SECTION 3 - ENABLING TECHNOLOGY

Genetics and Proteomics: Phenotype Issues and Methodological Approcaches

Background

Sleep behavior is extremely variable across and within animal species, suggesting the importance of genetically based differences. Limited genetic epidemiological data indicate that many sleep disorders have a strong genetic component. Advances in genetics and genomics have been spectacular and include sequencing the genomes of various organisms and high throughput studies using genetic arrays and polymorphic markers. Animal models of sleep and circadian disorders with selected genetic alterations are now being generated. Similar developments in the area of protein characterization and the more general field of proteomics are now rapidly developing. The field of sleep disorders medicine is now well positioned to take advantage of these new technologies.

A solid foundation in the area of phenotyping sleep and its disorders in both animals and humans is needed before proceeding with genetic analysis. The discovery of new methods and improvements in existing sleep recording techniques in humans are also needed. When performing genetic studies, it is important to consider potential study design limitations. The strength and location of linkage regions identified, for example, depends on the strength and precise phenotype selected. Thus, linkage regions may not be identified if the power of the study is insufficient, and large numbers may be required for such studies to be successfully accomplished. Even if linkage regions are identified, these may be large and contain many candidate genes. Sequencing of candidate genes may not yield mutations or may identify mutations that are not relevant to the phenotype. In this case, the use of complementary approaches such as DNA expression arrays and proteomics to identify novel genes of interest may be a powerful approach to identify relevant candidate genes.

Molecular correlates of sleep and diurnal rhythms would be important for a wide range of clinical studies. Much human research relies on blood samples, which are easily obtained, but often there is little knowledge about chronobiologic variations in the parameters being measured, and no regard for the time of day or the sleep history of the subject when the sample is taken. The impact of sleep and diurnal variation on other systems is exemplified by blood coagulation and thrombotic tendencies. Myocardial infarctions or strokes occur more often in the morning, and blood properties such as platelet aggregation may change during the day. It would be useful to have molecular markers to assess chronobiologic and sleep history variability.

Progress In The Last 5 Years

- There have been advances in technology development in the area of sleep phenotyping in mice. In the clinical arena, sleep recording devices have been made more portable and easier to use. Similar progress has been made in automated sleep scoring algorithms that utilize concepts such as neural networks, fuzzy math, and wavelet fitting, allowing for more rapid analysis and the possibility of high-throughput sleep phenotyping.

- Eight genes that significantly contribute to the generation of circadian periodicity have been isolated in mammals. Recently, studies in humans have shown, for the first time, a correspondence between human and animal sleep phenotypes. Most strikingly, a mutation in the gene HPER2, a gene known to be involved in the regulation of circadian rhythmicity in mammals, was demonstrated to cause Familial Advanced Sleep Phase Syndrome (FASPS) in a human family. Additionally a polymorphism in CLOCK, another gene involved in the generation of circadian rhythmicity, was found to influence morningness-eveningness tendencies in humans. These studies are likely to be extended, with the discovery of other human mutations and polymorphisms affecting circadian regulation.

- Similar progress has been made using a genetic approach in Narcolepsy (Section V). Using a positional cloning approach, mutations in the hypocretin receptor 2 gene) have been isolated in a canine model of Narcolepsy. The knocking-out of preprohypocretin, a gene initially believed to be involved in appetite regulation, led to the establishment of a murine model of narcolepsy. These findings were found to be directly applicable to human narcolepsy-cataplexy, as it has been now shown that most patients have a hypocretin deficiency. This last finding is remarkable as the disorder in humans is genetically complex and HLA-associated. These results demonstrate the importance of careful phenotyping of human sleep disorders to reduce disease heterogeneity and the importance of animal models.

- Genome screening and genetic association studies have been initiated in Sleep-Disordered Breathing (SDB) and Restless Legs Syndrome (RLS) (Section V). Significant linkage results have been reported and await confirmation. In the candidate gene area, an association between APOE e4 and sleep apnea has been reported and will need to be replicated.

Research Recommendations

- Develop new methods to measure sleep, circadian physiology and sleepiness in large numbers of animals and human subjects. One goal is to develop and validate surrogate measures. Another goal is to define normal sleep pattern variation in the general human population. Normative data will also be critical to define and validate existing or novel sleep disorder phenotypes. These data will be needed to elucidate corresponding genetic factors.

- Continue the study of animal models such as fruit fly, zebra fish and mice to enhance our understanding of physiology and circadian biology. Use of these models to study sleep or the regulation of rest/activity should be a priority, as this may lead to the discovery of novel sleep regulatory pathways. Powerful new genetic approaches, such as those used to discover circadian clock genes (e.g. mutagenesis screens), can be used to find new genes that are involved in the homeostatic need to sleep and in interactions between the circadian and sleep-wakefulness systems. Other approaches such as "quantitative trait loci" should be considered insofar as true "sleep knockouts" may be not viable in mutagenesis screens.

- Identify new disease phenotypes, including rare familial sleep disorders or subtypes of current sleep disorders based on treatment response or other characteristics. The study of multiplex families where sleep disorders appear to be segregating as a single gene could lead to the positional cloning of novel sleep disorder-related genes. This may facilitate our understanding of other, more common sleep disorders, as well as increase our understanding of the normal physiology of sleep.

- Study inter-individual differences in baseline sleep, circadian physiology and response to sleep deprivation in a large number of subjects to better define normal and pathological conditions. Recent results indicate large inter- individual differences in how people react to sleep deprivation. Additionally, subjective sleepiness varies significantly in patients with equivalent degrees of SDB and sleep fragmentation. The study of these inter-individual differences has both clinical and basic research relevance.

- Twin prevalence and segregation analysis studies need to be conducted for all sleep disorders across various populations in order to estimate heritability and environmental contributions for each sleep disorder. This will help prioritization and design of further genetic studies.

- Genome screening studies using classical family design and candidate gene-based research should be continued and extended. As most sleep disorders are genetically complex, large numbers will be needed and there is a need to encourage the blending of epidemiological and genetic designs. Ethnic variation in the expression and the genetic basis of various sleep disorders has been identified and will require further exploration. Studies of SDB, RLS, and other disorders such as hypersomnia will benefit from this approach.

- Genetic array and proteomic studies in selected tissue samples or protein-protein interaction experiments should be encouraged. The use of these new techniques can be extremely useful to discover novel components within a molecular or a disease pathway.

- Mouse models are increasingly used in genetic and behavioral studies, and have been created but not yet widely utilized in Narcolepsy. A large number of mice with various genetic alterations are being created in multiple laboratories but are rarely tested for sleep abnormalities. Finding sleep abnormalities in some of these models could lead to the discovery of novel sleep regulating pathways that may be involved in selected sleep disorders. To remedy this situation, there is a need for developing and distributing genetically determined animal models for sleep disorders. Collaborative efforts should be explored to phenotype sleep in mice models for investigators that are working outside of the field of sleep research.