Advances in post mortem molecularneurochemistry and neuropathology:examples from schizophrenia research

Paul J HarrisonUniversity Department of Psychiatry, Warneford Hospital and University Department of ClinicalNeurology (Neuropathology), Radcliffe Infirmary, Oxford, UK

Post mortem studies of psychiatric disorders have been revitalised by severaldevelopments. Molecular biology has provided tools for studying genes and theirexpression in post mortem brain tissue, an approach which facilitates integration ofmolecular genetics with neurochemistry and neuropathology.The techniques canbe used quantitatively as well as qualitatively, applications which have been aidedby developments in image analysis. Accompanying these advances has been animprovement in the robustness of the data as a result of greater attention toconfounding variables and other methodological improvements.These issues areillustrated by two recent areas of interest in schizophrenia research: the expressionof central dopamine D4 and 5-HT receptors, and synoptic pathology.

Postal address:DrPaulJ Harrison,

University Department ofPsychiatry, Warneford

Hospital, OxfordQX3 7 JX, UK

Examining the brain has always been a cornerstone of biologicalpsychiatry. For example, the recent profound advances in understandingof Alzheimer's disease aetiopathogenesis, although largely genetic innature, have their roots in pathochemical studies of Alzheimer's diseasebrain tissue. It was from this source that the composition of amyloidplaques and neurofibrillary tangles was revealed and the central role ofcholinergic deficits in producing the cognitive impairment of the diseasewas established. Such discoveries contributed significantly to theidentification of the disease-causing genes and to potential therapeuticavenues.

Post mortem studies have played similarly important roles in other'organic' psychiatric conditions, whereas for the 'functional' disorders,their contribution has been more modest and their status has waxedand waned. However, the post mortem approach is currently enjoying arenaissance in all areas of psychiatry. This has come about for severalreasons. Firstly, greater attention is being paid to research design andexecution, reducing the influence of confounding variables andincreasing the ability to identify genuine differences between the groupsbeing compared. Secondly, many molecular biological methods haveproved to be suitable for post mortem research, allowing genes and

British Medical Bulletin 1996;52 (No. 3)-.527-538 ©The British Council 1996

by guest on Novem

ber 4, 2016http://bm

b.oxfordjournals.org/D

ownloaded from

Biological psychiatry

gene expression to be studied in addition to the conventionalneurochemical study of receptors, enzymes and so on. Thirdly,experimental approaches have converged, blurring the distinctionbetween abnormalities of structure (neuropathology) and of function(neurochemistry).

These areas of progress are neither mutually exclusive nor exhaustive.However, they serve as a useful focus for a review of recent post mortembiological psychiatry research. The themes are developed and illustratedhere with reference to aspects of schizophrenia.

New tools for old questions

The techniques of molecular biology have been adopted rapidly for usein post mortem research1"3. The ability to detect mRNA allows geneexpression to be investigated and to complement studies which target theprotein products. mRNA-directed techniques have several attractions:molecular probes are more specific, reducing problems of cross-reactivity with related molecular species; they can be designed to detectisoforms or other mRNA variants which may be functionally importantbut undetectable at the protein level, and the abundance of an mRNAgives quantitative information about synthesis of the encoded protein. Arange of techniques for studying mRNAs is now available for postmortem brain research; these techniques vary in their sensitivity,anatomical resolution (from single cell to whole brain) and ease of use(see4-5; Table 1). The ability to retrieve intact mRNA from post mortemtissue also permits techniques such as subtractive hybridization6 and invitro translation/two-dimensional protein electrophoresis7 in which therepertoire of expressed genes can be investigated. In addition, thecloning and sequencing of genes, and the ability to express genes invarious systems, has facilitated the development of an ever increasingrange of antibodies against proteins of interest including receptors,which were previously only detectable by less specific ligands. Theiravailability allows individual gene products to be detected byimmunoblotting8 and immunocytochemistry9.

Many methodologies have been enhanced by the widespreadavailability of computerised image analysis2-10-11. These improve thepotential and practicality of accurate, rapid and automated measure-ments. They have contributed to an increasing preference for tissuesections rather than tissue homogenates as the starting material forresearch, to allow measurements over identified neuronal populations.Suitable images include autoradiograms generated from radioligandreceptor binding, in situ hybridization or immunoautoradiography,

528 British M«£cal Bulktin 1996^2 (No. 3)

by guest on Novem

ber 4, 2016http://bm

b.oxfordjournals.org/D

ownloaded from

Advances in post mortem molecular neurochemistry and neuropathology

Table 1 Methods for analysis of genss and gene expression in post mortem brain tissue

Technique Uses

DNA analyses

Porymerase chain reaction (PCR) Gena sequencing; identification of polymorphisms

RNA anal/ias

cDNA library Profile of expressed genes

In vtv hybridization Localisation and quantitation of cellular mRNA

Northern or ilot blotting Integrity and quantitation of tissue mRNA

Nudeas* protection assay Quantitation of tissue mRNA

Reverse transcriptase-PCR Quantitation of tissue mRNA

Subtroctive hybridization Identify genes expressed differentially between samples

Differential display As subtractive hybridization, but more sensitive

Protein analyses

Immunocytochemisrry Cellular or tissue localisation of a protein

Immunoautoradioaraphy Same principle as ICC, but more reliable quantitation

In vitro translation Biological ochvity of mRNA

Protein eloctrophoresis Pattern of proteins synthesized by mRNA

Receptor autorodiography Localisation and quantrtation of ligand bending sites

Homogenate binding Quantitation and characterization of binding sites

Enzymes* Levels and activities

OtherCulture of neurons and glia* In vitro studies of gene expression

Transmitters and metabolites* Studies of transmitter levels and turnover

'Techniques relatively susceptible to post mortem interval

whilst sections can also be analysed after conventional or histochemicalstaining. Furthermore, expression of mRNAs can be studied inindividual neurons and their environs using liquid emulsion autoradio-graphy and grain counting12-13. The options for research at this highlevel of anatomical resolution are further extended by confocalmicroscopy14.

Neuroanatomical and neuropathological studies have also benefitedfrom a different quantitative advance, that of stereological methods forcounting cells15. They allow robust and biologically meaningful assess-ments of absolute neuronal number to be made which overcome problemsassociated with the traditional assessment of neuronal density (neurons perunit area or volume)16; however, they do require a considerable investmentof time, material and money, which may not always be necessary17.

Taken together, these new tools greatly increase the range and powerof post mortem approaches to biological psychiatry. Examples of recentapplications of these techniques with regard to schizophrenia research aregiven later in this chapter.

BrfhshMeotca/BuWehn 1996;52 (No. 3) 529

by guest on Novem

ber 4, 2016http://bm

b.oxfordjournals.org/D

ownloaded from

Biological psychiatry

Recognition of confounders and improvements inexperimental design

Inherent in all post mortem studies, especially neurochemical ones, is alarge brain-to-brain variability. Much of this experimental noise can beattributed to a number of confounding factors operating in the peri-mortemperiod. These were recognised by neurochemists in the pre-moleculardays18, and have been rediscovered as being of equal if not greaterimportance for measurements of RNA and gene expression19. Howeverpowerful the newer methodologies may be, without adequate characterisa-tion of these factors they are still likely to result in data of doubtful value.

Post mortem interval is usually cited as the main problem with postmortem research. Certainly for some biochemical and ultrastructuralmeasurements it is. However, for the majority of purposes, routine postmortem intervals (at least 48 h) are acceptable and often do not have anydemonstrable effect on the abundance or integrity of the parameter beingstudied19. In fact, it is much more important to know how the person diedthan how long they were dead before the brain was processed. Apowerful influence is brain acidosis, a biochemical correlate of premortem features such as coma or hypoxia. Brain or cerebrospinal fluidpH strongly predicts the abundance of many mRNAs and other geneproducts20*21, such that knowledge of pH and matching betweencomparison groups should become a prerequisite for quantitative humanbrain neurochemistry. As an example, studies of P-amyloid precursorprotein (|3APP) expression in Alzheimer's disease have been notoriouslyinconsistent, probably because the mRNA is vulnerable to pre mortemevents22'23 and this had not been controlled for in earlier studies.Conversely, mRNA changes have recently been described in schizo-phrenia24 and Parkinson's disease25 which take brain pH into account.They may, therefore, be interpreted with more confidence as disease-associated findings.

Many other peri-mortem confounders can affect neurochemistry. Theimportance of standardised protocols for brain collection and processingis apparent and is recognised in most brain banks26. Regardingprocessing, it is noteworthy that conventional formalin fixed, waxembedded material is suitable for a surprising range of the techniqueslisted in Table 1, including extraction of nucleic acids. A recentlydiscovered trick is to microwave the tissue to enhance the success ofimrnunocytochemical and in situ hybridization methods27. In this andother ways, even archival brain material can be retrieved for molecularand imrnunochemical approaches.

The major peri-mortem factors are summarised in Table 2, withstraightforward ways to help overcome them presented in Table 3. It isuncertain how much of the residual neurochemical variance between

5 3 0 British Mtdical Bu/leh'n 19M;52 (No. 3)

by guest on Novem

ber 4, 2016http://bm

b.oxfordjournals.org/D

ownloaded from

Advances in post mortem molecular neurochemistry and neuropathology

Table 2 Confounding variables for post mortem brain studies

PER/MORTEM FACTORS

Pre-mortem (ogond state)

Addosb(pH)

Hypoxia

Fever

Coma

Sepsis

On ventilator

Seizures

Postmortem

Delay from death to processing (post mortem interval)

Decdh-to-refrigeration temperature and interval

Method of freezing or processing

Temperature ot storage

Duration of storage

Duration of tissue fixation

Mkrowaving and other post-processing procedures

OTHER FACTORS

Age

Sex

Race

Genotype

Side of brain

Drugs, alcohol and smoking

Medication

Season of death

Tune of day of death

Death by suicide

Coincidental pathology

For references and further details see

human brains is due to further unrecognised peri-mortem factors, andhow much reflects genuine individual differences, i.e. which are present invivo.

Examples of recent advances from post mortemschizophrenia research

Dopamine D4 receptors

Neurochemical studies of schizophrenia continue to be extremely diverse,reflecting the different hypotheses which abound concerning the nature ofthe abnormalities characterising the disease. However, the major focusremains neuroreceptors implicated by virtue of a belief that they or theirassociated transmitter systems are involved in the disease pathophysiol-ogy, or, more speculatively, as candidate genes.

Brihih Medico/ Bulletin 1996^2 (No. 3) 531

by guest on Novem

ber 4, 2016http://bm

b.oxfordjournals.org/D

ownloaded from

Biological psychiatry

Table 3 Improving experimental design for post mortem neurochemistry

Larger group ikes

Better documentation

Matching of groups for key variables in Table 2

Use of different techniques to measure a given parameter

Statistical manoeuvres (e.g. multiple regression)

Extensive study of normals before extension to disease comparisons

Use of appropriate animal models (e.g. for medication effects)

The prevailing dopamine (DA) hypothesis of schizophrenia gained a new-lease of life with the cloning of the DA receptor gene family, not leastbecause it produced a new molecular haystack amongst which adopaminergic needle might be located. Of the DA receptor gene products,D4 demanded immediate attention because of its high affinity forclozapine, its remarkable allelic variability and its cortical and limbicdistribution28'29. The question arose whether an elevation in D4 receptordensity might occur in schizophrenia, analogous to the earlier studies ofD2 receptors. Unfortunately no specific ligand exists for the D4 receptorand its mRNA has a limited expression in human brain which makes itsdetection problematic. Seeman and colleagues30, therefore, measuredindirectly the density of D4 receptor binding sites in striatal homogenatesusing two ligands: [3H]-emonapride (which detects D2, D3 and D4

receptors), and [3H]-raclopride in the presence of guanine nucleotide(which detects D2 and D3 receptors). The difference in binding betweenthe two ligands was used as an estimate of D4 receptor density. Theyfound that this was elevated about 6-fold in schizophrenics. Using similarsubtractive methodologies, others have confirmed an increase in putativeD4 receptors in the striatum in schizophrenia, though to a lesserdegree31-32. These findings have generated considerable interest, sincethey appear to resuscitate the DA hypothesis and provide a direct linkwith the mechanism of action of clozapine and perhaps other atypicalantipsychotics. However, matters have proved to be complicated.Reynolds and Mason33 cast doubt upon the results of the above studies,since using a competitive (rather than subtractive) method they found noevidence for detectable D4 receptors in human striatal tissue; Seeman andvan Tol34 have countered this criticism with their own reservations aboutthe suitability of the competition method. These controversies are likelyto be resolved only when D4 receptor expression is measured in otherways, either after development of a D4-specific ligand for use in vivo andin vitro, or a D4 receptor antibody, or by D4 receptor mRNA quantitationusing sensitive methods, such as RT-PCR or ribonuclease protectionassay. Regardless of technique, it remains to be established whetherelevations in D4-like receptors in schizophrenia are partially30-35, or

532 Brih'sri Medical Bulletin 1996^2 (No. 3)

by guest on Novem

ber 4, 2016http://bm

b.oxfordjournals.org/D

ownloaded from

Advances in post mortem molecular neurochemistry and neuropathology

wholly, an effect of neuroleptic treatment. Overall, the importance of D4

receptor involvement in schizophrenia remains unclear; it does notappear to be a candidate gene36, and its polymorphisms have proved tobe of less significance for the actions of clozapine than anticipated29'37'38.

5-HTreceptors

Interest in the 5-HT system in schizophrenia has been spurred on by anincreasing body of post mortem data39, by evidence that 5-HT receptorsmay be relevant (perhaps more so than the D4 receptor) to the mechanismof action of atypical antipsychotics40'41, and by genetic data42.

There are two changes in 5-HT receptor expression which seem tocharacterise the disease: an increase in 5-HT1A receptor binding sites infrontal cortex which is not accompanied by increased 5-HT1A receptormRNA43; and a loss of cortical 5-HT2A receptors, which is paralleled by aloss of its mRNA24'43. These data illustrate the value of combining thestudy of a receptor (or other protein) with that of its mRNA. Forexample, if the abundance of a receptor changes but its mRNA doesn't,altered translational or post-translational processes may be invoked,whereas if both receptor and mRNA change in parallel, transcriptionalregulation is most likely. (The third possibility, that the abundance of anmRNA changes but not that of the receptor, is harder to interpret; itsuggests an alteration in gene expression regulation which is compen-sated for at the protein level.) Correlation of findings from each approachwith analysis of receptor polymorphisms24 is a novel extension of postmortem work which will become of increasing importance as effortsbegin to be made to understand relationships between genotype,phenotype and treatment response41; one illustration is in Alzheimer'sdisease, where allelic variation in the apolipoprotein E (apoE) gene affectsapoE expression44 and the pathological features of the disease45, inaddition to its major influence on disease risk and age of onset46.

Synoptic pathology

The discovery that schizophrenia is associated with structural brainabnormalities is at the same time both a small step and a giant leap.Resolution of this long-standing controversy has eventually come aboutthrough a combination of imaging and post mortem studies47-48.However, the location and nature of the pathological changes remainuncertain49 and it is now necessary to identify their histological,neuronal, and molecular characteristics if there is to be a convergenceof structural, biochemical and genetic aspects of the disease.

British Mtdkal Bulletin 1996£2 (No. 3) 533

by guest on Novem

ber 4, 2016http://bm

b.oxfordjournals.org/D

ownloaded from

Biological psychiatry

One component of the neuropathology of schizophrenia appears to besynaptic alterations. Changes in the density, number, type, or structure ofsynapses is important both for neuronal plasticity and as a substrate ofbrain disorders, including Alzheimer's disease and temporal lobeepilepsy. Synaptic alterations have often been proposed to occur inschizophrenia as part of the neurodevelopmental anomaly whichputatively underlies the disease. However, testing of the prediction isdifficult since it has traditionally required electron microscopic investiga-tion, which is severely limited both by the availability of suitable tissueand by the resources needed to carry out a study of adequate size.

Recently, detection of specific presynaptic proteins, such as synapto-physin, has been shown to be a suitable marker of synaptic terminals and,moreover, their abundance can be used as a proxy measure of synapticdensity50. Changes in distribution and level of synaptophysin, measuredeither immunocytochemically or by immunoblotting, have thereby beenused to detect synaptic alterations in aging, degenerative disorders andafter experimental manipulations51.

We have used synaptophysin to address the question of synapticalterations in schizophrenia. Having first characterised the vulnerabilityof synaptophysin and its mRNA to peri-mortem factors52, wecompared hippocampal synaptophysin gene expression in schizophre-nics and matched controls. We found a significant loss of synaptophy-sin mRNA in most subfields in the cases, but only a non-significanttrend for the protein, measured in terms of tissue immunoreactivityafter immunocytochemical detection13. However, the latter method haslimitations as a quantitative technique so we turned to immuno-autoradiography, a combination of immunocytochemistry and auto-radiography and which has advantages in this regard. We were therebyable to demonstrate a loss of synaptophysin in schizophrenia53,supporting our mRNA data. Antipsychotics (haloperidol or clozapine)do not alter hippocampal synaptophysin expression in rats (13andunpublished observations), reducing the likelihood that the changes aresecondary to medication.

The abnormalities in synaptic density and/or plasticity implied by thesynaptophysin data is supported by several other recent findings in themedial temporal lobe in schizophrenia. These include reduced subicularneuronal size54, decreased density of staining of the mossy fibre pathwayfrom dentate gyrus to Ammon's horn55, a loss of embryonic nerve celladhesion molecule56 and differences in hippocampal cytoarchitecture57.However, other synaptic protein data are more equivocal58, indicatingthe need for further studies. Synaptophysin is thought to label all synapticterminals and it remains to be seen whether phenotypically-definedsubpopulations of neurons or synapses are differentially affected inschizophrenia; this question could be approached either by the use of

534 British Medical Bulletin 1996;52 (No. 3)

by guest on Novem

ber 4, 2016http://bm

b.oxfordjournals.org/D

ownloaded from

Advances in post mortem molecular neurochemistry and neuropathology

synaptic proteins limited to certain synaptic types59'60 or by double-labelling techniques.

The synaptophysin work illustrates several of the recent themes of postmortem research mentioned above: more careful attention to confoun-ders such as peri-mortem events and medication, the use of new tools(detection of synaptic protein gene products), the use of severalcomplementary methods to address a single question, and the applicationof image analysis for quantitation.

Summary and future directions

Conceptual and practical limitations of post mortem research will alwaysremain. These are particularly pertinent for schizophrenia and otherdiseases whose neurobiological signature is faint, and for which thequestion of disease heterogeneity and the validity of the syndromalcategory itself remains unanswered61. However, the influence of thelimitations can be minimised by a combination of good experimentaldesign, continuing technical developments, and increasing integrationwith in vivo approaches. If these principles are followed, not only are thedata more robust, but it becomes possible to develop testable hypothesesas well as merely cataloguing ever more detailed alterations62. In thismode, the post mortem approach retains its status as an essentialcomponent of biological psychiatry, and one which is complementary to,not replaced by, developments in functional imaging, peripheral markers,or molecular genetics.

Acknowledgements

I am a Wellcome Trust Senior Research Fellow in Clinical Science. Ithank the Wellcome Trust, Medical Research Council and StanleyFoundation for support, and my research group for their productivityand intellectual input.

References

1 Harrison PJ, Pearson RCA. Gene expression and mental disease. Psycbol Med 1989; 19: 813-9

2 Harrison PJ, Pearson RCA. Principles and practice of in situ hybridization histochemistry inneuropsychiatric research. In: Roberts GW, Polak JM (eds) Molecular Neuropatbology.Cambridge: Cambridge University Press, 1995; 38-60

535

by guest on Novem

ber 4, 2016http://bm

b.oxfordjournals.org/D

ownloaded from

Biological psychiatry

3 Harrison PJ. In vitro brain imaging: techniques for studying and localizing the pathogenesis ofpsychiatric diseases. In: Kerwin RW (ed) Neuroimagtng. Neurobiology and Psychiatry, vol 3.Cambridge: Cambridge University Press, 1995; 157-74

4 Burnet PWJ, Eastwood SL, Harrison PJ. Detection and quantitation of 5-HTj^ and 5-HT2Areceptor mRNAs in human hippocampus using a reverse transcriptase-polymerase chainreaction (RT-PCR) technique and their correlation with binding site densities and age. NeurosciLett 1994; 178: 85-9

5 Burnet PWJ, Eastwood SL, Lacey K, Harrison PJ. The distribution of 5 - H T J A and 5-HT2Areceptor mRNA in human brain. Brain Res 1995; 676: 157-68

6 Mulcrone J, Whatley SA, Ferrier IN, Marchbanks RM. A study of altered gene expression infrontal cortex from schizophrenic patients using differential screening. Schizophr Res 1995; 14:203-13

7 Perrett CW, Whatley SA, Ferrier IN, Marchbanks RM. Changes in relative levels of specificbrain mRNA species associated with schizophrenia and depression. Mol Brain Res 1992; 12:163-71

8 Blackstone CD, Levey AI, Martin LJ, Price DL, Huganir RL. Immunological detection ofglutamate receptor subtypes in human central nervous system. Ann Neurol 1992; 31: 680-3

9 Eastwood SL, Burnet PWJ, Beckwith J, Kerwin RW, Harrison PJ. AMPA glutamate receptorsand their flip and flop mRNAs in human hippocampus. Neuroreport 1994; 5: 1325—8

10 Baskin DG, Stahl WL. Fundamentals of quantitative autoradiography by computerdensitometry for in situ hybridization, with emphasis on P. / Histochem Cytochem 1993;41: 1767-73

11 Royston MC. Neuropathology by numbers: application of image analysis techniques toquantitative neuropathology. In: Roberts GE, Polak JM (eds) Molecular Neuropathology.Cambridge: Cambridge University Press, 1995; 174-84

12 Akbanan S, Kim JJ, Potkin SG et al. Gene expression for glutamic acid decarboxylase is reducedwithout loss of neurons in prefrontal cortex of schizophrenics. Arch Gen Psychiatry 1995; 52:258-66

13 Eastwood SL, Burnet PWJ, Harrison PJ. Altered synaptophysin expression as a marker ofsynaptic pathology in schizophrenia. Neurosaence 1995; 66: 309-19

14 Chen KS, Masliah E, Mallory M, Gage FH. Synaptic loss in cognitively impaired aged rats isameliorated by chronic human nerve growth factor infusion. Neuroscience 1995; 68: 19-27

15 West MJ. Advances in the study of age-related neuron loss. Semin Neurosa 1994; 6: 403—1116 Oorschot DE. Are you using neuronal densities, synaptic densities or neurochemical densities as

your definitive data? There is a better way to go. Prog Neurobiol 1994; 44: 233—4717 Ma SY, Collan Y, Rytt M, Rinne JO, Rinne UK. Cell counts in the substantia nigra: a

comparison of single section counts and disector counts in patients with Parkinson's disease andin controls. Neuropathol Appl Neurobiol 1995; 21: 10-7

18 Perry EK, Perry RH. Human brain neurochemistry: some postmortem problems. Life Sci 1983;33: 1733-43

19 Barton AJL, Pearson RCA, Najlerahim A, Harrison PJ. Pre and post mortem influences on brainRNA. / Neurochem 1993; 61: 1-11

20 Harrison PJ, Heath PR, Eastwood SL, Burnet PWJ, McDonald B, Pearson RCA. The relativeimportance of premortem acidosis and postmortem interval for human brain gene expressionstudies: selective mRNA vulnerability and a comparison with their encoded proteins. NeurosciLett 1995; 200: 151-4

21 Kingsbury AE, Foster OJF, Nisbet AP et al. Tissue pH as an indicator of mRNA preservation inhuman post-mortem brain. Mol Brain Res 1995; 28: 311-8

22 Harrison PJ, Barton AJL, Procter AW, Bowen DM, Pearson RCA. The effects of Alzheimer'sdisease, other dementias and premortem course upon amyloid B precursor protein messengerRNAs in frontal cortex. / Neurochem 1994; 62: 635-44

23 Procter AW, Francis FT, Holmes C et al. APP isoforms show correlations with neurones but notwith glia in human brain. Ada Neuropathol 1994; 88: 545-52

24 Burnet PWJ, Harrison PJ. Genetic variation of the 5-HT2A receptor and response to clozapine.Lancet 1995; 346: 909

536 Bnhtfi Medico/ Bulletin 1996;52 (No. 3)

by guest on Novem

ber 4, 2016http://bm

b.oxfordjournals.org/D

ownloaded from

Advances in post mortem molecular neurochemiitry and neuropathology

25 Nisbet AP, Foster OJF, Kingsbury A et al. Preproenkephalin and preprotachykinin messengerRNA expression in normal human basal ganglia and in Parkinson's disease. Neuroscience 1995;66: 361-76

26 Vonsattel JPG, Aizawa H, Ge P et al. An improved approach to prepare human brains forresearch. / Neuropathol Exp Neurot 1995; 54: 42-56

27 Login GR, Dvorak AM. Application of microwave fixation techniques in pathology toneuroscience studies: a review. / Neurosci Methods 1994; 55: 173—82

28 van Tol HHM, Bunzow JR, Guan HC et al. Cloning of the gene for a human dopamine D4receptor with a high affinity for the antipsychotic clozapine. Nature 1991; 350: 650-4

29 van Tol HHM, Wu CM, Guan H-C et al. Multiple dopamine D4 receptor variants m the humanpopulation. Nature 1992; 358: 149-52

30 Seeman P, Guan H-C, van Tol HHM. Dopamine D4 receptors elevated in schizophrenia. Nature1993; 365: 441-5

31 Murray AM, Hyde TM, Knable MB et al. Distribution of putative D4 dopamine receptors inpostmortem striatum from patients with schizophrenia. / Neurosci 1995; 15: 2186—91

32 Sumiyoshi T, Stockmeier CA, Overholser JC, Thompson PA, Meltzer HY. Dopamine D4receptors and effects of guanine nucleotides on [^H]raclopride binding in postmortem caudatenucleus of subjects with schizophrenia or major depression. Brain Res 1995; 681: 109-16

33 Reynolds GP, Mason SL. Are striatal dopamine D4 receptors increased in schizophrenia? /Neurochem 1994; 63: 1576-7

34 Seeman P, van Tol HHM. Dopamine D4-like receptor elevation in schizophrenia: cloned D2 andD4 receptors cannot be discriminated by raclopride competition against [ H]nemonapride. /Neurocbem 1995; 64: 1413-5

35 Schoots O, Seeman P, Guan H-C, Paterson AD, Van Tol HHM. Long-term halopendol elevatesdopamine D4 receptors by 2-fold in rats. Eur ] Pharmacol 1995; 289: 67-72

36 Sommer SS, Lind TJ, Heston LL, Sobell JL. Dopamine D4 receptor variants in unrelatedschizophrenic cases and controls. Am J Med Genet (Neuropsychiatr Genet) 1993; 48: 90-3

37 Rao PA, Pickar D, Gejman PV, Ram A, Gershon ES, Gelertner J. Allelic variation in the D4dopamine receptor (DRD4) gene does not predict response to clozapine. Arch Gen Psychiatry1994; 51: 912-7

38 Shaikh S, Collier D, Kerwin RW et al. Dopamine D4 receptor subtypes and response toclozapine. Lancet 1993; 341: 116

39 Roth BL, Meltzer HY. The role of serotonin m schizophrenia. In: Bloom FE, Kupfer DJ (eds)Psychopharmacology. The Fourth Generation of Progress. New York: Raven Press, 1995;1215-28

40 Stockmeier CA, DiCarlo JJ, Zhang Y, Thompson P, Meltzer HY. Characterization of typicaland atypical antipsychotic drugs based on in vivo occupancy of serotonin and dopamine2receptors. / Pharmacol Exp Ther 1993; 266: 1374-94

41 Arranz M, Collier D, Sodhi M et al. Association between clozapine response and allelic variationin 5-HT2A receptor gene. Lancet 1995; 346: 281-2

42 Harrison PJ, Geddes JR. Schizophrenia and the 5HT1A receptor gene. Lancet 1996; 347: 127443 Burnet PWJ, Eastwood SL, Harrison PJ. 5-HTM receptor mRNAs and binding site densities are

differentially affected in schizophrenia. Neuropsychopharmacol 1996; In press44 Bertrand P, Poirier J, Oda T, Finch CE, Pasinetti GM. Association of apolipoprotein E genotype

with brain levels of apolipoprotein E and apolipoprotein J (clusterin) in Alzheimer disease. MolBrain Res 1995; 33: 174-8

45 Ohm TG, Kirca M, Bohl J, Schnargl H, Gross W, Mrz W. Apolipoprotein E polymorphisminfluences not only cerebral senile plaque load but also Alzheimer-type neurofibrillary tangleformation. Neuroscience 1995; 66: 583-7

46 Stnttmatter WJ, Roses AD. Apolipoprotein E and Alzheimer disease. Proc Natl Acad Sci USA1995; 92: 4725-7

47 Harrison PJ. On the neuropathology of schizophrenia: neurodevelopmental, neurodegenerarive,or both? Neurodegeneration 1995; 4: 1-12

48 Weinberger DR. Schizophrenia: from neuropathology to neurodevelopment. Lancet 1995; 346:552-7

Brihj/i Medical BulhHn 1996;32 (No. 3) 537

by guest on Novem

ber 4, 2016http://bm

b.oxfordjournals.org/D

ownloaded from

Biological psychiatry

49 Shapiro RM. Regional neuropathology of schizophrenia. Where are we? Where are we going?Schizophr Res 1993; 10: 187-239

50 Masliah E, Terry RD, Alford M, DeTeresa R. Quantitative immunohistochemistry ofsynaptophysin in human neocortex: an alternative method to estimate density of presynapricterminals in paraffin sections. / Histochem Cytochem 1990; 38: 837-44

51 Masliah E, Terry RD. The role of synaptic proteins in the pathogenesis of disorders of thenervous system. Brain Pathol 1993; 3: 77-86

52 Eastwood SL, Burnet PWJ, McDonald B, Clinton J, Harrison PJ. Synaptophysin gene expressionin human brain: a quantitative in situ hybridization and immunocytochemical study.Neuroscience 1994; 59: 881-92

53 Eastwood SL, Harrison PJ. Decreased synaptophysin in the medial temporal lobe inschizophrenia demonstrated using immunoautoradiography. Neurosaence 1995; 69: 339—43

54 Arnold SE, Franz BR, Gur RC et al. Smaller neuron size in schizophrenia in hippocampalsubfields that mediate cortical-hippocampal interactions. Am] Psychiatry 152: 738—48

55 Goldsmith SK, Joyce JN. Alterations in hippocampal mossy fiber pathway in schizophrenia andAlzheimer's disease. Biol Psychiatry 1995; 37: 122-6

56 Barbeau D, Liang JJ, Robitaille Y, Quirion R, Srivastava LK. Decreased expression of theembryonic form of the nerve cell adhesion molecule in schizophrenic brains. Proc Natl Acad SaUSA 1995; 92: 2785-90

57 Harrison PJ, Eastwood SL, Esiri MM, Zaidel DW. Cytoarchitectural hippocampal asymmetriesin schizophrenia. Schizophr Res 1996; 18: 180

58 Browning MD, Dudek EM, Rapier JL, Leonard S, Freedman R. Significant reductions insynapsin but not synaptophysin specific activity in die brains of some schizophrenics. BiolPsychiatry 1993; 34: 529-35

59 Jorgensen OS. Neural cell adhesion molecule (NCAM) as a quantitative marker in synapticremodelling. Neurocbem Res 1995; 20: 533-47

60 Takahashi S, Yamamoto H, Matsuda Z et al. Identification of two highly homologouspresynapnc proteins distinctly located at the dendritic and somatic synapses. FEBS Lett 1995;368: 455-60

61 Benes FM. Post-mortem structural analyses of schizophrenic brain: study designs and theinterpretation of data. Psychtatr Dev 1988; 3: 213-26

62 Kerwin RW. Glutamate receptors, microtubule associated proteins and developmental anomalyin schizophrenia: an hypothesis. Psychol Med 1993; 23: 547-51

538 Bnh'sfi Medical Bulletin 1996£2 (No. 3)

by guest on Novem

ber 4, 2016http://bm

b.oxfordjournals.org/D

ownloaded from