SIR - We appreciate the comments from Leegwater and colleagues regarding our recently published article about the possibility that the L309M variant of the KIAA0027 (WKL1, MLC1) gene may cause catatonic schizophrenia (SCZD10, MIM 605419).1 We are aware of their report on the involvement of KIAA0027 in the pathogenesis of megalencephalic leukoencephalopathy with subcortical cysts (MLC, MIM 604004)2 and we agree with Leegwater and coworkers that the hypothesis of an involvement of KIAA0027 in schizophrenia requires confirmation in other affected families or patients. A patient with catatonic schizophrenia who is likely to be homo- or hemizygous for six adjacent rare genetic markers from the chromosome 22q13.33 region is currently being investigated for the presence of a deletion in the 3' region of KIAA0027.3 However, we cannot rule out that a mutation in a gene different from KIAA0027 in the 4 cM telomeric region of chromosome 22 causes catatonic schizophrenia. The neuronal cadherin gene CELSR1 which is exclusively expressed in brain has meanwhile been excluded from the candidate gene panel by mutational analysis. Investigation of other candidate genes from this chromosomal region should help to confirm or exclude a possible role of KIAA0027 in the pathogenesis of chromosome 22-related catatonic schizophrenia.
Homozygous and compound heterozygous mutations of KIAA0027 cause recessively inherited MLC, thus supporting the notion that the KIAA0027 protein plays a critical role in myelination and in diseases of the white matter. This potential function of KIAA0027 is interesting in the light of the increasingly appreciated hypothesis that some forms of schizophrenia-like psychosis represent leukoencephalopathies or myelinopathies. Several lines of evidence indicate that a variety of neurodegenerative disorders with a genetic, vascular, or inflammatory etiology that lead to white matter dysfunction and disrupted cortical-striatal-limbic connectivity are more or less commonly associated with psychotic episodes. For instance, symptoms of acute psychosis including complex auditory hallucinations and bizarre delusions are frequently observed in a neurological disorder distantly related to MLC, metachromatic leukodystrophy (MLD, MIM 250100) (for review see Hyde et al4). MLD is caused by mutations of the arylsulfatase A gene (ARSA), which is coincidentally located in the vicinity of KIAA0027 on chromosome 22q13.33. A microdeletion of chromosome 22q11 that causes velocardiofacial syndrome (VCFS, MIM 192430) which is associated with craniofacial anomalies, cardiac defects, learning disabilities, and psychiatric syndromes including schizophrenia-like psychosis appears to be related to disturbances in myelination and axonal integrity.5 Another example for a combination of psychotic symptoms rapidly progressing to dementia and white matter pathology is polycystic lipomembranous osteodysplasia with sclerosing leukoencephalopathy (PLOSL; MIM 221770) which is caused by loss-of-function mutations in a tyrosine kinase binding protein (TYROBP). Moreover, late-onset paranoid psychosis is frequently produced by leukoencephalopathy of vascular origin, and schizophrenia has been reported in a patient with cerebral autosomally dominant arteriopathy with subcortical infarcts and leucoencephalopathy (CADASIL, MIM 125310).6 Particularly catatonic and bipolar syndromes may occur in multiple sclerosis.7 Finally, psychotic episodes have frequently been observed in the Andermann syndrome (ACCPN, MIM 218000), an autosomal recessive disorder linked to a gene on chromosome 15q13-15 that is associated with mental retardation, agenesis of the corpus callosum, and peripheral neuropathy.8
The hypothesis that some forms of schizophrenia are disorders of interhemispheric connectivity is further supported by imaging studies (for review see David9). Using diffusion tensor imaging (DTI), a modified technique of magnet resonance imaging (MRI), several investigators demonstrated disrupted corpus callosum and adjacent white matter organization in the brain of patients with schizophrenia, thus indicating alteration of fiber tracts and/or reduced myelination.10,11 Interestingly, Hakak and coworkers reported deviation of the expression pattern of several genes participating in the process of myelination using microarray expression profiling of postmortem brain from schizophrenic patients.12
We are pleased that Leegwater and associates share our view that the precise function of the KIAA0027 protein remains elusive. In silico structural analyses indicate that KIAA0027 is a transmembrane protein involved in ion and/or substrate transport. Variable phenotypes due to different mutations in KIAA0027 may represent pleiotropy of a single Mendelian gene, whereas the reduced penetrance of catatonic syndromes is likely to be produced by genetic modifiers elsewhere in the genome. Unlike the MLC-causing homozygous and compound heterozygous mutations of KIAA0027, which are potentially loss-of-function mutations, the L309M variant of KIAA0027 identified in catatonic schizophrenia is uniquely located in the last transmembrane domain which stabilizes the  -helix conformation of a highly conserved leucine string and may thus lead to a gain-of-function of KIAA0027. This notion is further supported by reports describing different phenotypes caused by gain-of-function mutations of genes encoding ion channels compared to loss-of-function mutations (for review see Lester and Karschin13). Interestingly, the involvement of genes encoding channel proteins in both psychosis of the schizophrenic spectrum and myelinopathies combined with motor impairment has recently been reported. Bowen and colleagues14 reported a frameshift mutation in the calcium-activated potassium channel gene KCNN3 in a schizophrenic patient, and inactivation of the inwardly rectifying potassium channel Kcnj10 causes hypomyelination, vacuolation, axonal swelling, motor impairment, and tremor in Kcnj10 knockout mice.15
Like Leegwater and colleagues, we feel that additional studies on the role of white matter diseases and myelinopathies in the etiopathogenesis of psychosis requires additional studies and that elucidation of KIAA0027 function by biochemical, electrophysio- logic, and transgenic strategies may shed light on its role in brain pathology and behavioral phenotypes.
J Meyer1, SJ Scherer2, R Mössner and KP Lesch
1Department of Psychiatry and Psychotherapy, University of Würzburg, Würzburg, Germany; 2Department of Physiologic Chemistry I, University of Würzburg, 97080 Würzburg, Germany
|
