Issue Navigator

Volume 13 No. 02
Earn CME
Accepted Papers

Scientific Investigations

Normal Morning Melanin-Concentrating Hormone Levels and No Association with Rapid Eye Movement or Non-Rapid Eye Movement Sleep Parameters in Narcolepsy Type 1 and Type 2

Maren Schrölkamp, MSc1,2; Poul J. Jennum, MD, DMedSc3; Steen Gammeltoft, MD, DMedSc1; Anja Holm, PhD1; Birgitte R. Kornum, PhD1; Stine Knudsen, MD, PhD3,4
1Molecular Sleep Laboratory, Department of Clinical Biochemistry, Rigshospitalet, Glostrup, Denmark; 2FU-Berlin, Faculty Biology, Chemistry, Pharmacy, Takustr, Berlin, Germany; 3Danish Center for Sleep Medicine, University of Copenhagen, Rigshospitalet, Glostrup, Denmark; 4Norwegian Centre of Expertise for Neurodevelopmental Disorders and Hypersomnias (NevSom), Oslo University Hospital, Ullevål, Norway


Study Objectives:

Other than hypocretin-1 (HCRT-1) deficiency in narcolepsy type 1 (NT1), the neurochemical imbalance of NT1 and narcolepsy type 2 (NT2) with normal HCRT-1 levels is largely unknown. The neuropeptide melanin-concentrating hormone (MCH) is mainly secreted during sleep and is involved in rapid eye movement (REM) and non-rapid eye movement (NREM) sleep regulation. Hypocretin neurons reciprocally interact with MCH neurons. We hypothesized that altered MCH secretion contributes to the symptoms and sleep abnormalities of narcolepsy and that this is reflected in morning cerebrospinal fluid (CSF) MCH levels, in contrast to previously reported normal evening/afternoon levels.


Lumbar CSF and plasma were collected from 07:00 to 10:00 from 57 patients with narcolepsy (subtypes: 47 NT1; 10 NT2) diagnosed according to International Classification of Sleep Disorders, Third Edition (ICSD-3) and 20 healthy controls. HCRT-1 and MCH levels were quantified by radioimmunoassay and correlated with clinical symptoms, polysomnography (PSG), and Multiple Sleep Latency Test (MSLT) parameters.


CSF and plasma MCH levels were not significantly different between narcolepsy patients regardless of ICSD-3 subtype, HCRT-1 levels, or compared to controls. CSF MCH and HCRT-1 levels were not significantly correlated. Multivariate regression models of CSF MCH levels, age, sex, and body mass index predicting clinical, PSG, and MSLT parameters did not reveal any significant associations to CSF MCH levels.


Our study shows that MCH levels in CSF collected in the morning are normal in narcolepsy and not associated with the clinical symptoms, REM sleep abnormalities, nor number of muscle movements during REM or NREM sleep of the patients. We conclude that morning lumbar CSF MCH measurement is not an informative diagnostic marker for narcolepsy.


Schrölkamp M, Jennum PJ, Gammeltoft S, Holm A, Kornum BR, Knudsen S. Normal morning melanin-concentrating hormone levels and no association with rapid eye movement or non-rapid eye movement sleep parameters in narcolepsy type 1 and type 2. J Clin Sleep Med. 2017;13(2):235–243.


Narcolepsy is a neurological disorder that affects approximately 1 in every 2,000 individuals.1 It is characterized by symptoms of sleep-wake dysregulation (excessive daytime sleepiness, multiple sleep/wake stage shifts,2 multiple nightly awakenings3) and of rapid eye movement (REM) sleep dysregulation (early/sleep onset REM periods [(SOREMs]), cataplexy (muscle atonia triggered by emotions), hypnagogic hallucinations, sleep paralysis, and REM sleep behavior disorder).4,5 We have also shown motor/tonus dysregulation and autonomic dysfunction not only during REM sleep but also during non-rapid eye movement (NREM) sleep in narcolepsy.5,6


Current Knowledge/Study Rationale: Hypocretin/orexin neurons interact with the melanin-concentrating hormone (MCH) neurons also located in the hypothalamus. We speculated whether altered MCH function could contribute to the symptoms and sleep abnormalities of narcolepsy.

Study Impact: Our study is the first to systematically collect morning cerebrospinal fluid from patients of both narcolepsy subtypes (narcolepsy type 1, narcolepsy type 2) and healthy controls and measure MCH levels in the samples. We do not find altered morning MCH levels in narcolepsy, regardless of narcolepsy subtype or HCRT-1 status.

Narcolepsy type 1 (NT1) is according to the latest International Classification of Sleep Disorders, Third Edition (ICSD-3)4 classified as presence of excessive daytime sleepiness, clear cataplexy, and a mean sleep latency ≤ 8 min and two or more SOREMs at the Multiple Sleep Latency Test (MSLT) and/or excessive daytime sleepiness and low levels of cerebrospinal fluid (CSF) hypocretin-1 (HCRT-1) (< 110 pg/mL). Although the majority of NT1 cases have both clear cataplexy and low CSF HCRT-1 level, importantly, a fraction of ICSD-3 classified NT1 cases actually consist of patients with clear cataplexy but normal (or unknown) HCRT-1 level, or patients with low HCRT-1 level but no cataplexy. Narcolepsy type 2 (NT2) is classified as the presence of excessive daytime sleepiness, no clear cataplexy, a mean sleep latency ≤ 8 min and one or fewer SOREM at the MSLT and, if measured, normal levels of CSF HCRT-1 (> 200 pg/mL).4 NT1 is caused by the massive loss of hypocretin neurons in the lateral hypothalamus,7,8 which is associated with low or undetectable HCRT-1 levels in the CSF found in 90% to 100% of NT1 cases.3,9,10 Other neurons, for example nearby MCH neurons, are not lost, suggesting that the hypocretin neuron loss is selective.7,8 Animal models show that, in addition to some phasic activity during REM sleep, hypocretin neurons are primarily active during wakefulness,11 acting to regulate and excite the wake-promoting and REM sleep-inhibiting side of the sleep-wake regulatory neuronal network in the brainstem.12 Hypocretin deficiency in NT1 could therefore explain many of the symptoms found in that patient group.5

However, several pathophysiological matters in narcolepsy remain unexplained, such as: reasons for varying disease severity/symptoms in patients with similar CSF HCRT-1 levels3,9; the often extremely fragmented night-time sleep13; abnormal muscle tone and increased number of muscle movements not only in REM but also in NREM sleep5; and the 10% of NT1 patients and all narcolepsy type 2 (NT2) patients who have normal CSF HCRT-1 levels.3,9 The pathogenesis of NT2 is unclear and is, mainly based on differing findings about the association of HLA, believed to be a more pathophysiologically heterogeneous group than the NT1 group, in which 90% to 100% are HLA-DQB1*0602-positive.3,14 However, hypocretin neurons strongly interact with multiple neurons of the neuronal sleep-wake and REM/NREM sleep networks, so other neurochemical alterations in addition to hypocretin deficiency are probably involved in narcolepsy, as has been suggested for histamine.15,16

MCH is an inhibitory neuropeptide produced by neurons mainly located in the zona incerta and lateral hypothalamus. They are closely intermingled with the hypocretin neurons and both strongly project to the same brainstem sleep/wake and REM/NREM sleep-regulating nuclei.12 For several years, it has been strongly believed that MCH neurons are fundamental to sleep induction and regulation of REM sleep.

Studies in different rodent models have given some insight into the relationship of MCH neuron activity and sleep regulation. They are almost only active during sleep and fire maximally during REM sleep in a reciprocal pattern to hypocretin neuron firing.17 If stimulated during sleep, MCH neurons disinhibit the REM sleep-activating brainstem nuclei, resulting in conversion of NREM to REM sleep and prolongation of REM sleep.18,19 MCH neurons have recently also been shown to play a role in NREM regulation, whereby MCH neuron stimulation during wakefulness inhibits the arousal/wakefulness brainstem nuclei resulting in NREM sleep.20 Although this finding was not reproduced by another research group, they nevertheless confirmed the role of MCH in NREM regulation, because MCH neuron ablation resulted in a clear increase in wakefulness and a significant reduction in amount of NREM sleep (but not REM sleep).19

MCH neurons inhibit hypocretin neurons locally.21 HCRT-1 peptide, when applied to electrophysiological slice preparations, has previously been shown to activate MCH neurons.22 In contrast, a recent optogenetic study convincingly showed that hypocretin neuron stimulation only activates a small subset of MCH neurons but inhibits the majority (> 90%) of MCH neurons, probably through local gamma-aminobutyric acid (GABA)-ergic interneurons.19

Based on the aforementioned concepts, it can be speculated that (1) missing input from the hypocretin neurons in hypocretin-deficient NT1 (and possibly to some extent in NT2 with partial hypocretin neuron loss) could result in the chronic alteration (and possibly activation) of MCH secretion, and (2) altered MCH neuron activity could be a contributory factor in narcolepsy. However, these ideas were not supported by a previous report of non-differential CSF MCH levels and no symptom/sleep parameter association in NT1 patients, idiopathic hypersomnia patients, and neurological controls.23 As MCH brain concentrations in humans are generally low during the day and peak at sleep onset and during sleep,24 these previous data23 were limited by afternoon/evening CSF collection, when MCH secretion is expected to be low, thereby risking a false-negative result.

The delay of several hours observed in lumbar CSF neuropeptide levels compared with expected ventricular CSF levels24,25 implies that measuring morning lumbar CSF levels should better reflect night-time (sleep) ventricular CSF levels. We therefore aimed to establish whether morning MCH levels collected closer to the expected main night-time secretion period are higher in narcolepsy patients and associated with the phenotype(s) in a larger cohort of NT1 and NT2 patients and healthy controls.



After obtaining ethical approval and written informed consent, 41 Danish patients with confirmed NT1 and low HCRT-1 levels (CSF HCRT-1 ≤ 135 pg/mL), 6 Danish patients with NT1 with cataplexy but normal CSF HCRT-1 levels, and 10 Danish non-cataplectic NT2 patients with normal CSF HCRT-1 levels (CSF HCRT-1 > 200 pg/mL) were consecutively included at the Danish Center for Sleep Medicine, Glostrup Hospital, Denmark (patients reported in previous studies, see demographics in Table 1).3,5 Narcolepsy was diagnosed using the criteria of ICSD-3.4 Presence and onset of daytime and night-time narcoleptic symptoms were evaluated by semi-structured interview, questionnaires generally about sleep, hypnagogic hallucinations, sleep paralysis, and disrupted night sleep, and validated questionnaires especially for estimates of excessive daytime sleepiness,26 cataplexy,27 and REM sleep behavior disorder,28 as previously reported.3,5 Twenty asymptomatic healthy volunteer controls, matched with the patients by age, sex, and body mass index were recruited through, or because their subjective sleep or neurological complaints had been rejected objectively (controls reported in a previous study3). In the 20 healthy controls, CSF and plasma samples were available from 10 people, whereas only plasma or CSF was available from 5 people. Therefore, a total of 15 healthy control plasma samples (referred to as the plasma cohort, see Table 2) and 15 healthy control CSF samples (referred to as the CSF cohort, see Table 2) were analyzed.

Demographic, clinical, and paraclinical variables of the narcolepsy cohort.


table icon
Table 1

Demographic, clinical, and paraclinical variables of the narcolepsy cohort.

(more ...)

Cerebrospinal fluid melanin-concentrating hormone levels and plasma melanin-concentrating hormone levels in narcolepsy patients and normal controls.


table icon
Table 2

Cerebrospinal fluid melanin-concentrating hormone levels and plasma melanin-concentrating hormone levels in narcolepsy patients and normal controls.

(more ...)

Sample collection and CSF HCRT-1 measurement: Lumbar punctures and blood samples were performed in the morning between 07:00 and 10:00, and CSF and plasma were stored on ice for 15 to 30 min and then frozen at −80°C until analysis. Participants were instructed to fast before the collection. CSF HCRT-1 peptide levels were measured in duplicate using a radioimmunoassay from Phoenix Pharmaceuticals Inc., USA (for details, see study by Knudsen et al.3).

MCH Measurements

MCH peptide concentrations were measured in CSF and plasma using a radioimmunoassay from Phoenix Pharmaceuticals Inc., USA. The analyses were carried out according to the manufacturer's instructions, but with minor modifications: 480 μL of either CSF or plasma, each in duplicate, were dried using a SpeedVac (Heto-Holten A/S) and stored at 4°C until analysis. Dried samples were resuspended in 100 μl of radioimmunoassay (RIA)-buffer and vortexed on the day of the experiment. One hundred μL of rabbit anti-peptide serum was added and samples were stored at 4°C overnight. One hundred μL 125iodine-labeled MCH was added, samples vortexed and stored overnight at 4°C. Tracer solution was always adjusted to contain 10,000 cpm/100 μL. The following day, 100 μL of goat anti-rabbit serum and 100 μL of normal rabbit-serum, both included in the kit, were added and after vortexing, samples were left at room temperature for 90 min. Before centrifugation at 4°C, 1700×g for 25 min, 500 μL RIA-buffer was added to each sample. Supernatants were then aspirated and counts per minute were determined using a γ-counter. The standard curve was performed in triplicate and fitted with a four-parameter model. Values below the detection limit of the assay were left out. To determine interassay variation, a CSF reference (mean of 44 ± 22.1 pg/mL) was included in all three assay kits and used to normalize the data. The interassay coefficient of variation was 11.3% based on repeated samples, and 0.5% based on the calculated Hill slopes. This is similar to what has been reported for other RIAs.29,30

Statistical Analyses

SPSS version 22 was used for statistical analyses. Results are reported as mean ± standard deviation (SD), unless otherwise stated. Values of p < 0.05 were considered statistically significant. Data presented in Table 1 were analysed with the independent-samples t-test for continuous variables or the Mann-Whitney U test for ordinal variables. Equality of variances was tested using the F-test, and when necessary Welch correction was applied to the t-test. Data presented in Table 2 were analyzed by one-way analysis of variance with Tukey post hoc comparison between groups. Equality of variances was tested using Levenes test, and when necessary Kruskal-Wallis test was used instead of the analysis of variance. We excluded three outlier values of plasma MCH (very high levels of 335, 826, and 1,194 pg/mL) as they differed way more than 3.3 SD from the mean; two of those outliers were in the NT1 group and one in the NT2 group. Data presented in Table 3 were analyzed by binary logistic regression models to ascertain the effects of age, body mass index, sex, and low HCRT-1 and MCH level in CSF on the likelihood that participants have one of the following symptoms: cataplexy, hypnagogic hallucinations, sleep paralysis or clinical REM sleep behavior disorder (RBD). The Box-Tidwell procedure was used to test whether the continuous independent variables were linearly related to the logit of the dependent variable. This was indeed the case. Because we only had access to MSLT and PSG variables in some of the NT1 and NT2 patients, these variables were analyzed separately by linear regression (Table 4). Continuous data were natural log-transformed, if necessary, to meet the criteria for linearity, normal distribution, and variance homogeneity of residuals. This was necessary for age and all PSG variables.

Correlations between cerebrospinal fluid melanin-concentrating hormone level and categorical clinical rapid eye movement sleep manifestations and paraclinical rapid eye movement sleep polysomography/Multiple Sleep Latency Test characteristics.


table icon
Table 3

Correlations between cerebrospinal fluid melanin-concentrating hormone level and categorical clinical rapid eye movement sleep manifestations and paraclinical rapid eye movement sleep polysomography/Multiple Sleep Latency Test characteristics.

(more ...)

Correlations between cerebrospinal fluid melanin-concentrating hormone level and continuous clinical rapid eye movement sleep manifestations and paraclinical rapid eye movement sleep polysomnography/Multiple Sleep Latency Test characteristics.


table icon
Table 4

Correlations between cerebrospinal fluid melanin-concentrating hormone level and continuous clinical rapid eye movement sleep manifestations and paraclinical rapid eye movement sleep polysomnography/Multiple Sleep Latency Test characteristics.

(more ...)


Table 1 shows the narcolepsy patient demographics, clinical daytime and night-time symptoms, and MSLT data.

We have previously shown that hypocretin deficiency (low HCRT-1 levels), and not cataplexy, is the primary predictor for the core clinical manifestations and number of short and long muscle movements in REM and NREM sleep of narcolepsy,5 so we divided the patients into two groups with respect to their HCRT-1 level: one with low HCRT-1 levels (only NT1 patients), and the other with normal HCRT-1 levels (6 NT1 patients with cataplexy; 10 NT2 patients without cataplexy).

Narcolepsy patients with low HCRT-1 levels were significantly more likely to be carriers of the HLA-DQB1*0602-allele, to be sleep/wake disturbed (more awakenings at night) and to present abnormal daytime and night-time REM sleep manifestations (cataplexy prevalence and frequency, hypnagogic hallucination prevalence, sleep paralysis prevalence, number of SOREMs on MSLT) than the group of narcolepsy patients with normal HCRT-1 levels.

There were no significant differences in age, sex, body mass index, disease duration, subjective sleepiness score (Epworth Sleepiness Scale [ESS]), years since onset of cataplexy, presence of sleep paralysis, or hypnagogic hallucinations between the two groups. There was a nearly significant trend (p = 0.06) toward a higher prevalence of clinical RBD symptoms in patients with low HCRT-1 level.

Table 2 shows the characteristics of the narcolepsy patients and healthy controls with available CSF and/or plasma, together with CSF MCH level (CSF cohort) and plasma MCH level (plasma cohort) (illustrated in Figure 1).

MCH level in CSF and plasma in narcolepsy with low HCRT-1, with normal HCRT-1, and healthy controls.

(A) CSF MCH levels in patients with low HCRT-1 levels (n = 38), narcolepsy patients with normal CSF HCRT-1 levels (n = 15) and healthy controls (n = 15). (B) Plasma MCH levels in patients with low HCRT-1 (n = 29), narcolepsy patients with normal CSF HCRT-1 levels (n = 12) and healthy controls (n = 15). No significant differences are seen (one-way analysis of variance). CSF = cerebrospinal fluid, HCRT-1 = hypocretin-1, MCH = melanin-concentrating hormone.


Figure 1

MCH level in CSF and plasma in narcolepsy with low HCRT-1, with normal HCRT-1, and healthy controls.

(more ...)

The MCH levels in the CSF and plasma cohorts did not significantly differ between the group of patients with low HCRT-1 (36.7 ± 12.5 pg/mL CSF; 40.0 ± 19.3 pg/mL plasma), the group of narcolepsy patients with normal HCRT-1 levels (42.4 ± 24.5 pg/mL CSF; 32.3 ± 16.3 pg/mL plasma), or the group of healthy controls (36.2 ± 14.9 pg/mL CSF; 38.3 ± 22.1 pg/mL plasma) (p = 0.48 CSF; p = 0.52 plasma), (Figure 1). We also performed an analysis of the patients divided into NT1 and NT2 following the ICSD-3 classification,4 and also here there were no significant differences in MCH levels between the two groups in either CSF (NT1: 36.4 ± 12.8 pg/mL and NT2: 48.8 ± 28.0 pg/mL) nor plasma (NT1: 37.4 ± 19.2 pg/mL and NT2: 39.4 ± 16.2 pg/mL). Therefore, our results did not support our hypothesis of different levels of morning lumbar CSF MCH in narcolepsy patients.

None of the three groups (narcolepsy low HCRT-1 level, narcolepsy with normal HCRT-1 level, and healthy controls) had significant correlations (Pearson) between levels of CSF HCRT-1 and CSF MCH (Figure 2A) or between levels of CSF HCRT-1 and plasma MCH (Figure 2B). Moreover, we found no correlation between CSF MCH and plasma MCH levels in any of the groups or when combining all samples (Pearson correlation, r = 0.04, p = 0.79, all samples) (Figure 2C). Hence, these results do not support the idea of a simple, direct relationship between CSF HCRT-1 and MCH levels either, at least when measured at the lumbar level. Moreover, measurements of plasma MCH level also seem to be uninformative.

Correlations of CSF or plasma MCH level and CSF HCRT-1 level.

(A) Correlation between CSF MCH and CSF HCRT-1 level in narcolepsy patients (all NT1+NT2 patients combined, gray triangles, n = 53, r2 = 0.01, p = 0.42) and healthy controls (black triangles, n = 15, r2 = 0.12, p = 0.21). Pearson correlations were performed. CSF = cerebrospinal fluid; HCRT-1 = hypocretin-1; MCH = melanin-concentrating hormone. (B) Correlation between plasma MCH and CSF HCRT-1 level in narcolepsy patients (all NT1+NT2 combined, gray triangles, n = 41, r2 = 0.01, p = 0.60) and healthy controls (black triangles, n = 15, r2 = 0.05, p = 0.44). (C) Correlation between plasma MCH and CSF MCH levels in patients with low CSF HCRT-1 levels (n = 29, r2 = 0.13, p = 0.07); narcolepsy patients with normal CSF HCRT-1 levels (n = 12, r2 = 0.35, p = 0.06); and healthy controls (n = 15, r2 = 0.10, p = 0.36). CSF = cerebrospinal fluid, HCRT-1 = hypocretin-1, MCH = melanin-concentrating hormone.


Figure 2

Correlations of CSF or plasma MCH level and CSF HCRT-1 level.

(more ...)

Tables 3 and 4 show the results of the analysis of CSF MCH levels with respect to the demographic data, clinical REM manifestations, and PSG and MSLT REM sleep parameters of the narcolepsy patients.

There was a nonsignificant trend for correlation between high CSF MCH levels and the presence of hypnagogic hallucinations (p = 0.074) and low CSF MCH levels and presence of a SOREM on the PSG (p = 0.080). However, no significant correlations were found between CSF MCH levels and demographic data (body mass index, age, sex, presence of hypocretin deficiency), narcolepsy symptoms (disease duration, ESS duration (years passed since onset of ESS), ESS score, presence of cataplexy, cataplexy frequency, presence of sleep paralysis, presence of clinical REM sleep behavior disorder symptoms) or REM sleep parameters (PSG REM latency, REM sleep as a percentage of total sleep time, number of short and long muscle movements during REM sleep, or number of MSLT SOREMs). We developed a similar multivariate general linear model of CSF MCH compared with number of short and long muscle movements during NREM sleep and, once again, found no significant associations (data not shown).

To summarize, our results generally do not show a relationship between morning lumbar CSF MCH levels and clinical symptoms, REM sleep parameters, or number of muscle movements during REM or NREM sleep in narcolepsy.


The current study was designed to evaluate whether morning CSF levels of the NREM and REM activating neuropeptide, MCH, differ and are associated with disease parameters in human narcolepsy, with an additional focus on the association with ICSD-3 narcolepsy subtypes (low versus normal HCRT-1 level in the CSF). A close association exists between hypocretin deficiency and the muscle tonus abnormalities and REM sleep symptoms of narcolepsy, mainly cataplexy and REM sleep behaviour disorder.5,9 This was confirmed in the current study (Table 3).

We show for the first time that MCH levels collected by lumbar puncture in the morning do not differ between the two narcolepsy subtypes or compared with healthy controls. Moreover, MCH levels show no apparent association with most symptoms, REM or NREM sleep abnormalities of narcolepsy regardless of the ICSD-3 subtype. A trend was seen toward significant associations between MCH level and hypnagogic hallucinations and the presence of a SOREM on the PSG, but the effect size was very small. We also show that the MCH level, in contrast to the HCRT-1 level,5 is not associated with the clinical or motor/tonus abnormalities of RBD in narcolepsy, which is a novel finding. Our study included normal healthy controls, so we can for the first time report that the MCH levels detected are indeed normal in narcolepsy.

In accordance with our current findings, a single previous study by Peyron et al.23 reported similar CSF MCH levels and no association with clinical or sleep parameters in 16 patients with NT1, 6 patients with idiopathic hypersomnia, 2 patients with traumatic brain damage, and 14 diverse neurological controls. This study only included CSF collected during the afternoon/evening, in contrast to the current study, in which all samples were collected following a standard procedure in the morning. Moreover, unlike in our study, they did not include narcolepsy patients with normal CSF HCRT-1 levels. We here show that MCH levels from morning lumbar CSF samples have no diagnostic value in narcolepsy. Our finding that this is regardless of narcolepsy subtype is novel. Furthermore, our finding of normal morning CSF MCH levels, taken together with the similar results of afternoon/evening CSF samples of Peyron et al.,23 indicates that this is also independent of sample time collection. However, as shown by Blouin et al.,24 human brain MCH levels are highest at sleep onset/during sleep, so between group differences could exist at not-yet-measured other times of the day or night. Our study also includes data on peripheral MCH levels, and we find that these do not correlate with CSF levels of MCH. This is likely due to peripheral expression of MCH.31

Postmortem studies of narcolepsy have revealed normal numbers of MCH neurons, regardless of narcoleptic sub-type.8,32 However, in general, neuropeptide neuron number and secretion levels are not necessarily correlated. In animal studies, MCH neurons have been shown to exert their inhibitory effect partly by co-expression/secretion of GABA along with MCH.18 Moreover, there is mutual inhibition between MCH and hypocretin neurons, but, equally importantly perhaps, there is mutual inhibition between MCH neurons and the downstream arousal promoting histaminergic neurons of the tuberomammillary nucleus.18,33 Much higher number of histamine neurons have recently been found in NT1 patients,15,16 suggesting a possible compensatory upregulation due to lack of excitation caused by hypocretin deficiency (or alternatively an increased number of histamine neurons as part of the pathogenesis of hypocretin neuron cell death). We may speculate that the combined increased histaminergic tone and reduced MCH tone (due to histamine inhibition) during sleep could contribute to the fragmented sleep of narcolepsy. However, when measured at the lumbar level, CSF histamine levels in patients with narcolepsy were not higher than in controls3437 implying that, in spite of greater numbers of histamine-producing neurons, these are not able to compensate effectively for the lack of hypocretin stimulation. An alternative theory is that MCH neurons that become disinhibited due to hypocretin deficiency produce greater inhibition of histaminergic neurons. Parallel (preferably ventricular) CSF studies of several neuropeptides, including those downstream of hypocretin and MCH, would therefore be valuable.

Our study had some limitations. Five patients had short disease duration of up to 1 y, but in the majority of our cohort, as in the study by Peyron et al.,23 the mean disease duration was more than 10 y. We did not find any significant difference between MCH levels in samples from the five recent-onset patients and our remaining patient cohort, although this needs to be examined in a larger cohort. It is therefore still possible that patients closer to disease onset do initially present elevated/altered CSF MCH levels that later return to normal through the action of adaptive or compensatory neurochemical mechanisms. Recall bias should always be considered, and cannot completely be excluded regarding the subjective symptom reporting in some cases for example those with long disease duration. However, we did not see an effect of disease duration on MCH levels and objective PSG/MSLT sleep parameters. As previously stated, we find it plausible that our lumbar samples collected in the morning reflect the previous night's MCH secretion, because lumbar neuropeptide levels are generally thought to represent the previous 3 to 6 h, approximately, of neuropeptide release.24 This is supported by the observation of a delayed neuropeptide (HCRT-1) peak of several hours in the CSF collected at the lumbar level in humans25 compared with the ventricular CSF peak in squirrel monkeys.38 However, this contrasts with a study by Baumann et al.,39 which found no significant ventriculolumbar CSF HCRT-1 gradient in patients with traumatic brain damage nor in controls. That study was mainly concerned with the comparison of groups with ventricular or lumbar measurements, and only two patients in the study had simultaneous ventricular and lumbar measurements.39 Human MCH levels in the amygdala are known to be highest during the night, low during the day, and markedly increased during and after eating.24 Therefore, we cannot exclude the possibility that the generally low CSF MCH levels observed in our fasting patients and controls in morning CSF samples reflect MCH secreted in the morning and not during the night. Peyron et al.23 do report nominally higher MCH levels in their (supposedly nonfasting) afternoon/ evening samples. However, quantitative differences in MCH levels between laboratories are more likely to reflect interassay variability of the MCH RIA.

In summary, the current study reports the first parallel measurements of morning lumbar CSF MCH and CSF HCRT-1 levels in both ICSD-3 subtypes of narcolepsy compared with healthy controls. The main finding is that the morning MCH level is normal and not associated with the clinical symptoms, or REM and NREM sleep abnormalities of narcolepsy, regardless of the disease subtype (low or normal HCRT-1 levels). We therefore conclude that morning lumbar CSF MCH levels are not a useful additional diagnostic tool in narcolepsy, or in the subtyping of narcolepsy.


This was not an industry supported study. Dr. Kornum has received research support from UCB Pharma. Dr. Jennum has received research support from UCB Pharma. The other authors have indicated no financial conflicts of interest. The work was performed at the Danish Center for Sleep Medicine, and Molecular Sleep Laboratory, Department of Clinical Biochemistry, Rigshospitalet, Nordre Ringvej, Glostrup, Denmark.



cerebrospinal fluid


Epworth Sleepiness Scale


gamma-aminobutyric acid




narcolepsy type 1


narcolepsy type 2


melanin-concentrating hormone


Multiple Sleep Latency Test


non-rapid eye movement




REM sleep behavior disorder


rapid eye movement




sleep onset REM periods


Author contributions: SK, BRK, MS, SG, and PJ designed the study. MS and BRK measured and analysed the MCH levels. SG and SK analyzed the hypocretin data. SK included and characterized the patients and controls. PJJ analysed the sleep investigation data. MS, BRK, SK, and AH analyzed the data. SK, BRK and MS wrote the manuscript. PJJ, SG, and AH revised the manuscript.



Mignot E. Genetic and familial aspects of narcolepsy. Neurology. 1998;50 2 Suppl 1:S16–S22. [PubMed]


Sorensen GL, Knudsen S, Jennum P. Sleep transitions in hypocretin-deficient narcolepsy. Sleep. 2013;36(8):1173–1177. [PubMed Central][PubMed]


Knudsen S, Jennum P, Alving J, Sheikh SP, Gammeltoft S. Validation of the ICSD-2 criteria for CSF hypocretin-1 measurements in the diagnosis of narcolepsy in the Danish population. Sleep. 2010;33(2):169–176. [PubMed Central][PubMed]


American Academy of Sleep Medicine. International Classification of Sleep Disorders. 3rd ed. Darien, IL: American Academy of Sleep Medicine; 2014.


Knudsen S, Gammeltoft S, Jennum PJ. Rapid eye movement sleep behaviour disorder in patients with narcolepsy is associated with hypocretin-1 deficiency. Brain. 2010;133(Pt 2):568–579. [PubMed]


Sorensen GL, Knudsen S, Petersen ER, et al. Attenuated heart rate response is associated with hypocretin deficiency in patients with narcolepsy. Sleep. 2013;36(1):91–98. [PubMed Central][PubMed]


Peyron C, Faraco J, Rogers W, et al. A mutation in a case of early onset narcolepsy and a generalized absence of hypocretin peptides in human narcoleptic brains. Nat Med. 2000;6(9):991–997. [PubMed]


Thannickal TC, Moore RY, Nienhuis R, et al. Reduced number of hypocretin neurons in human narcolepsy. Neuron. 2000;27(3):469–474. [PubMed]


Mignot E, Lammers GJ, Ripley B, et al. The role of cerebrospinal fluid hypocretin measurement in the diagnosis of narcolepsy and other hypersomnias. Arch Neurol. 2002;59:1553–1562. [PubMed]


Nishino S, Ripley B, Overeem S, Lammers GJ, Mignot E. Hypocretin (orexin) deficiency in human narcolepsy. Lancet. 2000;355(9197):39–40. [PubMed]


Mileykovskiy BY, Kiyashchenko LI, Siegel JM. Behavioral correlates of activity in identified hypocretin/orexin neurons. Neuron. 2005;46(5):787–798. [PubMed]


Scammell TE. Overview of sleep: the neurologic processes of the sleep-wake cycle. J Clin Psychiatry. 2015;76(5):e13. [PubMed]


Roth T, Dauvilliers Y, Mignot E, et al. Disrupted nighttime sleep in narcolepsy. J Clin Sleep Med. 2013;9(9):955–965. [PubMed Central][PubMed]


Mignot E, Lin L, Rogers W, et al. Complex HLA-DR and -DQ interactions confer risk of narcolepsy-cataplexy in three ethnic groups. Am J Hum Genet. 2001;68(3):686–699. [PubMed Central][PubMed]


John J, Thannickal TC, McGregor R, et al. Greatly increased numbers of histamine cells in human narcolepsy with cataplexy. Ann Neurol. 2013;74(6):786–793. [PubMed]


Valko PO, Gavrilov YV, Yamamoto M, et al. Increase of histaminergic tuberomammillary neurons in narcolepsy. Ann Neurol. 2013;74(6):794–804. [PubMed]


Hassani OK, Lee MG, Jones BE. Melanin-concentrating hormone neurons discharge in a reciprocal manner to orexin neurons across the sleep-wake cycle. Proc Natl Acad Sci U S A. 2009;106(7):2418–2422. [PubMed Central][PubMed]


Jego S, Glasgow SD, Herrera CG, et al. Optogenetic identification of a rapid eye movement sleep modulatory circuit in the hypothalamus. Nat Neurosci. 2013;16(11):1637–1643. [PubMed Central][PubMed]


Tsunematsu T, Ueno T, Tabuchi S, et al. Optogenetic manipulation of activity and temporally controlled cell-specific ablation reveal a role for MCH neurons in sleep/wake regulation. J Neurosci. 2014;34(20):6896–6909. [PubMed Central][PubMed]


Konadhode RR, Pelluru D, Blanco-Centurion C, et al. Optogenetic stimulation of MCH neurons increases sleep. J Neurosci. 2013;33(25):10257–10263. [PubMed Central][PubMed]


Rao Y, Lu M, Ge F, et al. Regulation of synaptic efficacy in hypocretin/ orexin-containing neurons by melanin concentrating hormone in the lateral hypothalamus. J Neurosci. 2008;28(37):9101–9110. [PubMed Central][PubMed]


van den Pol AN, Acuna-Goycolea C, Clark KR, Ghosh PK. Physiological properties of hypothalamic MCH neurons identified with selective expression of reporter gene after recombinant virus infection. Neuron. 2004;42(4):635–652. [PubMed]


Peyron C, Valentin F, Bayard S, et al. Melanin concentrating hormone in central hypersomnia. Sleep Med. 2011;12(8):768–772. [PubMed]


Blouin AM, Fried I, Wilson CL, et al. Human hypocretin and melanin-concentrating hormone levels are linked to emotion and social interaction. Nat Commun. 2013;4:1547. [PubMed Central][PubMed]


Salomon RM, Ripley B, Kennedy JS, et al. Diurnal variation of cerebrospinal fluid hypocretin-1 (Orexin-A) levels in control and depressed subjects. Biol Psychiatry. 2003;54(2):96–104. [PubMed]


Johns MW. A new method for measuring daytime sleepiness: the Epworth sleepiness scale. Sleep. 1991;14(6):540–545. [PubMed]


nic-Labat S, Guilleminault C, Kraemer HC, Meehan J, Arrigoni J, Mignot E. Validation of a cataplexy questionnaire in 983 sleep-disorders patients. Sleep. 1999;22(1):77–87. [PubMed]


Stiasny-Kolster K, Mayer G, Schafer S, Moller JC, Heinzel-Gutenbrunner M, Oertel WH. The REM sleep behavior disorder screening questionnaire--a new diagnostic instrument. Mov Disord. 2007;22(16):2386–2393. [PubMed]


Mamikunian P, Ardill JE, O'Dorisio TM, et al. Validation of neurokinin a assays in the United States and Europe. Pancreas. 2011;40(7):1000–1005. [PubMed]


Wewer Albrechtsen NJ, Kuhre RE, Torang S, Holst JJ. The intestinal distribution pattern of appetite- and glucose regulatory peptides in mice, rats and pigs. BMC Res Notes. 2016;9:60. [PubMed Central][PubMed]


Viale A, Zhixing Y, Breton C, et al. The melanin-concentrating hormone gene in human: flanking region analysis, fine chromosome mapping, and tissue-specific expression. Brain Res Mol Brain Res. 1997;46(1-2):243–255. [PubMed]


Thannickal TC, Nienhuis R, Siegel JM. Localized loss of hypocretin (orexin) cells in narcolepsy without cataplexy. Sleep. 2009;32(8):993–998. [PubMed Central][PubMed]


Parks GS, Olivas ND, Ikrar T, et al. Histamine inhibits the melanin-concentrating hormone system: implications for sleep and arousal. J Physiol. 2014;592(10):2183–2196. [PubMed]


Bassetti CL, Baumann CR, Dauvilliers Y, Croyal M, Robert P, Schwartz JC. Cerebrospinal fluid histamine levels are decreased in patients with narcolepsy and excessive daytime sleepiness of other origin. J Sleep Res. 2010;19(4):620–623. [PubMed]


Dauvilliers Y, Delallee N, Jaussent I, et al. Normal cerebrospinal fluid histamine and tele-methylhistamine levels in hypersomnia conditions. Sleep. 2012;35(10):1359–1366. [PubMed Central][PubMed]


Kanbayashi T, Kodama T, Kondo H, et al. CSF histamine contents in narcolepsy, idiopathic hypersomnia and obstructive sleep apnea syndrome. Sleep. 2009;32(2):181–187. [PubMed Central][PubMed]


Nishino S, Sakurai E, Nevsimalova S, et al. Decreased CSF histamine in narcolepsy with and without low CSF hypocretin-1 in comparison to healthy controls. Sleep. 2009;32(2):175–180. [PubMed Central][PubMed]


Zeitzer JM, Buckmaster CL, Parker KJ, Hauck CM, Lyons DM, Mignot E. Circadian and homeostatic regulation of hypocretin in a primate model: implications for the consolidation of wakefulness. J Neurosci. 2003;23(8):3555–3560. [PubMed]


Baumann CR, Stocker R, Imhof HG, et al. Hypocretin-1 (orexin A) deficiency in acute traumatic brain injury. Neurology. 2005;65(1):147–149. [PubMed]