Concentrations of Somatostatinlike
GarthBissette, PhD; Erik Widerl\l=o"\v,MD, PhD; Helena Wall\l=e'\us,MD; Ingvar Karlsson, MD, PhD;
KurtEklund, MD; AndersForsman, MD, PhD;Charles B. Nemeroff, MD, PhD
\s=b\The concentration of somatostatinlike immunoreactivity
in cerebrospinalfluid (CSF) from normal, healthy volunteers
(n=10) and patients with DSM-IIIdiagnoses of major depres-
schizophrenia (n =11), or dementia (n=29) was
measured by a sensitive and specific radioimmunoassay.
Statistically significant decreases in CSF concentrations of
somatostatinlike immunoreactivitywere seen in all threepa-
tient populations when compared with controls. These find-
ingsconfirm previous reportsof decreased concentrations of
somatostatinlikeimmunoreactivityin the CSF ofpatientswith
depression and dementia and extend this observation to
patientswith schizophreniaas well. These findingsare con-
cordant with the view that reductions in somatostatinlike
immunoreactivity concentrations are associated with dis-
eases in which cognitivefunction is disturbed.
hypothalamus,1 manyof therequisiteneurotransmitter
criteria have been fulfilled for thistetradecapeptide.The
originalendocrine role of the inhibition ofgrowthhormone
the time that has elapsedsince theisolation, sequenc¬
ing,and characterization of somatostatin from the ovine
release has now been expanded to include inhibition of
release of thyroid-stimulatinghormone2 and avarietyof
peripheral peptidehormones including gastrin, insulin,
cholecystokinin, vasoactive-intestinal peptide,andgluca-
gon.3Infact,stomatostatin has been termed"panhibin"
because of itspropertyofinhibitingthe secretion ofmany
Theheterogeneousdistribution of somatostatinlike im¬
munoreactivityin both hypothalamicandextrahypotha-
lamic brainregions,4,5as well as in thegastrointestinal
tract,6,7 has established that this neuropeptide'srole in¬
cludes more than inhibition ofrelease in neuroendocrine
regulation. Subcellular localization of somatostatinlike in
the brain afterdensity-gradient centrifugationdisclosed
that thepeptideis concentrated in thesynaptosomalfrac¬
tion8 and that thepeptideis released from brain slicesby
depolarizingstimuli in acalcium-dependentmanner.9Soma¬
tostatin isenzymatically degraded by specificbrainpep-
tidases10 and thesedata,as well as thepresenceofhigh-
pellingevidencethat somatostatinis a neurotransmitter or
neuromodulator in themammalian centralnervous system.
Somatostatin has been localized in the normal human
brain1214 andcerebrospinalfluid (CSF).15The concentration
of somatostatin has beenreportedto be altered in central
nervoussystemtissue and/or the CSFin avarietyof neuro¬
psychiatrie disorders, including Huntington^ disease,14,1617
Parkinson'sdisease,18"20 depression,21"23 multiple sclerosis,24
and Alzheimer's disease.16,25"28
Thisstudywasperformedtocompare directlythe CSF
somatostatinlike immunoreactivitylevels in thesegroups
with the same antiserum in asingle assayand to confirm
and extend previous findings.Thisreportconcerns mea¬
surement of somatostatinlikeimmunoreactivityconcentra-
Acceptedforpublication July 29, 1986.
From the Departments ofPsychiatry (Drs Bissette and Nemeroff) and
Pharmacology (Dr Nemeroff), and the Center for Aging and Human
Development (Dr Nemeroff), Duke University Medical Center, Durham,
NC; the Department of Psychiatry and Neurochemistry, University of
Lund, Lund, Sweden (Dr Widerl\l=o"\v);the Psychiatric Research Center,
UniversityofUppsala, Uppsala, Sweden(Dr Walle\l=u'\s);theDepartmentof
Psychiatry and Neurochemistry (Dr Karlsson), and Lillhagen Hospital
DepartmentofPsychiatry (Dr Forsman), UniversityofGothenburg, His-
ings Backa, Gothenburg, Sweden; and SaterHospital, Sater, Sweden(Dr
Reprint requeststo DepartmentofPsychiatry, Box 3859, Duke Univer-
sityMedical Center, Durham, NC 27710 (Dr Bissette).
at University of Miami School of Medicine, on March 8, 2011 www.archgenpsychiatry.com Downloaded from
Concentration of somatostatinlike immmunoreactivity (SRIF-LI) in
cerebrospinalfluid(CSF)of normalcontrols,DSM-IIImajor depres¬
sives, DSM-III schizophrenics, and patientswith senile dementia
(DSM-III primary degenerative dementia)or multi-infarct dementia
or both. Statistical analysis was by analysis of variance and
Newman-Keuls test, as well asMann-Whitney U test. Compared
with controls, all three patient groupshad significant (P<.05 by
eithertest) decreases in SRIF-LI concentration.
tions in CSF frompatientswithdepression, dementia,and
schizophrenia,as well as innormal, healthyvolunteers.
SUBJECTS AND METHODS
The diagnostic evaluations and lumbar puncture procedures
were conducted at various Swedishpsychiatricclinics: thePsychi¬
atric Research Center, Uppsala (healthy volunteers and major
depressives); St Jörgen Hospital, Gothenburg (dementias); and
Sater Hospital, Sater (schizophrenics). The subjects had given
their informed consent and theinvestigationhad beenapproved by
the Ethics Committees at the medical faculties oftheUniversityof
Thehealthyvolunteergroup(n=10)comprisedfive men and five
women whose mean (±SEM)agewas 34.2±3.2years (range,21 to
55years).Thepatientsweredrug-freefor at least two weeks and
fulfilled DSM-III criteria29 for a diagnosis of major depression,
dementia, or schizophrenia. The patientswith major depression
(n=17) consisted of 11 men and six women with a mean age of
49.1±3.5 years (range, 25 to 70 years). The schizophrenic group
(n=10) comprisedseven women and three men with a meanageof
46.6 ±3.7 years (range, 27 to 66 years). The dementedgroup
(n=29) comprised both multi-infarct and primary degenerative
typesof dementia and included 20 women and nine men, with a
meanageof81.0±1.0years (range,73 to 92years).Allgroupswere
assessed forsignificantcorrelations between CSF concentrations
of somatostatin andage, sex, height, and weight. In addition, in
the depressed patients CSF concentrations of homovanillic acid,
5-hydroxyindoleacetic acid, and3-methoxy-4-hydroxyphenylgly-
col were determined and theseverityoftheirdepressive symptoms
was assessedusingthe HamiltonDepression RatingScale30 and the
ten-item subscale from the Clinical Psychopathological Rating
Scale fordepression.31 Theyalso received a thyrotropin-releasing
hormone stimulation test and a dexamethasone suppressiontest.
Statistically significantcorrelations between thesebiologicvaria¬
bles and CSF concentrations of somatostatin were alsosought.
Lumbarpunctureswere performedat 9 am in all subjects, as
previouslydescribed.32 The CSF was frozen and stored at -80°C
untilaliquotedfor the somatostatinlikeimmunoreactivityradioim¬
munoassay. Duplicate 400-µ . aliquots of unextracted CSF were
lyophilized and reconstituted with 200 µ , of radioimmunoassay
(lOmM sodium phosphate, 0.15M sodium chloride, 0.01% sodium
azide, and 0.1% gelatin; pH 7.6). Standard curves were prepared
with 400 µ ; of artificial CSF that had been lyophilized and
reconstituted in 200 µ , of radioimmunoassay buffer containing
serial dilutions from 5120 to 0.625 pgofsynthetic cyclicsomato-
Antiserum to somatostatin was raised in New Zealand white
rabbits. This antiserumrecognizesbothcyclic and linear somato-
statin,.u, somatostatin,.^ somatostatin,.^, and Tyr,-somatostatin
with equal affinity. It does not recognize Tyru-somatostatin or a
substituted D-Trp8-somatostatin, indicating that the amino acid
sequence8-11 is essential for activity with this antiserum. This
antiserum(100 µ was added toassaytubes inradioimmunoassay
buffercontaining1.5%normal rabbit serum at afinal concentration
of 1:8000 and can detect as little as 0.625 pg per tube with 50%
displacement of tracer at 85pg.Radioactive tracer is prepared
using eitherTyr°-orTyr^somatostatin,.,,,(5 µ-g) with the chlo-
ramine-T method. The somatostatin is reacted with 1.0 mCi of
iodine 125 in 20 µ , of 0.5M sodium phosphate (pH, 7.5), by the
addition of 20 µ ^ chloramine (15 mg/10 µ!., in 0.05 sodium
phosphate, pH7.5). After 20 s, the reaction is terminated with an
excess (300 µ of10% bovine serum albumin. Metabisulfite can be
used to terminate the reaction, but the generation of linear
somatostatin may compromiseantisera that do notrecognize the
linear form. The iodination mixture isappliedto acarboxymethyl-
cellulose column (1.0x10.0 cm) equilibrated with 0.002M am¬
monium acetate buffer (pH 4.6). The column is washed with ten
volumes of this acetate buffer and the iodinated somatostatin is
eluted with 2N acetic acid. Approximately25 000 cpmofthe trace
contained in the fractionexhibitingmaximum bindingis added to
each assay tube in a 50 µ , volume of radioimmunoassay buffer.
Tubes are incubated at 4°C for 24 hours and subsequentlyreceive
10 µ].,ofgoatanti-rabbit serum toprecipitatethe bound fraction.
Tubes are incubatedagainat 4°C for 24 hours and thencentrifuged.
Supernatantsareaspiratedandpelletsare counted for two minutes
each in agammacounter. Results are expressed aspicogramsof
somatostatinpermilliliter of CSF.
Allsampleswere measured in asingle assaywith anintra-assay
coefficient ofvariation of4.0%. Data wereanalyzedforstatistically
significantdifferencesbybothparametric (analysisofvariance and
Newman-Keuls) and nonparametric (Mann-Whitney U) tests.
The results are shown in the Figure. Thehealthyvolunteers had
mean (±SEM) CSF concentrations of116.1±15.7pg/mLofsomato¬
statinlike immunoreactivity (range, 22.6 to 174.5 pg/mL). The
CSF of depressed patients contained
immunoreactivity concentrations of 65.4 ±7.8 pg/mL of CSF
(range, 20.8 to 140.3 pg/mL). The demented patients exhibited
69.3±6.6 pg/mL of CSF (range, 16.4 to 139.5 pg/mL), with no
significant difference between the patients suspected of having
Alzheimer's disease and the dementias presumably due to other
causes such as multi-infarct dementia. Finally, CSF from the
schizophrenic patients contained a mean somatostatinlike immu¬
noreactivity concentration of 68.7±7.7 pg/mL of CSF (range,
25.2 to 81.0pg/mL).All threepatientgroupmeans were found to be
significantlylower than the controlgroupat the P<.05 level after
either ananalysisof variance followedbythe Newman-Keuls test
or the Mann-Whitney U test.
There were nostatistically significantcorelations between CSF
concentrations of somatostatin in the depressed patientgroupand
to a thyrotropin-releasinghormone challenge or postdexametha-
at University of Miami School of Medicine, on March 8, 2011 www.archgenpsychiatry.comDownloaded from
soné serum cortisol concentrations. There were also no statis¬
tically significantcorrelations for thedepressed patientsbetween
CSF concentrations of somastatin and CSF concentrations of ho-
movanillicacid, 5-hydroxyindoleacetic acid,and 3-methoxy-4-hy-
droxyphenylglycol, or severity ratings for depression based on
Hamilton or ClinicalPsychopathological Rating Scale scores. No
statistically significant correlation between CSF concentration
andage, sex, height, orweight was observed inany group.
Thesefindings supportthe conclusions reached in several
previousstudies that the mean concentration of CSF soma¬
tostatin is reduced inpatientswithdepressionandpatients
with dementia due to Alzheimer's disease, when such
trols. Severalgroups—Gemer and Yamada,21Rubinow et
al,22and Black et al23—have described decreased mean CSF
concentrations of somatostatin indepressed patientsthat
were57%, 63%,and 40% ofcontrols, respectively.The
decrease inmean CSF concentrations ofsomatostatin in our
depressed patientsis 56% of the meanhealthyvolunteer
concentration and is inagreementwith findingsin those
Rubinow33 recently reportedanegative correlation be¬
tween postdexamethasone serum cortisol concentrations
and CSF concentrations of somatostatin in agroupof
depressed patients. However,in thepresent study,neither
thedepressed patients exhibiting escapefrom dexametha¬
sonesuppression nor thosedepressed patientswith
normal cortisolresponse to thissynthetic glucocorticoid
showedany significantcorrelation betweenpost-dexa-
methasone suppressiontest serum cortisol and CSF con¬
centrations of somatostatin. While these findings seem to
bediscordant,there aremany possible confoundingvaria¬
bles, eg, diagnosticcriteria or antiserumspecificity,that
could explainthisapparent discrepancy.
Severalgroupshavepreviously reporteddecreased CSF
concentrations of somatostatin inpatientswithpresumed
Alzheimer's diseasecompared with controls. Oram et al25
observed a 75% reduction in CSF concentrations ofsomato¬
statin inpatientswith senile dementia (n=16) compared
with "controls" (n=17) with variousneurologicdiseases.
Serby et al26 reported a 45% reduction in somatostatin
concentrations in the CSF ofpatients diagnosedashaving
Alzheimer's disease (n=10) comparedwith various neu¬
rologiccontrols (n=9). Beai and Martin16 observed a 40%
decrease in immunoreactive somatostatin levelsinthe CSF
ofpatients diagnosedashavingAlzheimer's disease(n=21)
comparedwith controls(n=30),while Wood etal27 observed
a 60% decreased in somatostatin concentrations when com¬
paringthe CSF ofpatientswith adiagnosisof Alzheimer's
disease(n=11) with normalmyélographiecontrols (n=19).
In thepresent study,we have also observed a 40% decrease
in somatostatin concentrations in our populationof de¬
thechange.These observations alsoagreewith the well-
documented finding byourgroup34and others28,35,36 that
somatostatin concentrations are decreased inpostmortem
tissuesamplesfrom several cortical and subcortical brain
regionsinpatientswith Alzheimer's disease.
Conflictingresults have beenreportedinpreviousin¬
vestigationsthatsoughtto determine whether concentra¬
tions of somatostatin in CSF are altered inschizophrenia.
Gerner et al37 studied twoseparate populations of schizo¬
phrenic patients (National Institutes ofHealth, Bethesda,
Md,and UCLA) andreportedthatcomparedwith normal
California controls(n=31),theschizophrenicpatientsat the
National Institutes of Health (n=53)had a twofold eleva¬
tion in CSF somatostatin levels,while the schizophrenic
patientsat UCLA(n=19)had normal concentrations. Wood
et al27 did not observe a significantdifference in CSF
somatostatin concentrations in schizophrenic patients
(n=8) whencomparedwith controls with normalmyelo-
grams (n=19). Ourpopulationofschizophrenic patients
exhibited a mean decrease of approximately50% in CSF
somatostatin concentrations comparedwith thehealthy
volunteer controls. A recentreport byBeai and Martin38has
shown decreasedregionalbrain somatostatin concentra¬
tions in ratsreceiving long-term treatment with several
neuroleptic drugs.Ourpatientsweredrug-freeforat least
two weeks before a CSF samplewasobtained,as were the
patientsof Gemer et al,37but whether two weeks is suffi¬
cient to reverse aneuroleptic-inducedalteration of neu¬
ropeptideconcentration is unknown. The resolution of the
disparitybetween our results inschizophrenic patientsand
those of Wood et al27 and Gemer et al37 must await a future
studyin which CSF is withdrawn before and after treat¬
ment withneuroleptic drugs.
The concentration of somatostatin in the CSF has also
beenreportedto be reduced inpatientswithHuntington^
chorea,17Parkinson's disease with associateddementia,18
and activemultiplesclerosis.24 This concatenation ofresults
is concordant with thehypothesis that reduced levels of
somatostatin in the CSF maybe an indicator ofcognitive
dysfunction. Indeed, depression, schizophrenia, and de¬
mentia are all disorders that are characterizedbydisturbed
cognition. Insupportof thishypothesisis our recent
observation39 that injection ofcysteamine directlyinto the
aprofound impairmentofpassiveavoidance behavior in
rats. It is evident that future work must concentrate on the
neurobiologiemechanisms that result in decreased CSF
levels ofsomatostatin, eg,neuronaldegenerationand im¬
pairedrelease. It does notappearthat this indicator will
havespecific diagnostic utility.
This research was supported by the Swedish Medical Research Council
(B86-21X-07517-01A);National Institute of Mental Healthgrants MH-39415,
MH-40524, MH-40159,and NIAAG-05128;and the Scottish Rite Foundation.
Dr Nemeroff is the recipientof a Nanaline H. Duke Fellowship from Duke
UniversityMedical Center. J. S. Kizer, MD, University of North Carolina
School ofMedicine, Chapel Hill, provided the antiserum to somatostatin.
1. BrazeauP, ValeW, Burgus R, Ling N, ButcherM, RivierJ, Guille-
min R:Hypothalamicpolypeptidethat inhibits the secretion ofimmunoreac-
tivepituitary growthhormone. Science1973;197:77-79.
2. McCann SM: Physiology and pharmacology of LHRH and somato-
statin. Ann Rev Pharmacol1982;22:491-515.
3. Reichlin S: Somatostatin, inKrieger DT, Brownstein MJ, Martin JB
(eds): BrainPeptides. NewYork,JohnWiley& SonsInc, 1983, pp711-752.
4. BrownsteinM,ArimuraA,StaoH, Schally AV, KizerJS: Theregional
distribution of somatostatin in the rat brain. Endocrinology 1975;96:1456\x=req-\
5. Kobayashi RM,BrownM,Vale WW:Regionaldistribution ofneuroten-
sin and somatostatin in rat brain. Brain Res1977;126:584-588.
6. Luft R, Efendic S, Hokfelt T, Johansson O, Arimura A: Immu-
nohistochemical evidence for the localization of somatostatin-like immu-
noreactivity in a cell population of pancreatic islets. Med Biol 1974;52:
7. Schultzberg M, Drefus CF, GershonMD, Hokfelt T, Elde RP, Nils-
sonG, Said SI, Goldstein M: VIP, enkephalin, substance P and somato-
statin-likeimmunoreactivityin neurons intrinsic to the intestine: Immuno-
histochemical evidence fromorganotypictissue cultures. Brain Res 1978;
8. Epelbaum J, Brazeau P, Tsang D, BrawerJ, Martin JB: Subcellular
distribution ofradioimmunoassayablesomatostatin in rat brain. Brain Res
at University of Miami School of Medicine, on March 8, 2011 www.archgenpsychiatry.comDownloaded from
9. Iversen LL, Iversen SD, Bloom F, Douglas C, Brown M, Vale W:
Calcium-dependentrelease of somatostatin and neurotensin from rat brain
in vitro. Nature1978;273:161-163.
10. Marks N, Stern F: Inactivation of somatostatin (GH-RIH) and its
analogs by crude and partially purified rat brain extracts. FEBS Lett
11. Srikant CB, Patel YC: Somatostatinreceptors:Identification and
characterization in rat brain membranes. Proc Natl Acad Sci USA1981;78:
12. GeolaFL,YamadaT, WarwickRJ, Tourtelotte WW, Hershman JM:
Regionaldistribution of somatostatin-likeimmunoreactivityin the human
brain. Brain Res1981;229:35-42.
13. Cooper PE,FernstromMH,RorstadOP, LeemanSE,MartinJB: The
regionaldistribution ofsomatostatin,substance P and neurotensin in human
brain. Brain Res1981;218:219-232.
14. NemeroffCB, Youngblood WW, Manberg PJ, Prange AJ, Kizer JS:
Regionalbrain concentrations ofneuropeptides inHuntington'schorea and
15. Patel YC, RaoK, Reichlin S: Somatostatin in human cerebrospinal
fluid. NEngl J Med1977;296:529-533.
16. Beal MF, Martin JB: Somatostatin: Normal and abnormal observa-
tions in the central nervoussystemin Alzheimer'sdisease, in WurtmanRJ,
CorkinS, Growdon J (eds): Advances in Brain Research and Therapies.
Boston, Center for Brain Sciences and MetabolismTrust, 1984, pp229-259.
17. CramerH,KohlerJ, Ocpen G, Schomberg G,Schoter E:Huntington's
chorea: Measurements ofsomatostatin, substance P andcyclicnucleotides
in thecerebrospinalfluid. J Neurol1981;228:183-187.
18. Epelbaum J, Ruberg M, Moyse E, Javoy-Agid F, DuboisB, Agid Y:
Somatostatin and dementia in Parkinson's disease. Brain Res1983;278:376\x=req-\
19. Dupont E,ChristensenSE,HansenAP,OlivariusBF,Orskov H: Low
cerebrospinal fluid stomatostatin in Parkinson's disease: An irreversible
abnormality. Neurology 1982;32:312-314.
20. ChristensenSE, Dupont E, Mondrup K, OlivariusBF, Orskiv H:
Parkinson's disease andbenignessential tremor: Somatostatin-like immu-
noreactivity in the cerebrospinal fluid, in Hassler RG, Christ JF (eds):
Advances inNeurology. NewYork,RavenPress, 1984, pp325-331.
21. Gerner RH, Yamada T: Altered neuropeptide concentrations in
cerebrospinalfluid ofpsychiatric patients.Brain Res1982;238:298-302.
22. RubinowDR,GoldPW, PostRM, BallengerJC, Cowdry R, Ballinger
J, Reichlin S: CSF somatostatin in affective illness. Arch GenPsychiatry
23. Black PM, Ballantine HT, Carr DB, Beal MF, Martin JB: Beta-
endorphin and somatostatin concentrations in the cerebrospinal fluid of
patientswithdepressive disorder, abstract 467. Programand abstracts of
the 39th Annual Convention of the Society ofBiological Psychiatry, Los
Angeles, May 2-6, 1984.
24. SorensenKV, ChristensenSE, Dupont E, HansenAP, PedersenE,
Orskov H: Low somatostatin content in cerebrospinal fluid in multiple
sclerosis. Acta Neurol Scand 1980;61:186-191.
25. Oram JJ, EdwardsonJ, Millard PH: Investigation ofcerebrospinal
fluidneuropeptides inidiopathicsenile dementia. Gerontology 1981;27:216\x=req-\
26. Serby M, RichardsonSB, Twente S, SiekierskiJ, Rotrosen J: CSF
somatostatin in Alzheimer's disease. NeurobiolAging 1984;5:187-189.
27. WoodPL,EtienneP,LalS,GauthierS, Cajal S,Nair NPV: Reduced
lumbar CSF somatostatin levels in Alzheimer's disease. Life Sci1982;31:
28. RossorMN,EmsonPC, Mountjoy CQ,RothM,Iversen LL: Reduced
amounts of immunoreactive somatostatin in the temporalcortex in senile
dementia of the Alzheimertype.Neurosci Lett1980;29:373-377.
29. AmericanPsychiatric Association Committee on Nomenclature and
Statistics:Diagnostic and Statistical Manual ofMentalDisorders, ed 3.
Washington, DC, AmericanPsychiatric Association, 1980.
30. Hamilton M:Developmentof rating scale forprimary depressive
illness. Br J Soc ClinPsychol 1967;6:278-296.
31. Montgomery SA, AsbergM: A newdepressionscaledesignedto be
sensitive to change. Br JPsychiatry 1979;134:382-389.
32. Widerl\l=o"\vE, LindstromLH, Besev G, Manberg PJ, Nemeroff CB,
BreeseGR, KizerJS, PrangeAJ Jr: Subnormalcerebrospinalfluid levels of
neurotensin in asubgroupofschizophrenics:Normalization afterneurolep-
tic treatment. Am JPsychiatry 1982;139:1122-1126.
33. Rubinow DR: Cerebrospinal fluid somatostatin and psychiatric ill-
ness. BiolPsychiatry 1986;21:341-365.
34. Nemeroff CB, Bissette G, Reynolds GA, Kizer JS, Kilts CD,
Knight D, Daughtry G,Widerl\l=o"\vE:Analysisof theintegrityofneuropep-
tide-containing neuralsystemsin senile dementia of the Alzheimer'stype:
factor-containingneurons,inProceedings of the 42ndAnnualMeetingofthe
American GeriatricsSociety. NewYork,American GeriatricsSociety, 1985,
35. Davies P, Katzman R, Terry RD: Reduced somatostatin-like immu-
noreactivity in cerebral cortex from cases of Alzheimer's disease and
Alzheimer senile dementia. Nature1980;288:279-280.
36. Roberts GW, Ferrier IN,
AvenoDGC, Hamilton AJ, McGregor G, O'Shaughnessey D,PolakJ,Bloom
SR:Peptides,the limbic lobe andschizophrenia.Brain
37. Gerner RH, van Kammen DP, Ninan PT: Cerebrospinal fluid, cho-
lecystokinin, bombesin and somatostatin in schizophrenia and normals.
Prog Neuropsychopharmacol BiolPsychiatry 1984;9:73-82.
38. Beal MF, Martin JB: Effects ofneuroleptic drugson brain somato-
statin-likeimmunoreactivity. Neurosci Lett1984;47:125-130.
39. WalshTJ, EmerichDF, WinokurA, Banki C, BissetteG, Nemeroff
CB: Intrahippocampal injection ofcysteamine depletes somatostatin and
produces cognitive impairmentsin the rat. Soc Neurosci Abstr1985;11:621.
Crow TJ, Johnstone EC,
at University of Miami School of Medicine, on March 8, 2011 www.archgenpsychiatry.com Downloaded from