Dietary long chain PUFAs differentially affect hippocampal muscarinic 1 and serotonergic 1A receptors in experimental cerebral hypoperfusion

Brain Research 954 (2002) 32–41 www.elsevier.com / locate / bres

Research report

Dietary long chain PUFAs differentially affect hippocampal muscarinic 1 and serotonergic 1A receptors in experimental cerebral hypoperfusion Eszter Farkas a,c , *, Martijn C. de Wilde a,b , Amanda J. Kiliaan b , John Meijer b , Jan N. Keijser a , Paul G.M. Luiten a a

Group of Molecular Neurobiology, Department of Animal Physiology, University of Groningen, Groningen, The Netherlands b Numico Research, Wageningen, The Netherlands c ¨ Health Science Center, Faculty of Medicine, University of Szeged, Szeged, Hungary Department of Anatomy, Albert Szent-Gyorgyi Accepted 19 July 2002

Abstract The chronic dietary intake of essential polyunsaturated fatty acids (PUFAs) can modulate learning and memory by being incorporated into neuronal plasma membranes. Representatives of two PUFA families, the n-3 and n-6 types become integrated into membrane phospholipids, where the actual (n-6) /(n-3) ratio can determine membrane fluidity and thus the function of membrane-bound proteins. In the present experiment we studied hippocampal neurotransmitter receptors after chronic administration of n-3 PUFA enriched diets in a brain hypoperfusion model, which mimics decreased cerebral perfusion as it occurs in ageing and dementia. Male Wistar rats received experimental diets with a decreased (n-6) /(n-3) ratio from weaning on. Chronic experimental cerebral hypoperfusion was imposed by a permanent, bilateral occlusion of the common carotid arteries (2VO) at the age of 4 months. The experiment was terminated when the rats were 7 months old. Three receptor types, the muscarinic 1, serotonergic 1A and the glutaminergic NMDA receptors were labeled in hippocampal slices by autoradiographic methods. Image analysis demonstrated that 2VO increased muscarinic 1 and NMDA receptor density, specifically in the dentate gyrus and the CA3 region, respectively. The increased ratio of n-3 fatty acids in combination with additional dietary supplements enhanced the density of the serotonergic 1A and muscarinic 1 receptors, while n-3 fatty acids alone increased binding only to the muscarinic 1 receptors. Since the examined receptor types reacted differently to the diets, we concluded that besides changes in membrane fluidity, the biochemical regulation of receptor sensitivity might also play a role in increasing hippocampal receptor density.  2002 Elsevier Science B.V. All rights reserved. Theme: Development and regeneration Topic: Nutritional and prenatal factors Keywords: Dietary; Cerebral hypoperfusion; Neuronal membrane; Polyunsaturated fatty acid; Receptor autoradiography

1. Introduction A balanced intake of essential polyunsaturated fatty acids (PUFAs) of the n-3 and n-6 families, which become incorporated into membrane phospholipids, is crucial for the proper functioning of neuronal membranes. The opti*Corresponding author. University of Szeged, School of Medicine, Department of Anatomy, P.O. Box 427, 6701 Szeged, Hungary. Tel.: 136-62-545-669; fax: 136-62-545-707. E-mail address: [email protected] (E. Farkas).

mal ratio between n-3 and n-6 PUFA types gains special importance in neuronal signal transduction, since the biochemical composition of neuronal membranes can influence the activity of membrane-associated signaling proteins, such as receptors, G proteins or enzymes [36]. A typical representative of the n-3 PUFA family abundantly present in the central nervous system (CNS) is docosahexaenoic acid (DHA, 22:6n-3), which is synthesized by the elongation of a-linolenic acid (aLA, 18:3n-3) and eicosapentaenoic acid (EPA, 20:5n-3). Studies investigating the neural effects of n-3 PUFA deficient or sup-

0006-8993 / 02 / $ – see front matter  2002 Elsevier Science B.V. All rights reserved. PII: S0006-8993( 02 )03300-0

E. Farkas et al. / Brain Research 954 (2002) 32–41

plemented diets pay special attention to DHA, which is known to be taken up by the brain in preference to other fatty acids and has a very rapid turnover [15]. Arachidonic acid (AA, 20:4n-6), a prominent member of the n-6 PUFA family and an important functional PUFA of the CNS, is synthesized of linoleic acid (LA, 18:2n-6) [3]. The two types of essential PUFAs (n-3 and n-6) compete with each other for enzymatic conversion, therefore the ratio of the dietary intake of the n-3 and n-6 types has a determining influence on their biological effect in the brain [37]. More specifically, at the level of performance in learning and memory tasks, the deviation from an optimal (n-6) /(n-3) ratio has been shown to aggravate spatial learning in rats [35]. Another experimental approach indicated that an increased hippocampal and cortical DHA /AA ratio due to DHA supplementation negatively correlated with reference memory errors [8]. The precise cellular correlates of PUFA intake-affected changes of cognitive capacity have not been identified, but there is a strong indication that modified neuronal signaling may play an important role. The modification of signal processing properties of neurons has been suggested to depend on the fluidity of cellular membranes. Membrane fluidity can be altered by shifting the balance between the incorporated n-3 and n-6 PUFAs, but dietary PUFAs were also shown to reduce membrane-bound cholesterol that would enhance membrane rigidity when present in excess [36]. Experimental evidence that supplementation with DHA could improve membrane fluidity and fatty acid unsaturation index was found in cultured endothelial or retinoblastoma cells [12,32]. At the same time, the assumption that increased membrane fluidity improved ligand–receptor interaction was substantiated by in vitro ligand binding studies. Synaptosomal membrane preparations demonstrated that binding of serotonin to its receptors, or alternatively, QNB (quinuclidinyl benzylate) low-affinity binding to cholinergic muscarinic receptors significantly increased when membrane fluidity was experimentally enhanced [7,13]. Furthermore, elevated binding to the dopamine D2 receptors in the rat frontal cortex was established after a long-term, n-3 PUFA rich feeding regiment [1]. All these findings indicate that increased neuronal membrane fluidity augments receptor–ligand binding, which can reflect a better availability or exposure of the binding sites on receptor proteins, or an increased concentration of the receptors themselves inserted in the membranes. Recent epidemiological studies have drawn attention to the relationship between the source of dietary fat and the development of dementia [17]. These population based clinical studies associated the high intake of saturated fat and cholesterol with an increased risk for dementia, while high fish consumption (an essential source of n-3 PUFAs) was related to a reduced risk for cognitive decline [17,18,31]. Interestingly, dementia with a vascular component was the most strongly associated with high saturated dietary fat intake [18,31]. A well-documented cere-

33

brovascular factor accompanying ageing, and specially dementia is decreased cerebral blood flow (CBF) [6], which has been shown to correlate with a failing cognitive performance [19,24]. The question whether essential PUFAs could improve CBF parameters in particular was explored, as well. A positron emission tomography (PET) study has recently reported that dietary DHA intake facilitated an age-impaired CBF response to tactile stimulation in old monkeys [33]. The authors presumed that the improvement in CBF response was achieved by DHA acting on the cholinergic innervation of the cerebral vasculature. In the present study we employed the permanent, bilateral occlusion of the common carotid arteries of rats (2VO) to impose chronic cerebral hypoperfusion as a condition that mimics reduced CBF in dementia. By the use of combined EPA-, DHA- and AA-supplemented diets, we were first investigating the prospective improvement of cognitive and cerebrovascular parameters [4]. In the present paper we report on the density of hippocampal receptors in cerebral hypoperfusion under the given dietary conditions. We have chosen the hippocampus as the area of interest because we previously conducted behavioral experiments tackling hippocampus-related spatial learning in the same experimental setup [4]. Specifically, the NMDA, serotonin 1A (5HT 1A ) and cholinergic muscarinic 1 (M 1 ) receptors were labeled with autoradiographic methods, since the hippocampus is very rich in the listed receptor types that are involved in learning and memory. We hypothesized that reduced CBF would affect the density of the mentioned receptor types, which could be restored by an increased ratio of dietary n-3 PUFAs.

2. Materials and methods

2.1. Dietary design The experimental rat chow composition was designed by Numico Research, Wageningen, The Netherlands. In the present experiment, the neuronal effects of two PUFA enriched nutritional regimes were compared to placebo. The control food was essentially identical to the standard rat chow produced by Hope Farms (Woerden, The Netherlands). All three diets used here contained the same composition and amount of carbohydrates, proteins and minerals, contained the same caloric value and were manufactured in the form of regular food pellets. The experimental diet 1 and diet 2 were enriched by PUFAs, antioxidants, vitamins and particular extra additives. Table 1 shows the source of fatty acid additives of the diets, Table 2 lists the PUFA contents, while Table 3 summarizes the types and amounts of the other additives. Diet 1 differed from the control only in that diet 1 contained additional PUFAs, where the (n-6) /(n-3) ratio was reduced to 1.30 compared to 4.13 in the control diet. PUFA

E. Farkas et al. / Brain Research 954 (2002) 32–41

34

Table 1 The source of fatty acid additives of the experimental diets (g / 100 g) Component

Control

Diet 1

Diet 2

Soybean oil (Florin, Switzerland) Marinol C45 (source of EPA and DHA) (Loders Crocklaan, The Netherlands) Ropufa 50 (source of AA) (Roche, The Netherlands)

5

2.5

2.5

–

2.15

2.15

–

0.35

0.35

Table 2 The polyunsaturated fatty acid content of the experimental diets (g / 100 g) Fatty acid

Experimental diet Control

Diet 1

Diet 2

18:2n-6 (LA) 20:4n-6 (AA) Total n-6 18:3n-3 (ALA) 20:5n-3 (EPA) 22:6n-3 (DHA) Total n-3

0.640 0.000 0.640 0.155 0.000 0.000 0.155

1.321 0.118 1.439 0.137 0.589 0.382 1.108

1.661 0.118 1.779 0.184 0.589 0.382 1.155

(n-6) /(n-3)

4.13

1.30

1.54

additions in diet 1 were stabilized by antioxidants and vitamins (Table 3). Diet 2 was a complex diet enriched with PUFAs at an (n-6) /(n-3) ratio of 1.54, supplied with antioxidants and vitamins similar to diet 1, and provided with additional phospholipids and neurotransmitter precursors (see Tables 2 and 3 for details).

2.2. Experimental groups Male Wistar rats were weaned at the age of 4 weeks and randomly assigned to the three dietary groups: control, diet 1 or diet 2. Each dietary group contained 20 animals Table 3 Additional supplements of the experimental diets (g / 100 g) Nutrient

Component

Antioxidants and vitamins

b-Carotene Flavonoids Folate Selenium Vitamin B6 Vitamin B12 Vitamin C Vitamin E

Other

Control

Diet 1

Diet 2

0.0004 0.000019 0.00153 0.00005 0.0063

0.02 0.2 0.001 0.00004 0.00172 0.00012 0.2 0.3

0.02 0.2 0.001 0.00004 0.00172 0.00012 0.2 0.3

0.15

0.15

0.002 0.944 0.232

0.002 0.944 0.232

L-Acetylcarnitin

Choline Phosphatidylcholine Phosphatidylserine Co-Q10 Thiamin Tyrosine Tryptophan

0.6 0.4 0.2 0.2 0.2 0.2 1 1

group-housed in cages of 5. The animals were following the given nutritional regimes from weaning till the termination of the experiment (at the age of 7 months), and were offered food and water ad libitum. The animals were weighed weekly (Fig. 1). Experimental cerebral hypoperfusion was imposed at the age of 4 months by a permanent bilateral occlusion of the common carotid arteries (two vessel occlusion, 2VO). Half of the animals of each dietary group were assigned for 2VO while the other half served as sham-operated control (SHAM). The surgical procedure was similar to the one reported by De Jong et al. [2]. Briefly, the rats were anesthetized by isoflurane gas. The common carotid arteries were exposed via a longitudinal cervical incision, by separating the muscles lateral to the trachea (m. sternomastoideus and m. sternohyoideus). The vessels were carefully prepared free of the nerve bundles running next to them and were permanently tied up by surgical suture. The SHAM animals received the same treatment except that the surgical suture was not tied around the arteries. The wound was then closed and the animals were monitored until they recovered. A few animals died of respiratory failure shortly after surgery, probably due to nerve damage in the neck region. The composition of the final experimental groups is presented in Table 4. After a survival time of 3 months, the rats were deeply anesthetized with sodium pentobarbital (i.p.) and shortly perfused by an ice-cold saline solution of 5.8 mM EDTA. The brains were quickly removed and stored at 280 8C for further processing.

2.3. Receptor autoradiography The left hemispheres of the brains were cut to 20 mm coronal slices on a cryostat microtome and were thawmounted on gelatin-coated glass slides. The sections were de-moisturized overnight in the presence of silica gel and stored at 280 8C. The autoradiographic receptor labeling was performed at ligand concentrations saturating all binding sites. Receptor autoradiography for the 5HT 1A receptor was described elsewhere [22]. Briefly, the procedure began by pre-incubating the sections in a solution of 0.17 M Tris– HCl, 4 mM CaCl 2 , and 0.01% ascorbic acid (pH 7.6) for 3310 min at room temperature. Then the preparations were incubated in a solution of 1.5 nM [ 3 H]8-hydroxy2(di-n-propylamino)tetralin ([ 3 H]8-OH-DPAT, 221 mCi / nM, Amersham, TRK-850) and 10 mM pargyline dissolved in pre-incubation buffer for 1 h at room temperature. Finally the sections were rinsed in ice-cold pre-incubation buffer for 3390 s, in distilled water for 5 s, and were dried by a stream of cold air. The preparations were exposed to a 3 H-sensitive film (Hyperfilm, Amersham) for 8 weeks. M 1 autoradiography followed the technical guidelines of Spencer Jr. et al. [28]. The pre-incubation was performed by dipping the slides in a solution of 20 mM Tris, 20 mM Hepes and 10 mM MgCl 2 ?4H 2 O (pH 7.5) for 3315 min at

E. Farkas et al. / Brain Research 954 (2002) 32–41

35

Fig. 1. Growth curves of the experimental groups. The curves were statistically analyzed by repeated measurement ANOVA. No significant body weight differences were found between groups. Abbreviations: 2VO: bilateral common carotid artery occlusion, C: control diet, D1: diet 1, D2: diet 2, SHAM: sham-operated control animals.

room temperature. The sections were incubated by 1.5 nM [ 3 H]pirenzipine (79.3 mCi / nM, Amersham, NET-1050) mixed in pre-incubation buffer for 1 h at 30 8C. The slides were then washed for 3360 s in ice-cold pre-incubation buffer, 135 s in distilled water and were finally dried. The exposure to a 3 H-sensitive film took 5 weeks. N-Methyl-D-aspartate (NMDA) receptor radiography developed by Jaarsma et al. [16] followed similar steps. Pre-incubation was performed in a 50 mM Tris buffer (pH 8.0) for 3315 min at room temperature. The incubation solution contained 50 nM [ 3 H]CGP39653 (40.0 mCi / nmol, Amersham, NET-780), 2.5% GDH ( L-glutamic dehydrogenase, Sigma), 1.135 mM NAD (b-nicotinamide adenine

Table 4 The experimental groups Diet / cerebral hypoperfusion

SHAM

2VO

Control Diet 1 Diet 2

P-SHAM, n59 S1-SHAM, n57 S2-SHAM, n56

P-2VO, n510 S1-2VO, n57 S2-2VO, n510

Abbreviations: 2VO: permanent bilateral occlusion of the common carotid arteries, two-vessel occulion, SHAM: sham-operated control.

dinucleotide, Sigma) and 0.05% hydrazine (Sigma). The slices were incubated for 1 h at 4 8C. Subsequently, the slides were rinsed for 3330 s in ice-cold pre-incubation buffer and for 3 s in distilled water. Finally, the preparations were dried and exposed to a 3 H-sensitive film for 6 weeks. The films were developed with a Kodak D19 developer (5.8 mM Elon, 0.317 M Na 2 SO 3 , 0.082 M C 6 H 6 O 2 , 0.358 M Na 2 CO 3 , 3.33 mM citric acid and 6.75 mM K 2 S 2 O 5 ), and were fixed with a 30% solution of Na 2 S 2 O 3 ?H 2 O. Densitometric analysis of receptor binding was performed with the use of a computer assisted image analysis system (Quantimet 600, Leica, Cambridge, UK). The optical density of receptor labeling was expressed as tissue equivalent (nCi / mg brain tissue) according to autoradiographic [ 3 H]-microscales (Amersham, RPA-506). We investigated the following six dorsal hippocampal areas at the level of bregma 23.14 [25]: the stratum oriens and the stratum radiatum of the CA1 and the CA3 region, and the inner and outer molecular layers of the dentate gyrus. Three consecutive coronal sections per animal were examined in case of each receptor type. The entire profile of the listed hippocampal regions was outlined separately,

36

E. Farkas et al. / Brain Research 954 (2002) 32–41

and the density of the signal was measured. The density values obtained from the three sections were averaged and used for further statistics. We used a two-way analysis of variance (ANOVA) test (Univariate General Linear Model) of the program SPSS for statistical analysis. For the post hoc multiple comparison analysis of dietary effect the LSD correction was chosen. The effect of 2VO in the control C dietary group alone was additionally tested by one-way ANOVA.

3. Results Fig. 2 presenting the results of the NMDA receptor autoradiography depicts a slight but consistent, cerebral hypoperfusion-related increase of receptor binding in the Control dietary group in all analyzed areas. The approximate 10% increase in receptor density was statistically significant in the CA3 region. There were no detectable diet-induced changes between the control, diet 1 and diet 2

Fig. 2. NMDA receptor density in the rat dorsal hippocampus labeled with [ 3 H]CGP39653 autoradiography. Abbreviations: 2VO: bilateral common carotid artery occlusion, DG iml: dentate gyrus, inner molecular layer, DG oml: dentate gyrus outer molecular layer, C: control diet, D1: diet 1, D2: diet 2, SHAM: sham-operated control animals. The F and P values are calculated by two-way ANOVA and only those for dietary effect are indicated. The comparison of 2VO and SHAM in the C dietary group is expressed separately by one-way ANOVA analysis.

E. Farkas et al. / Brain Research 954 (2002) 32–41

groups in any of the investigated six hippocampal areas (Fig. 2). 5HT 1A receptor labeling demonstrated a region-specific and diet-dependent increase of receptor density (Fig. 3). Experimental reduction of cerebral blood flow alone did not seem to affect 5HT 1A density values. Further, the analysis of the data revealed that the dietary supplementation could enhance receptor density in four regions of the dorsal hippocampus, namely with 12% in the stratum oriens and 17% in the stratum radiatum of the CA1 area (Fig. 3A and B), and with 20% in the inner and 18% in the outer molecular layers of the dentate gyrus (DG) (Fig. 3E and F), but not in the CA3 segment (Fig. 3C and D). The graphs in Fig. 3 further illustrate that specifically diet 2 led to a remarkable and significant increase in 5HT 1A receptor density in the given areas. The consistent

37

direction of dietary effect across hippocampal regions strengthens the perception that diet 2 could effectively elevate 5HT 1A receptor density. The findings of the M 1 receptor experiment are summarized in Fig. 4. Here, chronic cerebral hypoperfusion alone induced a tendency for enhanced M 1 receptor binding in the control group, but the difference between controlSHAM and control-2VO was not significant except the DG inner molecular layer (iml). When looking at dietary effects, the graphs are exhibiting a profound increase of M 1 receptor density due to the experimental diets. Both diet 1 and diet 2 enhanced M 1 receptor density to a similar degree when compared to control. This effect was clearly visible in all six analyzed hippocampal regions and the increase was found highest with its 40% in the CA3 stratum oriens.

Fig. 3. 5HT 1A receptor density in the rat dorsal hippocampus labeled with [ 3 H]8-OH-DPAT autoradiography. Abbreviations: 2VO: bilateral common carotid artery occlusion, DG iml: dentate gyrus, inner molecular layer, DG oml: dentate gyrus outer molecular layer, C: control diet, D1: diet 1, D2: diet 2, SHAM: sham-operated control animals. The F and P values are calculated by two-way ANOVA and only those for dietary effect are indicated. The left side of each panel is presenting all the six experimental groups while the right side is illustrating the three dietary groups corresponding with the statistical post-hoc testing for diet effect.

38

E. Farkas et al. / Brain Research 954 (2002) 32–41

Fig. 4. M 1 receptor density in the rat dorsal hippocampus labeled with [ 3 H]pirenzipine autoradiography. Abbreviations: 2VO: bilateral common carotid artery occlusion, DG iml: dentate gyrus, inner molecular layer, DG oml: dentate gyrus outer molecular layer, C: control diet, D1: diet 1, D2: diet 2, SHAM: sham-operated control animals. The F and P values are calculated by two-way ANOVA and only those for dietary effect are indicated. The comparison of 2VO and SHAM in the C dietary group is separately expressed by one-way ANOVA analysis. The left side of each panel is presenting all the six experimental groups while the right side is illustrating the three dietary groups corresponding with the statistical post-hoc testing for diet effect.

In summary, chronically reduced cerebral blood flow showed no clear effect on receptor binding to NMDA, 5HT 1A or M 1 receptors. The experimental diets markedly and selectively augmented ligand binding. Irrespective of cerebral blood flow, the diet 1 feeding regime increased the density of the M 1 receptors, while diet 2 enhanced binding to both the M 1 and 5HT 1A receptors. Neither diet 1 nor diet 2 supplementation seemed to affect the density of the NMDA receptors.

4. Discussion Our results indicated that the density of the 5HT 1A receptors remained unaffected after 3 months of cerebral hypoperfusion. This finding stands in line with the outcome of previous experiments with 2VO rats, which

demonstrated that the concentration of serotonin and serotonin turnover quickly recovered to control level after imposing 2VO [30]. Examining the effects of the dietary regime, we found that diet 1 did not affect 5HT 1A receptor density significantly, while diet 2 caused a noteworthy increase in 5HT 1A binding. Based on these results, it appears that the here used dietary (n-6) /(n-3) PUFA ratio of 1.30 was not capable of markedly influencing the affinity of 5HT 1A receptors. Conversely, the complex composition of diet 2 effectively augmented 5HT 1A binding, which may mean that the here used fatty acid concentration only in combination with other additives like those listed in diet 2 can have an effect on 5HT 1A receptor profile. Alternatively, the changes may well be unrelated to fatty acids. Some of the extra additives in diet 2 could also account for the increased 5HT 1A receptor density regardless of PUFA

E. Farkas et al. / Brain Research 954 (2002) 32–41

supplementation, but the precise identification of the active compound(s) under the given experimental conditions is not feasible. Although serotonin binding was previously reported to follow changes in membrane fluidity, the viscosity of the membrane preparations in those experiments was directly adjusted with the help of a Tris–acetate buffer, and not by PUFA supplementation [13]. The discrepancy in the methodological approach renders the comparison of the previous results and those of ours inappropriate, because the dietary PUFA effect on membrane fluidity in vivo can hardly be correlated to the more dramatic, experimental viscosity manipulations in vitro. Our present study is the first to our best knowledge to tackle the question of dietary PUFA effect on 5HT 1A binding. The results suggest that PUFA intake in the given concentration alone does not exert a noticeable influence on 5HT 1A sensitivity. The present data indicate that the density of the M 1 receptors under control dietary circumstances slightly but consistently increased in the six analyzed hippocampal regions. In contrast, experimental cerebral hypoperfusion imposed by 2VO was previously described to cause a notable reduction of cholinergic function, determined by pharmacological challenges in learning tasks or muscarinic receptor binding assays [5,20,29]. However, the reduced muscarinic receptor binding in 2VO rats is most probably not irreversible, because M 1 stimulation could restore the 2VO-impaired learning capacity of rats as shown by another study [20]. These observations suggest that the functional binding characteristics rather than the expression of the receptor proteins are affected in cerebral hypoperfusion, which is also supported by the finding that the mRNA level for the M 1 receptor did not parallel the functional decline of M 1 binding in old animals [23]. Our data, that M 1 binding was moderately elevated due to cerebral hypoperfusion can thus reflect the dynamic nature of M 1 sensitivity, and may be associated with a compensatory mechanism in chronic cerebral hypoperfusion. The discrepancy with the previous 2VO study that reported diminishing binding to muscarinic receptors [29] may lie in the ligand QNB used there, which does not exclusively bind to M 1 but to other muscarinic receptors, as well. In contrast, we labeled solely the M 1 receptors in the 2VO paradigm, which may indicate that other muscarinic receptors than M 1 could be responsible for the decreased muscarinic receptor density observed by Tanaka et al. [29]. Both diet 1 and diet 2 in our experimental design led to a remarkable increase in M 1 binding. The enhanced M 1 density most probably reflects a dietary PUFA effect, since both diet 1 and diet 2 raised M 1 binding to an equal degree. It must be mentioned though that both diets contained an increased amount of antioxidants, as well. Nevertheless, our next set of experiments demonstrates that the diets significantly reduced n-6 / n-3 ratio in the brain (manuscript in preparation), which supports the hypothesis of PUFA-related receptor changes.

39

The initial M 1 density difference between SHAM and 2VO in the dietary control dietary group vanished in diet 1 and diet 2 supplemented groups. It seems likely that the effect of the dietary PUFA intake dominated over and concealed the 2VO-induced difference, which implies the possibility that the binding had reached a physiological maximum. An increased ligand binding to receptors in comparable experiments has traditionally been attributed to structural changes in the neuronal membranes and increased membrane fluidity [1,7,9,10,13,26]. Theoretically, a physical change in the state of the lipid phase of the membrane should then equally influence the availability of various kinds of receptors to their ligands. Curiously, the dependence of ligand binding on membrane viscosity of a variety of receptors was shown to differ. What is more, ligand binding to muscarinic receptors remained unaffected at increasing membrane viscosity [14]. Our results also show that receptors even from the same family (M 1 and 5HT 1A being both G protein coupled receptors) can react different to the dietary supply of PUFAs. Earlier evidence that muscarinic receptors could not be affected by changes in membrane viscosity, and our results that the M 1 receptor alone seemed to increase its ligand binding due to dietary PUFA intake asks for another, more specific explanation than physical alterations in membrane structure alone. A biologically more active and dynamic process like the biochemical regulation of the amount of available, sensitive receptor binding sites may offer an explanation. The density of binding sites can be regulated by protein kinase-mediated phosphorylation, which can desensitize the membrane bound receptors leading to reduced ligand binding. In case of the M 1 receptor, the involvement of protein kinase C (PKC) was proposed among other mediators of the desensitization process by phosphorylation [11,34]. The role of PKC in regulating receptor sensitivity seems rather intriguing for the interpretation of our M 1 binding data for the reason that DHA and EPA have been reported to be able to inhibit PKC activity [21,27]. By integrating the data from these publications we can postulate that if PKC activity, thus M 1 phosphorylation is inhibited by n-3 PUFAs, it should logically lead to an increased concentration of dephosphorylated, active M 1 binding sites. This line of reasoning bears functional importance in the view of our dietary paradigm, where the ratio of DHA and EPA intake was increased, and the density of M 1 receptors appeared to be simultaneously higher. Based on these arguments, we propose the theory that the enhanced M 1 binding detected here could possibly be caused by a process like a DHA / EPA-mediated inactivation of protein kinases. 5. Conclusions We have shown here that chronic experimental cerebral hypoperfusion could selectively modify the density of the

40

E. Farkas et al. / Brain Research 954 (2002) 32–41

excitatory M 1 receptors, and did not affect the inhibitory 5HT 1A receptors. The increase in M 1 receptor binding found here could reflect a process compensating an altered neuronal metabolism due to the suboptimal blood supply to the brain. The dietary PUFA supplementation in our experiment effectively increased hippocampal M 1 receptor density, enhanced 5HT 1A density only in combination with extra additives and did not affect the NMDA receptors. Comparing these results to previous reports that investigated the role of membrane fluidity in receptor ligand binding, we conclude that the PUFA effect demonstrated here is most probably only partially related to membrane viscosity changes. Rather, the role of other cellular regulatory mechanisms, such as the adjustment of receptor sensitivity by phosphorylation can be considered. As the strongest PUFA effect concerned the M 1 receptor, dietary PUFA intake may have further implications in learning and memory processes and may be relevant in the moderation of age-related cognitive dysfunction and dementia.

Acknowledgements The research presented here was financed by the Numico Dementia project No. 98 / 006.

References [1] S. Chalon, S. Delion-Vancassel, C. Belzung, D. Guilloteau, A.M. Leguisquet, J.C. Besnard, G. Durand, Dietary fish oil affects monoaminergic neurotransmission and behavior in rats, J. Nutr. 128 (12) (1998) 2512–2519. [2] G.I. De Jong, E. Farkas, C.M. Stienstra, J.R. Plass, J.N. Keijser, J.C. de la Torre, P.G.M. Luiten, Cerebral hypoperfusion yields capillary damage in the hippocampal CA1 area that correlates with spatial memory impairment, Neuroscience 91 (1) (1999) 203–210. [3] J. De Vries, N-3 polyunsaturated fatty acids, in: A.W. Teelken, J. Korf (Eds.), Neurochemistry, Plenum Press, New York, 1997, pp. 125–136. [4] M.C. de Wilde, E. Farkas, M. Gerrits, A.J. Kiliaan, P.G.M. Luiten, The effect of n-3 polyunsaturated fatty acid-rich diets on cognitive and cerebrovascular parameters in chronic cerebral hypoperfusion, Brain Res., in press. [5] T. Egashira, F. Takayama, Y. Yamanaka, Effects of bifemelane on muscarinic receptors and choline acetyltransferase in the brains of aged rats following chronic cerebral hypoperfusion induced by permanent occlusion of bilateral carotid arteries, Jpn. J. Pharmacol. 72 (1) (1996) 57–65. [6] E. Farkas, P.G.M. Luiten, Cerebral microvascular pathology in ageing and Alzheimer’s disease, Prog. Neurobiol. 64 (6) (2001) 575–611. [7] G. Freund, T.R. Brophy III, J.D. Scott, Membrane fluidization increases low-affinity muscarinic receptor binding in brain: changes with ageing, Exp. Gerontol. 21 (1) (1986) 37–44. [8] S. Gamoh, M. Hashimoto, K. Sugioka, M. Shahdat Hossain, N. Hata, Y. Misawa, S. Masumura, Chronic administration of docosahexaenoic acid improves reference memory-related learning ability in young rats, Neuroscience 93 (1) (1999) 237–241.

[9] C.M. Gelbmann, W.E. Muller, Chronic treatment with phosphatidylserine restores muscarinic cholinergic receptor deficits in the aged mouse brain, Neurobiol. Ageing 13 (1) (1992) 45–50. [10] C. Ghosh, R.M. Dick, S.F. Ali, Iron / ascorbate-induced lipid peroxidation changes membrane fluidity and muscarinic cholinergic receptor binding in rat frontal cortex, Neurochem. Int. 23 (5) (1993) 479–484. [11] K. Haga, K. Kameyama, T. Haga, U. Kikkawa, K. Shiozaki, H. Uchiyama, Phosphorylation of human m1 muscarinic acetylcholine receptors by G protein-coupled receptor kinase 2 and protein kinase C, J. Biol. Chem. 271 (5) (1996) 2776–2782. [12] M. Hashimoto, S. Hossain, H. Yamasaki, K. Yazawa, S. Masumura, Effects of eicosapentaenoic acid and docosahexaenoic acid on plasma membrane fluidity of aortic endothelial cells, Lipids 34 (12) (1999) 1297–1304. [13] D.S. Heron, M. Shinitzky, M. Hershkowitz, D. Samuel, Lipid fluidity markedly modulates the binding of serotonin to mouse brain membranes, Proc. Natl. Acad. Sci. USA 77 (12) (1980) 7463–7467. [14] M. Hershkowitz, D.S. Heron, D. Samuel, M. Shinitzky, The modulation of protein phosphorylation and receptor binding in synaptic membranes by changes in lipid fluidity: implications for ageing, Prog. Brain Res. 56 (1982) 419–434. [15] L.A. Horrocks, Y.K. Yeo, Health benefits of docosahexaenoic acid (DHA), Pharmacol. Res. 40 (3) (1999) 211–225. [16] D. Jaarsma, J.B. Sebens, J. Korf, Localization of NMDA and AMPA receptors in rat barrel field, Neurosci. Lett. 133 (2) (1991) 233–236. [17] S. Kalmijn, Fatty acid intake and the risk of dementia and cognitive decline: a review of clinical and epidemiological studies, J. Nutr. Health. Aging 4 (4) (2000) 202–207. [18] S. Kalmijn, L.J. Launer, A. Ott, J.C. Witteman, A. Hofman, M.M. Breteler, Dietary fat intake and the risk of incident dementia in the Rotterdam Study, Ann. Neurol. 42 (5) (1997) 776–782. [19] A. Komatani, K. Yamaguchi, Y. Sugai, T. Takanashi, M. Kera, M. Shinohara, S. Kawakatsu, Assessment of demented patients by dynamic SPECT of inhaled xenon-133, J. Nucl. Med. 29 (10) (1988) 1621–1626. [20] Y. Murakami, M. Ikenoya, K. Matsumoto, H. Li, H. Watanabe, Ameliorative effect of tacrine on spatial memory deficit in chronic two-vessel occluded rats is reversible and mediated by muscarinic M1 receptor stimulation, Behav. Brain. Res. 109 (1) (2000) 83–90. [21] S.S. Nair, J. Leitch, M.L. Garg, N-3 polyunsaturated fatty acid supplementation alters inositol phosphate metabolism and protein kinase C activity in adult porcine cardiac myocytes, J. Nutr. Biochem. 12 (1) (2001) 7–13. [22] C. Nyakas, B.J. Oosterink, J.N. Keijser, K. Felszeghy, G.I. De Jong, J. Korf, P.G.M. Luiten, Selective decline of 5-HT1A receptor binding sites in rat cortex, hippocampus and cholinergic basal forebrain nuclei during ageing, J. Chem. Neuroanat. 13 (1) (1997) 53–61. [23] N. Ogawa, M. Asanuma, Y. Kondo, S. Nishibayashi, A. Mori, Reduced choline acetyltransferase activity and muscarinic M1 receptor levels in aged Fisher 344 rat brains did not parallel their respective mRNA levels, Brain Res. 658 (1–2) (1994) 87–92. [24] T. Ohnishi, H. Hoshi, S. Nagamachi, S. Jinnouchi, L.G. Flores II, S. Futami, K. Watanabe, High-resolution SPECT to assess hippocampal perfusion in neuropsychiatric diseases, J. Nucl. Med. 36 (1995) 1163–1169. [25] G. Paxinos, C. Watson, The Rat Brain in Stereotaxic Coordinates, Academic Press, London, 1986. [26] K. Scheuer, S. Stoll, U. Paschke, R. Weigel, W.E. Muller, N-MethylD-aspartate receptor density and membrane fluidity as possible determinants of the decline of passive avoidance performance in ageing, Pharmacol. Biochem. Behav. 50 (1) (1995) 65–70. [27] H.F. Seung Kim, E.J. Weeber, J.D. Sweatt, A.L. Stoll, L.B. Marangell, Inhibitory effects of omega-3 fatty acids on protein kinase C activity in vitro, Mol. Psychiatry 6 (2) (2001) 246–248. ´ [28] D.G. Spencer Jr., E. Horvath, J. Traber, Direct autoradiographic determination of M1 and M2 muscarinic acetylcholine receptor

E. Farkas et al. / Brain Research 954 (2002) 32–41

[29]

[30]

[31]

[32]

[33]

distribution in the rat brain: relation to cholinergic nuclei and projections, Brain Res. 380 (1) (1986) 59–68. K. Tanaka, N. Ogawa, M. Asanuma, Y. Kondo, M. Nomura, Relationship between cholinergic dysfunction and discrimination learning disabilities in Wistar rats following chronic cerebral hypoperfusion, Brain Res. 729 (1) (1996) 55–65. K. Tanaka, N. Wada, N. Ogawa, Chronic cerebral hypoperfusion induces transient reversible monoaminergic changes in the rat brain, Neurochem. Res. 25 (2) (2000) 313–320. T. Terano, S. Fujishiro, T. Ban, K. Yamamoto, T. Tanaka, Y. Noguchi, Y. Tamura, K. Yazawa, T. Hirayama, Docosahexaenoic acid supplementation improves the moderately severe dementia from thrombotic cerebrovascular diseases, Lipids 34 (1999) S345–S346. M. Treen, R.D. Uauy, D.M. Jameson, V.L. Thomas, D.R. Hoffman, Effect of docosahexaenoic acid on membrane fluidity and function in intact cultured Y-79 retinoblastoma cells, Arch. Biochem. Biophys. 294 (2) (1992) 564–570. H. Tsukada, T. Kakiuchi, D. Fukumoto, S. Nishiyama, K. Koga,

[34]

[35]

[36] [37]

41

Docosahexaenoic acid (DHA) improves the age-related impairment of the coupling mechanism between neuronal activation and functional cerebral blood flow response: a PET study in conscious monkeys, Brain Res. 862 (1–2) (2000) 180–186. M.G. Waugh, R.A. Challiss, G. Berstein, S.R. Nahorski, A.B. Tobin, Agonist-induced desensitization and phosphorylation of m1-muscarinic receptors, Biochem. J. 338 (Pt. 1) (1999) 175–183. S. Yehuda, R.L. Carasso, Modulation of learning, pain thresholds, and thermoregulation in the rat by preparations of free purified alpha-linolenic and linoleic acids: determination of the optimal omega 3-to-omega 6 ratio, Proc. Natl. Acad. Sci. USA 90 (21) (1993) 10345–10349. S. Yehuda, S. Rabinovitz, R.L. Carasso, D.I. Mostofsky, Fatty acids and brain peptides, Peptides 19 (2) (1998) 407–419. S. Yehuda, S. Rabinovitz, D.I. Mostofsky, Essential fatty acids are mediators of brain biochemistry and cognitive functions, J. Neurosci. Res. 56 (6) (1999) 565–570.