The morning burst: Shifting daily patterns of ATP production in Drosophila and temporal windows for their improvement in ageing

Mitochondria produce energy for cell function via adenosine triphosphate (ATP) and are regulated by a molecular 24h clock. Here we use Drosophila melanogaster to reveal shifts in whole animal ATP production over 24h, showing a marked peak in the morning that declines around midday and remains low from then through to the following morning. Mitochondrial membrane potential and ATP production has been shown previously to improve after long wavelength exposure, but apparently not at all times. Hence, to explore this further we exposed flies to 670nm at different times. Exposures between 08.00 and 11.00 resulted in a significant increase in ATP, while exposures at other times had no effect. Within the morning window, not all times were equally effective, however, 670nm exposure mid-morning when ATP production was maximal did not increase ATP, possibly because mitochondria lacked spare capacity at this time. Hence, in the morning there is a complex dynamic relationship between long wavelength light and mitochondria. Mitochondrial function and the influence of long wavelengths are conserved across species from fly to human, and determining the time points for light administration to improve function in ageing and disease is of key importance. Our data progress this search and reveal the outline of these times.


38
Many physiological processes are regulated over a 24-hour period to adapt to environmental changes 39 of day and night (1,2). Such regulation includes the sleep-wake cycle, locomotor activity and feeding 40 during the active phase (3-11). It is able to keep running even under constant 12 /12-hour light/dark 41 environmental conditions.

42
There is evidence that the function of mitochondria, the major metabolic hubs that produce ATP for 43 cellular function, is also regulated by a 24-hour clock (12) This is not only in terms of their metabolic 44 function but also in terms of their complex dynamics that, include fusion and fission processes that are 45 central to the maintenance of their integrity and whose disruption is associated with ageing and 46 disease (13-15).

47
With age and many diseases mitochondrial respiration declines (16) and their circadian behaviour is 48 disrupted (13). However, aged mitochondrial respiration can be improved with longer wavelengths of 49 light (650-900nm). This range of wavelengths has been shown to improve mitochondrial membrane 50 potential (17) and ATP production (18). These improvements have been associated with increased 51 longevity and neuroprotection in ageing in insects and mammals (19)(20)(21). Recently, it has been 52 reported that such improvements cannot be achieved consistently over the 24-hour period, and that 53 there may be an optimum period of exposure that selectively improves mitochondrial function (22).

54
The precise time point(s) during the day for generating the maximum impact for improving 55 mitochondrial function is not known. In this study, we determine the shifting pattern of mitochondrial 56 activity by investigating ATP production over a 24-hour period in whole aged flies. Thereafter, we 57 explored whether there are specific periods of sensitivity when long wavelength light (670nm)  experimental group had a matched control whereby they were exposed to the same 12/12-hour 83 lighting. The fly vials were examined with a spectrometer and a radiometer prior to 670nm exposure.

84
They reduced the overall energy applied by <5% and did not modify the wavelength.

85
ADP/ATP. Flies exposed to 670nm were processed for changes in both ATP and the ADP/ATP In order to examine the effect of 670nm light on the varying patterns of ATP production, flies were 108 treated for 15 minutes daily over 7 days ( Figure 2) at six different time points that corresponded to 109 periods of change in the overall 24h patterns of ATP production ( Figure 1). The first two time points 110 were at 08.00 and 09.00, on the ascending slope of the peak of ATP production in the morning. Third

111
was on the top of the slope when the ATP level was at its maximum at 10.30. The fourth was at the 112 descending slope in the morning at 11:00. The fifth, corresponded to the early phase of the afternoon 113 plateau at 16:00 and sixth in the dark phase at 05:30 when ATP production was close to its minimum.

114
Changes in ATP production as a consequence of 670nm light exposure were confined to the morning 115 period. However, even within this period, 670nm applied at different exposure times did not result in 116 consistent changes, indicating that a more complex relationship exists between 670nm light exposure 117 and mitochondrial function than simply a sensitive/insensitive switch. There were significant 118 increases in ATP at 08.00 (P = 0.0003, increase of 27%) and 11:00 (P = 0.0084, increase of 11%).

119
However, at the peak of ATP production at 10.30 (Figure 1), 670nm light exposure was associated 120 with a decrease in ATP levels. This may be because at this point, mitochondrial ATP production was Finally, we examined whether the changes in ATP production during the day were mirrored also by 146 changes in ADP by examining ADP/ATP ratios at times when 670nm was applied. Overall, our 147 results showed that if ATP was not changed by 670nm exposure, neither was the ADP/ATP ratio.

148
However, when ATP was elevated during the 24hr cycle, the ADP/ATP ratio declined and when ATP 149 was reduced, the ADP/ATP ratio increased. The flies treated at 08.00 and 11:00 showed a significant 7 150 decrease (P= 0.0051 and P= 0.0003 respectively) in ADP/ATP ratios, likely to be due to ADP 151 conversion to ATP, and a release and utilisation of cellular energy.

152
There was a small, but significant increase in ADP/ATP ratios in flies treated with 670nm red light at 153 09:00 (P = 0.0265), reinforcing the notion that this time point may be one of transition and a 154 significant increase at 10.30 (P = 0.0189) compared to controls ( Figure 2B). Hence, at these time 155 points it is likely that ATP was being converted to ADP, potentially as an energy storage mechanism.

156
There were no significant differences in ADP/ATP ratios at the other times examined, 16.00 and 5.30 157 ( Figure 2B) when 670nm light had no impact on ATP levels ( Figure 2A).

160
This study had two major findings. First, that total ATP levels in flies vary considerably over a 24h hour and showed variations in peaks of production. Although the key difference was in the production, the authors suggested that that variation was due to the different activity patterns and 178 energy demands of each region. In essence, targeting one organ does not give one an overall 179 understanding of how ATP levels fluctuate over a 24-hour period in a whole organism and this is why 180 we chose to use the Drosophila fly model.

181
Our results showed that flies elevate ATP production early in the day, peaking around 10:30. This 182 may be for the normal active physiological function for the rest of the day. Hence, flies increase 183 mitochondrial function when lights were turned on, presumably related to an increased energy 184 demand and resumption of normal activity. This result may be reflective of their overall feeding 185 patterns. Xu et al. 2008 (33) have reported that, flies start to feed prior to or around the time of lights 186 ON in the morning but then decrease their feeding behaviour between 4-6-hours after lights ON. This 187 matched closely our peak ATP level at 10:30, which is 4.5-hours after lights ON and the general 188 decline in levels thereafter. Further, Xu et al. 2008 (33) showed a small peak in the fly feeding pattern 189 at 8-10-hours after lights ON just before onset of night, which is reflected broadly here in our finding