Journal of Affective Disorders 75 (2003) 65–69www.elsevier.com/ locate/ jad

Brief report

Studies on serotonergic markers in the human hippocampus:changes in subjects with bipolar disorder

*Brian Dean , Elizabeth Scarr, Geoffrey Pavey, David Copolov

The Rebecca L. Cooper Research Laboratories, The Mental Health Research Institute of Victoria, Parkville, Victoria, Australia

Received 23 April 2001; received in revised form 18 December 2001; accepted 4 January 2002

Abstract

Background: Various studies suggest the hippocampus and serotonergic systems are important in the pathology of bipolardisorder (BD). We therefore measured hippocampal serotonergic markers in post-mortem tissue from BD and control

3subjects.Methods: The density and affinity of [ H]citalopram binding to the serotonin transporter (SERT), as well as thedensity of the 5HT , 5HT , 5HT and 5HT receptors were measured.Results: The density of SERT and 5HT receptors2A 1A 1D 1F

3was no different in BD. There was a significant decrease in the affinity of [ H]citalopram binding to SERT in the stratumlacunosum-moleculare (S ) in BD (K mean6S.E.M.5 4.360.8 vs. 1.960.3 nM). Limitations: This study was completedlac d

using relatively small cohorts.Conclusions: There are no generalised changes in hippocampal serotonergic markers in thehippocampus from subjects with BD. There is a decreased affinity of radioligand binding toS SERT in subjects with BD.lac

2002 Published by Elsevier Science B.V.

Keywords: Bipolar disorder; Hippocampus; Post-mortem; Serotonin; Autoradiography

1. Introduction serotonergic neurons in hippocampus obtained post-mortem from subjects with BD. As we had previous-ly shown a change in the affinity of radioligandEvidence suggests that abnormalities in the tempo-binding to hippocampal SERT in schizophreniaral lobe (Benes et al., 1998; Eastwood and Harrison,(Dean et al., 1995), we measured the affinity and2000; Gruzelier et al., 1988; Soares and Mann, 1997)

3density of [ H]citalopram binding to SERT in BD.and serotonergic systems (Mahmood and Silver-stone, 2001) are important in the pathology ofbipolar disorder (BD). To attempt to unify theselines of evidence, we measured markers on

2. Methods

*Corresponding author. The Mental Health Research Institute The cohorts of BD and control subjects, tissueof Victoria, Locked Bag 11, Parkville, Victoria, 3052, Australia.

collection and processing were as described previ-Tel.: 161-3-9389-2940; fax: 61-3-9387-5061.ously (Dean et al., 2000).E-mail address: b.dean@papyrus.mhri.edu.au (B. Dean).

0165-0327/02/$ – see front matter 2002 Published by Elsevier Science B.V.doi:10.1016/S0165-0327(02)00021-6

66 B. Dean et al. / Journal of Affective Disorders 75 (2003) 65–69

2.1. Measurement of radioligand binding to human two-way ANOVA was used to determine if theretemporal lobe were significant differences in radioligand binding

between diagnostic cohorts or regions of the hip-The density of radioligand binding was analysed pocampus. A post-test Bonferonni Multiple Com-

from calibrated images generated on phosphoimag- parison was completed to identify differences ining plates. Regional localisation of binding was specific hippocampal regions between the bipolardetermined with reference to tissue sections stained and control subjects.with Cresyl violet (Disbrey and Rack, 1970).

Frozen hippocampal sections (53 20 mm at220 8C) were cut from each subject for each of the

3. Resultsthree receptor binding assays and for each con-

3centration of [ H]citalopram and thaw-mounted onComparisons of demographic and pharmacological

gelatin coated slides. In all cases, the index subjectdetails on the tissue donors for this study have been

and its control were processed together with thepublished previously (Dean et al., 2000).3operator blind to diagnosis. [ H]citalopram (New 3Two way ANOVA of the K for [ H]citalopramdEngland Nuclear) binding to the hippocampalbinding showed a significant variance in binding data

serotonin transporter (SERT) was measured essen-between subjects with BD and controls (F 5 40.7,

tially as described previously (Dean et al., 2000).d.f.5 1, 98, P , 0.001; Table 1). There were no

However, in this study, the difference between 3significant differences in theK for [ H]citalopram3 d[ H]citalopram binding at five concentrations (1–10binding between regions of the hippocampus (F 5

nM) in the absence (total) and presence (NSB) of 101.29, d.f.5 6, 98, P 5 0.27) nor was there a signifi-

mM paroxetine was measured.cant interaction between the two factors (F 5 0.46,3The [ H]ketanserin (New England Nuclear) bind-d.f.5 6, 98,P 5 0.84). A Bonferonni Multiple Com-

ing to the serotonin 2A receptor (5HT R) and2A parison test showed a significance increase in theK3 d[ H]8-hydroxy-DPAT (Amersham) binding to thefor radioligand binding inS from subjects with BDlac5HT R was measured as described previously 31A (Table 1; Fig. 1). The density for [ H]citalopram3(Dean et al., 2000). The [ H]sumatriptan (Amer-binding was not different between BD and control

sham) binding to the 5HT R and 5HT R was1D 1F subjects (F 50.3, d.f.5 1, 98, P 5 0.23; Table 1).measured using a method modified from a previous

Two way ANOVA showed there was no signifi-study (Pascual et al., 1996). Sections were washed in 3cant variation in binding of [ H]ketanserin (F 5 0.64,assay buffer (170 mM Tris containing 4 mM CaCl , 32 d.f.5 1, 126, P 50.42), [ H]8-hydroxy-DPAT (F 50.01% ascorbic acid and 10mM pargyline) for 30

0.39, d.f.5 1, 98, P 50.53), methiothepin displace-min at room temperature. The sections were then 3able [ H]sumatriptan (F 5 0.61, d.f.51, 70, P 53exposed to [ H]sumatriptan (4.5 nM) in the absence

0.43) or non-methiothepin displaceableor presence of methiothepin mesylate (1mM) or 3[ H]sumatriptan (F 5 0.88, d.f.5 1, 70, P 5 0.35;serotonin (1mM) for 60 min at room temperature.

Table 1) binding between BD and controls.3For [ H]citalopram, binding data were processedThere were no significant correlations between any

in the Radlig computer program, a Scatchard plotbinding parameter and any confounding factor.

was constructed and the affinity (K ) of radioligandd

binding calculated along with a Hill’s coefficient.3 3 3For [ H]citalopram (10 nM), [ H]ketanserin, [ H]8-

3hydroxy-DPAT and [ H]sumatriptan the density of 4. Discussionbinding sites was calculated as a single point satura-

3tion analysis study. In this study, [ H]citalopram binding to hippocam-pal SERT was essentially as reported in other studies

2.2. Statistical analysis (Cortes et al., 1988; Duncan et al., 1992; Gross-Isseroff et al., 1989; Tejani-Butt et al., 1995). As in

All statistical analyses were performed as de- the frontal cortex (Dean et al., 2000), there were noscribed previously (Dean et al., 2000) except that differences in the density of SERT in the hippocam-

B. Dean et al. / Journal of Affective Disorders 75 (2003) 65–69 67

Table 1Binding parameters (mean6S.E.M.) for a variety of radioligands to different regions of the hippocampus from subjects with bipolar disorder(BD) and control subjects

Radioligand CA1 CA2 CA3 Dent S Sub PHGlac

3[ H]Citalopram: K Control 3.460.6 3.960.8 3.660.8 2.960.6 1.960.3 3.060.3 2.460.4d

BD 4.060.5 4.660.7 4.860.8 4.460.8 4.360.8 4.160.7 3.460.5B Control 1961.1 2362.6 2863.9 2662.8 3764.6 2362.5 3664.3max

BD 1862.6 2663.3 2863.9 2763.9 5267.4 2062.8 2964.2

CA1A CA1B CA2A CA2B CA2C CA3 Dent Sub A Sub B3[ H]Ketanserin Control 5564.9 1262.2 2562.4 5065.3 1161.4 5466.4 5166.7 1161.7 2464.1

BD 4064.7 1162.7 2362.5 4867.1 1562.6 44611 51611 1362.7 2664.1

CA1 CA2 CA3 Dent Sub A Sub B Sub C3[ H]8-hydroxy-DPAT Control 8966.4 3466.9 2264.7 3764.3 6.560.8 4264.2 1161.2

BD 8167.6 4467.7 1362.2 3364.3 5.461.0 4564.3 1061.4

CA1 S Sub P PHGlac sub3[ H]sumatriptan Control 2.760.3 1060.9 8.560.8 1161.3 9.861.1

BD 3.460.5 1261.4 9.661.1 1061.1 1061.33[ H]sumatriptan–non-

methiothepin sensitive Control 3.960.2 4.460.2 4.560.2 3063.9 1261.3BD 2.760.4 3.860.5 3.860.4 2664.4 1261.8

CA, cornu ammonis; Dent, gyrus dentatus;S , stratum lacunosum-moleculare; Sub, subiculum;P , pre-subiculum; PHG, parahip-lac sub

pocampal gyrus.

raphy, ranged from 1.9 to 3.9 nM in the controlsubjects; which is similar to that reported in par-ticulate membrane from the human caudate (Plengeand Mellerup, 1991).

There was a decrease in the affinity of3[ H]citalopram binding to SERT inS from BDlac

subjects. We have proposed that changes in theaffinity of radioligand binding to hippocampal SERTis associated with a hyperserotonergic state in schiz-ophrenia (Scarr et al., 2001). This hypothesis isbased on the finding that decreases in the affinity ofradioligand binding to SERT in human platelets andin synaptosomes from rats treated with 5,7-dihydrox-ytryptamine is associated with a marked decrease inthe capacity of SERT to take up serotonin. Wehypothesised that the change in SERT we observed

3Fig. 1. The affinity [ H]citalopram binding (mean6S.E.M.) to in schizophrenia would therefore be associated withvarious regions of the hippocampus from BD and control subjects. decreased serotonin uptake and a relatively hyper-CA, cornu ammonis; Dent, gyrus dentatus;S , stratumlac serotonergic state in the hippocampus. Our data nowlacunosum-moleculare; Sub, subiculum, PHG, parahippocampal

suggest that there is a similar change in SERT in angyrus.area of the hippocampus from subjects with BD whowere psychotic at death and treated with anti-psy-chotic drugs. Importantly, we have shown that

pus from subjects with BD. The K for treating rats with anti-psychotic drugs does not alterd3[ H]citalopram binding reported in this, the first the affinity of radioligand binding to hippocampal

SERT (Naylor et al., 1996). Thus, it appears that astudy to measure this parameter using autoradiog-

68 B. Dean et al. / Journal of Affective Disorders 75 (2003) 65–69

conformational change in hippocampal SERT is Referencesmore likely associated with psychoses rather than aneffect of anti-psychotic drug treatment. Benes, F.M., Kwok, E.W., Vincent, S.L., Todtenkopf, M.S., 1998.

3 A reduction of non-pyramidal cells in sector CA2 of schizo-The [ H]ketanserin binding to the 5HT R and2A3 phrenics and manic depressives. Biol. Psychiatry 44, 88–97.[ H]8-hydroxy-DPAT binding to the 5HT R were1A Biegon, A., Kargman, S., Snyder, L., McEwen, B.S., 1986.

as described elsewhere (Biegon et al., 1986; Gross- Characterization and localization of serotonin receptors inIsseroff et al., 1990; Hall et al., 1997; Hoyer et al., human brain post-mortem. Brain Res. 363, 91–98.1986; Kohler et al., 1986). Thus, our data suggest Cortes, R., Soriano, E., Pazos, A., Probst, A., Palacios, J.M., 1988.

Autoradiography of anti-depressant binding sites in the humanthat, as in the frontal cortex (Dean et al., 2000), there3 3brain: localization using [ H]imipramine and [ H]paroxetine.is no change in the levels of these receptors in the

Neuroscience 27, 473–496.hippocampus in BD. This differs from a study using Dean, B., Opeskin, K., Pavey, G., Naylor, L., Hill, C., Keks, N.,

3PET that reported a decrease in the density of Copolov, D.L., 1995. The [ H]paroxetine binding is altered inhippocampal 5HT R in depressed BD subjects with the hippocampus but not the frontal cortex or caudate nucleus1A

from subjects with schizophrenia. J. Neurochem. 64, 1197–a family history of the disorder (Drevets et al.,1202.1999). Notably, subjects in our study were psychotic

Dean, B., Pavey, G., McLeod, M., Opeskin, K., Keks, N.,at death and thus the differing findings could support Copolov, D.L., 2000. A change in the density of

3 3the hypothesis that the level of 5HT Rs may change [ H]flumazenil, but not [ H]muscimol binding, in Brodmann’s1A

with changing phases of the illness. area 9 from subjects with bipolar disorder. J. Affect. Disord.3 66, 147–158.The [ H]sumatriptan binds to both the 5HT and1D

Disbrey, B.D., Rack, J.H., 1970. Histological Laboratory Methods.5HT R in human post-mortem brain tissue (Pascual1F E and S Livingstone, Edinburgh.et al., 1996), methiothepin binds to 5HT and1D Drevets, W.C., Frank, E., Price, J.C., Kupfer, D.J., Holt, D., Greer,5HT receptors (Pauwels, 1997) whilst serotonin P.J., Huang, Y., Gautier, C., Mathis, C., 1999. PET imaging of1B

binds to all 5HTR. Thus, methiothepin sensitive serotonin 1A receptor binding in depression. Biol. Psychiatry3 46, 1375–1387.[ H]sumatriptan binding provides a measure of

Duncan, G.E., Little, K.Y., Kirkman, J.A., Kaldas, R.S., Stumpf,5HT R whilst the methiothepin insensitive binding1D W.E., Breese, G.R., 1992. Autoradiographic characterization ofwould be predominately to the 5HT R. Hence, there 3 31F [ H]imipramine and [ H]citalopram binding in rat and humanwere no significant differences in the density of brain: species differences and relationships to serotonin in-either hippocampal 5HT R or 5HT R in BD. nervation patterns. Brain Res. 591, 181–197.1D 1F

Eastwood, S.L., Harrison, P.J., 2000. Hippocampal synapticIn conclusion, other than in SERT, our studiespathology in schizophrenia, bipolar disorder and major depres-have not revealed changes in hippocampal serotoner-sion: a study of complexin mRNAs. Mol. Psychiatry 5, 425–

gic markers in BD. However, data from this study 432.and a study using PET raise the possibility that the Gross-Isseroff, R., Israeli, M., Biegon, A., 1989. Autoradiographiclevels of 5HT R may fluctuate in different phases analysis of tritiated imipramine binding in the human brain1A

post-mortem: effects of suicide. Arch. Gen. Psychiatry 46,of BD. This could have important ramifications in237–241.understanding the neurobiology of the mood swings

Gross-Isseroff, R., Salama, D., Israeli, M., Biegon, A., 1990.in BD. Autoradiographic analysis of age-dependent changes in

serotonin 5-HT2 receptors of the human brain post-mortem.Brain Res. 519, 223–227.

Gruzelier, J., Seymour, K., Wilson, L., Jolley, A., Hirsch, S., 1988.Impairments on neuropsychologic tests of temporohippocampalAcknowledgementsand frontohippocampal functions and word fluency in remittingschizophrenia and affective disorders. Arch. Gen. Psychiatry

We acknowledge the assistance of Robyn Brad- 45, 623–629.bury, Dr Kenneth Opeskin, Professor Nicholas Keks Hall, H., Lundkvist, C., Halldin, C., Farde, L., Pike, V.W.,

McCarron, J.A., Fletcher, A., Cliffe, I.A., Barf, T., Wikstrom,and Ms Christine Hill. This work was supported inH., Sedvall, G., 1997. Autoradiographic localization of 5-part by The Carroll Family Trust, the NationalHT1A receptors in the post-mortem human brain using

Alliance for Research into Schizophrenia and De- 3 11[ H]WAY-100635 and [ C]way-100635. Brain Res. 745, 96–pression, National Health and Medical Research 108.Council Project Grant 000341. Hoyer, D., Pazos, A., Probst, A., Palacios, J.M., 1986. Serotonin

B. Dean et al. / Journal of Affective Disorders 75 (2003) 65–69 69

receptors in the human brain. I. Characterization and au- Pazos, A., 1996. Autoradiographic distribution of3toradiographic localization of 5-HT1A recognition sites. Ap- [ H]sumatriptan-binding sites in post-mortem human brain.

parent absence of 5-HT1B recognition sites. Brain Res. 376, Cephalalgia 16, 317–322.85–96. Pauwels, P.J., 1997. 5-HT 1B/D receptor antagonists. Gen.

Kohler, C., Radesater, A.C., Lang, W., Chan-Palay, V., 1986. Pharmacol. 29, 293–303.3Distribution of serotonin-1A receptors in the monkey and the Plenge, P., Mellerup, E.T., 1991. The [ H]citalopram binding to

post-mortem human hippocampal region. A quantitative au- brain and platelet membranes of human and rat. J. Neurochem.3toradiographic study using the selective agonist [ H]8-OH- 56, 248–252.

DPAT. Neurosci. Lett. 72, 43–48. Scarr, E., Copolov, D.L., Dean, B., 2001. A proposed pathologicalMahmood, T., Silverstone, T., 2001. Serotonin and bipolar model in the hippocampus of subjects with schizophrenia. Clin.

disorder. J. Affect. Disord. 66, 1–11. Exp. Pharmacol. Physiol. 28, 70–73.Naylor, L., Dean, B., Opeskin, K., Hill, C., Keks, N.A., Copolov, Soares, J.C., Mann, J.J., 1997. The anatomy of mood disorders—

D.L., 1996. Possible conformational changes in the serotonin Review of structural neuroimaging studies. Biol. Psychiatry 41,transporter in the hippocampus of subjects with schizophrenia 86–106.

3identified using [ H]paroxertine. J. Neural Transm. Gen. Sect. Tejani-Butt, S.M., Yang, J., Pawlyk, A.C., 1995. Altered serotonin103, 749–757. transporter sites in Alzheimer’s disease raphe and hippocam-

Pascual, J., del Arco, C., Romon, T., del Olmo, E., Castro, E., pus. Neuroreport 6, 1207–1210.