Behavioral effects of dietary cholesterol in rats tested in experimental models of mild stress and cognition tasks

www.e l sev i e r. com/ l oca te /eu roneu ro

European Neuropsychopharmacology (2008) 18, 462–471

Behavioral effects of dietary cholesterol in rats testedin experimental models of mild stress andcognition tasksVincenzo Micale a,b, Giovanni Scapagnini c, Claudia Colombrita b,Carmen Mazzola a, Daniel L. Alkon b, Filippo Drago a,⁎

a Department of Experimental and Clinical Pharmacology, University of Catania, Catania, Italyb Blanchette Rockefeller Neuroscience Institute, West Virginia University, Rockville, MD, United Statesc Institute of Neurological Sciences, Italian National Research Council, CNR, Catania, Italy

Received 4 February 2007; received in revised form 3 September 2007; accepted 21 November 2007

⁎ Corresponding author. Departmentfax: +39 095 7384238.

E-mail address: [email protected] (F. Dr

0924-977X/$ - see front matter © 200doi:10.1016/j.euroneuro.2007.11.006

Abstract

Abnormalities in serum cholesterol levels of patients with mood disorders have been identified inepidemiological studies. However, evidence for an influence of dietary cholesterol on behavioralmodels is poor. Here, we investigated the behavioral changes of Wistar male rats fed a 2%cholesterol-enriched diet for 2 months in experimental models of depression and anxiety, such asthe forced swim test (FST) paradigm and the novelty-induced grooming sampling test (NGT). Thecorrelation between behavioral depression and impaired cognitive capacity was also examinedtesting rats in the Morris watermaze (MWM) task one day after the FST. Different groups of rats fedvarious dietary regimens, were subjected to acute or repeated treatment (14 days) withclomipramine hydrochloride (50 or 25 mg/kg), diazepam (1 mg/kg) or with the peripheralbenzodiazepine receptors (PBRs) antagonist, isoquinoline PK11195 (1 mg/kg) injected intraper-itoneally (i.p.). Rats fed the cholesterol-enriched diet showed a significant decrease of groomingscore in the NGTand of immobility time in the FST in comparison to animals fed a standard diet.Furthermore, the anxiolytic and antidepressant effects of diazepam and clomipramine were notaffected by the different diets. Only after repeated treatment, PK11195 impaired theperformance of animals fed a standard diet in the FST, and exhibited an anxiolytic-like profilein animals fed either the cholesterol-enriched or the standard diet. The improved performance inthe FSTwas followed by a better learning performance in the acquisition phase of the MWM. Theseresults suggest that effects of cholesterol-enriched diet on the behavioral reaction of rats inexperimental models of mild stress may involve PBRs. They deserve attention in order to clarifythe clinical correlation between plasma cholesterol levels and mood disorders in humans.© 2007 Elsevier B.V. and ECNP. All rights reserved.

KEYWORDSCholesterol;Forced swim test;Antidepressant activity;Memory;Peripheral benzodiazepinereceptor

of Experimental and Clinical Pharmacology, Viale A. Doria 6, 95125 Catania, Italy. Tel.: +39 095 7384236;

ago).

7 Elsevier B.V. and ECNP. All rights reserved.

mailto:[email protected]

http://dx.doi.org/10.1016/j.euroneuro.2007.11.006

463Cholesterol and mood disorders

1. Introduction

Several studies (for reviews, see Papakostas et al., 2004a;Huang and Chen, 2005) have found low serum cholesterollevels (b160 mg/dl), including triglycerides (TG), total cho-esterol (TC) and high-density lipoprotein cholesterol (HDL-C)in patients with depression, anxiety, co-morbid depression andanxiety, suicidal ideation and current or past suicidal behavior.This findingwas not entirely believed to be the consequence ofdepression-related malnutrition, since confounding by weightchange and parameters of nutritional status was excluded insome studies (Morgan et al., 1993; Horsten et al., 1997).

Thebiochemical relationship betweencholesterol andmooddisorders is poorly understood. Cholesterol is an essential cellmembrane component and may be involved in the develop-ment, function and stability of synapse, modulating membraneprotein function (Pfrieger, 2003). The brain possesses a highcontent of lipids, and cerebral cholesterol levels are quitestable. It is not clear if dietary cholesterol directly influencesbrain cholesterol concentrations (Howland et al., 1998). Ingeneral, however, there is evidence that brain lipid content isinfluenced by dietary lipid intake and alterations in brain lipidsmakeup, in turn, cause behavioral and other neuropsychicconsequences (Oner et al., 1991; Farquharson et al., 1992;Young, 1993; Kaplan et al., 1994). In particular, rats fed dietarycholesterol showed increased cholesterol synthesis in thehippocampus (Koudinov and Koudinova, 2003).

Serotonin (5-HT) neurotransmission in the central nervoussystem is involved in the pathogenesis of depression. Adecrease in brain concentrations of 5-HT and its majormetabolite, 5-hydroxyindolacetic acid (5-HIAA) is commonlyobserved in animals and in patients experiencing stress anddepression, suggesting a dysfunction of the 5-HT system(Spreux-Varoquaux et al., 2001). Interestingly, a decrease inplasma 5-HT levels is observed in depressed patients with lowlevels of plasma lipids (Almeida-Montes et al., 2000). In vitrostudies have demonstrated that a decreased cholesterolconcentration in neuronal membranes could reduce 5-HTneurotransmission by effects on both pre- and postsynapticsites (Heron et al., 1980; Engelbergb, 1992; Scanlon et al.,2001). Furthermore, preclinical studies in monkeys havesupported the hypothesis of a link between low serumcholesterol concentrations, reduced central 5-HT neuro-transmission, and impulsive and aggressive behavior (Mul-doon et al., 1992; Kaplan et al., 1994). Recently, majordepressive disorder (MDD) patients with high-cholesterollevels exhibited poor response to fluoxetine or nortriptylinetreatment, probably due to impairment of 5-HTergic trans-mission (Sonawalla et al., 2002; Papakostas et al., 2003a,b).

Among other biological actions, the role as a cofactor forsignaling molecules has been described for cholesterol. It is aprecursor for steroid hormones synthesized in the brain andin the peripheral steroidogenic tissues (Claudepierre andPfrieger, 2003). For a number of these so-called “neuroster-oid”, the evidence exists that they are synthesized de novo inthe brain from sterol precursors independently from periph-eral endocrine sources (Rupprecht, 2003). The biosynthesisof steroids is promoted by activation of peripheral benzo-diazepine receptors (PBRs) inducing cholesterol to movefrom cellular stores into mitochondria with its increasedavailability for cytochrome P450 side-chain cleavage enzyme(P450scc), which catalyses the synthesis of pregnenolone

(Krüeger and Papadopulos, 1990). The involvement of PBRsin experimental models of anxiety, depression and cogni-tive deficit has been studied using specific ligands, such as4′-chlorodiazepam (Ro5-4864), its isoquinoline carboxamide(PK11195) and the 2-phenyl-imidazo[1,2-a]pyridine deriva-tives (Holmes and Drugan, 1991; Bitran et al., 2000; Gavioliet al., 2003; Serra et al., 2004). Furthermore, acute andchronic stress may influence the density of PBRs in varioustissues in an opposite manner. Acute stress thus is followedby an increased brain PBRs density and allopregnanoloneconcentration (Avital et al., 2001; Droogleever Fortuynet al., 2004; Serra et al., 2004); whereas repeated swimstress (Burgin et al., 1996), repeated foot shock (Druganet al., 1986) and food deprivation stress (Weizman et al.,1990) lead to a reduction in PBRs density.

In the light of the epidemiological and clinical datadescribed above (for reviews, see Papakostas et al., 2004a;Huang and Chen, 2005), the aim of the present study was toassess the effects of cholesterol-enriched diet on experi-mental animal models of mood disorders. For this purposerats fed diet enriched with 2% cholesterol for 2 months asdescribed by Koudinov and Koudinova (2003), were testedin experimental paradigms used to value depression- oranxiety-like behavior as the forced swim test (FST) procedureor the novelty-induced grooming sampling test (NGT),respectively. Since cholesterol is the precursor of steroidsthrough a biochemical process involv ing PBRswhich expression with the steroid synthesis is also influencedby acute or chronic stressors (Burgin et al., 1996; Avital et al.,2001), we also investigated whether the effects of choles-terol-enriched diet may influence the action of PK11195, aPBRs antagonist administered acutely or chronically.

Depression is commonly associated to a cognitive deficitthat can be observed as impaired learning and memory and adirect relationship between relief of depression andimprovement of the associated cognitive deficit has beendescribed (Sternberg and Jarvik, 1976; Willner, 1984;Bulbena and Berrios, 1993). The 5-HT system plays a role incognitive processes, particularly in learning and memory(Meneses, 1999) and some antidepressants as tricyclicsand selective serotonin reuptake inhibitors (SSRI) enhancelearning and memory in animal behavioral tests (Kumar andKulkarni, 1996; Fujishiro et al., 2002). Since PBRs receptorsare also involved in learning and memory processes and atthe same time given the relationship between cholesteroland 5-HTergic system (Heron et al., 1980; Holmes andDrugan, 1991; Engelbergb, 1992; Scanlon et al., 2001), theseprompted us to assess whether dietary cholesterol andPK11195 may affect the performance of rats in the Morriswater maze task after being tested in the FST procedure.

2. Materials and methods

2.1. Animals

Male rats of the Wistar strain (purchased from Charles River, USA)weighing 120–150 g at the beginning of the experiments were used.Rats were randomly assigned to one of the two dietary regimens.Animals were housed two into a cage under standard environmentalconditions: constant temperature of 23±1 °C, 60% humidity, 12-hlight/dark cycle (lights on between 08.00 and 20.00) with food and tapwater ad libitum.

464 V. Micale et al.

Dietary regimes were selected based on the results of Koudinovand Koudinova (2003), furthermore confirmed by Dufour et al.(2006) and applied to the rats as follows:

1. Standard (STD) diet (based on commercial rodent pellets with acholesterol content of 48 ppm (parts per million), Global Diet2018, Harlan Teklad, USA), ad libitum.

2. Cholesterol-enriched (CHO) diet (based on Global Diet 2018enriched with 2% cholesterol, TD.01383; Harlan Teklad, USA), adlibitum.

The STD and CHO diets also differ in terms of fat composition(STD: 5.0% by weight; CHO: 5.7% by weight) and calorie content(STD: 3.3 kcal/g; CHO: 3.1 kcal/g).

All procedures were conducted according to the NationalInstitutes of Health Animal Care and Use Committee guidelines,and they were approved by the Ethical Committee of the Institute.Efforts were made to minimize animal suffering and to reduce thenumber of animals used and duration of observation required toobtain consistent data.

2.2. Forced swim test (FST) procedure

For the FST procedure, rats were individually forced to swim insidevertical plexiglas cylinders containing 25 cm of water maintained at25 °C (Porsolt et al., 1978). After 15 min in the water they wereremoved and allowed to dry for 15 min in a heated container beforebeing returned to their home cages. They were replaced in thecylinders 24 h later and the total duration of immobility wasmeasured during a 5-min test. A rat was judged to be immobilewhenever it remained passively floating in the water in a slightlyhunched but upright position, its head just above the surface.

2.3. Novelty-induced grooming sampling test (NGT)

The NGTwas executed between 15.00 and 18.00 h, always under thesame environmental conditions according to the method describedelsewhere (Drago et al., 1980). The rats were placed individuallyinto plexiglas boxes (24×12×24 cm) in a low noise room. After 1 minof adaptation, the behavior of the rats was sampled every 15 s andthe occurrence of grooming was recorded in a session of 30 min. Theoccurrence of the following single elements of grooming was scoredas grooming: washing (vibrating movements of the fore paws in frontof the snout and liking of the same paws leading to a series of strokesalong the snout and semicircular movements over the top of thehead), scratching (scratching of the body by one of the limbs),licking (licking of the body fur, limbs and tail), and genital grooming(licking of genital area). Stretching and yawning episodes were notrecorded. Grooming behavior of all animals was recorded on a tapeusing a video camera (Hitachi videocam) and then scored in monitordisplay by two independent observers. The mean score of the twoobservations was used for statistical analysis. Data were acceptedwithin an inter-rate variability of maximum 10%.

2.4. Morris water maze task

The Morris water maze (MWM) consisted of a black circular pool(156 cm in diameter, 100 cm high), theoretically divided into fourequal quadrants for analysis purposes. The pool was filled to a depthof 60 cm with water (24±1 °C) and located in an environment withdistinct extra-maze cues. A 10-cm2 transparent square platform wasplaced 1.5 cm below the water surface. Avideo camera was mountedabove the center of the tank and all trials were recorded. The datawas analyzed by computerized image analysis system.

The water maze test used here was an adaptation of the MWMtask (Morris, 1984). For 4 days each rat was submitted to daily

session of four trials per day to find the submerged platform locatedin the center of a quadrant of the tank during this period of training.On each trial, the rats were placed in the water, facing the edge ofthe tank, in one of the four standard start locations (N, S, W, and E).The order of the start locations was random within daily sessions.The rats were allowed to swim 120 s and the latency to find theplatform was recorded. Once the rats located the platform, theywere permitted to remain on it for 10 s. If the rats did not find theplatform within this time, they were placed on it for 20 s by theexperimenter. After each trial, the rats were removed, dried in atowel and putted back in their home cages.

Acquisition probe trial used here was according to the methoddescribed elsewhere (Schrijver et al., 2004). Briefly, on days 2 and 4a probe trial of 60 s with the platform removed was administered30 min prior the normal training trials to assess memory formationfor the trained platform position. Percentage of the total time spentin the previous target quadrant was analyzed.

The day after the acquisition training, a reversal-training phasewas run in which animals were trained for 2 days (four trails per day)to find the hidden platform now located in the diagonally opposite(reversed) quadrant to the location in the acquisition phase. Latencyto find platform was determined in each trial. On the day after, allanimals were submitted to a reversal probe. Each rat was placed inthe water diagonally opposite the target quadrant and allowed toswim for 60 s. The time spent in the opposite platform quadrant areawas analyzed.

2.5. Drugs

All drugs in this study were supplied from Sigma (USA). Clomipraminehydrochloride was freshly diluted in physiological saline and injectedintraperitoneally (i.p.) in a dose of 25 or 50 mg/kg. Diazepam wasprepared as a suspension in physiological saline containing Tween80 (0.1%) and injected i.p. in a dose of 1 mg/kg. The 1-(2-chloro-phenyl)-N-methyl-N-(1-methylpropyl)-1-isoquinoline carboxamidecompound (PK11195), dissolved in absolute ethanol and freshlydiluted in saline containing 0.5% Tween 80 to the adequateconcentrations as described by Gavioli et al. (2003), was adminis-tered i.p. in a dose of 1 mg/kg. All these drugs were injected in atotal volume of 0.1 ml/100 g body weight.

2.6. Experimental design

Animals were fed STD or CHO diet for 2 months. At the end of thedietary regimen, different groups of rats (n=10 per group)were testedin the NGT or in the FST. Animals subjected to the NGT received onlyone injection i.p. of diazepam (1 mg/kg) or PK11195 (1 mg/kg) 30 minprior to behavioral testing. Two groups of control animals wereinjected i.p. with diazepam vehicle or PK11195 vehicle and subjectedto the same behavioral procedure. The rats subjected to the FSTreceived three i.p. injections of clomipramine (50 mg/kg) or PK11195(1 mg/kg) made 24, 5 and 1 h prior to the behavioral testing. Twogroups of control animals were injected i.p.with clomipramine vehicleor PK11195 vehicle, following the same procedure. The dose ofclomipramine was selected based on results of previous experiments(Drago et al., 2001; Consoli et al., 2005; Micale et al., 2006).Furthermore, this dose was considered as the minimum effective doseof clomipramine in the FST (Porsolt et al., 1978). The day after the FSTthe rats were subjected to the MWM task, as described in Section 2.4.

In the second experiment, different groups of rats fed STD or CHOdiet for 2 months, during the last 14 days of the dietary regimen,received a repeated i.p. treatment with PK11195 (1 mg/kg/day),clomipramine (25 mg/kg/day) or diazepam (1 mg/kg/day). Controlanimals received the vehicles used for the preparation of eachcompound. The different dose of clomipramine used for the chronictreatment was selected based on results of previous experiments, toavoid the anticholinergic side effects of the compound (Fujishiroet al., 2002).

Table 2 Effects of acute administration of diazepam(1 mg/kg) or PK11195 (1 mg/kg) on grooming response ofrats fed a different diet (standard diet or diet enrichedwith 2% cholesterol) for 2 months in the novelty-inducedgrooming sampling test (NGT)

Treatment Grooming score

STD+vehicle (20) 38.3±6.1STD+diazepam (10) 17.0±3.0 a

STD+PK11195 (10) 40.1±2.9CHO+vehicle (20) 24.2±2.7 b

CHO+diazepam (10) 16.3±1.7 a

CHO+PK11195 (10) 24.5±3.0 b

Abbreviations STD and CHO denote different groups of rats fed astandard diet or a standard diet enriched with 2% cholesterol for2 months, respectively. Diazepam (1 mg/kg), PK11195 (1 mg/kg)or their vehicles were administered i.p. 30 min prior tobehavioral testing. In this experiment, two groups of controlanimals were injected i.p. with diazepam or PK11195 vehicles. Assimilar results were obtained from these two control groups, datawere conglomerated.Values are mean±S.E.M. In parentheses the number of animalsper each group is indicated.a Significantly different as compared to STD+vehicle group

(Pb0.01, Dunnett's two-tailed test for multiple comparisons).b Significantly different as compared to STD+vehicle group

(Pb0.05, Dunnett's two-tailed test for multiple comparisons).


The rats subjected to the NGT received the last injection ofPK11195, diazepam or vehicles 30 min prior to the behavioral test.The rats subjected to the FSTreceived the last injection of PK11195,clomipramine or vehicles 1 h prior the behavioral testing. The dayafter the FST, the rats were subjected to the MWM task as describedbefore.

2.7. Statistical analysis of data

All data were analyzed using the two-way ANOVA (dietary regimenand treatment), followed by the post-hoc Dunnett's two-tailed testfor multiple comparisons. Alpha (α) was set at 0.05.

3. Results

At the end of the second month of dietary regimen, the weightof the rats fed a standard diet and cholesterol-enriched dietwas 465±10 g and 537±13 g, respectively. Statistical analysisshowed any significant variation in the post-diet average groupweights. Rat food intake was monitored daily, finding a con-sumption of 21.0±2.2 g (mean±S.D.) in rats fed a STD diet and23.2±1.8 g (mean±S.D.) in rats fed a CHO diet. Statisticalanalysis did not reveal any significant differences in food intakebetween the two groups (PN0.05).

In the first experiment, rats fed the CHO diet and treatedwith vehicle exhibited a decreased duration of immobility timein the FST procedure (Table 1). However, the different dietregimes did not affect the antidepressant effect of clomipra-mine as both STD and CHO groups treated with the antidepres-sant drug showed a similar decrease in the immobility time incomparison to both groups of rats injected with vehicle.Furthermore, the acute treatment with PK11195 did not modifythe immobility time of STD and CHO animals tested in the FST

Table 1 Effects of acute administration of clomipramine(50 mg/kg) or PK11195 (1 mg/kg) on immobility time of ratsfed a different diet (standard diet or diet enriched with 2%cholesterol) for 2 months in the forced swim test (FST)procedure

Treatment Duration of immobility

STD+vehicle (20) 192.9±10.2STD+clomipramine (10) 110.5±16.1 a


CHO+clomipramine (10) 99.2±9.1 a

CHO+PK11195 (10) 142.5±18.9 b

Abbreviations STD and CHO denote rats fed a standard diet or astandard diet enriched with 2% cholesterol for 2 months,respectively. Clomipramine (50 mg/kg), PK11195 (1 mg/kg) ortheir vehicles were administered i.p. 24, 5 and 1 h prior tobehavioral testing. In this experiment, two groups of controlanimals were injected i.p. with clomipramine or PK11195vehicles. As similar results were obtained from these two controlgroups, data were conglomerated.Values are mean±S.E.M. of time measures expressed in s. Inparentheses the number of animals per each group is indicated.a Significantly different as compared to STD+vehicle group

(Pb0.01, Dunnett's test two-tailed for multiple comparisons).b Significantly different as compared to STD+vehicle group


procedure. Difference in body weight dependent on the variousdietary regimens, did not affect the immobility time of rats. Asshown in Table 2, rats fed the CHO diet and treated with vehicleexhibited a significant decrease of grooming response to novelenvironment in the NGT. Acute treatment with PK11195 failed

Figure 1 Effects of acute administration of clomipramine(50 mg/kg) or PK11195 (1 mg/kg) in rats fed a different dietduring the acquisition phase in Morris water maze (MWM) task,made the day after the FST procedure. Abbreviations STD andCHO denote different groups of rats fed a standard diet or astandard diet enriched with 2% cholesterol for 2 months,respectively. Values are mean±S.E.M. of time measuresexpressed in s. In this experiment, two groups of control animalswere injected i.p. with clomipramine or PK11195 vehicles. Assimilar results were obtained from these two control groups,data were conglomerated. aSignificant difference vs. vehicle- orPK11195-treated STD animals (Pb0.05, Dunnett's two-tailed testfor multiple comparisons).

Table 3 Effects of repeated administration (14 days) ofclomipramine (25 mg/kg/day) or PK11195 (1 mg/kg/day) onimmobility time of rats fed a different diet (standard diet ordiet enriched with 2% cholesterol) for 2 months in the forcedswim test (FST) procedure

Treatment Duration of immobility

STD+vehicle (20) 195.05±14.53STD+clomipramine (10) 124.10±8.01 a


CHO+clomipramine (10) 122.15±6.00 a

CHO+PK11195 (10) 172.25±16.04

Abbreviations STD and CHO denote different groups of rats fed astandard diet or a standard diet enriched with 2% cholesterol for2 months, respectively. Clomipramine (25 mg/kg/day), PK11195(1 mg/kg/day) or their vehicles were administered i.p. for 14days prior to behavioral testing. In this experiment, two groups ofcontrol animals were injected i.p. with clomipramine or PK11195vehicles. As similar results were obtained from these two controlgroups, data were conglomerated.Values are mean±S.E.M. of time measures expressed in s. Inparentheses the number of animals per each group is indicated.a Significantly different as compared to STD+vehicle or

PK11195 groups (Pb0.01, Dunnett's two-tailed test for multiplecomparisons) and CHO+vehicle or PK11195 groups (Pb0.05,Dunnett's two-tailed test for multiple comparisons).b Significantly different as compared to STD+vehicle group

(Pb0.05, Dunnett's two-tailed test for multiple comparisons)and STD+PK11195 group (Pb0.01, Dunnett's two-tailed test formultiple comparisons).

Figure 2 Effects of acute administration of clomipramine(50 mg/kg) or PK11195 (1 mg/kg) in rats fed a different dietduring the 1st and 2nd probe trials made 30min prior the 2nd and4th training in MWM task. Abbreviations STD and CHO denotedifferent groups of rats fed a standard diet or a standard dietenriched with 2% cholesterol for 2 months, respectively. Valuesare mean±S.E.M. of time spent in quadrant target expressed in%. In this experiment, two groups of control animals wereinjected i.p. with clomipramine or PK11195 vehicles. As similarresults were obtained from these two control groups, data wereconglomerated. aSignificant difference vs. vehicle- or PK11195-treated STD animals in the time in quadrant target (Pb0.05,Dunnett's two-tailed test for multiple comparisons).

Table 4 Effects of repeated administration (14 days) ofdiazepam (1 mg/kg/day) or PK11195 (1 mg/kg/day) ongrooming response of rats fed a different diet for 2 months


to cause any significant change in grooming response of ratstested in the NGT and the effect of diazepam was not affectedby the various dietary regimens.

Figs. 1 and 2 show data on the performance of animals testedin the MWM task as assessed the day after the FST procedure.Experimental groups showing decreased immobility time in theFST procedure (every CHO group and STD animals treated withclomipramine), exhibited also an improved water maze perfor-mance. In fact, these animals showed decreased latency to findthe platform only in the 1st day of training, whereas there was

Figure 3 Effects of acute administration of clomipramine(50 mg/kg) or PK11195 (1 mg/kg) in rats fed a different dietduring the reversal-training trials and the reversal probe trial inMWM task. Abbreviations STD and CHO denote different groupsof rats fed a standard diet or a standard diet enriched with 2%cholesterol for 2 months, respectively. Values are mean±S.E.M.of time measures expressed in s (latency to find the platform)and of time spent in target quadrant expressed in %,respectively. In this experiment, two groups of control animalswere injected i.p. with clomipramine or PK11195 vehicles. Assimilar results were obtained from these two control groups,data were conglomerated.

(standard diet or diet enriched with 2% cholesterol) in thenovelty-induced grooming sampling test (NGT)

Treatment Grooming score

STD+vehicle (20) 39.8±4.12STD+diazepam (10) 14.2±6.3 a, b

STD+PK11195 (10) 21.7±2.9 a

CHO+vehicle (20) 26.5±8.7 c

CHO+diazepam (10) 17.3±4.8 a

CHO+PK11195 (10) 22.5±2.9 a

Abbreviations STD and CHO denote different groups of rats feda standard diet or a standard diet enriched with 2% cholesterolfor 2 months, respectively. Diazepam (1 mg/kg/day), PK11195(1 mg/kg/day) or their vehicles were administered for 14 daysprior to behavioral testing. In this experiment, two groups ofcontrol animals were injected i.p. with diazepam vehicle orPK11195 vehicle. As similar results were obtained from thesetwo control groups, data were conglomerated.Values are mean±S.E.M. In parentheses the number of animalsper each group is indicated.a Significantly different as compared to STD+vehicle group

(Pb0.01, Dunnett's two-tailed test for multiple comparisons).b Significantly different as compared to CHO+vehicle group

(Pb0.01, Dunnett's two-tailed test for multiple comparisons).c Significantly different as compared to STD+vehicle group


Figure 5 Effects of repeated administration (14 days) ofclomipramine (25 mg/kg/day) or PK11195 (1 mg/kg/day) in ratsfed a different diet during the 1st and 2nd probe trial made30min before the 2nd and 4th training in Morris water maze task.Abbreviations STD and CHO denote different groups of rats fed astandard diet or a standard diet enriched with 2% cholesterol for2 months, respectively. Values are mean±S.E.M. of time spent intarget quadrant expressed in %. In this experiment, two groups ofcontrol animals were injected i.p. with clomipramine or PK11195vehicles. As similar results were obtained from these two controlgroups, data were conglomerated. aSignificant difference vs.vehicle-treated STD animals (Pb0.01, Dunnett's two-tailed testfor multiple comparisons). bSignificant difference vs. vehicle-treated STD animals (Pb0.05, Dunnett's two-tailed test formultiple comparisons).


no significant difference in the remaining three training days(Fig. 1). Furthermore, they exhibited a better performance inthe probe trial made on day 2 (1st probe) but not in that madeon day 4 (2nd probe), as expressed by an increased time (%)spent in the target quadrant (Fig. 2). No significant differencewas found in the reversal-training phase and in the reversalprobe as expressed by the latency to find the platform and bythe time (%) spent in the previous target quadrant, respectively(Fig. 3).

In the second experiment, the animals fed STD diet andtreated with 1 mg/kg/day PK11195 i.p. for 14 days, showed anincreased immobility time in the FST procedure (Table 3). Incontrast, animals exposed to CHO diet failed to show anyalteration of immobility time in the FST after repeatedtreatment with the PBRs antagonist. Also, the behavioral effectgiven by i.p. repeated administration of clomipramine (25 mg/kg/day) for 14 days was not affected by CHO diet exposure. Asshown in Table 4, STD or CHO fed animals showed decreasedgrooming response to novel environment in the NGT afterrepeated injections of PK11195 (1 mg/kg/day) or diazepam(1 mg/kg/day) for 14 days. The anxiolytic effect of both di-azepam and PK11195 was not affected by the various dietaryregimens.

As shown in Figs. 4 and 5, rats treated chronically withPK11195 before the FST procedure, showed an impairedacquisition in the MWM (with a slight, non significant prevalenceof the animals exposed to STD diet), i.e. an increased latency tofind the platform in all training days except the fourth.Furthermore, both STD and CHO fed groups treated chronically

Figure 4 Effects of repeated administration (14 days) ofclomipramine (25 mg/kg/day) or PK11195 (1 mg/kg/day) in ratsfed a different diet during the acquisition phase in Morris watermaze task, made the day after the FST procedure. AbbreviationsSTD and CHO denote different groups of rats fed a standard dietor a standard diet enriched with 2% cholesterol for 2 months,respectively. Values are mean±S.E.M. of time measuresexpressed in s. In this experiment, two groups of control animalswere injected i.p. with clomipramine or PK11195 vehicles. Assimilar results were obtained from these two control groups,data were conglomerated. aSignificant difference vs. vehicle-treated STD animals (Pb0.05, Dunnett's two-tailed test formultiple comparisons). bSignificant difference vs. clomipra-mine-treated STD or CHO animals (Pb0.01, Dunnett's two-tailedtest for multiple comparisons). cSignificant difference vs.vehicle-treated STD animals (Pb0.05, Dunnett's two-tailed testfor multiple comparisons).

with clomipramine (25 mg/kg/day) exhibited a better perfor-mance than controls in the 1st and 2nd acquisition days (Fig. 4).They also exhibited a better performance in the probe trialmade on days 2 (1st probe) but not in that made on day 4 (2ndprobe), as expressed by an increased time (%) spent in targetquadrant (Fig. 5). No significant difference was found in thereversal-training phase and in the reversal probe as expressedby the latency to find the platform and by time (%) spent in theprevious target quadrant (Fig. 6).

Figure 6 Effects of repeated administration (14 days) ofclomipramine (25 mg/kg/day) or PK11195 (1 mg/kg/day) in ratsfed a different diet during the reversal-training trials and thereversal probe trial in MWM task. Abbreviations STD and CHOdenote different groups of rats fed a standard diet or a standarddiet enriched with 2% cholesterol for 2 months, respectively.Values are mean±S.E.M. of time measures expressed in s(latency to find the platform) and of time spent in targetquadrant expressed in %, respectively. In this experiment, twogroups of control animals were injected i.p. with clomipramineor PK11195 vehicles. As similar results were obtained from thesetwo control groups, data were conglomerated.


4. Discussion

The first finding of the present study is the reducedimmobility time of rats fed a diet enriched with 2%cholesterol, that persists even after adjustment for theparameter of the weight change, suggesting that the bodyweight did not influence the immobility time in the FST.Furthermore, they also exhibited a reduced grooming scorein NGT, as reduced anxiety index. In rodents, grooming is a‘maintenance’ behavior that may be specifically elicited insituations in which animals are in stress-induced conflict orfrustration (Bolles, 1960; Fentress, 1973). A typical conditionof such type is found in the NGT, where grooming may play adeactivating role in restoring homeostasis (Gispen andIsaacson, 1981). Among various experimental models ofdepression, helplessness models such as the FST have beenreported to highly fulfill face, construct and predictivecriteria established for experimental models of mentaldiseases and are widely used for preclinical studies onnovel antidepressant drugs (Porsolt et al., 1978; Willner,1984). In this study clomipramine was selected as goldstandard antidepressant compound, based on results ofprevious experiments (Drago et al., 2001; Consoli et al.,2005; Micale et al., 2006). Furthermore, it has been seenthat the traditional FST, the same used in our experiments, isunreliable in the detection of the effects of SSRIs, thus toenhance the sensitivity of the traditional FST in the rats toSSRIs several simple procedural modifications have beenmade (Cryan et al., 2002). The two different dietary regimeswere selected based on the results of Koudinov andKoudinova (2003). Recently, they were confirmed by Dufouret al. (2006) showing that an enriched CHO diet in rats for2 months elicited an increase both of blood cholesterol levelsand of hippocampal cholesterol concentration in comparisonto the rats fed a STD diet, lacking any significant effect onthe weight of the rats.

Additionally, we found that different diets did not affectthe effects of classical antidepressant or anxiolytic drugssuch as clomipramine or diazepam. Conversely, previousstudies showed that high-cholesterol levels (≥200 mg/dl) inpatients with major depressive disorders (MDD) are asso-ciated both with poor response to fluoxetine or nortriptylinetreatment (Sonawalla et al., 2002; Papakostas et al., 2003a)and higher risk of subsequent relapse during the continuationphase of treatment with fluoxetine (Papakostas et al.,2004b). The mechanisms underlying these findings are notfully understood, but the authors suggest that high-choles-terol levels could lead to a lower 5-HT receptor sensitivity or5-HT transporter activity in depressed patients, eitherdirectly by binding to the various membrane-bound 5-HTreceptors and transporter or indirectly by altering thefluidity of the neuronal membrane and thereby the con-formation of these structures (Papakostas et al., 2004b).These findings could suggest a more vulnerability of SSRIscompounds to higher cholesterol levels, but a limitation ofthis hypothesis is given by the lack of data concerning therelationship between higher cholesterol levels and risk ofsubsequent relapse during the continuation phase of treat-ment with nortriptyline. However, in our results we did notfind any influence of different diets on the effects ofclomipramine and diazepam, since the rats fed a cholesterol-enriched diet only for 2 months to avoid the biological

cholesterol homeostasis dysregulation with synaptic plasti-city impairment and neuronal degeneration seen after 4–6 months of diet, as described by Koudinov and Koudinova(2003).

It is well accepted that cholesterol serves a vital functionin the brain via its important role as a component of myelinsheath, precursor for neurosteroids synthesis, component ofall membranes with subsequent role both in the neurotrans-mitter release and in the receptor expression (Dietschy andTurley, 2001; Rupprecht, 2003). Epidemiological and clinicalstudies have described an association between lower serumcholesterol concentrations and negative mood, decrementsin cognitive functions and various types of aggressivebehavior in adults (see for reviews Papakostas et al.,2004a,b; Huang and Chen, 2005) and recently also in children(Zhang et al., 2005). Genetic disorders of cholesterolsynthesis are also associated with mental retardation andsuicidality (Tierney et al., 2001; Lalovic et al., 2004).Furthermore, Jow et al. (2006) have found low plasma leptinand cholesterol levels in patients with major depressivedisorders, in line with previous studies showing in psychiatricdisorders both a positive correlation between serum leptinconcentration and total cholesterol (Kaplan, 1998) and aninteraction between leptinergic and 5-HT systems in thecentral nervous system (CNS) (Leibowitz and Alexander,1998). While several epidemiological and clinical studiesshowed a connection between cholesterol and mood dis-orders, only a few animal studies have been made tocorroborate this relationship. Monkeys, fed a low-cholesteroldiet in comparison to their high-cholesterol counterparts,exhibited less affiliative interaction, higher levels ofaggression with a lower cerebrospinal fluid concentrationsof 5-HIAA and blunted parameters of 5-HTactivity in the CNS(Muldoon et al., 1992; Kaplan et al., 1994). Several reportssuggested that a reduction in serum cholesterol may directlylead to decreased brain 5-HT activity through a variety ofmechanisms, ranging from an alteration in 5-HT levels to 5-HT receptor concentration or 5-HT transporter activity(Engelbergb, 1992; Ringo et al., 1994; Steegmans et al.,1996; Buydens-Branchey et al., 2000).

Since brain cholesterol is largely synthesized in situ, withlittle transfer of cholesterol or its precursors from plasma,our findings are in agreement to data suggesting that brainlipid metabolism, particularly cholesterol synthesis, isinfluenced by dietary lipid intake and that alterations inthe brain lipids makeup are followed by neurochemical andbehavioral consequences (Farquharson et al., 1992; Young1993; Kaplan et al., 1994; Howland et al., 1998; Koudinovand Koudinova, 2003).

Cholesterol is also the precursor for all steroid hormonessynthesized in the brain and in peripheral steroidogenictissues through the activation of PBRs and subsequentmovement of cholesterol from the cellular stores intomitochondria. In the brain, PBRs are predominantly localizedon glial cells, where the so-called “neurosteroids” aresynthesized to affect memory and mood mechanisms(Rupprecht, 2003). PBRs contribute to the regulation ofseveral major biological systems involved in stress, asshowed by the opposite changes in PBRs density in varioustissues as a result of exposure to acute or chronic stress(Drugan et al., 1986; Weizman et al., 1990; Burgin et al.,1996; Droogleever Fortuyn et al., 2004; Serra et al., 2004). In


fact, elevation in anxiety level with decreased hippocampalPBRs density, followed by a reduced anxiety level withincreased receptor density 24 h later, was observed afteracute stress (Avital et al., 2001). In our experiments, acutestress induced by the FST pre-test (swim for 15 min) wasfollowed by an improvement in the performance of CHO ratsthat was not affected by the acute administration ofPK11195. Interestingly, the same rats showing a betterperformance in the FST procedure, exhibited a significantimprovement of the acquisition phase in the first day of theMWM task (48 h after the pre-test of the FST) and in the firstprobe trial made 30 min prior the second training trial (i.e.,72 h after the pre-test of the FST, in agreement with thefindings described by Avital et al., 2001). Thus, the behaviorof CHO rats could be probably linked to the increasedhippocampal PBRs density, followed by an increased translo-cation of cholesterol from outer to inner mitochondrialmembranes and an increased neurosteroid synthesis.

In contrast to acute treatment, repeated administrationof the PBRs antagonist was followed by a more complexpicture. It was found that chronic treatment with the PBRsantagonist induced a significant increase in immobility timeof STD rats, but not in CHO animals. It is known that ratsexposed to a chronic mild stress (CMS), as social isolation,show a decreased brain density of PBRs and low plasma andbrain neurosteroids levels. Furthermore, CMS per seincreases the sensitivity of these receptors to endogenousor exogenous ligands, followed by increased plasma ACTHconcentration that in turn stimulates cholesterol uptakefrom plasma with increased neurosteroid synthesis (Calogeroet al., 1990; Serra et al., 2004). These mechanisms could betaken into account to explain the present results. In fact,both STD and CHO animals were chronically stressed by a 15-day daily injection of the PBRs antagonist and then tested inthe FST. In relation to the oversensitivity of PBRs to PK11195,it is possible that CMS, due to repeated drug injection,increased plasma ACTH levels that in turn positivelymodulated cholesterol uptake available to neurosteroidsynthesis. Therefore, the different behavior between therats fed a different diet could be linked to less availability ofcholesterol in STD animals. However the mechanismsunderling these findings remain unclear.

The literature is equivocal on the functional conse-quences of PK11195 action on PBRs. Antagonist effects werefound as PK11195 blocked anticonvulsant, anxiolytic- andantidepressant-like effects following PBRs activation(Romeo et al., 1992; Gavioli et al., 2003). Others havereported partial or full agonist activities for PK11195, suchas increased audiogenic seizure threshold in mice (Bena-vides et al., 1984), decreased symptoms of anxiety inhumans (Ansseau et al., 1991) and stimulated neurosteroidproduction in C6-2B glioma cell line (Papadopoulos et al.,1992).

The hypothesis discussed above could also explain theresults on the MWM (rats showing an antidepressant-likebehavior in the FST procedure, exhibited a better perfor-mance in the first day of the acquisition phase and in the firstprobe trial of MWM). At the end of the MWM, all groupsshowed the same cognitive performance. Since it is wellaccepted both the relationship between cholesterol and 5-HT system and the involvement of 5-HT system in cognitiveprocesses (Spreux-Varoquaux et al., 2001), common denomi-

nator underlying the better behavioral performance ofanimals in the FST procedure and in the MWM task could bedue to an increased 5-HT tone.

It has been seen that exposure to a cholesterol dietincreases memory retention in rabbits (Sparks and Schreurs,2003) and in young rats (Dufour et al., 2006). Almost all braincholesterol is synthesized in situ and little or none of theperipheral cholesterol crosses the blood–brain barrier.However, Sparks et al. (1995) found increased cholesterollevels in the brain of rabbits fed a cholesterol diet.Cholesterol metabolites such as neurosteroids do not sufferfrom the same limitation as cholesterol to reach the brain,since they are synthesized in situ. Thus, another hypothesisto explain the effect of cholesterol-enriched diet on moodand memory is that elevations in serum cholesterol couldincrease the production of neurosteroids such as dehydroe-piandrosterone (DHEA), pregnenolone, 17β-estradiol thatacting in turn on the GABA, N-methyl-D-aspartate (NMDA),cholinergic or sigma opioid systems may exert anxiolytic,antidepressant or cognition enhancer effects (Rupprecht,2003). Common causal factors in mood disorders and inmemory impairments are stress and dysregulation of HPAwith increased glucocorticoid levels. Antidepressant drugs,which alter monoaminergic neurotransmission, normalizeHPA function. 5-HT tone, neurosteroids and HPA axis are alsoimplicated in the pathogenesis of anxiety and they couldunderlie the better performance of CHO rats in the NGT.Furthermore, the anxiolytic-like behavior showed by the ratstreated chronically with the PBRs antagonist is in line withthe results of Ansseau et al. (1991), suggesting a potentialanxiolytic activity of this compound.

Despite the exact relationships between lipid profiles,stress, HPA axis and mood disorders are still unknown, thepresent results support the hypothesis of a correlation be-tween low serum cholesterol and mood disorders. Howeverthe reader should be cautioned about the limited inter-pretations of the present findings. In fact, the nature of thisstudy does not allow establishing a causal relationshipbetween a dietary regimen for 2 months and phenotypingchanges, since only behavioral tests have been used.Furthermore more research should be done to investigatethe effects of low serum cholesterol on the neurosteroidssynthesis through the involvement of PBRs receptors inanimal and humans.

Role of funding source

No funding sources are recognized.

Contributors

This research has been made with the only contribution of theauthors' institutions, i.e. the Department of Experimental andClinical Pharmacology, University of Catania, Catania, Italy, theBlanchette Rockefeller Neuroscience Institute, West Virginia Uni-versity, Rockville, MD, and the Institute of Neurological Sciences,Italian National Research Council, CNR, Catania, Italy.

Conflict of interest

No conflict of interest exists for any of the authors.


References

Almeida-Montes, L.G., Valles-Sanchez, V., Moreno-Aguilar, J.,Chavez-Balderas, R.A., Garcia-Marin, J.A., Cortes Sotres, J.F.,Hheinze-Martin, G., 2000. Relation of serum cholesterol, lipid,serotonin and tryptophan levels to severity of depression and tosuicide attempts. J. Psychiatry Neurosci. 25, 371–377.

Ansseau, M., von Frenckell, R., Cerfontaine, J.L., Papart, P., 1991.Pilot study of PK 11195, a selective ligand for the peripheral-typebenzodiazepine binding sites, in patients with anxious ordepressive symptomatology. Pharmacopsychiatry 24, 8–12.

Avital, A., Richter-Levin, G., Leschiner, S., Spanier, I., Veenman, L.,Weizman, A., Gavish, M., 2001. Acute and repeated swim stresseffects on peripheral benzodiazepine receptors in the rathippocampus, adrenal, and kidney. Neuropsychopharmacology25, 669–678.

Benavides, J., Guilloux, F., Allam, D.E., Uzan, A., Mizoule, J.,Renault, C., Dubroeucq, M.C., Gueremy, C., Le Fur, G., 1984.Opposite effects of an agonist, RO5-4864, and an antagonist, PK11195, of the peripheral type benzodiazepine binding sites onaudiogenic seizures in DBA/2J mice. Life Sci. 34, 2613–2620.

Bitran, D., Foley, M., Audette, D., Leslie, N., Frye, C.A., 2000.Activation of peripheral mitochondrial benzodiazepine receptorsin the hippocampus stimulates allopregnanolone synthesis andproduces anxiolytic-like effects in the rat. Psychopharmacology151, 64–71.

Bolles, R.C., 1960. Grooming behavior in the rat. J. Comp. Physiol.Psychol. 53, 306–310.

Bulbena, A., Berrios, G.E., 1993. Cognitive function in the affectivedisorders: a prospective study. Psychopathology 26, 6–12.

Burgin, R., Weizman, R., Gavish, M., 1996. Repeated swim stress andperipheral-type benzodiazepine receptors. Neuropsychobiology33, 28–31.

Buydens-Branchey, L., Branchey, M., Hudson, J., Fergeson, P., 2000.Low HDL cholesterol, aggression and altered central serotonergicactivity. Psychiatry Res. 93, 93–102.

Calogero, A.E., Kamilaris, T.C., Bernardini, R., Johnson, E.O.,Chrousos, G.P., Gold, P.W., 1990. Effects of peripheral benzo-diazepine receptor ligands on hypothalamic–pituitary–adrenalaxis function in the rat. J. Pharmacol. Exp. Ther. 253, 729–737.

Claudepierre, T., Pfrieger, F.W., 2003. New aspects of cholesterol inthe central nervous system. Med. Sci. 19, 601–605.

Consoli, D., Fedotova, J., Micale, V., Sapronov, N.S., Drago, F., 2005.Stressors affect the response of male and female rats toclomipramine in a model of behavioral despair (forced swimtest). Eur. J. Pharmacol. 520, 100–107.

Cryan, J.F., Markou, A., Lucki, I., 2002. Assessing antidepressantactivity in rodents: recent developments and future needs.Trends Pharmacol. Sci. 23, 238–245.

Dietschy, J.M., Turley, S.D., 2001. Cholesterol metabolism in thebrain. Curr. Opin. Lipidol. 12, 105–112.

Drago, F., Canonico, P.L., Bitetti, R., Scapagnini, U., 1980. Systemicand intraventricular prolactin induces excessive grooming. Eur.J. Pharmacol. 65, 457–458.

Drago, F., Nicolosi, A., Micale, V., Lo Menzo, G., 2001. Placeboaffects the performance of rats in models of depression: is it agood control for behavioral experiments? Eur. Neuropsychophar-macol. 11, 209–213.

Droogleever Fortuyn, H.A., van Broekhoven, F., Span, P.N., Backstrom,T., Zitman, F.G., Verkes, R.J., 2004. Effects of PhD examinationstress on allopregnanolone and cortisol plasma levels andperipheral benzodiazepine receptor density. Psychoneuroendocri-nology 29, 1341–1344.

Drugan, R.C., Basile, A.S., Crawley, J.N., Paul, S.M., Skolnick, P.,1986. Inescapable shock reduces [3H]Ro 5-4864 binding to“peripheral-type” benzodiazepine receptors in the rat. Pharma-col. Biochem. Behav. 24, 1673–1677.

Dufour, F., Liu, Q.Y., Gusev, P., Alkon, D.L., Atzori, M., 2006.Cholesterol-enriched diet affects spatial learning and synapticfunction in hippocampal synapses. Brain Res. 1103, 88–98.

Engelbergb, H., 1992. Low serum cholesterol and suicide. Lancet339, 727–729.

Farquharson, J., Cockburn, F., Patrick, W.A., Jamieson, E.C., Logan,R.W., 1992. Infant cerebral cortex phospholipid fatty-acidcomposition and diet. Lancet 340, 810–813.

Fentress, J.C., 1973. Development of grooming in mice withamputated forelimbs. Science 179, 704–705.

Fujishiro, J., Imanishi, T., Onozawa, K., Tsushima, M., 2002.Comparison of the anticholinergic effects of the serotonergicantidepressants, paroxetine, fluvoxamine and clomipramine.Eur. J. Pharmacol. 454, 183–188.

Gavioli, E.C., Duarte, F.S., Bressan, E., Ferrara, P., Farges, R.C., DeLima, T.C., 2003. Antidepressant-like effect of Ro5-4864, aperipheral-type benzodiazepine receptor ligand, in forcedswimming test. Eur. J. Pharmacol. 471, 21–26.

Gispen, W.H., Isaacson, R.L., 1981. ACHT-induced excessive groom-ing in the rat. Pharmacol. Ther. 12, 209–246.

Heron, D.S., Shinitzky, M., Hershkowitz, M., Samuel, D., 1980.Lipid fluidity markedly modulates the binding of serotoninto mouse brain membranes. Proc. Natl. Acad. Sci. U. S. A. 77,7463–7467.

Holmes, P.V., Drugan, R.C., 1991. Differential effects of anxio-genic central and peripheral benzodiazepine receptor ligandsin tests of learning and memory. Psychopharmacology 104,249–254.

Horsten, M., Wamala, S.P., Vingerhoets, A., Orth-Gomer, K., 1997.Depressive symptoms, social support, and lipid profile in healthymiddle-aged women. Psychosom. Med. 59, 521–528.

Howland, D.S., Trusko, S.P., Savage, M.J., Reaume, A.G., Lang, D.M.,Hirsch, J.D., Maeda, N., Siman, R., Greenberg, B.D., Scott, R.W.,Flood, D.G., 1998. Modulation of secreted beta-amyloid pre-cursor protein and amyloid beta-peptide in brain by cholesterol.J. Biol. Chem. 273, 16576–16582.

Huang, T.L., Chen, J.F., 2005. Cholesterol and lipids in depression:stress, hypothalamo–pituitary–adrenocortical axis, and inflam-mation/immunity. Adv. Clin. Chem. 39, 81–105.

Jow, G.M., Yang, T.T., Chen, C.L., 2006. Leptin and cholesterol levelsare low in major depressive disorder, but high in schizophrenia.J. Affect. Disord. 90, 21–27.

Kaplan, L.M., 1998. Leptin, obesity, and liver disease. Gastroenter-ology 115, 997–1001.

Kaplan, J.R., Shively, C.A., Fontenot, M.B., Morgan, T.M., Howell,S.M., Manuck, S.B., Muldoon, M.F., Mann, J.J., 1994. Demon-stration of an association among dietary cholesterol, centralserotonergic activity, and social behavior in monkeys. Psycho-som. Med. 56, 479–484.

Koudinov, A.R., Koudinova, N.V., 2003. Cholesterol, synapticfunction and Alzheimer's disease. Pharmacopsychiatry 36,107–112.

Krüeger, K.E., Papadopulos, V., 1990. Peripheral-type benzodiaze-pine receptors mediate translocation of cholesterol from outer toinner mitochondrial membranes in adrenocortical cells. J. Biol.Chem. 265, 15015–15022.

Kumar, S., Kulkarni, S.K., 1996. Influence of antidepressant drugs onlearning and memory paradigms in mice. Indian J. Exp. Biol. 34,431–435.

Lalovic, A., Merkens, L., Russell, L., Arsenault-Lapierre, G., Nowaczyk,M.J., Porter, F.D., Steiner, R.D., Turecki, G., 2004. Cholesterolmetabolism and suicidality in Smith–Lemli–Opitz syndrome car-riers. Am. J. Psychiatry 161, 2123–2126.

Leibowitz, S.F., Alexander, J.T., 1998. Hypothalamic serotonin incontrol of eating behavior, meal size, and body weight. Biol.Psychiatry 44, 851–864.

Meneses, A., 1999. 5-HT and cognition. Neurosci. Biobehav. Rev. 23,1111–1125.


Micale, V., Arezzi, A., Rampello, L., Drago, F., 2006. Melatoninaffects the immobility time of rats in the forced swim test: therole of serotonin neurotransmission. Eur. Neuropsychopharma-col. 16, 538–545.

Morgan, R.E., Palinkas, L.A., Barrett-Connor, E.L., Wingard, D.L.,1993. Plasma cholesterol and depressive symptoms in older men.Lancet 341, 75–79.

Morris, R., 1984. Developments of a water-maze procedure forstudying spatial learning in the rat. J. Neurosci. Methods 11, 47–60.

Muldoon, M.F., Kaplan, J.R., Manuck, S.B., Mann, J.J., 1992. Effectsof a low-fat diet on brain serotonergic responsivity in cynomolgusmonkeys. Biol. Psychiatry 31, 739–742.

Oner, P., Bekpinar, S.,Oz, B., 1991. Alterations in some lipid componentsand Ca2+ ATPase activity in brain of rats fed an atherogenic diet.Res. Commun. Chem. Pathol. Pharmacol. 72, 337–345.

Papadopoulos, V., Guarneri, P., Kreuger, K.E., Guidotti, A., Costa, E.,1992. Pregnenolone biosynthesis in C6-2B glioma cell mitochon-dria: regulation by a mitochondrial diazepam binding inhibitorreceptor. Proc. Natl. Acad. Sci. U. S. A. 89, 5113–5117.

Papakostas, G.I., Petersen, T., Sonawalla, S.B., Merens, W.,Iosifescu, D.V., Alpert, J.E., Fava, M., Nierenberg, A.A., 2003a.Serum cholesterol in treatment-resistant depression. Neuropsy-chobiology 47, 146–151.

Papakostas, G.I., Petersen, T., Mischoulon, D., Hughes, M.E., Alpert,J.E., Nierenberg, A.A., Rosenbaum, J.F., Fava, M., 2003b. Serumcholesterol and serotonergic function in major depressivedisorder. Psychiatry Res. 118, 137–145.

Papakostas, G.I., Ongur, D., Iosifescu, D.V., Mischoulon, D., Fava, M.,2004a. Cholesterol in mood and anxiety disorders: review of theliterature and new hypotheses. Eur. Neuropsychopharmacol. 14,135–142.

Papakostas, G.I., Iosifescu, D.V., Petersen, T., Hamill, S.K., Alpert,J.E., Nierenberg, A.A., Rosenbaum, J.F., Fava, M., 2004b. Serumcholesterol in the continuation phase of pharmacotherapy withfluoxetine in remitted major depressive disorder. J. Clin.Psychopharmacol. 24, 467–469.

Pfrieger, F.W., 2003. Cholesterol homeostasis and function inneurons of the central nervous system. Cell. Mol. Life Sci. 60,1158–1171.

Porsolt, R.D., Anton, G., Blavet, N., Jalfre, M., 1978. Behaviouraldespair in rats: a new model sensitive to antidepressanttreatments. Eur. J. Pharmacol. 47, 379–391.

Ringo, D.L., Lindley, S.E., Faull, K.F., Faustman, W.O., 1994.Cholesterol and serotonin: seeking a possible link betweenblood cholesterol and CSF 5-HIAA. Biol. Psychiatry 35, 957–959.

Romeo, E., Auta, J., Kozikowski, A.P., Ma, D., Papadopoulos, V.,Puia, G., Costa, E., Guidotti, A., 1992. 2-Aryl-3-indoleaceta-mides (FGIN-1): a new class of potent and specific ligands for themitochondrial DBI receptor (MDR). J. Pharmacol. Exp. Ther. 262,971–978.

Rupprecht, R., 2003. Neuroactive steroids: mechanisms of actionand neuropsychopharmacological properties. Psychoneuroendo-crinology 28, 139–168.

Scanlon, S.M., Williams, D.C., Schloss, P., 2001. Membrane choles-terol modulates serotonin transporter activity. Biochemistry 40,10507–10513.

Schrijver, N.C., Pallier, P.N., Brown, V.J., Wurbel, H., 2004. Doubledissociation of social and environmental stimulation on spatiallearning and reversal learning in rats. Behav. Brain Res. 152,307–314.

Serra, M., Pisu, M.G., Floris, I., Floris, S., Cannas, E., Mossa, A.,Trapani, G., Latrofa, A., Purdy, R.H., Biggio, G., 2004. Socialisolation increases the response of peripheral benzodiazepinereceptors in the rat. Neurochem. Int. 45, 141–148.

Sonawalla, S.B., Papakostas, G.I., Petersen, T.J., Yeung, A.S., Smith,M.M., Sickinger, A.H., Gordon, J., Israel, J.A., Tedlow, J.R.,Lamon-Fava, S., Fava, M., 2002. Elevated cholesterol levelsassociated with nonresponse to fluoxetine treatment in majordepressive disorder. Psychosomatics 43, 310–316.

Sparks, D.L., Schreurs, B.G., 2003. Trace amounts of copper in waterinduce beta-amyloid plaques and learning deficits in a rabbitmodel of Alzheimer's disease. Proc. Natl. Acad. Sci. U. S. A. 100,11065–11069.

Sparks, D.L., Liu, H., Gross, D.R., Scheff, S.W., 1995. Increaseddensity of cortical apolipoprotein E immunoreactive neurons inrabbit brain after dietary administration of cholesterol. Neu-rosci. Lett. 187, 142–144.

Spreux-Varoquaux, O., Alvarez, J.C., Berlin, I., Batista, G.,Despierre, P.G., Gilton, A., Cremniter, D., 2001. Differentialabnormalities in plasma 5-HIAA and platelet serotonin concen-trations in violent suicide attempters: relationships with impul-sivity and depression. Life Sci. 69, 647–657.

Steegmans, P.H., Fekkes, D., Hoes, A.W., Bak, A.A., van der Does, E.,Grobbee, D.E., 1996. Low serum cholesterol concentration andserotonin metabolism in men. Br. Med. J. 312, 221.

Sternberg, D.E., Jarvik, M.E., 1976. Memory functions in depression.Arch. Gen. Psychiatry 33, 219–224.

Tierney, E., Nwokoro, N.A., Porter, F.D., Freund, L.S., Ghuman, J.K.,Kelley, R.I., 2001. Behavior phenotype in the RSH/Smith–Lemli–Opitz syndrome. Am. J. Med. Genet. 98, 191–200.

Weizman, A., Bidder, M., Fares, F., Gavish, M., 1990. Fooddeprivation modulates gamma-aminobutyric acid receptors andperipheral benzodiazepine binding sites in rats. Brain Res. 535,96–100.

Willner, P., 1984. Cognitive functioning in depression: a review oftheory and research. Psychol. Med. 14, 807–823.

Young, S.N., 1993. Cholesterol, heart disease and the brain: anopportunity in research and a disaster in public health education?J. Psychiatry Neurosci. 18, 1–3.

Zhang, J., McKeown, R.E., Hussey, J.R., Thompson, S.J., Woods,J.R., Ainsworth, B.E., 2005. Low HDL cholesterol is associatedwith suicide attempt among young healthy women: the ThirdNational Health and Nutrition Examination Survey. J. Affect.Disord. 89, 25–33.

Documents

Behavioral effects of dietary cholesterol in rats tested in experimental models of mild stress and cognition tasks