- Email: [email protected]
Specialization of mitochondrial and vascular oxidant modulated VEGFR in the denervated skeletal muscle
Cellular Signalling 25 (2013) 2106–2114
Contents lists available at ScienceDirect
Cellular Signalling journal homepage: www.elsevier.com/locate/cellsig
Specialization of mitochondrial and vascular oxidant modulated VEGFR in the denervated skeletal muscle Hui Zhao a, Han-Wei Huang b,⁎, Jun-Guo Wu b, Pei-Yan Huang b a b
Department of Integrative Medicine and Neurobiology, National Key Lab of Medical Neurobiology, Institute of Brain Sciences, Fudan University, Shanghai, China Department of Orthopaedics, 5th Hospital of Shanghai, Fudan University, Shanghai, China
a r t i c l e
i n f o
Article history: Received 2 April 2013 Received in revised form 17 June 2013 Accepted 25 June 2013 Available online 2 July 2013 Keywords: Muscle denervation Mitochondria XO pathway VEGFR-1 VEGFR-2
a b s t r a c t Denervation of skeletal muscles results in timely muscular inflammation and muscle-T cell interaction, the cellular events might orchestrate a local circuit involved with IL-1β and IL-15. In the present study, by a combination assay of nerve–muscle preparation, western blot, immuno-precipitation, and radioactive of enzyme activity, we confirmed that mitochondrial and vascular oxidants were considerably up-regulated following gastrocnemius denervation, which was due to gradual decay in mitochondrial biogenesis and XO pathway and accompanied by strengthened IL-1β-VEGFR-2 and IL-15-VEGFR-1 signaling. Intriguingly, these alterations could be triggered by the early established muscular inflammation. In contrast, with prolonged muscle denervation, settings of organelle interconnection were ultimately conveyed by ER bound PTP1B, which promoted VEGFR-1 signaling and contributed to VEGFR-2 activation, and the process could be modulated by mitochondrial and vascular oxidant. Importantly, VEGFR-2 could rescue the disruption of MuSK activity and AchR cluster exerted by IL-1β and IL-15, with PGC-1α and XO involvement. Altogether, extensive network centered on VEGFR-2 signaling was essentially contributed to early recovery processes regarding muscle denervation. Increasing knowledge of this mechanism might open up a conduit for functional response to muscle atrophy, and enable the development of better agents to combat the related disorders. © 2013 Published by Elsevier Inc.
1. Introduction Previously, we reported that muscle denervation could produce a stereotypic inflammation and muscle-T cell interaction [1,2], and the produced IL-1β and IL-15 were proposed to contribute to the complex and dynamic micro-environment [3–9]. As reported, in skeletal muscle, blood flow was closely related with muscle activity, and there was an increase in capillary density per area of muscle (capillary-to-fiber ratio, C:F) following exercise or exposure to hypoxia. Also, the anatomical pattern of muscle microvascularization changed dramatically after nerve transection, with defected capillaries served by muscle fibers [10–12]. VEGF, a 35- to 45-kDa peptide growth factor, has proved to be essential for the development of capillary networks during embryogenesis and the maintenance of capillary structures [13–16]. In skeletal muscle, deletion of VEGF gene resulted in greatly reduction in muscle capillarity [17,18]. Then, it was proposed that the VEGF signaling might be integral components for remodeling of vascular bed or impairment of microcirculation in the pathogenesis of muscle denervation. As well established, between two related but distinct receptors: VEGFR-1 and VEGFR-2, VEGF signaling through VEGFR-2 was the major pathway that activated angiogenesis by inducing the proliferation, ⁎ Corresponding author at: Department of Orthopaedics, 5th Hospital of Shanghai, Fudan University, 128# Ruili Rd. Shanghai, China. Tel.: +86 21 64308151. E-mail address: [email protected] (H.-W. Huang). 0898-6568/$ – see front matter © 2013 Published by Elsevier Inc. http://dx.doi.org/10.1016/j.cellsig.2013.06.014
survival, sprouting and migration of ECs [15], and therapeutic targeting of this pathway was already used to delay tumor related vasculature [19]. In contrast, VEGFR-1 displayed stable ER-Golgi localization, upon stimuli, and VEGFR-1 could be preferential sequestrated and seemed to act as a negative regulator of VEGF-mediated angiogenesis [20,21]. Recently, it has been suggested that Src kinases represented the predominant protein tyrosine kinase activity in VEGFR activation and MuSK directed AChR clustering [22,23], and the overlap modulation was proposed to form a viable link between functional vasculature and neuromuscular junction stabilization during muscle denervation. It was already identified that XO resided in the cytosol and was the primary source of ROS in the vasculature. Particularly, XO was predictive of superior potency for vascular inflammation and oxidative capacity in skeletal muscle [24–26], which could be up-regulated by proinflammatory stimuli (lipopolysaccharide, TNF-α, IL-1, and IL-6) [27,28]. Besides that, many studies addressed that muscle atrophy was associated with mitochondrial dysfunction and resulted in increased rate of mitochondrial reactive oxygen species production [29–32]. When rats were subjected to unilateral sciatic nerve transection, decrements in mitochondrial content, enzymatic activity of respiratory chain complexes, as well as the rate of mitochondrial ATP production were significantly decreased, which led to reduced muscle mass [33,34]. As such, the present study designed to examine whether XO activity and mitochondrial biogenesis could be triggered by muscle denervation, which was presumed to link to changes in VEGF-VEGFR signaling,
H. Zhao et al. / Cellular Signalling 25 (2013) 2106–2114
understanding the mechanism underlying these alterations would enable the development of better agents to combat muscular disorders. 2. Materials and methods 2.1. Tissue harvest Adult SD rats (160–180 g, all male) were divided into five groups (n = 5 in each group): unoperated controls, animals after 4 h, 1, 3 or 7 days following sciatic nerve transection. For muscle denervation, the right sciatic nerve was surgically exposed and transected after anesthesia with pentobarbital sodium (35 mg/kg, intraperitoneally). At the specified time-points the animals were euthanized, and a 1 × 1 cm block of gastrocnemius was removed and cleaned of fat and connective tissues, weighed, and stored for later analysis. Body temperature was maintained at 37°C throughout surgery, and there was no evidence of postoperative infection. This study was approved by Ethics Committee of Fudan University. All animal procedures were performed in strict accordance with the U.S. National Institutes of Health's Guide for the Care and Use of Laboratory Animals. 2.2. Myeloperoxidase MPO activity was measured spectrophotometrically. Briefly, samples were incubated with a substrate o-dianisidine hydrochloride. This reaction was carried out in a 96-well plate by adding 290 μl, 50 mM phosphate buffer, 3 μl substrate solution (containing 20 mg/ml o-dianisidine hydrochloride), and 3 μl H2O2 (20 mM). Sample (10 μl) was added to each well to start the reaction. Standard MPO (Sigma Chemical Co., St. Louis, MO) was used in parallel to determine MPO activity in the sample. The reaction was stopped by adding 3 μl sodium azide (30%). Plates were read at 460 nm. MPO activity was defined as nmol/mg protein. 2.3. Oxidant production Gastrocnemius was homogenized in 0.1 M phosphate buffer (pH 7.4) with a motor-driven Polytron glass homogenizer at 4 °C. The resulting homogenate was filtered through four layers of medical gauze to remove connective tissue debris. Muscle oxidant generation was determined within 30 min of tissue harvest in fresh tissue homogenates using dichlorofluorescein (DCF) as a probe. The basal buffer consisted of 0.1 M potassium phosphate at pH 7.4. The induced buffer additionally included 1.7 mM ADP, 0.1 mM NADPH, and 0.1 mM FeCl3. Both buffers included 5 μM 2′,7′-dichlorofluoresceindiacetate (DCFH-DA), which was made fresh in 1.25 mM methanol and kept in a dark room at 0 °C. For basal condition, following reagents were added: 2938 μl 0.1 M phosphate buffer, 50 μl filtered muscle homogenate, and 12 μl 1.25 mM DCFH-DA. Induced condition was replaced 90 μl of buffer with 50 μl FeCl3, 20 μl ADP, and 20 μl NADPH. Total reaction volume was 3.0 ml. A blank consisting of the appropriate buffer and 5.0 μM DCFH-DA without sample was used to correct for the auto-oxidation rate of DCFH-DA. The mixture was incubated in the dark for 15 min at 37 °C. DCF formation was determined with a Hitachi F-2000 fluorescence spectrophotometer (Hitachi Instruments, Co., San Jose, CA) with a thermostated cell compartment at excitation wavelength 488 nm and emission wavelength 525 nm. The units were expressed as pmol DCF/mg protein. 2.4. XO/XDH activity assay Measurement of XO and XDH activity in kidney tissue was based on the pterin-based assay. In brief, approximately 40 mg of tissues was homogenized in 1 ml assay buffer (50 mM K-phosphate, 1 mM ethylene-diaminetetraacetic acid, 0.5% dimethyl sulfoxide, and protease
2107
inhibitor cocktail, pH 7.4). The supernatant (150 μl) was co-incubated with 50 μl pterin solution (final concentration of 50 μM) or pterin with methylene blue solution (final concentration of 50 μM) to assay XO or both XO and XDH activities, respectively. Before and after 120-min incubation at 37°C, fluorometric assays were performed to calculate the production of isoxanthopterin. Protein concentration was measured by Pierce BCA Protein Assay Kit (Thermo Scientific Inc., Billerica, MA, USA). 2.5. Nerve–muscle preparations Gastrocnemius muscles, with their respective motor nerve, were dissected from SD rats under deep anesthesia. Nerve–muscle preparations were pinned in an experimental Sylgard-coated recording chamber filled with normal Ringer's solution containing (in mM): 124 NaCl, 5 KCl, 1.25 NaH2PO4, 2 MgCl2, 1NaHCO3, 2 CaCl2, 25 HEPES and 10 glucose, oxygenated with 100% O2. Treatments included in IL-1β (R&D systems, Minneapolis, MN, 20 ng/ml, 24 h), IL-1ra (10 ng/ml, 24 h), recombinant IL-15 (10μg/ml, 24 h, BioLegend, San Diego, CA), XO inhibitor allopurinol (1.0 mM, 24 h), PGC-1α antibody (20μg/ml, 24 h), or adenovirus infection (VEGFR-1/VEGFR-2, 5 × 109 plaqueforming units (PFU) dissolved in sterilized PBS, 72 h). A 2–3 mm broad band of muscle around innervated muscle was excised, representing the junctional region, which was used for the indicated assays. 2.6. Immunohistochemistry Serial sections (5 μm thick) of gastrocnemius were sectioned on a cryostat at 15 μm. Serial cross-sections were mounted on glass slides (Superfrost Plus) and labeled with antibody against VEGF (1:1000, Abcam, Cambridge, MA). Antibodies were diluted in 0.01 M PB with 3% normal goat serum, 1% BSA, 0.5% Triton X-100, and 0.05% sodium azide, pH 7.4. The primary antibodies were revealed using ABC kit. Sections were viewed under microscope (Leica L2000A). Images were acquired and processed by Leica QWin software. In separate experiments, we also tested different antigen retrieval techniques including microwave heating and SDS-pretreatment. 2.7. Discontinuous gradient centrifugation for lipid raft extraction Gastrocnemius was washed twice with phosphate-buffered saline (PBS, pH 7.4) at 4°C and then homogenized in 2 ml of 500 mM NaCO3 (pH 11.0) solution with complete protease inhibitor mixture (Roche Applied Science, Indianapolis, Indiana). The tissue suspension was further sonicated with one 30-s burst at setting 4 and one 30-s burst at setting 7 (Model W-220F; Ultrasonics, Inc., Bartlett, Illinois). The homogenate was adjusted to 45% sucrose by adding 2 ml of 90% (w/v) sucrose prepared in MBS buffer (25 mM MES, pH 6.5, and 0.15 M NaCl) and placed at the bottom of an ultracentrifuge tube. Four milliliters of 35% sucrose and 4 ml of 5% sucrose (both in MBS buffer containing 250 mM sodium carbonate) were then overlaid upon the sample to form a 5–45% discontinuous sucrose gradient. The sample was centrifuged at 32,000 rpm for 16 h in a Beckman ultracentrifuge in an SW-41Ti rotor. A light-scattering band at the 5% and 35% sucrose interface (Fraction 4–5) represented the lipid raft fractions, which was collected and used for the indicated assay. 2.8. Immunoprecipitation and western blotting Gastrocnemius was homogenized in lysis buffer containing 10 mM Tris–HCl (pH 7.5), 50 mM NaCl, 0.1% Triton X-100, and one complete protease inhibitor cocktail tablet (Roche Applied Science, Mannheim, Germany). After centrifugation at 14,000 rpm for 5 min, the supernatants were incubated with anti-pTyr, PTP1B (1:200; BD Transduction Laboratories, Lexington, Kentucky), anti-AchRα antibody
2108
H. Zhao et al. / Cellular Signalling 25 (2013) 2106–2114
(1:200; Santa Cruz Biotechnologies, Santa Cruz, CA at 4°C overnight with slow rotation. Sixty microliters of protein G-agarose beads (Invitrogen, Carlsbad, California) was added, and the mixture was further incubated at 4°C for 3 h with slow rotation. The protein G-agarose beads were then pelleted by centrifugation at 12,000 g for 15 min at 4°C and washed five times with wash buffer (50 mM Tris–HCl (pH 8.0), 150 mM NaCl). For western blot analysis, protein samples were resolved by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to polyvinylidene difluoride membranes (GE Healthcare, Little Chalfont, Buckinghamshire, UK). Membranes were probed with anti-caveolin-3 (1:1000, BD Transduction Laboratories), anti-VEGFR-1, anti-VEGFR-2 antibodies (1:500, Abcam), anti-AchRβ antibody (1:500; Santa Cruz Biotechnologies). Protein bands were detected using alkaline phosphatase conjugated secondary antibodies (1:5000) and ECF substrate (GE Healthcare, Piscataway, NJ), and then scanned using a Storm 860 imaging system (GE Healthcare). Band intensities were quantified and analyzed with ImageQuant software (GE Healthcare). PGC-1α (1:1000, Abcam, Cambridge, MA). 2.9. TaqMan Reverse Transcription (RT)-PCR for miRNA quantification Total RNA was isolated from gastrocnemius (50 mg) with TRIzol™ (Invitrogen, Carlsbad, VA) according to manufacturer's protocol. miR23a quantification was carried out by reverse transcribing total RNA using Taqman™ microRNA reverse transcription kit and subjected to real-time PCR using TaqMan™ MicroRNA Assay kit (Applied Biosystems, Carlsbad, CA). Reactions were performed using Stratagene Mx3000 instrument in triplicate. Real-time PCR data was analyzed using a ΔΔCt calculation. A p value of less than 0.05, when considering treated animals or cells vs. control group, was considered significant. 2.10. In vitro Src kinase assay Proteins in gastrocnemius lipid rafts or anti-MuSK antibody immunoprecipitation (1:200; Abcam) were precipitated with 5% (w/v) TCA. The resulting pellets were washed with acetone and incubated at 30°C with 5 μg of SRC substrate peptide (KVEKIGEGTYGVVYK, corresponding to amino acids 6–20 of p34cdc2; Upstate Biotechnology, Lake Placid, New York) in kinase buffer containing 5 μCi of [γ-32P]-adenosine triphosphate ([γ-32P]-ATP; PerkinElmer Life Sciences, Waltham, Massachusetts), 50 mM Tris–HCl (pH 7.5), 10 mM MgCl2, 10 mM MnCl2, 25μM ATPase, 1 mM dithiothreitol, and 100μM Na3VO4. After 30 min, the reaction was terminated by the addition of 10 μl of 40% (w/v) TCA, and samples were spotted on P81 cellulose phosphate paper (Upstate Biotech). The paper was washed three times with 1% (w/v) phosphoric acid and once with acetone. Radioactivity retained on the P81 paper was quantified by liquid scintillation counting. Blank counts (without tissue lysate) were subtracted from each result, and radioactivity (cpm) was converted to picomoles per minute (pmol/min). 2.11. PTP 1B assay Gastrocnemius was collected and homogenized in RIPA buffer (50 mM Tris [pH7.5], 150 mM NaCl, 2 mM EDTA, 1.0% Triton X-100 and complete protease inhibitor mixture). Equal amounts of protein were incubated at 4 °C for 2 h with anti-PTP1B antibody and then, 20 μl of protein G sepharose was added and incubated for additional 2 h with mixing. Immunoprecipitated complexes were washed twice in RIPA buffer, once with the assay buffer (25 mM imidazole [pH7.2], 0.1 mg/ml BSA, 10 mM DTT) and finally they were resuspended in 25 μl of assay buffer. The phosphatase substrate raytide was labeled at its unique tyrosine residue in the presence of [γ-32P] ATP. Assay mixtures (30 μl) containing the immunoprecipitated pellet and [γ-32P]-labeled raytide (1 × 105 cpm) were incubated at 30 °C for
2 h and the reaction was terminated by adding 750 μl of a charcoal mixture (0.9 M HCl, 90 mM sodium pyrophosphate, 2 mM NaH2PO4, 4% vol/vol Norit A). After centrifugation, the radioactivity in 400 μl of the supernatant was measured by scintillation counting. Blanks were determined by measuring the free γ-32P in reactions where the immunoprecipitates were either boiled or omitted, and these values were subtracted from the reaction values. 2.12. Recombinant adenoviruses Recombinant adenovirus expressing rat VEGFR-1 or VEGFR-2 was constructed by inserting into the adenoviral shuttle vector pDE1sp1A (Microbix Biosystems, Inc. Canada), and the insert was then switched to the adenoviral vector through LR recombination. Adenovirus was purified by CsCl2 gradients and PD-10 Sephadex chromatography. 2.13. Enzyme-linked immunoassay analysis of VEGFR levels Levels of VEGFR-1/2 production were measured by sandwich enzyme-linked immunoassay (ELISA) according to manufacturer's instructions (R&D Systems, Inc., Minneapolis, MN). A 96-well plate was coated with 2 μg/ml monoclonal anti-VEGFGR-1/2 antibodies at 4°C overnight and then blocked with 1% bovine serum albumin (BSA) in PBS for 1 h. The plates were washed three times with PBS containing 0.2% Tween 20 (PBST). Aliquots of tissue lysates were diluted to100μl with Hanks' balanced salt solution (HBSS) with calcium and magnesium, 10 mM 4-(2-hydroxyethyl)-1piperazineethanesulfonic acid (HEPES), and 1% fetal bovine serum, added to the plates, and then incubated for 2 h at room temperature. The plates were washed three times with PBS, and 100 μl aliquots of 0.1 μg/ml biotinylated VEGFR-1/2 affinity purified polyclonal antibody were added and incubated for 2 h. After further three washes with PBST, the immune complexes were colorimetrically detected using horseradish peroxidase (HRP)-streptavidin conjugate. The reaction was halted by the addition of 1 M H2SO4, and the absorbance at 450 nm was measured using a microplate reader (VersaMax; Molecular Devices, Sunnyvale, CA). 2.14. Statistical analyses All experiments were performed using 5 animals per group (n = 5). Data were represented as mean ± SEM and analyzed with Prism 5 software. For all data sets, normality and homocedasticity assumptions were reached, validating the application of the one-way ANOVA, followed by t test for multiple comparisons. Differences were considered significant for p b 0.05. 3. Results 3.1. Mitochondrial and vascular oxidant initiated during gastrocnemius denervation Rats were undergone sciatic nerve transection, 4 h, 1, 3 and 7 days later, and MPO activity, a granular-based heme enzyme present in neutrophils and macrophages, was examined in the denervated gastrocnemius. As shown in Fig. 1A, MPO activity was robustly elevated by factor of 3.3 and 3.4 respectively after 4 h and 1 day of operation. Afterwards, they were gradually decreased to control level. Furthermore, dynamic alteration in oxidant production in the denervated muscle was measured using DCF probe, in which the basal buffer condition (0.1 M potassium phosphate) measured predominantly nonmitochondrial sources of oxidant generation and maximum one could be stimulated by the induced buffer (including 1.7 mM ADP, 0.1 mM NADPH, 0.1 mM FeCl3). As illustrated, oxidant production both in the basal (Fig. 1B) and induced buffers (Fig. 1C) were rapidly rose around 2 folds over control at day 1,and the increase was
H. Zhao et al. / Cellular Signalling 25 (2013) 2106–2114
2109
Fig. 1. Mitochondrial and vascular oxidant initiated during gastrocnemius denervation. Rats were killed 4 h, 1, 3 and 7 days after sciatic nerve transection (n = 5 for each group). MPO activity in gastrocnemius homogenates was measured. data was converted to nmol/mg protein (A). Rats were killed 1, 3 and 7 days after sciatic nerve transection (n = 5 for each group), oxidant production in induced or basal buffer (B and C). XO and XOH activity were quantified (D). Data was converted to pmol or μU/mg protein. Nerve–muscle preparation was cultured for 1 day, then exposed to IL-1β or IL-15 for 24 h, DCF production in induced buffer (E and G) and XO/XDH activity were quantified (F and H). Data was converted to pmolor μU/mg protein. Values represent mean ± SEM for 5 independent experiments. *p b 0.05 vs Con. #p b 0.05 vs IL-1β or IL-15. Con: control; 4 h, 1, 3, 7d: 4 h, 1, 3, 7 days after surgery.
sustained up to 7 days following muscle denervation. Also, similar alterations occurred on XO and Xanthine dehydogenase (XDH) activity. Fig. 1D illustrated that they were up-regulated around 2 folds over control following muscle denervation. Given our previous description that gastrocnemius denervation could produce a stereotypic alteration of IL-1β involved inflammation and muscle-T cell interaction initiated IL-15 production, we consequently demonstrated in nerve–gastrocnemius preparation that IL-1β exposure resulted in an increase in maximum oxidant production (Fig. 1E), in parallel, XO and xanthine dehydogenase (XDH) activity rose to 3.3 and 3.1 folds over control (Fig. 1F). Likewise, it was demonstrated that IL-15 incubation could also specifically led to dramatic elevation of maximum oxidant production, and XO and XDH activities (Fig. 1G, H).
3.2. VEGF-VEGFR signaling orchestrated by gastrocnemius denervation Vascular endothelial growth factor (VEGF) was noticed recently to be crucial factor during muscle denervation [17,18]. Consequently, it was shown in Fig. 2A and B, by immunohistochemistry, that VEGF immuno-positive signals were quite weak in control group, however, they were robustly increased and widely distributed in gastrocnemius at day 1 following muscle denervation. Quantification analysis revealed that the expression levels rose to 4.3 folds over control. By subcellular fractionation and western blot analysis, it was demonstrated that VEGFR-2 expression within the lipid rafts was enhanced rapidly in the denervated muscle. Levels of expression rose to 4.3 folds over control at day 1. Afterwards, they were gradually decreased. Comparably, total VEGFR-1 expression was dramatically increased at day 7, and expression level rose to 3.5 folds over control (Fig. 2C). Moreover, in nerve–gastrocnemius, we found that IL-1β exposure resulted in the greatly increase in VEGFR-2 expression within the lipid rafts, and the effect could be partly inhibited by XO inhibition. In contrast, there were no detectable changes in total VEGFR-1 expression (Fig. 2D). For IL-15 incubation, VEGFR-2 expression within the lipid rafts did not display any significant alteration, while total VEGFR-1 expression dramatically rose to 3.6 folds over control. Similarly, they partly exhibited attenuation after XO inhibition (Fig. 2E).
3.3. The role of mitochondrial and vascular oxidant on VEGFR signaling during gastrocnemius denervation A number of physiological studies have established a role for peroxisome proliferators-activated receptor coactivator-1 α (PGC-1α) in mediating mitochondrial biogenesis, which was a target of miR23a and generally accompanied by expression of genes related with muscle regeneration [35–37]. By real-time PCR, and western blot analysis, Fig. 3A illustrated that miR23a expression was progressively up-regulated by factor of 1.2, 1.4 and 1.8 respectively at day 1, 3 and 7 following gastrocnemius; parallel to these changes, PGC-1α expression were dramatically reduced to 68.3, 61.1 and 50.1% of control respectively. Furthermore, in nerve–gastrocnemius preparation, it was demonstrated that VEGFR-2 was up-regulated by IL-1β, which could be partly attenuated by anti-miR23a or PGC-1α antibody, but completely reduced by concurrent treatment of PGC-1α antibody and XO inhibitor (Fig. 3B). Likewise, VEGFR-1 could be increased by IL-15, which was disrupted by treatment with PGC-1α antibody and XO inhibitor concurrently but not individually. Intriguingly, IL-1β could not result in dramatic change in VEGFR-1, and IL-15 on VEGFR-2 expression respectively (Fig. 3C). 3.4. Timely modulation of c-Src activity during gastrocnemius denervation It was proposed that the integrated VEGFR signaling might be centered on c-Src activation which was linked to spatial and temporal segregation of protein phosphatase. We then detected Src activity within the lipid rafts by [γ-32p] incorporation. As shown in Fig. 4A, Src kinase activity was rapidly increased around 4.0 folds over control at day 1 following gastrocnemius denervation, which was gradually returned to control level. PTP1B activity was also investigated by [γ-32p] incorporation. As demonstrated, PTP1B activity was remained at control level at day 1 and 3 following surgery, however, by day 7, which it was considerably strengthened to 3.7 folds over control (Fig. 4B). In nerve–gastrocnemius preparation, IL-1β exposure resulted in a great increase in c-Src but not PTP1B activity, and the effect was completely blocked by concurrently treatment with PGC-1α antibody
2110
H. Zhao et al. / Cellular Signalling 25 (2013) 2106–2114
Fig. 2. VEGF-VEGFR signaling orchestrated by gastrocnemius denervation. Rats were killed 1, 3 and 7 days after sciatic nerve transection (n = 5 for each group). Consecutive cryosections of gastrocnemius were immuno-stained with antibody against VEGF, scale bar = 200 μm (A). Panel B depicts quantitative analysis of VEGF-staining intensities. Gastrocnemius was collected and homogenated, VEGFR-2 within the lipid rafts and total VEGFR-1 expression were determined by western blot analysis (C). Nerve–muscle preparation was cultured for 1 day, then exposed to IL-1β or IL-15 for 24 h, VEGFR-2 within the lipid rafts and total VEGFR-1 expression in the presence/absence of XO inhibitor were determined by western blot analysis (D and E). Data were normalized and calculated as percentage of control. Each value represents mean ± SEM for 5 independent experiments. *p b 0.05 vs Con. #p b 0.05 vs IL-1β or IL-15.
and XO inhibitor (Fig. 4C). Instead, IL-15 led to robust up-regulation of PTP1B but not c-Src activity, and the effect was attenuated by PGC-1α antibody and XO inhibitor (Fig. 4D). 3.5. Association of c-Src activity with VEGFR signaling during gastrocnemius denervation VEGFR signaling has been conclusively identified to be modulated by tyrosine phosphorylation. Consequently, by immunoprecipitation in which anti-pTyr as immunoprecipitated antibody and antiVEGFR-2 as immunoblot antibody, it was shown that tyrosine phophorylated VEGFR-2 rose to around 3.6 folds over control at day 1 following gastrocnemius denervation, which was gradually decreased afterwards. Comparably, there were no dramatic changes in VEGFR-1 signaling (Fig. 5A). In nerve–gastrocnemius preparation, by immunoprecipitation, it was shown that tyrosine phophorylated VEGFR-2 was robustly up-regulated around 3.3 folds over control after IL-1β but not IL-15 exposure, and the effect was blocked by both PGC-1α antibody and XO inhibitor (Fig. 5B and D). In contrast, the interaction of PTP1B and VEGFR-1 was increased by IL-15 but not IL-1β incubation. Coincidently, the effect was dependent on PGC-1α and XO (Fig. 5C and E). 3.6. Modulation of neuromuscular junction by VEGFR signaling Neuromuscular junction might be initiated by local environment, which was proposed to be related with mitochondrial and vascular oxidative responses. As demonstrated in Fig. 6A, junctional PGC-1α expression was robustly increased around 4.0 folds over control at day 1 following gastrocnemius denervation. After that, they were
gradually decreased. In the same time, junctional XO and XDH activities were rapidly up-regulated around 3.5 and 3.7 folds over control (Fig. 6B). In nerve–gastrocnemius preparation, by ELISA assay, we demonstrated that junctional VEGFR-2 could be up-regulated by IL-1β while VEGFR-1 by IL-15, and the effects were partly or completely reversed by PGC-1α antibody and XO inhibitor individually or concurrently (Fig. 6C and D). As well established, NMJ was a peripheral cholinergic synapse that conveys signals from the motor neurons to the muscle cells, which required muscle-specific kinase (MuSK) activation and AchR cluster. In the present observation, MuSK activity and AchR cluster were slightly reduced by IL-1β exposure. In contrast, IL-15 led to their dramatically attenuation: MuSK activity decreased to 56.3 ± 5.7% control, and AchR cluster to 58.8 ± 7.9% control. These effects could be rescued by adenovirus over-expressing VEGFR-2 but not VEGFR-1 (Fig. 6E and F). 4. Discussion In previous study, we characterized that rapid muscular inflammatory and muscle-T cell interaction could be subsequently initiated by gastrocnemius denervation, and the produced IL-1β and IL-15 were contributed primarily to the dynamic microenvironment [1,2]. Herein, we further demonstrated that there were sustainable upregulation of oxidant generation including basal and mitochondrial oxidant production and XO/XDH activity following muscle denervation. Coincidently, in nerve–muscle preparation, following IL-1β and IL-15 treatment, both mitochondrial oxidant production and XO/XDH activity were considerably increased. Therefore, these enzymatic changes reasonably reflected distinct variables in oxidative
H. Zhao et al. / Cellular Signalling 25 (2013) 2106–2114
2111
Fig. 3. The role of mitochondrial and vascular oxidant on VEGFR signaling during gastrocnemius denervation. Rats were killed 1, 3 and 7 days after sciatic nerve transection (n = 5 for each group). Gastrocnemius was collected and homogenized. miR23a and PGC-1α were detected using real-time PCR and western blot analysis respectively (A). Nerve–muscle preparation was cultured for 1 day, then exposed to IL-1β or IL-15 for 24 h, VEGFR-2 within the lipid rafts and total VEGFR-1 expression in the presence/absence of XO inhibitor, or anti-miR23a/PGC-1α antibody, was determined by western blot analysis (B and C). Data were normalized and calculated as percentage of control. Each value represents mean ± SEM for 5 independent experiments. *p b 0.05 vs Con. #p b 0.05 vs IL-1β or IL-15.
state during muscle denervation [24,26,38]. We then proposed a teleological argument that reactive oxygen species were involved in both rapid muscular inflammation and later muscle-T cell interaction.
It was already reported that muscle atrophy was associated with mitochondrial dysfunction and resulted in an increased rate of mitochondrial reactive oxygen species production [29–32], decrements in
2112
H. Zhao et al. / Cellular Signalling 25 (2013) 2106–2114
Fig. 4. Timely modulation of c-Src activity during gastrocnemius denervation. Rats were killed 1, 3 and 7 days after sciatic nerve transection (n = 5 for each group). Gastrocnemius was homogenized and separated by discontinuous sucrose centrifugation. SRC kinase activity within the lipid rafts was assayed using an SRC-specific substrate peptide (A). In the same time, PTP1B activity in gastrocnemius was determined by PTP1B-specific substrate peptide (B). Nerve–muscle preparation was cultured for 1 day, then exposed to IL-1β or IL-15 for 24 h. Src activity within the lipid rafts and PTP1B activity in the presence/absence of individual or concurrent of XO inhibitor and PGC-1α antibody were determined by [32p] incorporation (C, D). Data were normalized and calculated as percentage of control, each value represents mean ± SEM for 5 independent experiments. *p b 0.05 vs Con. # p b 0.05 vs IL-1β or IL-15.
mitochondrial content, enzymatic activity of respiratory chain complexes. Particularly, the rate of mitochondrial ATP production was significantly decreased following sciatic nerve transection, which was in parallel with the reduction of muscle mass by 40–65% [33,34]. Besides
that, XO pathways could also be triggered by muscular disorder, which was associated with vascular oxidative stress [39–41]. Two inter-convertible forms known as XO and XDH could oxidize NADH and resulted in ROS formation [42]. Therefore, our observation, in
Fig. 5. Association of c-Src activity with VEGFR signaling during gastrocnemius denervation. Rats were killed 1, 3 and 7 days after sciatic nerve transection (n = 5 for each group). Gastrocnemius was collected and homogenized. Immuno-precipitation was used to detect VEGFR phosphorylation. The immuno-precipitated antibody was anti-c-pTyr and the immuno-blotting antibody was VEGFR-1/2 (A). Nerve–muscle preparation was cultured for 1 day, then exposed to IL-1β or IL-15 for 24 h in the presence/absence of individual or concurrent of XO inhibitor and PGC-1α antibody. Immuno-precipitation was used to detect VEGFR phosphorylation. The immuno-precipitated antibody was anti-c-pTyr and the immuno-blotting antibody was VEGFR-1/2 (B, D). Immuno-precipitation was used to detect the interaction of PTP1B and VEGFR-1. The immuno-precipitated antibody was anti-PTP1B and the immuno-blotting antibody was VEGFR-1 (C and E). Data were normalized and calculated as percentage of control. Each value represents mean ± SEM for 5 independent experiments. *p b 0.05 vs Con. #p b 0.05 vs IL-1β or IL-15.
H. Zhao et al. / Cellular Signalling 25 (2013) 2106–2114
2113
Fig. 6. Modulation of neuromuscular junction by VEGFR signaling. Rats were killed 1, 3 and 7 days after sciatic nerve transection (n = 5 for each group). Gastrocnemius NMJ was separated and homogenized. PGC-1α expression and XO/XDH activity were determined by western blot analysis and fluorometric assays respectively (A, B). Nerve–muscle preparation was cultured for 1 day, then exposed to IL-1β or IL-15 for 24 h. In the presence/absence of individual or concurrent of XO inhibitor and PGC-1α antibody, junctional VEGFR-1/2 was determined by ELISA assay (C, D). Nerve–muscle preparation was infected with adeno-VEGFR-1 or VEGFR-2 for 48 h, then exposed to IL-1β or IL-15 for another 24 h. Junctional MuSK was determined by in vitro kinase assay, AchR cluster by immuno-precipitation in which AchRα was used as immuno-precipitated antibody, AchRβ as immuno-blotting antibody (E, F). Data were normalized and calculated as percentage of control. Each value represents mean ± SEM for 5 independent experiments. *p b 0.05 vs Con. #p b 0.05 vs IL-1β or IL-15.
conjunction with above findings, implied that mitochondrial and vascular functions were actually targeted and appeared to be associated with processes initiated by muscle denervation. We further demonstrated that VEGF expression was up-regulated at day 1 following muscle denervation, and immuno-positive densities were mainly localized to the capillary area. Also, there was a timely up-regulation of VEGFR-2 (1 day) and VEGFR-1 (7 day). Intriguingly, mitochondrial biogenetic gene miR23a was increased and its target PGC-1α was decreased. All above alterations could essentially be affected by IL-1β and IL-15 respectively. As such, the upper estimate was that muscle denervation triggered a general mitochondrial transcription [43–47], which together with vascular oxidant, contributed to the dynamic and complex alterations in VEGFR signaling. Notably, the event linked to VEGFR-2 tyrosine phosphorylation was the activation of vasodilation, permeability and angiogenic processes [48]. In contrast, VEGFR-1 displayed stable ER-Golgi localization that could be related with PTP1B [49–51]. In the present study, we demonstrated that c-Src and PTP1B activities were subsequently activated during muscle denervation, and the association of VEGFR-2 and c-Src, VEGFR-1 and PTP1B underwent vigorous up-regulation in consequence. Intriguingly, these alterations could
be inferred to IL-1β and IL-15 respectively and involved XO pathway and mitochondrial transcription. Then, the data indicated a dual VEGFR system, in which up-regulation of VEGFR-2 might result in temporary vasculogenesis or increase in vascular permeability in the early stage of muscle denervation; however, along with T cells invasion into the muscle, VEGFR-1 could be targeted and stimulated, which seemed to related to PTP1B involved organelle interconnection. A substantial body of data already shed light on the strong association of VEGFR signaling with muscle function, for example, VEGFR-2 appeared to be the main signaling route [52–54], whose signaling might overlap with alterations in motor endplate and terminal axons, thus precluding characterization of neuromuscular junction (NMJ) stability during muscle denervation [55,56]. Coincidently, we demonstrated herein that junctional PGC-1α expression and XO/XDH activity were rapidly increased when challenged with muscle denervation. Besides that, the alterations involved VEGFR-2 or -1 could be modulated by IL-1β or IL-15. Importantly, VEGFR-2 functioned to rescue the damage on NMJ exerted by IL-1β or IL-15. Then, we realized that the damage on muscular function could be deteriorated along with muscle denervation, and VEGFR-2 was proposed to be important for the maintenance of NMJ.
2114
H. Zhao et al. / Cellular Signalling 25 (2013) 2106–2114
5. Conclusion Collectively, we demonstrated a dramatic up-regulation of mitochondrial and vascular oxidant production following muscle denervation, which was due to gradual disorder in mitochondrial biogenesis and XO/XDH activity, and accompanied by timely up-regulation of VEGFR-1 and VEGFR-2. It is of interest in this context that VEGFR signaling could be specifically initiated by IL-1β and IL-15 with involvement of mitochondrial and vascular oxidant production. Moreover, we demonstrated a preferentially disruption of NMJ function by IL-15, which could be rescued by VEGFR-2 but not VEGFR-1 activation. Thus, extensive network centered on VEGFR-2 signaling might be a key component necessary for the early recovery processes regarding muscle denervation, which was considered a strong attractant for the treatment on muscular disorders.
References [1] H. Hanwei, H. Zhao, Muscle & Nerve 42 (2010) 70–77, (Pubmed: 20544909). [2] H. Zhao, H. Huang, Cytokine 60 (2012) 608–615, (Pubmed: 23017227). [3] M.H. Laughlin, R.B. Armstrong, American Journal of Physiology 243 (1982) H296–H306, (Pubmed: 7114239). [4] P. Andersen, B. Saltin, The Journal of Physiology 366 (1985) 233–249, (Pubmed: 4057091). [5] J.M. Dawson, O. Hudlicka, International Journal of Microcirculation, Clinical and Experimental 8 (1989) 53–69, (Pubmed: 2722411). [6] X.J. Musacchia, J.M. Steffen, R.D. Fell, M.J. Dombrowski, Journal of Applied Physiology 69 (1990) 2248–2253, (Pubmed: 2077023). [7] B.M. Hather, G.R. Adams, P.A. Tesch, G.A. Dudley, Journal of Applied Physiology 72 (1992) 1493–1498, (Pubmed: 1534323). [8] O. Hudlicka, M. Brown, S. Egginton, Physiological Reviews 72 (1992) 369–417, (Pubmed: 1372998). [9] K. Tyml, O. Mathieu-Costello, E. Noble, Microvascular Research 49 (1995) 17–32, (Pubmed: 7746162). [10] O. Hudlicka, D. Pette, H. Staudte, Pflügers Archiv 343 (1973) 341–356, (Pubmed: 4796826). [11] H. Hoppeler, H. Howald, P. Cerretelli, Experientia 46 (1990) 1185–1187, (Pubmed: 2253720). [12] K. Tyml, O. Mathieu-Costello, L. Cheng, E.G. Noble, Journal of Applied Physiology 87 (1999) 1496–1505, (Pubmed: 10517784). [13] K. Tang, E.C. Breen, H.P. Gerber, N.M. Ferrara, P.D. Wagner, Physiological Genomics 18 (2004) 63–69, (Pubmed: 15084712). [14] A.F. Bruns, L. Bao, J.H. Walker, S. Ponnambalam, Biochemical Society Transactions 37 (Pt 6) (2009) 1193–1197, (Pubmed: 19909245). [15] P.D. Wagner, Biochemical Society Transactions 39 (2011) 1556–1559, (Pubmed: 22103486). [16] W.S. Jones, B.D. Duscha, J.L. Robbins, N.N. Duggan, J.G. Regensteiner, W.E. Kraus, W.R. Hiatt, A.O. Dokun, B.H. Annex, Vascular Medicine 17 (2012) 94–100, (Pubmed:22402934). [17] R. Gianni-Barrera, M. Trani, S. Reginato, A. Banfi, Biochemical Society Transactions 39 (2011) 1644–1648, (Pubmed: 22103501). [18] H. Singh, T.A. Tahir, D.O. Alawo, E. Issa, N.P. Brindle, Biochemical Society Transactions 39 (2011) 1592–1596, (Pubmed: 22103492). [19] B. Sitohy, J.A. Nagy, H.F. Dvorak, Cancer Research 72 (2012) 1909–1914, (Pubmed: 22508695). [20] S. Mittar, C. Ulyatt, G.J. Howell, A.F. Bruns, I. Zachary, J.H. Walker, S. Ponnambalam, Experimental Cell Research 315 (2009) 877–889, (Pubmed: 19162007). [21] J. Gutierrez, E. Brandan, Molecular and Cellular Biology 30 (2010) 1634–1649, (Pubmed: 20100867). [22] C.L. Smith, P. Mittaud, E.D. Prescott, C. Fuhrer, S.J. Burden, Journal of Neuroscience 21 (2001) 3151–3160, (Pubmed: 11312300). [23] K.A. Rivas-Plata, J.R. Kraas, S.M. Saleh, S.L. Swope, Journal of Neuroscience 21 (2001) 3806–3818, (Pubmed: 11356869). [24] B. Klitzman, D.N. Damon, R.J. Gorczynski, B.R. Duling, Circulation Research 51 (1982) 711–721, (Pubmed: 7139886).
[25] J.M. Dopp, N.R. Philippi, N.J. Marcus, E.B. Olson, C.E. Bird, J.J. Moran, S.W. Mueller, B.J. Morgan, Respiration 82 (2011) 458–467, (Pubmed: 21846958). [26] M.J. Jackson, Antioxidants & Redox Signaling 15 (2011) 2477–2486, (Pubmed: 21699411). [27] C. Zhang, X. Xu, B.J. Potter, W. Wang, L. Kuo, L. Michael, G.J. Bagby, W.M. Chilian, Arteriosclerosis, Thrombosis, and Vascular Biology 26 (2006) 475–480, (Pubmed: 16385082). [28] S. Basuroy, S. Bhattacharya, C.W. Leffler, H. Parfenova, American Journal of Physiology. Cell Physiology 296 (2009) C422–C432, (Pubmed: 19118162). [29] S. Brickson, J. Hollander, D.T. Corr, L.L. Ji, T.M. Best, Medicine & Science in Sports & Exercise 33 (2001) 2010–2015, (Pubmed: 11740292). [30] M.F. O'Leary, D.A. Hood, Autophagy 5 (2009) 230–231, (Pubmed: 19098460). [31] B. Chabi, P.J. Adhihetty, M.F. O'Leary, K.J. Menzies, D.A. Hood, Journal of Applied Physiology 107 (2009) 1730–1735, (Pubmed: 19797682). [32] P.M. Abruzzo, S. di Tullio, C. Marchionni, S. Belia, G. Fano, S. Zampieri, U. Carraro, H. Kern, G. Sgarbi, G. Lenaz, M. Marini, Free Radical Research 44 (2010) 563–576, (Pubmed: 20298122). [33] K.L. Wicks, D.A. Hood, American Journal of Physiology 260 (4 Pt 1) (1991) C841–C850, (Pubmed: 1850197). [34] P.M. Siu, S.E. Always, The Journal of Physiology 565 (Pt 1) (2005) 309–323, (Pubmed: 15774533). [35] J. Norrbom, C.J. Sundberg, H. Ameln, W.E. Kraus, E. Jansson, T. Gustafsson, Journal of Applied Physiology 96 (2004) 189–194, (Pubmed: 12972445). [36] A. Wagatsuma, N. Kotake, K. Mabuchi, S. Yamada, Journal of Physiology and Biochemistry 67 (2011) 359–370, (Pubmed: 21394548). [37] A. Wagatsuma, N. Kotake, S. Yamada, Molecular and Cellular Biochemistry 349 (2011) 139–147, (Pubmed: 21739150). [38] J.M. Marshall, H.C. Tandon, The Journal of Physiology 350 (1984) 447–459, (Pubmed: 6747856). [39] P.M. Hassoun, F.S. Yu, C.G. Cote, J.J. Zulueta, R. Sawhney, K.A. Skinner, H.B. Skinner, D.A. Parks, J.J. Lanzillo, American Journal of Respiratory and Critical Care Medicine 158 (1998) 299–305. [40] Y.W. Lee, W.H. Lee, P.H. Kim, Inflammation Research 59 (2010) 755–765, (Pubmed: 20349326). [41] S. Wind, K. Beuerlein, M.E. Armitage, A. Taye, A.H. Kumar, D. Janowitz, C. Neff, A.M. Shah, K. Wingler, H.H. Schmidt, Hypertension 56 (2010) 490–497, (Pubmed: 20606112). [42] H. Ishikita, B.T. Eger, K. Okamoto, T. Nishino, E.F. Pai, Journal of the American Chemical Society 134 (2012) 999–1009, (Pubmed: 22145797). [43] N.H. Herzberg, E. Middelkoop, M. Adorf, H.L. Dekker, M.J. Van Galen, M. Van den Berg, P.A. Bolhuis, C. Van den Bogert, European Journal of Cell Biology 61 (1993) 400–408, (Pubmed: 8223726). [44] N. Gleyzer, K. Vercauteren, R.C. Scarpulla, Molecular and Cellular Biology 25 (2005) 1354–1366, (Pubmed: 15684387). [45] R.C. Scarpulla, Physiological Reviews 88 (2008) 611–638, (Pubmed: 18391175). [46] C.M. LeMoine, S.C. Lougheed, C.D. Moyes, Journal of Molecular Evolution 70 (2010) 492–505, (Pubmed: 20443112). [47] M. Picard, T. Taivassalo, G. Gouspillou, R.T. Hepple, The Journal of Physiology 589 (Pt 18) (2011) 4413–4421, (Pubmed: 21708903). [48] O.T. Hernandez-Hernandez, T.K. Gonzalez-Garcia, I. Camacho-Arroyo, The Journal of Steroid Biochemistry and Molecular Biology 132 (2012) 127–134, (Pubmed: 22542550). [49] J.V. Frangioni, P.H. Beahm, V. Shifrin, C.A. Jost, B.G. Neel, Cell 68 (1992) 545–560, (Pubmed: 1739967). [50] F.G. Haj, B. Markova, L.D. Klaman, F.D. Bohmer, B.G. Neel, Journal of Biological Chemistry 278 (2003) 739–744, (Pubmed: 12424235). [51] Y. Romsicki, M. Reece, J.Y. Gauthier, E. Asante-Appiah, B.P. Kennedy, Journal of Biological Chemistry 279 (2004) 12868–12875, (Pubmed: 14722096). [52] M. Marino, F. Scuderi, C. Provenzano, E. Bartoccioni, Gene Therapy 18 (2011) 109–116, (Pubmed: 20927136). [53] R. Ciurea, C. Margaritescu, C. Simionescu, A. Stepan, M. Ciurea, Romanian Journal of Morphology and Embryology 52 (2011) 1227–1232, (Pubmed: 22203927). [54] M.P. Turunen, S. Yla-Herttuala, Cardiovascular Research 90 (2011) 441–446, (Pubmed: 21498421). [55] C.K. Magill, A. Tong, D. Kawamura, A. Hayashi, D.A. Hunter, A. Parsadanian, S.E. Mackinnon, T.M. Myckatyn, Experimental Neurology 207 (2007) 64–74, (Pubmed: 17628540). [56] Q. Wang, B. Zhang, Y.E. Wang, W.C. Xiong, L. Mei, Neurosignals 16 (2008) 246–253, (Pubmed: 18253062).
Contents lists available at ScienceDirect
Cellular Signalling journal homepage: www.elsevier.com/locate/cellsig
Specialization of mitochondrial and vascular oxidant modulated VEGFR in the denervated skeletal muscle Hui Zhao a, Han-Wei Huang b,⁎, Jun-Guo Wu b, Pei-Yan Huang b a b
Department of Integrative Medicine and Neurobiology, National Key Lab of Medical Neurobiology, Institute of Brain Sciences, Fudan University, Shanghai, China Department of Orthopaedics, 5th Hospital of Shanghai, Fudan University, Shanghai, China
a r t i c l e
i n f o
Article history: Received 2 April 2013 Received in revised form 17 June 2013 Accepted 25 June 2013 Available online 2 July 2013 Keywords: Muscle denervation Mitochondria XO pathway VEGFR-1 VEGFR-2
a b s t r a c t Denervation of skeletal muscles results in timely muscular inflammation and muscle-T cell interaction, the cellular events might orchestrate a local circuit involved with IL-1β and IL-15. In the present study, by a combination assay of nerve–muscle preparation, western blot, immuno-precipitation, and radioactive of enzyme activity, we confirmed that mitochondrial and vascular oxidants were considerably up-regulated following gastrocnemius denervation, which was due to gradual decay in mitochondrial biogenesis and XO pathway and accompanied by strengthened IL-1β-VEGFR-2 and IL-15-VEGFR-1 signaling. Intriguingly, these alterations could be triggered by the early established muscular inflammation. In contrast, with prolonged muscle denervation, settings of organelle interconnection were ultimately conveyed by ER bound PTP1B, which promoted VEGFR-1 signaling and contributed to VEGFR-2 activation, and the process could be modulated by mitochondrial and vascular oxidant. Importantly, VEGFR-2 could rescue the disruption of MuSK activity and AchR cluster exerted by IL-1β and IL-15, with PGC-1α and XO involvement. Altogether, extensive network centered on VEGFR-2 signaling was essentially contributed to early recovery processes regarding muscle denervation. Increasing knowledge of this mechanism might open up a conduit for functional response to muscle atrophy, and enable the development of better agents to combat the related disorders. © 2013 Published by Elsevier Inc.
1. Introduction Previously, we reported that muscle denervation could produce a stereotypic inflammation and muscle-T cell interaction [1,2], and the produced IL-1β and IL-15 were proposed to contribute to the complex and dynamic micro-environment [3–9]. As reported, in skeletal muscle, blood flow was closely related with muscle activity, and there was an increase in capillary density per area of muscle (capillary-to-fiber ratio, C:F) following exercise or exposure to hypoxia. Also, the anatomical pattern of muscle microvascularization changed dramatically after nerve transection, with defected capillaries served by muscle fibers [10–12]. VEGF, a 35- to 45-kDa peptide growth factor, has proved to be essential for the development of capillary networks during embryogenesis and the maintenance of capillary structures [13–16]. In skeletal muscle, deletion of VEGF gene resulted in greatly reduction in muscle capillarity [17,18]. Then, it was proposed that the VEGF signaling might be integral components for remodeling of vascular bed or impairment of microcirculation in the pathogenesis of muscle denervation. As well established, between two related but distinct receptors: VEGFR-1 and VEGFR-2, VEGF signaling through VEGFR-2 was the major pathway that activated angiogenesis by inducing the proliferation, ⁎ Corresponding author at: Department of Orthopaedics, 5th Hospital of Shanghai, Fudan University, 128# Ruili Rd. Shanghai, China. Tel.: +86 21 64308151. E-mail address: [email protected] (H.-W. Huang). 0898-6568/$ – see front matter © 2013 Published by Elsevier Inc. http://dx.doi.org/10.1016/j.cellsig.2013.06.014
survival, sprouting and migration of ECs [15], and therapeutic targeting of this pathway was already used to delay tumor related vasculature [19]. In contrast, VEGFR-1 displayed stable ER-Golgi localization, upon stimuli, and VEGFR-1 could be preferential sequestrated and seemed to act as a negative regulator of VEGF-mediated angiogenesis [20,21]. Recently, it has been suggested that Src kinases represented the predominant protein tyrosine kinase activity in VEGFR activation and MuSK directed AChR clustering [22,23], and the overlap modulation was proposed to form a viable link between functional vasculature and neuromuscular junction stabilization during muscle denervation. It was already identified that XO resided in the cytosol and was the primary source of ROS in the vasculature. Particularly, XO was predictive of superior potency for vascular inflammation and oxidative capacity in skeletal muscle [24–26], which could be up-regulated by proinflammatory stimuli (lipopolysaccharide, TNF-α, IL-1, and IL-6) [27,28]. Besides that, many studies addressed that muscle atrophy was associated with mitochondrial dysfunction and resulted in increased rate of mitochondrial reactive oxygen species production [29–32]. When rats were subjected to unilateral sciatic nerve transection, decrements in mitochondrial content, enzymatic activity of respiratory chain complexes, as well as the rate of mitochondrial ATP production were significantly decreased, which led to reduced muscle mass [33,34]. As such, the present study designed to examine whether XO activity and mitochondrial biogenesis could be triggered by muscle denervation, which was presumed to link to changes in VEGF-VEGFR signaling,
H. Zhao et al. / Cellular Signalling 25 (2013) 2106–2114
understanding the mechanism underlying these alterations would enable the development of better agents to combat muscular disorders. 2. Materials and methods 2.1. Tissue harvest Adult SD rats (160–180 g, all male) were divided into five groups (n = 5 in each group): unoperated controls, animals after 4 h, 1, 3 or 7 days following sciatic nerve transection. For muscle denervation, the right sciatic nerve was surgically exposed and transected after anesthesia with pentobarbital sodium (35 mg/kg, intraperitoneally). At the specified time-points the animals were euthanized, and a 1 × 1 cm block of gastrocnemius was removed and cleaned of fat and connective tissues, weighed, and stored for later analysis. Body temperature was maintained at 37°C throughout surgery, and there was no evidence of postoperative infection. This study was approved by Ethics Committee of Fudan University. All animal procedures were performed in strict accordance with the U.S. National Institutes of Health's Guide for the Care and Use of Laboratory Animals. 2.2. Myeloperoxidase MPO activity was measured spectrophotometrically. Briefly, samples were incubated with a substrate o-dianisidine hydrochloride. This reaction was carried out in a 96-well plate by adding 290 μl, 50 mM phosphate buffer, 3 μl substrate solution (containing 20 mg/ml o-dianisidine hydrochloride), and 3 μl H2O2 (20 mM). Sample (10 μl) was added to each well to start the reaction. Standard MPO (Sigma Chemical Co., St. Louis, MO) was used in parallel to determine MPO activity in the sample. The reaction was stopped by adding 3 μl sodium azide (30%). Plates were read at 460 nm. MPO activity was defined as nmol/mg protein. 2.3. Oxidant production Gastrocnemius was homogenized in 0.1 M phosphate buffer (pH 7.4) with a motor-driven Polytron glass homogenizer at 4 °C. The resulting homogenate was filtered through four layers of medical gauze to remove connective tissue debris. Muscle oxidant generation was determined within 30 min of tissue harvest in fresh tissue homogenates using dichlorofluorescein (DCF) as a probe. The basal buffer consisted of 0.1 M potassium phosphate at pH 7.4. The induced buffer additionally included 1.7 mM ADP, 0.1 mM NADPH, and 0.1 mM FeCl3. Both buffers included 5 μM 2′,7′-dichlorofluoresceindiacetate (DCFH-DA), which was made fresh in 1.25 mM methanol and kept in a dark room at 0 °C. For basal condition, following reagents were added: 2938 μl 0.1 M phosphate buffer, 50 μl filtered muscle homogenate, and 12 μl 1.25 mM DCFH-DA. Induced condition was replaced 90 μl of buffer with 50 μl FeCl3, 20 μl ADP, and 20 μl NADPH. Total reaction volume was 3.0 ml. A blank consisting of the appropriate buffer and 5.0 μM DCFH-DA without sample was used to correct for the auto-oxidation rate of DCFH-DA. The mixture was incubated in the dark for 15 min at 37 °C. DCF formation was determined with a Hitachi F-2000 fluorescence spectrophotometer (Hitachi Instruments, Co., San Jose, CA) with a thermostated cell compartment at excitation wavelength 488 nm and emission wavelength 525 nm. The units were expressed as pmol DCF/mg protein. 2.4. XO/XDH activity assay Measurement of XO and XDH activity in kidney tissue was based on the pterin-based assay. In brief, approximately 40 mg of tissues was homogenized in 1 ml assay buffer (50 mM K-phosphate, 1 mM ethylene-diaminetetraacetic acid, 0.5% dimethyl sulfoxide, and protease
2107
inhibitor cocktail, pH 7.4). The supernatant (150 μl) was co-incubated with 50 μl pterin solution (final concentration of 50 μM) or pterin with methylene blue solution (final concentration of 50 μM) to assay XO or both XO and XDH activities, respectively. Before and after 120-min incubation at 37°C, fluorometric assays were performed to calculate the production of isoxanthopterin. Protein concentration was measured by Pierce BCA Protein Assay Kit (Thermo Scientific Inc., Billerica, MA, USA). 2.5. Nerve–muscle preparations Gastrocnemius muscles, with their respective motor nerve, were dissected from SD rats under deep anesthesia. Nerve–muscle preparations were pinned in an experimental Sylgard-coated recording chamber filled with normal Ringer's solution containing (in mM): 124 NaCl, 5 KCl, 1.25 NaH2PO4, 2 MgCl2, 1NaHCO3, 2 CaCl2, 25 HEPES and 10 glucose, oxygenated with 100% O2. Treatments included in IL-1β (R&D systems, Minneapolis, MN, 20 ng/ml, 24 h), IL-1ra (10 ng/ml, 24 h), recombinant IL-15 (10μg/ml, 24 h, BioLegend, San Diego, CA), XO inhibitor allopurinol (1.0 mM, 24 h), PGC-1α antibody (20μg/ml, 24 h), or adenovirus infection (VEGFR-1/VEGFR-2, 5 × 109 plaqueforming units (PFU) dissolved in sterilized PBS, 72 h). A 2–3 mm broad band of muscle around innervated muscle was excised, representing the junctional region, which was used for the indicated assays. 2.6. Immunohistochemistry Serial sections (5 μm thick) of gastrocnemius were sectioned on a cryostat at 15 μm. Serial cross-sections were mounted on glass slides (Superfrost Plus) and labeled with antibody against VEGF (1:1000, Abcam, Cambridge, MA). Antibodies were diluted in 0.01 M PB with 3% normal goat serum, 1% BSA, 0.5% Triton X-100, and 0.05% sodium azide, pH 7.4. The primary antibodies were revealed using ABC kit. Sections were viewed under microscope (Leica L2000A). Images were acquired and processed by Leica QWin software. In separate experiments, we also tested different antigen retrieval techniques including microwave heating and SDS-pretreatment. 2.7. Discontinuous gradient centrifugation for lipid raft extraction Gastrocnemius was washed twice with phosphate-buffered saline (PBS, pH 7.4) at 4°C and then homogenized in 2 ml of 500 mM NaCO3 (pH 11.0) solution with complete protease inhibitor mixture (Roche Applied Science, Indianapolis, Indiana). The tissue suspension was further sonicated with one 30-s burst at setting 4 and one 30-s burst at setting 7 (Model W-220F; Ultrasonics, Inc., Bartlett, Illinois). The homogenate was adjusted to 45% sucrose by adding 2 ml of 90% (w/v) sucrose prepared in MBS buffer (25 mM MES, pH 6.5, and 0.15 M NaCl) and placed at the bottom of an ultracentrifuge tube. Four milliliters of 35% sucrose and 4 ml of 5% sucrose (both in MBS buffer containing 250 mM sodium carbonate) were then overlaid upon the sample to form a 5–45% discontinuous sucrose gradient. The sample was centrifuged at 32,000 rpm for 16 h in a Beckman ultracentrifuge in an SW-41Ti rotor. A light-scattering band at the 5% and 35% sucrose interface (Fraction 4–5) represented the lipid raft fractions, which was collected and used for the indicated assay. 2.8. Immunoprecipitation and western blotting Gastrocnemius was homogenized in lysis buffer containing 10 mM Tris–HCl (pH 7.5), 50 mM NaCl, 0.1% Triton X-100, and one complete protease inhibitor cocktail tablet (Roche Applied Science, Mannheim, Germany). After centrifugation at 14,000 rpm for 5 min, the supernatants were incubated with anti-pTyr, PTP1B (1:200; BD Transduction Laboratories, Lexington, Kentucky), anti-AchRα antibody
2108
H. Zhao et al. / Cellular Signalling 25 (2013) 2106–2114
(1:200; Santa Cruz Biotechnologies, Santa Cruz, CA at 4°C overnight with slow rotation. Sixty microliters of protein G-agarose beads (Invitrogen, Carlsbad, California) was added, and the mixture was further incubated at 4°C for 3 h with slow rotation. The protein G-agarose beads were then pelleted by centrifugation at 12,000 g for 15 min at 4°C and washed five times with wash buffer (50 mM Tris–HCl (pH 8.0), 150 mM NaCl). For western blot analysis, protein samples were resolved by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to polyvinylidene difluoride membranes (GE Healthcare, Little Chalfont, Buckinghamshire, UK). Membranes were probed with anti-caveolin-3 (1:1000, BD Transduction Laboratories), anti-VEGFR-1, anti-VEGFR-2 antibodies (1:500, Abcam), anti-AchRβ antibody (1:500; Santa Cruz Biotechnologies). Protein bands were detected using alkaline phosphatase conjugated secondary antibodies (1:5000) and ECF substrate (GE Healthcare, Piscataway, NJ), and then scanned using a Storm 860 imaging system (GE Healthcare). Band intensities were quantified and analyzed with ImageQuant software (GE Healthcare). PGC-1α (1:1000, Abcam, Cambridge, MA). 2.9. TaqMan Reverse Transcription (RT)-PCR for miRNA quantification Total RNA was isolated from gastrocnemius (50 mg) with TRIzol™ (Invitrogen, Carlsbad, VA) according to manufacturer's protocol. miR23a quantification was carried out by reverse transcribing total RNA using Taqman™ microRNA reverse transcription kit and subjected to real-time PCR using TaqMan™ MicroRNA Assay kit (Applied Biosystems, Carlsbad, CA). Reactions were performed using Stratagene Mx3000 instrument in triplicate. Real-time PCR data was analyzed using a ΔΔCt calculation. A p value of less than 0.05, when considering treated animals or cells vs. control group, was considered significant. 2.10. In vitro Src kinase assay Proteins in gastrocnemius lipid rafts or anti-MuSK antibody immunoprecipitation (1:200; Abcam) were precipitated with 5% (w/v) TCA. The resulting pellets were washed with acetone and incubated at 30°C with 5 μg of SRC substrate peptide (KVEKIGEGTYGVVYK, corresponding to amino acids 6–20 of p34cdc2; Upstate Biotechnology, Lake Placid, New York) in kinase buffer containing 5 μCi of [γ-32P]-adenosine triphosphate ([γ-32P]-ATP; PerkinElmer Life Sciences, Waltham, Massachusetts), 50 mM Tris–HCl (pH 7.5), 10 mM MgCl2, 10 mM MnCl2, 25μM ATPase, 1 mM dithiothreitol, and 100μM Na3VO4. After 30 min, the reaction was terminated by the addition of 10 μl of 40% (w/v) TCA, and samples were spotted on P81 cellulose phosphate paper (Upstate Biotech). The paper was washed three times with 1% (w/v) phosphoric acid and once with acetone. Radioactivity retained on the P81 paper was quantified by liquid scintillation counting. Blank counts (without tissue lysate) were subtracted from each result, and radioactivity (cpm) was converted to picomoles per minute (pmol/min). 2.11. PTP 1B assay Gastrocnemius was collected and homogenized in RIPA buffer (50 mM Tris [pH7.5], 150 mM NaCl, 2 mM EDTA, 1.0% Triton X-100 and complete protease inhibitor mixture). Equal amounts of protein were incubated at 4 °C for 2 h with anti-PTP1B antibody and then, 20 μl of protein G sepharose was added and incubated for additional 2 h with mixing. Immunoprecipitated complexes were washed twice in RIPA buffer, once with the assay buffer (25 mM imidazole [pH7.2], 0.1 mg/ml BSA, 10 mM DTT) and finally they were resuspended in 25 μl of assay buffer. The phosphatase substrate raytide was labeled at its unique tyrosine residue in the presence of [γ-32P] ATP. Assay mixtures (30 μl) containing the immunoprecipitated pellet and [γ-32P]-labeled raytide (1 × 105 cpm) were incubated at 30 °C for
2 h and the reaction was terminated by adding 750 μl of a charcoal mixture (0.9 M HCl, 90 mM sodium pyrophosphate, 2 mM NaH2PO4, 4% vol/vol Norit A). After centrifugation, the radioactivity in 400 μl of the supernatant was measured by scintillation counting. Blanks were determined by measuring the free γ-32P in reactions where the immunoprecipitates were either boiled or omitted, and these values were subtracted from the reaction values. 2.12. Recombinant adenoviruses Recombinant adenovirus expressing rat VEGFR-1 or VEGFR-2 was constructed by inserting into the adenoviral shuttle vector pDE1sp1A (Microbix Biosystems, Inc. Canada), and the insert was then switched to the adenoviral vector through LR recombination. Adenovirus was purified by CsCl2 gradients and PD-10 Sephadex chromatography. 2.13. Enzyme-linked immunoassay analysis of VEGFR levels Levels of VEGFR-1/2 production were measured by sandwich enzyme-linked immunoassay (ELISA) according to manufacturer's instructions (R&D Systems, Inc., Minneapolis, MN). A 96-well plate was coated with 2 μg/ml monoclonal anti-VEGFGR-1/2 antibodies at 4°C overnight and then blocked with 1% bovine serum albumin (BSA) in PBS for 1 h. The plates were washed three times with PBS containing 0.2% Tween 20 (PBST). Aliquots of tissue lysates were diluted to100μl with Hanks' balanced salt solution (HBSS) with calcium and magnesium, 10 mM 4-(2-hydroxyethyl)-1piperazineethanesulfonic acid (HEPES), and 1% fetal bovine serum, added to the plates, and then incubated for 2 h at room temperature. The plates were washed three times with PBS, and 100 μl aliquots of 0.1 μg/ml biotinylated VEGFR-1/2 affinity purified polyclonal antibody were added and incubated for 2 h. After further three washes with PBST, the immune complexes were colorimetrically detected using horseradish peroxidase (HRP)-streptavidin conjugate. The reaction was halted by the addition of 1 M H2SO4, and the absorbance at 450 nm was measured using a microplate reader (VersaMax; Molecular Devices, Sunnyvale, CA). 2.14. Statistical analyses All experiments were performed using 5 animals per group (n = 5). Data were represented as mean ± SEM and analyzed with Prism 5 software. For all data sets, normality and homocedasticity assumptions were reached, validating the application of the one-way ANOVA, followed by t test for multiple comparisons. Differences were considered significant for p b 0.05. 3. Results 3.1. Mitochondrial and vascular oxidant initiated during gastrocnemius denervation Rats were undergone sciatic nerve transection, 4 h, 1, 3 and 7 days later, and MPO activity, a granular-based heme enzyme present in neutrophils and macrophages, was examined in the denervated gastrocnemius. As shown in Fig. 1A, MPO activity was robustly elevated by factor of 3.3 and 3.4 respectively after 4 h and 1 day of operation. Afterwards, they were gradually decreased to control level. Furthermore, dynamic alteration in oxidant production in the denervated muscle was measured using DCF probe, in which the basal buffer condition (0.1 M potassium phosphate) measured predominantly nonmitochondrial sources of oxidant generation and maximum one could be stimulated by the induced buffer (including 1.7 mM ADP, 0.1 mM NADPH, 0.1 mM FeCl3). As illustrated, oxidant production both in the basal (Fig. 1B) and induced buffers (Fig. 1C) were rapidly rose around 2 folds over control at day 1,and the increase was
H. Zhao et al. / Cellular Signalling 25 (2013) 2106–2114
2109
Fig. 1. Mitochondrial and vascular oxidant initiated during gastrocnemius denervation. Rats were killed 4 h, 1, 3 and 7 days after sciatic nerve transection (n = 5 for each group). MPO activity in gastrocnemius homogenates was measured. data was converted to nmol/mg protein (A). Rats were killed 1, 3 and 7 days after sciatic nerve transection (n = 5 for each group), oxidant production in induced or basal buffer (B and C). XO and XOH activity were quantified (D). Data was converted to pmol or μU/mg protein. Nerve–muscle preparation was cultured for 1 day, then exposed to IL-1β or IL-15 for 24 h, DCF production in induced buffer (E and G) and XO/XDH activity were quantified (F and H). Data was converted to pmolor μU/mg protein. Values represent mean ± SEM for 5 independent experiments. *p b 0.05 vs Con. #p b 0.05 vs IL-1β or IL-15. Con: control; 4 h, 1, 3, 7d: 4 h, 1, 3, 7 days after surgery.
sustained up to 7 days following muscle denervation. Also, similar alterations occurred on XO and Xanthine dehydogenase (XDH) activity. Fig. 1D illustrated that they were up-regulated around 2 folds over control following muscle denervation. Given our previous description that gastrocnemius denervation could produce a stereotypic alteration of IL-1β involved inflammation and muscle-T cell interaction initiated IL-15 production, we consequently demonstrated in nerve–gastrocnemius preparation that IL-1β exposure resulted in an increase in maximum oxidant production (Fig. 1E), in parallel, XO and xanthine dehydogenase (XDH) activity rose to 3.3 and 3.1 folds over control (Fig. 1F). Likewise, it was demonstrated that IL-15 incubation could also specifically led to dramatic elevation of maximum oxidant production, and XO and XDH activities (Fig. 1G, H).
3.2. VEGF-VEGFR signaling orchestrated by gastrocnemius denervation Vascular endothelial growth factor (VEGF) was noticed recently to be crucial factor during muscle denervation [17,18]. Consequently, it was shown in Fig. 2A and B, by immunohistochemistry, that VEGF immuno-positive signals were quite weak in control group, however, they were robustly increased and widely distributed in gastrocnemius at day 1 following muscle denervation. Quantification analysis revealed that the expression levels rose to 4.3 folds over control. By subcellular fractionation and western blot analysis, it was demonstrated that VEGFR-2 expression within the lipid rafts was enhanced rapidly in the denervated muscle. Levels of expression rose to 4.3 folds over control at day 1. Afterwards, they were gradually decreased. Comparably, total VEGFR-1 expression was dramatically increased at day 7, and expression level rose to 3.5 folds over control (Fig. 2C). Moreover, in nerve–gastrocnemius, we found that IL-1β exposure resulted in the greatly increase in VEGFR-2 expression within the lipid rafts, and the effect could be partly inhibited by XO inhibition. In contrast, there were no detectable changes in total VEGFR-1 expression (Fig. 2D). For IL-15 incubation, VEGFR-2 expression within the lipid rafts did not display any significant alteration, while total VEGFR-1 expression dramatically rose to 3.6 folds over control. Similarly, they partly exhibited attenuation after XO inhibition (Fig. 2E).
3.3. The role of mitochondrial and vascular oxidant on VEGFR signaling during gastrocnemius denervation A number of physiological studies have established a role for peroxisome proliferators-activated receptor coactivator-1 α (PGC-1α) in mediating mitochondrial biogenesis, which was a target of miR23a and generally accompanied by expression of genes related with muscle regeneration [35–37]. By real-time PCR, and western blot analysis, Fig. 3A illustrated that miR23a expression was progressively up-regulated by factor of 1.2, 1.4 and 1.8 respectively at day 1, 3 and 7 following gastrocnemius; parallel to these changes, PGC-1α expression were dramatically reduced to 68.3, 61.1 and 50.1% of control respectively. Furthermore, in nerve–gastrocnemius preparation, it was demonstrated that VEGFR-2 was up-regulated by IL-1β, which could be partly attenuated by anti-miR23a or PGC-1α antibody, but completely reduced by concurrent treatment of PGC-1α antibody and XO inhibitor (Fig. 3B). Likewise, VEGFR-1 could be increased by IL-15, which was disrupted by treatment with PGC-1α antibody and XO inhibitor concurrently but not individually. Intriguingly, IL-1β could not result in dramatic change in VEGFR-1, and IL-15 on VEGFR-2 expression respectively (Fig. 3C). 3.4. Timely modulation of c-Src activity during gastrocnemius denervation It was proposed that the integrated VEGFR signaling might be centered on c-Src activation which was linked to spatial and temporal segregation of protein phosphatase. We then detected Src activity within the lipid rafts by [γ-32p] incorporation. As shown in Fig. 4A, Src kinase activity was rapidly increased around 4.0 folds over control at day 1 following gastrocnemius denervation, which was gradually returned to control level. PTP1B activity was also investigated by [γ-32p] incorporation. As demonstrated, PTP1B activity was remained at control level at day 1 and 3 following surgery, however, by day 7, which it was considerably strengthened to 3.7 folds over control (Fig. 4B). In nerve–gastrocnemius preparation, IL-1β exposure resulted in a great increase in c-Src but not PTP1B activity, and the effect was completely blocked by concurrently treatment with PGC-1α antibody
2110
H. Zhao et al. / Cellular Signalling 25 (2013) 2106–2114
Fig. 2. VEGF-VEGFR signaling orchestrated by gastrocnemius denervation. Rats were killed 1, 3 and 7 days after sciatic nerve transection (n = 5 for each group). Consecutive cryosections of gastrocnemius were immuno-stained with antibody against VEGF, scale bar = 200 μm (A). Panel B depicts quantitative analysis of VEGF-staining intensities. Gastrocnemius was collected and homogenated, VEGFR-2 within the lipid rafts and total VEGFR-1 expression were determined by western blot analysis (C). Nerve–muscle preparation was cultured for 1 day, then exposed to IL-1β or IL-15 for 24 h, VEGFR-2 within the lipid rafts and total VEGFR-1 expression in the presence/absence of XO inhibitor were determined by western blot analysis (D and E). Data were normalized and calculated as percentage of control. Each value represents mean ± SEM for 5 independent experiments. *p b 0.05 vs Con. #p b 0.05 vs IL-1β or IL-15.
and XO inhibitor (Fig. 4C). Instead, IL-15 led to robust up-regulation of PTP1B but not c-Src activity, and the effect was attenuated by PGC-1α antibody and XO inhibitor (Fig. 4D). 3.5. Association of c-Src activity with VEGFR signaling during gastrocnemius denervation VEGFR signaling has been conclusively identified to be modulated by tyrosine phosphorylation. Consequently, by immunoprecipitation in which anti-pTyr as immunoprecipitated antibody and antiVEGFR-2 as immunoblot antibody, it was shown that tyrosine phophorylated VEGFR-2 rose to around 3.6 folds over control at day 1 following gastrocnemius denervation, which was gradually decreased afterwards. Comparably, there were no dramatic changes in VEGFR-1 signaling (Fig. 5A). In nerve–gastrocnemius preparation, by immunoprecipitation, it was shown that tyrosine phophorylated VEGFR-2 was robustly up-regulated around 3.3 folds over control after IL-1β but not IL-15 exposure, and the effect was blocked by both PGC-1α antibody and XO inhibitor (Fig. 5B and D). In contrast, the interaction of PTP1B and VEGFR-1 was increased by IL-15 but not IL-1β incubation. Coincidently, the effect was dependent on PGC-1α and XO (Fig. 5C and E). 3.6. Modulation of neuromuscular junction by VEGFR signaling Neuromuscular junction might be initiated by local environment, which was proposed to be related with mitochondrial and vascular oxidative responses. As demonstrated in Fig. 6A, junctional PGC-1α expression was robustly increased around 4.0 folds over control at day 1 following gastrocnemius denervation. After that, they were
gradually decreased. In the same time, junctional XO and XDH activities were rapidly up-regulated around 3.5 and 3.7 folds over control (Fig. 6B). In nerve–gastrocnemius preparation, by ELISA assay, we demonstrated that junctional VEGFR-2 could be up-regulated by IL-1β while VEGFR-1 by IL-15, and the effects were partly or completely reversed by PGC-1α antibody and XO inhibitor individually or concurrently (Fig. 6C and D). As well established, NMJ was a peripheral cholinergic synapse that conveys signals from the motor neurons to the muscle cells, which required muscle-specific kinase (MuSK) activation and AchR cluster. In the present observation, MuSK activity and AchR cluster were slightly reduced by IL-1β exposure. In contrast, IL-15 led to their dramatically attenuation: MuSK activity decreased to 56.3 ± 5.7% control, and AchR cluster to 58.8 ± 7.9% control. These effects could be rescued by adenovirus over-expressing VEGFR-2 but not VEGFR-1 (Fig. 6E and F). 4. Discussion In previous study, we characterized that rapid muscular inflammatory and muscle-T cell interaction could be subsequently initiated by gastrocnemius denervation, and the produced IL-1β and IL-15 were contributed primarily to the dynamic microenvironment [1,2]. Herein, we further demonstrated that there were sustainable upregulation of oxidant generation including basal and mitochondrial oxidant production and XO/XDH activity following muscle denervation. Coincidently, in nerve–muscle preparation, following IL-1β and IL-15 treatment, both mitochondrial oxidant production and XO/XDH activity were considerably increased. Therefore, these enzymatic changes reasonably reflected distinct variables in oxidative
H. Zhao et al. / Cellular Signalling 25 (2013) 2106–2114
2111
Fig. 3. The role of mitochondrial and vascular oxidant on VEGFR signaling during gastrocnemius denervation. Rats were killed 1, 3 and 7 days after sciatic nerve transection (n = 5 for each group). Gastrocnemius was collected and homogenized. miR23a and PGC-1α were detected using real-time PCR and western blot analysis respectively (A). Nerve–muscle preparation was cultured for 1 day, then exposed to IL-1β or IL-15 for 24 h, VEGFR-2 within the lipid rafts and total VEGFR-1 expression in the presence/absence of XO inhibitor, or anti-miR23a/PGC-1α antibody, was determined by western blot analysis (B and C). Data were normalized and calculated as percentage of control. Each value represents mean ± SEM for 5 independent experiments. *p b 0.05 vs Con. #p b 0.05 vs IL-1β or IL-15.
state during muscle denervation [24,26,38]. We then proposed a teleological argument that reactive oxygen species were involved in both rapid muscular inflammation and later muscle-T cell interaction.
It was already reported that muscle atrophy was associated with mitochondrial dysfunction and resulted in an increased rate of mitochondrial reactive oxygen species production [29–32], decrements in
2112
H. Zhao et al. / Cellular Signalling 25 (2013) 2106–2114
Fig. 4. Timely modulation of c-Src activity during gastrocnemius denervation. Rats were killed 1, 3 and 7 days after sciatic nerve transection (n = 5 for each group). Gastrocnemius was homogenized and separated by discontinuous sucrose centrifugation. SRC kinase activity within the lipid rafts was assayed using an SRC-specific substrate peptide (A). In the same time, PTP1B activity in gastrocnemius was determined by PTP1B-specific substrate peptide (B). Nerve–muscle preparation was cultured for 1 day, then exposed to IL-1β or IL-15 for 24 h. Src activity within the lipid rafts and PTP1B activity in the presence/absence of individual or concurrent of XO inhibitor and PGC-1α antibody were determined by [32p] incorporation (C, D). Data were normalized and calculated as percentage of control, each value represents mean ± SEM for 5 independent experiments. *p b 0.05 vs Con. # p b 0.05 vs IL-1β or IL-15.
mitochondrial content, enzymatic activity of respiratory chain complexes. Particularly, the rate of mitochondrial ATP production was significantly decreased following sciatic nerve transection, which was in parallel with the reduction of muscle mass by 40–65% [33,34]. Besides
that, XO pathways could also be triggered by muscular disorder, which was associated with vascular oxidative stress [39–41]. Two inter-convertible forms known as XO and XDH could oxidize NADH and resulted in ROS formation [42]. Therefore, our observation, in
Fig. 5. Association of c-Src activity with VEGFR signaling during gastrocnemius denervation. Rats were killed 1, 3 and 7 days after sciatic nerve transection (n = 5 for each group). Gastrocnemius was collected and homogenized. Immuno-precipitation was used to detect VEGFR phosphorylation. The immuno-precipitated antibody was anti-c-pTyr and the immuno-blotting antibody was VEGFR-1/2 (A). Nerve–muscle preparation was cultured for 1 day, then exposed to IL-1β or IL-15 for 24 h in the presence/absence of individual or concurrent of XO inhibitor and PGC-1α antibody. Immuno-precipitation was used to detect VEGFR phosphorylation. The immuno-precipitated antibody was anti-c-pTyr and the immuno-blotting antibody was VEGFR-1/2 (B, D). Immuno-precipitation was used to detect the interaction of PTP1B and VEGFR-1. The immuno-precipitated antibody was anti-PTP1B and the immuno-blotting antibody was VEGFR-1 (C and E). Data were normalized and calculated as percentage of control. Each value represents mean ± SEM for 5 independent experiments. *p b 0.05 vs Con. #p b 0.05 vs IL-1β or IL-15.
H. Zhao et al. / Cellular Signalling 25 (2013) 2106–2114
2113
Fig. 6. Modulation of neuromuscular junction by VEGFR signaling. Rats were killed 1, 3 and 7 days after sciatic nerve transection (n = 5 for each group). Gastrocnemius NMJ was separated and homogenized. PGC-1α expression and XO/XDH activity were determined by western blot analysis and fluorometric assays respectively (A, B). Nerve–muscle preparation was cultured for 1 day, then exposed to IL-1β or IL-15 for 24 h. In the presence/absence of individual or concurrent of XO inhibitor and PGC-1α antibody, junctional VEGFR-1/2 was determined by ELISA assay (C, D). Nerve–muscle preparation was infected with adeno-VEGFR-1 or VEGFR-2 for 48 h, then exposed to IL-1β or IL-15 for another 24 h. Junctional MuSK was determined by in vitro kinase assay, AchR cluster by immuno-precipitation in which AchRα was used as immuno-precipitated antibody, AchRβ as immuno-blotting antibody (E, F). Data were normalized and calculated as percentage of control. Each value represents mean ± SEM for 5 independent experiments. *p b 0.05 vs Con. #p b 0.05 vs IL-1β or IL-15.
conjunction with above findings, implied that mitochondrial and vascular functions were actually targeted and appeared to be associated with processes initiated by muscle denervation. We further demonstrated that VEGF expression was up-regulated at day 1 following muscle denervation, and immuno-positive densities were mainly localized to the capillary area. Also, there was a timely up-regulation of VEGFR-2 (1 day) and VEGFR-1 (7 day). Intriguingly, mitochondrial biogenetic gene miR23a was increased and its target PGC-1α was decreased. All above alterations could essentially be affected by IL-1β and IL-15 respectively. As such, the upper estimate was that muscle denervation triggered a general mitochondrial transcription [43–47], which together with vascular oxidant, contributed to the dynamic and complex alterations in VEGFR signaling. Notably, the event linked to VEGFR-2 tyrosine phosphorylation was the activation of vasodilation, permeability and angiogenic processes [48]. In contrast, VEGFR-1 displayed stable ER-Golgi localization that could be related with PTP1B [49–51]. In the present study, we demonstrated that c-Src and PTP1B activities were subsequently activated during muscle denervation, and the association of VEGFR-2 and c-Src, VEGFR-1 and PTP1B underwent vigorous up-regulation in consequence. Intriguingly, these alterations could
be inferred to IL-1β and IL-15 respectively and involved XO pathway and mitochondrial transcription. Then, the data indicated a dual VEGFR system, in which up-regulation of VEGFR-2 might result in temporary vasculogenesis or increase in vascular permeability in the early stage of muscle denervation; however, along with T cells invasion into the muscle, VEGFR-1 could be targeted and stimulated, which seemed to related to PTP1B involved organelle interconnection. A substantial body of data already shed light on the strong association of VEGFR signaling with muscle function, for example, VEGFR-2 appeared to be the main signaling route [52–54], whose signaling might overlap with alterations in motor endplate and terminal axons, thus precluding characterization of neuromuscular junction (NMJ) stability during muscle denervation [55,56]. Coincidently, we demonstrated herein that junctional PGC-1α expression and XO/XDH activity were rapidly increased when challenged with muscle denervation. Besides that, the alterations involved VEGFR-2 or -1 could be modulated by IL-1β or IL-15. Importantly, VEGFR-2 functioned to rescue the damage on NMJ exerted by IL-1β or IL-15. Then, we realized that the damage on muscular function could be deteriorated along with muscle denervation, and VEGFR-2 was proposed to be important for the maintenance of NMJ.
2114
H. Zhao et al. / Cellular Signalling 25 (2013) 2106–2114
5. Conclusion Collectively, we demonstrated a dramatic up-regulation of mitochondrial and vascular oxidant production following muscle denervation, which was due to gradual disorder in mitochondrial biogenesis and XO/XDH activity, and accompanied by timely up-regulation of VEGFR-1 and VEGFR-2. It is of interest in this context that VEGFR signaling could be specifically initiated by IL-1β and IL-15 with involvement of mitochondrial and vascular oxidant production. Moreover, we demonstrated a preferentially disruption of NMJ function by IL-15, which could be rescued by VEGFR-2 but not VEGFR-1 activation. Thus, extensive network centered on VEGFR-2 signaling might be a key component necessary for the early recovery processes regarding muscle denervation, which was considered a strong attractant for the treatment on muscular disorders.
References [1] H. Hanwei, H. Zhao, Muscle & Nerve 42 (2010) 70–77, (Pubmed: 20544909). [2] H. Zhao, H. Huang, Cytokine 60 (2012) 608–615, (Pubmed: 23017227). [3] M.H. Laughlin, R.B. Armstrong, American Journal of Physiology 243 (1982) H296–H306, (Pubmed: 7114239). [4] P. Andersen, B. Saltin, The Journal of Physiology 366 (1985) 233–249, (Pubmed: 4057091). [5] J.M. Dawson, O. Hudlicka, International Journal of Microcirculation, Clinical and Experimental 8 (1989) 53–69, (Pubmed: 2722411). [6] X.J. Musacchia, J.M. Steffen, R.D. Fell, M.J. Dombrowski, Journal of Applied Physiology 69 (1990) 2248–2253, (Pubmed: 2077023). [7] B.M. Hather, G.R. Adams, P.A. Tesch, G.A. Dudley, Journal of Applied Physiology 72 (1992) 1493–1498, (Pubmed: 1534323). [8] O. Hudlicka, M. Brown, S. Egginton, Physiological Reviews 72 (1992) 369–417, (Pubmed: 1372998). [9] K. Tyml, O. Mathieu-Costello, E. Noble, Microvascular Research 49 (1995) 17–32, (Pubmed: 7746162). [10] O. Hudlicka, D. Pette, H. Staudte, Pflügers Archiv 343 (1973) 341–356, (Pubmed: 4796826). [11] H. Hoppeler, H. Howald, P. Cerretelli, Experientia 46 (1990) 1185–1187, (Pubmed: 2253720). [12] K. Tyml, O. Mathieu-Costello, L. Cheng, E.G. Noble, Journal of Applied Physiology 87 (1999) 1496–1505, (Pubmed: 10517784). [13] K. Tang, E.C. Breen, H.P. Gerber, N.M. Ferrara, P.D. Wagner, Physiological Genomics 18 (2004) 63–69, (Pubmed: 15084712). [14] A.F. Bruns, L. Bao, J.H. Walker, S. Ponnambalam, Biochemical Society Transactions 37 (Pt 6) (2009) 1193–1197, (Pubmed: 19909245). [15] P.D. Wagner, Biochemical Society Transactions 39 (2011) 1556–1559, (Pubmed: 22103486). [16] W.S. Jones, B.D. Duscha, J.L. Robbins, N.N. Duggan, J.G. Regensteiner, W.E. Kraus, W.R. Hiatt, A.O. Dokun, B.H. Annex, Vascular Medicine 17 (2012) 94–100, (Pubmed:22402934). [17] R. Gianni-Barrera, M. Trani, S. Reginato, A. Banfi, Biochemical Society Transactions 39 (2011) 1644–1648, (Pubmed: 22103501). [18] H. Singh, T.A. Tahir, D.O. Alawo, E. Issa, N.P. Brindle, Biochemical Society Transactions 39 (2011) 1592–1596, (Pubmed: 22103492). [19] B. Sitohy, J.A. Nagy, H.F. Dvorak, Cancer Research 72 (2012) 1909–1914, (Pubmed: 22508695). [20] S. Mittar, C. Ulyatt, G.J. Howell, A.F. Bruns, I. Zachary, J.H. Walker, S. Ponnambalam, Experimental Cell Research 315 (2009) 877–889, (Pubmed: 19162007). [21] J. Gutierrez, E. Brandan, Molecular and Cellular Biology 30 (2010) 1634–1649, (Pubmed: 20100867). [22] C.L. Smith, P. Mittaud, E.D. Prescott, C. Fuhrer, S.J. Burden, Journal of Neuroscience 21 (2001) 3151–3160, (Pubmed: 11312300). [23] K.A. Rivas-Plata, J.R. Kraas, S.M. Saleh, S.L. Swope, Journal of Neuroscience 21 (2001) 3806–3818, (Pubmed: 11356869). [24] B. Klitzman, D.N. Damon, R.J. Gorczynski, B.R. Duling, Circulation Research 51 (1982) 711–721, (Pubmed: 7139886).
[25] J.M. Dopp, N.R. Philippi, N.J. Marcus, E.B. Olson, C.E. Bird, J.J. Moran, S.W. Mueller, B.J. Morgan, Respiration 82 (2011) 458–467, (Pubmed: 21846958). [26] M.J. Jackson, Antioxidants & Redox Signaling 15 (2011) 2477–2486, (Pubmed: 21699411). [27] C. Zhang, X. Xu, B.J. Potter, W. Wang, L. Kuo, L. Michael, G.J. Bagby, W.M. Chilian, Arteriosclerosis, Thrombosis, and Vascular Biology 26 (2006) 475–480, (Pubmed: 16385082). [28] S. Basuroy, S. Bhattacharya, C.W. Leffler, H. Parfenova, American Journal of Physiology. Cell Physiology 296 (2009) C422–C432, (Pubmed: 19118162). [29] S. Brickson, J. Hollander, D.T. Corr, L.L. Ji, T.M. Best, Medicine & Science in Sports & Exercise 33 (2001) 2010–2015, (Pubmed: 11740292). [30] M.F. O'Leary, D.A. Hood, Autophagy 5 (2009) 230–231, (Pubmed: 19098460). [31] B. Chabi, P.J. Adhihetty, M.F. O'Leary, K.J. Menzies, D.A. Hood, Journal of Applied Physiology 107 (2009) 1730–1735, (Pubmed: 19797682). [32] P.M. Abruzzo, S. di Tullio, C. Marchionni, S. Belia, G. Fano, S. Zampieri, U. Carraro, H. Kern, G. Sgarbi, G. Lenaz, M. Marini, Free Radical Research 44 (2010) 563–576, (Pubmed: 20298122). [33] K.L. Wicks, D.A. Hood, American Journal of Physiology 260 (4 Pt 1) (1991) C841–C850, (Pubmed: 1850197). [34] P.M. Siu, S.E. Always, The Journal of Physiology 565 (Pt 1) (2005) 309–323, (Pubmed: 15774533). [35] J. Norrbom, C.J. Sundberg, H. Ameln, W.E. Kraus, E. Jansson, T. Gustafsson, Journal of Applied Physiology 96 (2004) 189–194, (Pubmed: 12972445). [36] A. Wagatsuma, N. Kotake, K. Mabuchi, S. Yamada, Journal of Physiology and Biochemistry 67 (2011) 359–370, (Pubmed: 21394548). [37] A. Wagatsuma, N. Kotake, S. Yamada, Molecular and Cellular Biochemistry 349 (2011) 139–147, (Pubmed: 21739150). [38] J.M. Marshall, H.C. Tandon, The Journal of Physiology 350 (1984) 447–459, (Pubmed: 6747856). [39] P.M. Hassoun, F.S. Yu, C.G. Cote, J.J. Zulueta, R. Sawhney, K.A. Skinner, H.B. Skinner, D.A. Parks, J.J. Lanzillo, American Journal of Respiratory and Critical Care Medicine 158 (1998) 299–305. [40] Y.W. Lee, W.H. Lee, P.H. Kim, Inflammation Research 59 (2010) 755–765, (Pubmed: 20349326). [41] S. Wind, K. Beuerlein, M.E. Armitage, A. Taye, A.H. Kumar, D. Janowitz, C. Neff, A.M. Shah, K. Wingler, H.H. Schmidt, Hypertension 56 (2010) 490–497, (Pubmed: 20606112). [42] H. Ishikita, B.T. Eger, K. Okamoto, T. Nishino, E.F. Pai, Journal of the American Chemical Society 134 (2012) 999–1009, (Pubmed: 22145797). [43] N.H. Herzberg, E. Middelkoop, M. Adorf, H.L. Dekker, M.J. Van Galen, M. Van den Berg, P.A. Bolhuis, C. Van den Bogert, European Journal of Cell Biology 61 (1993) 400–408, (Pubmed: 8223726). [44] N. Gleyzer, K. Vercauteren, R.C. Scarpulla, Molecular and Cellular Biology 25 (2005) 1354–1366, (Pubmed: 15684387). [45] R.C. Scarpulla, Physiological Reviews 88 (2008) 611–638, (Pubmed: 18391175). [46] C.M. LeMoine, S.C. Lougheed, C.D. Moyes, Journal of Molecular Evolution 70 (2010) 492–505, (Pubmed: 20443112). [47] M. Picard, T. Taivassalo, G. Gouspillou, R.T. Hepple, The Journal of Physiology 589 (Pt 18) (2011) 4413–4421, (Pubmed: 21708903). [48] O.T. Hernandez-Hernandez, T.K. Gonzalez-Garcia, I. Camacho-Arroyo, The Journal of Steroid Biochemistry and Molecular Biology 132 (2012) 127–134, (Pubmed: 22542550). [49] J.V. Frangioni, P.H. Beahm, V. Shifrin, C.A. Jost, B.G. Neel, Cell 68 (1992) 545–560, (Pubmed: 1739967). [50] F.G. Haj, B. Markova, L.D. Klaman, F.D. Bohmer, B.G. Neel, Journal of Biological Chemistry 278 (2003) 739–744, (Pubmed: 12424235). [51] Y. Romsicki, M. Reece, J.Y. Gauthier, E. Asante-Appiah, B.P. Kennedy, Journal of Biological Chemistry 279 (2004) 12868–12875, (Pubmed: 14722096). [52] M. Marino, F. Scuderi, C. Provenzano, E. Bartoccioni, Gene Therapy 18 (2011) 109–116, (Pubmed: 20927136). [53] R. Ciurea, C. Margaritescu, C. Simionescu, A. Stepan, M. Ciurea, Romanian Journal of Morphology and Embryology 52 (2011) 1227–1232, (Pubmed: 22203927). [54] M.P. Turunen, S. Yla-Herttuala, Cardiovascular Research 90 (2011) 441–446, (Pubmed: 21498421). [55] C.K. Magill, A. Tong, D. Kawamura, A. Hayashi, D.A. Hunter, A. Parsadanian, S.E. Mackinnon, T.M. Myckatyn, Experimental Neurology 207 (2007) 64–74, (Pubmed: 17628540). [56] Q. Wang, B. Zhang, Y.E. Wang, W.C. Xiong, L. Mei, Neurosignals 16 (2008) 246–253, (Pubmed: 18253062).