A form of DISC1 enriched in nucleus: altered subcellular distribution in orbitofrontal cortex in psychosis and substance/alcohol abuse

Abstract

Disrupted-In-Schizophrenia 1 (DISC1) was identified as the sole gene whose ORF is truncated and cosegregates with major mental illnesses in a Scottish family. DISC1 has also been suggested, by association and linkage studies, to be a susceptibility gene for schizophrenia (SZ) in independent populations. However, no analysis of DISC1 protein in human brains, especially those of patients with SZ, has yet been conducted. Here we performed a biochemical analysis of DISC1 protein in a well characterized set of autopsied brains, including brains of patients with SZ, bipolar disorder, and major depression (MD), as well as normal control brains. We identified an isoform of DISC1 by using MS and demonstrated that it is enriched in the nucleus of HeLa cells. In the orbitofrontal cortex, the subcellular distribution of this DISC1 isoform, assessed by the nuclear to cytoplasmic ratio in the immunoreactivity of the isoform, is significantly changed in brains from patients with SZ and MD. This altered distribution is also observed in those subjects with substance and alcohol abuse. The changes in MD brains are significantly influenced by substance/alcohol abuse as well as postmortem interval; however, the alteration in SZ brains is free from brain-associated confounding factors, although an interaction with substance/alcohol abuse cannot be completely ruled out. These results suggest that DISC1 may be implicated in psychiatric conditions in other populations than the unique Scottish family.

Fig. 1.

Characterization of human DISC1 protein. Western blotting of human autopsied brains, HeLa cells, and SH-SY5Y cells with an anti-DISC1 antibody (C2) is shown. DISC1 is expressed in two distinct bands at 95–100 kDa and 70–85 kDa in human brains. The signal from 70 to 85 kDa is subdivided into three categories: a major band in the middle at 75–85 kDa occurs in all of the samples with an additional upper signal in HeLa cells and a minor lower signal in HeLa cells and human brains. Similar results were obtained with antibodies from K. Millar and T. Katayama (N.S. and A.S., unpublished data).

Fig. 2.

A pool of DISC1 in the nucleus. (a) DISC1 in the crude nucleus obtained by a classic method. In human autopsied brains, the DISC1 signal at 95–100 kDa is exclusively enriched in the S, but the signal at 75–85 kDa occurs in both the P and S fractions. (b) DISC1 in a fraction enriched with nuclear transcription factors. In HeLa cells and SH-SY5Y cells, the major band at 75–85 kDa focused on in the present study is distributed in both the nuclear fraction (N) and the cytosolic fraction (C). α-Tubulin and polyADP polymerase are used for cytosolic and nuclear markers, respectively.

Fig. 3.

Immunoreactivity of DISC1 at 75–85 kDa in the total tissue extracts. (a) No change in the level of DISC1 in the total homogenates among brains from patients with SZ, BP, and MD as well as NORM brains. Relative ratios to the intensity of the band from one control sample were plotted. Each symbol represents the data from individual samples. Means ± SD are shown. (b) The immunoreactivity of GAPDH, a constant expression marker, shows no difference in the levels among four groups.

Fig. 4.

Aberrant subcellular distribution of DISC1 in patient brains with SZ and MD. (a) Representative Western blot of DISC1 in the P and S fractions. Subcellular distribution of DISC1 in human brains was analyzed, which was further characterized in Figs. 2a and 5. (b) The signal ratio of DISC1 from the P to the S fraction (the P:S ratio) is increased in patient brains with SZ and MD groups (*, P = 0.018; and **, P = 0.0091, respectively). Each symbol represents the data from individual samples. Means ± SD are shown.