Most organisms possess “biological chronometers” in the form of circadian clocks (Pittendrigh 1960). Organisms track time in their local environment by entraining their circadian clocks to the light/dark (LD) cycles, a phenomenon that is of paramount importance for their survival under fluctuating environments (Daan and Aschoff 2001). In epidemiological studies, it was repeatedly shown that sleep disturbance and shift work (long-term night shifts and work schedules in which employees change or rotate shifts) disrupt normal biological rhythms and have negative impacts on health. Deterioration of health can manifest in the short-term as sleep disorders, jet-lag syndrome and accidents; in the long term, shift work is associated with a number of health problems including cardiovascular disease, impaired glucose and lipid metabolism, gastrointestinal discomfort, reproductive difficulties, and breast cancer (Boivin et al. 2007; Garbarino 2006; Ellingsen et al. 2007; Knutsson 2008). The results of a number of animal studies also indicate the importance of normal circadian rhythmicity in health and survival. It has been repeatedly shown that circadian dysfunction causes a reduction in Drosophila life span (Klarsfeld and Rouyer 1998; Kumar et al. 2005; Pittendrigh and Minis 1972). Shortened life span was reported for cycle mutant male flies compared to wild-type flies (Hendricks et al. 2003). The authors argued that the reduction in life span among the cycle mutant males was due to the lack of sleep (or rest) rather than to circadian dysfunction per se. Kumar et al. (2005) assayed locomotor activity behavior and life span among adult flies kept under constant dark conditions in the laboratory, wherein they were categorized as rhythmic if their activity/rest schedules followed circadian (approximately 24 h) patterns, and as arrhythmic if their activity/rest schedules did not display any pattern. The rhythmic flies lived significantly longer than the arrhythmic ones. However, enhanced life span also may be achieved by manipulating environmental conditions or by administering treatments that promote proper functioning of circadian clocks (Hurd and Ralph 1998).       

It was repeatedly shown that circadian clocks play an important role in determining the timing of key ontogenetic events.  Both genetic and environmental manipulations of circadian clock parameters have been shown to affect many life-history related traits (Prasad and Joshi 2003; Sharma 2003). In Drosophila melanogaster, for example, LD regime has been shown to affect pre-adult development time (Sheeba et al. 1999), lifetime fecundity (Sheeba et al. 2000), and adult life span (Hendricks 2003; Klarsfeld and Rouyer 1998; Kumar et al. 2005; Pittendrigh and Minis 1972; Sheeba et al. 2000).

In the present study, we investigated whether the duration of LD period could influence the pre-adult developmental time and adult life span in Drosophila melanogaster.

Materials and methods

Stock and cultivation methods

Wild-type Oregon R flies were used in the experiment. Flies were reared and maintained as adults at constant temperature 25°C (±1°C), relative humidity at ~70% (±5%), on standard cornmeal-sugar-yeast-agar medium. Flies were reared from egg to eclosion in 0.2-liter glass jars (9.0 cm height × 5.5 cm diameter). Egg collections were performed on sufficient animals (150-200 10-12-day-old flies per 0.2-liter jar, 20 jars in total) so as to obtain a number of eggs adequate for the entire study from a single oviposition. Females were allowed to lay eggs for one hour and then dumped. Immediately after egg-laying, we counted the number of eggs in each jar. To avoid the unfavourable effects of larval crowding, ie increased duration of development and decreased imaginal body size (Economos and Lints 1984), 4 overcrovded jars were removed and only jars with normal egg density (~ 80-100 eggs per jar, 16 jars in total) were used.

Flies in 8 jars were reared to adulthood under normal 12:12 LD cycle (light: 12 h; dark: 12 h) with light from 8 a.m. to 8 p.m. (period T24), and flies in 8 jars were reared to adulthood under long-period 24:24 LD cycle (light: 24 h; dark: 24 h) with light from 8 a.m. of one day to 8 a.m. of the next day (period T48).  Flies reared under T24 period were divided in two groups which were then maintained as adults under both T24 and T48 periods. Similarly, flies reared under T48 period were also maintained as adults under both T24 and T48 periods. The light phase in these treatments was achieved by means of fluorescent white light sources (with a light intensity of ~ 100 lux). The design of the experiment is presented in Table 1.

Measurement of egg-to-adult developmental time and viability

In each T24 and T48 developmental LD regime, pre-adult viability measured as percentage of viable eggs (i.e. ratio of living imagoes to number of eggs laid) and egg-to-adult development time (measured from the midpoint of the egg-laying period to the midpoint of the adult eclosion in each jar) were estimated in 8 jars. Adult eclosion (based on the number of empty pupae) was recorded two-hourly during the whole period of emergence.

Longevity test

For determination of adult life span, flies with identical duration of development (10 days) were collected within 24 h after emergence, etherized, separated according to sex, and placed in groups of 25 in glass vials (2.2 cm in diameter ´ 10 cm high), containing 3 ml of standard fly culture medium. The flies were transferred to new vials containing fresh medium three times a week. Dead flies were removed and recorded on the same terms until the last death.  Six replicates (149-150 flies in total) were used for life span testing in each group. One male and one female were lost during the transfers.

Statistics

The Student’s t-test was used to compare the differences in developmental time, and z-test for significance of difference between proportions was used to compare the differences in pre-adult viability between flies reared under normal (T24) and long-period (T48) LD cycles. The effect of sex, pre-adult LD and adult LD on Drosophila longevity was assessed by a  three-way analysis of variance (ANOVA), followed by Tukey's honestly significant difference (HSD) test for post-hoc comparisons among means.

Results

In accordance with our previous results (Voitenko et al. 2006), the flies reared during pre-adult stages at the long-period (T48) LD cycle had significantly prolonged developmental time compared to those of flies reared at normal 24-h period (t = 4.06, d.f. = 6, P < 0.01) (Table 2). Flies reared in these conditions have decreased pre-adult viability. This decrease in proportion of survived insects was small in magnitude (95%  in T24 flies vs. 91% in T48 flies) but statistically significant (z = 2.92, P < 0.01).

There is enough evidence that circadian clocks of insects are labile and pre-adult light regimes can influence the adult circadian organization (so-called “after-effects”). For checking a possible after-effect of pre-adult LD regime, in our study flies were reared and maintained under both identical (T24/T24 and T48/T48) or different (T24/T48 and T48/T24) length of LD cycle during the pre-adult and adult stages. The data obtained showed no after-effect of pre-adult LD regime on adult life span. A three-way ANOVA demonstrated a significant effect of sex [F(1,1190) = 155.6, P < 0 .001] and  adult LD regime [F(1,1190) = 48.2, P < 0 .001] but no significant effect of pre-adult LD regime [F(1,1190) = 1.98, P = 0.16] on the flies’ longevity; all interactions were also nonsignificant (P > 0.05 for all). Irrespective of the duration of pre-adult LD period, the flies maintained as adults at extended LD period (T48) tend to live longer than those maintained at normal LD period (T24) (Figure 1).

Discussion

A set of hypotheses was proposed earlier to explain the association between chronobiological conditions and ageing rate. Circadian dysfunction could reduce life span because it cause desynchronization of various cyclic metabolic processes within the body (Daan and Aschoff 1982). Massie and Whitney’s hypothesis postulates that visible light can be a major factor in the ageing process for Drosophila and that photochemical effects may contribute to senescence in other organisms (Massie and Whitney 1991). According to the redusome hypothesis recently proposed by Olovnikov (2003), the aging of an organism is determined by the shortening of chronomeres (small perichromosomal linear DNA molecules). The rate of their shortening could be influenced by the organism’s biological rhythms. The author suggests that infradian (longer than 24 hours) biorhythms of neuroendocrine system activity (so-called T-rhythms) are the best candidates for the role of these rhythms, and peaks of T-rhythms are used as pacemaker signals to keep the life-long “clockwork” of the brain running (Olovnikov 2003). The “ticking” of this clock is realized by the periodically repeated shortening of chronomeres in postmitotic neuroendocrine cells, which occurs just at the maxima of T-rhythms. Subsequent development and deepening of Olovnikov's hypothesis should include data on circadian (about 24-hour) and other oscillations (Anisimov 2003).

It is generally believed that faster clocks speed up development and cause reduction in life span, while slower clocks slow down development and lengthen life span (Kumar 2005; Kyriacou et al. 1990; Massie et al. 1993; Paranjpe and Sharma 2005; Paranjpe et al. 2005). The endogenous genetically determined timers were shown to be involved in regulating the pre-adult development time of D. melanogaster. The per^s mutants, which have short 19-hr endogenous (free-running) period of circadian rhythms (tau), develop faster from eggs to adult than the wild-type; per^l mutants, which have long 28-hr tau, complete development more slowly than the wild-type flies (Paranjpe and Sharma 2005). Some controversial results were obtained regarding the role of circadian clocks in life span determination. A few studies have shown that genetic and environmental manipulations in circadian clocks accelerate senescence, thereby reducing life span. The heterozygous hamsters (tau = 22 h) lived significantly shorter than the wild-type animals (tau = 24 h) under a LD schedule of 14:10 h, whereas the homozygous animals (tau = 20 h) survived equally well as the wild-type animals (Hurd and Ralph 1998). However, in another study, tau mutant homozygous hamsters survived significantly longer than the wild-type controls under constant darkness, whereas no difference in survivorship was recorded between the heterozygous and the wild-type animals (Oklejewicz and Daan 2002). Male Drosophila melanogaster mutants per^s (short-period, tau = 19 h), and per^l (long-period, tau = 29 h) were compared to the wild type in two different LD schedules. Life span was used as a global index of physiological adaptation. The life span of the mutants was significantly reduced by up to 15% for the flies whose period differs most from that of the wild type (Klarsfeld and Rouyer 1998). Drosophila melanogaster, which had been reared under standard conditions, were exposed, on the first day of adult life, to four environments as follows: (1) a 24-hr day consisting of 12 hr light and 12 hr dark; (2) a 21-hr day (10.5 hr light, 10.5 hr dark); (3) a 27-hr day (13.5 hr light, 13.5 hr dark); and (4) constant light. The experiment was repeated four times. In all four experiments the flies on a 24-hr day lived significantly longer than the flies in the other environments (Pittendrigh and Minis 1972). Thus, longevity is greater when organisms are maintained in environments with periodicity close to their endogenous periodicity, a phenomenon termed circadian resonance by Pittendrigh and Minis (1972).

Contrary to previous studies, we found that life span of flies maintained under 48-h LD period was significantly increased compared to those maintained under “normal” 24-h LD period. One plausible explanation for this contrary finding is that 48-h LD period, which is exactly double the normal period, is the sort of change that is most likely not disrupt normal activity cycle. It can be viewed as a harmonic oscillator. A useful example of such oscillator is a child’s swing (Betts 1989). A child on a swing oscillates at a natural frequency that depends on the length of the swing. In order to maintain or increase the amplitude of the swing the frequency of the pushes must be timed to coincide with the frequency of the swings. The amplitude of the swing oscillation could still increase if the natural frequency is an integer multiple of the input frequency. The pushes could be supplied after every second swing and still increase the amplitude. However, the amplitude of the swings would decrease if the pushes were supplied out of sequence. The other possible explanation is that the 48-h LD period could be a natural component of endogenous rhythmicity in Drosophila as against those periods used in previous studies. Indeed, it has been shown in some studies that the labile period of evening oscillator sometimes lengthens to synchronize with the stable period of morning oscillator in a 2:1 mode, to cause the day-skipping (double-length “circabidian” rhythms). In rodent studies, it has been shown that under the periodic food restriction some rats showed circabidian rhythms, i.e., an activity band appeared at every second food presentation (Honma et al. 1992). In human beings, a shift from circadian to circabidian periods or vice versa was repeatedly observed in the rhythm of sleep and wakefulness under temporal and social isolation (Aschoff 1994; Chandrashekaran et al. 1997; Honma and Honma 1988). Among insects, circabidian rhythmicity was found in the flight activity of mosquito Culiseta incidens recorded in constant darkness as well as in cockroach (Clopton 1984; Page 1989). Taken together, these results suggest that circabidian rhythmicity is a natural component of endogenous multi-oscillator system for many organisms including insects. If it is so, the 48-h LD cycle applied in our study could be synchronized with endogenous rhythms of the flies.

A potential limitation in generalizing the results obtained in present study could be that the effect of the duration of LD period on life span can include changes in flies’ spontaneous locomotor activity and metabolic rate. However, it seems impossible because in our study, in both regimes studied (T24 and T48) the light was on 50% of the time, thus, the insects in all experimental groups were exposed to essentially the same total illumination over the long duration of their lives. Besides, according to a number of studies, spontaneous locomotor activity and life span are not correlated in D. melanogaster (Le Bourg 1987; Lints et al. 1984; Melvin et al. 2007; Van Voorhies et al. 2004). We plan to monitored adult rest/activity rhythm in the flies maintained under normal-period (T24) and long-period (T48) conditions in our future investigations.

ACKNOWLEDGEMENTS