell differentiation. (A) ES cells were differentiated using the hanging drop method as described in Materials and Methods. At 7, 9, and 11 d of differentiation, cells were fixed and MHC was detected by immunofluorescence (green) and DAPI (blue) was performed (40x magnification). (B) Quantification of MHC positive area within the field was performed using NIH ImageJ software. Data are reported as the mean ± SD for each representative experiment. (C) Lysates from ES cells were collected at day 9 of differentiation and Western-blotted with anti-myogenin and anti-tubulin antibodies.

View larger version (67K): [in this window] [in a new window]   Figure 5. VEGF receptor expression and activation during skeletal myogenesis. (A) Lysates from BREC and C2C12 cells were Western blotted for VEGFR1 and VEGFR2. Tubulin levels were utilized as a loading control. (B) Lysates from C2C12 cells grown in 2% horse serum with (VEGF) or without (control) the 25 ng/ml VEGF (2 min) were immunoprecipitated with anti-VEGFR2 antibodies and Western blotted with anti-PY20 and anti-VEGF-R2 antibodies were performed to detect VEGFR2 phosphorylation. (C) Lysates from C2C12 cells in proliferative conditions (GM) and days 1 through 4 of myogenic differentiation (DM1-4) were subjected to Western analysis to detect levels of VEGFR1 and VEGFR2. Detection of MHC levels demonstrated myogenic differentiation and tubulin protein levels were used as a loading control.

VEGF Stimulation of C2C12 Cells Enhances Myogenic Differentiation
Given that C2C12 cells provide a well-documented model in which to examine the effects of VEGF on myogenesis, we utilized phase-contrast microscopy to quantify myotube formation during C2C12 myogenic differentiation in response to VEGF stimulation or inhibition. As observed in Figure 6A and C, the addition of 25 ng/ml VEGF to differentiating C2C12 cells resulted in an 30% increase in the number of myotubes compared with untreated control cells. Moreover, inhibition of autocrine VEGF signaling via the expression of sFlt1 led to an 70% reduction in myotube numbers relative to AV-control (Figure 6A and C). We confirmed these results using fluorescent detection of the skeletal muscle–specific differentiation marker, MHC. Similar to our findings measuring myotube numbers, addition of VEGF led to an increase in MHC-positive cells compared with cells that received control treatment (Figure 6B), whereas inhibition of autocrine VEGF signaling using sFlt1 resulted in decreased MHC-positive cells relative to controls (Figure 6B). Additionally, VEGF stimulation of differentiating C2C12 cells resulted in a significant increase in myotube hypertrophy, although sFlt1 inhibition of VEGF signaling led to a decrease in myotube size compared with controls (Figure 6D). These results strongly suggest that VEGF signals to C2C12 cells through an autocrine mechanism.

View larger version (69K): [in this window] [in a new window]   Figure 6. VEGF regulates C2C12 skeletal myogenic differentiation. (A and B) C2C12 cells were differentiated for 3 d in the following conditions: control, VEGF, AV-control, or AV-sFlt1. Phase-contrast images (40x magnification; A) and immunofluorescent detection of MHC (MHC; green) and DAPI (blue; 100x magnification; B) were collected at day 3 of differentiation. (C) Quantification of myotube number per field was performed using NIH ImageJ software. Data are reported as the mean ± SD for each representative experiment. (D) Quantification of relative MHC-positive cell size using NIH Image J software (red dashed line indicates median size of MHC positive control or AV-control cells).

View larger version (62K): [in this window] [in a new window]   Figure 7. VEGF serves multiple roles in C2C12 cells. (A) Lysates were collected from C2C12 cells grown at 3 d of differentiation and treated with 25 ng/ml VEGF, 100 ng/ml sFlt1, or left untreated and subsequently subjected to Western analysis to detect steady-state levels of MHC. Tubulin was utilized as a loading control. (B) C2C12 cells were grown to 100% confluence, treated as in A and wounded with a sterile pipette tip to remove cells. Photographs were taken (40x magnification) at t = 0 and t = 12 h (control, VEGF, and sFlt1) after the injury. (C) Fluorescence-activated cell sorter analysis for annexin V:propidium iodide staining of C2C12 cells at day 2 of differentiation and treated 100 ng/ml sFlt1. (D) Flow cytometric analysis of CFDASE stability in proliferating C2C12 cells treated with 25 ng/ml VEGF or 100 ng/ml sFlt1 (red dashed line indicates median fluorescence of control cells).

In addition to its action on the endothelium, VEGF has been heavily implicated as a survival factor for nonvascular cells (Zachary, 2001; Darland et al., 2003; Nishijima et al., 2007). To determine if the promyogenic effect of VEGF signaling might occur through its anti-apoptotic signaling, differentiated C2C12 cells were treated with 100 ng/ml recombinant sFlt1 for 48 h and examined for apoptosis via Annexin V:propidium iodide staining and subsequent flow cytometric analysis. As demonstrated in Figure 7C, sFlt1-treated cells exhibited an 50% increase in the number of apoptotic cells compared with control cells, suggesting that VEGF signaling is essential for cell survival during C2C12 differentiation, an effect that may contribute to the promyogenic effects of VEGF.

Muscles are formed by fusion of individual postmitotic myoblasts to form multinucleated myotubes; however, during skeletal muscle regeneration, which is associated with routine maintenance, hypertrophy, and repair, satellite stem cells are activated from their quiescent state, proliferate, and finally differentiate (Adams, 2006; Zammit et al., 2006). Because VEGF is a mitogen for ECs (Leung et al., 1989), we tested whether it might also regulate myoblast proliferation. Using CFDASE cell labeling as a measure of cell proliferation, we observed that sFlt1-treated cells retained significantly more CFDASE than control cells, indicating that VEGF neutralization reduced the division rate of C2C12 cells (Figure 7D). We found no difference in CFDASE retention between control and VEGF-treated cells (data not shown); however, this was not surprising given that the cells were grown in 10% FBS and were likely replicating at maximum capacity. Given the tendency of C2C12 cells to undergo differentiation and thereby enter G0 at lower serum concentrations, we did not test the mitogenic effects of VEGF at low serum concentrations.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The formation of vascular systems is essential for normal growth, differentiation, and organ function where the vascular bed is specialized to meet the needs of the organ it supplies. To achieve the high degree of structural organization necessary to serve the specific tissue requirements, blood vessel formation is tightly regulated by the specific expression and secretion of factors such as VEGF. VEGF has been shown to be expressed at varying levels in a cell-specific manner in virtually all vascularized adult tissues (Maharaj et al., 2006) and is up-regulated in a number of cell types in response to hypoxic conditions (Liu et al., 1995; Forsythe et al., 1996; Mazure et al., 1996). The data presented in this report demonstrate a novel regulatory mechanism by which VEGF is up-regulated during skeletal muscle differentiation by the myogenic regulatory factor, MyoD. Moreover, although VEGF is well established as a modulator of blood vessel formation (Ferrara et al., 2003), the expression of VEGF receptors by many cell types in addition to EC points to nonvascular roles for VEGF in the adult (Maharaj et al., 2006; D'Amore, 2007). Indeed, our data suggest that VEGF mediates skeletal myogenesis and that VEGF expressed by differentiating myogenic precursors functions in an autocrine manner.

Through differential mRNA splicing, the single murine VEGF gene gives rise to at least three isoforms including, VEGF₁₂₀, VEGF₁₆₄, and VEGF₁₈₈ (Ferrara et al., 1992; Shima et al., 1996). VEGF₁₂₀ lacks a heparan sulfate binding site and is therefore freely diffusible, whereas VEGF_188, with two heparan sulfate binding sites, associates with the cell surface and extracellular matrix, and VEGF₁₆₄ has intermediate properties (Park et al., 1993; Ferrara and Davis-Smyth, 1997). The various VEGF isoforms play distinct roles in vascular development, and their patterns of expression vary during organ development and in the adult (Ng et al., 2001; Robinson and Stringer, 2001; Skryabin et al., 2004). Our data demonstrating VEGF₁₈₈ and VEGF₁₆₄ expression in skeletal muscle is consistent with a number of other studies in human, rat, and mouse (Ng et al., 2001; Ishii et al., 2002; Jensen et al., 2004; Maharaj et al., 2006). Given that skeletal muscle is subjected to a considerable amount of daily stresses including exercise-induced hypoxia (Hoppeler, 1999) and structural damage (Russell et al., 1992), necessitating angiogenesis, VEGF expression in this tissue is not surprising. In addition, we demonstrate that multipotent cell lines (C2C12 and 10T1/2), which under the appropriate conditions are molecularly similar to myogenic satellite stem cells, express all three VEGF isoforms. These data suggest that myogenic precursors, in addition to expressing the sequestered forms of VEGF (VEGF₁₈₈ and VEGF₁₆₄), may express diffusible VEGF_120, which could function to attract distant vessels to sites of myoblast differentiation. Although multiple reports demonstrate variable tissue-specific isoform expression patterns for VEGF, the mechanism controlling this process remains unknown.

HIF1 serves as a master regulator for the expression of a number of genes involved in the hypoxic response, including glucose transporters and glycolytic enzymes as well as VEGF (Damert et al., 1997; Minchenko et al., 2002; Mobasheri et al., 2005), and hypoxic regulation of VEGF expression via HIF-1 is the best characterized mechanism of VEGF control. Interestingly, deletion of the hypoxia-response element in the VEGF promoter in mice results in a viable mouse with only minor phenotypes (Oosthuyse et al., 2001). In contrast, VEGF–/– and +/– mice are embryonic lethal (Ferrara et al., 1996). Thus, the dispensable nature of HIF-1 suggests that HIF-1 is much less important in the regulation of VEGF than previously suspected. Furthermore, there are a small but growing number of reports documenting mechanisms that modulate VEGF expression in normoxic conditions. For instance, myogenic Akt signaling controls both muscle fiber hypertrophy and VEGF synthesis, independent of HIF-1 activity, illustrating a mechanism through which blood vessel recruitment can be coupled to normal tissue growth (Takahashi et al., 2002). Moreover, analysis of VEGF mRNA levels across an array of adult tissues demonstrates a wide variability in the expression pattern for VEGF (Ng et al., 2001), suggesting complex transcriptional modulation of this gene. TESS promoter analysis (http://www.cbil.upenn.edu/cgi-bin/tess/tess) of the human VEGF promoter reveals a wide array of cell-type specific putative transcriptional regulatory sites for transcription factors, such as c-Myb (involved in proliferation and differentiation of hematopoetic progenitor cells), myogenin (involved in differentiation of skeletal muscle), T-cell factor (TCF; involved in T-cell specific differentiation), and peroxisome proliferator-activated receptor (PPAR; transcription factor specific for liver, kidney, heart, and brown adipose tissue), to name only a few.

Our observation of a number of putative CANNTG MyoD-binding sites within 9 kb upstream of the VEGF transcriptional start site motivated us to investigate if skeletal muscle-specific regulation of VEGF exists independent of hypoxic regulation. In skeletal muscle, MyoD functions in an instructive chromatin context and directly regulates genes expressed throughout the myogenic program (McKinsey et al., 2001; Pownall et al., 2002). For example, expression of MyoD is sufficient to convert fibroblast and adipoblast cells into skeletal muscle cells (Tapscott et al., 1988). Indeed, our analysis revealed that VEGF is transcriptionally up-regulated during skeletal muscle differentiation via the myogenic transcription factor MyoD and its binding partners, E12 and E47. Many studies have demonstrated VEGF up-regulation in skeletal muscle regeneration, transplantation, and exercise models (Rissanen et al., 2002; Smythe et al., 2002; Arsic et al., 2004; Prior et al., 2004; Yan et al., 2005; Wagatsuma, 2007; Xia et al., 2006), and our report expands on that knowledge by providing a specific mechanism by which VEGF is up-regulated during normal skeletal muscle differentiation. Indeed, in light of the many putative tissue-specific transcription factor binding sites located in the VEGF promoter, we speculate that the differentiation-dependent expression of VEGF that we have documented in skeletal muscle may be a paradigm for VEGF regulation in adult tissues.

The most studied aspects of VEGF in skeletal muscle involves its role in inducing angiogenesis during muscle formation and after bouts of exercise (Wagner, 2001; Haas, 2002; Bloor, 2005). Two contradicting studies have previously attempted to examine the effects of VEGF on skeletal myogenesis. One study suggests that VEGF strongly affects myocyte migration and survival but reports no effect on myogenic marker expression (Germani et al., 2003). The interpretation of these results remains complicated because the authors report very early expression of MHC, in contrast to a large number of reports that demonstrate that MHC is only detectable by Western blotting in C2C12 cells at 2 to 3 d of differentiation (Miller, 1990; Andres and Walsh, 1996). Another report demonstrated increased proliferation and decreased apoptosis in C2C12 cells in response to exogenous VEGF, as well as an increase in myotube number, size, and multinucleation (Arsic et al., 2004); however, this group did not examine a possible autocrine role for VEGF.

We have comprehensively examined the role and regulation of VEGF during skeletal myogenesis and demonstrate that VEGF stimulation increases myotube number, myogenic marker expression, and myotube size, whereas inhibition of VEGF signaling decreases these processes. These data indicate that VEGF may be essential for proper myogenic differentiation. Given the ubiquitous distribution of VEGF, we speculate that VEGF may perform similar autocrine roles for a number of developmental processes. For instance, VEGF is overexpressed in chondrogenic and osteogenic differentiation cascades, and inhibition of VEGF activity results in reduced bone differentiation, suggesting that this factor plays an important role during cartilage and bone formation (Mayer et al., 2005; Bluteau et al., 2007). Moreover, differentiating kidney podocytes strongly up-regulate VEGF and VEGFR2, and treatment with VEGF reduces podocyte apoptosis 40%, suggesting VEGF promotes survival in these cells through VEGFR2 (Guan et al., 2006). Genetic overexpression of VEGF in adult rats results in an approximately twofold increase in hippocampal neurogenesis associated with improved cognition, whereas inhibition of VEGF expression by RNA interference completely blocks the environmental induction of neurogenesis, suggesting that VEGF serves as a mediator of neurogenesis and cognition (During and Cao, 2006). It seems clear that future studies will elucidate more instances where VEGF is utilized for the function of nonvascular cell types.

The coordinated regulation of the VEGF receptors during skeletal myogenesis possibly reflects the underlying biology of the system. Our data suggest that proliferating, undifferentiated C2C12 cells predominantly express VEGFR1; however, differentiating C2C12 cells down-regulate VEGFR1 and dramatically increase the expression of VEGF2. Because the differentiated C2C12 cells have completely exited the cell cycle, VEGFR2 signaling no longer leads to cell proliferation but instead appears to contribute to myogenic differentiation. A similar induction of VEGFR2 expression has been observed in differentiation of podocytes (Guan et al., 2006). The expression of VEGFR2 has been shown in other systems to be induced by VEGF (Enholm et al., 2001), and given our data indicate that VEGF is up-regulated during myogenic differentiation, this may account for the observed increased VEGFR2 expression. Our findings that VEGF can mediate myogenic differentiation and survival are consistent with an autocrine role for VEGFR2 signaling in differentiating and differentiated skeletal muscle. At the same time, the VEGF produced by the differentiating myocytes is necessary for the vascularization of the developing skeletal muscle.

Although these results are important in elucidating the molecular mechanisms controlling tissue-specific vascular development and maintenance, they also present broader medical implications for anti-angiogenic treatment for diseases such as cancer and macular degeneration. Patients undergoing long-term anti-VEGF therapy may experience unexpected side effects due to interference with normal vessel regrowth/stability in skeletal muscle tissue and/or prolonged muscle regeneration time after injury. For example, although developmental differentiation of skeletal muscle does not occur in the adult, day-to-day wear and tear of adult skeletal muscle accounts for 1–2% replacement of muscle fibers per week (Schmalbruch and Lewis, 2000)—processes that involve satellite stem cell activation and subsequent myogenic differentiation and that could be susceptible to anti-VEGF therapies. Consistent with this speculation, patients treated with Avastin report asthenia, which, though nonspecific, may result from reduced muscle regeneration. Because of the low rate of skeletal muscle turnover/repair, abnormalities might be expected to require extended treatment to develop. In addition, because patients receiving Avastin often exhibit a spectrum of medical issues, subtle changes in muscle function might be difficult to detect. A beneficial effect of VEGF overexpression has been reported for skeletal muscle regeneration through both proangiogenic effects and myogenic regenerative effects (Arsic et al., 2004; Messina et al., 2007). These observations suggest that anti-VEGF drugs could inhibit muscle regeneration through dual mechanisms—blocking angiogenesis and blocking satellite stem cell activation and subsequent myogenesis.

	ACKNOWLEDGMENTS

 
We thank Dr. M. Horwitz from the University of Washington, Seattle, WA, for the use of the dominant negative MyoD plasmid construct and Dr. M. Liu from the Texas A&M University Health Science Center, Houston, TX, for the use of the wild-type MyoD plasmid construct. This study was supported by the National Institutes of Health Grants EY05318, EY015435, and CA45548. P.A.D. is a Research to Prevent Blindness Senior Scientific Investigator.

	Footnotes

 
This was published online ahead of print in MBC in Press (http://www.molbiolcell.org/cgi/doi/10.1091/mbc.E07-09-0856) on December 19, 2007.

Address correspondence to: Patricia A. D'Amore (patricia.damore{at}schepens.harvard.edu)

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Adams, G. R. (2006). Satellite cell proliferation and skeletal muscle hypertrophy. Appl. Physiol. Nutr. Metab 31, 782–790.[CrossRef][Medline]

Andres, V., and Walsh, K. (1996). Myogenin expression, cell cycle withdrawal, and phenotypic differentiation are temporally separable events that precede cell fusion upon myogenesis. J. Cell Biol 132, 657–666.[Abstract/Free Full Text]

Arsic, N., Zacchigna, S., Zentilin, L., Ramirez-Correa, G., Pattarini, L., Salvi, A., Sinagra, G., and Giacca, M. (2004). Vascular endothelial growth factor stimulates skeletal muscle regeneration in vivo. Mol. Ther 10, 844–854.[CrossRef][Medline]

Bacic, M., Edwards, N. A., and Merrill, M. J. (1995). Differential expression of vascular endothelial growth factor (vascular permeability factor) forms in rat tissues. Growth Factors 12, 11–15.[Medline]

Bains, W., Ponte, P., Blau, H., and Kedes, L. (1984). Cardiac actin is the major actin gene product in skeletal muscle cell differentiation in vitro. Mol. Cell. Biol 4, 1449–1453.[Abstract/Free Full Text]

Ballaun, C., Weninger, W., Uthman, A., Weich, H., and Tschachler, E. (1995). Human keratinocytes express the three major splice forms of vascular endothelial growth factor. J. Invest. Dermatol 104, 7–10.[CrossRef][Medline]

Bloor, C. M. (2005). Angiogenesis during exercise and training. Angiogenesis 8, 263–271.[Medline]

Bluteau, G., Julien, M., Magne, D., Mallein-Gerin, F., Weiss, P., Daculsi, G., and Guicheux, J. (2007). VEGF and VEGF receptors are differentially expressed in chondrocytes. Bone 40, 568–576.[Medline]

Byun, J. H., Park, B. W., Kim, J. R., and Lee, J. H. (2007). Expression of vascular endothelial growth factor and its receptors after mandibular distraction osteogenesis. Int. J. Oral Maxillofac. Surg 36, 338–344.[CrossRef][Medline]

Castellon, R., Hamdi, H. K., Sacerio, I., Aoki, A. M., Kenney, M. C., and Ljubimov, A. V. (2002). Effects of angiogenic growth factor combinations on retinal endothelial cells. Exp. Eye Res 74, 523–535.[CrossRef][Medline]

Cheng, S. Y., Nagane, M., Huang, H. S., and Cavenee, W. K. (1997). Intracerebral tumor-associated hemorrhage caused by overexpression of the vascular endothelial growth factor isoforms VEGF121 and VEGF165 but not VEGF189. Proc. Natl. Acad. Sci. USA 94, 12081–12087.[Abstract/Free Full Text]

Christ, B., and Brand-Saberi, B. (2002). Limb muscle development. Int. J. Dev. Biol 46, 905–914.[Medline]

D'Amore, P. A. (2007). Vascular endothelial cell growth factor-a: not just for endothelial cells anymore. Am J. Pathol 171, 14–18.[Free Full Text]

Damert, A., Machein, M., Breier, G., Fujita, M. Q., Hanahan, D., Risau, W., and Plate, K. H. (1997). Up-regulation of vascular endothelial growth factor expression in a rat glioma is conferred by two distinct hypoxia-driven mechanisms. Cancer Res 57, 3860–3864.[Abstract/Free Full Text]

Darland, D. C., Massingham, L. J., Smith, S. R., Piek, E., Saint-Geniez, M., and D'Amore, P. A. (2003). Pericyte production of cell-associated VEGF is differentiation-dependent and is associated with endothelial survival. Dev. Biol 264, 275–288.[CrossRef][Medline]

Davis, R. L., Weintraub, H., and Lassar, A. B. (1987). Expression of a single transfected cDNA converts fibroblasts to myoblasts. Cell 51, 987–1000.[CrossRef][Medline]

During, M. J., and Cao, L. (2006). VEGF, a mediator of the effect of experience on hippocampal neurogenesis. Curr. Alzheimer Res 3, 29–33.[CrossRef][Medline]

Enholm, B., Karpanen, T., Jeltsch, M., Kubo, H., Stenback, F., Prevo, R., Jackson, D. G., Yla-Herttuala, S., and Alitalo, K. (2001). Adenoviral expression of vascular endothelial growth factor-C induces lymphangiogenesis in the skin. Circ. Res 88, 623–629.[Abstract/Free Full Text]

Esser, S., Wolburg, K., Wolburg, H., Breier, G., Kurzchalia, T., and Risau, W. (1998). Vascular endothelial growth factor induces endothelial fenestrations in vitro. J. Cell Biol 140, 947–959.[Abstract/Free Full Text]

Ferrara, N., Houck, K., Jakeman, L., and Leung, D. W. (1992). Molecular and biological properties of the vascular endothelial growth factor family of proteins. Endocr. Rev 13, 18–32.[CrossRef][Medline]

Ferrara, N., Carver-Moore, K., Chen, H., Dowd, M., Lu, L., O'Shea, K. S., Powell-Braxton, L., Hillan, K. J., and Moore, M. W. (1996). Heterozygous embryonic lethality induced by targeted inactivation of the VEGF gene. Nature 380, 439–442.[CrossRef][Medline]

Ferrara, N., and Davis-Smyth, T. (1997). The biology of vascular endothelial growth factor. Endocr. Rev 18, 4–25.[Abstract/Free Full Text]

Ferrara, N. (1999). Molecular and biological properties of vascular endothelial growth factor. J. Mol. Med 77, 527–543.[CrossRef][Medline]

Ferrara, N., Gerber, H. P., and LeCouter, J. (2003). The biology of VEGF and its receptors. Nat. Med 9, 669–676.[CrossRef][Medline]

Fons, P., Herault, J. P., Delesque, N., Tuyaret, J., Bono, F., and Herbert, J. M. (2004). VEGF-R2 and neuropilin-1 are involved in VEGF-A-induced differentiation of human bone marrow progenitor cells. J. Cell. Physiol 200, 351–359.[CrossRef][Medline]

Forsythe, J. A., Jiang, B. H., Iyer, N. V., Agani, F., Leung, S. W., Koos, R. D., and Semenza, G. L. (1996). Activation of vascular endothelial growth factor gene transcription by hypoxia-inducible factor 1. Mol. Cell. Biol 16, 4604–4613.[Abstract]

Germani, A., Di Carlo, A., Mangoni, A., Straino, S., Giacinti, C., Turrini, P., Biglioli, P., and Capogrossi, M. C. (2003). Vascular endothelial growth factor modulates skeletal myoblast function. Am J. Pathol 163, 1417–1428.[Abstract/Free Full Text]

Grosskreutz, C. L., Anand-Apte, B., Duplaa, C., Quinn, T. P., Terman, B. I., Zetter, B., and D'Amore, P. A. (1999). Vascular endothelial growth factor-induced migration of vascular smooth muscle cells in vitro. Microvasc Res 58, 128–136.[CrossRef][Medline]

Gruber, B. L., Marchese, M. J., and Kew, R. (1995). Angiogenic factors stimulate mast-cell migration. Blood 86, 2488–2493.[Abstract/Free Full Text]

Guan, F., Villegas, G., Teichman, J., Mundel, P., and Tufro, A. (2006). Autocrine VEGF-A system in podocytes regulates podocin and its interaction with CD2AP. Am J. Physiol. Renal Physiol 291, F422–F428.[Abstract/Free Full Text]

Gustafsson, T., and Kraus, W. E. (2001). Exercise-induced angiogenesis-related growth and transcription factors in skeletal muscle, and their modification in muscle pathology. Front. Biosci 6, D75–D89.[Medline]