Narcolepsy 1


A number sign (#) is used with this entry because of evidence that narcolepsy-1 (NRCLP1) is caused by heterozygous mutation in the HCRT gene (602358) on chromosome 17q21. One such patient has been reported.


Adie (1926) first delineated narcolepsy as a separate and specific entity. It is a sleep disorder characterized by attacks of disabling daytime drowsiness and low alertness. The normal physiologic components of rapid eye movement (REM) sleep, dreaming and loss of muscle tone, are separated and also occur while the subject is awake, resulting in half-sleep dreams and episodes of skeletal muscle paralysis and atonia (cataplexy and sleep paralysis). Unlike normal sleep, that of narcolepsy often begins with REM activity and the time taken to fall asleep is shorter than normal.

In contrast to animal models, human narcolepsy is not a simple genetic disorder. Most human cases of narcolepsy are sporadic and carry a specific HLA haplotype (Peyron et al., 2000). Familial cases are the exception rather than the rule, and monozygotic twins show only partial concordance (25 to 31%) (Mignot, 1998).

Genetic Heterogeneity of Narcolepsy

Additional narcolepsy loci have been mapped to chromosomes 4 (NRCLP2; 605841), 21q (NRCLP3; 609039), 22q13 (NRCLP4; 612417), 14q11 (NRCLP5; 612851), and 19p13.2 (NRCLP6; 614223). NRCLP7 (614250) is caused by mutation in the MOG gene (159465) on chromosome 6p22. Resistance to narcolepsy is associated with minor alleles of a SNP and a marker in the NLC1A gene (610259) on chromosome 21q22.

Clinical Features

In 3 generations of a family, Daly and Yoss (1959) found 12 definite and 3 possible cases. Whereas about two-thirds of all cases of narcolepsy (sleeping attacks) are associated with cataplexy (paroxysmal attacks of weakness or frank paralysis, associated especially with strong emotion), only 3 of the 12 affected persons in this family displayed cataplexy. Furthermore, in these cases the weakness was mild.

In a later publication, Yoss (1970) reported studies with infrared pupillography in narcolepsy families, leading to the conclusion that narcolepsy is polygenic, i.e., that the affected persons are at one end of a spectrum. When a person is awake and alert in total darkness, his pupils are large. During sleep the pupils are small. The pupils are intermediate in size when the subject is between these two extremes. This is the basis of infrared pupillography as a gauge of wakefulness. The author suggested that it would be very unusual for 2 persons with philagrypnia (ability to stay alert with little sleep) to have an offspring with narcolepsy. Based on the findings of abnormalities on pupillometry (Yoss et al., 1969), a disturbance of the central autonomic nervous system in narcolepsy had been suggested. Hublin et al. (1994) proved that this was not the case in their studies of 22 unmedicated narcoleptics with extensive tests of autonomic function, all of which proved to be normal.

Thannickal et al. (2000) studied the hypothalamus of 16 human brains, including those of 4 narcoleptics. The human narcoleptics had an 85 to 95% reduction in the number of HCRT neurons. Melanin-concentrating hormone (176795) neurons, which are intermixed with HCRT cells in the normal brain, were not reduced in number, indicating that cell loss was relatively specific for HCRT neurons. The presence of gliosis in the hypocretin cell region is consistent with a degenerative process being the cause of the HCRT loss in narcolepsy.

Nishino et al. (2000) measured immunoreactive HCRT in the cerebrospinal fluid of 9 patients with narcolepsy and 8 age-matched controls. HCRT1 was detectable in all controls; in 7 of 9 patients, HCRT concentrations were below the detection limit of the assay. The authors proposed that an HLA-associated autoimmune-mediated destruction of HCRT-containing neurons in the lateral hypothalamus might produce narcolepsy in these patients.

In 31 patients with narcolepsy, Dalal et al. (2001) found reduced or undetectable levels of CSF hypocretin compared to controls. Plasma levels of hypocretin, however, were at normal levels, similar to controls, suggesting that systemic hypocretin derived from CNS-independent sources is preserved in narcolepsy. The authors noted that a potential autoimmune mechanism for the disorder is unlikely to be directed against the hypocretin molecule.

Dauvilliers et al. (2001) collected data on age at onset and severity of narcolepsy in 317 patients with well-defined narcolepsy-cataplexy from Montpellier, France, and in 202 patients from Montreal, Canada. The mean age at onset was 23.4 years in Montpellier and 24.4 years in Montreal. However, the age at onset was bimodal in these 2 independent patient populations: a first peak occurred at 14.7 years, and a second peak at 35 years. Age at onset clearly differentiated patients with a positive family history of narcolepsy (early onset) from those without a family history. Other clinical and polygraphic findings suggested that young age at onset is associated with increased severity of the condition (higher frequency of cataplexy and decreased mean sleep latency on the Multiple Sleep Latency Test). Dauvilliers et al. (2001) suggested that age at onset is genetically determined.

Arii et al. (2004) found very low CSF hypocretin-1 levels in 6 of 6 children with narcolepsy ranging in age from 6 to 16 years. All were DR2-positive. Decreased levels of CSF hypocretin-1 were also found in children with Guillain-Barre syndrome (139393), head trauma, brain tumor, and CNS infection. The authors concluded that measurement of CSF hypocretin-1 is diagnostically useful in children.

Clinical Management

Dauvilliers et al. (2009) reported a 28-year-old woman with narcolepsy who had complete reversal of symptomatology after intravenous immunoglobulin (IVIg) infusion. She was positive for HLA-DRB1*1501 and HLA-DQB1*0602. Nighttime polysomnography before treatment showed a mean sleep latency of 5 minutes and 2 sleep-onset REM periods. CSF hypocretin-1 level was undetectable. Polysomnograpy after IVIg treatment showed substantial improvement, with a mean sleep latency of 8.6 minutes and no sleep-onset REM periods. A second lumbar puncture showed a normal hypocretin-1 level. The findings indicated that narcolepsy may be an autoimmune disease. Dauvilliers et al. (2009) hypothesized that an acute and focal inflammatory process may have blocked hypocretin production without neurons being destroyed and that such a putative autoimmune process could be reversible with IVIg treatment at disease onset.


Gelardi and Brown (1967) reported a family in which 11 persons in 4 generations had cataplexy. Three may have had narcolepsy. No instance of male-to-male transmission occurred in the pedigree.

In a study of 50 persons with narcolepsy-cataplexy, Baraitser and Parkes (1978) found that 52% had an affected first-degree relative and that 41.9% of the sibs of those probands with an affected parent were similarly affected. In one-third of instances in which 2 sibs were affected, a parent was affected. After correction for age, 41.2% of children were affected.

The results of the family study of Mueller-Eckhardt et al. (1986) were consistent with a dominant mode of inheritance with incomplete penetrance of a hypothetical disease susceptibility gene.

In a clinic population of 334 unrelated narcoleptic patients, Guilleminault et al. (1989) found that 40% of probands had at least 1 family member with an isolated daytime sleepiness complaint and 6% had positive family history of narcolepsy. In only 2 families were 3 or more relatives affected. Family members often shared the same HLA-DR2 haplotype as the proband but did not have narcolepsy.


To find direct evidence for the autoimmune hypothesis of narcolepsy, Smith et al. (2004) injected mice with purified IgG from serum of 9 patients with narcolepsy. Bladder detrusor muscle strips from the mice showed enhanced contractile responses to the cholinergic muscarinic agonist carbachol and to endogenously released acetylcholine on electrical field stimulation compared to muscle strips from mice injected with IgG from control individuals. Functional activity was present in the IgG from each patient with narcolepsy. There was no increase in activity of the vas deferens, a model of sympathetic neurotransmission. Smith et al. (2004) concluded that patients with narcolepsy have a functional IgG autoantibody that enhances postganglionic cholinergic neurotransmission.

Using detailed immunohistochemical studies, Crocker et al. (2005) and Blouin et al. (2005) independently demonstrated that patients with narcolepsy have a marked reduction (5 to 11% of normal) of orexin-producing neurons in the posterior, lateral, dorsal, and dorsomedial hypothalamic nuclei compared to normal controls. The findings of both studies indicated that narcolepsy is associated with a loss of the orexin-producing neurons themselves, rather than a failure to produce the orexin protein. The results were consistent with selective neurodegeneration of these cells or an autoimmune process.

Latorre et al. (2018) used sensitive cellular screens and detected hypocretin-specific CD4+ T cells in all 19 narcolepsy patients tested. T cells specific for tribbles homolog-2 (TRIB2; 609462), another self-antigen of hypocretin neurons, were found in 8 of 13 patients. Autoreactive CD4+ T cells were polyclonal, targeted multiple epitopes, were restricted primarily by HLA-DR (see 142860), and did not crossreact with influenza antigens. Hypocretin-specific CD8+ T cells were also detected in the blood and cerebrospinal fluid of several patients with narcolepsy. Autoreactive clonotypes were serially detected in the blood of the same, and even of different, patients, but not in healthy control individuals. Latorre et al. (2018) concluded that their findings solidified the autoimmune etiology of narcolepsy.

Molecular Genetics


Peyron et al. (2000) identified a dominant, presumably de novo mutation of the HCRT gene in a single case of early-onset narcolepsy (602358.0001).

Association with the HLA Region on Chromosome 6p21

Nearly 100% of individuals of European descent with narcolepsy carry the HLA haplotype DRB5*0101-DRB1*1501-DQA1*0102-DQB1*0602. However, 15 to 25% of individuals in the general population also carry this risk haplotype, suggesting that it is necessary but not sufficient for development of the disorder (summary by Hor et al., 2010).

Some reported findings are consistent with an immunologically mediated destruction of hypocretin-containing cells in human narcolepsy (Mignot et al., 2001).

Langdon et al. (1984) found that all of 37 patients were HLA-DR2 compared to 21.5% of 200 normal controls. They pointed out that this is the strongest HLA-disease association yet found. Studies with DNA probes will be of great interest; a subtype of DR2 may be responsible. The molecular defect in narcolepsy may be elucidated by this line of research. Conventional linkage studies would be worthwhile.

In Japan, Juji et al. (1984) found that all patients with narcolepsy were DR2 positive. Matsuki et al. (1985) studied HLA and complement types in 111 Japanese patients with narcolepsy and in 6 multiple case families. They found that B35-DR2, B15-DR2, and B51-DR2 were the most frequent haplotypes in Japanese narcoleptics whereas these were rare in the normal population. The most frequent haplotype of HLA-DR2 in Japanese had a frequency in narcoleptics only one-third of that in controls. It is a different haplotype, A3-Cw7-B7-DR2-DQw1 (see HLA-DQB1, 604305), that is found most frequently in Caucasoid narcoleptics.

In family studies, Matsuki et al. (1985) found 4 persons with no signs of narcolepsy among 19 subjects with the disease susceptibility haplotypes, suggesting incomplete penetrance. Mueller-Eckhardt et al. (1986) found that 57 out of 58 unrelated German narcoleptics were positive for DR2 and DQw1. One patient with typical signs of narcolepsy was found to be negative for these 2 specificities. In an addendum they called attention to 2 other cases of DR2-negative narcolepsy. In a study of narcolepsy among Israeli Jews, Wilner et al. (1988) found by conventional HLA typing that all 7 narcoleptic patients studied carried the HLA-DR2 haplotype. Analysis of RFLPs showed that all 7 had the RFLP pattern seen in the DR2,Dw2 haplotype. The frequency of this haplotype in the healthy Israeli population is 3.2%. Family studies were not done in this population.

Although most patients with narcolepsy had the DR2 haplotype, Guilleminault et al. (1989) found 2 new DR2-negative narcoleptics and predicted that as many as 9% of unrelated North-American white patients with narcolepsy will be DR2 negative. Among the 19 instances of a positive family history of narcolepsy, there was 1 instance of affected father and son. Singh et al. (1990) presented data relevant to the role of DR2 in this disorder from a study of 3 families. Kuwata et al. (1991) indicated that they had typed HLA antigens for 264 patients with narcolepsy since their first report in 1984 (Juji et al., 1984). All the patients were positive for the DRw15 subtype of HLA-DR2 and the DQw6 subtype of DQw1. Nucleotide sequencing and pulsed field gel electrophoresis revealed no completely specific change in the DR/DQ region that could explain this susceptibility. Mignot et al. (1991) indicated that narcolepsy is associated with the MHC haplotype HLA-DRw15 (DR2), Dw2, DQw6 (DQw1), which is present in 32.8% of Caucasians and 7.7% of Asian normal controls, respectively, compared with 90 to 95% of Caucasian and 100% of Asian narcoleptic patients. No tests of immunopathology have been found abnormal in narcoleptic patients. DNA studies have not revealed differences between DRw15 and DQw6 genes of narcoleptics and those of normals.

Matsuki et al. (1992) reviewed briefly the evidence indicating that a specific allele of DQw6, namely DQB1*0602, has been found in all narcoleptic patients tested, demonstrating that the disease susceptibility gene for narcolepsy is this one or a gene located close to it rather than DRw15 (DR2). Thus, DQ rather than DR is the marker gene to search for in the diagnosis of narcolepsy. Multiplex families with narcolepsy not linked to HLA have been reported (Guilleminault et al., 1989; Singh et al., 1990).

Mignot et al. (1997) HLA-typed 509 patients enrolled in a clinical trial of the drug modafinil and analyzed the results in relation to cataplexy, a symptom of narcolepsy characterized by muscle weakness triggered by emotions. The results showed that the HLA association (with DQB1*0602) is as tight as previously reported (85 to 95%) when cataplexy is clinically typical or severe. They also found that patients with mild, atypical, or no cataplexy had a significantly increased DQB1*0602 frequency (40 to 60%) in comparison with ethnically matched controls (24%).

Siegel (1999) reviewed the nature of narcolepsy and the hypocretin (HCRT; 602358) system. The hypocretins (which are also called orexins) are a pair of neuropeptides produced from a single precursor protein (de Lecea et al., 1998). Nishino et al. (2000) measured immunoreactive HCRT in the cerebrospinal fluid of 9 patients with narcolepsy and 8 age-matched controls. All patients were positive for HLA-DR2/DQB1*0602. HCRT1 was detectable in all controls; in 7 of 9 patients, HCRT concentrations were below the detection limit of the assay. The authors proposed that an HLA-associated autoimmune-mediated destruction of HCRT-containing neurons in the lateral hypothalamus might produce narcolepsy in these patients.

Mignot et al. (2001) investigated the influence of HLA class II alleles, in addition to HLA-DQB1*0602, on susceptibility to narcolepsy. In African Americans, white Americans, and Japanese, a strong effect of DQB1*0602 homozygosity was observed. They found that 9 HLA class II alleles carried in trans with DQB1*0602 influenced disease predisposition. Two DQ and 4 DR alleles were associated with significantly higher relative risks; 3 DQ alleles were found to be protective. They interpreted the results to indicate that complex HLA-DR and -DQ interactions contribute to the genetic predisposition to human narcolepsy but that additional susceptibility loci are also most likely involved. Together with the discoveries concerning the role of hypocretin in narcolepsy, the findings were considered consistent with an immunologically mediated destruction of hypocretin-containing cells in this disorder.

Dauvilliers et al. (2004) reported a pair of monozygotic twin girls who were discordant for narcolepsy and for CSF hypocretin levels. At age 11 years, the affected twin developed recurrent excessive daytime sleepiness with frequent sleep attacks at school, as well as sleep paralysis and hypnagogic hallucinations. A rapid weight gain was noted at disease onset. CSF hypocretin was below the level of detection. No mutations were identified in the hypocretin and both hypocretin receptor genes (HCRTR1, 602392; HCRTR2, 602393) The unaffected twin had no sleep symptoms, normal levels of CSF hypocretin, and no weight gain. Both girls were positive for the HLA-DQB1*0602 allele. Dauvilliers et al. (2004) concluded that there is a strong environmental triggering effect in the development of narcolepsy and suggested that DQB1*0602 may confer susceptibility.

Khatami et al. (2004) reported a pair of monozygotic twin women who were concordant for narcolepsy and HLA-DQB1*0602. Onset in both sisters occurred around age 7 to 9 years, with full manifestation during adolescence. Although both reported cataplexy, the most severe features were daytime sleepiness and sleep paralysis. Complete cataplexy was uncommon. Both sisters had normal CSF hypocretin levels, and no mutations were identified in the hypocretin or both hypocretin receptor genes. Khatami et al. (2004) suggested that HLA type and hypocretin signaling may be independently associated with the development of narcolepsy.

In a genomewide association study of 562 European individuals with narcolepsy and 702 ethnically matched controls, with independent replication in 370 cases and 495 controls, all of whom were heterozygous for the risk DRB1*1501-DQB1*0602 haplotype, Hor et al. (2010) found a significant association with a protective variant rs2858884 located 8.8-kb upstream of HLA-DQA2 (613503) (p = 2.94 x 10(-8); odds ratio of 0.56). The frequency of the minor C allele was higher (17%) in the control population than in individuals with narcolepsy (10%), suggesting a protective effect. Further analysis revealed that rs2858884 is strongly linked to DRB1*03-DQB1*02 (p less than 4 x 10(-43)) and DRB1*1301-DQB1*0603 (p less than 3 x 10(-7)). Narcolepsy patients almost never carried the DRB1*1301-DQB1*0603 haplotype in trans with the HLA risk haplotype (p less than 6 x 10(-14)). This protective HLA haplotype further suggested a causal involvement of the HLA region in narcolepsy susceptibility.

Biochemical Features

The human MHC class II molecule encoded by DQA1*0102/DQB1*0602 (termed DQ0602) confers strong susceptibility to narcolepsy but dominant protection against type I diabetes (222100). To elucidate the molecular features underlying these contrasting genetic properties, Siebold et al. (2004) determined the crystal structure of the DQ0602 molecule at 1.8-angstrom resolution. Structural comparisons to homologous DQ molecules with differential disease associations highlighted a previously unrecognized interplay between the volume of the P6 pocket and the specificity of the P9 pocket, which implies that presentation of the expanded peptide repertoire is critical for dominant protection against type I diabetes. In narcolepsy, the volume of the P4 pocket appears central to the susceptibility, suggesting that the presentation of a specific peptide population plays a major role.

Population Genetics

The frequency in the United States is estimated to be between 0.050% and 0.067%.

Narcolepsy affects more than 1 in 2,000 individuals (Blouin et al., 2005).

Animal Model

Hereditary narcolepsy has been described in several animal species. Motoyama et al. (1989) could not establish linkage to the major histocompatibility complex or to a specific MHC-related RFLP in the canine disease. Mignot et al. (1991) reported on studies in a colony of narcoleptic dogs in which the disorder is transmitted as an autosomal recessive trait with full penetrance, designated canarc-1. The same gene has been found in both Doberman and Labrador breeds (Foutz et al., 1979; Baker et al., 1982). As in the human disorder, affected animals are excessively drowsy, have a short sleep latency during the day and fragmented sleep during the night, and display the disease hallmark, cataplexy (episodes of muscle weakness induced by emotions). Mignot et al. (1991) demonstrated that the canarc-1 gene is not located within the dog MHC cluster but is tightly linked to a polymorphic band with strong homology with a human switch region of the mu immunoglobulin heavy chain gene.

Faraco et al. (1999) isolated genomic clones encompassing the canarc-1 marker and the variable heavy chain immunoglobulin region in canines. They presented data indicating that the mu-switch-like marker is not part of the canine immunoglobulin machinery.

Ostrander and Giniger (1999) discussed narcolepsy in dogs and mice. They published a partial pedigree of a Doberman pinscher family in which narcolepsy was autosomal recessive and fully penetrant, as published by Lin et al. (1999). Lin et al. (1999) mapped and cloned the responsible gene. The region of canine chromosome 12 to which the canarc-1 locus was mapped was found to bear orthologous synteny with the well-mapped region of human 6p21. This greatly facilitated the development of a BAC contig across the region and the identification of the gene encoding the hypocretin type 2 receptor (HCRTR2; 602393) as a plausible candidate. By genomic sequencing of the Hcrtr2 gene of the narcoleptic Doberman, Lin et al. (1999) identified an insertion that resulted in aberrant splicing and a truncated transcript. They identified a different deletion in the Hcrtr2 transcript in the narcoleptic Labrador. Lin et al. (1999) speculated that these changes disrupt the proper membrane localization or transduction functions of this receptor. Chemelli et al. (1999) engineered a mouse model of narcolepsy that independently implicated the same genetic pathway. Physiologic and pharmacologic studies of the Doberman pinschers suggested a close similarity between the canarc-1 phenotype and human narcolepsy (Nishino and Mignot, 1997).


Familial narcolepsy has been known since Westphal (1877) described an affected mother and son.