Despite extensive effort for many years, the etiology of major psychiatric diseases remains unknown. A recent study by Baysal et al. has argued against the ALG9 gene variants in causing psychosis. Due to its disruption by a balanced t(9p24;11q23) translocation that segregates with the disorder in a family, it was proposed to be a primary candidate gene causing psychosis. In addition, a recent review article by Pickard et al., entitled "Cytogenetics and gene discovery in psychiatric disorders," highlighted the importance of studies of chromosome rearrangements in finding disease-causing mutations. However, achieving the goal of finding genes by conventional association studies and by investigating chromosome rearrangements remains elusive. Here we discuss a fundamentally different explanation from the usual one considered by workers in the field concerning chromosome aberrations and psychoses etiology. We hypothesize how chromosome aberrations might cause disease but the gene at the rearrangement breakpoint is irrelevant for the etiology. Moreover, we discuss subsequently published findings that help scrutinize validity of the two very different hypotheses considered in the psychiatric genetics field. In sum, we alert the readers to the complexities of interpreting phenotypes associated with rearrangements.

---

BOTH the recent study (BAYSAL et al. 2006) and a recent review (PICKARD et al. 2005) of the field clearly point out that the gene mutations causing major psychiatric schizophrenia and bipolar diseases have not been identified. Numerous chromosome regions were first identified by linkage analysis but all fell by the wayside as none of them has been definitively implicated in subsequent studies. Lack of replication of studies has been the norm rather than an exception in the field. Therefore, the above quoted studies take a different tactic. PICKARD et al. (2005) searched literature for an association of chromosomal rearrangements with psychoses as a way to identify disease-conferring mutations. This search identified several rearrangements. Often, rearrangements have caused gene-interrupting mutations and therefore are hypothesized to cause psychoses by gene haplo-insufficiency. Although it is logical to consider it as an explanation of genetic etiology, experimental support for the haplo-insufficiency hypothesis has been lacking. It should be noted that mutations caused by chromosome rearrangements by this hypothesis are considered to be dominant. Should this supposition be correct, their identification should have been accomplished years ago because dominant mutations are relatively easy to map. Nonetheless, fueled by lack of any other success in the field, stock of genes located at the chromosome breakpoint goes up immediately. Such genes are considered to be candidate genes for psychoses and are justifiably subjected to verification with subsequent linkage-analysis studies. However, none of those genes has panned out when disease inheritance in large families in the general population is investigated by whole-genome linkage analysis. The latest of such studies quoted above (BAYSAL et al. 2006) has also argued against the Disrupted in bipolar disorder 1 locus, initially so named because of its partial association with disease and not because it has been demonstrated to have caused the disorder. This locus was originally defined by the balanced translocation t(9p24;11q23); it encodes the mannosyltransferase-encoding ALG9 gene. This latest study concerning this translocation showed that the sequence variations in the ALG9 gene in the general population are not associated with predisposition to psychosis (BAYSAL et al. 2006). Invariably such studies hypothesize that breakpoints might perturb local gene expression at the region by long-range "position effects" and also propose to stay the course to perform further linkage analysis on much larger numbers of diseased families selected at large to increase chances of finding disease-causing mutations. Overall, the logical strategy, which had been very successful in identifying disease-causing oncogenes associated with chromosome rearrangements, has failed to yield success in the field of psychiatric genetics. Does another hypothesis exist to explain association of chromosome rearrangements with psychosis? Moreover, it remains unexplained as to why only one-half of heterozygous translocation carriers are diseased (see below).

Here we discuss an explanation different from the one proposed by workers in the psychiatric genetics field that specifically concerns chromosome rearrangements. Because recent studies (KLAR 2002, 2004a,b) that advanced an unconventional genetic explanation for the etiology of psychoses concerning chromosome rearrangements were not discussed in the recent articles referenced above, their discussion is presented here (see also KLAR 2003; MILLAR et al. 2003). One of those studies has also reported a similar review of the literature concerning association of chromosomal rearrangements with psychoses (KLAR 2004a), and it had proposed an explanation very different from that of the haplo-insufficiency hypothesis advanced by all authors who first described chromosome aberrations.

These studies proposed that while the rearrangements provide the best evidence supporting genetic etiology for psychosis in certain families, they only help identify the relevant chromosome, and do not necessarily implicate gene(s) at the breakpoint, if indeed they were disrupted by the breakpoint. According to the recently proposed somatic strand-specific imprinting and selective chromatid segregation (SSIS) hypothesis (KLAR 2002, 2004a), the disease is caused by anomalies of brain laterality development because of chromosome rearrangements that affect normally biased chromatid segregation in cell division, but without a mutation of any gene (see Figure 1). It is hypothesized that sister chromatids inherently differ because one inherits "older" "Watson" (W in Figure 1) and "newer" "Crick" (C) strand and the sister chromatid inherits newer W and older C strand. A somatically installed epigenetic modification of the relevant gene for brain hemispheric laterality development "differentiates" sister chromatids of a chromosome 11 pair during replication at a specific cell division when brain hemispheric laterality is initially determined in the embryo. This is followed by selective segregation of differentiated sister chromatids to daughter cells. Such a specialized mitosis in brain hemisphere-generating progenitor cell will produce developmentally nonequivalent daughter cells that will subsequently generate nonequivalent brain hemispheres. That is, an asymmetric cell division is proposed to be the basis of brain laterality development and, accordingly, psychosis is proposed to result from anomalous hemispheric development. In short, this hypothesis highlights the importance of chromosomally borne somatically installed epigenetic mechanisms to accomplish cellular differentiation required for development in general. This hypothesis was first proposed as a mechanism to explain left-right body axis development of vertebrates (KLAR 1994).

View larger version (30K): In this window In a new window Download PPT slide   FIGURE 1.— The SSIS hypothesis. The hypothesis is proposed as a mechanism for brain hemispheric laterality development by chromosome 11 chromatids differentiation and their selective segregation to affect an asymmetric cell division. The hypothetical brain-laterality determining gene is expressed (ON "epiallele") in the chromatid that inherits "older" arbitrarily named "Watson" (W) strand, while it remains epigenetically silenced (OFF epiallele) in the sister chromatid inheriting the older "Crick" (C) strand in a brain laterality-generating progenitor cell during embryogenesis. A hypothetical segregator factor functions through the centromere for executing selective chromatid segregation. Consequently, both older W-containing chromatids from the progenitor cell are segregated to a daughter cell located on the left side of the embryo, while those with older C-containing strands are delivered to the other daughter cell. This results in production of nonequivalent sister cells that ultimately lead to the development of nonequivalent brain hemispheres in individuals containing both wild-type chromosome 11 copies. The hypothesis suggests that sister chromatids of chromosome 11 are segregated in a biased mode, while other partner chromosomes in the translocations are segregated in a random fashion. Consequently, chromatids containing the chromosome 11 epialleles in the rearranged chromosome will be randomly segregated to daughter cells because of their attachment to the centromere of a chromosome that follows a random mode of segregation. Thus, one-half of the translocation heterozygote individuals will develop lateralized healthy brains and the other half should be diseased. That is, the model predicts that heterozygous chromosome 11 balanced translocations should cause disease in one-half of carriers. For simplicity, the DNA chains are drawn as strait lines. In daughter cells, the older W strands used as DNA replication templates in the progenitor cell are in green, the older C in red, the younger C in brown, and the younger W in blue. The color-coded arrows reflect segregation of color-matched-strand-containing chromatids to the indicated daughter cell.

 
As most rearrangements involve only one to three cases of psychosis (KLAR 2002, 2004a; PICKARD et al. 2005), such cases may be incidental and unrelated to chromosome rearrangements. However, three translocations affecting numerous individuals exist with junction regions covering 40% of the long arm of chromosome 11 and that are coupled with portions of chromosome 1, 6, or 9 in different families. Chromosome 11 chromatids are proposed in mitosis to follow the selective segregation mode because of attachment to their centromeres, but in theory, the centromeres of chromosomes 1, 6, and 9 follow the random distribution mode. Consequently, normally selective segregation of epigenetic states of the brain laterality-establishing gene located distal to chromosome 11 breakpoints (see Figure 1) should be randomized by the rearrangement. This will lead to generation of equivalent sister cells causing disease because the brain hemispheres' laterality development will be compromised. Most strikingly, each chromosome 11 translocation is associated with psychosis but only 50% of heterozygous translocation carriers are diseased. The model precisely predicts this outcome should the rearranged chromosome with centromere of chromosome 1, 6, or 9 segregate by the random chromatid segregation mode, while the wild-type chromosome 11 copies follow the selective segregation mode (KLAR 2004a). It is therefore predicted that only one-half of translocation carriers in the heterozygous configuration should be diseased resulting from symmetric cell division (Figure 1) in them. Accordingly, it will be interesting to compare brain laterality of diseased with healthy translocation-carrying individuals in these families. In sum, the disorder is caused by genetics in translocation-carrying families, not by conventional mutations in genes, but by disruption of selective segregation of chromosome 11 "epialleles" that control brain laterality development.

Similar arguments were advanced previously (KLAR 2004a) to explain t(1q42;11q14) and t(6q14;11q25) translocations. Consistent with this hypothesis, no gene is interrupted by both junction regions in the t(6q14;11q25) translocation (JEFFRIES et al. 2003). Also, common variations in the ALG9 gene that is disrupted in the t(9p24;11q23) translocation are not associated with the disorder, a very informative result that is presented in the newest study that concerns this specific translocation (BAYSAL et al. 2006). Their article is the latest one that notes the lack of evidence for breakpoint lesions being responsible for psychiatric disorders. We therefore propose that the SSIS hypothesis should be entertained as an explanation for this translocation as well. Because most cases of psychoses are sporadic, their etiology remains unknown. By extension of the strand-segregation hypothesis, it is possible that low levels of spontaneous aberrant generation and/or epiallele distribution in the brain laterality-generating progenitor cell might cause disease in general cases. However, our discussion here is limited to explain only cases involving chromosome rearrangements. In general, the literature concludes in favor of genetic etiology because of psychoses's prevalence in some families. This conclusion is inherently flawed due to investigators selection bias because those families were chosen in the first place because of high incidence of disease in them. Therefore, translocations associated with psychoses in families discussed above provide the best support for the genetic etiology; our hypothesis provides an alternative view for how they might cause disease.

The process of strand-specific imprinting proposed in the model as a mechanism for cellular differentiation is demonstrated only in fission yeast (KLAR 1987, 1990, 2007) and we recently suggested the existence of selective chromatid segregation phenomenon in mouse cells (ARMAKOLAS and KLAR 2006, 2007). Although discovery of these unrelated biological phenomena, one in yeast and the other in mouse cells, do not directly bear on the mechanism of brain laterality development in humans, these findings support the evolution of two novel features of the model proposed for brain laterality development (see Figure 1). As the conventional paradigm of focusing on finding a disease-causing gene mutation has failed for a long time, and realizing that the lack of success is not for lack of trying, discussion of alternative hypotheses is warranted. In this regard, we agree with the recent summary of the American Society of Human Genetics conference that highlighted the lack of progress in this area where the editorial notes "Having witnessed a lack of consensus in the formal review of genetics association studies submitted to this journal over the last 2-3 years ..." (ANONYMOUS 2006). Lacking success in the field by approaches limited to scrutinizing conventional models, and to help design future studies, the SSIS hypothesis should be considered as an alternative mechanism to explain chromosome rearrangements that confer psychoses. The observations of 50% disease prevalence and the breakpoint region not segregating with the disease have verified two key predictions of the SSIS hypothesis. Clearly further analysis of disease-causing chromosome aberrations is urgently needed to determine the genetic basis of these debilitating disorders. In short, we encourage the readers and workers in the field of psychiatric genetics to be circumspect when interpreting disease models that assume the genes underlying a disease to be located at the rearrangement breakpoint.

The Intramural Research Program of the National Institutes of Health, National Cancer Institute, Center for Cancer Research, supports research in our laboratory.

ANONYMOUS, 2006 Human genetics in the Crescent City. Nat. Genet. 38: 1221.[CrossRef][Medline]

ARMAKOLAS, A., and A. J. S. KLAR, 2006 Cell type regulates selective segregation of mouse chromosome 7 DNA strands in mitosis. Science 311: 1146–1149.[Abstract/Free Full Text]

ARMAKOLAS, A., and A. J. S. KLAR, 2007 Left-right dynein motor implicated in selective chromatid segregation in mouse cells. Science 315: 100–101.[Abstract/Free Full Text]

BAYSAL, B. E., J. E. WILLETT-BROZICK, S. A. BACANU, S. DETERA-WADLEIGH and V. L. NIMGAONKAR, 2006 Common variations in ALG9 are not associated with bipolar I disorder: a family-based study. Behav. Brain Funct. 2: 25–31.[CrossRef][Medline]

JEFFRIES, A. R., A. J. MUNGALL, E. DAWSON, K. HALLS, C. F. LANGFORD et al., 2003 beta-1,3-Glucuronyltransferase-1 gene implicated as a candidate for a schizophrenia-like psychosis through molecular analysis of a balanced translocation. Mol. Psychiatry 8: 654–663.[CrossRef][Medline]

KLAR, A. J. S., 1987 Differentiated parental DNA strands confer developmental asymmetry on daughter cells in fission yeast. Nature 326: 466–470.[CrossRef][Medline]

KLAR, A. J. S., 1990 The developmental fate of fission yeast cells is determined by the pattern of inheritance of parental and grandparental DNA strands. EMBO J. 9: 1407–1415.[Medline]

KLAR, A. J. S., 1994 A model for specification of the left-right axis in vertebrates. Trends Genet. 10: 392–396.[CrossRef][Medline]

KLAR, A. J. S., 2002 The chromosome 1;11 translocation provides the best evidence supporting genetic etiology for schizophrenia and bipolar affective disorders. Genetics 160: 1745–1747.[Abstract/Free Full Text]

KLAR, A. J. S., 2003 Response to Millar et al.: critique of chromosome 1;11 translocation causing psychosis. Genetics 163: 837–838.[Free Full Text]

KLAR, A. J. S., 2004a A genetic mechanism implicates chromosome 11 in schizophrenia and bipolar diseases. Genetics 167: 1833–1840.[Abstract/Free Full Text]

KLAR, A. J. S., 2004b An epigenetic hypothesis for human brain laterality, handedness and psychosis development. Cold Spring Harbor Symp. Quant. Biol. 69: 499–506.[CrossRef][Medline]

KLAR, A. J. S., 2007 Lessons learned from studies of fission yeast mating-type switching and silencing. Ann. Rev. Genet. 41: 213–236.[Medline]

MILLAR, J. K., P. A. THOMSON, N. R. WRAY, W. J. MUIR, D. H. BLACKWOOD et al., 2003 Response to Amar J. Klar: the chromosome 1;11 translocation provides the best evidence supporting genetic etiology for schizophrenia and bipolar affective disorders. Genetics 163: 833–835.[Free Full Text]

PICKARD, B. S., J. K. MILLAR, D. J. PORTEOUS, W. J. MUIR and D. H. BLACKWOOD, 2005 Cytogenetics and gene discovery in psychiatric disorders. Pharmacogenomics J. 5: 81–88.[CrossRef][Medline]

Communicating editor: J. A. BIRCHLER

International Access Link

Online ISS 1091-6490
Copyright 2008 by the Genetics Society of America
phone: 412-268-1812   fax: 412-268-1813   email: genetics-gsa{at}andrew.cmu.edu