50% in postmortem brain tissue obtained from patients diagnosed with schizophrenia. Interestingly, no significant changes are observed in GAD65 mRNA and immunoreactivity levels. We hypothesize that sequences within the reelin gene are responsible for the observed down-regulation of its’ mRNA and protein. We have been studying the promoter and 5' untranslated region (UTR) of the human reelin gene to explore determinants operative in modulating its expression. Within the 5' UTR, we have found a polymorphic CGG repeat which lies immedi- ately upstream of the ATG initiation codon. Seven different alleles have been identified thus far which vary in the length of the repeat unit. Transient transfection results obtained using genomic DNA containing 514 bp upstream of the reelin RNA start site are active in driving expression of a reporter construct following transfection into human neuroblastoma cells (approximately 50 fold over the reporter-less vec- tor). These sequences are also active in driving expression in human colon carcinoma cells, but at reduced levels in comparison to the neuronal cells (10 fold above the promoter-less construct). Additional data will be presented to provide an appropriate framework for under- standing the regulatory motifs operative in mediating transcriptional regulation of the reelin gene. These data may allow us to formulate hypotheses as to the mechanism by which reelin down regulation occurs in the brains of schizophrenia and bipolar disorder patients. 139. ASSESSING PREPULSE INHIBITION AND P50 SUPPRESSION IN A SINGLE TEST SESSION G.A. Light, D.L. Braff University of California, San Diego Schizophrenia has been conceptualized as a disorder with deficits in early sensory gating processes. Two distinct operational measures have been developed to assess gating functioning: prepulse inhibition (PPI) of the acoustic startle response and P50 event-related potential (ERP) suppres- sion. Both PPI of startle and P50 suppression are of interest in schizophrenia research as each measure identifies deficits in the time- linked information processing of rapidly presented stimuli. Deficient inhibition in the PPI and P50 paradigms are among the most consistent findings of information processing deficits in the schizophrenia literature. In the present study of 15 normal subjects, we sought to determine if PPI and P50 suppression could be assessed in a single paradigm and if variables obtained in this session yield equivalent values to those previously reported. Thirty 116dB pulse alone trials and 30 60 ms prepulse trials (15dB above background) were administered to subjects to assess PPI. In addition, 120 auditory click pairs (89dB) were also presented during this session for recording ERP suppression measures. Preliminary analyses indicate that PPI (mean = 62%), P50 suppression (mean = 90%), and N100 suppression (mean = 65%) measured in a single session appear to produce equivalent values to those obtained in previous reports of normal subjects. The relationships between PPI and P50 suppression will also be presented. Further studies are needed to determine if PPI and P50 suppression sampled in close temporal proximity (i.e. single-session) will be useful for assessing gating deficits in patients with schizophrenia. 140. LONG-TERM TREATMENT WITH QUETIAPINE IMPROVES COGNITIVE FUNCTION IN SCHIZOPHRENIA S.E. Purdon, A. Malla, A. Labelle, W. Litt University of Alberta, Alberta Hospital, Edmonton, Alberta, P5J 2J7, Canada The purpose of this study was to assess the efficacy of quetiapine in the reduction of cognitive impairment in schizophrenia. After a 48-hour washout period, 25 patients with schizophrenia were randomly assigned to double- blind treatment with quetiapine or haloperidol for 6 months. Patients were evaluated after 8 weeks and 6 months using ratings scales for psychotic symptoms, mood, and extrapyramidal symptoms (EPS). Standardised neu- ropsychological measures of the following six cognitive domains were also used: fine motor skills, attention span, verbal reasoning and fluency, visuospatial construction and fluency, executive skills and visuomotor tracking, and immediate recall of verbal and non-verbal materials. The mean dosages of quetiapine and haloperidol were 468 mg/day and 16 mg/day, respectively. Within 8 weeks, quetiapine improved psychosis and mood without inducing EPS. Quetiapine also enhanced cognitive skills, particu- larly verbal reasoning and fluency skills and immediate recall, and with long-term treatment also improved executive skills, visuomotor tracking, and the average of the six cognitive domains. While haloperidol improved the general clinical status at 8 weeks and 6 months, there were no specific improvements in the positive syndrome, the negative syndrome, depression ratings, mood or cognitive skills. These preliminary results illustrate the potential value of quetiapine in the improvement of cognitive impairment in schizophrenia and establish the importance of further research with this promising atypical antipsychotic. 141. LONG-TERM QUETIAPINE FOR PSYCHOSIS IN PARKINSON’S DISEASE IN PLACE OF OTHER ANTIPSYCHOTICS J.L. Juncos (1), M.L. Evatt (1), R.D. Jewart (1), V.J. Roberts (1), C.D. Wood (1), L.S. Potter (2), H.-C. Jou (2), P.P. Yeung (2) (1) Emory University School of Medicine, Atlanta, GA, 30329-5102 (2) AstraZeneca, Wilmington, DE, 19850-5437 Treating psychosis in patients with Parkinson’s disease (PD) is extremely challenging because conventional antipsychotics commonly worsen mo- tor dysfunction, while atypical antipsychotics, such as clozapine, risperi- done, and olanzapine, may be associated with anticholinergic, extrapy- ramidal (EPS), or hematologic side effects. Quetiapine (‘SEROQUEL’) has been shown to be effective and well tolerated in patients with PD and psychosis. A single-center, long-term (24 week), open-label trial evalu- ated the safety, tolerability, and efficacy of quetiapine for treating psychosis in patients with parkinsonism who failed treatment with clozapine, risperidone or olanzapine. Failure could be due to lack of efficacy, intolerable side effects, or lack of compliance with blood monitoring. Twenty-nine patients (mean age 73 years) with parkinsonian and psychotic symptoms received up to 400 mg/day of quetiapine, dosed according to clinical response and tolerability. Assessments included the Brief Psychiatric Rating Scale (BPRS), Clinical Global Impression (CGI) Severity of Illness, Neuropsychiatric Inventory (NPI), and Unified Parkinson’s Disease Rating Scale (UPDRS). The mean BPRS total score improved by 30% and remained significantly improved throughout the 24-week trial (p = 0.004). The mean NPI psychosis subscale (delusions, hallucinations and agitation/aggression) score had 50% improvement (p = 0.028). The mean UPDRS total and motor subscale scores did not 42S Thursday Abstracts BIOL PSYCHIATRY 2000;47:1S–173S