Review articleNarcolepsy and the HLA region
Section snippets
Clinical and epidemiological aspects of narcolepsy
Narcolepsy–cataplexy is a neurological disorder affecting 0.02–0.18% of the general population Honda, 1979, Mignot, 1998. The syndrome is characterized by excessive daytime sleepiness, disturbed nocturnal sleep and abnormalities in rapid eye movement (REM) sleep (such as cataplexy, sleep paralysis, and hypnagogic hallucinations) Aldrich, 1990, Bassetti and Aldrich, 1996. It is currently diagnosed using nocturnal polysomnography and the demonstration of sleepiness and REM transitions during the
Structure and function of the HLA complex
Major Histocompatibility Complex (MHC) genes play a key role in the recognition and processing of foreign antigens by the immune system. The human MHC system is also called human leukocyte antigen (HLA), based on its initial description relevant to tissue transplantation. The HLA complex maps on chromosome 6 (6p21.31) and is divided into three subregions, MHC class I, II, and III (Fig. 1). Several genes in the class I and II regions are involved in the processing and presentation of antigens to
HLA-DR2 and narcolepsy–cataplexy
Early HLA polymorphism studies used serologically based HLA typing techniques and panels of poorly selective antibodies recognizing HLA dimers expressed on the surface of white blood cells (Fig. 2). In the early 1980s, a significant association of narcolepsy–cataplexy with HLA class I Bw35 was discovered in Japanese (Juji et al., 1981). In Caucasians, HLA-Bw7 but not Bw35 was found to be increased (Seignalet and Billiard, 1984). Studies were soon extended to the class II region, with the
HLA-DQB1*0602 is a better marker than HLA-DR2 for narcolepsy–cataplexy
The finding that the association with HLA-DR2 was less tight in African-Americans suggested interethnic differences. Since 1983, further sequencing studies of HLA complex genes have indicated higher allelic diversity than suspected using serological typing techniques. HLA-DR2, for example, was first split into DR15 and DR16 using improved serological reagents but HLA-DRB1 sequencing techniques have indicated 11 subtypes of DR15, now labeled DRB1*1501 to DRB1*1511. DQ1 was first split into DQ5
Sequencing and microsatellite marker studies in the HLA-DQ region
90–100% of narcolepsy patients with definite cataplexy share HLA-DQA1*0102-DQB1*0602 Matsuki et al., 1992, Mignot et al., 1994, Rogers et al., 1997. This allele combination is found in 12–38% of the control population in various ethnic groups Honda et al., 1983a, Honda et al., 1983b, Matsuki et al., 1992, Rogers et al., 1997, Lin et al., 1997. HLA-DRB1, DQA1 and DQB1 genes have been sequenced in normal and narcoleptic Caucasian and Japanese subjects and no differences were detected Lock et al.,
Complex HLA-DR and DQ interactions in narcolepsy–cataplexy
Recent studies indicate that a simple dominant effect for DQB1*0602 does not fully explain HLA-associated susceptibility in narcolepsy. In Caucasian-Americans and African-Americans, HLA-DQB1*0602 homozygosity increases two to four times the risk of developing narcolepsy with or without cataplexy (Pelin et al., 1998). This result led us to suspect that HLA class II alleles other than DQB1*0602 may influence disease susceptibility. To address this question, a collaborative study was initiated in
Other studies in the HLA region
The genetic contribution of HLA-DQ to narcolepsy susceptibility was also estimated by use of lambda statistics. In all cases, Lambda HLA was estimated to be low, with values ranging from 2 to 4. Results indicate that complex HLA-DR and -DQ interactions contribute to the genetic predisposition to human narcolepsy but that additional susceptibility loci are most likely involved (Mignot et al., 2001). Association studies were not only restricted to polymorphisms of HLA genes and pseudogenes (DRB1,
Human narcolepsy and autoimmunity
As indicated above, most of the other diseases known to be tightly HLA-associated (e.g. rheumatoid arthritis, insulin-dependent diabetes mellitus, celiac disease, and multiple sclerosis) are autoimmune in nature (Klein and Sato, 2000b). The DQB1*0602 association thus initially suggested that narcolepsy may result from an autoimmune injury within the central nervous system. The peripubertal onset of the disorder, together with the reported low concordance rate in monozygotic twins and the
Family studies in human narcolepsy
Narcolepsy has been reported to cluster in families since 1877 (Westphal, 1877). The risk of a first degree relative has consistently been shown to be 0.9–2.3% for narcolepsy–cataplexy Kessler, 1976, Kales et al., 1982, Honda et al., 1983a, Honda et al., 1983b, Guilleminault et al., 1989, Billiard et al., 1994, Nevsimalova et al., 1997. This value is 10–40 times higher than the general population prevalence (Mignot, 1998), suggesting predisposing genetic factors. Monozygotic twins are often
Canine narcolepsy and MHC studies
Canine and human narcolepsy are similar at the clinical and genetic epidemiological level. Canine narcolepsy affect many different breeds and, like in humans, most cases do not have a family history. Strikingly, however, the disorder is genetically transmitted as a single autosomal recessive gene, known as canarc-1 in Doberman pinschers and Labrador retrievers (Baker et al., 1982). Canarc-1 homozygous animals exhibit the full syndrome: cataplexy, fragmented sleep, and excessive daytime
Positional cloning of canarc-1
The lack of available genetic map in dogs led us to initially use candidate genes (such as immune and neurotransmitter system genes) and minisatellite markers (DNA fingerprinting using repetitive DNA probes) in our linkage study. In 1991, a polymorphic band crossreacting with a human immunoglobulin μ heavy-chain segment (Sμ) was found to segregate with narcolepsy. This suggested the possibility that the immune system was involved in the pathophysiology of the canine model, paralleling the
Hypocretin/orexin abnormalities in canine narcolepsy
Hcrtr2 (also called Orexin-receptor 2) encodes a G protein-coupled receptor with a high affinity for the neuropeptides system hypocretin/orexin. Hypocretins (hypocretin-1 and 2) are neurotransmitters processed from a common precursor, preprohypocretin. These peptides are produced exclusively by a well-defined group of neurons in the lateral hypothalamus. The endogenous ligands bind and activate two closely related G protein-coupled receptors: Hcrtr1 and Hcrtr2. Initial studies pointed out a
Genetic studies of the hypocretin system in human narcolepsy
Most human cases occur sporadically and are HLA-associated (Mignot, 1998). Mutations of the Hypocretin system were thus not expected to be a frequent cause of narcolepsy. An extended mutation screening study of the Hcrt, Hcrtr1, and Hcrtr2 loci was carried out and confirmed this result (Peyron et al., 2000). Patients screened included both sporadic and family cases with and without HLA-DQB1*0602. Frequent Hcrtr1 and Hcrtr2 polymorphisms were observed but no disease-causing mutations were
Histopathology and immunopathology studies in human narcolepsy
Hcrt positive cell bodies are localized in the posterior hypothalamus not far from the third ventricle Peyron, 1998, Nambu et al., 1999. Hypocretin-1 levels were measured in the CSF of narcolepsy patients and control subjects. The results showed that all controls have detectable Hypocretin-1 levels, while in the CSF of seven out of nine sporadic HLA-DQB1*0602 positive narcoleptic patients, Hcrt-1 was undetectable (Nishino et al., 2000). In an extended follow-up study including 38 narcoleptic
Perspectives in narcolepsy research and therapy
For many years, narcolepsy has been thought to be an autoimmune disorder. The finding that hypocretin-containing cells cannot be detected in the brain of narcoleptic patients, together with the report of perifornical gliosis, suggest that these cells may be a primary target for this process. Only 10,000–80,000 hypocretin-containing cells exist in the human brain and this low number may increase vulnerability to an autoimmune injury. Surprisingly, HLA-DR expression was found to be normal in the
Acknowledgements
Supported by NIH grants (NS23724, NS33797, HL59601) to Emmanuel Mignot. Marcel Hungs is supported by the Deutsche Forschungsgemeinschaft (HU 827/2-1).
References (96)
- et al.
Susceptibility to multiple sclerosis mediated by HLA-DRB1 is influenced by a second gene telomeric of the TNF cluster
Hum. Immunol.
(1999) - et al.
Canine model of narcolepsy: genetic and developmental determinants
Exp. Neurol.
(1982) - et al.
Narcolepsy
Neurol. Clin.
(1996) - et al.
Extraordinary association between HLA-DR2 and narcolepsy
Lancet
(1985) - et al.
Ankylosing spondylitis and HL-A 27
Lancet
(1973) - et al.
Autoimmune hypothesis in narcolepsy
Neurophysiol. Clin.
(1993) - et al.
Narcolepsy in orexin knockout mice: molecular genetics of sleep regulation
Cell
(1999) - et al.
Narcolepsy without unique MHC class II antigen association: studies in the canine model
Hum. Immunol.
(1989) - et al.
Genetic markers in narcolepsy
Lancet
(1984) - et al.
The sleep disorder canine narcolepsy is caused by a mutation in the hypocretin (orexin) receptor 2
Cell
(1999)