- Email: [email protected]
Chronic n-3 polyunsaturated fatty acid deficiency alters dopamine vesicle density in the rat frontal cortex
Neuroscience Letters 284 (2000) 25±28
www.elsevier.com/locate/neulet
Chronic n-3 polyunsaturated fatty acid de®ciency alters dopamine vesicle density in the rat frontal cortex L. Zimmer a, S. Delpal b, D. Guilloteau a, J, AõÈoun b, G. Durand b, S. Chalon a,* a
ISERM U316, Laboratoire de Biophysique MeÂdicale et Pharmaceutique, Faculte de Pharmacie, 31 Avenue Monge, 37200 Tours, France b INRA, Laboratoire de Nutrition et SeÂcurite Alimentaire, INRA, Jouy-en-Josas, France Received 28 October 1999; received in revised form 24 February 2000; accepted 24 February 2000
Abstract We studied the effects of a chronic de®ciency in n-3 polyunsaturated fatty acids (n-3 PUFA) on the vesicle dopaminergic compartment in the frontal cortex of rats. Electronic micrographic analysis showed that the synaptic density and the clear vesicle density were similar in de®cient and control rats. However, dopaminergic immunolabeling revealed a signi®cantly decreased number of gold-labeled vesicles in the dopaminergic presynaptic terminals of the de®cient rats. These ®ndings demonstrate that dopamine cortical vesicles are speci®cally decreased in n-3 PUFA de®ciency. The mechanism leading to this modi®cation could involve several abnormalities (vesicle turn-over, membrane ¯uidity, vesicular monoamine transporter). This reduction in the dopaminergic vesicle pool constitutes the ®rst structural support for the previously described modi®cations of dopamine metabolism in the frontal cortex. Such changes in dopamine neurotransmission could be involved in behavioral abnormalities occurring in n-3 PUFA de®cient rats. q 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Dopamine; Electron microscopy; Frontal cortex; n-3 Polyunsaturated fatty acid; Vesicular compartment
It is known that a diet de®cient in alpha-linolenic acid, the precursor of n-3 polyunsaturated fatty acid (PUFA), greatly affects the fatty acid composition of membrane phospholipids in the central nervous system [1]. Moreover n-3 PUFA deprivation in the rodent over several generations results in overall changes in behavior, including performance on learning tasks and reactivity to stimuli or rewards, whereas locomotion remains unchanged [5]. Although it is dif®cult to link responses to behavioral tests with speci®c neurochemical pathways, we have recently proposed that the effects of n-3 PUFA de®ciency on the rat behavior could be mediated through the cortical dopaminergic system [3,4], which is known to be a major modulator of learning [2]. Rats with n-3 PUFA de®ciency show decreased levels of endogenous dopamine and decreased D2 receptors in the frontal cortex but not in the striatum or cerebellum [3,4]. In addition, microdialysis ®ndings have revealed that the vesicular dopamine is decreased and the metabolite pathway is enhanced [18]. The binding of the vesicular monoamine
* Corresponding author. Tel.: 133-02-47-36-72-18; fax: 133-0247-36-72-24. E-mail address: [email protected] (S. Chalon).
transporter (VMAT2), studied using [ 3H]-dihydrotetrabenazine, is also markedly decreased [19]. It can therefore be proposed that dietary n-3 PUFA de®ciency induces abnormalities in the dopamine vesicles in the frontal cortex. In order to check this proposal, using electronic microscopy we ®rst examined whether ultrastructural changes occurred in cortical neurons after chronic n-3 PUFA de®ciency. We then quanti®ed the dopamine contained in the vesicular compartment using dopamine immunolabeling. Two generations of female Wistar rats originating from the Laboratoire de Nutrition et SeÂcurite Alimentaire (INRA, Jouy-en-Josas, France) were fed a diet containing 6/100 g fat in the form of African peanut oil speci®cally de®cient in a-linolenic acid (precursor of n-3 PUFA), as previously described [3,4]. Two weeks before mating, female rats originating from the second generation of a-linolenic acid-de®cient rats were divided into two groups. The ®rst group (`de®cient rats') received the n-3 PUFA de®cient diet and the second group (`control rats') received a diet in which peanut oil was replaced by a mixture of 60% peanut oil and 40% rapeseed oil (containing 200 mg of a-linolenic acid per 100 g of diet). At weaning the male progeny of these two groups of female rats received the same diet as
0304-3940/00/$ - see front matter q 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S03 04 - 394 0( 0 0) 00 95 0- 2
26
L. Zimmer et al. / Neuroscience Letters 284 (2000) 25±28
their respective dams. Experiments were performed on 250± 300 g male rats (2±3 months of age) from both dietary groups. The experimental procedures were in compliance with guidelines from the European Communities Council directives 86/609/EEC. Four control rats and four de®cient rats were anesthetized with sodium pentobarbital (50 mg/kg, i.p.) and perfused through the ascending aorta with 40 ml of heparin (1000 units/ml heparin in 0.15 M NaCl), and 50 ml of a mixture of 2.5% glutaraldehyde, 1% paraformaldehyde and 0.1% picric acid in phosphate buffer, pH 7.3. After ®xation, the brains were removed and post-®xed in the same solution for 2 h at 48C. Sections were then obtained (50 mm) through the frontal cortex (A.P. coordinates 1.7 from Bregma according to [10]) using a Vibratome. Light microscopy was used to localise the areas of interest (Layers V±VI of prefrontal cortex) and adjacent sections were prepared for electron microscopy. Embedding was performed by osmium-free techniques, as previously described [12], and characterised by immunoreactivity preservation. Brie¯y, the sections were incubated for 40 min in 1% tannic acid in 0.1 M maleate buffer (MB), incubated in the dark for 40 min in 1% uranyl acetate in MB and ®nally incubated for 20 min in 0.5% platinum chloride in MB. Sections were dehydrated in ethanol and incubated for 5 min in 100% propylene oxide. Epon resin was added to make a mixture with propylene oxide and was gently mixed for 2 h. The next morning sections were sandwiched between strips of ACLAR plastic ®lm and polymerized for 24 h at 608C. Thin sections were cut and collected on 300-mesh uncoated nickel grids. Post-embedding immunocytochemistry was performed as follows: grids were washed with Tris-buffered saline containing 0.1% Triton X-100 (TBS/T), pH 7.6, incubated overnight with a primary dopamine antibody (Rabbit polyclonal antidopamine 1:800, Af®niti, U.K.), rinsed in TBS/T (pH 8.2) incubated with a secondary antibody (goat anti-rabbit IgG 1:50 conjugated to 10-nm gold particles (Amersham, France)) in TBS/T (pH 8.2) for 1 h, then rinsed and dried. For structural numeration, grids were examined with a Philips 201 transmission electron microscope at 80 kV. Photographs were usually taken at £20 000 and represented a 6 £ 6 mm area. Synaptic junctions and small synaptic vesicles were identi®ed on the criteria of Peters et al. [11]. Fifty images were examined for four de®cient rats and 53 images for four control rats (10±13 images per rat for each group) to determine: (i) the number of synaptic junctions in the photograph area (6 mm 2); (ii) the number of small clear vesicles per terminal axon (synaptic vesicle density was de®ned as synaptic vesicle counts per terminal axon). The results are reported as the mean ^ SEM, and statistical signi®cance, calculated using Student's t-test, was de®ned as P , 0:05. Validity of the dopamine-labeling was established by observation of the distribution of immunogold particles in the subcellular organelles. Each photograph represented 6
mm 2. Two particle counts were performed for each group: (i) the total number of immunogold particles (plasmalemmal and intracellular location) per 6 mm 2-®eld; and (ii) the number of immunogold particles located in the vesicle compartment per dopaminergic labeled terminal, de®ning the dopamine-labeled vesicle density. The results are reported as the mean ^ SEM, and statistical signi®cance, calculated using Student's t-test, was de®ned as P , 0:05. The synapses were mostly of symmetrical type, according to [14]. The average synapse density was identical in both the de®cient group and the control group: 15 ^ 3 vs. 17 ^ 4 synapses per 6 mm 2-section on average. We focused on the number of presynaptic vesicles adjacent to a dense postsynaptic structure. The cortical terminals contained many clear round vesicles of 50 nm and a few large dense-core vesicles of approximately 90 nm diameter. The photomicrographs in Fig. 1 are typical examples of cortical synapse terminals. Histograms of clear vesicles per axon terminal are shown in Fig. 1. We examined 177 terminals in four de®cient rats and 148 terminals in control rats ( < 40 terminals per rat). The average clear vesicle density of the de®cient group (60 ^ 12 vesicles/terminal) was equivalent to that of the control group (65 ^ 18 vesicles/terminal). Immunogold labeling for dopamine was detected within extracellular and intracellular compartments in the cortex of de®cient and control rats. Labeling was absent from sections in which the primary antiserum had been omitted (results not shown). A smaller number of dopamine-labeled gold particles per section was detected in all the photographic ®elds in de®cient rats than in control rats. The average number of immuno gold particles per synapse was signi®cantly lower (40%, P , 0:05) in de®cient rats than in
Fig. 1. Left: histograms of synaptic vesicles in de®cient and control groups. Four rats were used in each group, and a total of 177 or 148 terminals (NTM, number of terminals measured) were measured. Vesicle densities were similar in both groups. Right: electron photomicrographs of frontal cortex synapses (£20 000). Bar represents 0.2 mm.
L. Zimmer et al. / Neuroscience Letters 284 (2000) 25±28
control rats (251 ^ 44 vs. 402 ^ 68 dopamine-particles per 6 mm 2-®eld in de®cient and control rats, respectively). We also focused on dopaminergic presynapses containing at least one dopamine-labeled vesicle. We detected on average 12 dopamine-labeled terminals per photograph in de®cient and control rats. Because dopaminergic ®bres are sparingly distributed in the frontal cortex [14], this number represented about 25% of total terminals in these sections. The percentage of dopaminergic terminals with higher dopamine-labeled vesicle densities was lower in the de®cient group compared to the control group. The histograms in Fig. 2 demonstrate the differences in the distribution of dopamine-labeled vesicle densities between the two dietary groups. Both histograms ®tted a normal distribution, although the average dopamine-labeled vesicle densities were 6.2 ^ 1.8 vs. 15.9 ^ 3.9 in the de®cient and control groups, respectively. The difference between groups was statistically signi®cant (P , 0:05). In agreement with several published ®ndings, we observed cortical terminals which contained mainly small clear vesicles and a few large dense-core vesicles [11,14]. The synaptic vesicles containing dopamine belong to the clear vesicle family. The similarity in synapse density and
Fig. 2. Histograms of density of dopamine-labeled synaptic vesicles in the frontal cortex of de®cient and control rats. A total of 45 dopamine-labeled terminals were analysed in the de®cient group, and 41 dopamine-labeled terminals in the control group. Dopamine-labeled vesicle densities were signi®cantly lower in the de®cient group compared to the control group (P , 0:05, Student's t-test for unpaired values).
27
vesicle density between the two dietary groups might indicate the same cortical innervation. However, there was a decrease in the number of dopamine-labeled vesicles in the de®cient group. The total number of dopamine-labeled gold particles was smaller in de®cient than in control rats and this difference could be caused by the lower level of stored dopamine. Our study demonstrates therefore that the dopamine vesicle compartment is decreased in the cortex of de®cient rats. It is known that modi®cation of the vesicular compartment in¯uences neurotransmitter metabolism and has a dramatic in¯uence on transmitter release [7]. The decrease in the number of dopamine vesicles may disturb the quantal size of dopamine (number of transmitter molecules released from individual synaptic vesicles). The present study thus constitutes the ®rst direct evidence explaining the decrease in dopamine release [19] and enhanced metabolite formation in de®cient rats [18]. Evidence from several sources indicates that the frontal cortex is involved in memory, behavioral ¯exibility and exploratory activity [9]. Structural changes in cortical synapses have been reported to be linked to changes in learning situations [8]. Moreover, it is known that cortical dopamine is involved in cognitive processes [2]. As previously described, the learning performance of n-3 PUFA de®cient rats and mice is decreased [1,5,15]. The reduction in dopamine available in the cortical vesicle pool can therefore be related to such inappropriate behavioural responses. The inadequate reserves of newly synthesised dopamine might not be suf®cient to maintain high release during stimulated cognitive processes [6]. In addition, it has already be shown that the synaptic vesicles of n-3 PUFA de®cient rats decrease in the hippocampus after a learning situation, in contrast to rats fed an equilibrated diet [15]. In our study, we did not observe any modi®cation of the total number of clear vesicles in the cortex, but the rats were not subjected to learning tests. The possible vesicle modi®cations might therefore have occurred in the frontal cortex of de®cient rats after a learning session. Moreover, the speci®city of the reduction in dopaminelabeled vesicles we observed in our conditions might be explained by an increase in the reactivity of the dopaminergic system to environmental stimuli or stress (e.g. manipulation before euthanasia) in de®cient rats. The changes in dopamine vesicle density may result from several mechanisms. The ®rst mechanism could involve changes in vesicle turnover which depend on membrane movement to recycling sites [7]. According to Yoshida [15,16], enzymatic membrane modi®cations after n-3 PUFA de®ciency might alter the processes of vesicle restoration after dopamine release. The second mechanism could be changes in the physical properties of neuronal membranes. Modi®cation of dietary intake of n-3 PUFA in¯uences the content of the membrane phospholipid bilayer and has a considerable impact on synaptic membrane ¯uidity [13,17]. Decrease in membrane microviscosity could therefore reduce the formation of dopamine
28
L. Zimmer et al. / Neuroscience Letters 284 (2000) 25±28
vesicles. However both these hypotheses, which imply nonspeci®c mechanisms, seem insuf®cient to explain our results. We found no differences in the proportion of clear vesicles per termination between de®cient and control rats but speci®c modi®cations in the dopaminergic vesicle pool. A plausible mechanism could therefore involve the vesicular monoamine transporter (VMAT2). In dopaminergic nerve endings, dopamine is stored in synaptic vesicles via the VMAT2. We previously observed that the autoradiographic labeling of cortical VMAT2 was reduced in de®cient rats [19]. Although this VMAT2 level could also be a consequence of the decreased number of dopamine vesicles, it cannot be excluded that altered PUFA membrane composition modi®es the structure of the VMAT2, and thus decreases internalisation in storage vesicles. Further studies are necessary to explain the mechanism by which `nonspeci®c' molecules (n-3 PUFA) interact with speci®c molecules (i.e. VMAT2) in speci®c areas. Our results constitute direct evidence that there are differences in the dopaminergic vesicle compartment in the frontal cortex of n-3 PUFA de®cient rats. Decreased dopamine vesicle density could therefore be responsible for lower dopamine release during cognitive processes. This study provides the ®rst structural evidence for disturbance of dopamine cortical neurotransmission after n-3 PUFA de®ciency and possibly forms the basis for understanding behavioral abnormalities in de®cient rats.
[6] [7] [8] [9] [10] [11] [12]
[13]
[14]
[15]
We thank Serge Dahirel for his excellent technical assistance, and Doreen Raine for editorial assistance. [1] Bourre, J.M., Francois, M., Youyou, A., Dumont, O., Piciotti, M., Pascal, G. and Durand, G., The effects on dietary alphalinolenic acid on the composition of nerve membranes, enzymatic activity, amplitude of electrophysiological parameters, resistance to poisons and performance of learning tasks in rats, J. Nutr., 119 (1989) 1880±1892. [2] Brozoski, T.J., Brown, R.M., Rosvold, H.E. and Goldman, P.S., Cognitive de®cit caused by regional depletion of dopamine in prefrontal cortex of Rhesus monkey, Science, 205 (1979) 929±932. [3] Delion, S., Chalon, S., HeÂrault, J., Guilloteau, D., Besnard, J.C. and Durand, G., Chronic dietary alpha-linolenic acid de®ciency alters dopaminergic and serotoninergic neurotransmission in rats, J. Nutr., 124 (1994) 2466±2476. [4] Delion, S., Chalon, S., Guilloteau, D., Besnard, J.C. and Durand, G., Alpha-linolenic acid dietary de®ciency alters age-related changes of dopaminergic and serotoninergic neurotransmission in the rat frontal cortex, J. Neurochem., 66 (1996) 1582±1591. [5] Frances, H., Monier, C. and Bourre, J.M., Effects of dietary
[16]
[17]
[18]
[19]
alpha-linolenic acid de®ciency on neuromuscular and cognitive functions in mice, Life Sci., 57 (1995) 1935±1947. Garris, P.A., Collins, L.B., Jones, S.R. and Wightman, R.M., Evoked extracellular dopamine in vivo in the medial prefrontal cortex, J. Neurochem., 61 (1993) 637±647. Kelly, R.B., Storage and release of neurotransmitters, Cell, 10 (1993) 43±53. Klintsova, A.Y. and Greenough, W.T., Synaptic plasticity in cortical systems, Curr. Opin. Neurobiol., 9 (1999) 203±208. Mitchell, J.B. and Laiacona, I., The medial frontal cortex and temporal memory: tests using spontaneous exploratory behaviour in the rat, Behav. Brain Res., 97 (1998) 107±113. Paxinos, G. and Watson, C., The Rat Brain in Stereotaxic Coordinates, Academic Press, New York, 1982. Peters, A., Palay, S.L. and Webster, H.D., The Fine Structure of the Nervous System, Oxford University Press, New York, 1991. Phend, K.D., Rustioni, A. and Weinberg, R.J., An osmiumfree method of epon embedment that preserves both ultrastructure and antigenicity for post-embedding immunocytochemistry, J. Histochem. Cytochem., 43 (1995) 283±292. Suzuki, H., Park, S.J., Tamura, M. and Ando, S., Effect of the long-term feeding of dietary lipids on the learning ability, fatty acid composition of brain stem phospholipids and synaptic membrane ¯uidity in adult mice: a comparison of sardine oil diet with palm oil diet, Mech. Age Dev., 101 (1998) 119±128. Van Eden, C.G., Hoorneman, E.M., Buijs, R.M., Matthijssen, M.A., Geffard, M. and Uylings, H.B., Immunocytochemical localization of dopamine in the prefrontal cortex of the rat at the light and electron microscopical level, Neuroscience, 22 (1987) 849±862. Yoshida, S., Yasuda, A., Kawazato, K., Sakai, K., Shimada, T., Takeshita, M., Yuasa, S., Kobayashi, T., Watanabe, S. and Okuyama, H., Synaptic vesicle ultrastructural changes in the rat hippocampus induced by a combination of alphalinolenate de®ciency and a learning task, J. Neurochem., 68 (1997) 1261±1268. Yoshida, S., Miyazaki, M., Takeshita, M., Yuasa, S., Kobayashi, T., Watanabe, S. and Okuyama, H., Functional changes of rat brain microsomal membrane surface after learning task depending on dietary fatty acids, J. Neurochem., 68 (1997) 1269±1277. Zerouga, M., BeaugeÂ, F., Niel, E., Durand, G. and Bourre, J.M., Interactive effects of dietary (n-3) polyunsaturated fatty acids and chronic ethanol intoxication on synaptic membrane lipid composition and ¯uidity in rats, Biochim. Biophys. Acta, 1086 (1991) 295±304. Zimmer, L., Hembert, S., Durand, G., Breton, P., Guilloteau, D., Besnard, J.C. and Chalon, S., Chronic n-3 polyunsaturated fatty acid diet-de®ciency acts on dopamine metabolism in the rat frontal cortex: a microdialysis study, Neurosci. Lett., 240 (1998) 177±181. Zimmer, L., Breton, P., Durand, G., Guilloteau, D., Besnard, J.C. and Chalon, S., Prominent role of n-3 polyunsaturated fatty acids in cortical dopamine metabolism, Nutr. Neurosci., 2 (1999) 257±265.
www.elsevier.com/locate/neulet
Chronic n-3 polyunsaturated fatty acid de®ciency alters dopamine vesicle density in the rat frontal cortex L. Zimmer a, S. Delpal b, D. Guilloteau a, J, AõÈoun b, G. Durand b, S. Chalon a,* a
ISERM U316, Laboratoire de Biophysique MeÂdicale et Pharmaceutique, Faculte de Pharmacie, 31 Avenue Monge, 37200 Tours, France b INRA, Laboratoire de Nutrition et SeÂcurite Alimentaire, INRA, Jouy-en-Josas, France Received 28 October 1999; received in revised form 24 February 2000; accepted 24 February 2000
Abstract We studied the effects of a chronic de®ciency in n-3 polyunsaturated fatty acids (n-3 PUFA) on the vesicle dopaminergic compartment in the frontal cortex of rats. Electronic micrographic analysis showed that the synaptic density and the clear vesicle density were similar in de®cient and control rats. However, dopaminergic immunolabeling revealed a signi®cantly decreased number of gold-labeled vesicles in the dopaminergic presynaptic terminals of the de®cient rats. These ®ndings demonstrate that dopamine cortical vesicles are speci®cally decreased in n-3 PUFA de®ciency. The mechanism leading to this modi®cation could involve several abnormalities (vesicle turn-over, membrane ¯uidity, vesicular monoamine transporter). This reduction in the dopaminergic vesicle pool constitutes the ®rst structural support for the previously described modi®cations of dopamine metabolism in the frontal cortex. Such changes in dopamine neurotransmission could be involved in behavioral abnormalities occurring in n-3 PUFA de®cient rats. q 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Dopamine; Electron microscopy; Frontal cortex; n-3 Polyunsaturated fatty acid; Vesicular compartment
It is known that a diet de®cient in alpha-linolenic acid, the precursor of n-3 polyunsaturated fatty acid (PUFA), greatly affects the fatty acid composition of membrane phospholipids in the central nervous system [1]. Moreover n-3 PUFA deprivation in the rodent over several generations results in overall changes in behavior, including performance on learning tasks and reactivity to stimuli or rewards, whereas locomotion remains unchanged [5]. Although it is dif®cult to link responses to behavioral tests with speci®c neurochemical pathways, we have recently proposed that the effects of n-3 PUFA de®ciency on the rat behavior could be mediated through the cortical dopaminergic system [3,4], which is known to be a major modulator of learning [2]. Rats with n-3 PUFA de®ciency show decreased levels of endogenous dopamine and decreased D2 receptors in the frontal cortex but not in the striatum or cerebellum [3,4]. In addition, microdialysis ®ndings have revealed that the vesicular dopamine is decreased and the metabolite pathway is enhanced [18]. The binding of the vesicular monoamine
* Corresponding author. Tel.: 133-02-47-36-72-18; fax: 133-0247-36-72-24. E-mail address: [email protected] (S. Chalon).
transporter (VMAT2), studied using [ 3H]-dihydrotetrabenazine, is also markedly decreased [19]. It can therefore be proposed that dietary n-3 PUFA de®ciency induces abnormalities in the dopamine vesicles in the frontal cortex. In order to check this proposal, using electronic microscopy we ®rst examined whether ultrastructural changes occurred in cortical neurons after chronic n-3 PUFA de®ciency. We then quanti®ed the dopamine contained in the vesicular compartment using dopamine immunolabeling. Two generations of female Wistar rats originating from the Laboratoire de Nutrition et SeÂcurite Alimentaire (INRA, Jouy-en-Josas, France) were fed a diet containing 6/100 g fat in the form of African peanut oil speci®cally de®cient in a-linolenic acid (precursor of n-3 PUFA), as previously described [3,4]. Two weeks before mating, female rats originating from the second generation of a-linolenic acid-de®cient rats were divided into two groups. The ®rst group (`de®cient rats') received the n-3 PUFA de®cient diet and the second group (`control rats') received a diet in which peanut oil was replaced by a mixture of 60% peanut oil and 40% rapeseed oil (containing 200 mg of a-linolenic acid per 100 g of diet). At weaning the male progeny of these two groups of female rats received the same diet as
0304-3940/00/$ - see front matter q 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S03 04 - 394 0( 0 0) 00 95 0- 2
26
L. Zimmer et al. / Neuroscience Letters 284 (2000) 25±28
their respective dams. Experiments were performed on 250± 300 g male rats (2±3 months of age) from both dietary groups. The experimental procedures were in compliance with guidelines from the European Communities Council directives 86/609/EEC. Four control rats and four de®cient rats were anesthetized with sodium pentobarbital (50 mg/kg, i.p.) and perfused through the ascending aorta with 40 ml of heparin (1000 units/ml heparin in 0.15 M NaCl), and 50 ml of a mixture of 2.5% glutaraldehyde, 1% paraformaldehyde and 0.1% picric acid in phosphate buffer, pH 7.3. After ®xation, the brains were removed and post-®xed in the same solution for 2 h at 48C. Sections were then obtained (50 mm) through the frontal cortex (A.P. coordinates 1.7 from Bregma according to [10]) using a Vibratome. Light microscopy was used to localise the areas of interest (Layers V±VI of prefrontal cortex) and adjacent sections were prepared for electron microscopy. Embedding was performed by osmium-free techniques, as previously described [12], and characterised by immunoreactivity preservation. Brie¯y, the sections were incubated for 40 min in 1% tannic acid in 0.1 M maleate buffer (MB), incubated in the dark for 40 min in 1% uranyl acetate in MB and ®nally incubated for 20 min in 0.5% platinum chloride in MB. Sections were dehydrated in ethanol and incubated for 5 min in 100% propylene oxide. Epon resin was added to make a mixture with propylene oxide and was gently mixed for 2 h. The next morning sections were sandwiched between strips of ACLAR plastic ®lm and polymerized for 24 h at 608C. Thin sections were cut and collected on 300-mesh uncoated nickel grids. Post-embedding immunocytochemistry was performed as follows: grids were washed with Tris-buffered saline containing 0.1% Triton X-100 (TBS/T), pH 7.6, incubated overnight with a primary dopamine antibody (Rabbit polyclonal antidopamine 1:800, Af®niti, U.K.), rinsed in TBS/T (pH 8.2) incubated with a secondary antibody (goat anti-rabbit IgG 1:50 conjugated to 10-nm gold particles (Amersham, France)) in TBS/T (pH 8.2) for 1 h, then rinsed and dried. For structural numeration, grids were examined with a Philips 201 transmission electron microscope at 80 kV. Photographs were usually taken at £20 000 and represented a 6 £ 6 mm area. Synaptic junctions and small synaptic vesicles were identi®ed on the criteria of Peters et al. [11]. Fifty images were examined for four de®cient rats and 53 images for four control rats (10±13 images per rat for each group) to determine: (i) the number of synaptic junctions in the photograph area (6 mm 2); (ii) the number of small clear vesicles per terminal axon (synaptic vesicle density was de®ned as synaptic vesicle counts per terminal axon). The results are reported as the mean ^ SEM, and statistical signi®cance, calculated using Student's t-test, was de®ned as P , 0:05. Validity of the dopamine-labeling was established by observation of the distribution of immunogold particles in the subcellular organelles. Each photograph represented 6
mm 2. Two particle counts were performed for each group: (i) the total number of immunogold particles (plasmalemmal and intracellular location) per 6 mm 2-®eld; and (ii) the number of immunogold particles located in the vesicle compartment per dopaminergic labeled terminal, de®ning the dopamine-labeled vesicle density. The results are reported as the mean ^ SEM, and statistical signi®cance, calculated using Student's t-test, was de®ned as P , 0:05. The synapses were mostly of symmetrical type, according to [14]. The average synapse density was identical in both the de®cient group and the control group: 15 ^ 3 vs. 17 ^ 4 synapses per 6 mm 2-section on average. We focused on the number of presynaptic vesicles adjacent to a dense postsynaptic structure. The cortical terminals contained many clear round vesicles of 50 nm and a few large dense-core vesicles of approximately 90 nm diameter. The photomicrographs in Fig. 1 are typical examples of cortical synapse terminals. Histograms of clear vesicles per axon terminal are shown in Fig. 1. We examined 177 terminals in four de®cient rats and 148 terminals in control rats ( < 40 terminals per rat). The average clear vesicle density of the de®cient group (60 ^ 12 vesicles/terminal) was equivalent to that of the control group (65 ^ 18 vesicles/terminal). Immunogold labeling for dopamine was detected within extracellular and intracellular compartments in the cortex of de®cient and control rats. Labeling was absent from sections in which the primary antiserum had been omitted (results not shown). A smaller number of dopamine-labeled gold particles per section was detected in all the photographic ®elds in de®cient rats than in control rats. The average number of immuno gold particles per synapse was signi®cantly lower (40%, P , 0:05) in de®cient rats than in
Fig. 1. Left: histograms of synaptic vesicles in de®cient and control groups. Four rats were used in each group, and a total of 177 or 148 terminals (NTM, number of terminals measured) were measured. Vesicle densities were similar in both groups. Right: electron photomicrographs of frontal cortex synapses (£20 000). Bar represents 0.2 mm.
L. Zimmer et al. / Neuroscience Letters 284 (2000) 25±28
control rats (251 ^ 44 vs. 402 ^ 68 dopamine-particles per 6 mm 2-®eld in de®cient and control rats, respectively). We also focused on dopaminergic presynapses containing at least one dopamine-labeled vesicle. We detected on average 12 dopamine-labeled terminals per photograph in de®cient and control rats. Because dopaminergic ®bres are sparingly distributed in the frontal cortex [14], this number represented about 25% of total terminals in these sections. The percentage of dopaminergic terminals with higher dopamine-labeled vesicle densities was lower in the de®cient group compared to the control group. The histograms in Fig. 2 demonstrate the differences in the distribution of dopamine-labeled vesicle densities between the two dietary groups. Both histograms ®tted a normal distribution, although the average dopamine-labeled vesicle densities were 6.2 ^ 1.8 vs. 15.9 ^ 3.9 in the de®cient and control groups, respectively. The difference between groups was statistically signi®cant (P , 0:05). In agreement with several published ®ndings, we observed cortical terminals which contained mainly small clear vesicles and a few large dense-core vesicles [11,14]. The synaptic vesicles containing dopamine belong to the clear vesicle family. The similarity in synapse density and
Fig. 2. Histograms of density of dopamine-labeled synaptic vesicles in the frontal cortex of de®cient and control rats. A total of 45 dopamine-labeled terminals were analysed in the de®cient group, and 41 dopamine-labeled terminals in the control group. Dopamine-labeled vesicle densities were signi®cantly lower in the de®cient group compared to the control group (P , 0:05, Student's t-test for unpaired values).
27
vesicle density between the two dietary groups might indicate the same cortical innervation. However, there was a decrease in the number of dopamine-labeled vesicles in the de®cient group. The total number of dopamine-labeled gold particles was smaller in de®cient than in control rats and this difference could be caused by the lower level of stored dopamine. Our study demonstrates therefore that the dopamine vesicle compartment is decreased in the cortex of de®cient rats. It is known that modi®cation of the vesicular compartment in¯uences neurotransmitter metabolism and has a dramatic in¯uence on transmitter release [7]. The decrease in the number of dopamine vesicles may disturb the quantal size of dopamine (number of transmitter molecules released from individual synaptic vesicles). The present study thus constitutes the ®rst direct evidence explaining the decrease in dopamine release [19] and enhanced metabolite formation in de®cient rats [18]. Evidence from several sources indicates that the frontal cortex is involved in memory, behavioral ¯exibility and exploratory activity [9]. Structural changes in cortical synapses have been reported to be linked to changes in learning situations [8]. Moreover, it is known that cortical dopamine is involved in cognitive processes [2]. As previously described, the learning performance of n-3 PUFA de®cient rats and mice is decreased [1,5,15]. The reduction in dopamine available in the cortical vesicle pool can therefore be related to such inappropriate behavioural responses. The inadequate reserves of newly synthesised dopamine might not be suf®cient to maintain high release during stimulated cognitive processes [6]. In addition, it has already be shown that the synaptic vesicles of n-3 PUFA de®cient rats decrease in the hippocampus after a learning situation, in contrast to rats fed an equilibrated diet [15]. In our study, we did not observe any modi®cation of the total number of clear vesicles in the cortex, but the rats were not subjected to learning tests. The possible vesicle modi®cations might therefore have occurred in the frontal cortex of de®cient rats after a learning session. Moreover, the speci®city of the reduction in dopaminelabeled vesicles we observed in our conditions might be explained by an increase in the reactivity of the dopaminergic system to environmental stimuli or stress (e.g. manipulation before euthanasia) in de®cient rats. The changes in dopamine vesicle density may result from several mechanisms. The ®rst mechanism could involve changes in vesicle turnover which depend on membrane movement to recycling sites [7]. According to Yoshida [15,16], enzymatic membrane modi®cations after n-3 PUFA de®ciency might alter the processes of vesicle restoration after dopamine release. The second mechanism could be changes in the physical properties of neuronal membranes. Modi®cation of dietary intake of n-3 PUFA in¯uences the content of the membrane phospholipid bilayer and has a considerable impact on synaptic membrane ¯uidity [13,17]. Decrease in membrane microviscosity could therefore reduce the formation of dopamine
28
L. Zimmer et al. / Neuroscience Letters 284 (2000) 25±28
vesicles. However both these hypotheses, which imply nonspeci®c mechanisms, seem insuf®cient to explain our results. We found no differences in the proportion of clear vesicles per termination between de®cient and control rats but speci®c modi®cations in the dopaminergic vesicle pool. A plausible mechanism could therefore involve the vesicular monoamine transporter (VMAT2). In dopaminergic nerve endings, dopamine is stored in synaptic vesicles via the VMAT2. We previously observed that the autoradiographic labeling of cortical VMAT2 was reduced in de®cient rats [19]. Although this VMAT2 level could also be a consequence of the decreased number of dopamine vesicles, it cannot be excluded that altered PUFA membrane composition modi®es the structure of the VMAT2, and thus decreases internalisation in storage vesicles. Further studies are necessary to explain the mechanism by which `nonspeci®c' molecules (n-3 PUFA) interact with speci®c molecules (i.e. VMAT2) in speci®c areas. Our results constitute direct evidence that there are differences in the dopaminergic vesicle compartment in the frontal cortex of n-3 PUFA de®cient rats. Decreased dopamine vesicle density could therefore be responsible for lower dopamine release during cognitive processes. This study provides the ®rst structural evidence for disturbance of dopamine cortical neurotransmission after n-3 PUFA de®ciency and possibly forms the basis for understanding behavioral abnormalities in de®cient rats.
[6] [7] [8] [9] [10] [11] [12]
[13]
[14]
[15]
We thank Serge Dahirel for his excellent technical assistance, and Doreen Raine for editorial assistance. [1] Bourre, J.M., Francois, M., Youyou, A., Dumont, O., Piciotti, M., Pascal, G. and Durand, G., The effects on dietary alphalinolenic acid on the composition of nerve membranes, enzymatic activity, amplitude of electrophysiological parameters, resistance to poisons and performance of learning tasks in rats, J. Nutr., 119 (1989) 1880±1892. [2] Brozoski, T.J., Brown, R.M., Rosvold, H.E. and Goldman, P.S., Cognitive de®cit caused by regional depletion of dopamine in prefrontal cortex of Rhesus monkey, Science, 205 (1979) 929±932. [3] Delion, S., Chalon, S., HeÂrault, J., Guilloteau, D., Besnard, J.C. and Durand, G., Chronic dietary alpha-linolenic acid de®ciency alters dopaminergic and serotoninergic neurotransmission in rats, J. Nutr., 124 (1994) 2466±2476. [4] Delion, S., Chalon, S., Guilloteau, D., Besnard, J.C. and Durand, G., Alpha-linolenic acid dietary de®ciency alters age-related changes of dopaminergic and serotoninergic neurotransmission in the rat frontal cortex, J. Neurochem., 66 (1996) 1582±1591. [5] Frances, H., Monier, C. and Bourre, J.M., Effects of dietary
[16]
[17]
[18]
[19]
alpha-linolenic acid de®ciency on neuromuscular and cognitive functions in mice, Life Sci., 57 (1995) 1935±1947. Garris, P.A., Collins, L.B., Jones, S.R. and Wightman, R.M., Evoked extracellular dopamine in vivo in the medial prefrontal cortex, J. Neurochem., 61 (1993) 637±647. Kelly, R.B., Storage and release of neurotransmitters, Cell, 10 (1993) 43±53. Klintsova, A.Y. and Greenough, W.T., Synaptic plasticity in cortical systems, Curr. Opin. Neurobiol., 9 (1999) 203±208. Mitchell, J.B. and Laiacona, I., The medial frontal cortex and temporal memory: tests using spontaneous exploratory behaviour in the rat, Behav. Brain Res., 97 (1998) 107±113. Paxinos, G. and Watson, C., The Rat Brain in Stereotaxic Coordinates, Academic Press, New York, 1982. Peters, A., Palay, S.L. and Webster, H.D., The Fine Structure of the Nervous System, Oxford University Press, New York, 1991. Phend, K.D., Rustioni, A. and Weinberg, R.J., An osmiumfree method of epon embedment that preserves both ultrastructure and antigenicity for post-embedding immunocytochemistry, J. Histochem. Cytochem., 43 (1995) 283±292. Suzuki, H., Park, S.J., Tamura, M. and Ando, S., Effect of the long-term feeding of dietary lipids on the learning ability, fatty acid composition of brain stem phospholipids and synaptic membrane ¯uidity in adult mice: a comparison of sardine oil diet with palm oil diet, Mech. Age Dev., 101 (1998) 119±128. Van Eden, C.G., Hoorneman, E.M., Buijs, R.M., Matthijssen, M.A., Geffard, M. and Uylings, H.B., Immunocytochemical localization of dopamine in the prefrontal cortex of the rat at the light and electron microscopical level, Neuroscience, 22 (1987) 849±862. Yoshida, S., Yasuda, A., Kawazato, K., Sakai, K., Shimada, T., Takeshita, M., Yuasa, S., Kobayashi, T., Watanabe, S. and Okuyama, H., Synaptic vesicle ultrastructural changes in the rat hippocampus induced by a combination of alphalinolenate de®ciency and a learning task, J. Neurochem., 68 (1997) 1261±1268. Yoshida, S., Miyazaki, M., Takeshita, M., Yuasa, S., Kobayashi, T., Watanabe, S. and Okuyama, H., Functional changes of rat brain microsomal membrane surface after learning task depending on dietary fatty acids, J. Neurochem., 68 (1997) 1269±1277. Zerouga, M., BeaugeÂ, F., Niel, E., Durand, G. and Bourre, J.M., Interactive effects of dietary (n-3) polyunsaturated fatty acids and chronic ethanol intoxication on synaptic membrane lipid composition and ¯uidity in rats, Biochim. Biophys. Acta, 1086 (1991) 295±304. Zimmer, L., Hembert, S., Durand, G., Breton, P., Guilloteau, D., Besnard, J.C. and Chalon, S., Chronic n-3 polyunsaturated fatty acid diet-de®ciency acts on dopamine metabolism in the rat frontal cortex: a microdialysis study, Neurosci. Lett., 240 (1998) 177±181. Zimmer, L., Breton, P., Durand, G., Guilloteau, D., Besnard, J.C. and Chalon, S., Prominent role of n-3 polyunsaturated fatty acids in cortical dopamine metabolism, Nutr. Neurosci., 2 (1999) 257±265.