Trex1 overexpression leads to longer lifespans and fragmented sleep in Drosophila melanogaster

TL;DR

Overexpressing the Trex1 gene in fruit flies leads to longer male lifespans but disrupts sleep patterns and circadian rhythms.

Contribution

This study reveals the first in vivo effects of Trex1 overexpression on lifespan and sleep in Drosophila.

Findings

Trex1 overexpression significantly extended male fly lifespans but not female lifespans.

Trex1 overexpression caused fragmented sleep and reduced circadian rhythm robustness.

The effects on lifespan and sleep appear to involve distinct mechanisms.

Abstract

Three-prime repair exonuclease 1 (Trex1) prevents cytosolic DNA accumulation and immune activation, yet the physiological consequences of increased Trex1 expression in vivo remain unclear. In this study, we used Drosophila melanogaster, a model well suited for quantitative analyses of aging, sleep, and circadian rhythms, to generate flies that ubiquitously overexpress murine Trex1 under the Act5C-GAL4 driver, given that a clear Trex1 ortholog has not been identified in flies. Trex1 overexpression significantly extended the lifespan of male flies (40.55 ± 1.10 days in controls; n = 60 vs. 44.98 ± 1.59 days in Trex1; n = 57, p < 0.05, Bonferroni corrected), whereas female flies showed a modest but statistically non-significant increase in lifespan. Interestingly, Trex1 overexpressing flies exhibited more fragmented sleep and reduced circadian rhythm robustness compared with controls. These findings suggest that the mechanisms by which Trex1 overexpression influences lifespan and those regulating sleep and circadian stability may be functionally separable.

Figures

A schematic of method for creating Trex1 overexpression flies. Male Act5C-GAL4 flies were bred with virgin female UAS-Trex1 flies, with the control flies distinguished from overexpression flies using the CyO (Curly wings) genetic marker (A). Protein expression levels were confirmed by immunoblot analysis (B)

Results of lifespan assay in Trex1 overexpression. The Trex1 overexpression and control populations were monitored for 55 days for both females (A) and males (B), with the number of deaths recorded every other day until Day 30, after which records were made daily. The LD50 value and significance was calculated via OASIS2 software. The impact of paraquat on male lifespan was also measured (C), with a similar LD50 calculation made using OASIS2 software

Results of sleep assays reveal significant differences between Trex1 overexpression and control flies with respect to bout length (A, B), bout number (C, D), and sleep duration (E, F) for both female and male flies. Data was collected on Trikinetic monitors and evaluated using ClockLab software. Error bars denote standard error of the mean, and p* < 0.05 when comparing Trex1 overexpression and control lines of the same sex and age. Per three-way ANOVA, it was found that there was a significant line effect (*p* < 0.0001) for the length of average bouts and bout numbers, for both day and night.

Results of the sleep deprivation assay where deprivation was induced via 12-hour continuous shaking, with comparisons between the pre-deprivation, deprivation, and rebound states. Sleep duration was separated into day (A, B), night (C, D), and total (E, F) for both males and females. Error bars denote standard error of the mean. Per three-way ANOVA, there was a significant line effect for night sleep (*p* < 0.0009) in the perturbed flies that was not seen in the still flies. Additionally, there is a trend toward significance with regard to genotype in both night sleep and total sleep for the r

Full text

Introduction

Aging is accompanied by progressive physiological decline across multiple organ systems, and chronic low-grade inflammation has been recognized as an important contributor to this process [1]. Three-prime repair exonuclease 1 (Trex1), also known as DNase III, is a protein found in the cytoplasm and degrades both single- and double-stranded DNA [2, 3], and plays a role in regulating cellular autoimmune pathways [4, 5] through cGAS which plays a role in type-1 interferon activation. Trex1 deficiency has been implicated in the inflammatory myocarditis, cardiomyopathy and circulatory failure [6]. The TREX1 D18N mutation, which exhibits dysfunctional dsDNA-degrading activity, causes autoimmune diseases within the spectrum of lupus-like diseases [7]. In addition, TREX1 deficiency induces persistent DNA damage and replication stress, leading to sustained activation of the DNA damage response and progressive cellular senescence [8, 9]. Although the cellular and molecular consequences of Trex1 deficiency have been extensively characterized, very little is known about how increased Trex1 expression affects organismal aging. In particular, it remains unclear whether elevating Trex1 levels can modulate lifespan in an in vivo system, representing an important research gap in our understanding of Trex1’s broader physiological roles.

In the past, several researchers have used Drosophila melanogaster as a model to examine the role of inflammation in physiology, disease and aging [10]. Advantages of using fly models to study the physiological effects of gene overexpression include its short lifespan, well-documented and easily manipulated physiological systems, and the ability to generate large data sets due to the organism’s high fecundity. In particular, the effect of aging on fly’s sleep cycle and circadian rhythm is well characterized [11] and reflective of sleep patterns in humans, wherein sleep and circadian rhythm become more fragmented with age [12].

Based on this rationale, we formulated the testable hypothesis that increased Trex1 expression would modulate aging phenotypes, including lifespan and age-associated alterations in sleep continuity and circadian rhythm stability. To evaluate this hypothesis, we generated transgenic flies that ubiquitously overexpress murine Trex1 under the Act5C-GAL4 driver and systematically examined the effects of Trex1 overexpression on lifespan, oxidative stress sensitivity, sleep parameters, and circadian rhythm robustness. In this study, we found that Trex1 overexpression extended lifespan through mechanisms independent of oxidative stress. Interestingly, and contrary to our initial hypothesis, Trex1 overexpressing flies exhibited more fragmented sleep patterns and reduced circadian rhythm robustness compared with controls. These findings suggest that the mechanisms through which Trex1 influences lifespan may be distinct from those governing sleep architecture and circadian stability.

Materials and methods

Generation of Transgenic flies

The full-length mouse Trex1 gene (Mus musculus, transcript variant 1, isoform 1; NCBI RefSeq accession NM_011637) was cloned into the pUAST vector containing a 5× UAS sequence, and transgenic Drosophila melanogaster carrying the pUAST-Trex1 construct were generated by BestGene Inc. (Chino Hills, CA). The inserted Trex1 transgene was fully sequence-verified to confirm that the entire coding region matched the reference full-length sequence. The Trex1 transgene was inserted into the second chromosome, and fly stocks were maintained on standard BDSC cornmeal medium. To induce ubiquitous Trex1 overexpression, virgin UAS-Trex1 females were crossed with Act5C-GAL4 males (Bloomington Drosophila Stock Center, BL#3953), in which GAL4 is driven by the actin promoter. Trex1 overexpressing flies were identified based on the absence of the curly wing phenotype associated with the CyO second-chromosome balancer and the absence of white-eye pigmentation.

Protein expression in Transgenic flies

To ascertain the presence of murine Trex1 in the transgenic flies, protein was isolated from Trex1 transgenic flies using RIPA lysis buffer (EMD Millipore) containing protease inhibitors (Roche). Total protein concentrations were determined using the Coomassie Plus Protein Assay (Thermo Fisher Scientific) and equal amounts of protein were subjected to immunoblot analysis. Immunoblotting was performed with Trex1 antibody (Cat No. 611987, BD Biosciences), and β-actin (Cat No. ab8224, abcam) was used as a loading control.

Lifespan assay

Drosophila lifespans were measured in male and female virgins emerging within a 3-day period. Flies of each sex and genotype were separated and kept on BDSC cornmeal food at 10 flies to a same-sex vial in standard light-dark cycles of 12:12 h. Flies were monitored, and deaths recorded every other day up to day 30, after which they were monitored every day. The flies were transferred to new vials with fresh food every 3 days. The flies were kept in a controlled incubator with 12 h:12 h light: dark cycles, temperature (25˚C) and humidity (60%). Survival analysis was performed using OASIS2 software [13].

Sleep and circadian rhythm assay

Sleep assays were conducted using Drosophila Activity Monitors (DAM2, Trikinetics, Waltham, MA), placed in incubators at constant temperature (25˚C) and humidity (60%). The sleep patterns of young (7 days) and old (35 days) flies were measured. Sleep, circadian rhythm, and sleep deprivation assays were conducted according to methods previously described [14]. Sleep and circadian rhythm data were gathered sequentially, with the monitors at a 12 h:12 h light: dark cycle to measure sleep for the first 4 days, and then switched over to constant darkness (DD) to measure circadian rhythm for another 10 days.

Data from the DAM monitors were consolidated using DAM scan software and analyzed using Sleep Analysis v6-1 software (R. S. Barnes, personal communication) for sleep and ClockLab (Actimetrics, Wilmette, IL) for circadian rhythm. The Sleep Analysis V6-1 program calculates night and day sleep duration in minutes, night and day sleep bout number, and night and day average sleep bout length; it also calculates sleep latency, the time in minutes to the first sleep bout after lights are turned off and waking activity, the number of activity counts per minute spent awake. Inactivity recorded for 5 min or more was regarded as sleep. Using ClockLab program, the length of the circadian rhythm was determined using Chi-square periodogram analysis, and the rhythm strength was measured by the amplitude of a Fast Fourier Transform (FFT). Significance was calculated using a three-way ANOVA analysis with respect to sex, age and genotype.

Sleep deprivation for 7 day old flies was invoked using mechanical perturbation via Trikinetics Mounted Monitor, which gave a 5 s perturbation every minute at intensity 2. Data were recorded for five days, with perturbation occurring on day 3 for 12 h during the dark cycle. Duplicates were set up on a stable shelf in the same incubator as the non-deprived control. Data was analyzed using the same methods as the sleep assay, with the sleep loss and rebound calculated by comparing the deprived and non-perturbed populations on day 3 night for sleep loss and the cumulative hours slept on days 4 and 5 for sleep rebound. Significance was calculated by using a three-way ANOVA analysis with respect to sex, genotype, and perturbation.

Paraquat assay

Paraquat, a pesticide that increases levels of reactive oxygen species (ROS) within an organism, was used to test the Trex1 overexpression fly’s resistance to oxidative stress [11, 15, 16]. After 2 days of starvation, male Trex1 overexpression (N = 72) and control (N = 90) flies were exposed to 20 mM of paraquat via filter paper, where they were also provided 5% sucrose with yeast sprinkle to control for starvation. Observations were made daily, and dead flies were counted. Survival analysis was made using OASIS2 [13].

Results

To assess the effects of increased Trex1 expression in vivo, we generated transgenic Drosophila melanogaster overexpressing murine Trex1. The UAS-Trex1 transgene, inserted on the second chromosome, was crossed with Act5C-GAL4 to achieve ubiquitous expression under the actin promoter. Trex1 protein overexpression in F1 progeny was confirmed by immunoblot analysis (Fig. 1A and B). All experiments were performed using a single, sequence-verified insertion line, and Trex1 overexpression levels were reproduced consistently across independent biological replicates.

Fig. 1A schematic of method for creating Trex1 overexpression flies. Male Act5C-GAL4 flies were bred with virgin female UAS-Trex1 flies, with the control flies distinguished from overexpression flies using the CyO (Curly wings) genetic marker (A). Protein expression levels were confirmed by immunoblot analysis (B)

Overexpression of Trex1 increased the lifespan of Drosophila melanogaster populations compared to the control. The female of Trex1 transgenic flies lifespan was increased from 37.35 ± 1.67 days in the control line to 40.13 ± 1.54 days in the Trex1 overexpression line, though the difference is not significant (p > 0.05 Bonferroni corrected) (Fig. 2A); however the male lifespan was increased significantly from 40.55 ± 1.1 days to 44.98 ± 1.59 days (p < 0.05 Bonferroni corrected) (Fig. 2B).

Fig. 2. Results of lifespan assay in Trex1 overexpression. The Trex1 overexpression and control populations were monitored for 55 days for both females (A) and males (B), with the number of deaths recorded every other day until Day 30, after which records were made daily. The LD50 value and significance was calculated via OASIS2 software. The impact of paraquat on male lifespan was also measured (C), with a similar LD50 calculation made using OASIS2 software

Reactive oxygen species (ROS) can induce DNA damage, protein misfolding and aggregation through altering their structure and function [17–19]. Such oxidative stress has been shown to accelerate aging and shorten lifespan by driving the accumulation of cytoplasmic DNA, which in turn promotes cellular senescence [16, 20, 21]. To further characterize the potential roles of Trex1 in oxidative stress-induced lifespan defects, Trex1 overexpression (N = 72) and control (N = 90) flies were exposed to paraquat which potent ROS inducer. Trex1 overexpression did not have a significant impact on oxidative stress resistance, as the overexpression and control fly populations had similar paraquat-induced death curves with an average lifespan of 4.6 ± 0.3 days for the overexpression flies and 4.5 ± 0.2 for the control flies (Fig. 2C). These results suggest that Trex1 overexpression does not confer an advantage in acute oxidative stress resistance, consistent with the observation that oxidized self-DNA is relatively resistant to Trex1-mediated degradation [22]. Together, these findings indicate that Trex1 can extend lifespan through mechanisms that are largely independent of classical oxidative stress resistance.

Aging [11] and DNA damage [23–25] are known to affect the circadian clock and sleep architecture. Accordingly, we used the DAM2 monitoring system to quantify standard sleep parameters in Trex1 overexpressing flies, including daytime and nighttime sleep duration, the number of sleep bouts, and mean sleep bout length. As per three-way ANOVA, the Trex1 overexpression flies had a significant effect on average sleep bout length and the number of sleep bouts for both day and night, as well as for the duration of sleep during the night as compared to those in the controls (Fig. 3). Post hoc analysis revealed that the mean sleep bout length of Trex1 overexpressing flies was significantly reduced compared to the control group in female flies during the daytime under both young (7-day) and old (35-day) conditions (Fig. 3A). In male flies, a consistent reduction was observed across all age and circadian phase conditions, except during the daytime in the old group (Fig. 3B). In contrast, the number of sleep bouts was significantly increased under all conditions except during the nighttime in young female groups (Fig. 3C and D). Collectively, these results indicate that the observed shortening of bout length, together with the concomitant increase in bout number, act in a complementary manner, suggesting a marked enhancement of sleep fragmentation in Trex1 overexpressing flies compared with controls.

Fig. 3. Results of sleep assays reveal significant differences between Trex1 overexpression and control flies with respect to bout length (A,** B**), bout number (C,** D**), and sleep duration (E,** F**) for both female and male flies. Data was collected on Trikinetic monitors and evaluated using ClockLab software. Error bars denote standard error of the mean, and *p < 0.05 when comparing Trex1 overexpression and control lines of the same sex and age. Per three-way ANOVA, it was found that there was a significant line effect (p < 0.0001) for the length of average bouts and bout numbers, for both day and night. Night sleep duration was also significant (p < 0.002) and total sleep duration (p < 0.01) but was not significant for day sleep duration

In the sleep deprivation assay, there was a significant genotype effect in the duration of sleep during the night of the shaking that was not seen in the undisturbed flies. Additionally, there was a stronger tendency towards significance in total sleep duration in the rebound period seen in the shaken flies than the disturbed flies, but it was not less than 0.05 (Fig. 4).

Fig. 4. Results of the sleep deprivation assay where deprivation was induced via 12-hour continuous shaking, with comparisons between the pre-deprivation, deprivation, and rebound states. Sleep duration was separated into day (A,** B**), night (C,** D**), and total (E,** F**) for both males and females. Error bars denote standard error of the mean. Per three-way ANOVA, there was a significant line effect for night sleep (p < 0.0009) in the perturbed flies that was not seen in the still flies. Additionally, there is a trend toward significance with regard to genotype in both night sleep and total sleep for the rebound values of the shaken flies

The circadian rhythms of the flies were also assayed under DD conditions for 10 days. There was no discernible difference in the period length of the flies’ rest-activity rhythms seen via three-way ANOVA. However, there was a significant (p < 0.001) difference in rest-activity robustness, measured by fast Fourier transformation (FFT) amplitude in female flies; the Trex1 overexpression flies had a significantly lower amplitude of 0.0053 A.U. in young and as well as in old ages 0.0029 A.U. when old, as compared to the similarly aged control flies − 0.0094 A.U. (young) and 0.0050 A.U. (old). Old male flies also exhibited a difference in FFT amplitude, with overexpression flies having an average amplitude of 0.0021 A.U. compared to the control value of 0.0029 A.U. (Table 1), though the value was not significant. This indicates that Trex1 overexpression may lead to weakened rest-activity rhythms in the older individuals as compared to the similar-aged controls.

Table 1. Value of circadian rhythm periods calculated via Chi-square test and Lomb-Scargle approximation, as well as the fast fourier transformation (FFT) amplitudes, used as a measure of cycle stability. Circadian rhythm was assayed for both male and female flies at young (7 days) and old (35 days) age

Discussion

Our results showed that Trex1 overexpression in Drosophila melanogaster led to a significant improvement in male lifespan. However, we found that Trex1 overexpression flies had more frequent bouts of sleep and slept for a shorter time within each bout compared to the control. Additionally, the Trex1 overexpression flies had a greater sensitivity to sleep deprivation, as seen in the differences in the amount of sleep lost during perturbation, which was significant during the night, and the amount of sleep required for rebound, which was not significant. This phenotype is generally associated with old age, which is counterintuitive to the effect Trex1 has on fly lifespan.

These findings suggest that the benefits of increased longevity may come with a trade-off in neural and behavioral homeostasis. In other words, although sleep exhibits clear aging-associated characteristics, the pathways regulating sleep may operate at least partially independently of the broader aging process. Because Drosophila circadian rhythms are regulated not only by photoreceptive input but also by metabolic signals [26, 27], Trex1 overexpression may have disrupted signaling processes involved in maintaining sleep continuity and circadian rhythm robustness. One study found stress response genes protective against sleep deprivation in Drosophila [28], as mutations in heat shock protein 83 (Hsp83) conferred the same sensitivity to sleep deprivation as a loss-of-function clock mutant fly (cyc^01^). Additionally, it was found in a human study that short-term sleep deprivation decreased the level of redox metabolite found in human subjects [29]. Our study suggests that the converse relationship may also be true: overexpressing Trex1, which may decrease inflammation, is associated with increased sensitivity to sleep deprivation [30–33]. At present, it remains unclear whether the behavioral alterations associated with Trex1 overexpression arise from direct neuronal mechanisms, such as changes in DNA metabolism within neural cells, or whether they are mediated indirectly through systemic immune or metabolic signals.

Future studies are needed to investigate the potential role of Trex1 in a mouse model, which would allow systems-level evaluation of how Trex1 overexpression influences cellular signaling, as Trex1 is native to mice and embedded within their complex signaling architecture In particular, it will be important to determine whether the discrepancy observed in Drosophila between lifespan extension and aging-related changes in sleep and circadian rhythm is conserved in mammals. Moreover, behavioral assessments such as tests of memory, appetite, and motivational or sexual behaviors may help determine whether Trex1 influences organismal functions beyond the lifespan extension observed in flies.

In addition, it will be important to clarify the tissue-specific, circuit-level, and molecular mechanisms underlying the sleep and circadian phenotypes observed in Trex1 overexpressing flies. Because this study used the Act5C-GAL4 driver to induce ubiquitous Trex1 expression, it remains difficult to determine whether lifespan extension and alterations in sleep architecture originate from the same tissues or from distinct anatomical sites. Selectively manipulating Trex1 expression in neurons, glia, or peripheral metabolic tissues would help identify which tissues contribute to lifespan extension and which are responsible for sleep fragmentation and reduced rhythm robustness. Furthermore, examining how Trex1 expression varies across the sleep cycle, together with transcriptomic and metabolomic analyses, may help elucidate the downstream molecular pathways linking Trex1 activity to sleep homeostasis, energy metabolism, and neural excitability.