John M. Davis, Eminio Costa,

Alesandro Guidotti, M. Dneil Smalheiser, Dennis Grayson, Patricia Tuiting,

Cristeen Pesold and Yogesh Dwivedi

University of Illinois at Chicago

Costa and his coworkers studying

postmortem brains, principally of the Stanley Foundation collection, find that

schizophrenia and bipolar patients display a 50% decrease in Reelin protein, in

reelin mRNA, in the number of reelin stained neurons, in GAD67 protein and in

GAD67 mRNA in comparison to normal controls or non-psychotically depressed

patients. The number of neurons, the level of neuron specific enolase, a marker

for neurons, the levels of GAD67 protein and mRNA are normal.  Furthermore

these changes are not related to lifetime neuroleptic dose, postmortem interval

etc. and if an artifact accounted for this finding, then it is an artifact

common to three brain banks and common to three radically different

assays.  Costa’s laboratory has breed reelin mice (which have no reelin)

with normal mice to produce reelin heterozygous reeler mice.  These mice

are grossly normal, but demonstrate a 50% decrease in reelin, in GAD67 and in

the number of dendritic spines plus an increase in neuro-packing density and a

decrease in cortical thickness due to a decrease in neuropil.  The

heterozygous mice histology is remarkably similar to that of

schizophrenia.  We hypothesize that a reelin defect, may produce a

reduction in synaptic spines which produces the decreased grey matter and

increased ventricular size observed in schizophrenia and bipolar disease. 

We find that heterozygous GAD67 knockout mice have a 50% decrease in GAD67 mRNA

expression but normal reelin had no neuropil hypoplasia or decrease in dendritic

spine expression.  Enlarged ventricles are associated with the following

cluster of measures: negative symptoms, cognitive impairment (e.g. working

memory, etc.), abnormal activation (blood flow, or glucose utilization) in the

prefrontal cortex (PFC) by cognitive stimulation and decreased NMR spectroscopic

markers (such as NAA signals) for neural tissues, the Enlarged Ventricles-PFC

Integrity cluster.  It is possible that one could diagnose a Reelin

deficiency at birth (or before) and develop a drug that would alter the

regulation of reelin in such a manner that it would stimulate Reelin

function.  Many workers have found GABAergic abnormalities in postmortem

brains from schizophrenics.  We will review what is known of the effects of

GABAergic drugs (GABAa and GABAb) on schizophrenia.  While benzodiazepines

have not long-term beneficial effects on schizophrenia, brief administration

(several days) does have a substantial beneficial effect in

schizophrenics.  It is possible that schizophrenics rapidly develop

tolerance to this effect.  The possibility that benzodiazepine partial

agonists, for which tolerance does not develop, might be effective

antipsychotics will be discussed.