Age Related Changes in Sleep and Relation to Altered Immunity

A focus of our own interest, as indeed for a number of other groups, has been the age-related changes which occur in sleep performance, and their interrelationship with simultaneous changes known to occur with immune functioning with age. Human sleep in old age is characterized by a number of changes, including sleep fragmentation and reductions in sleep efficiency and amounts of visually scored slow-wave and REM sleep, as well as amplitude of the diurnal sleep/wake rhythm. While acknowledging that a large body of evidence suggests that major declines in immunity with age are related to altered functions in T lymphocytes (and or in their numbers/subset distribution) (Globerson and Effros 2000), it is also evident that significant changes in cytokine responses also occur with age, and include changes in production of the very cytokines discussed above which can impact on sleep in humans (experimental animals) (Caruso et al. 1996; Gorczynski et al. 1997; Gorczynski, Dubiski, Munder, Cinader and Westphal, 1993; Hobbs et al. 1993; Sindermann, Kruse, Frercks, Schutz and Kirchner, 1993). One could thus consider the possibility that approaches taken to restore immune functioning in the elderly, if they included addressing similar cytokine changes, might thus also be reflected in the restoration of normal sleep behavior.

A model organism for the study of circadian rhythmicity and the effects of age on the circadian system has been the golden hamster. Surprisingly, nothing is known about the effects of advanced age on sleep in this species. As a first step in determining the effects of aging on sleep in the golden hamster, Naylor et al. recorded sleep over a 24-h period in young (3 months) and old (17 to 18 months) golden hamsters entrained to a 14:10 light:dark (LD) cycle (Naylor, Buxton, Bergmann, Easton, Zee, and Turek 1998). Aged hamsters were found to exhibit small increases in overall NREM sleep time, with no significant differences in REM sleep, median sleep episode length, or the number of arousals. The most striking differences between the sleep of young and old hamsters however was in NREM delta sleep, with older animals showing approximately one-third less delta sleep per NREM epoch than young hamsters, a phenomenon also reported earlier for humans, but not, interestingly, in another animal used popularly in such research, the rat. In these animals a characteristic of older age is a decrease in the mean duration of sleep bouts, an increase in the number of sleep bouts, and a modest reduction of REM sleep. From comparisons of sleep behavior in young adult rats (3 months), middle-aged (12 months) animals, and older (24 months) rats during 24 h under constant dim light, the most significant changes in reductions in high-voltage NREM sleep

("HS2"), the mean length of sleep bouts, and REM-onset duration were already evident by one year of age (the "rat equivalent" of midlife).

A major feature of sleep alteration with age is the altered circadian rhythmicity of the sleep (Weinert 2000). Both in aging animals and humans, all rhythm characters change, with the most prominent changes being a decrease in the amplitude and the diminished ability to synchronize with a periodic environment. Susceptibility to both photic and nonphotic cues is decreased, and in consequence, both internal and external temporal orders are disturbed both under steady-state conditions and, even more dramatically, following changes in the periodic environment. There are potentially multiple causes for these changes, many of which may be related to alterations within the SCN itself. As but one example, the number of functioning neurons decreases with advancing age as indeed likely does the coupling between them. In consequence, the SCN is significantly less able to produce stable rhythms and to transmit timing information to target sites. Interestingly it is thought that this change occurs in stepwise fashion, such that initially only the ability to synchronize with the periodic environment is diminished, while the rhythms themselves continue to be well represented. This being so it should in principle be possible to treat (some) age-dependent sleep disturbances pharmacologically; or by increasing the magnitude of the light-dark (LD) zeitgeber; or by strengthening feedback effects (e.g., by increasing the daily activity). At least in some preliminary studies these predictions seem to hold true (Weinert 2000).

As a further attempt to understand altered sleep with age, Mazzoccoli et al. explored whether age-related changes in neurotransmitters, hormones and cytokines, and in particular age-related changes in the 24-h hormonal and nonhormonal rhythms of their production, might be in some way responsible for altered sleep and/or immune functioning with age (Mazzoccoli et al. 1997). Cortisol, melatonin, thyrotropin-releasing hormone (TRH), thyroid-stimulating hormone (TSH), free thyroxine (T4), growth hormone (GH), insulin-like growth factor I (IGF-I), and IL-2 serum levels were measured, along with a detailed lymphocyte subpopulation analysis, on blood samples collected over a 24-h period from healthy young subjects (aged 36 to 58 years) and healthy older subjects (aged 65 to 78 years). The values of CD20+ (B cells) and CD25+ (activated T cells, expressing the alpha chain of the IL-2r) were higher in elderly subjects. In contrast, there was no statistically significant difference in the observed values of CD2 (total lymphocytes), CD4 (helper/inducer T cells), CD8 (suppressor/cytotoxic T cells), CD16 (natural killer cells), cortisol, melatonin, TRH, TSH, FT4, GE-I, IGF-I, and IL-2. However, while the group of elderly subjects continued to show a circadian rhythmicity for changes in CD2+, CD8+, CD16+, and CD25+ cells, as well as in cortisol, melatonin, TSH and GH there were significant phase changes in the cycling observed, leading the authors to conclude that not only is aging associated with enhanced responsiveness of the T-cell compartment, but that significant alterations in temporal architecture of the neuro-endocrine-immune system also occur with age. Despite these clear difference however, the response to total sleep deprivation, a powerful stimulus for sleep, does not apparently change significantly with age, despite the aforementioned decreased sleep continuity, slow wave sleep (SWS), growth hormone (GH) release and an increased hypothalamo-pituitary-adrenocortical system activity which is a characteristic of sleep in aging populations.

Independently, Prinz et al. (2000) have addressed in some detail the issue of whether the increased cortisol levels at the circadian nadir which accompany aging, reflecting an increased stress responsivity and a longer-lasting glucocorticoid increase, are responsible for individual differences in age-related sleep impairments. In this particular study they compared sleep, cortisol, and sleep-cortisol correlations under baseline and "stress" conditions in men and women, with the mildly stressful procedure being the insertion of a 24-h indwelling intravenous catheter. Healthy, nonobese subjects (60 women and 28 men) were included in the study, with mean ages in years of the female/male population being 70.6 (±6.2) and 72.3 (±5.7), respectively. 24h urines were assayed for cortisol, serum for IL-ip, and EEGs were also assessed by polysomnography and EEG power spectral analysis. These data showed that healthy older subjects (men and women) with the highest levels of free cortisol (24-h urine level) under a mild stress condition had impaired sleep, as reflected in lower sleep efficiency, less time in stages 2 to 4 sleep, and more EEG beta activity during NREM sleep. Interestingly, men had the highest levels of free urine cortisol under both baseline and mild stressful conditions, and cortisol and sleep correlated most strongly in men (accounting for ~36% of the variance in NREM sleep stress responses). In the female (but not the male population) higher cortisol levels in response to stress were associated with increased circulating levels of IL-ip, explaining ~24% of the variance in a subset of women. These intriguing data suggest that future research in this area must pay attention to gender-related effects in the elderly when we attempt to make correlations between altered neurohormonal/immunological parameters and sleep physiology. A recent study by Hawkley et al. has also made note of the profound effect of cumulative stressors of various types (social, psychological, physical) on immune impairment in the elderly, though no attempt was made to correlate those changes with altered sleep physiology (Hawkley and Cacioppo 2004).

What about studies which have directly assessed whether the sleep changes occurring during immune challenge are themselves affected by age? In one example of this type of investigation, sleep alterations induced by the administration of lipopolysaccharide (LPS) in young and middle-aged rats were examined (Schiffelholz and Lancel 2001). After vehicle challenge, middle-aged rats exhibited less pre-REM sleep as well as REM sleep itself, due to both a smaller number and shorter duration of REM sleep episodes, than young rats (see above). While LPS was observed to elevate body temperature, increase non-REM sleep, and suppress both pre-REM sleep and REM sleep in both young and middle-aged rats, the effects were not identical. Thus, in the younger animals, LPS enhanced slow-wave activity in the EEG within non-REM sleep, presumably reflecting an increase in sleep intensity, while it attenuated the same measures in the older animals. These striking observations are the first to hint at the possibility that the alteration in sleep in response to infection and immune challenge may be different in young versus aged cohorts. This hypothesis was tested directly in a model study by Imeri et al., using intracerebroventricular injections of IL-ip given to young and aged rats whose subsequent sleep-wake behavior was analyzed (Imeri, Ceccarelli, Mariotti, Manfridi, Opp, and Mancia 2004). Under basal conditions and in the absence of an immune challenge, the sleep patterns of young (3 months) and aged rats (25 to 27 months) were not significantly different. However, in young animals, IL-ip (2.5 ng) enhanced non-REM sleep, inhibited REM sleep, and induced fever. In contrast, in the aged animals, IL-ip did not alter non-REM sleep, although changes in REM sleep and brain temperature were equivalent to those seen in young animals. It has been postulated that enhanced non-REM sleep is important in facilitating recovery from microbial infection, which leads to the provocative conclusion that this alteration in non-REM sleep unique to the elderly population may contribute to the increased infection-induced morbidity and mortality of aged organisms. Our own research in sleep and aging has focused around two other of our areas of interest. In the one case we have reported on the role of a novel extract prepared from fetal sheep liver (CLP) in inducing changes in cytokine production from aged mice (Gorczynski et al. 2005; Gorczynski et al. 1997; Gorczynski et al. 1993). While our detailed reports to date only refer to altered cytokine production following CLP injection, it is important to note in the context of this discussion on sleep changes in the elderly that anecdotal records from patients treated with CLP intramuscularly (im) in the clinic clearly suggest a significant sleep-promoting effect also of this material (Waelli, unpublished observations). Exhaustive studies in aged mice showed an increased background production of inflammatory-type cytokines (TNF-a, IL-ip, and IL-6) compared with young mice of the same strain, and mitogen-induced cytokine production from lymphocytes of aged mice showed a bias toward type-2 cytokine production, rather than the type-1 cytokines derived predominantly from cells of young mice (Gorczynski et al. 1997). However, ongoing administration of CLP over a 4-week period resulted in restoration of both of these profiles toward those seen in younger animals.

The active moieties in CLP are now being defined, and apparently represent a complex mixture of (at least) LPS, fetal hemoglobin, MIF, and GSH. In addition to these investigations, we have published at length on the role of a molecule, CD200, in the regulation of nonspecific inflammation, and other acquired immune responses in animal model systems, and in man. Other groups have also shown that in a CD200 KO mouse, increased inflammation is seen within the brain and CNS, consistent with a physiological role of this molecule in regulating inflammatory processes within the CNS. We thus postulated that both the background levels of (inflammatory) cytokine production in aged animals, and the augmentation of those levels following peripheral LPS administration, might be regulated as a reflection of alterations in expression of CD200 which would again be pertinent to observations concerning a sleep-controlling effect of CLP.

In the first series of studies, we analyzed in young and aged groups of C57BL/6 mice, resting levels of expression in the brain both of CD200 and a number of the inflammatory cytokines reported to be key to promoting sleep behavior both in man and in experimental animals. Those levels were then measured in similar groups of mice 48 h following ip administration of LPS as a mimic of exposure to an infectious insult. Typical data for one of three such studies are shown in Fig. 7.1.

Comparison of expression of various mRNAs in brain of young (8wk) or aged (80wk) BL/6 mice at rest and 2d following peripheral (ip) administration of 5^/mouse LPS

0.50

0.00

Comparison of expression of various mRNAs in brain of young (8wk) or aged (80wk) BL/6 mice at rest and 2d following peripheral (ip) administration of 5^/mouse LPS

TNFi Il-1 p IL-6 CD200 TGBJJ

Super Serenity Sleepers

Super Serenity Sleepers

Do You Have Problem Getting A Good Night Sleep? Learn To Sleep Like A Cat At Night And Run Like A Lion When You Wake Up.

Get My Free Ebook


Post a comment