Is Low Histamine a Fundamental Cause of Sleepiness in Narcolepsy and Idiopathic Hypersomnia?

Abstract

THE NEUROBIOLOGY OF SLEEPINESS IS STILL VERY MUCH A MYSTERY. NARCOLEPSY WITH CATAPLEXY IS CAUSED BY A LOSS OF THE NEURONS PRODUCING the hypocretin/orexin neuropeptides, but for other disorders such as obstructive sleep apnea and idiopathic hypersomnia, the neural basis of sleepiness remains unknown. Furthermore, it is still unclear whether the sleepiness of narcolepsy is a direct consequence of hypocretin deficiency or whether low hypocretin tone reduces activity in other wake-promoting systems. In this issue of Sleep, Nishino, Kanbayashi and their colleagues report that levels of the wake-promoting neurotransmitter histamine are often low in patients with narcolepsy and idiopathic hypersomnia.1,2 These are the first measurements of histamine in patients with sleep disorders, and they may lead to a better understanding of sleepiness in general. Histamine is one of the most well-established wake-promoting neurotransmitters. About 25 years ago, Panula and Watanabe independently discovered that neurons of the tuberomammillary nucleus are the only source of histamine in the brain.3,4 Soon after, it became clear that these histamine-producing neurons are very active during wakefulness and inactive during sleep, especially REM sleep.5–7 Extracellular concentrations of histamine vary with a clear diurnal rhythm, with high levels during the wake period and low levels during sleep.8–10 Histamine-deficient mice have less wakefulness at the beginning of the usual active period, and mice lacking H1 receptors have fewer awakenings from sleep.11,12 Most importantly, anti-histamines (H1 receptor antagonists) promote sleep across species including humans, leading many researchers to hypothesize that histamine is a key wake-promoting neurotransmitter.13–15 What has been lacking is any information on histamine signaling in clinical sleep disorders. The papers from Nishino, Kanbayashi and colleagues now provide the first perspectives on histamine in people with sleep disorders. In a study of patients with narcolepsy compared to age-matched, healthy controls, Nishino found that on average, narcolepsy patients with low hypocretin levels have lumbar CSF histamine concentrations about 44% of normal.1 In a companion paper, Kanbayashi reports that not only were histamine levels low in patients with narcolepsy but also in patients with idiopathic hypersomnia.2 This later finding is particularly novel as (by definition!) nothing is known about the causes of sleepiness in idiopathic hypersomnia. Furthermore, this histamine deficiency may be specific to narcolepsy and idiopathic hypersomnia as patients with obstructive sleep apnea had relatively normal histamine levels.2 Both studies found some of the lowest histamine levels in narcolepsy patients with low hypocretin, and as hypocretin excites the tuberomammillary neurons, one might predict that low histamine is simply a consequence of reduced hypocretin signaling. However, the two studies are not entirely consistent: In narcolepsy patients with normal hypocretin levels, Kanbayashi describes severely reduced histamine levels (29-34% of normal), but Nishino reports a much more modest reduction (68-78% of normal). As histamine levels are low in patients with idiopathic hypersomnia who have normal hypocretin signaling, it seems likely that factors other than hypocretin deficiency underlie the reductions in histamine. These observations have several clinical and scientific implications. Animal studies have demonstrated the importance of histamine in promoting wake, and these new clinical studies suggest that reduced histamine may contribute to the sleepiness of narcolepsy and idiopathic hypersomnia. In fact, Kanbayashi hypothesizes that low histamine may be specific to “CNS hypersomnias” as histamine levels were normal in patients with OSA. If so, then perhaps histamine signaling is low in other central causes of hypersomnia such as Kleine-Levin Syndrome or perhaps multi-system atrophy. Still, one should not rush off to treat patients with drugs that enhance histamine signaling because neither study found that histamine levels correlated with MSLT sleep latencies. In fact, histamine levels may be low as a consequence of sleepiness, and much work remains to be done to determine whether reduced histamine signaling contributes to the sleepiness of these disorders. In addition, these observations are the first indication of a neurochemical abnormality in well-characterized patients with idiopathic hypersomnia. Reduced histamine signaling is an attractive explanation for the sleepiness of idiopathic hypersomnia as many individuals with this disorder have great difficulty rousing from sleep in the morning, and one of the most striking abnormalities in mice lacking histamine is their reduction in wake at the beginning of the usual active period.11 Patients with idiopathic hypersomnia could have a reduction in the number of histamine-producing neurons, but it seems more likely that the tuberomammillary neurons are pathologically underactive as up to one quarter of idiopathic hypersomnia patients spontaneously improve over time.16 Studies of less well-characterized idiopathic hypersomnia patients in the 1980s suggested some disruption of dopamine signaling, but those results were controversial.17–19 This new finding of reduced histamine signaling may be the first useful toehold in understanding the neurochemistry of idiopathic hypersomnia. These observations are clinically novel, yet both studies have some technical limitations. One major concern is that histamine levels were quite variable, differing substantially within groups and overlapping considerably between patients and controls. In addition, the levels of histamine in these lumbar CSF specimens were only about one-quarter of those reported in animal studies that measured extracellular histamine concentrations in brain.8–10 These variable and presumably low histamine levels were probably not due to technical problems with the histamine assay as it appears to have great sensitivity and low inter-assay variability. More likely, histamine levels were low because histamine has a short half life (probably about 5-30 min),20 and it can take hours for ventricular CSF to make its way down to the lumbar sac. This slow transit and subsequent stasis of lumbar CSF may limit the amount one can learn about histamine signaling when studying levels so far from the source. These studies are an excellent beginning, but much more needs to be done before we can fully understand the connections between low histamine signaling and sleepiness. If histamine metabolites such as tele-methylhistamine are more stable than histamine itself, then lumbar CSF might be more informative. It may also be possible to reduce variability by using a within-subject design to determine the effects of drugs or sleep deprivation using continuous or frequent lumbar sampling. To circumvent the limitations of lumbar CSF, one could measure histamine in cisternal CSF or in microdialysis samples under certain clinical circumstances. Alternatively, PET or other neuroimaging techniques may allow one to measure central histamine tone when probed by drugs binding to histamine receptors.21 As these techniques improve our ability to assay and understand central histamine signaling, we can then address whether enhancement of histamine signaling is an effective treatment for people with narcolepsy, idiopathic hypersomnia, and other sleep disorders.