Bioi Cell (1991) 71, 57-65 © Elsevier,Paris
57 Original article
Mitochondrial maturation during neuronal differentiation in vivo and in vitro Laurence Cordeau-Lossouarn, Jean-Luc Vayssi~re, Jean-Christophe Franfois Gros, Bernard Croizat
Larcher,
Laboratoire de Biochimie Ceilulaire, URA 1115 CNRS, Coll~ge de France, 11 place Marcelin-Berthelot, 75231 Paris Cedex 05, France (Received 31 October 1990; accepted 11 February 1991) Summary - The evolution of the mitochondrion has been followed within differentiating neuronal cells, both in primary cultures of neurons from fetal rat cortex and during rat brain cortex maturation. Changes in total mitochondrial proteins (mt-proteins) were evaluated, and qualitative changes in the mt-proteins pattern were analyzed using the Western blot technique. The evolution of mtprotein contents in cultured neurons resembles what is observed during rat brain maturation. The mitochondrion exhibits pronounced changes in the course of neurogenesis, in particular, bursts of mitochondrial masses accompanying the successive steps of neurogenesis are observed. There are indications that protein equipment of mitochondria during neuronal development undergoes variations. Although more work is required to establish the significance of these correlations, the present data might suggest an important role of the mitochondrion in neurogenesis. primary culture of neurons I neurogenesisI quantitative Western blot I miteehondrial maturation
Introduction Mitochondria are near ubiquitous organelles of eukaryotic cells. While their principal function is oxidative phosphorylation, they also contribute to the biosynthesis of pyrimidines, aminoacids, phospholipids, nucleotides, folate coenzymes, heme, urea and many other metabolites [8]. It is thus not surprising that mitochondria found in various animal tissues show considerable differences in physiological activity, depending on the specific metabolic requirements [18]. In agreement with this statement, work from our laboratory has previously shown that mitochondriai proteins purified from various tissues and analyzed by 2-dimensional NEPHGE display a certain degree of tissue specificity [24]. Mitochondriai morphology also varies; the organelles are capable of changing their shape and size [18]. Since all adult mitochondria are derived from a small and homogeneous maternal population, mitochondrial maturation must occur during ontogenesis. This is expressed by quantitative and qualitative changes within one cell type during the organogenesis. However, little is known about the physiological and development regulations of mitochondriai gene expression [2]. I n our laboratory attention has been focused on the mitochondrion within differentiating neuronal cell. This stems from the following reasons. Due to their electrophysiological activity, mature neurons are large energy consumers. Total ATP synthesis proceeds via mitochondrial respiration [4]. Furthermore, it has been suggested that, in the brain, a close relationship would exist between the development of the enzymatic acAbbreviations:mt, mitochondrial;NSE, neuron specificenolase; SYN, synaptophysin;68K HF, 68 kDa neurofdamentpolypepfide;NEPHGE, non-equilibrium pH gel electrophoresis; Ara-C, arabinosyl-cytosine; CNS, central nervous system; COX, cytochromec oxidase
tivities which are responsible for energy metabolism and the onset of neurological functions [13, 15]. It has also been shown that a correlation exists between the mode of synaptic input and the level of oxidative metabolism [15]. In a previous work, we described a neurospecific phenotype for mt-proteins from rat brain cortex, according to their electrophoretical behaviors [24]. Here, we report on a molecular study of mitochondrial maturation, both during rat brain cortex development and during the differentiation of primary cultures of neurons from fetal rat cortex.
Materials and methods Cell culture Primary cultures of neurons were established from 15th fetal day rat brain cortex, according to [3], with several modifications. Briefly, the tissue was removed from embryos under sterile conditions, mechanically dissociated and seeded (about 107 cells per well), in 60 mm diameter wells (Falcon), precoated with poly L-ornithine. Neurons were grown in D-MEM/Ham's FI2 medium (1:1) supplemented with 7.5% fetal calf serum, 2.5% horse serum, glucose (0.6%), HEPES (5 10.3 M), glutamine (0.3% w/v), penicillin (103 U/ml) and streptomycin (0.5 mg/ml). AraC was added (1/zg/ml) after 3 days of culture, then cells were fed once a week. Immunocytofluorescence Cells grown on glass coverslips were fixed in different ways depending upon the type of antibodies utilized. For immunocytofluorescent visualization of neurofilaments, cells were extracted with Triton XI00 in a microtubule stabilization buffer (100 mM PIPES, 5 mM EGTA, 2 mM MgCl2, pH 6.8) before f'Lxationin methanol 80°7o( - 20°C, 20 tnin). We used monoclonal mouse antibodies (anti-68 kDa neurofilament poly-
58
L Cordeau-Lossouarn et al
peptides, Amersham). For immunocytofluorimetric characterization of neuron-specific enolase (NSE), cells were fixed with 4% paraformaldehyde in PBS. A polyclonal rabbit antibody was used. Synaptophysin immuno-reactivity was visualized with a mouse monoclonal antibody after fixation in 100% acetone ( - 20°C, 10 rain). The immunoreactions were revealed by an indirect antibody method [25]. Controls were treated in the same way, but the first antibody was omitted.
Electron microscopy For electron microscopy, cells grown on coverslips were processed as in [21].
Mitochondria preparation Cultured neurons and brain cortex mitochondria are prepared as described in [16].
2-dimensional gel electrophoresis The procedure followed was essentially that described by O'Farrell [17]. The first dimension separations were performed as basic non-equilibrium pH gradient electrophoresis (NEPHGE) for 5 h at 500 V in 120 mm 4% polyacrylamide gels containing 2% ampholines (1.6% pH 3.5-10 LKB, 0.4% pH 5 - 8 LKB). NEPHGE electrophoresis appears to be the adequate technique to visualize on 2-D electrophoregrams the basic spots which are the major components of mt-proteins. The SDS polyacrylamide gel for the second dimension conrained 15% acrylamide and 0.2% bisacrylamide. Proteins were revealed by silver staining as in [22].
Nonactin treatment of cultured cells Neurons were labelled with radioactive methionine (with or without nonactin 2.5/~M) in D-MEM/Ham's FI2 medium, lacking methionine (J Boy) and complemented as described in "cell cultures". 35S-Methionine was used at a concentration of 60/~Ci/ml and incubation was carried out for 2 h. Prelabeled cells were then homogenized, mitochondria were purified and mt-proteins separated on 2-dimensional NEPHGE. Electrophoregrams were then autoradiographed with Kodak AX films.
Quantitative western blot analysis Crude homogenates preparation Cerebral cortices: several rats were sacrificed by decapitation in each preparation. The cortices were homogenized with isolation buffer (sucrose 0.32 M, TES 5 raM, EDTA-Na 0.5 raM, pH 7.4) in a DOUNCE (20 up and down strokes). Primary cultures of neurons: cells were rinsed and harvested in isolation buffer, then homogenized in a Thomas potter (150 up and down strokes). All steps were carried out in the presence of protease inhibitors. Gel electrophoresis and immunoblotting One-dimensional SDS-PAGE was performed on 15% acrylamide, 0.2% bisacrylamide gels. For Western blotting, proteins were transfered to nitrocellulose as described in [23]. Blots were reacted with the first antibody and immunoreactivity revealed by using biotinylated anti-IgG, followed by 3SS-streptavidin (both from Amersham). For quantitation, immunoblot autoradiograms were scanned with a Vernon photometer, peak areas measured and integrated.
Fig 1. Morphological development of cultured neurons dissociated from fetal rat brain (detailed culture conditions are described in the text). Neurons are shown after: 2 h (A); 4 days (B); 12 days (C) and 26 days (D) in vitro.
Mitochondrial maturation during neurogenesis
Primary antibodies used Anti-neurospecific proteins: mouse monoclonal antibody to synaptophysin (clone SY 38), from Boehringer Mannheim Biochemica; rabbit polyclonal antibody to NSE (a gift from Dr N Lamand~). Anti-mitochondriai proteins: antibodies to subunit VI (mouse Ig G, monoclonal) and to subunit II and III (rabbit IgG, polyclonal) of cytochrome c oxidase (gifts from Pr Kadenbach); antibodies to FcATPsynthetase (rabbit IgG, polyclonal) (gift from Pr Vignais).
Results
Characterization of primary culturesfrom rat brain cortex Figure 1 shows the morphological characteristics of cells from fetal rat cortex after either, 2 h (fig IA), 4 days (fig IB), 12 days (fig IC) and 26 days (fig ID) cultivation in the conditions described in Material and methods. The neuronal character of the cells was assessed, in the course of differentiation, by the immunoreactivity of tissue specific markers such as neurofilaments, NSE and synaptophysin (fig 2). Neurofilament (fig 2 a-d) and NSE (fig 2 o-h) antigens were distributed cytologically both in cell bodies and neurites (fig 2 a-d). Synaptophysin was detectable after three days in vitro and its distribution displayed a characteristic punctiform appearance (fig 2 i-d). Observa-
59
tion by electron microscopy of 17 and 23 days old neurons (fig 3) revealed that they were still in good condition at least from an ultrastructural view point. One can note: large Golgi areas, highly organized cytoskeleton, presence of endoplasmic reticulum and numerous mitochondria. To obtain more precise developmental correlates of the differentiation taking place in vivo and in vitro and to challenge the extent to which these two neurogenic conditions are comparable, NSE and synaptophysin were elected as neurospecific protein markers [12, 20] and their quantitative changes examined either during brain cortex maturation or during the in vitro development of rat primary neurons from the CNS. These changes were estimated by scanning Western blot autoradiograms. Results are shown in figure 4. One can see that both in rat brain and in neuron cultures, NSE shows a modest but rather regular increase with time. In the adult rat and after 21 days in culture, one observes a 4.5-fold increase relative to the reference day which correponds to the 15th day of fetal life and to 24 h after plating the cultures dishes. In contrast, synaptophysin rapidly and considerably increased (50-fold and 30-fold in brain cortex and cultures, respectively).
Changes in total mt-proteins contents during neuronal differentiation It has been reported that the K ÷ ionophore nonactin specifically inhibits the electrochemical gradient within the
Fig 2. Immunocytofluorescent characterization of cultured neurons after various times in vitro. Neurons are stained with different antibodies against: 68KNF (a: 3 days, b: 7 days, c: 15 days, d: 21 days); NSE (e: 3 days, f: 7 days, g: 15 days; h: 21 days) and SYN (i: 3 days, j: 7 days, k: 15 days, 1:21 days).
60
L Cordeau-Lossouarn et al
0
" o~'~e,,.~~ .o~.
Fig 3. Electron microscopy analysis of cultured neurons. A: day 17 ( x 13 000); B: day 17( x 6 300); C: day 23 ( x 6 000); D: day 23 ( x 3 000); E: day 23 ( x 3 000). M - mitochondria; G - Golgi centers; r -- endoplasmic reticulum; n = neuronal cytoskeleton.
Mitochondrial maturation during neurogenesis
mitochondrion, thus preventing the translocation into the matrix of the mt-proteins synthesized in the cytosol [1]. Furthermore, we have observed that in the presence of nonactin, the mt-proteim COX If/Ill encoded by mt-DNA are no more translated. These proteins, arrowheaded in figure 5C are not recovered in figure 5N. Changes in ionic ~trength within the mitochondriai matrix due to a massive concentration of K + might account for the inhibition of tn~nslation on the mr-ribosomes. As far as preparations originating from cultured neurons were concerned, the technique of Nicholls [16] was used to prepare mitochondriai fractions. In order to asses the degree of purity of these fractions, cells were incubated with or without nonactin, and the proteins were then labelled with 3SS-methionine. The autoradiograms corresponding to the NEPHGE of purified mitochondria from nonactin treated (N) or control (C) cells were analyzed (fig 5). In the presence of nonactin, radioactive mrproteins which were synthesized on the cytoplasmic polysomes were not incorporated into the pre-existing organelles, but recovered into the cytosolic fraction [6]; while mt-proteins encoded by mt-DNA were not further translated, as shown in figure 5N. Under these conditions, the labelled spots present on the autoradiogram as shown in figure 5N, correspond to the newly synthesized cytoplasmic proteins (not mitochondrial) which have co-purified with pre-existing non-labelled mtproteins. One can calculate that the amount of radioactivity found on the autoradiogram (fig 5N) from nonactin treated cells represents 5O/oof the total radioactivity associated with the mr-proteins from control cells (fig 5C). This percentage remained constant at every stage of the culture.
61
Concerning the mitochondriai fractions derived from brain cortex from rats of various ages, the technique of Nicolls [14] resulted in preparations with a high degree of purity. Yet, in order to probe for eventual subcellular con-
eo®
,t I
OwjvCavtm
t
Fig 4. SYN and NSE evolution during brain cortex maturation (A) and cultured neurons differentiation (B). Each point corresponded to the value found at day x, relatively to that found at fetal day 15 (fig A) or on the first day in vitro (fig B); each point represents the means + SEM of n -- 3 independent experiments (SEM varies between I to 5% of the value determined by scanning Western blot autoradiograms, see Material and methods), s _ s SYlq; o--o NSE. FI5 - fetal day 15; B = birth; A - adult (60 day).
Fig 5. Autoradiograms from electrophoregrams of mitochondria purified from cultured neurons labelled in the absence or presence of nonactin. Neurons were labelled with 35S-methionine in absence (C) or in the presence (N) of mitochondrial inhibitor (Nonactin). Mitochondria were purified and mt-proteins separated on 2-dimensional NEPHGE. Spots arrowed on autoradiogram (C) corresponded to proteins encoded by mt-DNA, identified as in [5] (cox ll/llI = subunits II and I11 of cytochrome c oxidase). These proteins, not synthesized in the presence of nonactin, were not recovered on autoradiogram (I'4) (circles). Remaining spots on autoradiogram (N) corresponded to non-mitochondrial proteins, two of them were arrowed on (N): aTUB = • isotubulins, A = actin.
62
L Cordeau-Lossouarn et al
taminations, various specific marker enzymes were measured. They were found negative (data not shown). Moreover, we could not detect any of the contaminating proteins shown in figure 5N for rot-preparation of cultured neurons on electrophoregrams of mt-proteins issued from neuronal cortex (fig 6). The protein content of pellets containing mitochondrial fractions were measured and estimated as a percentage of the total protein content from corresponding crude extracts (cerebral cortices and neuronal cultures homogeneized at various differentiation stages: fig 7). We have compared the kinetics obtained in vivo (fig 7A) and in vitro (fig 7B), assuming that day 0 of the cultures corresponds to the 15th day of fetal life for the rat cortex. They exhibited a similar pattern, especially when comparing the evolution between day 0 and day 13 in vitro on one hand, and between the 15th day of fetal life and the 8th day of neonatal life on the other hand. Quantitatively, we observed an 8 and 10-fold increase to mitochondrial masses for the brain cortex and neuron cultures respectively, at the end of the experiments.
PM
OH 8,o
-'~--
Mitochondrial proteins changes in brain cortex and neuronal cultures
In an attempt to elucidate what accurately reflected these important increases in the mitochondrial masses, we analysed the changes occurring in some mitochondrial enzymes involved in the energy metabolism. To this purpose, we selected the ~ subunit of F~ATPase, the nuclear DNA encoded subunit VI and the mt-DNA encoded II/III subunits of the cytochrome oxidase. Equal quantities of proteins from crude extracts (prepared from cerebral cortex from rats of different ages or from cultivated neurons at different stages of differentiation) were applied on slab-gels. The proteins were then transferred onto nitrocellulose membranes, processed for radioimmunological reaction and autoradiographied. Autoradiograms were scanned and antigens indirectly quantified (scanning value at one day taken as a percentage of that found at reference day: figure 8). As shown in figure 8B, striking differences were observed in the relative increases between the different mt-
4,5_.~H+
pH i
56K-
36K-
2OK-
6,5K-
Fig 6. Two-dimensional electrophoregram of mitochondria purified from adult rat brain. Two mitochondrial proteins are arrowed: FI - ~ subunit of F I component of F0F, ATP synthetase; Cox II/III = subunits II and III of cytochrome c oxidase; E~,, E,/3, E 2, E3: PDH complex subunits.
Mitochondrial maturation during neurogenesis proteins during in vitro neurogenesis. The changes in every mr-protein display a wide range of specificity. The changes found in vitro are partially recovered during in vivo differentiation (fig 8A): the cell content of ~ subunit of FiATPase displayed a 5-fold increase at the 20th day of post-natal life and after 25 days in vitro; - the II/III subunits of the COX were increased 15-fold both at the 20th day of post-natal life and after 25 days in vitro; - in vivo and in vitro kinetics of COX VI revealed different patterns. The cell content of COX VI showed a 20-fold increase in vivo and 15-fold in vitro.
Discussion
Changes in mitochondrial mass during neurogenesis
During brain maturation, or in a population of primary neurons maintained in conditions that permit their in vitro differentiation, the mt-mass, roughly equivalent to the ratio of mr-proteins versus total proteins, increases considerably by at least an order of magnitude. This increase grossly obeys a biphasic kinetics in vitro (fig 7B) and possibly also in vivo (fig 7A), with two phases of accumulation separated by a plateau. Interestingly, the first accumulation (from day 0 to day 2 in vitro and from the 15th to the 20th day of fetal life) and the plateau (between day 3 and day 13 in vitro and between birth and the 8th day of post-natal life) evolve similarly in both systems, whereas, the second phase (from day 13 to day 29 in vitro and from day 8 to day 60 in rico) is significantly more rapid in the cultures. A possible explanation can be found in the fact that in brain, the second phase coincides with the proliferation of the glial cells [9, 19]. These cells contain fewer mitochondria than neurons [26] and thus their multiplication would mask any specific neuronal increase of the mitochondriai mass. One must note, however that our neuron cultures contain about 5 to 10e/o of astrocytes, necessary for long-term survival of neurons [21], but that under our experimental conditions, mitosis of these cells is inhibited with Ara-C. With these considerations in mind, it appears likely that quantitative changes in the mitochondrial protein content of the neuronal cell are very similar in the developing cultures and during brain maturation. Changes in mitochondrial protein components during neurogenesis
The kinetics of increase in individual mt-proteins such as the ~ subunit of the F, component of the F0F I ATP synthetase and some of the subunits of cytochrome oxidase were not identical. In other words, not only did each protein fail to follow a modulation which was superimposable over the kinetics of the total mitochondrial mass but striking differences were observed in the relative increases between different mr-proteins, regardless of their genetic origin (mt-DNA or nuclear-DNA encoded). For example, we could not that both in vivo and in vitro, the kinetics of accumulation of the ~ subunit of the F l component reached a plateau while the amount of COX subunits was still increasing. In rico, one could observe that COX Vl increase far more than is the case in vitro. An explanation would possibly lie in the fact already reported [9], that new COX Vl isoforms, detected by our antibody, will appear dur-
63
ing brain maturation. These isoforms would not be detected during differentiation of cultured neurons. The lack of coordination observed during the synthesis of various mt-proteins synthesis may suggest that the mitochondrial protein equipment would vary in the course of neurogenesis, in response to various developmental or physiological requirements of the cell. For example, in [26], it was suggested that COX synthesis is increased in correlation with the development of neuronal activity. The significance of these large increase of mt-masses neurogenesis will have to be re-evaluated after having specified whether these changes are correlated with accompanying modifications in the size or in the number of the mitochondria (or in both). Ultrastructural analyses are presently underway. Similarly, investigations concerning the metabolic activities of mitochondria, sampled at different phases of neuronal development, could be of interest in view of the unbalanced synthesis observed in the various mt-proteins.
"®
f
2,5 -
~0 t l,S
i,° i
Fig 7. Evolution of mt-protein content. Percentage of mt-proteins versus total proteins were measured during brain cortex maturation (A) and cultured neurons differentiation 01D. Values represent the means + SEM of n = 5 independent experiments (SEM varies between 1 to 5% of the values). FI5 = fetal day 15; B = birth; A = adult (60 days).
64
L Cordeau-Lossouarn et al
Mitochondrial maturation and neuronal differentiation
A second point deserving comments concerns the evolution of the mitochondrial mass, in relation to the successive steps of neurogenesis, in vitro and in vivo. During the course of neuronal differentiation, in vitro, and under the conditions utilized in this work, it can be noted that the first phase of accumulation (fig 7B) corresponded to the formation of growth cones and to the appearance of processes, between the day of plating and the 3rd or 4th day in vitro (fig 1A, B). Then the plateau phase (from day 3 to day 13, fig 7B) temporally coincided with the establishment of the neuronal network (from day 4 to day 12, fig IB, C). In the second phase of accumulation (namely between day 10 and day 26 after seeding, fig 7B), dramatic morphological changes do occur (fig 1C, D), cell bodies collect into spherical aggregates interconnected with thick bundles of neuronal processes. During this phase, the neurons become electrophysiologically competent, as revealed by post-synaptic activity measurement with a patch-clamp technique (manuscript in preparation).
®
In brain cortex, the kinetics of accumulation of the mitochondrial mass can be correlated with successive steps of brain maturation which have been described elsewhere [8]. The first phase of accumulation and the plateau phase (from foetal day 15 to post-natal day 7, fig 6A) span over a period coinciding with migration of neurons, sprouting of neurites and formation of transient connections. The second phase of rot-mass accumulation could correspond to the final maturation of synapses in the cortex (from post-natal day 18 to adult ie 60 days). Although the present work still raises various questions, particularly with regard to the cytological and biological status of mitochondria during the successive bursts of mitochondrial masses accompanying neurogenesis, Jt is clear that the mitochondrion is the site of very pronounced changes in the course of this phenomenon, both quantitative and probably also qualitative. In particular, whether the physiological significance of the changes in mitochondrion observed during neuronal differentiation is uniquely related to different demands in the bioenergetics of the cell or not, remains to be elucidated. In this line of thought, recent models [7] ascribe a rather specific role of mrproteins and of mitochondria themselves in the dimerization of = and ~ tubulin subunits as well as in microtubule assembly, raising the challenging possibility that the mitochondrion might be directly involved in the formation of neurotubules and thus in process formation. Since the evolution of the m t mass in primary cultures of neurons from the rat CNS largely resembles what is observed in vivo, during brain maturation, and given their post-mitotic nature, these cultured neurons might constitute an interesting model to analyse the regulatory mechanisms involved in mitochondriogenesis, and the correlation between the latter and cell differentiation.
Acknowledgments
I
F~
i
D~
~ /
i
In vim •
This work was supported by a grant from Association Franqalse contre les Myopathies. We thank M Basseville for her excellent technical assistance, and Dr H O'Hayon for electron microscopy analysis.
ill
.® References
'
I
!
s
.,
I
i
I
lo
15
20
..
I
H
Daysi inii vnro
Fig 8. Changes in specific mt-proteins. A: in rico, B: in ~,itro. Each point corresponded to the value found at day x, relative to that found at fetal day 15 (.4,),or on the first day in vitro (B); each point represents the means :I: SEM of n = 3 independent experiments (SEM varies between 1 to 5% of the value determined by scanning Western blot autoradiograms, see Materials and methods), o--o Cox II/III; A--A Cox Vl; D--O ~FI; FI5 = fetal day 15; B = birth; A = adult (60 days).
1 Anderson L (1981) Identification of mitochondrial proteins and some of their precursors in two-dimensional electrophoretic maps of human cells. Proc Natl Acad Sci USA 78, 2407-2411 2 Attardi G, Schatz G (1988) Biogenesis of mitochondria. Ann Rev Cell Biol 4, 289-333 3 Berger B, Di Porzio U, Dnguet MC, Gay M, Vigny A, Glowinski J, Prochiantz A (1982) Long-term development of mesencephalic dopaminergic neurons of mouse embryos in dissociated primary cultures: morphological and histochemical characteristics. Neurosci 7, 193-205 4 Dngani F, Marzatico F, Curti D (1988)Oxidative metabolism of non-synaptic mitochondria isolated from rat brain hippocampus: a comparative regional study. JNeurochem 50, 1233-1236 5 Darley-Usmar VM, Rickwood D, Wilson MT (1987) Mitochondria: a practical approach. IRL Press, Oxford, Washington DC 6 Gupta RS, Austin RC (1987) Mitochondrial matrix localization of a protein altered in mutants resistant to the microtubule inhibitor podophyllotoxin. Eur J Cell Bio145, 170-176
Mitochondrial maturation during neurogenes~;s 7 Gupta RS (1990) Mitochondria, molecular chaperone proteins and the in vivo assembly of microtubules. Trends Biochem $ci 15, 415-418 8 Hay R, Bohni P, Gasser S (1984) How mitochondria import proteins. BBA 779, 65-87 9 Hattori T, Hamada S, Masaki T, Kushima R, Ohmura M, Kataoka Y, Ikehara Y, Kita M, Magami Y (1989) Turnover of glial cells in the central nervous system. Biomed Res 10, 355-363 10 Jacobson M (1978) Developmental Neurobiology. Plenum Press, NY/London 11 Kadenbach B, Khu-Netwig L, Buge U (1987) Evolution of a regulatory enzyme cytochrome c oxidase (complexe IV) Curt Topics Bioener 15, 113-161 12 Kato K, Suzuki F, Watanahe T, Semba R, Keino H (1984) Developmental profile of three enolase isozymes in rat brain determination from one-cell embryo to adult brain. Neurochem Int 6, 51-54 13 Leong SF, Lai JCK, Lim L, Clark JB (1981) Ener.gymetabollsing enzymes in brain regions of adult and aging rat. Y Neurochem 37, 1548-1556 14 Malloch GDA, Munday LA, Olson MS, Clark JB (1986) Comparative development of the pyruvate dehydrogenase complex and citrate syuthase in rat brain mitochondria. Biochem Y 238, 729-736 15 Mjaatvedt AlE, Wong-Riley MTT (1988) Relationship between synaptogenesis and cytochrome oxidase activity in purkinje cells of the developing rat cerebellum. J Comp Neurol 277, 155-182 16 Nicholls MG (1978) Calcium transport and proton electrochemical gradient in mitochondria from guinea-pig cerebral cortex and rat heart. Biochem J 170, 511-522
65
17 O'Farell PZ, Goodman HM, O'Farell PH (1977) High resolution two-dimensional electrophoresis of basic as well as acidic proteins. Cell 12, 1133-1142 18 Pollak JK, Sutton R (1980) The differentiation of animal mitochondria during development. Trends Biochem $ci 5, 23-27 19 Reichenbac A (1989) Gila: Neuron index: Review and Hypothesis to account for different values in various mammals. Glia 2, 71-77 20 Schilling K, Scherbaum C, Pilgrim C (1988) Developmenhal changes of neuron-specific enolase and neurofilament proteins in primary neural culture. Histochem 89, 295-299 21 Sebben M, Gabrion J, Manzoni O, Sladeczek F, Grii C, Bockaert J, Dumuis A (1990) Estabfishment of a long-term primary culture of striatal neurons. Dev Brain Res 52, 229-239 22 Thompson S (1987) Simpfifiedand reproducible silver staining of proteins in polyacrylamide gels. Med $ci Res 15, 1253-1255 23 Towbin H, Staehelin T, Gordon J (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some appfications. Proc Natl Acad $ci USA 76, 4350-4354 24 Vayssi6re JL, Cordeau-Lossouarn L, Larcher JC, Gros F, Croizat B (1989) Tissue-specific mitochondrial proteins. Biochimie 71,787-791 25 Vielkind U, Swierenga SH (1989) A simple fixation procedure for immunofluorescent detection of different o/toskeletal components within the same cell. Histochem 91, 81-88 26 Wong~Riley MTT (1989) Cytochrome oxidase: an endogenous metabofic marker for neuronal activity. TINS 12, 94-101
57 Original article
Mitochondrial maturation during neuronal differentiation in vivo and in vitro Laurence Cordeau-Lossouarn, Jean-Luc Vayssi~re, Jean-Christophe Franfois Gros, Bernard Croizat
Larcher,
Laboratoire de Biochimie Ceilulaire, URA 1115 CNRS, Coll~ge de France, 11 place Marcelin-Berthelot, 75231 Paris Cedex 05, France (Received 31 October 1990; accepted 11 February 1991) Summary - The evolution of the mitochondrion has been followed within differentiating neuronal cells, both in primary cultures of neurons from fetal rat cortex and during rat brain cortex maturation. Changes in total mitochondrial proteins (mt-proteins) were evaluated, and qualitative changes in the mt-proteins pattern were analyzed using the Western blot technique. The evolution of mtprotein contents in cultured neurons resembles what is observed during rat brain maturation. The mitochondrion exhibits pronounced changes in the course of neurogenesis, in particular, bursts of mitochondrial masses accompanying the successive steps of neurogenesis are observed. There are indications that protein equipment of mitochondria during neuronal development undergoes variations. Although more work is required to establish the significance of these correlations, the present data might suggest an important role of the mitochondrion in neurogenesis. primary culture of neurons I neurogenesisI quantitative Western blot I miteehondrial maturation
Introduction Mitochondria are near ubiquitous organelles of eukaryotic cells. While their principal function is oxidative phosphorylation, they also contribute to the biosynthesis of pyrimidines, aminoacids, phospholipids, nucleotides, folate coenzymes, heme, urea and many other metabolites [8]. It is thus not surprising that mitochondria found in various animal tissues show considerable differences in physiological activity, depending on the specific metabolic requirements [18]. In agreement with this statement, work from our laboratory has previously shown that mitochondriai proteins purified from various tissues and analyzed by 2-dimensional NEPHGE display a certain degree of tissue specificity [24]. Mitochondriai morphology also varies; the organelles are capable of changing their shape and size [18]. Since all adult mitochondria are derived from a small and homogeneous maternal population, mitochondrial maturation must occur during ontogenesis. This is expressed by quantitative and qualitative changes within one cell type during the organogenesis. However, little is known about the physiological and development regulations of mitochondriai gene expression [2]. I n our laboratory attention has been focused on the mitochondrion within differentiating neuronal cell. This stems from the following reasons. Due to their electrophysiological activity, mature neurons are large energy consumers. Total ATP synthesis proceeds via mitochondrial respiration [4]. Furthermore, it has been suggested that, in the brain, a close relationship would exist between the development of the enzymatic acAbbreviations:mt, mitochondrial;NSE, neuron specificenolase; SYN, synaptophysin;68K HF, 68 kDa neurofdamentpolypepfide;NEPHGE, non-equilibrium pH gel electrophoresis; Ara-C, arabinosyl-cytosine; CNS, central nervous system; COX, cytochromec oxidase
tivities which are responsible for energy metabolism and the onset of neurological functions [13, 15]. It has also been shown that a correlation exists between the mode of synaptic input and the level of oxidative metabolism [15]. In a previous work, we described a neurospecific phenotype for mt-proteins from rat brain cortex, according to their electrophoretical behaviors [24]. Here, we report on a molecular study of mitochondrial maturation, both during rat brain cortex development and during the differentiation of primary cultures of neurons from fetal rat cortex.
Materials and methods Cell culture Primary cultures of neurons were established from 15th fetal day rat brain cortex, according to [3], with several modifications. Briefly, the tissue was removed from embryos under sterile conditions, mechanically dissociated and seeded (about 107 cells per well), in 60 mm diameter wells (Falcon), precoated with poly L-ornithine. Neurons were grown in D-MEM/Ham's FI2 medium (1:1) supplemented with 7.5% fetal calf serum, 2.5% horse serum, glucose (0.6%), HEPES (5 10.3 M), glutamine (0.3% w/v), penicillin (103 U/ml) and streptomycin (0.5 mg/ml). AraC was added (1/zg/ml) after 3 days of culture, then cells were fed once a week. Immunocytofluorescence Cells grown on glass coverslips were fixed in different ways depending upon the type of antibodies utilized. For immunocytofluorescent visualization of neurofilaments, cells were extracted with Triton XI00 in a microtubule stabilization buffer (100 mM PIPES, 5 mM EGTA, 2 mM MgCl2, pH 6.8) before f'Lxationin methanol 80°7o( - 20°C, 20 tnin). We used monoclonal mouse antibodies (anti-68 kDa neurofilament poly-
58
L Cordeau-Lossouarn et al
peptides, Amersham). For immunocytofluorimetric characterization of neuron-specific enolase (NSE), cells were fixed with 4% paraformaldehyde in PBS. A polyclonal rabbit antibody was used. Synaptophysin immuno-reactivity was visualized with a mouse monoclonal antibody after fixation in 100% acetone ( - 20°C, 10 rain). The immunoreactions were revealed by an indirect antibody method [25]. Controls were treated in the same way, but the first antibody was omitted.
Electron microscopy For electron microscopy, cells grown on coverslips were processed as in [21].
Mitochondria preparation Cultured neurons and brain cortex mitochondria are prepared as described in [16].
2-dimensional gel electrophoresis The procedure followed was essentially that described by O'Farrell [17]. The first dimension separations were performed as basic non-equilibrium pH gradient electrophoresis (NEPHGE) for 5 h at 500 V in 120 mm 4% polyacrylamide gels containing 2% ampholines (1.6% pH 3.5-10 LKB, 0.4% pH 5 - 8 LKB). NEPHGE electrophoresis appears to be the adequate technique to visualize on 2-D electrophoregrams the basic spots which are the major components of mt-proteins. The SDS polyacrylamide gel for the second dimension conrained 15% acrylamide and 0.2% bisacrylamide. Proteins were revealed by silver staining as in [22].
Nonactin treatment of cultured cells Neurons were labelled with radioactive methionine (with or without nonactin 2.5/~M) in D-MEM/Ham's FI2 medium, lacking methionine (J Boy) and complemented as described in "cell cultures". 35S-Methionine was used at a concentration of 60/~Ci/ml and incubation was carried out for 2 h. Prelabeled cells were then homogenized, mitochondria were purified and mt-proteins separated on 2-dimensional NEPHGE. Electrophoregrams were then autoradiographed with Kodak AX films.
Quantitative western blot analysis Crude homogenates preparation Cerebral cortices: several rats were sacrificed by decapitation in each preparation. The cortices were homogenized with isolation buffer (sucrose 0.32 M, TES 5 raM, EDTA-Na 0.5 raM, pH 7.4) in a DOUNCE (20 up and down strokes). Primary cultures of neurons: cells were rinsed and harvested in isolation buffer, then homogenized in a Thomas potter (150 up and down strokes). All steps were carried out in the presence of protease inhibitors. Gel electrophoresis and immunoblotting One-dimensional SDS-PAGE was performed on 15% acrylamide, 0.2% bisacrylamide gels. For Western blotting, proteins were transfered to nitrocellulose as described in [23]. Blots were reacted with the first antibody and immunoreactivity revealed by using biotinylated anti-IgG, followed by 3SS-streptavidin (both from Amersham). For quantitation, immunoblot autoradiograms were scanned with a Vernon photometer, peak areas measured and integrated.
Fig 1. Morphological development of cultured neurons dissociated from fetal rat brain (detailed culture conditions are described in the text). Neurons are shown after: 2 h (A); 4 days (B); 12 days (C) and 26 days (D) in vitro.
Mitochondrial maturation during neurogenesis
Primary antibodies used Anti-neurospecific proteins: mouse monoclonal antibody to synaptophysin (clone SY 38), from Boehringer Mannheim Biochemica; rabbit polyclonal antibody to NSE (a gift from Dr N Lamand~). Anti-mitochondriai proteins: antibodies to subunit VI (mouse Ig G, monoclonal) and to subunit II and III (rabbit IgG, polyclonal) of cytochrome c oxidase (gifts from Pr Kadenbach); antibodies to FcATPsynthetase (rabbit IgG, polyclonal) (gift from Pr Vignais).
Results
Characterization of primary culturesfrom rat brain cortex Figure 1 shows the morphological characteristics of cells from fetal rat cortex after either, 2 h (fig IA), 4 days (fig IB), 12 days (fig IC) and 26 days (fig ID) cultivation in the conditions described in Material and methods. The neuronal character of the cells was assessed, in the course of differentiation, by the immunoreactivity of tissue specific markers such as neurofilaments, NSE and synaptophysin (fig 2). Neurofilament (fig 2 a-d) and NSE (fig 2 o-h) antigens were distributed cytologically both in cell bodies and neurites (fig 2 a-d). Synaptophysin was detectable after three days in vitro and its distribution displayed a characteristic punctiform appearance (fig 2 i-d). Observa-
59
tion by electron microscopy of 17 and 23 days old neurons (fig 3) revealed that they were still in good condition at least from an ultrastructural view point. One can note: large Golgi areas, highly organized cytoskeleton, presence of endoplasmic reticulum and numerous mitochondria. To obtain more precise developmental correlates of the differentiation taking place in vivo and in vitro and to challenge the extent to which these two neurogenic conditions are comparable, NSE and synaptophysin were elected as neurospecific protein markers [12, 20] and their quantitative changes examined either during brain cortex maturation or during the in vitro development of rat primary neurons from the CNS. These changes were estimated by scanning Western blot autoradiograms. Results are shown in figure 4. One can see that both in rat brain and in neuron cultures, NSE shows a modest but rather regular increase with time. In the adult rat and after 21 days in culture, one observes a 4.5-fold increase relative to the reference day which correponds to the 15th day of fetal life and to 24 h after plating the cultures dishes. In contrast, synaptophysin rapidly and considerably increased (50-fold and 30-fold in brain cortex and cultures, respectively).
Changes in total mt-proteins contents during neuronal differentiation It has been reported that the K ÷ ionophore nonactin specifically inhibits the electrochemical gradient within the
Fig 2. Immunocytofluorescent characterization of cultured neurons after various times in vitro. Neurons are stained with different antibodies against: 68KNF (a: 3 days, b: 7 days, c: 15 days, d: 21 days); NSE (e: 3 days, f: 7 days, g: 15 days; h: 21 days) and SYN (i: 3 days, j: 7 days, k: 15 days, 1:21 days).
60
L Cordeau-Lossouarn et al
0
" o~'~e,,.~~ .o~.
Fig 3. Electron microscopy analysis of cultured neurons. A: day 17 ( x 13 000); B: day 17( x 6 300); C: day 23 ( x 6 000); D: day 23 ( x 3 000); E: day 23 ( x 3 000). M - mitochondria; G - Golgi centers; r -- endoplasmic reticulum; n = neuronal cytoskeleton.
Mitochondrial maturation during neurogenesis
mitochondrion, thus preventing the translocation into the matrix of the mt-proteins synthesized in the cytosol [1]. Furthermore, we have observed that in the presence of nonactin, the mt-proteim COX If/Ill encoded by mt-DNA are no more translated. These proteins, arrowheaded in figure 5C are not recovered in figure 5N. Changes in ionic ~trength within the mitochondriai matrix due to a massive concentration of K + might account for the inhibition of tn~nslation on the mr-ribosomes. As far as preparations originating from cultured neurons were concerned, the technique of Nicholls [16] was used to prepare mitochondriai fractions. In order to asses the degree of purity of these fractions, cells were incubated with or without nonactin, and the proteins were then labelled with 3SS-methionine. The autoradiograms corresponding to the NEPHGE of purified mitochondria from nonactin treated (N) or control (C) cells were analyzed (fig 5). In the presence of nonactin, radioactive mrproteins which were synthesized on the cytoplasmic polysomes were not incorporated into the pre-existing organelles, but recovered into the cytosolic fraction [6]; while mt-proteins encoded by mt-DNA were not further translated, as shown in figure 5N. Under these conditions, the labelled spots present on the autoradiogram as shown in figure 5N, correspond to the newly synthesized cytoplasmic proteins (not mitochondrial) which have co-purified with pre-existing non-labelled mtproteins. One can calculate that the amount of radioactivity found on the autoradiogram (fig 5N) from nonactin treated cells represents 5O/oof the total radioactivity associated with the mr-proteins from control cells (fig 5C). This percentage remained constant at every stage of the culture.
61
Concerning the mitochondriai fractions derived from brain cortex from rats of various ages, the technique of Nicolls [14] resulted in preparations with a high degree of purity. Yet, in order to probe for eventual subcellular con-
eo®
,t I
OwjvCavtm
t
Fig 4. SYN and NSE evolution during brain cortex maturation (A) and cultured neurons differentiation (B). Each point corresponded to the value found at day x, relatively to that found at fetal day 15 (fig A) or on the first day in vitro (fig B); each point represents the means + SEM of n -- 3 independent experiments (SEM varies between I to 5% of the value determined by scanning Western blot autoradiograms, see Material and methods), s _ s SYlq; o--o NSE. FI5 - fetal day 15; B = birth; A - adult (60 day).
Fig 5. Autoradiograms from electrophoregrams of mitochondria purified from cultured neurons labelled in the absence or presence of nonactin. Neurons were labelled with 35S-methionine in absence (C) or in the presence (N) of mitochondrial inhibitor (Nonactin). Mitochondria were purified and mt-proteins separated on 2-dimensional NEPHGE. Spots arrowed on autoradiogram (C) corresponded to proteins encoded by mt-DNA, identified as in [5] (cox ll/llI = subunits II and I11 of cytochrome c oxidase). These proteins, not synthesized in the presence of nonactin, were not recovered on autoradiogram (I'4) (circles). Remaining spots on autoradiogram (N) corresponded to non-mitochondrial proteins, two of them were arrowed on (N): aTUB = • isotubulins, A = actin.
62
L Cordeau-Lossouarn et al
taminations, various specific marker enzymes were measured. They were found negative (data not shown). Moreover, we could not detect any of the contaminating proteins shown in figure 5N for rot-preparation of cultured neurons on electrophoregrams of mt-proteins issued from neuronal cortex (fig 6). The protein content of pellets containing mitochondrial fractions were measured and estimated as a percentage of the total protein content from corresponding crude extracts (cerebral cortices and neuronal cultures homogeneized at various differentiation stages: fig 7). We have compared the kinetics obtained in vivo (fig 7A) and in vitro (fig 7B), assuming that day 0 of the cultures corresponds to the 15th day of fetal life for the rat cortex. They exhibited a similar pattern, especially when comparing the evolution between day 0 and day 13 in vitro on one hand, and between the 15th day of fetal life and the 8th day of neonatal life on the other hand. Quantitatively, we observed an 8 and 10-fold increase to mitochondrial masses for the brain cortex and neuron cultures respectively, at the end of the experiments.
PM
OH 8,o
-'~--
Mitochondrial proteins changes in brain cortex and neuronal cultures
In an attempt to elucidate what accurately reflected these important increases in the mitochondrial masses, we analysed the changes occurring in some mitochondrial enzymes involved in the energy metabolism. To this purpose, we selected the ~ subunit of F~ATPase, the nuclear DNA encoded subunit VI and the mt-DNA encoded II/III subunits of the cytochrome oxidase. Equal quantities of proteins from crude extracts (prepared from cerebral cortex from rats of different ages or from cultivated neurons at different stages of differentiation) were applied on slab-gels. The proteins were then transferred onto nitrocellulose membranes, processed for radioimmunological reaction and autoradiographied. Autoradiograms were scanned and antigens indirectly quantified (scanning value at one day taken as a percentage of that found at reference day: figure 8). As shown in figure 8B, striking differences were observed in the relative increases between the different mt-
4,5_.~H+
pH i
56K-
36K-
2OK-
6,5K-
Fig 6. Two-dimensional electrophoregram of mitochondria purified from adult rat brain. Two mitochondrial proteins are arrowed: FI - ~ subunit of F I component of F0F, ATP synthetase; Cox II/III = subunits II and III of cytochrome c oxidase; E~,, E,/3, E 2, E3: PDH complex subunits.
Mitochondrial maturation during neurogenesis proteins during in vitro neurogenesis. The changes in every mr-protein display a wide range of specificity. The changes found in vitro are partially recovered during in vivo differentiation (fig 8A): the cell content of ~ subunit of FiATPase displayed a 5-fold increase at the 20th day of post-natal life and after 25 days in vitro; - the II/III subunits of the COX were increased 15-fold both at the 20th day of post-natal life and after 25 days in vitro; - in vivo and in vitro kinetics of COX VI revealed different patterns. The cell content of COX VI showed a 20-fold increase in vivo and 15-fold in vitro.
Discussion
Changes in mitochondrial mass during neurogenesis
During brain maturation, or in a population of primary neurons maintained in conditions that permit their in vitro differentiation, the mt-mass, roughly equivalent to the ratio of mr-proteins versus total proteins, increases considerably by at least an order of magnitude. This increase grossly obeys a biphasic kinetics in vitro (fig 7B) and possibly also in vivo (fig 7A), with two phases of accumulation separated by a plateau. Interestingly, the first accumulation (from day 0 to day 2 in vitro and from the 15th to the 20th day of fetal life) and the plateau (between day 3 and day 13 in vitro and between birth and the 8th day of post-natal life) evolve similarly in both systems, whereas, the second phase (from day 13 to day 29 in vitro and from day 8 to day 60 in rico) is significantly more rapid in the cultures. A possible explanation can be found in the fact that in brain, the second phase coincides with the proliferation of the glial cells [9, 19]. These cells contain fewer mitochondria than neurons [26] and thus their multiplication would mask any specific neuronal increase of the mitochondriai mass. One must note, however that our neuron cultures contain about 5 to 10e/o of astrocytes, necessary for long-term survival of neurons [21], but that under our experimental conditions, mitosis of these cells is inhibited with Ara-C. With these considerations in mind, it appears likely that quantitative changes in the mitochondrial protein content of the neuronal cell are very similar in the developing cultures and during brain maturation. Changes in mitochondrial protein components during neurogenesis
The kinetics of increase in individual mt-proteins such as the ~ subunit of the F, component of the F0F I ATP synthetase and some of the subunits of cytochrome oxidase were not identical. In other words, not only did each protein fail to follow a modulation which was superimposable over the kinetics of the total mitochondrial mass but striking differences were observed in the relative increases between different mr-proteins, regardless of their genetic origin (mt-DNA or nuclear-DNA encoded). For example, we could not that both in vivo and in vitro, the kinetics of accumulation of the ~ subunit of the F l component reached a plateau while the amount of COX subunits was still increasing. In rico, one could observe that COX Vl increase far more than is the case in vitro. An explanation would possibly lie in the fact already reported [9], that new COX Vl isoforms, detected by our antibody, will appear dur-
63
ing brain maturation. These isoforms would not be detected during differentiation of cultured neurons. The lack of coordination observed during the synthesis of various mt-proteins synthesis may suggest that the mitochondrial protein equipment would vary in the course of neurogenesis, in response to various developmental or physiological requirements of the cell. For example, in [26], it was suggested that COX synthesis is increased in correlation with the development of neuronal activity. The significance of these large increase of mt-masses neurogenesis will have to be re-evaluated after having specified whether these changes are correlated with accompanying modifications in the size or in the number of the mitochondria (or in both). Ultrastructural analyses are presently underway. Similarly, investigations concerning the metabolic activities of mitochondria, sampled at different phases of neuronal development, could be of interest in view of the unbalanced synthesis observed in the various mt-proteins.
"®
f
2,5 -
~0 t l,S
i,° i
Fig 7. Evolution of mt-protein content. Percentage of mt-proteins versus total proteins were measured during brain cortex maturation (A) and cultured neurons differentiation 01D. Values represent the means + SEM of n = 5 independent experiments (SEM varies between 1 to 5% of the values). FI5 = fetal day 15; B = birth; A = adult (60 days).
64
L Cordeau-Lossouarn et al
Mitochondrial maturation and neuronal differentiation
A second point deserving comments concerns the evolution of the mitochondrial mass, in relation to the successive steps of neurogenesis, in vitro and in vivo. During the course of neuronal differentiation, in vitro, and under the conditions utilized in this work, it can be noted that the first phase of accumulation (fig 7B) corresponded to the formation of growth cones and to the appearance of processes, between the day of plating and the 3rd or 4th day in vitro (fig 1A, B). Then the plateau phase (from day 3 to day 13, fig 7B) temporally coincided with the establishment of the neuronal network (from day 4 to day 12, fig IB, C). In the second phase of accumulation (namely between day 10 and day 26 after seeding, fig 7B), dramatic morphological changes do occur (fig 1C, D), cell bodies collect into spherical aggregates interconnected with thick bundles of neuronal processes. During this phase, the neurons become electrophysiologically competent, as revealed by post-synaptic activity measurement with a patch-clamp technique (manuscript in preparation).
®
In brain cortex, the kinetics of accumulation of the mitochondrial mass can be correlated with successive steps of brain maturation which have been described elsewhere [8]. The first phase of accumulation and the plateau phase (from foetal day 15 to post-natal day 7, fig 6A) span over a period coinciding with migration of neurons, sprouting of neurites and formation of transient connections. The second phase of rot-mass accumulation could correspond to the final maturation of synapses in the cortex (from post-natal day 18 to adult ie 60 days). Although the present work still raises various questions, particularly with regard to the cytological and biological status of mitochondria during the successive bursts of mitochondrial masses accompanying neurogenesis, Jt is clear that the mitochondrion is the site of very pronounced changes in the course of this phenomenon, both quantitative and probably also qualitative. In particular, whether the physiological significance of the changes in mitochondrion observed during neuronal differentiation is uniquely related to different demands in the bioenergetics of the cell or not, remains to be elucidated. In this line of thought, recent models [7] ascribe a rather specific role of mrproteins and of mitochondria themselves in the dimerization of = and ~ tubulin subunits as well as in microtubule assembly, raising the challenging possibility that the mitochondrion might be directly involved in the formation of neurotubules and thus in process formation. Since the evolution of the m t mass in primary cultures of neurons from the rat CNS largely resembles what is observed in vivo, during brain maturation, and given their post-mitotic nature, these cultured neurons might constitute an interesting model to analyse the regulatory mechanisms involved in mitochondriogenesis, and the correlation between the latter and cell differentiation.
Acknowledgments
I
F~
i
D~
~ /
i
In vim •
This work was supported by a grant from Association Franqalse contre les Myopathies. We thank M Basseville for her excellent technical assistance, and Dr H O'Hayon for electron microscopy analysis.
ill
.® References
'
I
!
s
.,
I
i
I
lo
15
20
..
I
H
Daysi inii vnro
Fig 8. Changes in specific mt-proteins. A: in rico, B: in ~,itro. Each point corresponded to the value found at day x, relative to that found at fetal day 15 (.4,),or on the first day in vitro (B); each point represents the means :I: SEM of n = 3 independent experiments (SEM varies between 1 to 5% of the value determined by scanning Western blot autoradiograms, see Materials and methods), o--o Cox II/III; A--A Cox Vl; D--O ~FI; FI5 = fetal day 15; B = birth; A = adult (60 days).
1 Anderson L (1981) Identification of mitochondrial proteins and some of their precursors in two-dimensional electrophoretic maps of human cells. Proc Natl Acad Sci USA 78, 2407-2411 2 Attardi G, Schatz G (1988) Biogenesis of mitochondria. Ann Rev Cell Biol 4, 289-333 3 Berger B, Di Porzio U, Dnguet MC, Gay M, Vigny A, Glowinski J, Prochiantz A (1982) Long-term development of mesencephalic dopaminergic neurons of mouse embryos in dissociated primary cultures: morphological and histochemical characteristics. Neurosci 7, 193-205 4 Dngani F, Marzatico F, Curti D (1988)Oxidative metabolism of non-synaptic mitochondria isolated from rat brain hippocampus: a comparative regional study. JNeurochem 50, 1233-1236 5 Darley-Usmar VM, Rickwood D, Wilson MT (1987) Mitochondria: a practical approach. IRL Press, Oxford, Washington DC 6 Gupta RS, Austin RC (1987) Mitochondrial matrix localization of a protein altered in mutants resistant to the microtubule inhibitor podophyllotoxin. Eur J Cell Bio145, 170-176
Mitochondrial maturation during neurogenes~;s 7 Gupta RS (1990) Mitochondria, molecular chaperone proteins and the in vivo assembly of microtubules. Trends Biochem $ci 15, 415-418 8 Hay R, Bohni P, Gasser S (1984) How mitochondria import proteins. BBA 779, 65-87 9 Hattori T, Hamada S, Masaki T, Kushima R, Ohmura M, Kataoka Y, Ikehara Y, Kita M, Magami Y (1989) Turnover of glial cells in the central nervous system. Biomed Res 10, 355-363 10 Jacobson M (1978) Developmental Neurobiology. Plenum Press, NY/London 11 Kadenbach B, Khu-Netwig L, Buge U (1987) Evolution of a regulatory enzyme cytochrome c oxidase (complexe IV) Curt Topics Bioener 15, 113-161 12 Kato K, Suzuki F, Watanahe T, Semba R, Keino H (1984) Developmental profile of three enolase isozymes in rat brain determination from one-cell embryo to adult brain. Neurochem Int 6, 51-54 13 Leong SF, Lai JCK, Lim L, Clark JB (1981) Ener.gymetabollsing enzymes in brain regions of adult and aging rat. Y Neurochem 37, 1548-1556 14 Malloch GDA, Munday LA, Olson MS, Clark JB (1986) Comparative development of the pyruvate dehydrogenase complex and citrate syuthase in rat brain mitochondria. Biochem Y 238, 729-736 15 Mjaatvedt AlE, Wong-Riley MTT (1988) Relationship between synaptogenesis and cytochrome oxidase activity in purkinje cells of the developing rat cerebellum. J Comp Neurol 277, 155-182 16 Nicholls MG (1978) Calcium transport and proton electrochemical gradient in mitochondria from guinea-pig cerebral cortex and rat heart. Biochem J 170, 511-522
65
17 O'Farell PZ, Goodman HM, O'Farell PH (1977) High resolution two-dimensional electrophoresis of basic as well as acidic proteins. Cell 12, 1133-1142 18 Pollak JK, Sutton R (1980) The differentiation of animal mitochondria during development. Trends Biochem $ci 5, 23-27 19 Reichenbac A (1989) Gila: Neuron index: Review and Hypothesis to account for different values in various mammals. Glia 2, 71-77 20 Schilling K, Scherbaum C, Pilgrim C (1988) Developmenhal changes of neuron-specific enolase and neurofilament proteins in primary neural culture. Histochem 89, 295-299 21 Sebben M, Gabrion J, Manzoni O, Sladeczek F, Grii C, Bockaert J, Dumuis A (1990) Estabfishment of a long-term primary culture of striatal neurons. Dev Brain Res 52, 229-239 22 Thompson S (1987) Simpfifiedand reproducible silver staining of proteins in polyacrylamide gels. Med $ci Res 15, 1253-1255 23 Towbin H, Staehelin T, Gordon J (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some appfications. Proc Natl Acad $ci USA 76, 4350-4354 24 Vayssi6re JL, Cordeau-Lossouarn L, Larcher JC, Gros F, Croizat B (1989) Tissue-specific mitochondrial proteins. Biochimie 71,787-791 25 Vielkind U, Swierenga SH (1989) A simple fixation procedure for immunofluorescent detection of different o/toskeletal components within the same cell. Histochem 91, 81-88 26 Wong~Riley MTT (1989) Cytochrome oxidase: an endogenous metabofic marker for neuronal activity. TINS 12, 94-101
