CENTRAL NERVOUS SYSTEM AND SYSTEMIC QUINOLINIC ACID IN SCHIZOPHRENIA
LD Bobo*, RH Yolken, EF Torrey, JD Paltan-Ortiz, MM Herman, NC Briggs,
H Velasco, M Zito, MP Heyes
Stanley Research Foundation, NAMI, Washington DC
Immunologic marker elevations in schizophrenia have been reported. Quinolinic acid (QUIN), a potential neurotoxic product of immune-activated macrophages and microglia, has been used as a surrogate marker of inflammation. The involvement of QUIN in psychosis was explored in a postmortem study of brain, CSF, serum and liver QUIN levels compared with DSM IV diagnosis of either schizophrenia, bipolar disorder, unipolar depression without psychosis and unaffected controls. Exclusive criteria included those with confirmed bacterial or viral infection at the time of death, autoimmune disorders, or who are taking steroids. CSF and serum albumin were measured by nephelometry, and a CSF:serum ratio>0.01 was used as a cut-off to indicate moderate-marked barrier permeability. Hemaglobin was determined colorimetrically. QUIN was measured by negative ion capture mass spectroscopy-HPLC in nm/1 for CSF and sera, and in pm/g wet weight of issue. Brain pH was also determined in the cerebellum and frontal cortex. Parafin-embedded sections of frontal cortex and cerebellum were evaluated for gliosis, mononuclear cell infiltrates, neuronal inclusions, neuritic plaques and tangles, and cerebellum was examined for Purkinje cell axonal swellings. The effect of pre- and postmortem characteristics on levels were also evaluated. Cerebellar QUIN levels differed between disease categories (Kruskal-Wallis, p=0.038), and increased cerebellar QUIN levels were elevated in schizophrenia compared to unaffected and unipolar depression (p<0.001 and p=0.03, Fisher’s LSD). In the schizophrenia group 41.2% (7/17) of values were >2 SD from mean of unaffected subjects. QUIN levels were not elevated in the frontal cortex, serum or liver in association with a DSM IV group. CSF QUIN was elevated in schizophrenia and bipolar groups compared to unaffected controls (p=0.03 and p=0.01, respectively). CSF hemoglobin levels were not related to CSF QUIN levels. CSF and cerebellar QUIN levels were related to elevated blood brain barrier permeability. Increased gliosis and mononuclear cell infiltrates were not noted, but cerebellar torpedo levels were directly related to cerebellar QUIN levels in schizophrenia (R=0.30, n=71, p=0.008). Cerebellum pH was significantly related with DSM IV in association with cerebellar QUIN (ANCOVA, n=76, F=6.26, df=3, p<0.001). In the schizophrenia group, a significant inverse relationship was seen between cerebellar QUIN and pH. We conclude that increases in cerebellar and CSF QUIN levels are found in a subgroup of patients with psychosis. In order to assess whether other immuno-inflammatory mediators are elevated, cytokine and MHC Class II levels are being measured in the cerebellum.