logical Psychiatry 30 (2006) 684–693www.elsevier.com/locate/pnpbp

Progress in Neuro-Psychopharmacology & Bio

Effects of maternal separation on neuropetide Y and calcitonin gene-relatedpeptide in “depressed” Flinders Sensitive Line rats: A study of

gene–environment interactions

Gitta Wörtwein a, Henriette Husum a, Weronica Andersson b,Tom G. Bolwig a, Aleksander A. Mathé b,⁎

a Laboratory of Neuropsychiatry, Rigshospitalet, 2100 Copenhagen, Denmarkb Karolinska Institutet, Clinical Neuroscience, Psychiatry M56, Karolinska University Hospital, Huddinge, SE-14186 Stockholm, Sweden

Accepted 30 January 2006Available online 6 March 2006

Abstract

Interactions between genetic vulnerability to stress/depression and early life experience may play a crucial role in the pathogenesis of mooddisorders. Here we explore this hypothesis by superimposing early life trauma in the form of maternal deprivation for 180min per day frompostnatal day 2 to 14 onto a genetic model of depression/susceptibility to depression, Flinders Sensitive Line (FSL) and their controls, FlindersResistant Line (FRL) rats. We investigate effects on neuropeptide Y (NPY) and calcitonin gene-related peptide (CGRP) like immunoreactivity (LI)in 10 brain regions as these neuropeptides are affected by antidepressants and are altered in cerebrospinal fluid of depressed patients. NPY-LI wasreduced while CGRP-LI was elevated in hippocampus and frontal cortex of “genetically depressed” FSL rats. The two peptides displayed asignificant negative correlation in these regions that was strongest in the FSL strain. Maternal deprivation exacerbated the strain difference inhippocampal CGRP-LI, while it was without effect on NPY-LI. FSL rats had higher tissue concentration of both neuropeptides in periaqueductalgrey and higher NPY-LI in caudate/putamen. Maternal deprivation selectively raised CGRP-LI in amygdala of the FRL control stain. Thus, in twobrain regions implicated in the neurobiology of depression, hippocampus and frontal cortex, changes in CGRP-LI and NPY-LI were in oppositedirection, and CGRP-LI appears to be more responsive to adverse experience. Our findings thus support the hypothesis that genetic dispositionand developmental stress may contribute to the susceptibility to depression by exerting selective neuropeptide- and brain region-specific effects onadult neurobiology.© 2006 Elsevier Inc. All rights reserved.

Keywords: Calcitonin gene-related peptide; Depression; Maternal deprivation; Neuropeptide Y; Rat

1. Introduction

In spite of extensive basic and clinical research, thepathogenesis of mood disorders is still not elucidated. In

Abbreviations: AMY, amygdala; CGRP, calcitonin gene-related peptide;CNS, central nervous system; CPU, caudate–putamen; CSF, cerebrospinal fluid;EC, entorhinal cortex; FC, frontal cortex; FRL, Flinders Resistant Line; FSL,Flinders Sensitive Line; HPA, hypothalamic–pituitary–adrenal; HPC, hippo-campus; HT, hypothalamus; icv, intracerebroventricular; LI, like immunor-eactiviy; MS, maternal separation; NA, nucleus accumbens; NPY, neuropeptideY; OC, occipital cortex; PAG, periaqueductal grey; PND, postnatal day; REM,rapid eye movement sleep; TC, temporal cortex.⁎ Corresponding author. Tel.: +46 8 52487972; fax: +46 8 52488929.E-mail address: aleksander.mathe@cns.ki.se (A.A. Mathé).

0278-5846/$ - see front matter © 2006 Elsevier Inc. All rights reserved.doi:10.1016/j.pnpbp.2006.01.027

addition to the statistically well-documented role of geneticpredisposition, clinical studies indicate that psychosocial factorsare also of importance, at least in precipitating the first episodeof recurrent depressive illness (Post, 1992). Increasing appre-ciation of the effects of early life experiences on neurobiologicaldevelopment further suggests that the interaction betweengenetic vulnerability to stress/depression and early lifeexperience may play a crucial role in the pathogenesis ofmood disorders (Liu et al., 2000; Weaver et al., 2004). Humanretrospective studies likewise have shown that adverse events inearly life can predispose to depression and to anxiety disordersin adult life. Such events include disruptions of bonding to theprimary, maternal caregiver (Kendler et al., 1992; Servant andParquet, 1994; Agid et al., 1999).

685G. Wörtwein et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 30 (2006) 684–693

An animal model developed to study these phenomenashows that maternal separation (MS) of rat pups during thestress-hyporesponsive period of early postnatal developmentaffects behavior and physiology of the offspring (Sapolsky andMeaney, 1986). MS results in increased adult stress reactivity,accompanied by augmented fluctuations in the secretion ofcorticotropin releasing hormone, adrenocorticotropin, andcorticosterone (reviewed in Heim and Nemeroff, 1999).However, MS has so far been carried out only in geneticallynormal animals, thus not taking into account the role geneticloading may play in mood disorders.

Consequently, to address this issue, the authors designed anovel paradigm in which early life trauma was superimposedonto a genetic model of depression/susceptibility to depression,Flinders Sensitive Line (FSL) and their controls, FlindersResistant Line (FRL) rats. We recently reported that thismanipulation further increased depression like behavior (Porsoltswim test) in the FSL animals, while leaving the FRL strainunaffected (El Khoury et al., in press).

We chose the FSL model since the FSL strain has good faceand predictive validity for a genetic animal model of depression.Originally, FSL and FRL rats were selected for high (FSL) andlow (FRL) sensitivity to anticholinesterase agents, based on theobservation that some depressed patients are more sensitive tocholinergic agonists (Overstreet et al., 1979). This resulted in astrain with reduced body weight, lower activity levels, increasedtotal rapid eye movement (REM) sleep, decreased REM sleeplatency, increased immobility in the Porsolt swim test, cognitivedeficits, and “anhedonia” (as inferred by e.g. sucrose consump-tion test) after chronic mild stress (Overstreet, 1993; Overstreet,2002). Some of those deficits can be reversed by chronic, butnot acute, antidepressant treatment (Schiller et al., 1992;Overstreet et al., 1995, 2004). FSL rats also show alterationsin parameters reflecting catecholaminergic system activity,namely increased tyrosine hydroxylase mRNA levels in theventral tegmental area (Serova et al., 1998), higher dopamine,serotonin and norepinephrine levels in various brain regions, anabnormality that could be normalized by chronic desimipraminetreatment (Zangen et al., 1997, 1999), higher tissue content butlower extracellular dopamine levels in the nucleus accumbensand absence of serotonin–dopamine interaction in this brainregion (Zangen et al., 2001; Dremencov et al., 2004).

Thus far there is no firm evidence that a dysregulation of themonoaminergic systems plays a necessary role in the patho-genesis of depression. Consequently, other neural systems arelikely to be involved. Neuropeptides have a number ofmodulatory/transmitter roles in the central nervous system(CNS). Neuropetide Y (NPY) and calcitonin gene-relatedpeptide (CGRP) have been shown to interact with classicalneurotransmitters and are affected by antidepressant andantipsychotic treatment modalities (Deutch and Roth, 1987;Drumheller et al., 1992; Stenfors et al., 1994; Mathé et al., 1997;Gruber et al., 2001, 2002).

Neuropeptide Y (NPY), a member of the pancreaticpolypeptide family, affects a variety of functions in the centralnervous system. It modulates neuronal activity and hypotha-lamic–pituitary–adrenal (HPA) axis function, influences circa-

dian rhythms and is involved in memory processing, anxietyand food intake (Heilig and Widerlov, 1995; Woldbye et al.,1997; Sørensen et al., 2004). Extensive evidence points tointerrelationships between NPY and the dopaminergic system.Thus, intracerebroventricular (icv) administration of NPYincreased the tissue concentration of dopamine in variousareas of the rat brain and its release from striatum (Heilig et al.,1990; Kerkerian-Le Goff et al., 1992; Drumheller et al., 1994).Correspondingly, acute activation of the dopaminergic systemby administration of D-amphetamine increased basal extracel-lular NPY efflux in the ventral striatum, an effect that could beblocked by dopamine receptor antagonists (Gruber et al., 2002).Also, chronic administration of antidopaminergic agentsdecreased basal extracellular NPY efflux in the ventral striatumand had brain region and drug specific effects on NPY-likeimmunoreactivity (-LI) tissue concentration (Gruber and Mathé,2000; for review see Obuchowicz et al., 2004). NPYconcentrations in cerebrospinal fluid (CSF) and plasma fromdepressed patients and suicide attempters are often reduced andstudies reported increases in NPY-LI in CSF from depressedpatients following a course of electroconvulsive therapy orpharmacological treatment (Widerlöv et al., 1988; Gjerris et al.,1992; Nilsson et al., 1996; Hashimoto et al., 1996; Mathé et al.,1996b; Westrin et al., 1999; Heilig et al., 2004; Nikisch et al.,2005). NPY expression, both mRNA and protein, in thehippocampus of genetic as well as environmental models ofdepression is reduced (Mathé et al., 1998; Holmes et al., 1998;Jimenez Vasquez et al., 2000a; Husum and Mathé, 2002;Husum et al., 2002; Lim et al., 2003). Central administration ofNPY reduces immobility in the Porsolt swim test and anxietylike behavior in rodents (Heilig et al., 1989; Husum et al., 2000;Stogner and Holmes, 2000; Redrobe et al., 2002; Mathé andGruber, 2004). Various antidepressant treatment modalitiesincrease NPY-like immunoreactivity or NPY mRNA in severalbrain regions (Heilig et al., 1988; Zachrisson et al., 1995a,b;Baker et al., 1996; Caberlotto et al., 1998; Jimenez Vasquez etal., 2000a; Husum et al., 2000). Taken together these datasupport the hypothesis that NPY plays a role in thepathophysiology of depression and that the therapeutic effectsof antidepressant treatment modalities are in part mediated viathe NPY system (Mathé, 1999; Mathé et al., in press).

Calcitonin-gene-related peptide (CGRP), a 37-amino acidpeptide resulting from alternative splicing of the calcitoningene, is widely distributed in the brain and interacts with theclassical neurotransmitters dopamine and noradrenaline (vanRossum et al., 1997). In rat, high concentrations of CGRPimmunoreactivity are measured in the ventral tegmental area,while CGRP fibers are found most abundantly in the frontalcortex, amygdala, and nucleus accumbens (Orazzo et al., 1993;Kresse et al., 1995). CGRP receptors are often localized ondopaminergic neurons (Wimalawansa, 1996, 1997). CGRPfibers in the amygdala and in the bed nucleus of the striaterminals innervate neurons containing CRF and neurotensin(Honkaniemi et al., 1990; Ju, 1991; Harrigan et al., 1994). It isthus not surprising that CGRP has been shown to respond tostress (Guidobono et al., 1991; Husum et al., 2002). Increases incalcitonin gene-related peptide-like immunoreactivity (CGRP-

686 G. Wörtwein et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 30 (2006) 684–693

LI) in the CSF of depressed patients have been reported (Mathéet al., 1994, 2002). In rodents CGRP potentiates fear relatedbehaviors (Kovacs and Telegdy, 1994, 1995; Poore andHelmstetter, 1996) and has anorectic effects (Lutz et al.,1997), actions which are directly opposite to those of NPY(Heilig and Widerlöv, 1995). CGRP reduces motor activity andinduces catalepsy, while coadministration of CGRP increaseshaloperidol-induced catalepsy and decreases apomorphine-induced hypermotility, actions known to be related to dopamine(Clementi et al., 1992; Jolicoeur et al., 1992; Kovacs et al.,1999). In agreement with these observations, it has been shownthat both psychotomimetics and antipsychotic drugs affectCGRP regional brain concentrations and release in vivo (Mathéet al., 1996a, 1997; Gruber et al., 2001; Angelucci et al., 2001).Moreover, administered directly into rat CNS, CGRP markedlyaffects dopamine release and metabolism in selected brainregions (Deutch and Roth, 1987; Drumheller et al., 1992;Jolicoeur et al., 1992). Taken together, these findings indicate apotential role for CGRP in dopamine-related CNS disorders.

In view of the above, the authors examined NPY-LI andCGRP-LI in selected brain regions of adult, genetically“depressed” FSL rats that had been subjected to maternaldeprivation in infancy.

2. Materials and methods

2.1. Experimental animals

Parent FSL and FRL rats were selected from the rat colonymaintained at the Karolinska Institutet, Stockholm. Thebreeding pairs were housed in standard macrolon cages untilpregnancy was noticeable (approximately gestational day 12)after which the pregnant rats (9 FRL, 7 FSL) were isolated andprovided with nesting material. The day of delivery wasdesignated postnatal day (PND) 0. A group of rat pups that wasnot handled at any time apart from the weekly change ofbedding material was designated TNH. All the other pups weresexed and culled into standardized litters of 8–10 pups (mixedsexes) on PND2. Within each strain, litters were randomlyassigned to one of the following two procedures which tookplace during PND2–14: no handling (NH) or maternalseparation for 180min/day (MS180). The separation procedurewas as follows: upon leaving the nest, the dam was placed inanother cage and the pups were moved as a litter into a neonatalincubator with fresh sawdust. The incubator was set to maintainan ambient temperature of 32±0.5°C (PND2–5) or 30±0.5°C(PND6–14). Upon returning the pups to the nest, the dam wasreturned to the home cage. All pups were weaned on PND22and from then onwards kept 4–5 rats/cage (single gender) understandard conditions with free access to rat chow and tap water ina 12-h day/night cycle (lights on at 7:00 a.m.) until adolescence.Six groups of male rats were included in the study; FRL: TNH(n=11), NH (n=9), MS180 (n=12); FSL: TNH (n=7), NH(n=6), MS180 (n=9). The studies were approved by Stock-holm's Ethical Committee for Protection of Animals and wereconducted in accordance with the Karolinska Institutets Guide-lines for the Care and Use of Laboratory Animals.

2.2. Biochemistry

On PND90, the rats were weighed and sacrificed by 2 s highenergy focused microwave irradiation as this proceduremarkedly increases the recovery of peptides (Mathé et al.,1990b). The brains were quickly removed and stored at −80°C.Later, frozen brains were cut into 2 mm slices using amicrotome and the following brain regions were isolated bymicropuncture on ice: frontal cortex (FC), caudate/putamen(CPU), nucleus accumbens (NA), hypothalamus (HT), temporalcortex (TC), entorhinal cortex (EC), occipital cortex (OC),hippocampus (HPC), amygdala (AMY), and periaqueductalgrey (PAG). The brain tissues were subsequently homogenized,twice extracted by boiling in acetic acid and water, lyophilisedand finally subjected to radioimmunoassay analysis for NPY-LI(antibody RIN7180, Peninsula) and CGRP-LI (antibodyRIN6006, Peninsula), as described in detail previously(Husum et al., 2003).

2.3. Statistical analysis

Student's t-tests showed no significant differences betweenthe TNH and NH groups in any of the reported parameters.Consequently TNH and NH animals from each strain werepooled and are subsequently referred to as ‘CONTROL’.Animals with values more than 3 interquartile ranges above orbelow the 75th percentile and 25th percentile respectively wereexcluded from statistical analysis (SPSS Inc.CIU, 2004). Two-way analysis of variance was used with factors strain (FSL, FRL)and neonatal handling (CONTROL, MS180). Post hoc Tukeytests were employed when appropriate. Additionally, Pearson'scorrelation coefficient was calculated for all brain areas. Level ofstatistical significance was set at P<0.05. The data are presentedas mean pmol/gram wet weight tissue±S.E.M.

3. Results

3.1. Bodyweight

On PND90, the FSL rats weighed significantly less than FRL(338±48g vs. 371±31g, respectively, P<0.001).

3.2. Neuropeptide Y

As shown in Fig. 1a, in the frontal cortex NPY-LI levels weresignificantly lower in FSL animals (F(1,50)=5.539, p=0.023)and this effect was due to a difference between control FSL andFRL animals (p=0.027). Maternal separation had no significanteffect on NPY-LI in this brain region and there was nointeraction between the two factors.

Hippocampal NPY-LI levels were marginally lower in FSLcompared to FRL animals (F(1,50)=3.654, p=0.062) and thisstrain difference was due to lower levels of NPY-LI in controlFSL animals (p=0.043). Maternal separation non-significantlylowered hippocampal NPY-LI levels (F(1,50) = 3.643,p=0.062). There was no interaction between the two factors(Fig. 1b).

Fig. 1. Effects of maternal separation on NPY-LI concentration in brain regions of “genetically depressed” Flinders Sensitive Line and control strain Flinders ResistantLine rats. Represented are means±S.E.M. MS180: maternal separation from postnatal day 2 to 14 for 180min per day. CONTROL: no manipulation from postnatalday 2 to 14. ⁎ indicates a strain difference (FSL differ from FRL of the same handling group, ⁎P<0.05, ⁎⁎P<0.01). Maternal separation had no significant effects.

687G. Wörtwein et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 30 (2006) 684–693

In the caudate/putamen, NPY-LI was significantly higher inFSL compared to FRL animals (F(1,50)=7.757, p=0.008) andthis strain difference was due to higher levels of NPY-LI incontrol FSL animals (p=0.003). Maternal separation had nosignificant effect on NPY-LI in this brain region and there wasno interaction between the two factors (Fig. 1c).

In the periaqueductal grey NPY-LI was marginally higher inthe FSL animals (F(1,50)=3.576, p=0.065) and this differencewas most pronounced in the control groups (p=0.038) asmaternal separation tended to increase NPY-LI in this brainregion only in the FRL animals. Maternal separation had nosignificant effect on NPY-LI in this brain region and there wasno interaction between the two factors (Fig. 1d). NPY-LIconcentrations were not affected by strain or maternaldeprivation in other investigated regions.

3.3. Calcitonin gene-related peptide

In the frontal cortex CGRP-LI levels were significantlyhigher in FSL animals (F(1,44)=19.380, p=0.00007). Thisdifference was mainly due to the control animals (p=0.0001),

while the same trend was observed in maternally separatedanimals (p=0.052). There was no significant effect of maternaldeprivation and no interaction between the two factors (Fig. 2a).

In the hippocampus CGRP-LI levels were significantlyhigher in FSL animals (F(1,49)=22.681, p=0.00002). Thisdifference was observed in both control and maternallyseparated animals (p=0.002 and p=0.006, respectively).Maternal separation increased CGRP-LI levels in this region(F(1,49)=7.078, p=0.011). There was no significant interactionbetween the two factors, however separate analysis of the twostrains revealed that the effect of maternal separation wasstrongest in the FSL animals (p=0.060) (Fig. 2b).

In the amygdala CGRP-LI levels were significantlyaffected by maternal deprivation (F(1,47)=12.448, p=0.001)and there was a significant interaction between the factorsstrain and maternal deprivation (F(1,47)=7.682, p=0.008).Post-hoc analysis revealed that maternal deprivation increasedamygdala CGRP-LI levels in FRL animals only (p=0.00001)(Fig. 2c).

In the periaqueductal grey CGRP-LI levels were signifi-cantly higher in FSL animals (F(1,42)=21.641, p=0.00003).

Fig. 2. Effects of maternal separation on CGRP-LI concentration in brain regions of “genetically depressed” Flinders Sensitive Line and control strain FlindersResistant Line rats. Represented are means±S.E.M. MS180: maternal separation from postnatal day 2 to 14 for 180 min per day. CONTROL: no manipulation frompostnatal day 2 to 14. ⁎⁎⁎ (P<0.001) indicates a strain difference (FSL differ from FRL, ANOVA). § indicates a maternal separation effect (MS180 differ fromCONTROL of the same strain, Tukey tests, (§)P<0.1, §§§P<0.001).

688 G. Wörtwein et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 30 (2006) 684–693

This difference was mainly due to the two control groups(p=0.00001). Maternal separation had no significant effect onCGRP-LI in this brain region and there was no interaction

Table 1Coefficient of correlation between NPY-LI and CGRP-LI in regions of the ratbrain

Region All animals

N Pearson's r P-value

Hippocampus 53 −0.03313 0.0154Frontal cortex 48 −0.2982 0.0395Occipital cortex 47 −0.0134 0.9285Entorhinal cortex 33 0.0682 0.7057Temporal cortex 53 0.1861 0.1865Amygdala 51 0.3054 0.0311Caudate/putamen 51 0.3069 0.0285Nucleus accumbens 52 0.3691 0.0071Periaqueductal grey 50 0.5051 0.0003Hypothalamus 54 0.6959 0.0000

between the two factors (Fig. 2d). CGRP-LI concentrationswere not affected by strain or maternal deprivation in otherregions investigated.

Table 2Coefficient of correlation between NPY-LI and CGRP-LI in regions of the ratbrain for the two strains separately

Region FRL FSL

N Pearson's r P-value N Pearson's r P-value

Hippocampus 31 −0.1911 0.3032 22 −0.3144 0.1541Frontal cortex 29 −0.0813 0.6749 19 −0.2866 0.2342Occipital cortex 26 0.1905 0.3511 21 −0.2444 0.2857Entorhinal cortex 17 0.1038 0.6919 16 −0.0039 0.9886Temporal cortex 30 0.0181 0.9243 22 −0.3468 0.1138Amygdala 28 0.1817 0.3548 22 −0.5059 0.0163Caudate/putamen 31 0.0610 0.7446 20 −0.4780 0.0330Nucleus accumbens 32 0.3199 0.0743 20 −0.5840 0.0069Periaqueductal grey 27 0.7635 0.0000 19 −0.2130 0.3814Hypothalamus 29 0.7147 0.0000 21 −0.7094 0.0003

689G. Wörtwein et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 30 (2006) 684–693

3.4. Correlations

In the hippocampus and frontal cortex NPY-LI and CGRP-LIshowed a significant negative correlation (Table 1). In theamygdala, caudate/putamen, nucleus accumbens, periaqueduc-tal grey, and hypothalamus the two peptides showed asignificant positive correlation (Table 1). When the correlationswere calculated for the two strains separately, it was found thatNPY-LI and CGRP-LI showed a stronger negative correlation inthe frontal cortex and hippocampus of the “depressed” FSL rats(Table 2).

4. Discussion

To the best of our knowledge this is the first animal modelbased study of gene–environment interaction where early lifetrauma was superimposed onto a genetic model of depression/susceptibility to depression. We investigated effects on NPYandCGRP concentrations in discrete brain regions in adult life. Inthe hippocampus, NPY-LI was reduced in “geneticallydepressed” FSL animals, compared to the FRL strain, whichis in agreement with our previous findings (Caberlotto et al.,1998; Jimenez Vasquez et al., 2000a,b; Husum et al., 2001,2003). This strain difference resided with the control that is non-maternally separated animals. In animals that were subjected tomaternal deprivation, the FSL–FRL difference was no longersignificant. This is consistent with earlier results, showing thatmaternal deprivation reduced NPY-LI in the hippocampus ofadult, genetically normal animals (Jimenez-Vasquez et al.,2001; Husum and Mathé, 2002; Lim et al., 2003). Also inagreement with these results is the finding that the stress ofrepeated injections/handling lowered hippocampal NPY-LImore in adult FRL than FSL animals (Husum et al., 2003).The decreased hippocampal NPY-LI concentrations in FSL ratsare consistent with previous findings in both FSL and FawnHooded rats, another genetic animal model of depression(Mathé et al., 1998; Caberlotto et al., 1999; Jimenez Vasquez etal., 2000a,b; Husum et al., 2001, 2003), and with some humanstudies showing reduced NPY-LI concentrations in plasma andCSF from depressed patients. (Widerlöv et al., 1988; Gjerris etal., 1992; Nilsson et al., 1996; Hashimoto et al., 1996; Mathé etal., 1996b; Westrin et al., 1999; Heilig et al., 2004). It has beensuggested that hippocampal NPY has an antistress action sincerats overexpressing NPY in the hippocampus are less sensitiveto stress and fear (Thorsell et al., 2000). As a corollary, thereverse is also likely to be operative, namely lower hippocampalNPY levels render the FSL rats more sensitive to stress. Thebiological correlates of such a behaviorally enhanced reactivityto stress are not clear. However, independent data from severallaboratories show brain region specific interrelationshipsbetween dopamine and NPY as well as dopamine and CGRP(which will be discussed in the context of the CGRP results). Insome brain regions there appears to be an inverse dopamine–NPY relationship (that is they inhibit each others release) whilein others they appear to act in parallel (for review seeObuchowicz et al., 2004). Altered dopamine release andmetabolism have been reported in FSL compared to normal

control rats (Zangen et al., 1997, 1999, 2001; Dremencov et al.,2004) and thus it is conceivable that a dysregulated NPY–dopamine interaction will account for the observed phenomena.Currently, no data are available to allow reasonable speculationsas to whether NPY or dopamine might constitute the primarychange.

In the frontal cortex, baseline levels of NPY-LI in FSLanimals were also lowered in line with one but not all earlierfindings (Jimenez Vasquez et al., 2000a,b; Husum et al., 2001,2003). Maternal deprivation was previously found to have noeffect or to lower NPY-LI in the frontal cortex of normal rats(Jimenez-Vasquez et al., 2001; Husum et al., 2002). The reasonfor these discrepancies may involve differences in theseparation paradigms and dissection techniques used.

In the caudate/putamen of FSL animals NPY-LI waselevated, in contrast to the decreased NPY-LI concentrationsfound in hippocampus and frontal cortex. This finding is inagreement with some but not all previous reports (JimenezVasquez et al., 2000a; Husum et al., 2001, 2003). As the fronto-subcortical circuits, particularly the limbic–striatal–pallidal–thalamic and the limbic–thalamo–cortical circuit have beenimplicated in the pathophysiology of depression (Cummings,1993), our finding of decreased NPY-LI in the frontal cortex andincreased NPY-LI in the caudate/putamen of the FSL strainfurther supports the hypothesis of a dysfunction of the NPYsystem in depression.

Levels of NPY-LI in the periaqueductal grey of an animalmodel of depression have not previously been reported.Interestingly, NPY-LI concentrations were elevated in thisregion in FSL animals, compared to their FRL counterparts.This brain region is believed to be involved in the neurobiologyof anxiety disorders, specifically panic anxiety (Graeff et al.,1993), and has been implicated in NPY's effects on anxiety-likebehavior (Kask et al., 2002). In one publication, FSL rats werereported to display normal anxiety like behavior (Schiller et al.,1991) and, consequently, confirmation of both behavioral andbiochemical results is necessary to further evaluate this issue. Inaddition, this finding could also be discussed in the context ofnociception, which has been linked via chronic stress-inducedHPA dysfunction to depressive symptomatology (Blackburn-Munro and Blackburn-Munro, 2001). It is possible that the FSLstrain is hypersensitive to pain, as NPY possesses antinocicep-tive effects in the PAG of normal rats (Wang et al., 2000). Thussummarizing the NPY results, by employing a finer dissectionprocedure and studying more brain regions we have replicatedand extended previous data regarding baseline NPY-LI straindifferences between the “genetically depressed” FSL andcontrol FRL rats. Another finding was that maternal deprivationdid not increase the strain differences in NPY-LI concentrationsin the hippocampus; in fact it had only minimal effect in the FSLanimals.

With regard to CGRP, this is the first paper reportingmeasurements of CGRP-LI in selected brain regions of a geneticanimal model of depression. In similarity to other neuropep-tides, strain differences were observed; CGRP-LI levels tendedto be higher in the FSL rats. The stress of early maternaldeprivation had region specific, selective effects resulting in

690 G. Wörtwein et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 30 (2006) 684–693

further increase in the CGRP-LI differences between the FSLand FRL animals, which is consistent with the reportedsensitivity of CGRP to stress in genetically normal animals(Guidobono et al., 1991; Husum et al., 2002). CGRP-LI washigher in the frontal cortex and hippocampus of the FSLanimals. These changes were in the opposite direction to thoseobserved for NPY-LI. CGRP-LI and NPY-LI were significantlynegatively correlated in these brain areas. Maternal deprivationadditionally increased the CGRP-LI levels in the hippocampusof the FSL animals. Our findings are therefore consistent with ahypothesis of additive effects of the genetic background and theenvironmental manipulation on CGRP-LI in the hippocampus,a brain region implicated in the neurobiology of depression.CGRP-LI was also elevated in the periaqueductal grey of theFSL animals, but maternal deprivation had no further effect. Incontrast, in the amygdala maternal deprivation affected CGRP-LI only in the FRL strain. This points to specific roles for thisneuropeptide in different brain regions, implicated in theneurobiology of depression and anxiety, respectively.

NPY and CGRP appear to exert opposing behavioraleffects in some models. While NPY is an orexigenic andanxiolytic peptide (Stanley and Leibowitz, 1984; Levine andMorley, 1984; Heilig et al., 1989; Sørensen et al., 2004),CGRP potentiates fear related behaviors (Kovacs and Telegdy,1994, 1995; Poore and Helmstetter, 1996) and has anorecticeffects (Lutz et al., 1997). In the hippocampus and frontalcortex NPY and CGRP interact with dopamine neurotrans-mission in a complex but often reciprocal manner. ICVadministration of NPY increased the tissue concentration ofdopamine in the frontal cortex of genetically normal animals,while icv CGRP had no effect (Heilig et al., 1990;Drumheller et al., 1992). ICV administration of NPY alsoincreased the tissue concentration of dopamine in thehippocampus (Drumheller et al., 1994). Dopaminergic deaf-ferentation decreased expression of NPY mRNA in frontalcortex, suggesting that dopamine stimulates NPY expressionin this region (Lindefors et al., 1990), however two otherstudies found no effect of a 6-hydroxydopamine inducedlesion of dopaminergic terminals or reserpine treatment onNPY-LI protein levels in the frontal cortex or hippocampus(Schon et al., 1986; Maeda et al., 1993). Acute administrationof amphetamine increased NPY and CGRP efflux in thefrontal cortex and CGRP efflux in the hippocampus but hadno effect on NPY tissue levels in the hippocampus (Mathé etal., 1996a, unpublished data). Chronic treatment with indirectdopaminergic agonists downregulated NPY in the frontalcortex and hippocampus (Wahlestedt et al., 1991; Maeda etal., 1993). The effect of such a treatment regimen on CGRPhas not been investigated. Chronic treatment with dopami-nergic antagonists does not affect NPY levels in thehippocampus, but upregulates CGRP levels (Angelucci etal., 2001; Obuchowicz et al., 2004). The reverse is true forthe frontal cortex, where chronic treatment with dopaminergicantagonists leaves CGRP levels unaffected but upregulatesNPY-LI (Angelucci et al., 2001; Obuchowicz et al., 2004).FSL rats show increased tyrosine hydroxylase mRNA levelsin the ventral tegmental area (Serova et al., 1998), higher

dopamine levels in the hippocampus, and higher dopamineturnover in the prefrontal cortex (Zangen et al., 1999). On thebasis of this evidence we speculate that our finding of astronger negative correlation between CGRP-LI and NPY-LIin frontal cortex and hippocampus in “depressed” FSL ratscompared to their FRL controls might be related to thedopamine abnormalities observed in FSL rats and might be animportant aspect of their phenotype.

5. Conclusions

To the best of our knowledge this is the first study of gene–environment interaction employing a paradigm of earlymaternal deprivation superimposed on a model of geneticvulnerability to depression and investigating effects on brainneurochemistry in adult life. Our results indicate that alteredNPY levels in selected brain regions are a hallmark ofdepression and that genetic vulnerability might exert a greatereffect than environmental factors on NPY levels. We alsoprovide evidence for altered CGRP in the same model. The twomost salient points are (1) in brain regions important forprocessing and expression of emotion and for cognitivefunction, the hippocampus and frontal cortex, respectively, thechanges in CGRP-LI were in the opposite direction to thosefound for NPY, and (2) CGRP appears to be more responsive toadverse environmental manipulation. This divergence betweenNPY and CGRP in direction of change and susceptibility togenetic versus environmental factors is in line with previouslyreported plasma and CSF levels in depressed patients (Widerlövet al., 1988; Gjerris et al., 1992; Mathé et al., 1994; Nilsson etal., 1996; Hashimoto et al., 1996; Westrin et al., 1999; Heilig etal., 2004) as well as with the observation that all antidepressanttreatment modalities examined so far elevate NPY but notCGRP (Stenfors et al., 1989; Mathé et al., 1990a, 1998; JimenezVasquez et al., 2000a; Husum et al., 2003). Our findings thussupport the hypothesis that genetic disposition and develop-mental stress may contribute to the susceptibility to depressionby exerting selective neuropeptide- and brain region-specificeffects on adult neurobiology.

Acknowledgments

Supported by the Swedish Medical Research Council grant10414, The Lundbeck Foundation, and Karolinska Institutet.We thank Anders Knall for expert technical assistance.

References

Agid O, Shapira B, Zislin J, Ritsner M, Hanin B, Murad H, et al. Environmentand vulnerability to major psychiatric illness: a case control study of earlyparental loss in major depression, bipolar disorder and schizophrenia. MolPsychiatry 1999;4:163–72.

Agelucci F, Gruber SH, Mathé AA. A pilot study of rat brain regionaldistribution of calcitonin, katacalcin and calcitonin gene-related peptidebefore and after antipsychotic treatment. Neuropeptides 2001;35:285–91.

Baker RA, Herkenham M, Brady LS. Effects of long-term treatment withantidepressant drugs on proopiomelanocortin and neuropeptide Y mRNAexpression in the hypothalamic arcuate nucleus of rats. J Neuroendocrinol1996;8:337–43.

691G. Wörtwein et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 30 (2006) 684–693

Blackburn-Munro G, Blackburn-Munro RE. Chronic pain, chronic stress anddepression: coincidence or consequence? J Neuroendocrinol 2001;13:1009–23.

Caberlotto L, Fuxe K, Overstreet DH, Gerrard P, Hurd YL. Alterations inneuropeptide Yand Y1 receptor mRNA expression in brains from an animalmodel of depression: region specific adaptation after fluoxetine treatment.Brain Res Mol Brain Res 1998;59:58–65.

Caberlotto L, Jimenez P, Overstreet DH, Hurd YL, Mathé AA, Fuxe K.Alterations in neuropeptide Y levels and Y1 binding sites in the FlindersSensitive Line rats, a genetic animal model of depression. Neurosci Lett1999;265:191–4.

Clementi G, Grassi M, Valerio C, Prato A, Fiore CE, Drago F. Effects ofcalcitonin gene-related peptide on extrapyramidal motor system. PharmacolBiochem Behav 1992;42:545–8.

Cummings JL. The neuroanatomy of depression. J Clin Psychiatry1993;54:14–20 [Suppl].

Deutch AY, Roth RH. Calcitonin gene-related peptide in the ventral tegmentalarea: selective modulation of prefrontal cortical dopamine metabolism.Neurosci Lett 1987;74:169–74.

Dremencov E, Gispan-Herman I, Rosenstein M, Mendelman A, Overstreet DH,Zohar J, et al. The serotonin–dopamine interaction is critical for fast-onsetaction of antidepressant treatment: in vivo studies in an animal model ofdepression. Prog Neuropsychopharmacol Biol Psychiatry 2004;28:141–7.

Drumheller A, Menard D, Fournier A, Jolicoeur FB. Neurochemical effects ofCGRP. Ann N YAcad Sci 1992;657:546–8.

Drumheller A, Bouali SM, Fournier A, St Pierre S, Jolicoeur FB. Neurochemicaleffects of neuropeptide Y (NPY) and NPY2-36. Neuropeptides 1994;27:291–6.

El Khoury A, Gruber SHM, Mørk A, Mathé AA. Adult life behavioralconsequences of early maternal separation are alleviated by escitalopramtreatment in a rat model of depression. Prog Neuropsychopharmacol BiolPsychiatry in press [Electronic publication ahead of print, Jan 13].

Gjerris A, Widerlöv E, Werdelin L, Ekman R. Cerebrospinal fluid concentra-tions of neuropeptide Y in depressed patients and in controls. J PsychiatryNeurosci 1992;17:23–7.

Graeff FG, Silveira MC, Nogueira RL, Audi EA, Oliveira RM. Role of theamygdala and periaqueductal gray in anxiety and panic. Behav Brain Res1993;58:123–31.

Gruber SH, Mathé AA. Effects of typical and atypical antipsychotics onneuropeptide Y in rat brain tissue and microdialysates from ventral striatum.J Neurosci Res 2000;61:458–63.

Gruber SH, Nomikos GG, Mathé AA. Dopamine receptor antagonists preventthe D-amphetamine-induced increase in calcitonin gene-related peptidelevels in ventral striatum. J Neurosci Res 2001;64:606–11.

Gruber SH, Nomikos GG, Mathé AA. D-Amphetamine-induced increase inneurotensin and neuropeptide Y outflow in the ventral striatum is mediatedvia stimulation of dopamine D1 and D2/3 receptors. J Neurosci Res2002;69:133–9.

Guidobono F, Netti C, Pecile A, Gritti I, Mancia M. Stress-related changes incalcitonin gene-related peptide binding sites in the cat central nervoussystem. Neuropeptides 1991;19:57–63.

Harrigan EA, Magnuson DJ, Thunstedt GM, Gray TS. Corticotropin releasingfactor neurons are innervated by calcitonin gene-related peptide terminals inthe rat central amygdaloid nucleus. Brain Res Bull 1994;33:529–34.

Hashimoto H, Onishi H, Koide S, Kai T, Yamagami S. Plasma neuropeptide Y inpatients with major depressive disorder. Neurosci Lett 1996;216:57–60.

Heilig M, Widerlöv E. Neurobiology and clinical aspects of neuropeptide Y. CritRev Neurobiol 1995;9:115–36.

Heilig M, Wahlestedt C, Ekman R, Widerlöv E. Antidepressant drugs increasethe concentration of neuropeptide Y (NPY)-like immunoreactivity in the ratbrain. Eur J Pharmacol 1988;147:465–7.

Heilig M, Soderpalm B, Engel JA, Widerlöv E. Centrally administeredneuropeptide Y (NPY) produces anxiolytic-like effects in animal anxietymodels. Psychopharmacology (Berl) 1989;98:524–9.

Heilig M, Vecsei L, Wahlestedt C, Alling C, Widerlöv E. Effects of centrallyadministered neuropeptide Y (NPY) and NPY13–36 on the brainmonoaminergic systems of the rat. J Neural Transm Gen Sect 1990;79:193–208.

Heilig M, Zachrisson O, Thorsell A, Ehnvall A, Mottagui-Tabar S, Sjögren M,et al. Decreased cerebrospinal fluid neuropeptide Y (NPY) in patients withtreatment refractory unipolar major depression: preliminary evidence forassociation with preproNPY gene polymorphism. J Psychiatr Res2004;38:113–21.

Heim C, Nemeroff CB. The impact of early adverse experiences on brainsystems involved in the pathophysiology of anxiety and affective disorders.Biol Psychiatry 1999;46:1509–22.

Holmes PV, Davis RC, Masini CV, Primeaux SD. Effects of olfactorybulbectomy on neuropeptide gene expression in the rat olfactory/limbicsystem. Neuroscience 1998;86:587–96.

Honkaniemi J, Pelto-Huikko M, Isola J, Rechardt L. Simultaneous localizationof calcitonin gene-related peptide and neurotensin in rat central amygdaloidnucleus. Neurosci Lett 1990;113:1–6.

HusumH,Mathé AA. Early life stress changes concentrations of neuropeptide Yand corticotropin-releasing hormone in adult rat brain. Lithium treatmentmodifies these changes. Neuropsychopharmacology 2002;27:756–64.

Husum H, Mikkelsen JD, Hogg S, Mathé AA, Mörk A. Involvement ofhippocampal neuropeptide Y in mediating the chronic actions of lithium,electroconvulsive stimulation and citalopram. Neuropharmacology2000;39:1463–73.

Husum H, Vasquez PA, Mathé AA. Changed concentrations of tachykinins andneuropeptide Y in brain of a rat model of depression: lithium treatmentnormalizes tachykinins. Neuropsychopharmacology 2001;24:183–91.

Husum H, Termeer E, Mathé AA, Bolwig TG, Ellenbroek BA. Early maternaldeprivation alters hippocampal levels of neuropeptide Yand calcitonin-generelated peptide in adult rats. Neuropharmacology 2002;42:798–806.

Husum H, Van Kammen D, Termeer E, Bolwig G, Mathé AA. Topiramatenormalizes hippocampal NPY-LI in flinders sensitive line ‘depressed’ ratsand upregulates NPY, galanin, and CRH-LI in the hypothalamus:implications for mood-stabilizing and weight loss-inducing effects.Neuropsychopharmacology 2003;28:1292–9.

Jimenez Vasquez PA, Salmi P, Ahlenius S, Mathé AA. Neuropeptide Y in brainsof the Flinders Sensitive Line rat, a model of depression. Effects ofelectroconvulsive stimuli and D-amphetamine on peptide concentrations andlocomotion. Behav Brain Res 2000a;111:115–23.

Jimenez-Vasquez PA, Overstreet DH, Mathé AA. Neuropeptide Y in male andfemale brains of Flinders Sensitive Line, a rat model of depression. Effectsof electroconvulsive stimuli. J Psychiatr Res 2000b;34:405–12.

Jimenez-Vasquez PA, Mathé AA, Thomas JD, Riley EP, Ehlers CL. Earlymaternal separation alters neuropeptide Y concentrations in selected brainregions in adult rats. Brain Res Dev Brain Res 2001;131:149–52.

Jolicoeur FB, Menard D, Fournier A, St Pierre S. Structure–activity analysis ofCGRP's neurobehavioral effects. Ann N YAcad Sci 1992;657:155–63.

Ju G. Gene-related peptide-like immunoreactivity and its relation withneurotensin- and corticotropin-releasing hormone-like immunoreactiveneurons in the bed nuclei of the stria terminalis in the rat. Brain Res Bull1991;27:617–24.

Kask A, Harro J, von Horsten S, Redrobe JP, Dumont Y, Quirion R. Theneurocircuitry and receptor subtypes mediating anxiolytic-like effects ofneuropeptide Y. Neurosci Biobehav Rev 2002;26:259–83.

Kendler KS, Neale MC, Kessler RC, Heath AC, Eaves LJ. Childhood parentalloss and adult psychopathology in women. A twin study perspective. ArchGen Psychiatry 1992;49:109–16.

Kerkerian-Le Goff L, Forni C, Samuel D, Bloc A, Dusticier N, Nieoullon A.Intracerebroventricular administration of neuropeptide Y affects parametersof dopamine, glutamate and GABA activities in the rat striatum. Brain ResBull 1992;28:187–93.

Kovacs A, Telegdy G. Behavioural impairment induced by calcitonin gene-related peptide (CGRP) antiserum in passive avoidance reflex in rats.Neuropeptides 1994;26:233–6.

Kovacs A, Telegdy G. Effects of CGRP on active avoidance behavior in rats.Physiol Behav 1995;58:429–35.

Kovacs A, Telegdy G, Toth G, Penke B. Neurotransmitter-mediated open-fieldbehavioral action of CGRP. Life Sci 1999;64:733–40.

Kresse A, Jacobowitz DM, Skofitsch G. Detailed mapping of CGRP mRNAexpression in the rat central nervous system: comparison with previousimmunocytochemical findings. Brain Res Bull 1995;36:261–74.

692 G. Wörtwein et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 30 (2006) 684–693

Levine AS, Morley JE. Neuropeptide Y: a potent inducer of consummatorybehavior in rats. Peptides 1984;5:1025–9.

Lim S, Ryu YH, Kim ST, Hong MS, Park HJ. Acupuncture increasesneuropeptide Y expression in hippocampus of maternally-separated rats.Neurosci Lett 2003;343:49–52.

Lindefors N, Brene S, Herrera-Marschitz M, Persson H. Regulation ofneuropeptide Y gene expression in rat brain. Ann N Y Acad Sci1990;611:175–85.

Liu D, Diorio J, Day JC, Francis DD, Meaney MJ. Maternal care, hippocampalsynaptogenesis and cognitive development in rats. Nat Neurosci2000;3:799–806.

Lutz TA, Rossi R, Althaus J, Del Prete E, Scharrer E. Evidence for aphysiological role of central calcitonin gene-related peptide (CGRP)receptors in the control of food intake in rats. Neurosci Lett1997;230:159–62.

Maeda K, Kawata E, Sakai K, Chihara K. Effects of putative cognitive function-enhancing drugs and dopaminergic agents on somatostatin and neuropeptideY in rat brain. Eur J Pharmacol 1993;233:227–35.

Mathé AA. Neuropeptides and electroconvulsive treatment. J ECT1999;15:60–75.

Mathé AA, Gruber SHM. Neuropeptide Y has marked antidepressant propertiesin a rat model of depression. Neuropsychopharmacology 2004;29(Suppl):s154.

Mathé AA, Jousisto-Hanson J, Stenfors C, Theodorsson E. Effect of lithium ontachykinins, calcitonin gene-related peptide, and neuropeptide Y in rat brain.J Neurosci Res 1990a;26:233–7.

Mathé AA, Stenfors C, Brodin E, Theodorsson E. Neuropeptides in brain:effects of microwave irradiation and decapitation. Life Sci 1990b;46:287–93.

Mathé AA, Ågren H, Lindström L, Theodorsson E. Increased concentration ofcalcitonin gene-related peptide in cerebrospinal fluid of depressed patients.A possible trait marker of major depressive disorder. Neurosci Lett1994;182:138–42.

Mathé AA, Hertel P, Nomikos GG, Gruber S, Mathé JM, Svensson TH. Thepsychotomimetic drugs D-amphetamine and phencyclidine release calcitoningene-related peptide in the limbic forebrain of the rat. J Neurosci Res1996a;46:316–23.

Mathé AA, Rudorfer MV, Stenfors C, Manji HK, Potter WZ,Theodorsson E. Effects of electroconvulsive treatment on somatostatin,neuropeptide Y, endothelin, and neurokinin A concentrations in cere-brospinal fluid of depressed patients: a pilot study. Depression 1996b;3:250–6.

Mathé AA, Gruber S, Jimenez PA, Theodorsson E, Stenfors C. Effects ofelectroconvulsive stimuli and MK-801 on neuropeptide Y, neurokinin A,and calcitonin gene-related peptide in rat brain. Neurochem Res1997;22:629–36.

Mathé AA, Jimenez PA, Theodorsson E, Stenfors C. Neuropeptide Y,neurokinin A and neurotensin in brain regions of Fawn Hooded “depressed”,Wistar, and Sprague Dawley rats. Effects of electroconvulsive stimuli. ProgNeuropsychopharmacol Biol Psychiatry 1998;22:529–46.

Mathé AA, Ågren H, Wallin A, Blennow K. Calcitonin gene-related peptideand calcitonin in the CSF of patients with dementia and depression:possible disease markers. Prog Neuropsychopharmacol Biol Psychiatry2002;26:41–8.

Mathé AA, AndersonW, Angelucci F, El Khoury AE, Gruber S, Husum H, et al.Search for biological correlates of depression and mechanisms of action ofantidepressant treatment modalities. Do neuropeptides play a role? PhysiolBrain [in press].

Nikisch G, Ågren H, Chin B, Czernik A, Baumann P, Mathé AA. NeuropeptideY and corticotropin-releasing hormone in CSF mark response toantidepressive treatment with citalopram. Int J Neuropsychopharmacol2005;8:403–10.

Nilsson C, Karlsson G, Blennow K, Heilig M, Ekman R. Differences in theneuropeptide Y-like immunoreactivity of the plasma and platelets of humanvolunteers and depressed patients. Peptides 1996;17:359–62.

Obuchowicz E, Krysiak R, Herman ZS. Does neuropeptide Y (NPY) mediatethe effects of psychotropic drugs? Neurosci Biobehav Rev 2004;28:595–610.

Orazzo C, Pieribone VA, Ceccatelli S, Terenius L, Hökfelt T. CGRP-likeimmunoreactivity in A11 dopamine neurons projecting to the spinal cordand a note on CGRP-CCK cross-reactivity. Brain Res 1993;600:39–48.

Overstreet DH. The Flinders Sensitive Line rats: a genetic animal model ofdepression. Neurosci Biobehav Rev 1993;17:51–68.

Overstreet DH. Behavioral characteristics of rat lines selected for differentialhypothermic responses to cholinergic or serotonergic agonists. Behav Genet2002;32:335–48.

Overstreet DH, Russell RW, Helps SC, Messenger M. Selective breeding forsensitivity to the anticholinesterase DFP. Psychopharmacology (Berl)1979;65:15–20.

Overstreet DH, Pucilowski O, Rezvani AH, Janowsky DS. Administration ofantidepressants, diazepam and psychomotor stimulants further confirms theutility of Flinders Sensitive Line rats as an animal model of depression.Psychopharmacology (Berl) 1995;121:27–37.

Overstreet DH, Keeney A, Hogg S. Antidepressant effects of citalopram andCRF receptor antagonist CP-154,526 in a rat model of depression. Eur JPharmacol 2004;492:195–201.

Poore LH, Helmstetter FJ. The effects of central injections of calcitoningene-related peptide on fear-related behavior. Neurobiol Learn Mem1996;66:241–5.

Post RM. Transduction of psychosocial stress into the neurobiology of recurrentaffective disorder. Am J Psychiatry 1992;149:999–1010.

Redrobe JP, Dumont Y, Fournier A, Quirion R. The neuropeptide Y (NPY) Y1receptor subtype mediates NPY-induced antidepressant-like activity in themouse forced swimming test. Neuropsychopharmacology 2002;26:615–24.

Sapolsky RM, Meaney MJ. Maturation of the adrenocortical stress response:neuroendocrine control mechanisms and the stress hyporesponsive period.Brain Res 1986;396:64–76.

Schiller GD, Daws LC, Overstreet DH, Orbach J. Lack of anxiety in an animalmodel of depression with cholinergic supersensitivity. Brain Res Bull1991;26:433–5.

Schiller GD, Pucilowski O, Wienicke C, Overstreet DH. Immobility-reducingeffects of antidepressants in a genetic animal model of depression. Brain ResBull 1992;28:821–3.

Schon F, Allen JM, Yeats JC, Kent A, Kelly JS, Bloom SR. The effect of 6-hydroxydopamine, reserpine and cold stress on the neuropeptide Y contentof the rat central nervous system. Neuroscience 1986;19:1247–50.

Serova L, Sabban EL, Zangen A, Overstreet DH, Yadid G. Altered geneexpression for catecholamine biosynthetic enzymes and stress response inrat genetic model of depression. Brain Res Mol Brain Res 1998;63:133–8.

Servant D, Parquet PJ. Early life events and panic disorder: course of illness andcomorbidity. Prog Neuropsychopharmacol Biol Psychiatry 1994;18:373–9.

Sørensen G, Lindberg C, Wörtwein G, Bolwig TG, Woldbye DP. Differentialroles for neuropeptide Y Y1 and Y5 receptors in anxiety and sedation. JNeurosci Res 2004;77:723–9.

SPSS Inc.CIU. SPSS base version 11.5 user manual. Chicago (IL, USA): SPSSInc.; 2004.

Stanley BG, Leibowitz SF. Neuropeptide Y: stimulation of feeding and drinkingby injection into the paraventricular nucleus. Life Sci 1984;35:2635–42.

Stenfors C, Theodorsson E, Mathé AA. Effect of repeated electroconvulsivetreatment on regional concentrations of tachykinins, neurotensin, vasoactiveintestinal polypeptide, neuropeptide Y, and galanin in rat brain. J NeurosciRes 1989;24:445–50.

Stenfors C, Mathé AA, Theodorsson E. Repeated electroconvulsive stimuli:changes in neuropeptide Y, neurotensin and tachykinin concentrations intime. Prog Neuropsychopharmacol Biol Psychiatry 1994;18:201–9.

Stogner KA, Holmes PV. Neuropeptide-Y exerts antidepressant-like effects inthe forced swim test in rats. Eur J Pharmacol 2000;387:R9–R10.

Thorsell A, Michalkiewicz M, Dumont Y, Quirion R, Caberlotto L, RimondiniR, et al. Behavioral insensitivity to restraint stress, absent fear suppression ofbehavior and impaired spatial learning in transgenic rats with hippocampalneuropeptide Y overexpression. Proc Natl Acad Sci U S A 2000;97:12852–7.

van Rossum D, Hanisch UK, Quirion R. Neuroanatomical localization,pharmacological characterization and functions of CGRP, related peptidesand their receptors. Neurosci Biobehav Rev 1997;21:649–78.

693G. Wörtwein et al. / Progress in Neuro-Psychopharmacology & Biological Psychiatry 30 (2006) 684–693

Wahlestedt C, Karoum F, Jaskiw G, Wyatt RJ, Larhammar D, Ekman R, et al.Cocaine-induced reduction of brain neuropeptide Y synthesis dependent onmedial prefrontal cortex. Proc Natl Acad Sci U S A 1991;88:2078–82.

Wang JZ, Lundeberg T, Yu L. Antinociceptive effects induced by intra-periaqueductal grey administration of neuropeptide Y in rats. Brain Res2000;859:361–3.

Weaver IC, Cervoni N, Champagne FA, D'Alessio AC, Sharma S, Seckl JR,et al. Epigenetic programming by maternal behavior. Nat Neurosci 2004;7:847–54.

Westrin A, Ekman R, Träskman-Bendz L. Alterations of corticotropin releasinghormone (CRH) and neuropeptide Y (NPY) plasma levels in mood disorderpatients with a recent suicide attempt. Eur Neuropsychopharmacol1999;9:205–11.

Widerlöv E, Lindström LH, Wahlestedt C, Ekman R. Neuropeptide Y andpeptide YYas possible cerebrospinal fluid markers for major depression andschizophrenia, respectively. J Psychiatr Res 1988;22:69–79.

Wimalawansa SJ. Calcitonin gene-related peptide and its receptors: moleculargenetics, physiology, pathophysiology, and therapeutic potentials. EndocrRev 1996;17:533–85.

Wimalawansa SJ. Amylin, calcitonin gene-related peptide, calcitonin, andadrenomedullin: a peptide superfamily. Crit Rev Neurobiol 1997;11:167–239.

Woldbye DP, Larsen PJ, Mikkelsen JD, Klemp K, Madsen TM, Bolwig TG.Powerful inhibition of kainic acid seizures by neuropeptide Y via Y5-likereceptors. Nat Med 1997;3:761–4.

Zachrisson O, Mathé AA, Stenfors C, Lindefors N. Limbic effects of repeatedelectroconvulsive stimulation on neuropeptide Y and somatostatin mRNAexpression in the rat brain. Brain Res Mol Brain Res 1995a;31:71–85.

Zachrisson O, Mathé AA, Stenfors C, Lindefors N. Region-specific effects ofchronic lithium administration on neuropeptide Y and somatostatin mRNAexpression in the rat brain. Neurosci Lett 1995b;194:89–92.

Zangen A, Overstreet DH, Yadid G. High serotonin and 5-hydroxyindoleaceticacid levels in limbic brain regions in a rat model of depression:normalization by chronic antidepressant treatment. J Neurochem1997;69:2477–83.

Zangen A, Overstreet DH, Yadid G. Increased catecholamine levels in specificbrain regions of a rat model of depression: normalization by chronicantidepressant treatment. Brain Res 1999;824:243–50.

Zangen A, Nakash R, Overstreet DH, Yadid G. Association between depressivebehavior and absence of serotonin–dopamine interaction in the nucleusaccumbens. Psychopharmacology (Berl) 2001;155:434–9.