INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

AGING AFFECTS MANY FUNCTIONAL aspects of circadian rhythms, including the phase relationship between overt circadian rhythms and the environmental light-dark cycle, decreased amplitude of rhythms, increased fragmentation, and reduced sensitivity of the circadian pacemaker to timing signals (7, 11, 12, 18, 21, 26, 29). Of these age-related changes, perhaps the most pronounced is the dramatic reduction in the ability of the aged circadian pacemaker to respond to so-called "nonphotic" signals (25, 26). In contrast to light, which stimulates phase shifts in rodents only during the subjective night and very early subjective day, nonphotic signals (e.g., dark pulses, benzodiazepine injections, access to a novel wheel) induce robust phase advances when presented during the middle of the subjective day (1, 4, 10, 17, 23). As shown by Van Reeth and colleagues (26), old hamsters (16-28 mo) show greatly diminished phase shifts to either triazolam injections or 6-h dark pulses compared with young adult hamsters (2-4 mo old).

Loss of sensitivity to nonphotic signals during aging may contribute to problems with circadian rhythms in older humans. If older individuals are less sensitive to timing signals, then they would be expected to exhibit more difficulties with shift work and jet lag than younger individuals. Also, because cataracts and lens opacity increase with age (20, 28) and light exposure often decreases (3), nonphotic signals may have a greater importance for resetting circadian rhythms in older adults than in younger individuals.

The objective of the present study was to determine the age of onset of the loss of sensitivity to dark pulses in golden hamsters, a well-established animal model for the study of circadian rhythms and aging. Also, because phase shifts to nonphotic stimuli are often associated with increased levels of activity (16, 24, 27), we sought to determine if age-related changes in nonphotic phase shifts were associated with changes in the amount of wheel running.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Animals and housing. Male Syrian hamsters (Harlan, HsdHan:AURA) of two ages (young, ~3 mo old and old, ~18 mo old) were used for these studies. [The average life span of Syrian hamsters is ~2 yr (6).] All of the hamsters had been maintained in our animal facility with exposure to a 14:10-h light-dark (14L:10D) photoperiod (lights on at 0600) for at least 1 wk before experimentation. Food (Teklad, Amway) and water were available continuously. During the experiment, hamsters were individually housed in polypropylene cages (10.25 × 8.25 × 18.75 in.).

Experimental design. At intervals of ~2 mo, the hamsters were tested for the ability to exhibit a phase shift in response to a dark pulse during exposure to continuous illumination. Before the dark pulse, each hamster was transferred to a plastic cage equipped with a running wheel (13.5-in. diameter) and was exposed to constant illumination (20-80 lx at cage level) for 2-3 wk. Wheel running activity was continuously recorded by magnetic switches and a computerized system, the Chronobiology Kit (Stanford Software Systems). Dark pulses were given for 6 h beginning at circadian time (CT) 6. Phase shifts, the period length (tau) before and after the dark pulse, and the amount of running (number of wheel revolutions) during the dark pulse were calculated 10-15 days after the dark pulse. Two to four weeks after the dark pulse, each hamster was transferred to cage without a running wheel and exposed to a 14L:10D photoperiod (lights on at 0600) for ~1 mo.

Data analysis. The computer program Clocklab was used to calculate phase shifts, period length, and wheel revolutions. The statistical significance of differences between age groups over time was assessed by repeated-measures ANOVA. In the case of significant F values (P < 0.05), post hoc comparisons were made using Fisher's least significant difference procedure for pairwise comparisons. Several other types of statistical analyses were also employed, as described below.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Effect of dark pulses and age on rhythm integrity and phase shifts. Six-hour dark pulses commencing at CT 6 induced abrupt splitting of locomotor activity rhythms in many of the young hamsters (Figs. 1 and 2). The percentage of young hamsters exhibiting split rhythms ranged from 6 to 56% over the four dark pulses. Split rhythms were never observed in old hamsters after a dark pulse. A Cochran-Mantel-Haenszel test indicated that there was a significant (P < 0.001) difference between old and young animals in the incidence of splitting over time. (The average age of each group of hamsters at the time of each dark pulse is shown in Table 1.)

View larger version (24K): [in this window] [in a new window]   Fig. 1.   Actograms showing the effect of dark pulses on circadian wheel running rhythms during exposure to continuous light. The actograms are double plotted. The x-axis represents clock time; the y-axis represents the day of the study. * Time of the onset of a 6-h dark pulse. Actograms represent young hamsters (7-8 mo old, A and B) and an old hamster (23 mo old, C). Splitting of the wheel running rhythm was exhibited by some young hamsters (e.g., B) but never by old hamsters.

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Percentage of animals in each age group exhibiting split wheel running rhythms after a dark pulse (DP). The number of animals in each age group varies among dark pulses, for several reasons. In the young age group, 5 animals spontaneously developed split rhythms before the 3rd dark pulses could be conducted. In the old age group, 3 animals died (1 of natural causes after the first dark pulse, and 2 were killed after the 4th dark pulse due to ill health) and 1 animal was removed from the study because its activity was very sparse and irregular.

View larger version (18K): [in this window] [in a new window]   Fig. 3.   The effect of age on the magnitude of phase shifts induced by a 6-h dark pulse commencing at circadian time 6. Values shown represent the mean ± SE of the phase shift. There was no significant effect of age on the magnitude of the dark pulse.

Effect of aging on activity levels and tau. Wheel running data were analyzed during the week before each 6-h dark pulse and during each 6-h dark pulse. The repeated-measures ANOVA indicated that there was a significant interaction between the age group and time (before or during dark pulse). Post hoc analyses revealed that there was a significant difference between the young and old hamsters during the week before the dark pulse but not during the dark pulse (Fig. 4).

View larger version (22K): [in this window] [in a new window]   Fig. 4.   The effect of age on the amount of wheel running. A: cumulative wheel revolutions during the week before each dark pulse. B: wheel revolutions during each 6-h dark pulse. Values shown represent the mean ± SE.

View larger version (26K): [in this window] [in a new window]   Fig. 5.   The effect of age on period length (tau). A: period length during the week before each dark pulse was presented. B: period length after each dark pulse. Values shown represent the mean ± SE.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The results of the current study did not did not reveal any significant effect of aging on the magnitude of the phase shifts induced by dark pulses in Syrian hamsters, in contrast to a previous study (26). In the current study, old hamsters as well as young hamsters showed robust phase advances in response to a dark pulse presented from CT 6-12. Also, old hamsters showed the same amount of running (numbers of wheel revolutions) during the dark pulses as the young hamsters, although they ran significantly less than the young hamsters during the week before each dark pulse. Thus our results do not confirm the earlier report by Van Reeth et al. (26) that aging reduces dark pulse induction of circadian phase shifts, and our results indicate that dark pulses stimulate activity and phase shifts equally well in young and old hamsters. Several differences between the present and previous study might explain the discrepant results. The hamsters used in the current study were obtained from Harlan Sprague Dawley, whereas the subjects of the study by Van Reeth et al. came from Charles River Lakeview. Also, the hamsters in the Van Reeth et al. study were 16-28 mo old. The present study tested hamsters in the middle of this age range (~19-25 mo of age) but not those at either extreme, in part because the normal life span of Syrian hamsters is ~22-24 mo (6). Thus it appears that loss of sensitivity to this particular nonphotic signal is not a consistent characteristic of the last quarter of the average life span in this species.

The most interesting finding of the present study was the acute induction of splitting by a dark pulse in young but not in old hamsters. In hamsters that were 5-11 mo of age, 6-56% of the animals exhibited splitting after a dark pulse. Most of the hamsters that exhibited splitting immediately after a dark pulse showed this response repeatedly, whereas none of the old hamsters ever exhibited acute splitting after a dark pulse. Although many previous studies have shown spontaneous splitting of activity rhythms in rodents during long-term exposure to constant light (2, 9, 13, 14, 22), we are aware of only one report (1) of acute splitting after a dark pulse. In this report, this phenomenon apparently occurred in only a few hamsters. Our finding that aging affects acute splitting after a dark pulse is reminiscent of an earlier report that aging affects the incidence of spontaneous splitting during chronic exposure to constant light (8). In this previous study, ~60% of young hamsters (5 mo old) exhibited split activity rhythms after 90 days of constant light exposure, in contrast to only 5% of old hamsters (~14 mo old) after 120 days of constant light exposure (8).

In cross-sectional aging studies such as this one, it is possible that any apparent age-related change may actually reflect some sort of life history effect, rather than an effect of biological aging per se. In the present study, although both the old and young hamsters lived in two facilities [the animal supplier's facilities (Harlan) and the Department of Laboratory Animal Medicine at the University of Kentucky], the old hamsters spent a longer time period in each facility. The possibility that long-term exposure to some condition (e.g., light intensity) in one or both of these facilities may have affected subsequent responses to dark pulses cannot be ruled out. However, the report that hamsters of another strain, received from a different supplier and housed at another university, also exhibited age differences in spontaneous splitting of rhythms during exposure to constant light (10) would argue against the concept that some specific feature of life history causes age-related changes in splitting.

Another issue related to cross-sectional aging studies concerns the fact that the old group of animals represents the survivors, i.e., only the healthiest animals, whereas the young group represents a broader mix of individuals. Information from the animal supplier (Harlan) indicates that the mortality rate for hamsters from birth to 12 mo of age is ~2-3%. According to our records, the mortality rate in our animal facility, between 12 and 18 mo of age, is also ~3%. Therefore, because ~94% of a hamster population would still be expected to be alive at 18 mo of age, the characteristics of the population have probably not changed dramatically during this time.

The factor(s) responsible for the inhibition of splitting by aging are unknown; indeed, the mechanisms inducing splitting are not well understood. It should be noted that splitting is not only associated with activity rhythms, but also with drinking rhythms, feeding rhythms, the ovulatory luteinizing hormone surge, and suprachiasmatic nucleus (SCN) electrical activity rhythms, and that splitting has been observed in several species (2, 22, 30). Several factors, besides age, have been reported to affect the incidence of splitting, including light intensity, duration of exposure to constant light, and the integrity of the geniculohypothalamic tract (5, 8, 13, 15). In our study, light intensity and duration of exposure to constant light were similar among all the hamsters. Also, the finding that phase shifts in response to dark pulses were not different among the age groups suggests that the perception of the light intensity or of the light-to-dark transitions did not vary with age. It is possible that the integrity or function of the geniculohypothalamic tract changes with age and influences the propensity to exhibit splitting. Previous studies of rats have shown that the SCN content of neuropeptide Y, the major neurotransmitter released by the geniculohypothalamic tract, decreases during aging (19). Furthermore, lesions of the hamster intergeniculate leaflet have been shown to reduce the incidence of splitting to 14%, compared with 75% in sham-operated animals (5).

The ability to split is considered a basic property of the circadian pacemaker. During splitting of circadian rhythms, two circadian oscillators, called the morning and evening oscillators, which are initially synchronized, separate and assume a new temporal relationship, 180° out of phase with each other (14). Thus there are two distinct, stable modes of coupling of these oscillators, one corresponding to the intact condition of rhythms and one corresponding to the split condition (14). Our results and those of Morin (8) suggest that the circadian oscillators can assume both coupling states in a large proportion of young hamsters but only one coupling state in old hamsters.

In conclusion, old male Syrian hamsters of the Sprague-Dawley strain respond robustly to 6-h dark pulses by showing increased wheel running activity and circadian phase advances. In these parameters, the old hamsters were not different from the young hamsters. In contrast to these similarities, old hamsters were significantly less likely than young hamsters to exhibit split circadian rhythms immediately after a dark pulse. Pittendrigh and Daan (14) suggested that the ability of morning and evening oscillators to separate facilitates adaptation to seasonal changes in day length. In support of this concept, Puchalski and Lynch (15) showed that the incidence of splitting in Siberian hamsters is greater in populations that exhibit short-photoperiod-induced seasonal adaptations than in those that do not show these adaptations. Thus the apparent lack of separability of morning and evening oscillators observed in old hamsters may contribute to their altered pattern of entrainment during short photoperiod, compared with young hamsters (21). Furthermore, reduced flexibility of coupling of circadian oscillators may contribute to problems associated with jet lag and shift work in the older human population.