Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 Dog Narcolepsy

Narcolepsy has been well described in several mammalian species including dogs, mice and humans. Similar to human narcolepsy, main symptoms of dog narcolepsy are excessive daytime sleepiness (EDS) and cataplexy. Similar to that of humans, dog cataplexy is characterized by sudden muscle atonia without loss of consciousness. The animals usually remain alert and can follow movement with their eyes throughout the episode. The episodes typically last from several seconds to 30 min, often occurring when the dogs are running, eating, playing, excited, doing sexual activity. Food and play with other dogs are the well-documented paradigms used to trigger cataplexy. During cataplexy, dogs collapse onto its side or belly, and all physical movement ceases. Dogs usually come out of an episode in response to external stimuli, such as loud sounds. Occasionally, dogs fall asleep after attacks. Narcoleptic dogs sometimes suffer from sudden loss of consciousness or sleep attack, phenotypes that suggest excessive daytime sleepiness.

Like human cases, most dog narcolepsy is sporadic, and CSF orexin A levels are low or undetectable, suggesting that postnatal degeneration of orexin neurons due to similar pathophysiology as human cases. However, rare hereditary cases were found in Labrador retrievers, poodles, dachshunds, and Doberman pinschers.

Autosomal recessive transmission with full penetrance for canine narcolepsy was first established in Labrador retrievers and Doberman pinschers (Foutz et al. 1979). Although one of the predisposing genetic factors of human narcolepsy is a HLA-DQ allele, HLA-DQB1*0602, genetic linkage analysis had excluded the linkage between the dog narcolepsy and the canine major histocompatibility complex (Mignot et al. 1991).

In 1999, by screening BAC clones containing a genetic marker a team of Stanford University led by Mignot showed that loss of function mutations in OX2R (hcrtr-2) gene are responsible for familial narcolepsy-cataplexy in dogs, supporting the idea that deficiency in orexin (hypocretin) signaling might have an important role in pathophysiology of narcolepsy.

The OX2R gene of narcoleptic Dobermans contained a 226-bp insertion corresponding to a common canine short interspersed elements (SINE) repeat element located 35 bp upstream of the exon 4. The insertion of the SINE displaces a putative lariat branchpoint sequence ocated at position −40 through −46 upstream of the 3’-splice site. The mRNA potentially encodes a protein with 38 amino acids deletion within the fifth transmembrane domain followed by a frameshift, making the transcript functionally null. In narcoleptic Labrador retrievers, a separate line, G to A transition in the 5’-splice junction consensus sequence was found. Therefore, the exon 5 is spliced directly to exon 7, omitting exon 6. The C terminus of the OX2R of narcoleptic Labrador was truncated and did not include a seventh transmembrane domain. In both cases, these structural abnormalities of OX2R disrupt proper function of OX2R, suggesting the importance of OX2R-mediated signaling in the stabilization of wakefulness.

2 Rodent Narcolepsy

To date, several genetically-modified narcoleptic rodent models have been established, including orexin −/− mice, orexin-ataxin-3 mice/rats, OX2R-deficient mice, OX1R/OX2R -double knockout mice, and orexin-tTA;TetO DTA mice. Sleep/wakefulness sates of these mice are similar, although each shows some distinguishing characteristics. In this section, I will firstly mention about their overall phenotypes, and then discuss about phenotypes of each model (Figs. 1 and 2).

Fig. 1
figure 1

Representative hypnograms of wild-type, OX1R −/−, OX2R −/− and double receptor deficient mice over 12 h of the dark phase obtained by concatenating 16 s epoch EEG/EMG stage scores. The height of the horizontal line above baseline indicates the vigilance state of the mouse at the time. Baseline, W represents periods of wakefulness; S non-REM sleep; R REM sleep. Arrowheads highlight direct transitions from wakefulness to REM sleep

Full size image
Fig. 2
figure 2

Durations (upper panels) and total amounts (lower panels) of awake, non-REM sleep, and REM sleep states (means and SEM) for wild type and orexin/ataxin-3 (Tg) mice in light and dark periods. Narcoleptic mice consistently show significantly increased REM sleep times during the dark phase compared to normal mice. *, p < 0.05

Full size image

2.1 Phenotypes of Narcoleptic Mice

Orexin −/− mice, orexin/ataxin-3 mice, and OX1R/OX2R double orexin receptor deficient (OX1R −/−; OX2R −/−) mice showed the almost identical sleep/wakefulness characteristics, including abrupt behavioral arrest (cataplexy) and fragmentation of sleep/wakefulness states. This section describes these phenotypes (Table 1).

Table 1 Phenotypes of rodent narcolepsy models produced by genetic engineering
Full size table

These phenotypes caused by genetic modifications are strikingly similar to symptomes seen in human narcolepsy. However, it should be noted that familial transmission of human narcolepsy is very rare in humans, and even in these rare cases penetrance is far less than 100 %. No mutation has been found so far either in the prepro-orexin or orexin receptor genes of human narcolepsy patients, except an unusually severe, early onset case, which is associated with mutation in the prepro-orexin gene that impairs peptide trafficking and processing (Peyron et al. 2000).

Behavioral arrests (Cataplexy): These narcoleptic (orexin −/−, orexin/ataxin-3, and OX1R −/− ;OX2R −/−) mice exhibited substantial numbers of behavioral arrests during the dark periods. They were specifically recognized by the abrupt cessation of purposeful motor activity associated with sudden, sustained change in posture that was maintained throughout the episode, ending abruptly with complete resumption of purposeful motor activity. Characteristics of abrupt arrests were very different from those of quiet behavioral states or normal transitions into sleep. Each episode usually lasted for a short period, mostly less than a minute. Occasionally, gait disturbance lasting several seconds was observed immediately before episodes. Side-to-side rocking, without change in overall posture, frequently occurred several seconds after the start of the arrest.

Detailed observations of behaviors during EEG/EMG recordings found that abrupt arrests in narcoleptic mice occurred at timing of the direct transitions from wakefulness to REM sleep or during pre-REM phase immediately after a waking period. (The pre-REM phase shows EEG with high-amplitude spindle oscillations superimposed on non-REM sleep background, and these spindles are observed only during the transition phase immediately prior to REM sleep in wild-type mice.) The direct or very rapid transitions from wakefulness to REM sleep are the most prominent characteristic of EEG/EMG in narcoleptic mice, which were observed almost exclusively in the dark phase, and were never observed in wild-type mice.

Abrupt arrests in orexin −/− mice were ameliorated by systemic administration of clomipramine, an agent used for treatment of human cataplexy. While administration of caffeine, a psychostimulant used to treat excessive sleepiness in human narcolepsy, produce a mild exacerbation of abrupt arrest frequency.

These behavioral, electrophysiological, and pharmacological characteristics of abrupt arrests in narcoleptic mice suggest that these attacks are the counterpart of cataplexy observed in human narcolepsy patients. Cataplexy has been well known to be triggered by strong emotions such as laughing, anger, fear, surprise, and excitement in human narcolepsy patients. In narcoleptic mice, abrupt arrests are very often preceded by purposeful motor activity such as excited ambulation, grooming, burrowing, and climbing. Palatable foods such as chocolates can also trigger cataplexy. The dramatic increase in the number of abrupt arrests noted in the group-housed mice as compared with the individually-housed littermates suggests that social interaction may enhance this attack. Chasing, tail biting, and social grooming were often observed to immediately precede the attacks in the group-housed mice.

Only approximately one-third of human patients experience full loss of muscle tone causing collapse to the floor with the majority having partial cataplexy evidenced by jaw sagging, head bobbing, arm dropping, ptosis, or dysarthria (Parkes et al. 1974). Unambiguous full postural collapse was frequently observed in young orexin −/− mice, while adults tended to collapse onto their ventral surface at odd angles, suggesting some residual muscle tone. Cataplexy is not always instantaneous in human patients but can progress over several seconds with some patients experiencing gait disturbance known as “zig-zag walking”. Similarly, gait ataxia immediately preceding almost one third of abrupt arrests in narcoleptic mice.

Sleepiness and sleep attack: One of the prominent features of sleep/wakefulness patterns in narcoleptic mice was shortened durations of both wakefulness and NREM sleep in the dark phase, causing increased fragmentation of sleep/wake cycle. The shortened durations of wakefulness suggests sleepiness, which is the cardinal symptom of narcolepsy. OX2R −/− mice also showed sleep/wake fragmentation, while occurrence of REM sleep-related abnormalities was very rare as compared to orexin −/−, orexin/ataxin-3 or double orexin receptor knockout mice. This fragmentation was accompanied by statistically insignificant tendency toward reduced amounts of wakefulness and increased amounts of NREM sleep during the dark phase.

Presumptive excessive sleepiness in orexin −/− mice was also shown by detailed analyses of “gradual arrests” in orexin −/− and OX2R −/− mice, which can be interpreted analogous to “sleep attacks” in human narcolepsy. Gradual arrests typically began during quiet wakefulness and could be easily distinguished from the normal onset of resting behavior by lack of stereotypic preparation for sleep (e.g., nesting and/or assumption of a curled or hunched posture, with limbs drawn under the body) and a characteristic ratchet-like “nodding” of the head over a period of several seconds, with a transition to a collapsed posture. Gradual arrests in both orexin −/− and OX2R −/− mice resemble the sleep attacks in human narcolepsy. Unlike cataplexy, the gradual arrests are not associated with strong emotions or muscle atonia, and are reduced by psychostimulants such as amphetamines, modafinil, and caffeine.

Systemic administration of caffeine dose-dependently suppressed gradual arrests, while administration of an anticataplectic agent clomipramine did not affect the frequency of gradual arrests in both orexin −/− and OX2R −/− mice.

EEG/EMG recordings with simultaneous video-capture further differentiated gradual arrests from abrupt ones. As described above, abrupt arrests were accompanied by direct transition from wakefulness to REM sleep in narcoleptic mice. In contrast, EEG/EMG correlates of gradual arrests in narcoleptic mice invariably revealed the onset of attenuated muscle tone, but not atonia, and an EEG transition from wakefulness to NREM sleep. Gradual arrests were occasionally accompanied by apparent automatic behavior, which is continuation of semipurposeful motor activity after the onset of light sleep such as stereotypic chewing of food.

2.2 Characteristic of Each Model

2.2.1 Pepeo-Orexin Gene Knock Out (Orexin −/−) Mice

Conventional behavioral tests revealed that orexin −/− mice did not show any overt abnormalities during the light period, except mild decrease in food intake  (Chemelli et al. 1999). However, infrared video-recordings at the dark period to examine abnormalities on feeding behavior of orexin −/− mice at night, when mice are normally most active, found frequent periods of behavioral arrest. No serum electrolyte imbalance or hypoglycemia was observed in orexin −/− mice. Electroencephalograph/electromyographic (EEG/EMG) recordings from orexin −/− mice showed no evidence of epileptic seizure during the attack. Instead, these EEG/EMG recordings revealed abnormal intrusions of REM sleep-like episodes into wakefulness and fragmentation of sleep/wakefulness cycle (Chemelli et al. 1999). Markedly reduced latency to REM sleep and increase in REM sleep during the dark phase were also observed. These characteristics of orexin −/− mice were strikingly similar to characteristics human narcolepsy.

2.2.2 Orexin-Ataxin-3 Mice

Orexin levels in the CSF and postmortem brains of narcolepsy patients was dramatically decreased (Nishino et al. 2000; Peyron et al. 2000; Thannickal et al. 2000). However, in most cases of human narcolepsy, onset of symptoms occurs during adolescence. Based on the late onset, as well as a strong association of human narcolepsy with certain HLA alleles (Kadotani et al. 1998; Singh et al. 2013), it has been thought that narcolepsy result from acquired, selective autoimmune degeneration of orexin neurons.

In order to mimic the postnatal degeneration of orexin neurons in human narcolepsy, we generated transgenic mice (orexin/ataxin-3 transgenic mice) in which orexin neurons are ablated by orexinergic-specific expression of a truncated Machado-Joseph disease gene (ataxin-3) fragment with expanded polyglutamine stretch as a toxic trans gene (Hara et al. 2001). Number of orexin neurons decreased after their birth, and at 12 weeks of age, over 99 % of orexin neurons were lost. Orexin/ataxin-3 transgenic mice exhibited behavioral and electrophysiological defects that are essentially the same as those displayed by orexin −/− mice, including cataplexy-like behavioral arrests, direct transitions from wakefulness to REM sleep, shortened latency to REM sleep, fragmentation of sleep/wakefulness states, and increases in REM sleep time and duration in the dark period. However, the behavioral arrests seen in orexin/ataxin-3 transgenic mice typically began about 6 weeks of age, while arrests were observed in some orexin −/− mice earlier than 3 weeks of age. Thus, orexin/ataxin-3 transgenic mice have etiology and the course of disease similar to human narcolepsy. This model has been widely used as an useful narcoleptic model.

The fact that orexin/ataxin-3 transgenic mice exhibited similar phenotype with that of orexin −/− mice also revealed that orexin is the primarily important factor for the maintenance of the sleep/wakefulness state by orexin neurons, although orexin neurons produce other neurotransmitters such as glutamate, dynorphins, and neurotensin (Chou et al. 2001; Rosin et al. 2003; Furutani et al. 2013).

We utilized this mouse model of narcolepsy to rescue its narcoleptic phenotype in orexin/ataxin-3 transgenic mice by genetic (overexpression of orexin peptides throughout the brain) and pharmacological (i.c.v. administration of orexin A) means (Mieda et al. 2004), demonstrating that mice retain the ability to respond to orexin neuropeptides even if they lack endogenous orexin neurons and that a temporally regulated and spatially targeted secretion of orexins is not necessary to prevent narcoleptic symptoms. This finding suggests that orexin receptor agonists would be of potential value for treating human narcolepsy.

2.2.3 Orexin/Ataxin-3 Rats

Transgenic rats made with the same transgene as orexin/ataxin-3 transgenic mice was also generated (orexin/ataxin-3 transgenic rats) (Beuckmann et al. 2004). Similar to orexin/ataxin-3 mice, number of orexin neurons was gradually reduced after their birth, and after 17 weeks of age orexin in the hypothalamus became undetectable. Orexin/ataxin-3 transgenic rats showed a typical narcoleptic phenotype, with a decreased latency to REM sleep, increased REM sleep time in dark period, direct transitions from wakefulness to REM sleep, and a marked fragmentation of vigilance states. Brief episodes of muscle atonia and postural collapse resembling cataplexy were also observed while rats maintained the EEG characteristics of wakefulness, suggesting they were conscious during these episodes as human patients are during cataplexy.

2.2.4 OX2R-Deficient Mice

In infrared videotaping studies in the dark phase revealed that OX2R −/− mice also exhibited behavioral arrests, or cataplexy (Willie et al. 2003). However, its frequency was much less than in orexin −/− mice (31-fold lower frequency in OX2R −/− mice as compared to orexin −/− mice). OX2R −/− mice appeared to exhibit different type behavioral arrests with onsets that were more gradual in nature (gradual arrests). Orexin −/− mice and orexin/ataxin-3 mice also found to exhibit the gradual arrests with the frequency similar to OX2R −/− mice in addition to numbers of cataplexy-like, abrupt arrests.

OX2R −/− mice and orexin −/− mice are similarly affected by gradual arrests (“sleep attacks”) (Willie et al. 2003), but OX2R −/− mice show a lower degree of disrupted wakefulness compared with double receptor-deficient (OX1R −/−; OX2R −/−) mice (Willie et al. 2003; Sakurai 2007; Hondo et al. 2010). In particular, OX2R −/− mice are only mildly affected by cataplexy and direct transitions to REM sleep from awake states, as compared with orexin −/− mice and double receptor KO mice(Chemelli et al. 1999; Willie et al. 2003).

2.2.5 OX1R/OX2R Gene Double Knockout Mice

Orexin −/− mice and double receptor-knockout mice have the virtually same phenotype (Willie et al. 2003). Considering the fact that OX2R −/− mice are only mildly affected by cataplexy and direct transitions to REM sleep from awake states, these observations suggest that OX2R plays a pivotal role in maintain wakefulness, but OX1R has additional effects on sleep-wake regulation. The gating of REM sleep is likely to critically involve both receptors. The contributions of both receptors in the sleep/wakefulness states regulation are described in detail in our separate chapter (Mieda).

2.2.6 Orexin-TTA;TetO DTA Mice

In this model, diphtheria toxin A (DTA) was expressed in orexin neurons under control of the Tet-off system. Doxycycline removal from the diet of adult orexin-tTA;TetO DTA mice induce orexin neurodegeneration with 80 % cell loss after one week of tet-off condition, and resulted in apparent sleep/wakefulness fragmentation. However, cataplexy was only observed 14 d after the removal of doxycycline when less than 5 % of the orexin neurons remained. Cataplexy frequency increased for at least 11 weeks after doxycycline removal. These suggest that very low level of orexin could inhibit cataplexy. Body weight was reported to be increased without a change in food consumption, as in human narcolepsy and orexin/ataxin-3 mice.