SEARCH

SEARCH BY CITATION

Keywords:

  • bioenergetics;
  • FoF1-ATP synthase;
  • myelin;
  • OXPHOS

Abstract

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References
  8. Supporting Information

Myelin sheath is the proteolipid membrane wrapping the axons of CNS and PNS. We have shown data suggesting that CNS myelin conducts oxidative phosphorylation (OXPHOS), challenging its role in limiting the axonal energy expenditure. Here, we focused on PNS myelin. Samples were: (i) isolated myelin vesicles (IMV) from sciatic nerves, (ii) mitochondria from primary Schwann cell cultures, and (iii) sciatic nerve sections, from wild type or Charcot-Marie-Tooth type 1A (CMT1A) rats. The latter used as a model of dys-demyelination. O2 consumption and activity of OXPHOS proteins from wild type (Wt) or CMT1A sciatic nerves showed some differences. In particular, O2 consumption by IMV from Wt and CMT1A 1-month-old rats was comparable, while it was severely impaired in IMV from adult affected animals. Mitochondria extracted from CMT1A Schwann cell did not show any dysfunction. Transmission electron microscopy studies demonstrated an increased mitochondrial density in dys-demyelinated axons, as to compensate for the loss of respiration by myelin. Confocal immunohistochemistry showed the expression of OXPHOS proteins in the myelin sheath, both in Wt and dys-demyelinated nerves. These revealed an abnormal morphology. Taken together these results support the idea that also PNS myelin conducts OXPHOS to sustain axonal function.

Abbreviations used
ANT

adenosine nucleotide translocase

CMT1A

Charcot-Marie-Tooth type 1A

COX II

cytochrome c oxydase, subunit II

EM

electron microscopy

ETC

electron transfer chain

IMV

isolated myelin vesicles

MPZ

myelin protein zero

OXPHOS

oxidative phosphorylation

PBS

phosphate-buffered saline

SC

Schwann cells

TIM

inner membrane translocase, subunit 8A

TOM

outer membrane translocase, subunit 20

Wt

wild type

The Nervous System (NS) displays a great energy demand, and critically depends on oxygen (O2) supply (Ames 2000). Faster transmitting axons are surrounded by myelin, a multilamellar membrane produced by highly specialized glial cells, i.e., Schwann cells (SC) in PNS and oligodendrocytes in CNS. Besides being an insulating wrap, myelin has been proposed to play a trophic role, as its loss causes axonal degeneration (Ferguson et al. 1997; Trapp and Nave 2008). We have recently proposed the existence of an extramitochondrial aerobic metabolism in myelin from bovine forebrain (Ravera et al. 2011, 2009; Morelli et al. 2011; and in retinal rod outer segment disks (Panfoli et al. 2008, 2009), both devoid of mitochondria. The expression of mitochondria-derived proteins of the oxidative phosphorylation (OXPHOS) was reported in many proteomic analyses of CNS (Taylor et al. 2004),(Vanrobaeys et al. 2005; Werner et al. 2007; Yamaguchi et al. 2008; Ishii et al. 2009; Jahn et al. 2009) and in the recent analysis of PNS myelin proteome (Patzig et al. 2011). The presence of proteins involved in aerobic respiration in the myelin fraction was also reported by large-scale proteomics screening of cerebrospinal fluid children during initial presentation of CNS inflammation (Dhaunchak et al. 2012). Consistently, it has been suggested that CNS myelin would facilitate nerve impulse transmission supporting the axonal mitochondria in supplying ATP rather than by limiting energy consumption (Ravera et al. 2009, 2011; Morelli et al. 2011). In this article, we focused on PNS myelin to investigate whether it plays the same energetic role as the CNS one. A transgenic rat, reproducing Charcot-Marie-Tooth type 1A (CMT1A) disease, was utilized as a model of dys-demyelination (Sereda et al. 1996). Isolated myelin vesicles (IMV) from rat sciatic nerves were assayed for their respiratory properties by biochemical and confocal scanning or transmission electron microscopy techniques. IMV from the CMT1A sciatic nerves lose the capacity to synthesize ATP and consume oxygen, with time. Moreover, an increment in mitochondrial density was observed in dys-demyelinated axons. Overall data are consistent with the idea of a pivotal role of myelin, besides axonal mitochondria, in conducting oxidative phosphorylation in PNS, and its impairment in a model of dys-demyelination.

Materials and methods

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References
  8. Supporting Information

Antibodies

Primary antibodies (Ab) were: monoclonal antibody against myelin protein zero (MPZ) (P07 extracellular domain, Astexx Ltd. & Co. KEG, Graz, Austria); polyclonal antibodies against: adenosine nucleotide translocase (ANT) (Q-18, sc-9300 Santa Cruz Biotechnology, Santa Cruz, CA, USA); mitochondrial import inner membrane translocase, subunit 8A (TIM) (N-20, sc-17052, Santa Cruz Biotechnology); mitochondrial import outer membrane translocase, subunit 20 (TOM) (C-20, sc-11021, Santa Cruz Biotechnology); FoF1-ATP synthase α-subunit (A9585, Sigma-Aldrich, St. Louis, MO, USA); NADH-ubiquinone oxidoreductase subunit (ND4L) (H-94, sc-20665, Santa Cruz Biotechnology), and cytochrome c oxydase, subunit II (COX II) (Q-18, sc-23983, Santa Cruz Biotechnology). Secondary Ab for immunofluorescence were: Alexa-488-conjugated anti-rabbit or anti-goat (Molecular Probes, Grand Island, NY, USA) or Cy3-conjugated anti mouse (Sigma-Aldrich). Horseradish peroxidase-conjugated mouse or rabbit IgG Ab (Santa Cruz Biotechnology) were used for western blot experiments;

Animal model

Transgenic Sprague–Dawley rats over-expressing PMP22 (CMT1A rats) were genotyped by PCR (Sereda et al. 1996). Heterozygous animals and normal age-matched littermates were used for experiments. Rearing conditions were consistent with the guidelines of the Italian Health Ministry relating to the use and storage of transgenic organisms. Experiments were carried out in compliance with animal care requirements requested by Italian law (law D.L. 27.1.1992 n. 116, in agreement with the European Union directive 86/609/CEE).

Primary Schwann cell cultures

SC for primary cultures were isolated from 3-day-old rat sciatic nerves, as previously described (Brockes et al. 1979; Nobbio et al. 2004). SC were grown in Dulbecco's modified Eagle's medium/F12 medium (Invitrogen, Grand Island, NY, USA) supplemented with 10% fetal calf serum (FCS), penicillin and streptomycin for 48 h. Cytosine arabinoside (Ara-C) 10−5 M was added after 48 h and the treatment prolonged for further 48 h resulting in SC cultures that were 99% pure and were used for biochemical experiments.

Isolation of Schwann cell mitochondria-enriched fractions

Mitochondria-enriched fractions from SC cultures were isolated by standard differential centrifugation techniques, as described (Ravera et al. 2007). SC were homogenated in 0.25 M sucrose, 5 mM N-2-hydroxyethylpiperazine-N1-2-ethanesulfonic acid (HEPES) pH 7.2, protease inhibitor cocktail, 30 nM Cyclosporin A, 30 nM 2-Aminoethoxydiphenyl borate and 0.060 mg/mL Ampicillin, and centrifuged at 700 g for 10 min, in Heraeus centrifuge. Pellet was discarded, and supernatant centrifuged at 10 900 g for 10 min. Pellet was retained as a mitochondria-enriched preparation, stored on ice and used within 4 h of isolation. All steps were performed at 4°C.

Myelin isolation

IMV were purified from sciatic nerves of 1- or 12-month-old wild type (Wt) or CMT1A rats by a modification of the method of Norton and Poduslo (Haley et al. 1981; Larocca and Norton 2007; Norton and Poduslo 1973). Briefly, sciatic nerves were homogenized by a Potter-Elvehjem procedure in 20 vol. (w/v) of 0.32 M sucrose in 2 mM EGTA and 100 mM Tris-HCl pH7.5. Myelin enriched fractions were obtained by differential centrifugation after hypotonic shock of the nerve homogenate. The sciatic nerve homogenate was layered onto 0.85 M sucrose in 2 mM EGTA and centrifuged at 75 000 g for 30 min in a TL 100 ultracentrifuge (Beckman Coulter, Brea, CA, USA). Protease inhibitor cocktail (Sigma-Aldrich), 30 μg/mL 5-fluorouracil, and 20 μg/mL ampicillin were present throughout. Oxygen consumption and ATP synthesis were observed in the absence of mitochondrial permeability transition pore inhibitors (Berman et al. 2000), as cyclosporin A or 2-aminoethoxydiphenyl borate (Chinopoulos et al. 2003; Crompton et al. 1988). The method utilized to obtain IMV minimizes contamination. In fact, during the first step the membrane fraction is separated from the cellular organelles that are found in the precipitate. Moreover, shocks in distilled water eliminate the contaminating mitochondria, if any, both physically and functionally. In fact, we have shown that mitochondria subjected to hypotonic treatment do not respire nor synthesize ATP (Panfoli et al. 2009).

Western blot

Equal protein amounts (20 μg) from both Wt and transgenic animal samples were loaded onto gradient 8–12% polyacrylamide gels and separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis. Proteins were transferred to nitrocellulose membranes (Amersham Biosciences, Glattbrugg, Switzerland) and probed overnight with primary antibodies against MBP (1 : 10 000); PMP22 (1 : 200); MPZ (1 : 400); ANT (1 : 400); TIM (1 : 200); TOM (1 : 200); Adenylate kinase, isoform 3 (AK3) (1 : 200). Secondary Horseradish peroxidase-conjugated Abs were diluted in phosphate-buffered saline (PBS) (1 : 5000 for anti-mouse IgG; 1 : 7500 for anti-rabbit and anti-goat IgG). Bands were visualized and quantified by ChemiDoc XRS system (Bio-Rad Laboratories, Hercules, CA, USA). Total protein content of mitochondria-enriched fractions IMV or sciatic nerves homogenates was determined by Bradford method, using bovine serum albumin as a standard protein.

Adenylate kinase, isoform 3 assay

Spectrophotometric assay of AK3, a protein typical of mitochondrial matrix, was performed according to Tomasselli et al. (1979).

Mitochondrial DNA quantification in IMV

To assess the presence/absence of mitochondrial DNA, a PCR analysis against a typical sequence of bovine mitochondrial DNA was carried out. The primer sequences were designed by the on-line Primer3 software program (Rozen and Skaletsky, 2000) and were: forward 5′-GCTAAGACCCAAACTGGGATT-3′, reverse 5′-AGCCCATTTCTTCCCATTTC-3′. Amplification reactions were carried out in a total volume of 20 μL, in a reaction mixture containing 5 × PCR-Buffer, 2 mM MgCl2, 1 pmol of each primer, 0.2 mM deoxynucleoside triphosphates, and 1 U of AmpliTaq DNA polymerase (PerkinElmer, Emeryville, CA, USA). The samples were subjected to a PCR program consisting of 1 cycle at 95°C for 2 min, then 12–27 cycles at 95°C for 30 s for denaturing, 56°C for 45 s for annealing, and 72°C for 30 s for extension, finally 1 cycle at 72°C for 7 min. The PCR products were separated onto 1% agarose gels in Tris-acetate-EDTA buffer and stained with ethidium bromide (0.4 mg/mL).

Oxygraphic measurements

O2 consumption by IMV was assayed in a thermostatically controlled oxygraph equipped with an amperometric electrode (Unisense – Microrespiration, Unisense A/S, Copenhagen, Denmark), according to manufacturer's instructions, as previously described (Ravera et al. 2009). To observe uncoupled respiration rates, 30 μM nigericin was added before the addition of 0.7 mM NADH (to stimulate the ETC. I + III + IV) and 20 mM succinate (to stimulate the ETC. II +III +IV). To verify that the oxygen consumption was because of the electron transfer chain (ETC.) activity, 40 μM rotenone (ETC. I inhibitor) and 50 μM Antimycin A (ETC. III inhibitor) were added. Respiratory rates are expressed as μM O2/min/mg.

Analysis of respiratory complexes

The ETC. complexes were tested on IMV by spectrophotometric assays: ETC. I (NADH-ubiquinone oxidoreductase) following the reduction of ferricyanide at 420 nm (Sottocasa et al. 1967); ETC. II (Succinic dehydrogenase) following the reduction of dichloroindophenol at 600 nm, (Janssen et al. 2007); ETC. III (cytochrome c reductase) following the reduction of oxidized cytochrome c at 550 nm (Sottocasa et al. 1967); ETC. IV (cytochrome c oxydase) following the oxidation of Ascorbate-reduced cytochrome c at 550 nm (Baracca et al. 2003).

ATP synthesis assay

ATP synthesis by IMV was assayed by the luciferin/luciferase chemiluminescent method (Roche Applied Science, Indianapolis, IN, USA), according to Sgarbi et al. (2006). Samples (0.01 mg protein/mL) were incubated for 5 min at 37°C in 10 mM Tris/HCl (pH 7.4), 100 mM KCl, 1 mM EGTA, 2.5 mM EDTA, 1 mM NaH2PO4, 2 mM MgCl2, 0.7 mM NADH, 20 mM succinate and 25 μg/mL ampicillin. To inhibit adenylate kinase and Na+/K+ ATPase activity, 0.2 mM di(adenosine-5) penta-phosphate and 0.6 mM ouabain were added, respectively. ATP synthesis was induced by adding 0.5 mM ADP to the sample, at the same pH of the mixture. 10 μM Oligomycin, 0.5 mM potassium cyanide or 50 μM antimycin A (Ravera et al. 2009) were added to the incubation mixture as inhibitors. ATP standard solutions between 10−9 and 10−7 M were used for calibration (Roche Diagnostics Corp., Indianapolis, IN, USA).

Immunohistochemistry

Sciatic nerves were fixed in 4% paraformaldehyde in PBS (pH 7.4) for 12 h, then dehydrated and embedded in Paraplast. Transversal sections (5 μm thick) were cut, rehydrated, dewaxed, and incubated with rabbit antibodies against FoF1-ATP synthase (1 : 100), ND4L (1 : 100) or COX II (1 : 100) together with a mouse antibody against MPZ, in a moist chamber at 4°C, overnight. After washing, sections were incubated with the secondary antibodies: Alexa-488-conjugated anti-rabbit and Cy3-conjugated anti-mouse. Sections were then washed in PBS, mounted in a glycerol/PBS (1 : 1) solution, and examined under an inverted Leica TCS SP5 AOBS confocal laser scanning microscope, equipped with a Leica 63XPL APO NA 1.40 oil immersion objective (Leica Microsystems CMS, Mannheim, Germany).

Electron microscopy

For transmission electron microscopy (EM) examination, 6-month-old CMT1A rats were killed and sciatic nerves were quickly removed. Specimens were fixed in 2.5% glutaraldehyde in cacodylate buffer, pH 7.4, for 2 h, post-fixed with 1% osmium tetroxide in cacodylate buffer, pH 7.4, for 1 h, dehydrated in alcohol and embedded in epoxy resin. Ultrathin sections, stained with 5% uranyl acetate and lead citrate, were then prepared from these specimens, and examined under a Zeiss EM 109 (Oberkochen, Germany).

Statistical analysis

Data are expressed as Mean ± SD Differences between groups were determined using unpaired Student's t-test. Multiple-parameter comparisons were performed by one-way anova followed by Bonferroni's post hoc test. A value of p < 0.05 was considered statistically significant. GraphPad Prism 5.0 Software (GraphPad Software Inc., La Jolla, CA, USA) was used for statistical analysis.

Results

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References
  8. Supporting Information

Characterization of IMV from sciatic nerve

The method utilized to purify myelin from sciatic nerves (Larocca and Norton 2007) IMV by hypotonic shock of the homogenate, minimizing contamination by subcellular organelle. In fact the membrane fraction and the organelle lay in different layers of the sucrose step gradient (see 'Materials and methods'). Fig. 1 shows a characterization of the IMV purified from Wt sciatic nerves from 12-month-old rats conducted by semiquantitative WB analysis. Panel (a) shows the protein pattern of mitochondria-enriched fractions (lane 1) sciatic nerve homogenate (lane 2) and IMV (lane 3), as stained with Colloidal Blue Coomassie. Semiquantitative WB analyses were conducted with antibodies against MBP(Panel b) a marker of myelin; PMP22 (Panel c) and MPZ (Panel d), markers of peripheral myelin; ANT (Panel e) and TIM (Panel f), two proteins typical of the inner mitochondrial membrane; TOM (Panel g), a marker of the outer mitochondrial membrane; AK3 (Panel h), a protein typical of mitochondrial matrix. The intensity of the chemiluminescent WB signal of myelin markers increased in parallel to the isolation of myelin. In contrast, in IMV the signal of typical mitochondrial proteins is absent. Data are confirmed by the densitometric analysis of chemiluminescent signal, reported in Panel (i). To confirm the absence of mitochondrial matrix proteins, AK3 was also assayed. Panel (j) shows that AK3 activity was present in mitochondria-enriched fractions and in the pellet obtained after the first step of WT or CMT1A myelin isolation (composed by nuclei, mitochondria, and cellular debris), decreased in sciatic nerve homogenate and was absent in isolated myelin fractions.

image

Figure 1. Characterization of rat sciatic nerve-derived fractions. Figure shows the semiquantitative WB analysis of typical myelin proteins, as MBP, PMP22, myelin protein zero (MPZ) and typical mitochondrial proteins, as adenosine nucleotide translocase (ANT), inner membrane translocase, subunit 8A (TIM), outer membrane translocase, subunit 20 (TOM) and Adenylate kinase, isoform 3 (AK3) in mitochondria-enriched fractions (lane 1), sciatic nerve homogenate (lane 2) and isolated myelin vesicles (IMV, lane 3) from 12-month-old wild type (Wt) rats. Panel (a) reports the protein pattern stained with Colloidal Blue Coomassie. In panels (b–h) WB analysis against MBP, PMP22, MPZ, ANT,TIM, TOM and AK3 is shown, respectively. The densitometric analysis of WB signals is reported in Panel (i). To further assess the absence of contamination from mitochondrial matrix we checked for AK3 activity in PNS IMV. AK3 activity was strongly impaired in SN, and undetectable in IMV compared to MF (Panel j). Panel (k) shows the amplification of a specific mitochondrial DNA sequence. The signal is present only in the mitochondrial sample, used as positive control, and absent in IMV. Each Panel is representative of at least 10 different experiments.

Download figure to PowerPoint

To demonstrate the absence of mitochondrial DNA in isolated myelin vesicles, a PCR analysis was performed, using primers specific for a mitochondrial DNA sequence (see 'Materials and methods'). Panel (k) shows that it was already possible to observe the signal after 10 cycles of amplification exclusively in the mitochondrial sample. In contrast, in IMV the signal was still absent after 30 cycles.

OXPHOS proteins are functional in IMV from Wt rats, but impaired in a model of peripheral dys-demyelination

Having reported that CNS myelin expresses the five complexes of respiration and is able to consume O2 (Ravera et al. 2009), we tested IMV from sciatic nerve of Wt 12-month-old rats for their ability to consume O2, as SC and oligodendrocytes do not share the same embryonic origin. O2 consumption of nigericin-uncoupled on IMV from sciatic nerves from Wt and CMT1A rats aged 1 and 12 months, respectively, was measured at 23C in an amperometric electrode (Unisense A/S, Denmark). Results are in Fig. 2. The maximal respiratory fluxes elicited by NADH and succinate were measured as rates sensitive to inhibition by rotenone and antimycin A, respectively, to demonstrate that the oxygen consumption is because of the OXPHOS proteins. Panels (a, and c) show the respiratory rates IMV from 12-month-old or 1-month-old Wt rats. O2 consumption in IMV from CMT1A 12-month-old rats was severely impaired (Panel b). In contrast, respiratory rates of IMV from Wt and CMT1A 1-month-old rats, and from Wt 12-month-old rats appear comparable (Panels a, c, and d). Data suggest that peripheral myelin sheath is metabolically functional in the first stages of the dys-demyelinating process, but with the progression of the pathology, it becomes unable to conduct oxidative phosphorylation. Mitochondria-enriched fractions from primary cultures of both wild type and CMT1A SC were used as controls to exclude mitochondrial contribution in the impairment of respiration in affected IMV. Respiratory rates of mitochondria from primary cultures of SC cells from both Wt and CMT1A 12-month-old rats were also assayed. Panels (e and f) show that these are fully functional.

image

Figure 2. Oxygen consumption. (a–f) Amperometric tracings of nigericin-uncoupled respiratory rates in isolated myelin vesicles (IMV) from Wt and CMT1A rat sciatic nerves, and mitochondria-enriched fractions extracted from Schwann cell (SC). (a) O2 consumption by IMV coming from Wt 12-months-old rats. (b) O2 consumption was severely impaired in IMV from CMT1A 12-months animals. No difference was found in O2 consumption by Wt IMV and CMT1A IMV in the early stages of myelination (1-month-old rats) (c, d) and by mitochondria extracted from primary SC cultures from both 12-months-old normal and CMT1A rats (e, f).

Download figure to PowerPoint

IMV ETC. I and IV are functional in Wt but are impaired in dys-demyelinated animals

The above data suggest that the impairment in O2 consumption observed in IMV from 12-month-old CMT1A rats is not because of damage of SC mitochondria, but to structural myelin defects. To assess whether the ETC. is active in myelin isolated from Wt or affected 12-month-old animals, activity of ETC. I (NADH-Ubiquinone oxidoreductase, Panel a) and IV (cytochrome c oxydase, Panel b), was assayed in IMV, in the presence or absence of specific inhibitors (rotenone for I and cyanide for IV). Activity of I and IV was reduced in 12-month-old CMT1A (Fig. 3, Panels a and b) but not in IMV from 1-month-aged CMT1A rats (data not shown). In particular, NADH oxidoreductase lost about 86% of its activity (Panel a, third column), while cytochrome c oxydase, about 65% (Panels d, third column). These data are consistent with the impairment in O2 consumption by IMV from CMT1A 12-month-old rats (see Fig. 2, Panel b). Inhibition by rotenone or cyanide ranged around 80% and 98%, respectively, suggesting that the activities are specific.

image

Figure 3. Activity of electron transfer chain (ETC.) I and IV. (a, b) Sciatic-nerve-derived isolated myelin vesicles (IMV) from Wt rats show the typical mitochondrial activity ETC. I and IV. In fact, the specific inhibition of each of the two complexes by rotenone and cyanide, respectively, dramatically reduced the respiratory properties of the Wt IMV compared to the untreated normal controls (ETC. I: mean ± SD: 0.37 ± 0.05 vs. 1.82 ± 0.11 U/mg, respectively; n = 5; ***p < 0.0001), (ETC. IV: Mean ± SD: 0.0024 ± 0.0005 vs. 0.09 ± 0.008 U/mg, respectively; n = 5; ***p < 0.0001). Interestingly, IMV from CMT1A rats showed a significantly reduced activity of ETC. I and IV compared to the Wt ones (ETC. I: mean ± SD: 0.26 ± 0.04 vs. 1.82 ± 0.11 U/mg, respectively; n = 5; ***p < 0.0001), (ETC. IV: mean ± SD: 0.03 ± 0.004 vs. 0.09 ± 0.008 U/mg, respectively; n = 5; ***p < 0.0001), resembling the respiratory rates observed in IMV from Wt rats following specific inhibition of complexes I and IV by rotenone and cyanide, respectively.

Download figure to PowerPoint

IMV synthesize ATP

In coupled conditions, the ETC. activity is linked to ATP synthase functioning, i.e. O2 consumption elicits an ATP production, similarly to mitochondria. Fig. 4, Panel (a), shows the extra vesicular ATP synthesis by IMV from Wt 12-month-old rats. Activity accounted for 37 nmol ATP/min/mg of total proteins and was sensitive to the FoF1-ATPase/H+-pump inhibitor oligomycin (90%), suggesting specificity of ATP production. Moreover, ATP synthesis was inhibited by KCN (99%) and Antimycin A (64%), inhibitors of ETC. IV and III, respectively, indicating that ATP synthesis is coupled with the electron transport chain. Panel (b) shows that, ATP synthase activity in IMV from affected 12-month-old rats is significantly impaired, as compared to the Wt at each time point.

image

Figure 4. ATP synthesis in isolated myelin vesicles (IMV) from sciatic nerves. IMV from Wt rats synthesize ATP through FoF1-ATP synthase activity (a). In fact, in the presence of classical inhibitors of mitochondrial ATP synthase, they lose this ability (oligomycin: Mean ± SD: 38 ± 4 vs. 4.3 ± 0.4 nmol ATP/min/mg; n = 5; ***p < 0.0001), (KCN: mean ± SD: 38 ± 4 vs. 0.4 ± 0.05 nmol ATP/min/mg; n = 5; ***p < 0.0001), antimycin A: mean ± SD: 38 ± 4 vs. 13.6 ± 0.8 nmol ATP/min/mg; n = 5; ***p < 0.0001). (b) Time course of ATP synthesis was significantly impaired in IMV from CMT1A adult rats, as compared to the Wt ones at each time point (30 s: Mean ± SD: 18.13 ± 1.9 vs. 3.3 ± 0.3 nmol ATP/min/mg; n = 4; ***p < 0.0001), (60 s: mean ± SD: 37.7 ± 2.1 vs. 6.1 ± 0.7 nmol ATP/min/mg; n = 4; ***p < 0.0001), (90 s: mean ± SD: 55.3 ± 5.1 vs. 8.6 ± 0.8 nmol ATP/min/mg; n = 4; ***p < 0.0001).

Download figure to PowerPoint

The OXPHOS proteins colocalize with MPZ in 12-month-old Wt rat sciatic nerves and in the model of dys-demyelination

The possibility to assay the ETC. in IMV implies the expression in the PNS myelin sheath of the mitochondrial respiratory complexes, as reported in its CNS counterpart (Ravera et al. 2009). To verify this hypothesis an immunohistochemical analysis with antibodies against MPZ, ND4L (a subunit of ETC. I) (Panel a), COX-II (a subunit of ETC. IV) (Panel b) or β-subunit of FoF1-ATP synthase (Panel c) was conducted on transversal sections of sciatic nerves from 12-month-old Wt or CMT1A rats. The green fluorescent signal of each of the OXPHOS proteins assayed colocalizes, in normal and CMT1A nerves, with the red fluorescent signal of MPZ suggesting that these mitochondrial subunits are present in the myelin sheath of both conditions. As expected, CMT1A tissue shows fewer myelinated fibers and some derangement of the myelin sheath in the surviving fibers. Localization studies then suggest that the impairment of Oxygen consumption in IMV of 12-months-old-CMT1A rats (see Figs 2-4) is not because of the absence of the OXPHOS proteins, rather to their altered function.

Mitochondrial density is increased in demyelinated fibers of affected sciatic nerves

The above data prompted us to verify whether mitochondria are recruited into the affected axons. The histogram (Figure S1) shows that the number of axonal mitochondria in demyelinated fibers from CMT1A rat was significantly increased, as compared to the number of mitochondria in the normally myelinated fibers of the Wt counterparts. It has been reported both in acquired and hereditary demyelinating disorders that mitochondria are recruited to the demyelinated regions to meet the increased energy requirements necessary to maintain conduction as an adaptive change (Trapp and Nave 2008; Hogan et al. 2009; Saporta et al. 2009).

Discussion

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References
  8. Supporting Information

Data show that IMV purified from rat sciatic nerve consume O2 (Fig. 2), and display functional ETC. proteins and ATP synthase, sensitive to classical mitochondrial inhibitors (Figs 3 and 4), supporting the hypothesis that PNS myelin may metabolically supply the axons with aerobically synthesized ATP, similarly to what reported for CNS (Ravera et al. 2009; Morelli et al. 2011). Therefore, myelin protein complement would be determined by its functional requirement of ATP supplier, regardless of the embryonic origin of the producing cell. A mitochondrial contamination of IMV has to be considered, however, absence of ANT, TIM, TOM, and AK3, as well as the absence of mitochondrial DNA (Fig. 1) suggests that our IMV preparations are negligibly contaminated. Moreover, to be fully functional, a metabolic process such as ATP synthesis associated to O2 consumption, needs a perfectly coupled system, which would not be represented by mere contaminating mitochondrial vesicles. Oligodendrocytes-synthesized CNS myelin has been proposed to act as an energy supplier for the axons (Ravera et al. 2009; Morelli et al. 2011), with a pivotal role of connexins in radial transport of the aerobically synthesized ATP to the axoplasm (Adriano et al. 2011). Data are also consistent with several proteomic studies, (Taylor et al. 2004; Vanrobaeys et al. 2005; Werner et al. 2007; Yamaguchi et al. 2008; Ishii et al. 2009; Jahn et al. 2009; Patzig et al. 2011; Dhaunchak et al. 2012), reporting the presence of proteins involved in aerobic respiration in purified myelin. The proteomic analysis of IMV from PNS, isolated by the same technique used here, suggested that some of the identified mitochondrial proteins may be additional myelin components (Werner et al. 2007). Also, glucose metabolism was considered over-represented in the myelin proteome (Patzig et al. 2011).

The idea of myelin sheath conducting an oxidative phosphorylation agrees with the observation that IMV from an animal model of dys/demyelination, the CMT1A rat, display an impaired O2 consumption, and OXPHOS protein activity at later stages (12-months-old animals), but similar to Wt in the early stages of myelination (1-month-old animals) (Figs 2-4). O2 consumption is impaired in IMV from 12-month-old CMT1A rats, with respect to Wt animals, but not in 1-month-old CMT1A rats. However, mitochondria from cultured Schwann cells from Wt and CMT1A rats (both 12-months old) showed normal O2 consumption rates (Fig. 2). Activity of the ETC. I and IV in IMV from sciatic nerve (Fig. 3) is consistent with the reports that myelin contains cardiolipin, (Heape et al. 1986), a lipid necessary for COX function (Zhang and Sejnowski 2000).The sensitivity of ATP synthesis to KCN, that was also able to inhibit ETC. IV (Fig. 3) and to antimycin A, an inhibitor of ETC. III (Fig. 4), suggests that ATP synthesis by IMV is coupled to the formation of a proton gradient, similarly to mitochondria (Boyer 1997). Confocal immunofluorescence imaging of sciatic nerve sections (Fig. 5) is suggestive of a colocalization of OXPHOS proteins with MPZ, in myelin from either Wt or CMT1A rats. Even though confocal microscopy does not possess the necessary resolution, such data are similar to those reported by the same technique in CNS (Ravera et al. 2009). A specific interaction between Schwann cells and axons was reported at the nodes of Ranvier (Gatzinsky et al. 2003), which may function in disposal of organelles. Further experiments may tell whether the OXPHOS proteins are true components of myelin sheath. Nevertheless, Fig. 5 would suggest that the impairment in O2 consumption, ATP synthesis and ETC. I and IV activity in CMT1A 12-month-old rats may not depend on the absence of the OXPHOS proteins, but to the dys/demyelinated phenotype, displaying a defective sheath. Indeed OXPHOS proteins were found ectopically expressed in many different cell lines membranes and in some cases they were active (Champagne et al. 2006; Panfoli et al. 2011a,b). Several reports suggest that mitochondrial dynamics involve a close contact between organelles like mitochondria and endoplasmic reticulum (Crotty and Ledbetter 1973; Hales 2004; Soltys and Gupta 2000; Giorgi et al. 2010). In this view, it was speculated that mitochondrial proteins may be transferred to the sheath during its development (Morelli et al. 2011). The phenotype of mitochondrial disorders (Nicholls and Budd 2000), a group of human diseases characterized by defects of the OXPHOS, primarily affect visual system, CNS, and PNS (Zeviani and Di Donato 2004), also supports the idea hypothesis that myelin exerts its trophic function by expressing OXPHOS proteins. Interestingly, mutations or knockout of ANT, a mitochondrial inner membrane protein not directly involved in oxidative phosphorylation, but in ATP delivery from the mitochondrion to the cytosol (Yang et al. 2007), do not affect the nervous tissue, but result in viable phenotypes involving cardiac and muscle dysfunction (Sharer 2005; Jang and Lee 2006). Along this line, it was shown that the energy budget for the myelinated axon is consistent with its reduced cost of action potential (however, assuming that myelinated axons use 23% of the ATP consumed by unmyelinated axons (Morelli and Panfoli 2012), but it was not excluded that some ATP production is driven by metabolic cooperation with glia (Harris and Attwell 2012a).

image

Figure 5. Immunofluorescence microscopy reveals electron transfer chain expression in rat sciatic nerves. A normal distribution of myelin protein zero (MPZ) was detected in 12-month-old rat sciatic nerves from both Wt and CMT1A animals, Panels (a–c, red signal). ND4L (a subunit of mitochondrial NADH oxidoreductase, Panel a), cytochrome c oxydase, subunit II (a subunit of mitochondrial Cytochrome oxydase, Panel b) and FoF1-ATP synthase β subunit (Panel c) are expressed in myelin (green signals), colocalizing with MPZ (Panels a–c, yellow signal). This suggests that the oxidative phosphorylation proteins decrement is not because of absence, but a lack of functionality. Images were recorded onto an inverted Leica TCS SP5 AOBS confocal laser scanning microscope.

Download figure to PowerPoint

Our observation that myelin from a model of dys/demyelination is not able to aerobically produce ATP in spite of a normal functioning of Schwann mitochondria, even in the presence of an increased mitochondrial number, supports the idea of an energetic role played by peripheral myelin. Supposing that in dys-demyelinating disorders, like CMT1A, a considerable impairment in myelin function is because of structural and functional alterations (Nobbio et al. 2004; Avila et al. 2005), these abnormalities might compromise the activity of the OXPHOS proteins, which would become unable to energetically support the axon. Recently, the hypothesis of myelin sheath able to conduct aerobic metabolism has been challenged (Harris and Attwell 2012b), however, other Authors reported an extramitochondrial ATP synthesis and expression of the ETC. and ATP synthase on the plasma membrane of cancer cells, a site characterized by an inverted membrane potential with respect to that present in the inner mitochondrial membrane (Chang et al. 2012). Also, a less compact myelin sheath, might negatively influence the radial diffusion of small molecules as ATP between the axonal and perinuclear SC cytoplasm (Balice-Gordon et al. 1998; Adriano et al. 2011). As recently suggested, an incorrectly assembled myelin can indeed be ‘toxic’ for the axon (Nave 2010). Consistently, in sciatic nerves from affected 12-month-old rats transmission EM showed a significantly increased number of mitochondria (Figure S1) with respect to matching unaffected controls. Therefore, based on our hypothesis that myelin is responsible for an aerobic metabolism in peripheral myelinated fibers, it is tempting to speculate that recruitment of mitochondria at the site of demyelination, represents an attempt to compensate for the aliquot of energy supply because of myelin.

In conclusion, the idea of a new energetic function of myelin is emerging (Ames 2000; Morelli et al. 2011). Our observation that IMV from peripheral nerves express functional ETC. proteins and FoF1-ATP synthase, as well as O2 consumption, ATP synthesis and OXPHOS functionality strongly support this hypothesis.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References
  8. Supporting Information

This study was supported by a Grant from the ‘Compagnia di San Paolo’- Neuroscience Program, for the research project entitled: ‘Energetic metabolism in myelinated axon: a new trophic role of myelin sheath’. The Authors have no conflict of interest to declare.

References

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References
  8. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References
  8. Supporting Information
FilenameFormatSizeDescription
jnc12253-sup-0001-FigS1.pdfapplication/PDF38KFigure S1. Evaluation of mitochondrial recruitment in CMT1A sciatic nerves.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.