| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Department of Neuroendocrinology, University of Lübeck, Germany;
Department of Internal Medicine, University of Lübeck, Germany;
Department of Medical Biochemistry and Immunology, The School of Medicine, Cardiff; University, Cardiff, Wales, UK; and
Department of Biochemistry, Christian-Albrechts University of Kiel, Kiel, Germany
1Correspondence: Department of Neuroendocrinology, University of Lübeck, Ratzeburger Allee 160, Haus 23a, Lübeck 23538, Germany. E-mail: born{at}kfg.uni-luebeck.de
ABSTRACT
Sleep is commonly considered to support immune defense. The underlying sleep-immune interaction appears to rely critically on cytokines, like interleukin-6 (IL-6), that combine effects on immune and neuronal functions. The IL-6 signal is conveyed in two ways: it stimulates a restricted group of (mostly immune) cells via membrane-bound IL-6 receptors (mIL-6R) by forming a complex with soluble IL-6R (sIL-6R), and it stimulates (via membrane-bound gp130) a great variety of other cell types—a process termed trans-signaling. Focusing on the receptor side of IL-6 signaling, we examined the effect of sleep on sIL-6R plasma concentrations, mIL-6R expression, plasma sgp130, and numbers of IL-6-producing monocytes in healthy humans who were tested during a regular sleep-wake cycle and 24 h of wakefulness while blood was sampled repeatedly. Sleep strongly enhanced concentrations of sIL-6R, exceeding wake levels by 70% at the end of sleep. This rise was due to an increase in the PC (proteolytic cleavage) rather than the DS (differentially spliced) variant of sIL-6R. Sleep did not affect IL-6-producing monocytes, mIL-6R density, or sgp130 concentrations. The selective increase in sIL-6R implicates an enhanced trans-signaling capacity whereby sleep distinctly widens the profile of IL-6 actions, enabling an integrated influence on brain and peripheral organs.—Dimitrov, S., Lange, T., Benedict, C., Nowell, M. A., Jones, S. A., Scheller, J., Rose-John, S., Born, J. Sleep enhances IL-6 trans-signaling in humans.
Key Words: neuroimmunology • cytokine receptors • cytokines • hormones
THERE IS INCREASING evidence pointing to a supportive role of sleep for immunity (1
2
3
4)
. This function of sleep is based on fine-tuned interactions between central nervous sleep processes and immune functions, which appear to rely critically on specific signal molecules, such as the proinflammatory cytokines IL-6, IL-1, and TNF-
, that are partially controlled by sleep, and themselves exert regulatory influences on both immune and sleep functions (4
5
6
7)
. For IL-6, a sleep-wake associated release has been demonstrated, with increased IL-6 expression in blood and brain during the sleep period (i.e., in rats during daytime and in humans during the night) (8
9
10
11)
. Deprivation of sleep increases plasma IL-6 concentrations in humans and rats (12
, 13)
. Moreover, fatigue is associated with enhanced IL-6 plasma concentrations in conditions of disturbed sleep, and fatigue increases after IL-6 administration in healthy humans (14
, 15)
.
Despite this evidence for a sleep-dependent regulation of the IL-6 signal, studies so far have focused exclusively on changes in IL-6 and have neglected to consider changes in IL-6 receptor levels that ultimately determine which cells are activated by IL-6 and to what extent (16)
. IL-6 binds to two separate receptors: membrane-bound receptors (mIL-6R) and receptors that are released as soluble molecules (sIL-6R) into the plasma. The membrane-bound IL-6/interleukin-6R complex in turn associates with two molecules of the signal transducing unit gp130, leading to initiation of an intracellular signal (17)
. The sIL-6R binds IL-6 to form an agonistic complex and, as such, contrasts with most other cytokines, where their soluble receptors act as antagonists and compete with the membrane-bound receptor for ligand binding. The sIL-6R in complex with IL-6 can stimulate a great diversity of cells that express only membrane-bound gp130 (18
19
20)
. This process by which IL-6 stimulates cells lacking the mIL-6R is termed trans-signaling (16)
.
Here we sought to further characterize the relationship between sleep and IL-6 signaling by concentrating on a differentiated analysis of sleep-dependent fluctuations in the IL-6 receptor mechanism. In parallel, we studied in humans the diurnal variations of 1) sIL-6R concentrations in plasma, 2) the expression of mIL-6R in different leukocyte subpopulations, 3) plasma sgp130 (the natural antagonist of IL-6 trans-signaling), and 4) IL-6 production by monocytes. Since two isoforms of the sIL-6R are known originating, respectively, from shedding of the mIL-6R from the cell surface after proteolytic cleavage (PC-sIL-6R) and by differential mRNA splicing (DS-sIL-6R) (21)
, we also measured concentrations of DS-sIL-6R to dissociate effects on these two isoforms. Sleep is characterized by distinct changes in the release of several hormones with well-known influences on immune functions (2)
. To assess whether such hormonal changes may mediate effects of sleep on IL-6 signaling, we monitored plasma concentrations of growth hormone (GH), prolactin, and cortisol. Concentrations of C-reactive protein (CRP) were also assessed. To separate effects of sleep from circadian rhythm, subjects were examined twice: during a regular wake-sleep cycle and during a 24 h period of continuous wakefulness.
MATERIALS AND METHODS
Subjects
Fifteen physically and mentally healthy men participated in the study (mean age 25 years, range 21–30 years, mean body mass index 23.5 kg/m2, range 20–25 kg/m2). All individuals were nonsmokers presenting with a normal sleep pattern and were on no medications. None had a medical history of any relevant chronic disease or psychological disorders. Acute illness was excluded by physical examination and routine laboratory investigation, including chemistry panel, C-reactive protein <6 mg/l, and white blood cell (WBC) count <9000/µl.
The men were synchronized by daily activities and nocturnal rest. They had a regular sleep-wake rhythm for at least 6 wk before the experiments. During the week preceding the study they were required to turn lights off for nocturnal sleep between 23:00 and 23:30 h, to get up by 7:00 h the next morning, and not to take any naps during the day. The presence of signs of sleep disturbances, including apnea and nocturnal myoclonus, was excluded by interview and by recordings during a separate adaptation night. All subjects were adjusted to the experimental setting by spending at least one adaptation night in the laboratory that took place prior to the proper experiment and included the attachment of all electrodes for sleep recordings. The study was approved by the Ethics Committee of the University of Lübeck. All men gave written informed consent and were paid for their participation.
Experimental design and procedure
Experiments were performed according to a within-subject crossover design. Each man participated in two experimental conditions, each starting at 20:00 h and ending 24 h later at 20:00 h the next day. One condition ("sleep") included a regular sleep-wake cycle; in the other condition ("wake") subjects remained awake throughout the 24 h experimental period. Both experimental sessions for a subject were separated by at least 4 wk, and the order of conditions was balanced across subjects. To exclude possible psychological effects deriving from anticipating a night of wakefulness or of regular sleep, subjects were generally kept misinformed about the actual condition established on a particular night until 23:00 h (i.e., the time lights were turned off in the sleep condition).
On experimental nights, subjects arrived at the laboratory at 18:00 h for standard polysomnographic recordings and blood supply preparations. Sleep was allowed between 23:00 h (lights off) and 7:00 h in the morning. In the wake condition, subjects stayed awake in bed in a half-supine position between 23:00 and 7:00 h. During this period they were watching TV, listening to music, and talking to the experimenter at normal room light (
300 lx). For both sessions, standardized meals were provided at appropriate times for breakfast (8:00 h), lunch (12:00 h), and dinner (18:00 h).
In both conditions, blood was sampled first at 20:00 h, then every 1.5 h between 23:00 and 8:00 h, and every 3 h between 8:00 and 20:00 h the next day. Blood was sampled via an i.v. forearm catheter that was connected to a long thin tube, allowing blood collection from an adjacent room without disturbing the subject’s sleep. To prevent clotting, 700 ml of saline solution were infused throughout the 24 h experimental period. The total volume of blood sampled during a session was 250 ml. Blood samples were always processed immediately after sampling.
Sleep, self-reported fatigue, and mood
Sleep stages were determined off-line from polysomnographic recordings following standard criteria (22)
. For each night, sleep onset (with reference to lights off at 23:00 h), total sleep time, and the time as well as percentage of total sleep time spent in the different sleep stages (wake, stages 1, 2, 3, and 4 and rapid eye movement (REM) sleep) were determined. Slow wave sleep (SWS) was defined by the sum of stage 3 and 4 sleep. Latencies of sleep stage 2, SWS, and REM sleep were assessed with reference to sleep onset. Subjective feelings of tiredness, activation, and mood were assessed three times (at 22:00, 8:00, and 20:00 h) by a checklist of adjectives (23)
.
mIL-6R density on leukocyte subpopulations
mIL-6R density on leukocytes was determined by FACS analysis. Blood collected into sodium-heparin vacutainers (BD Biosciences, San Jose, CA, USA) was immediately labeled with saturating amounts of PE-conjugated anti-IL-6R (CD126+) antibody (Ab), which recognizes mIL-6R (clone M5, BD Biosciences). Data were acquired from 10,000 gated events. Irrelevant Ab from the same subclass was used as a negative control. To determine the distribution of mIL-6R on leukocyte subpopulations, granulocytes were identified based on forward and side light scatter characteristics and CD45 expression. Identification of monocytes was based on CD14 expression and that of T cells on CD3 expression. Samples were analyzed using a FACS Calibur flow cytometer (BD Biosciences). Before measurements, the FL2 fluorescence channel was calibrated using QuantiBRITE beads (BD Biosciences). Results were then analyzed using the QuantiCALC software (Verity Software House, Topsham, ME, USA). mIL-6R density was expressed in terms of total Ab binding capacity (ABC) (24)
on granulocytes, monocytes, and T cells, and as a percentage of the CD126+CD3+ fraction of CD3+ T cells.
Analysis of plasma sIL-6R and sgp130 levels
Blood was centrifuged immediately after sampling and the plasma was stored at –70°C until analysis. sIL-6R and sgp130 levels were measured by ELISA kits (R&D Systems, Minneapolis, MN, USA). Assay sensitivity was 6.5 pg/ml for sIL-6R and 80 pg/ml for sgp130. In supplementary analyses, plasma IL-6 concentrations were measured using an ELISA kit (R&D Systems) measuring both free circulating IL-6 and IL-6 bound to sIL-6R at a sensitivity of 0.7 pg/ml. DS-sIL-6R was measured as described (25)
. Briefly, microtiter plates were coated overnight with 2 µg/ml anti-DS-sIL-6R Ab (clone 2F3), then blocked with 5% BSA-PBS for 24 h. After washing, test samples (200 µl) were added to each well and incubated at 37°C for 3 h. Secondary Ab (clone BAF 227, R&D Systems) was then added and DS-sIL-6R was detected by commercial streptavidin peroxidase (Amersham Biosciences, Piscataway, NJ, USA). The color was developed by TMB substrate. DS-sIL-6R concentrations were determined using the baculovirus-expressed form of the receptor as a control standard. Assay sensitivity was 15 pg/ml.
Intracellular IL-6 production of monocytes
The stimulated production of IL-6 was evaluated at the single cell level by multiparametric flow cytometry. Immediately after drawing, heparinized venous blood was diluted 1:1 with RPMI 1640 medium in 15 ml cones-bottomed Falkon tubes (BD Biosciences) preloaded with LPS at a final concentration of 100 pg/ml under laminar airflow. Tubes were incubated at 37°C in a 5% CO2 atmosphere for 6 h. For stopping cytokine excretion, brefeldin A (Sigma-Aldrich, St. Louis, MO, USA) was added during the last 4 h at a final concentration of 10 µg/ml. Intracellular IL-6 staining was performed according to the manufacturer’s instructions using antibodies and reagents from BD Biosciences. Immediately after drawing, cells were labeled at room temperature for 15 min using directly conjugated antibodies CD14/FITC and human leukocyte antigen (HLA)-DR/PerCP. Red blood cells were lysed with 2 ml lysis buffer for 10 min. After centrifugation for 5 min at 500 g and removal of supernatant, probes were fixed and permeabilized with Cytofix buffer and Perm/Wash buffer, respectively. Then probes were incubated with intracellular cytokine IL-6/PE (clone MQ2–6A3). Purified Ab in excess was used to discriminate between positive and negative cells. After washing, cells were resuspended in PBS with 1% formaldehyde and analyzed within 2 h. At least 10,000 CD14+HLA-DR positive cells (monocytes) were acquired and subsequently analyzed for the expression of IL-6 on a FACS Calibur using CellQuest Software (Becton Dickinson, San Jose, CA, USA). Results for cytokine positive cells were expressed as percentages of the total CD14+HLA-DR-positive cells (monocytes).
CRP and hormones
CRP and hormone concentrations were determined in serum samples that were kept frozen at –70°C until assay. The sensitivity and intra- and interassay coefficients of variation (CV) of the assays (all Immulite, DPC Biemann GmbH, Bad Nauheim, Germany) were as follows: CRP (sensitivity: 0.02 mg/dl, intra- and interassay CV:<10%); cortisol (0.2 µg/dl,<10%); GH (0.01 ng/ml,<6.5%); and prolactin (0.16 ng/ml,<8.2%).
Statistical analysis
Data are presented as means ± SE. Statistical analysis was based on repeated measures ANOVA, including the factors "sleep/wake" (reflecting the condition) and time (reflecting the different time points of measurement). Post hoc contrasts were applied to analyze differences at specific time points. Cosinor analysis was performed with Chronolab, a software package for chronobiological time series analysis (26)
. In addition, Pearson correlation analysis was performed for selected intervals during regular sleep to identify relationships between the occurrence of sleep stages and levels of blood parameters. Correlations were also calculated between measures of fatigue and, respectively, sIL-6R concentration, and production of IL-6 by monocytes. If normal distribution could not be assumed, Spearman rank correlations were calculated.
RESULTS
Normal sleep during the regular sleep-wake cycle and increased fatigue during continuous wakefulness
Polysomnographical recordings assured that the subjects’ sleep was typical for laboratory conditions. Figure 1
shows a representative sleep profile from one subject in combination with plasma concentrations of sIL-6R and sgp130, IL-6-producing monocyte counts, and plasma GH concentrations. In all subjects, SWS predominated during the early part whereas REM sleep dominated during the second half of nocturnal sleep. Mean (±SE) sleep onset latency was 19.6 ± 3.9 min. Mean values were for sleep time, 431.5 ± 15.9 min; time in stage 1 sleep, 28.6 ± 6.3 min; stage 2 sleep, 232.8 ± 13.3 min; SWS, 75.5 ± 6.3 min; and REM sleep, 71.0 ± 5.6 min. The mean latency (with reference to sleep onset) was 18.6 ± 3.4 min for SWS and 119.7 ± 14.7 min for REM sleep.
|
Self-report measurements noted increased tiredness and decreased activation in the morning (8:00 h) and evening (20:00 h) after the night of wakefulness compared with self-reports conducted after regular nocturnal sleep (8:00 h: tiredness: 4.4±0.47 vs. 2.3±0.35, feeling of activation: 6.2±0.35 vs. 7.5±0.47, P<0.01, 20:00 h: tiredness: 5.5±0.51 vs. 2.7±0.42, feeling of activation: 5.8±0.41 vs. 7.3±0.47, P<0.01).
mIL-6R density on leukocytes shows circadian variation but is not affected by sleep
Diurnal expression patterns of mIL-6R (CD126+) were studied in granulocytes (selected based on forward and side characteristics), monocytes (CD14+), and T cells (CD3+) using flow cytometry. The whole fractions of granulocytes and monocytes, and 80% of the T cells were positive for mIL-6R. There was no difference between the sleep and wake conditions in the IL-6R density expressed in terms of total Ab binding capacity in any of the three subpopulations. In the case of T cells, the percentage of CD126+ T cells (CD126+CD3+) remained unaffected by sleep (P>0.46, for respective comparisons, Fig. 2
). However, IL-6R density in all three subpopulations as well as the percentage of CD126+ T cells showed a distinct circadian variation that was independent of sleep or wakefulness. The maximum of membrane-bound IL-6R expression was generally found during the late part of the night (Table 1
).
|
|
Sleep strongly increases sIL-6R concentration without changing sgp130 concentration
Compared with the wake condition, sleep markedly increased plasma concentrations of sIL-6R (Fig. 3
A). The rise was most pronounced during late sleep (i.e., after 2:00 h) and peaked shortly after morning awakening (F(11,154)=4.3, P<0.01 for sleep/wakextime). Comparisons at single time points revealed significance for the difference between the effects of sleep and wakefulness at 6:30 h and 8:00 h (P<0.01) and approached significance at 3:30 h and 5:00 h (P<0.09). Seven of the 15 subjects showed an 2-fold increase in sIL-6R concentrations during the second half of sleep and around morning awakening (Fig. 1B
shows data from one subject); the other subjects exhibited smaller or no differences in sIL-6R concentration during this period.
|
Cosinor analyses confirmed the absence of any circadian rhythm under conditions of continuous wakefulness (P>0.41). Applied to the data of the sleep condition a systematic fluctuation across the day was detected (P<0.01), with the fitted cosine curve showing a mean level (mesor) of 30.7 ± 2.1 ng/ml, a (peak-to-trough) amplitude of 3.7 ± 1.3 ng/ml and an acrophase (peak time) at 9:36 h ± 92 min.
Plasma concentrations of sgp130, which antagonizes the IL-6/sIL-6R complex, were not consistently influenced by sleep (P>0.1 for sleep/wakextime, Fig. 3B
). To further exclude any masking effects of sgp130 on the changes in sIL-6R, we examined the sIL-6R/sgp130 ratio. These data confirmed (P<0.05) the observed sleep-dependent increase in sIL-6R concentration. Moreover, any individual changes in sgp130 concentrations during sleep were completely unrelated to the strong rise in sIL-6R concentrations as revealed by Pearson correlation analysis (r=–0.02, P>0.9, for difference values between the first and last 3 h period of sleep). Cosinor analyses confirmed (P>0.13) that the sgp130 concentration was also not affected by changes in circadian rhythm.
The sleep-dependent rise in sIL-6R concentration pertains to its PC variant rather than to its DS variant
The greatest portion of sIL-6R in plasma of healthy humans originates from shedding of the mIL-6R from the cell surface after proteolytic cleavage (PC-sIL-6R), whereas a minor amount derives from differential mRNA splicing (DS-sIL-6R). To determine which of the two isoforms contributes to the sleep-dependent elevation in sIL-6R, we additionally measured DS-sIL-6R levels. In contrast to total sIL-6R concentrations, DS-sIL-6R concentrations during the night were comparable between the sleep and wake condition (Fig. 3C
). However, in the sleep condition, DS-sIL-6R concentrations increased during daytime after sleep, indicating a stimulating action of sleep that is substantially delayed (F(11,154)=2.5, P<0.05 for sleep/wakextime). Pairwise comparisons between the sleep and wake condition at single time points revealed a consistent enhancement in DS-sIL-6R levels after sleep at 14:00 h (P<0.01, for differences at 8:00 h and 11:00 h, P<0.08). In line with these results, cosinor analyses failed to reveal any circadian variation in DS-sIL-6R concentrations during the wake condition (P>0.9). However, applied to the data of the sleep condition, this analysis indicated a distinct variation (P<0.01), with the fitted cosine curve showing a mean level (mesor) of 101.9 ± 19.1 pg/ml, a peak-to-trough amplitude of 29.4 ± 8.4 pg/ml, and its peak time in the early afternoon (15:08 h±78 min).
IL-6-producing monocyte counts show circadian but not sleep-dependent changes
To assess systemic IL-6 activity the number of IL-6-producing monocytes representing the main source of IL-6 in blood was determined by flow cytometry. The percentage (with reference to the total number) of monocytes producing IL-6 showed a clear-cut 24 h variation during continuous wakefulness and also during the regular sleep-wake cycle (P<0.01, Fig. 3D
). The cosine curve fitted to the data during continuous wakefulness indicated a mean level (mesor) of 55.8 ± 3.1%, a (peak-to-trough) amplitude of 8.7 ± 2.2%, and a nocturnal peak time (acrophase) at 2:55 h ± 37 min. However, sleep did not affect the proportion of IL-6-producing monocytes (P>0.5, for respective ANOVA comparisons).
Supplementary analyses aimed to assess plasma concentrations of IL-6 as an additional measure of IL-6 activity. However, results did not show systematic variations except on four occasions, when the catheter was changed in the daytime period (due to difficulties of blood sampling); this was followed in each case by an immediate drop in plasma IL-6 concentrations (last sample before change of catheter: 23.3±6.8 pg/ml, first sample after: 4.8±3.8 pg/ml, P<0.05). This observation corroborates previous studies indicating that IL-6 measured in blood sampled via indwelling i.v. catheters can be confounded by local release of the cytokine (27)
.
Sleep and circadian rhythms regulate plasma concentrations of growth hormone, prolactin and cortisol, but not of CRP
Concentrations of CRP were not influenced by sleep (P>0.56) or show any circadian fluctuations (P>0.39; Fig. 4
A). Compared with nocturnal wakefulness, regular sleep was, as expected, characterized by a distinct pattern of endocrine changes. Concentrations of GH were increased selectively during early sleep (between 0:30 and 2:00 h; F(11,154)=9.9, P<0.001, for sleep/wakextime, Fig. 4B
). Plasma concentration of prolactin were strongly enhanced during sleep throughout the night compared with the wake condition (F(11,154)=11.2, P<0.001, Fig. 4C
). Cortisol concentrations did not differ between sleep and wake condition (P>0.20), but showed a pronounced circadian rhythm with peak concentrations around 8:00 h (P<0.001, Fig. 4D
). In a supplementary analysis, we assessed melatonin concentrations. These were on average slightly higher during sleep than nocturnal wakefulness, but not approaching any significance (43.0±7.7 vs. 37.2±5.6 pg/ml, P>0.25, for all relevant comparisons).
|
sIL-6R concentrations during sleep are correlated with time in specific sleep stages and sleep-associated hormone release
To explore whether the sleep-associated increase in plasma concentrations of sIL-6R was linked to a particular sleep stage or hormonal release, we calculated Pearson correlation coefficients between the time spent in different sleep stages and the average plasma concentrations of hormones and of sIL-6R. For this analysis, sIL-6R concentrations were averaged across 3 time intervals, reflecting its temporal dynamics during sleep: 1) 20:00–23:00 h, defining the presleep baseline, 2) 0:30–2:00 h, defining the period of early sleep where sIL-6R concentrations were low in conjunction with a prevalence of SWS and high levels of GH, and 3) 3:30–8:00 h, defining late sleep and the period around awakening, characterized by distinctly enhanced levels of sIL-6R concurring with high amounts of REM sleep.
Baseline values of sIL-6R did not correlate with any measure of sleep architecture. However, average levels of sIL-6R during early sleep correlated negatively with time in SWS (r=–0.66, P<0.01, Fig. 5
A) as well as with the average level of GH during early sleep (r=–0.52, P<0.05, Fig. 5B
). Average sIL-6R levels during late night were positively correlated with time in REM sleep (r=0.54, P<0.05; Fig. 5C
). Average levels of DS-sIL-6R during late sleep (where these are at a minimum) correlated negatively with average GH levels during early sleep (r=–0.52, P<0.05).
|
Additional analyses indicated that sIL-6R concentrations were positively correlated with reported tiredness and negatively with feelings of activation; this relation was most pronounced in the morning after a night of wakefulness (08:00 h, tiredness: r=0.56, P<0.05; activation: r=–0.65, P<0.05). There were no consistent correlations between these self-report measures and the production of IL-6 by monocytes (r<0.41, P>0.2).
DISCUSSION
We examined effects of nocturnal sleep and circadian rhythms on IL-6 signaling, including measures of IL-6 activity and IL-6 receptors in human blood. The major finding is that sleep compared with continuous wakefulness increased concentrations of sIL-6R on average by >70%. The increase in sIL-6R covered the period of late night sleep with predominant REM sleep and persisted during the first hour after morning awakening. It was not accompanied by parallel effects of sleep on membrane-bound IL-6R or plasma concentrations of sgp130, which acts as a negative regulator of IL-6 trans-signaling by blocking sIL-6R-mediated events. Also, sleep did not affect the percentages of IL-6-producing monocytes. However, this measure revealed a distinct circadian rhythm, with maximum IL-6 activity observed during nighttime. The selective and profound increase in sIL-6R concentrations observed during late sleep, coinciding with a circadian increase in IL-6 activity during nighttime, implicates that sleep enhances the capacity for IL-6 trans-signaling. The development of this alternative mode of IL-6 activation would therefore enable cell types to be activated that normally remain unresponsive to IL-6 itself (20
, 28)
.
Our finding of a distinctly enhanced percentage of IL-6-producing monocytes during the night compared with daytime is highly consistent with earlier sleep studies performed in animals (8
, 11)
and humans (9)
. However, our data also clearly indicate that the nocturnal rise in IL-6-producing monocytes reflects a circadian oscillation, which does not depend on the subject’s vigilance state. The strong circadian rhythm in IL-6 activity might partly reflect inhibiting influences of glucocorticoid release, showing a reversed rhythm (29
, 30)
. However, since our analysis focused on the production of IL-6 by monocytes, our findings do not rule out that sleep affects circulating levels of IL-6, which in many studies has been used to indicate IL-6 activity. Thus, Redwine et al. (10)
revealed a sleep-dependent increase in circulating IL-6 levels, whereas IL-6 levels remained low during a night of sleep loss. Indeed, cells other than monocytes (e.g., adipocytes, muscle cells) contribute substantially to circulating IL-6 levels, accounting for the full systemic activity of IL-6. The contribution of adipocytes, for example, has been estimated to be as much as 30% (31
, 32)
. Since we found plasma IL-6 concentrations in our experiments to be confounded, most likely due to local release at the site of catheterization (27)
, we concentrated on the assessment of stimulated cytokine production, which in previous human studies proved sensitive to effects of sleep and circadian rhythm (e.g., ref. 33
, 34
). In fact, restriction to monocyte IL-6 production could also account for our failure to find any correlation between reported tiredness and IL-6 activity. An association between fatigue and circulating levels of IL-6 was consistently demonstrated in healthy humans after sleep deprivation and administration of IL-6 as well as in patients with sleep disturbances, and has encouraged the concept that circulating IL-6 is a mediator of sleepiness that accumulates as a result of increased homeostatic sleep pressure (13
, 15
, 35
, 36)
. If circulating IL-6 concentrations reflecting IL-6 from different sources, including monocytes, adipocytes, and muscle cells, more sensitively reflect effects of fatigue, this could well explain why we did not detect an increase in IL-6 specifically produced by monocytes in the wake condition where fatigue increased during the experimental epoch. In signaling fatigue, circulating IL-6 appears to be supported by sIL-6R, since concentrations of sIL-6R in our experiments were indeed positively correlated with reported tiredness (for related results, see ref. 37
).
The effect of the IL-6 signal in fact depends essentially on the regulation of its receptor molecules, the soluble types of which are present in the circulation at a >1000-fold higher concentration than IL-6 itself. Nevertheless, possible influences of sleep and circadian rhythm on IL-6 receptor molecules have so far been neglected. Our results show a circadian increase during nighttime not only for the monocyte production of IL-6 (peaking 3:00 h), but also for the expression of membrane-bound IL-6R in leukocytes, which overall peaked somewhat later (between 3:30 and 10:00 h, depending on the subpopulation). Together, these changes speak for a convergent circadian influence that enhances IL-6 signal strength for those cells carrying mIL-6R during nighttime, which is independent of sleep.
In contrast to changes in mIL-6R, increases in circulating sIL-6R were completely dependent on sleep. The 70% increase in concentrations of sIL-6R, as observed here during late sleep and around awakening, is expected to decrease free IL-6 levels by 40% and to increase concomitantly the concentration of IL-6/sIL-6R complex by 30% (38)
. The buffering capacity of sgp130, which was not affected by sleep, is estimated to be rather limited in this normal range of concentrations. Hence, the increase in sIL-6R concentration during sleep is expected to result in enhanced IL-6 trans-signaling. In light of the striking increase in sIL-6R concentrations on sleep, we examined whether this effect differed for the two known isoforms of the sIL-6R. The DS-sIL-6R isoform originates from translation of differentially spliced mRNA. The other, the PC-sIL-6R, which under normal conditions is the predominant form, is produced by shedding (proteolytic cleavage) of the mIL-6R from the cell surface (19)
. Abs raised against the unique COOH-terminal sequence of DS-sIL-6R enabled us to determine the relative abundance of each isoform and its contribution to the sleep-dependent rise in sIL-6R concentration. Like total sIL-6R concentrations, levels of DS-sIL-6R did not show a circadian rhythm (during continuous wakefulness), but were found to be selectively enhanced by sleep. However, in contrast to the sleep-induced rise in total sIL-6R concentrations that peaked around morning awakening, the rise in DS-sIL-6R concentrations was delayed by 6 h, peaking at 14:00 h. This rules out a contribution of DS-sIL-6R to the rise in total sIL-6R during late sleep, which accordingly reflects increased shedding of PC-sIL-6R.
Nevertheless, it is possible that the sleep-induced rise in both PC-sIL-6R and DS-sIL-6R originates from a common though unknown sleep-associated factor stimulating both isoforms. The release of PC-sIL-6R and DS-sIL-6R differs with respect to the kinetics, and different mechanisms control their production. Activators of sIL-6R shedding are typically rapid, and significant increases in PC-sIL-6R can be seen within 30–120 min of stimulation. In contrast, secretion of DS-sIL-6R is relatively slow, needs de novo synthesis, and enhanced concentrations of this isoform typically do not occur until 8–24 h after activation (39)
. Thus, sleep might stimulate production and release of these receptors at an early stage, which evolves into increased circulating receptor concentrations at different times.
To identify possible mediators involved in the up-regulation of sIL-6R during sleep, we measured plasma concentrations of CRP and several hormones with known sleep-dependent patterns of release. CRP at concentrations observed during acute infection (5 mg/dl) has been shown to induce large increases in sIL-6R production by proteolytic shedding of receptors from neutrophils (40)
. Our data, however, do not provide any evidence for an involvement of CRP in generating the sleep-dependent increase in sIL-6R concentrations, since its plasma concentration remained stable over the 24 h interval in both conditions. The sleep-dependent peak in GH concentration was found to be negatively correlated with sIL-6R concentrations during early sleep, an observation that is consistent with reports pointing at a suppressing influence of GH on sIL-6R (41)
. A contribution of GH to the regulation of sIL-6R concentration is further suggested by the fact that sIL-6R concentrations were also negatively correlated with SWS, which is known to be closely linked to somatotropic secretory activity (42
, 43)
. Time in REM sleep was the only variable that correlated with sIL-6R levels during late sleep, indicating an association between high amounts of REM sleep and high sIL-6R levels. Whether this relation reflects a stimulating effect of REM sleep on sIL-6R shedding, or vice versa, a stimulation of REM sleep through enhanced sIL-6R shedding remains to be clarified. Melatonin, another well-known immunoactive hormone whose release is controlled by light, in our experiments did not differ substantially between nocturnal sleep and wake conditions. However, other hormones and cytokines not measured here could contribute to the sleep-dependent increase in sIL-6R. Thus, TNF- and IL-1ß are known to enhance IL-6R shedding and to promote fatigue and sleep (6
, 44
, 45)
. However, there is no evidence that release of these cytokines is increased by sleep (2
, 3)
.
Our central finding of a sleep-induced increase in sIL-6R implicates an enhanced IL-6 trans-signaling capacity (i.e., that sleep enhances the effects of IL-6 on all cells expressing membrane gp130). Increased IL-6 binding to sIL-6R is expected to be paralleled by a relative decrease of free IL-6 available for binding to membrane IL-6R. Its remarkable size makes the sleep-dependent rise in sIL-6R concentration relevant to various disease conditions. Within the immune system, the shift toward increased membrane gp130-mediated actions presumably induces substantial changes in the pattern of adhesion molecule expression, which in turn enforces proinflammatory actions at local sites of mononuclear cell infiltration (21
, 46
47
48
49
50)
. sIL-6R/IL-6 complexes are likewise critical for maintaining chronic inflammatory diseases such as peritonitis, rheumatoid arthritis, allergic asthma, and colon cancer (16
, 51)
. Our finding of distinctly enhanced sIL-6R concentrations around the time of morning awakening agrees well with clinical observations of exacerbated symptoms in diseases during this period (52
, 53)
. Within the central nervous system, the sleep-induced rise in sIL-6R via enhanced binding to membrane gp130 is expected to potentiate effects on brain structures regulating body temperature and general activity (37)
. In this way, IL-6 is likely to gain increased (feedback) control over central nervous mechanisms regulating sleep and associated plastic neural processes (54
55
56)
. Our results identify IL-6 trans-signaling as a critical mechanism by which sleep presumably greatly extends the influence of IL-6 on multiple organ systems, thereby integrating actions on immune and central nervous functions.
ACKNOWLEDGMENTS
This work was supported by the Deutsche Forschungsgemeinschaft SFB654. We are grateful to C. Otten, A. Otterbein, T. Kriesen, and E. Böschen for technical assistance, and to Dr. S. Horiuchi, Tokyo Medical and Dental University, for providing us with 2F3 monoclonal antibody (mAb) for detection of DS-sIL-6R.
Received for publication January 24, 2006. Accepted for publication May 8, 2006.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  M.-F. O'Connor, S. J. Motivala, E. M. Valladares, R. Olmstead, and M. R. Irwin Sex differences in monocyte expression of IL-6: role of autonomic mechanisms Am J Physiol Regulatory Integrative Comp Physiol, July 1, 2007; 293(1): R145 - R151. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Lange, S. Dimitrov, H.-L. Fehm, J. Westermann, and J. Born Shift of monocyte function toward cellular immunity during sleep. Arch Intern Med, September 18, 2006; 166(16): 1695 - 1700. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
