The distribution of 5-HT1A and 5-HT2A receptor mRNA in human brain

BRAIN RESEARCH ELSEVIER

Brain Research 676 (1995) 157-168

Research report

The distribution of 5-HT1A and 5-HTzA receptor mRNA in human brain P.W.J. Burnet *, S.L. Eastwood, K. Lacey, P.J. Harrison University Department of Clinical Neurology (Neuropathology), Gibson Building, Radcliffe Infirmary NHS Trust, Woodstock Road and University Department of Psychiatry, Warneford Hospital, Oxford, OX2 6HE, UK Accepted 27 December 1994

Abstract

We have examined the distribution of 5-HT1A and 5-HT2A receptor mRNAs in post-mortem human hippocampus, neocortex, raphe nuclei, cerebellum and basal ganglia using in situ hybridization histochemistry. Receptor transcripts in brains from two males and two females (mean age + S.D. = 70 _+4 years; post-mortem interval = 29 _ 6 h) were visualised with 35S-radiolabelled synthetic oligodeoxyribonucleiic acid probes. In the hippocampus, 5-HT1A receptor mRNA was present in all fields, especially CA1. In the parahippocampal gyrus and neocortical regions 5-HT1A receptor mRNA was enhanced in superficial and middle laminae. 5-HT1A receptor ml;tNA was particularly abundant in the raphe and other serotonergic cell groups of the brainstem. The analysis of emulsion dipped sections showed 5-HT1A receptor mRNA to be concentrated in pyramidal neurons, together with the granule cells of the dentate gyms. In neocortical areas lamina III pyramidal neurons were more heavily labelled than those in lamina V. There was no evidence of glial expression of 5-HT1A receptor mRNA in grey matter or white matter compartments. 5-HTEA receptor mRNA was present in all neocortical areas examined, where it was located in pyramidal neurons, of lamina V more than in those of lamina III, as well as in putative interneurons, especially within lamina IVc of the striate cortex. 5-HT2A receptor mRNA was observed at minimal levels in the hippocampus and not in the raphe. Neither 5-HT1A nor 5-HT2A receptor mRNA were detected in the cerebellum, substantia nigra or striatum. The ability to detect these transcripts at the regional and cellular level will help reveal important details of the 5-HT receptor system in the human brain. This includes the investigation of their putative roles in the normal chemoarchitecture and in pathophysiological brain processes.

Keywords: Serotonin receptor; Human brain; Hybridization histochemistry, in situ; Serotonin; Messenger RNA

1. Introduction

The existence of a heterogenous population of brain 5-HT (5-hydroxytryptamine., serotonin) receptors and encoding genes [7] has illustrated the complexity of the 5-HT system and its potential roles in the pathophysiology and treatment of neuropsychiatric diseases. The most ubiquitous and extensively studied 5-HT receptors in the human brain are the 5-HTIA receptor (5-HTIAR) and 5-HT2A rec,eptor (5-HT2AR). Evidence from post-mortem studies implicates these receptors in Alzheimer's disease [12] depression [11,22] schizophrenia [23,25] and possibly suicide [11,22,23,27]. Additionally, the receptors may halve a role in the action of drugs which alleviate some; of the symptoms of these disorders, since marked alterations in the densities of

* Corresponding author. Fax: (4.4) (1865) 224 508. 0006-8993/95/$09.50 © 1995 Elsevier Science B.V. All rights reserved SSDI 0 0 0 6 - 8 9 9 3 ( 9 5 ) 0 0 1 0 4 - 2

5-HT1AR and 5-HT2AR binding sites occur in various regions of the rat brain after the administration of antidepressant and neuroleptic drugs (for review see [43]). The regulation of 5-HT1AR and 5-HT2AR binding site densities and m R N A s can be studied by complementing standard techniques such as quantitative autoradiography with molecular techniques such as in situ hybridization histochemistry ( I S H H ) [8]. I S H H is a semi-quantitative tool which allows the localization of the m R N A to individual cell populations and, if required, to individual neurons. The study of 5-HT receptor m R N A s in the human brain using I S H H could therefore assist in the identification of the receptor subtypes affected in various neuropsychiatric diseases. In addition, specific hypotheses have been developed which are predicated upon the cellular localization of these receptors. For example, it has been suggested that cortical pyramidal neuron loss or dysfunction may

158

P.w.J. Burnet et al. /Brain Research 676 (1995) 157-168

underlie the cognitive impairment of Alzheimer's disease [16]. Since 5-HTIAR are thought to be localized to these pyramidal neurons [15,16], with their activation producing hyperpolarization of the cell, selective antagonism of these receptors has been proposed as a therapeutic tool for the alle.viation of cognitive deficits in Alzheimer's disease [16]. The predicted localization of 5-HTIA R and 5-HT2AR in the human brain are based in part upon data obtained from rat brain using immunocytochemistry (ICC) [17,19,20,30,31], electro-

physiology [1,2,18] and I S H H [8,10,35,36] as well as from receptor autoradiography [8,10]. However, to date the cellular localization of 5-HT receptors has not been unequivocally determined in human brain. We have reported the presence of 5-HT1A R and 5-HT2AR m R N A s in the human hippocampus using a quantitative reverse transcription-polymerase chain reaction ( R T - P C R ) technique [9]. The aim of the present study was to begin to map in detail the regional and cellular distribution of 5-HT1AR and 5-HT2AR m R N A s

Fig. 1. The distribution of 5-HT1ARmRNA in rat and human hippocampus. Coronal sections of rat brain (A) and human hipppocampal sections (B) were hybridized usingasS-labelled oligonucleotide probes complementary to sequences of rat and human 5-HT1ARmRNA. Hybridization of rat brain coronal (C) and human hippocampal (D) sections with oligonucleotide sense probes produced minimal signals. DG, dentate gyrus; CA1, CA1 subfield of the stratum pyramidale; PHG, parahippocampal gyrus.

P.W.J. Burnet et a l . / Brain Research 676 (1995) 157-168

in the human brain, using ISHH. Preliminary communications concerning the distribution of these mRNAs have recently appeared [4,216].

2. Materials and methods

2.1. Tissue collection and processing Brains from two male and two female subjects with no clinical history of neurological or psychiatric disease were used. Neuropathological examination was unre-

159

markable. The mean age and post-mortem interval ( + S.D.) were 70 + 4 years and 29 + 6 h, respectively. Blocks of tissue were taken from the left hemisphere from medial temporal lobe, superior temporal gyrus (Brodmann area 22), striate cortex (Brodmann area 17), orbitofrontal cortex (Brodmann area 11), cerebellar cortex, raphe nuclei, striatum and substantia nigra. The dissected tissue was frozen on a dry-ice/alcohol slurry and stored at - 70 ° C. Sectioning of blocks was carried out at - 2 0 ° C. Eighteen micrometer cryostat cut sections were thaw mounted onto gelatin-subbed slides and pretreated for ISHH. To allow compar-

B

D

"

4V

.. cDR .PnO

"

+

B9

MnR

Fig. 2. The distribution of 5-HTIAR mRNA in human neocortical regions and brainstem. A: orbitofrontal cortex (Brodmann area 11). B: striate cortex (Brodmann area 17). C: superior temporal gyrus (Brodmann area 22). D: brainstem at the level of the raphe. 4V, fourth ventricle; cDR, caudal dorsal raphe; MnR, median raphe; PnO, oral pontine nuclei; B9, supralemniscal region.

160

P.W..J. Burnet et al./ Brain Research 676 (1995) 157-168

B

P.w.J. Burnet et al. /Brain Research 676 (1995) 157-168

isons with the distribution of the mRNAs in rodent, brains from adult male Sprague-Dawley rats (Harlan, Bicester, UK) were processed in parallel with the human tissues. 2.2. In situ hybridization histochemistry

Oligodeoxyribonucleic acid probes complementary to bases 82-123 and bases 866-907 of the human 5-HTIAR gene [24] and to bases 350-379 and bases 471-500 of the human 5-HT2AR gene [39] were synthesized and high-performance liquid chromatographypurified (Oswel DNA Services, Edinburgh, UK). All probes were 100% homologous to equivalent rat gene sequences. No significant homologies to any other transcripts were identified on the European Molecular Biology Laboratory (EMBL) data base. The two probes for each receptor mRNA were mixed at equimolar concentrations and 3'-tail labelled with [aasS]dATP (1500 Ci/mmol), in a 1:10 molar ratio using terminal deoxynucleotidyl transferase and standard labelling buffer. The labelled probe was added to hybridization buffer (4 X SSC, 50% deionised formamide, 10% dextran sulphate, 5 ×Denhardt's solution, 200 ~ g / m l salmon sperm DNA, 100 tzg/ml poly(A), 25 mM sodium phosphate, 1 mlVl sodium pyrophosphate and 100 mM dithiothreitol) at a final concentration of 2 × 10 4 cpm/p.1. A total volume of 100/zl was added to each section. Sections were covered with Nescofilm (BDH, Poole, UK) and placed in a chamber humidified with 4 × SSC and 50% formamide and incubated overnight at 35° C. Post-hybridization washes were carried out in 1 x SSC for 5-HT2AR and 0.5 x SSC for 5-HTIAR (since the melt temperatures of these probes were greater than those of the 5-HT2AR probes) at 55°C for 3 x 20 min followed by 2 x 60 min at room temperature. After a final rinse in water, slides were air-dried and apposed to X-ray film (Hyperfilm betamax, Amersham, UK) at room temperature for 6-8 weeks. Additional sections were dipped in LM1 emulsion (Amersham, UK) and kept at 4° C for 4-8 weeks before being developed and lightly counterstained with cresyl violet. Experimental controls included: (1) hybridization of sections using oligonucleotides in the sense orientation, (2) pretreatment of some sections with RNase A (100/zg/ml in PBS at 37° C for 60 min), and (3) incubation in the presence of 20-fold excess unlabelled probe, in addition to (4) concurrent use of the rat sections to confirm the predicted distribution of each mRNA in the rat forebrain.

161

3. Results 3.1. 5-HTIARR m R N A distribution

Fig. 1 shows representative autoradiograms generated when coronal sections of rat brain at the level of the hippocampus and human hippocampal sections were hybridized with antisense or sense oligonucleotide probes against 5-HT1AR mRNA. In the rat brain, 5-HTIAR mRNA was predominant in the hippocampus where it was characteristically localized to the dentate gyrus, CA1 and CA3 subfields and present at lower levels in CA2 (Fig. 1A). In the human hippocampus 5-HT1AR mRNA was predominant in CA1 with a weaker signal observed in the dentate gyrus, other CA subfields and subiculum (Fig. 1B). Over the parahippocampal gyrus, signal was greater in superficial than deep laminae. Minimal signals were generated using 5-HTIAR mRNA sense probes (Fig. 1C and D) or when antisense probes were hybridized to RNase-pretreated human hippocampal sections (data not shown). These controls, in tandem with the anticipated result of parallel ISHH in rat brain, confirms the specificity of the oligonucleotide probes for detection of 5-HT1AR mRNA in human brain. Fig. 2 demonstrates the distribution of 5-HT1AR mRNA in various neocortical regions and in the raphe nuclei of the human brain. It was present in all neocortical areas studied, but was consistently detectable only in superficial laminae (Fig. 2A-C). A high concentration of 5-HT1AR mRNA was observed in cells scattered throughout the brainstem raphe region. 5-HT~AR mRNA was not reliably detected in the cerebellum, substantia nigra or striatum (data not presented). Only background hybridization signals were observed in the white matter. The emulsion-dipped sections hybridized with 5HTIAR mRNA oligonucleotides (Fig. 3) revealed clusters of silver grains over dentate gyrus granule cells (Fig. 3A), pyramidal neurons of the CA subfields (Fig. 3B,C) and subiculum. In the parahippocampal gyms, grains were concentrated over neurons in laminae I I / I I I (Fig. 3D) and to a lesser extent in laminae V / V I (Fig. 3E). Grain clusters were also observed over~pyramidal neurons in the neocortical regions, especially in lamina III, corresponding to the film images; in addition, some signal was also detected over lamina V pyramidal neurons (data not presented). There was no clear evidence of 5-HT~AR mRNA signal over other cell types in the neocortex such as interneurons or grey

Fig. 3. Cellular distributionof 5-HTIARmRNA in human brain using liquid emulsionautoradiography.A: stratum granulosumof dentate gyrus. B: stratum pyramidale of CA3 subfield. C'. stratum pyramidale of CA1 subfield. D: lamina III of parahippocampal gyrus. E: lamina V of parahippocampal gyms. F: parahippocampal white matter. Note the absence of grain clustering over small cells, putativelyglial (arrows in F). Bar = 50/.~m.

162

P.W.J. Burnet et al. / Brain Research 676 (1995) 157-168

matter glia; however, this possibility can not be wholly excluded (see section 4.1). Neither did there appear to be any specific labelling of clustering of cells in the subcortical white matter (Fig. 3F). 3.2. 5-HT2AR m R N A distribution

Fig. 4 shows representative autoradiograms generated when 5-HT2AR antisensense or sense probes were hybridized to sections of human hippocampus and coronal sections of rat brain. The distribution of 5-

HTzA R m R N A in the rat brain was characteristically localized to the deeper laminae of the cortex, with low signals in the hippocampus (Fig. 4A). 5-HT2AR m R N A was barely detectable in the human hippocampus but strong signal was present over the middle and deep laminae of the parahippocampal gyrus (Fig. 4B). The minimal signal seen after sense strand hybridization (Fig. 4C,D), ribonuclease (not shown), plus the anticipated result in rat brain (Fig. 4A), confirms the specificity and sensitivity of the 5-HT2AR probes. The distribution of 5-HT2AR m R N A in neocortex

Fig. 4. The distribution of 5-HT~R mRNA in rat brain and human hippocampus. Coronal sections of rat brain (A) and human hippocampal sections (B) were hybridized using radiolabelled oligonucleotide probes complementary to sequences of rat and human 5-HTzAR mRNA. Hybridization of similar rat brain coronal (C) and human hippocampal (D) sections with oligonucleotide probes in the sense orientation produced minimal signals. CA1, CA1 subfield of stratum pyramidale.

P.W.J. Burnet et al. /Brain Research 676 (1995) 157-168

and raphe nuclei of the human brain is demonstrated in Fig. 5. In area 11 5-.HT2AR mRNA signal was concentrated in two band,;, probably corresponding to lamina III and V (Fig. 5A). In the striate cortex (area 17) 5-HTEAR mRNA was enriched in the in lamina IVc with moderate levels elsewhere (Fig. 5B). Area 22 presented a similar picture to area 11 (Fig. 5C). 5HTzAR mRNA was not reliably detected in the raphe (Fig. 5D), cerebellum, substantia nigra or striatum (data not presented). The pattern of 5-HTzAR mRNA distribution at the cellular level was more complex than that of 5-HTIAR mRNA (Fig. 6). The differential laminar distribution

163

evident in each neocortical region on film (Fig. 4A-C) was confirmed in the emulsion-dipped sections. The majority of pyramidal neurons in lamina III (Fig. 6B) and lamina V (Fig. 6C) were clearly labelled. Many non-pyramidal cells, presumably interneurons, were also labelled to a moderate degree (Fig. 6A). In the case of area 17, grains were most abundant over putative interneurons in lamina IVc with progressively lighter labelling towards the pial surface and white matter (data not shown). As with 5-HTIA R mRNA, we found no evidence for 5-HT2AR mRNA-positive glial cells. In the rat brain 5-HT2AR mRNA was localized to

A

¢

D

i

Fig. 5. The distribution of 5-HTzAR mRNA in human neocortical regions and brainstem. A: orbitofrontal cortex (Brodmann area 11). B: striate cortex (Brodmann area 17). C: superior temporal gyrus (Brodmann area 22). D: brainstem at the level of the raphe. Note the absence of signal in the region of the raphe nuclei; this section is adjacent to that illustrated in Fig. 2D.

164

P.W.J. Burnet et al. / Brain Research 676 (1995) 117-168

Fig. 6. Cellular distribution of 5-HT2AR mRNA in human and rat brain. A: lamina II of the superior temporal gyrus (area 22). B: lamina III of area 22. C: lamina V of area 22. D: subcortical white matter. E: lamina V of the rat parietal cortex. F: rat piriform cortex, lamina II. Bar = 50 /xm. Pyramidal neurons are delineated by arrows (C,D); a non-pyramidal cell is indicated by an arrow head (A). Note that some cells of both morphologies are unlabelled (see text).

n e o c o r t i c a l p y r a m i d a l n e u r o n s ( F i g . 6 E ) a n d to int e r n e u r o n s in t h e p i r i f o r m c o r t e x (Figi 6F). H o w e v e r , as in t h e h u m a n b r a i n , a c o n s i d e r a b l e n u m b e r o f cells

o f b o t h m o r p h o l o g i e s w e r e n o t l a b e l l e d w i t h g r a i n s in excess o f b a c k g r o u n d ( e g Fig. 6E). The relative abundance of 5-HT1AR and 5-HT2AR

P. W.J. Burnet et al. / Brain Research 676 (1995) 157-168 Table 1 T h e relative distribution of 5-HT1A R and 5-HTEA R m R N A in hum a n brain as detected by film autoradiography Brain region

Relative m R N A abundance 5-HT1AR

5-HT2AR

Dentate gyrus CA3 CA2 CA1 Subiculum Parahippocampal gyrus BA ll(SL) BA 11(DL) BA 17(SL) BA 17(DL) BA 22(SL) BA 22(DL) Raphe Cerebellum Striatum

+ + + ++ ++ ++ ++ +/ ++ +/ ++ +/ ++ 0

+/ +/+/ +/ +/ ++ + ++ + ++ + ++ 0 0

0

0

Substantia nigra

0

0

Relative abundance is rated as: 0 = absent; + / - = b a r e l y detectable; + = l o w / m o d e r a t e ; + + = strong. SL = superficial layers and D L = deep layers of the neocortex.

Table 2 T h e relative cellular distribution of 5 - H T m R and 5-HT2AR m R N A in h u m a n brain as demonstrated on emulsion-dipped sections

Cell type Granule cells (dentate gyrus) Pyramidal neurons (hippocampus) Pyramidal neurons (cortical) Interneurons (cortical) Grey matter glia a (cortical) White matter glia a (cortical)

Relative m R N A abundance 5-HT1AR

5-HTEA R

+ + + + + 0 0 0

0 +/ + + + 0 0

a W e did not attempt to distinguish glial subtypes. Relative abundance is rated as: 0 = a b s e n t ; + / - = b a r e l y detectable; + = l o w / m o d e r a t e ; + + = strong.

mRNAs in the various regions and cell types of the human brain are summarized in Table 1 and Table 2.

4. D i s c u s s i o n

Our results confirm that 5-HTIAR and 5 - H T 2 A R m R N A can be detected in post-mortem human brain [4,9,26,29], and provide the first detailed data as to their regional and cellular distribution.

4.1. 5-HT1AR mRNA distribution The distribution of 5-HTIA R m R N A in human brain was found to be broadly similar to that in the rat (Figs. 1, 2). However, some differences were observed. In the human hippocampus, CA1 contains a higher level of

165

the receptor transcript compared to the dentate gyrus (Fig. 1B). In the rat hippocampus the converse is true (Fig. 1A) [8,10,35]. In the human neocortex 5-HTIA R m R N A is enriched in the superficial and middle laminae (Fig. 2) whilst in the rat neocortex the transcript is more abundant in deep laminae [8,10]. The high abundance of 5-HTIA R m R N A in the CA1 subfield of the human hippocampus and superficial laminae of the neocortex parallels the distribution of 5-HT1AR binding site densities [14,33] and is in agreement with a recent Northern blot study [29]. The localization of binding sites preferentially to lamina II probably reflects their distribution on the apical dendrites of lamina III pyramidal neurons. The earlier studies, together with the current data, have demonstrated 5-HTtAR binding sites and m R N A at similar levels in CA2 as in other subfields of the human Ammon's horn whereas in the rat significantly lower levels are observed in CA2 compared to other CA subfields. We have previously demonstrated a significant correlation between the levels of 5-HT1AR m R N A and 5-HT1A R binding sites in the human hippocampus [9]. The present study complements our previous data by demonstrating that not only abundance but neuroanatomical localization of 5HTIAR m R N A in this instance correlates with that of receptor binding sites. At the cellular level, the presence of 5-HT1AR m R N A in hippocampal pyramidal neurons and granule cells of the dentate gyrus (Fig. 3) is in agreement with its cellular distribution in rat brain [35]. Contrary to previous receptor autoradiographic studies [14,33], however, we were unable consistently to demonstrate the presence of 5-HT1A R m R N A in the deeper laminae of the human neocortical regions from images generated on film. This discrepancy may in part result from binding sites in the deep laminae residing on the processes of neurons with their soma elsewhere. More likely, it reflects a lower level of m R N A expression which is close to our current film detection threshold, since some brains do show signal overlying lamina V on film (data not shown) as well as in the dipped sections (Fig. 3E). Previous studies have demonstrated 5-HT1AR m R N A in pyramidal neurons of the rat neocortex [35]; the cellular localization of 5-HTIAR m R N A in the human neocortex (Fig. 2) is therefore as predicted from rodent studies despite the differential laminar distribution between the two species. Its preferential expression by pyramidal neurons, especially in lamina III, provides support for theories involving 5-HT1AR upon these neurons in Alzheimer's disease [15]. The abundant expression of 5-HTIA R m R N A in neurons within the raphe and adjacent neuronal groups, putatively 5-HT containing ceils [5], parallels the distribution of 5-HTIA R binding sites in this region [40] and supports the importance of this receptor in presynaptic 5-HT neurons. Its distribution in the human brainstem

166

P.w.J. Burnet et aL / Brain Research 676 (1995) 157-168

is also similar to that of the 5-HT transporter mRNA recently reported [3]. The presence of 5-HTIA R in astroglia has been suggested both in culture and in vivo [21,41,42], although other studies have not confirmed this [19,20]. Here, inspection of emulsion-dipped human and rat sections failed to reveal specific grain clustering over putative glia within white or grey matter. Nor did the density or distribution of grains over the neuropil suggest 5-HTIAR mRNA in uncounterstained glial processes There are two possibilities which could explain these negative findings. Firstly, the abundance of 5HT1AR mRNA in glial cells may be below the detection limit of the current ISHH technique in human brain. Secondly, 5-HT~AR gene expression in glia might only occur at particular developmental stages or following certain stimuli such as brain injury. For example, the presence of receptor protein observed using specific antibodies [42] may represent 'inactive' 5-HTIARS in cytoplasmic stores which become depleted once these receptors are expressed on the cell membrane and become 'active'. The cytoplasmic stores would then be replenished, a process which would likely involve increased 5-HT1AR mRNA transcription. This, of course, is speculation and futher studies are required to establish the presence of 5-HT1AR in glial cells in human brain. Finally, the lower levels of 5-HTIAR mRNA observed in human than in rat brain (compare Fig. 1A with 1B) contrasts with the higher levels of 5-HTIA R binding sites detected autoradiographically in human compared to rat neocortex especially in superficial laminae [33]. This is unlikely to be due to sensitivity of the mRNA to post-mortem delay since, as with other mRNAs [6], its abundance is not related to post-mortem intervals of up to 42 h or more [9]. The species difference in abundance of 5-HTIAR mRNA relative to its receptor protein which is therefore implied by these data may reflect differences in the efficiency of translation or in the turnover of the encoded receptor in the human compared to the rodent brain. 4.2. 5-HT2A R m R N A distribution

The identity of the neuronal types expressing the 5-HT2AR in vivo has previously been unclear. ICC studies have suggested that they are expressed by interneurons [17], especially GABAergic interneurons [31], which is in keeping with electrophysiological data in the piriform cortex of the rat [18]. This view is supported by the preservation of 5-HT2AR binding site densities in the rat neocortex after intrastriatal injection of volkensin which destroys lamina V pyramidal neurons [15]. On the other hand, different electrophysiological studies [1,2] and ISHH studies [17,36,37] have inferred that pyramidal neurons contribute to the expression of the 5-HT2AR. The emulsion-dipped sec-

tions (Fig. 6B,C,E) show clearly that cells with pyramidal and non-pyramidal morphologies express the receptor in both species. The size, shape, staining characteristics and laminar distribution of the non-pyramidal cells, exemplified in lamina IVc of human striate cortex and in rat piriform cortex, (Fig. 6F), imply that these cells are interneurons. Conversely, it is unlikely that these latter cells are grey matter glia, though this remains a possibility in the light of some gene expression data [13,21]. Overall therefore, our data strongly suggest that both pyramidal neurons and interneurons express 5-HT2AR mRNA. Our ISHH data (see also [17,37]), which appear to show 5-HT2AR mRNA at approximately the same levels and in the same proportion of pyramidal neurons as in non-pyramidal neurons contrasts with the ICC data indicating that the receptor is rarely observed in pyramidal neurons [30]. Several reasons might account for the detection in pyramidal neurons of 5-HT2AR mRNA but not of the encoded receptor protein. Firstly, ISHH in this instance may be more sensitive than ICC, perhaps due to relative levels and accessibility of the mRNA compared to the protein epitope. Secondly, it is possible that the mRNA in pyramidal neurons is not translated, or gives rise to low steady-state numbers of receptors. Thirdly, the ISHH signal might be a false positive. This seems unlikely, given our use of a range of experimental controls, and since the artefact would need to affect pyramidal neurons selectively. If we accept the presence of 5-HT2ARS in pyramidal neurons as well as in interneurons, then constraints are put upon the nature and site of inhibitory actions proposed in recent models to be mediated by the receptor [18]. Additional complexity is provided by the fact that some neurons of both types were negative for 5-HT2AR mRNA (Fig. 6). While this may be a methodological problem, it may also indicate the presence of subpopulations of pyramidal and non-pyramidal neurons with regard to their 5-HTaAR mRNA expression status. As with the discrepancy between the ISHH and immunocytochemical data, this possibility can only be addressed by more extensive studies and double-labelling techniques. Like the 5-HTIAR, the regional and laminar distribution of 5-HT2AR expression in the human brain shares broad similarities with that in the rat brain. In both species, hippocampal expression of the receptor is very low, (Fig. 4A,B) whereas in parahippocampal and other cortical areas it is relatively abundant, especially in the deeper laminae (Figs. 4B and 5A-C) [8,17]. The laminar distribution of 5-HT2AR mRNA in the cortex closely parallels that of 5-HT2AR binding sites in each area, as well as the relative abundance between areas [34]. The latter observation complements our previous finding that 5-HT2AR mRNA abundance and 5-HT2AR binding site densities are correlated in the medial

P.W.J. Burnet et al./ Brain Research 676 (1995) 157-168

temporal lobe [9]. Area 17 had a higher overall level of signal than the other cortical regions, especially lamina IVc (compare Fig. 5B with 5A, 5C). However, striate cortex has a higher neuronal density than other neocortical areas, and lamina IVc has the highest neuronal density of all [32,38]. Thus, the strength of signal seen on film autoradiograms of 5-HT2AR mRNA is likely to result largely from cell packing density rather than from greater expression per cell. This question could be adressed directly by 'per cell' grain counting. Such an analysis was beyond the scope of this study. We did not find any labelling in the raphe after hybridization with the 5-HT2AR mRNA probes. In contrast a recent abstract reports the presence of scattered 5-HT2AR mRNA-positive neurons in this region [4]. We are currently conducting a more thorough examination of the human brainstem to address this discrepancy.

4.3. Summary Data concerning the distribution of 5-HT1AR and 5-HT2AR mRNAs in the human brain have been presented. Such data provide support and constraints for 5-HTIAR and 5-HT2AR involvement in models of brain function and dysfunction whose validity depends upon their precise cellular localization. In this respect, our study has made three important findings. Firstly, both receptors are expressed by pyramidal neurons of the hippocampus and neocortex, although the relative abundance of the mRNAs differs between pyramidal neuron populations according to lamina or region. Secondly, 5-HT2AR mRNA is present additionally in nonpyramidal neurons. Its presence in neurons of both major classes would appear to reconcile electrophysiological and immunocytochemical data which have been inconclusive regarding which cell types express this receptor. Thirdly, we found no clear evidence for either mRNA in non-neuronal cells, although negative conclusions of this kind are always problematic. More extensive and quantitative studies in human brain will be needed to clarify these issues. In this regard, ISHH together with other methods - - such as ICC and RT-PCR- will together enable the distribution and abundance of 5-HTIAR and 5-HT2AR gene expression to be determined and allow their putative alterations in various neuropsychiatric disorders to be revealed.

Acknowledgements We thank Dr. Brendan McDonald for neuropathological data. P.W.J.B. is a Medical Research Council (MRC) Training Fellow. P.J.H. is a Wellcome Senior Research Fellow in Clinical Science.

167

References [1] Araneda, R. and Andrade, R., 5-Hydroxytryptamine 2 and 5-hydroxytryptaminelA receptors mediate opposing responses on membrane excitability in the rat association cortex, Neuroscience, 40 (1991) 399-412. [2] Ashby Jr. C.R., Edwards, E. and Wang, R.Y., Electrophysiological evidence for a functional interaction between 5-HT1A and 5-HT2A receptors in the rat medial prefrontal cortex: an iontophoretic study, Synapse, 17 (1994) 173-181. [3] Austin, M.C., Bradley, C.C., Mann, J.J. and Blakely, R.D., Expression of serotonin transporter messenger RNA in the human brain, J. Neurochem., 62 (1994) 2362-2367. [4] Austin, M.C., Wekel, J,A. and Mann, J.J., Localization of 5HTzA receptor mRNA in neurons of the human brain stem, Soc. Neurosci Abstr., 20 (1994) 1546. [5] Baker, K.G., Halliday, G.M., Halasz, P., Hornung, J.-P., Geffen, L.B., Cotton, R.G.H. and Tork, I., Cytoarchitecture of serotonin-synthesizing neurons in the pontine tegmentum of the human brain, Synapse, 7 (1991) 301-320. [6] Barton, A.J.L., Pearson, R.C.A., Najlerahim, A. and Harrison, P.J., Pre- and post-mortem influences on brain RNA, J. Neurochem., 61 (1993) 1-11. [7] Boess, F.G. and Martin, I.L. Molecular biology of 5-HT receptors, Neuropharmacology, 33 (1994) 275-317. [8] Burnet, P.W.J., Michelson, D., Smith, M.A., Gold, P.W. and Sternberg, E,M., The effect of chronic imipramine administration on the densities of 5-HTtA and 5-HT 2 receptors and the abundance of 5-HT receptor and transporter mRNA in the cortex, hippocampus and dorsal raphe of three strains of rat, Brain Res., 638 (1994) 311-324. [9] Burnet, P.W.J., Eastwood, S.L. and Harrison, P.J., Detection and quantitation of 5-HT1A and 5-HT2A receptor mRNAs in human hippocampus using a reverse transcriptase-polymerase chain reaction (RT-PCR) technique and their correlation with binding site densities and age, Neurosci. Lett., 178 (1994) 85-89. [10] Chalmers, D.T. and Watson, S.J., Comparative anatomical distribution of 5-HT1A receptor mRNA and 5-HTIA binding in rat brain- a comparative in situ hybridization/in vitro receptor autoradiographic study, Brain Res., 561 (1991) 51-60. [11] Cheetham, S,C., Crompton, M.R., Katona, C.LE. and Horton, R.W., Brain 5-HT t binding sites in depressed suicides, Psychopharmacology, 102 (1990) 544-548. [12] Cross, A.J., Crow, T.J., Ferrier, I.N., Johnson, J.A., Bloom, S.R. and Corsellis, J.A., Serotonin receptor changes in dementia of Alzheimer type, J. Neurochem., 43 (1984) 1574-81. [13] Ding, D., Toth, M., Zhou, Y., Parks, C., Hoffman, B. and Shenk, T., Glial cell specific expression of the serotonin 2 receptor gene: selective reactivation of a repressed promoter, Mol. Brain Res., 20 (1993) 181-191. [14] Dillon, K.A., Gross-Isseroff, R., Israeli, M. and Biegon, A., Autoradiographic analysis of serotonin 5-HT1A receptor binding in the human brain post-mortem: effects of age and alcohol, Brain Res., 554 (1991) 56-64. [15] Francis, P.T., Pangalos, M.N,, Pearson, R.C.A., Middlemiss, D.N., Stratmann, G.C. and B o w e n , D . M . , 5HydroxytryptaminelA but not 5-hydroxytryptamine 2 receptors are enriched on neocortical pyramidal neurons destroyed by intrastriatal volkensin, J. Pharmacol. Exp. Ther., 261 (1992) 1273-1281. [16] Francis, P.T., Sims, N.R., Procter, A.W. and Bowen, D.M., Cortical pyramidal neuron loss may cause glutamatergic hypoactivity and cognitive impairment in Alzheimer's disease: investigative and therapeutic perspectives, J. Neurochem., 60 (1993) 1589-1604. [17] Garlow, S.J., Morilak, D.A., Dean, R.R., Roth, B.L. and Cia-

168

[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[29]

[30]

P.W.J. Burnet et al./Brain Research 676 (1995) 157-168 ranello, R.D., Production and characterization of a specific 5-HT 2 receptor antibody, Brain Res., 615 (1993) 113-120. Gellman, R.L. and Aghajanian, G.K., Serotonin 2 receptormediated excitation of interneurons in piriform cortex: Antagonism by atypical antipsychotic drugs, Neuroscience, 58 (1994) 515-525. Gerard, C., Langlois, X., Gingrich, J., Doucet, E., Verge, D., Kia, H.K., Raisman, R., Gozlan, H., El Mestikawy, S. and Hamon, M., Production and characterization of polyclonal antibodies recognizing the intracytoplasmic loop of the 5-HydroxytryptaminelA receptor, Neuroscience, 62 (1994) 721-739. Hillion, J., Catelon, J., Riad, M., Hamon, M. and De Vitry, F., Neuronal localization of 5-HTIA receptor mRNA and protein in rat embryonic brainstem cultures, Dev. Brain Res., 79 (1994) 195 -202. Hirst, W.D., Rattray, M.A.N., Price, G.W. and Wilkin, G.P., Expression of 5-HT1A, 5-HT2A and 5-HT2C receptors in astrocytes, Soc. Neurosci. Abstr., 20 (1994) 1547. Hrdina, P.D., Demeter, E., Vu, T.B., Sotonyi, P. and Palkovits, M., 5-HT uptake sites and 5-HT 2 receptors in brain of antidepressant-free suicide victims/depressives: Increase in 5-HT 2 sites in cortex and amygdala, Brain Res., 614 (1993) 37-44. Joyce, J.N., Lexow, N., Winokur, A., Casanova, M.F. and Kleinman, J.E., A quantitative autoradiographic study of serotonin uptake sites, serotoninlA and serotonin 2 receptors in striatum and cortex of schizophrenics, non-schizophrenic suicides and controls, Neuropsychopharmacology, 8 (1992) 315-336. Kobilka, B.K., Frielle, T., Collins, S., Yang-Feng, T., Kobilka, T.S., Francke, U., Lefkowitz, R.J. and Caron, M.G., An intronless gene encoding a potential member of the family of receptors coupled to guanine nucleotide regulatory proteins, Nature, 329 (1987) 75-79. Laruelle, M., Abi-Dargham, A., Casanova, M.F., Toti, R., Weinberger, D.R. and Kleinman, J.E., Selective abnormalities of prefrontal serotonergic receptors in schizophrenia. A postmortem study, Arch. Gen. Psychiatry, 5 (1993) 810-818. Lopez, J.F., Chalmers, D.T., Sima, A.A.F. and Watson, S.J., Distribution of 5-HT1A receptor mRNA in the human hippocampus, Soc. Neurosci. Abstr., 20 (1994) 1547. Lowther, S., De Paermentier, F., Crompton, M.R., Katona, C.L.E. and Horton, R.W., Brain 5-HTz receptors in suicide victims: violence of death, depression and effects of antidepressant treatment, Brain Res., 642 (1994) 281-289. Marazziti, D., Marracci, S., Palego, L., Rotondo, A., Mazzanti, C., Nardi, I., Ladinsky, H., Girado, E., Borsini, F. and Cassano, G.B., Localization and gene expression of serotonin-lA (5HT1A) receptors in human brain post-mortem, Brain Res., 658 (1994) 55-59. Morilak, D.A., Garlow, S.J. and Ciaranello, R.D., Immunocyto-

[31]

[32]

[33]

[34]

[35]

[36]

[37]

[38] [39]

[40]

[41]

[42]

[43]

chemical localization and description of neurons expressing serotonin 2 receptors in the rat brain, Neuroscience, 54 (1993) 701-717. Morilak, D.A., Somogyi, P., Lujan-Miras, R. and Ciaranello, R.D., Neurons expressing 5-HT 2 receptors in the rat brain: neurochemical identification of cell types by immunocytochemistry, Neuropsychopharmacology, 11 (1994) 157-166. O'Kusky, J. and Colonnier, M.A., A laminar analysis of the number of neurons, glia and synapses in the visual cortex (area 17) of adult macaque monkeys, J. Comp. Neurol., 210 (1982) 278-290. Pazos, A., Probst, A. and Palacios, J.M., Serotonin receptors in the human brain. III. Autoradiographic mapping of serotonin-1 receptors, Neuroscience, 21 (1987) 97-122. Pazos, A., Probst, A. and Palacios, J.M., Serotonin receptors in the human brain. IV. Autoradiographic mapping of serotonin-2 receptors, Neuroscience, 21 (1987) 123-139. Pompeiano, M., Palacios, J.M. and Mengod, G., Distribution and cellular localization of mRNA coding for 5-HTIA receptor in the rat brain: correlation with receptor binding, J. Neurosci., 12 (1992) 440-453. Pompeiano, M., Palacios, J.M. and Mengod, G., Distribution of the serotonin 5-HT 2 receptor family mRNAs: comparison between 5-HT2A and 5-HT2c receptors, Mol. Brain Res., 23 (1994) 163-178. Rahman, S., McLean, J.H., Darby-King, A., Paterno, G., Reynolds, J.N. and Neuman, R.S., Loss of cortical serotonin 5-HTzA signal transduction in senescent rats: reversal following inhibition of protein kinase C, Neuroscience, in press. Rockel, A.J., Hiorns, R.W. and Powell, T.P.S., The basic uniformity in structure of the neocortex, Brain, 103 (1980) 221-244. Saltzman, A.G., Morse, B., Whitman, M.M., Ivanshchenko, Y., Jaye, M. and Felder, S., Cloning of the human serotonin 5-HT 2 and 5-HTxc receptor subtypes, Biochem. Biophys. Res. Commun., 181 (1991) 1469-1478. Shapiro, L.A., Stockmeier, C.A., Lowy, M.T. and Singer, J.F., Distribution of the serotoninlA receptor, serotonin uptake site and tryptophan hydroxylase-like immunoreactivity (TpH-Li) in the human dorsal raphe, Soc. Neurosci. Abstr., 20 (1994) 1547. Whitaker-Azmitia, P.M. and Azmitia, E.C. Stimulation of astroglial serotonin receptors produces culture media which regulate growth of serotonergic neurons, Brain Res., 497 (1989) 80-85. Whitaker-Azmitia, P.M., Clarke, C. and Azmitia, E.C., Localization of 5-HTIA receptors to astroglial cells in adult rats: implications for neuronal-glial interactions and psychoactive drug mechanism of action, Synapse, 14 (1993) 201-205. Zifa, E. and Fillion, G., 5-Hydroxytryptamine receptors, Pharmacol. Rev., 44 (1992) 401-458.