17 August 2006. The case for subtyping schizophrenia by genetic study receives support in a paper published in the August issue of the Archives of General Psychiatry. Hugh Gurling of University College London and his multi-institution team found that variations in the gene for pericentriolar material 1 (PCM1) on chromosome 8 are associated with risk of schizophrenia. The researchers also provide imaging evidence that having these schizophrenia-associated variations in PCM1 leads to a different pattern of brain abnormality than is seen in other cases of schizophrenia.
Hunting for genes of small effect
Linkage studies of schizophrenia have pointed to a number of different loci on different chromosomes as being involved in the disorder, and meta-analyses suggest that the inconsistencies reflect heterogeneity of susceptibility genes acting in different individuals and families (see, e.g., Lewis et al., 2003). Similarly, association studies have not consistently replicated the involvement of individual genes. The conclusion geneticists have been able to draw from these findings is that there are going to be several different modes of transmission for different subtypes of schizophrenia.
Chromosome 8 seems to be a particularly good hunting ground for genes involved in schizophrenia. Several regions of the chromosome have been implicated (see, most recently, Walss-Bass et al., 2006), and neuregulin 1 (NRG1), one of the genes of most interest (see SRF related news story), is located in the same region of the chromosome as PCM1 (8p21-22). Also in this region are PPP3CC (calcineurin), DRP2 (dihydropyrimidinase-related protein 2), and FZD3 (frizzled), all of which have been linked to increased likelihood of having schizophrenia.
In their current study, Gurling’s team first revisited their earlier genomewide scan of 13 families with multiple cases of schizophrenia, which had found linkage to 8p21-22 (Gurling et al., 2001). Reanalyzing the data, the researchers found that variation at the microsatellite marker D8S261, located within the PCM1 gene, was associated with schizophrenia. They replicated the association with D8S261 in a case-control sample of 450 patients and 450 controls from London and surroundings, along with statistically significant association between schizophrenia and an additional microsatellite marker and several single nucleotide polymorphism (SNP) markers within the gene. Furthermore, several haplotypes consisting of three or four of these markers showed strongly significant associations with schizophrenia.
Interestingly, the researchers also found an association between the NRG1 gene and schizophrenia in this sample, but the evidence was not as strong as that seen for PCM1. They also report that they found no evidence for susceptibility from the PPP3CC or FZD3 genes.
The researchers went on to confirm the association of D8S261 variation with schizophrenia in a separate U.S. population of 100 schizophrenia and schizoaffective disorder patients and their parents, but they failed to replicate the results in a Scottish case-control sample of 200 cases and 200 controls.
Imaging “PCM1 schizophrenia”
Just what is this PCM1, and how might variations in the gene contribute to schizophrenia? The protein product is a player at the centrosome, the organizing hub of microtubules, meaning that it can assist in, or disrupt, a range of cellular processes from cell division and migration to axonal transport to synaptic function. PCM1 has been known to associate with other molecules that have been investigated in schizophrenia, including Nudel and Lis1 (see Guo et al., 2006), which of course makes it tempting to speculate that it might have ties to DISC1 (see SRF related news story).
In the second phase of the study, Gurling and colleagues chose to look for gross anatomical brain effects of PCM1 genetic variation, using voxel morphometry MRI. They selected two genetically distinct schizophrenia subgroups from the London case-control study for structural imaging: one group whose PCM1 genes harbored two of the schizophrenia-associated microsatellite or SNP marker variants (termed the SZ8 group; n = 14) and another group that displayed none of the marker alleles associated with schizophrenia (SZ0; n = 13).
Consistent with many previous studies, both patient groups showed significant reductions in gray matter compared to controls, both in absolute terms and relative to white matter, but no differences in white matter volume. The two schizophrenia groups also did not differ from each other on these global measures. However, there were clear regional differences. The SZ8 group had significantly lower gray matter volume, relative to controls, in the orbitofrontal cortex, bilaterally. By contrast, the SZ0 group did not differ from controls in this region, but showed lower gray matter volumes in a number of other brain areas.
When the two schizophrenia groups were compared directly, medial orbitofrontal cortex had significantly lower volume in SZ8 versus SZ0 patients. Conversely, left hippocampus, right dorsolateral prefrontal cortex, and the temporal poles of both hemispheres, were reduced in volume in SZ0 versus SZ8 patients.
The pattern of morphologic differences led the authors to speculate that it might be possible to distinguish PCM1-associated schizophrenia from the remainder of patients diagnosed with schizophrenia. For example, orbitofrontal cortex is known to be critical for reward-related or motivational processing. Deficits in these domains might contribute to the constellation of deficits (including greater affective and cognitive deficits, along with poorer outcome) found with greater frequency in families with linkage to 8p21-22, compared to families with linkage to other chromosomal loci (Kendler et al., 2000). But, as the authors point out, the association of PCM1 with schizophrenia needs confirmation in other samples.—Hakon Heimer.
Gurling HM, Critchley H, Datta SR, McQuillin A, Blaveri E, Thirumalai S, Pimm J, Krasucki R, Kalsi G, Quested D, Lawrence J, Bass N, Choudhury K, Puri V, O'Daly O, Curtis D, Blackwood D, Muir W, Malhotra AK, Buchanan RW, Good CD, Frackowiak RS, Dolan RJ. Genetic association and brain morphology studies and the chromosome 8p22 pericentriolar material 1 (PCM1) gene in susceptibility to schizophrenia. Arch Gen Psychiatry. 2006 Aug;63(8):844-54. Abstract