Search

The method of DNA restriction fragment analysis using gene probes for the proopiomelanocortin (POMC) gene was employed to detect possible molecular variation in the POMC gene in schizophrenia and bipolar illness. No gross structural abnormalities in restriction fragments were observed with the set of restriction enzymes used. Two allelic restriction sites were observed giving rise to fragment length polymorphisms. One of these is a new polymorphism, not previously reported, which will be of value as a linkage marker. The associations between the two DNA polymorphisms that are closely linked to the POMC gene and both schizophrenia and bipolar disorder were investigated. No association was found, thus adding weight to the evidence that there are no alterations in the POMC gene in schizophrenia and bipolar illness.

It is argued that further research to achieve more detailed diagnostic systems in many psychiatric disorders is unlikely to be productive without taking genetic effects into account. Even when this is done, for example when carrying out segregation analysis to determine a mode of genetic transmission, mental illnesses often pose specific problems that preclude accurate analysis. Because techniques in molecular biology and genetics have made it possible to study gene effects in human disease systematically it should now be possible to specify the genes that are involved. When this has been achieved then a diagnostic system based on genetic causation can develop. This will have the advantage of helping to pinpoint environmental factors more accurately. Specific strategies will need to be adopted to overcome uncertain modes of inheritance, incomplete or non-penetrance of disease alleles and disease heterogeneity. Highly speculative hypotheses can be put forward for a locus causing Alzheimer's disease on a portion of the long arm of chromosome 21. For autism it is plausible that there is a disease locus at or near the fragile X site on the X chromosome. A locus for manic depression has been very tentatively mapped using DNA markers to chromosome 11 and in a small proportion of families DNA markers have also shown some evidence for X linkage. Schizophrenia does not seem to be associated with any favoured loci. Candidate genes for schizophrenia include those encoding dopamine, other neurotransmitter receptors or enzymes and various neuropeptides such as enkephalin and beta endorphin.

Three out of 4 adoption and 2 out of 3 twin studies imply a genetic contribution to alcoholism in males. A recent study of moderate drinking in male twins suggests that familial factors account for some two-thirds of the variance in alcohol consumption, and that the genetic component slightly outweighs the effects of the within family culture. There has been little support for the notion of a genetic entity which expresses itself as alcoholism in males and depressive illness in females, but among normal male twins anxiety and depression are genetically correlated with alcohol consumption. Increased liability to form acetaldehyde and the potentially additive tetrahydroisoquinolines could be one factor predisposing to alcoholism but human studies have been less convincing than their animal counterparts.

Hypervariable regions of DNA exist at many discrete loci in the human genome. Variations in the length of specific classes of hypervariable DNA termed "minisatellite" sequences may be detected by hybridisation of radioactive probes composed of a common 10-15 base pair repeat or "core" sequence which is shared by each of the minisatellite loci. Such a procedure was used to perform linkage analysis on a kindred in which a disease allele causing manic depression appeared to be segregating as an autosomal dominant. Results indicate no apparent linkage between a manic depression allele and at least 20 hypervariable genetic loci which acted as markers. However, because of the hypervariability of the minisatellite loci these results do not preclude similar analyses of other manic depressive pedigrees being productive because each such family will have distinct observable minisatellite alleles at loci specific to that family. The method could be usefully applied to the genetic analysis of other psychiatric disorders where heterogeneity is suspected and where a single autosomal major gene locus appears to be responsible for causing multiple cases within a single kindred.

A study of 74 twin pairs with alcoholic probands from the Maudsley Hospital is reported. Pairwise concordance for alcoholism as categorised by the SADS-L Research Diagnostic Criteria is similar in MZ twins (29%, 8/22) and DZ twins (33%, 13/39). Concordance for all diagnoses other than alcoholism is however significantly greater in MZ twins (48%, 13/27) than DZ twins (21%, 8/39, P less than 0.001). Concordance in MZ and DZ twins for all diagnoses other than alcoholism, including depression, did not coincide with concordance for alcoholism. Data on criminality revealed that 21% (32/148) of the twin sample had non-alcohol, non-traffic offences on record at the UK Home Office. Of the 32 with criminal records, 28 were alcoholic probands and pairwise concordance for criminality was found in only 1 MZ and 2 DZ pairs.

21 pairs of monozygotic twins discordant for alcoholism and heavy drinking were examined by computerised tomography. Brain and ventricular volumes were calculated by the use of a semiautomated method with the rater blind to twinship and drinking history. 11 severely dependent alcoholics had larger ventricular volumes and ventricle/brain ratios than did their normal cotwins. These changes correlated best with the length of time the alcoholic twin had been drinking eight centilitres pure alcohol, equivalent to a bottle of wine, a day.

Schizophrenia is a common disorder with a life time prevalence of approximately 1 per cent. The illness often develops in young adults, who were previously normal, and is characterized by a constellation of symptoms including hallucinations and delusions (psychotic symptoms) and symptoms such as severely inappropriate emotional responses, a disorder of thinking and concentration, erratic behaviour as well as social and occupational deterioration. A considerable proportion of the variance in the liability to develop schizophrenia may be genetic, but segregation analysis, to establish a mode of transmission, has not produced a consistent result. One of these studies was carried out in Iceland and made use of the large family size and extensive geneaological information present in that country. Here we demonstrate genetic linkage of two DNA polymorphisms on the long arm of human chromosome 5 to schizophrenia in seven British and Icelandic families with multiple affected members. The results indicate the existence of a gene locus with a dominant schizophrenia-susceptibility allele. Inheritance of the allele in the families studied suggests that it may also predispose to psychiatric conditions such as schizophrenia spectrum disorders and a variety of other disorders. This report provides the first strong evidence for the involvement of a single gene in the causation of schizophrenia.

Recently a linkage study on five Icelandic and two English pedigrees has provided evidence for a dominant gene for schizophrenia on 5q11-13 (ref. 1). In that study, families with bipolar illness were not included. Using the same probes, two similar but independent investigations on one Swedish pedigree and on fifteen Scottish families excluded linkage to schizophrenia. To evaluate whether the susceptibility gene on 5q11-13 is a common cause of schizophrenia in other populations, we examined five affected North American pedigrees using probes to the D5S39, D5S76 and dihydrofolate reductase loci. Two families in the present series had cases of bipolar disorder. We found that linkage can be excluded by multipoint analysis. These results, taken together, suggest that the disease gene on 5q11-13 does not account for most cases of familial schizophrenia.

A clinico-radiological assessment of seven high-dose long-term heroin addicts was undertaken to look for the presence of neuropsychological and CT scan pathology. Definite abnormalities were found in both neuropsychological and CT scan assessments in some individuals but the pattern of CT scan and neuropsychological deficit in relation to heroin addiction was not constant.

The twin register of the Maudsley Hospital was used to select a series of monozygotic (MZ) and same-sexed dizygotic (DZ) twins who had been given an ICD8 diagnosis related to alcoholism. They and their co-twins were traced. Medical and drinking histories were compiled (from records and by interview) for 56 twin-pairs, to permit current and retrospective diagnosis of any neurotic disorders. RDC diagnoses of panic disorder, generalised anxiety disorder, obsessive-compulsive disorder and phobias were more common in the alcoholic probands and the co-twins who were also alcoholics than in the normal-drinking co-twins. The Eysenck Personality Questionnaire (EPQ) and the Severity of Alcohol Dependence Questionnaire (SADQ) were completed by 54 individuals. Neuroticism scores were significantly higher for both male and female alcoholics than for their normal-drinking co-twins; and intra-pair differences in neuroticism were significantly correlated with intra-pair differences in severity of dependence. These results suggest that both clinically diagnosed neurotic illness and high neuroticism scores are more often a consequence than a cause of alcoholism.

Techniques in molecular biology and genetics have made it possible to systematically study gene effects in human disease. The number of gene clusters specifically encoding human brain structure and function is probably about 1,600 or half of all clusters. Evolutionary effects such as linkage disequilibrium and conservation of exons (DNA encoding structural proteins) as well as the fact that there are a tractable number of gene clusters involved, tend to make it quite likely that DNA pathology or DNA variation (polymorphism) predisposing to mental illness can be detected. Genes involved in mental illness can be detected either by studying DNA obtained from blood samples (genomic DNA) directly or by the analysis of mRNA and proteins from suitable cell or tissue preparations. The study of gene expression in the human brain is still in its infancy, nevertheless there are some hints that non-poly-adenylated mRNAs may be important in brain development and certain transcribed sequences may have a specific role in gene expression of the brain. The advantage of studying genomic DNA by the use of linkage and association analysis in multiply affected families is that it will, in the end, almost certainly yield a positive result for a disease with a substantial genetic input. Analysis of gene products from tissues such as brain could in theory detect specific disease genes but the approach will also identify genes secondarily affected by the disease process. Differentiation of genes that are primarily causing mental illness from those that are secondarily affected can be carried out by using such candidate genes as linkage markers in multiply affected families.

Manic depression is a severe cyclic mental illness that can be unipolar or bipolar and has a lifetime risk of approximately 7 per 1,000 in most populations. Families with multiple cases of manic depression have been described that are compatible with both autosomal dominant and X-linked modes of genetic transmission. Psychoactive antidepressant and stimulant drugs that help to ameliorate depression and mania are thought to act by affecting catecholamine neurotransmitter systems such as adrenaline, noradrenaline and dopamine, amongst others. Mutations affecting the tyrosine hydroxylase (TH) gene, which encodes the rate-limiting enzyme for the synthesis of these three neurotransmitters, might therefore be responsible for causing the manic depressive phenotype. We have studied three Icelandic kindreds amongst whom it appears that a single autosomal dominant disease allele is segregating. In these families there were 44 cases amongst 73 individuals at risk. Genetic linkage studies were carried out using clones encoding tyrosine hydroxylase the variable portion of the Harvey-ras-1 (HRAS1) locus and the variable region of the insulin gene (INS). All three markers are closely linked on chromosome 11 and were used to observe the segregation of restriction fragment length polymorphisms (RFLPs) in the three affected kindreds. We found no evidence for linkage to these markers in any of the three families. In contrast, Gerhard et al. found linkage between manic depression and HRAS1 in a single large Amish kindred. We conclude that there is genetic heterogeneity of linkage in manic depression. Therefore mutations at different loci are responsible for the manic depressive phenotype in the Amish and in Iceland.

Recent research has shown that there are X-linked and possibly chromosome 11-linked forms of manic depression as well as at least one other autosomal form. Segregation analyses of large affected families and the finding of genetic linkage between chromosome specific markers and manic depression mutations provide strong evidence that bipolar as well as unipolar forms of manic depression (MD) within the same family are inherited as a dominant gene disorder. This clarification of the etiology of certain types of depression should bring changed attitudes within psychiatry and may serve to stimulate discussion of the role of evolutionary mechanisms. From a clinical point of view, it has now become possible to determine whether clinical (phenotypic) variation reflects the underlying genotypic heterogeneity of linkage. A preliminary analysis of data from four recent studies shows that there is no clear correlation between such clinical features as the ratio of unipolar to bipolar cases and the genotypic form of manic depression. Further recombinant DNA research, proven to be successful in other genetic diseases, can soon be applied to manic depression. The specific problems posed by manic depression for these techniques are discussed.