The C. elegans strains: the Bristol N2 (wild type), JK1107 (glp-1), GA480 (sod-2/-3), OK2040 (mpst-1), TK22 (mev-1), and CB1370 (daf-2) were used in this study and were obtained from the Caenorhabditis Genetics Centre (University of Minnesota, Minneapolis, MN, United States). All C. elegans strains were grown on nematode growth medium (NGM) agar plates at 20°C except for JK1107 (glp-1) strain, which was grown at 25.5°C to prevent progeny. Preparation of NGM agar plates was as previously described in Stiernagle (2006). All experiments used day 4 post-hatching age-synchronized nematode obtained by hypochlorite bleaching.

Lifespan of C. elegans was observed as described previously under blinded conditions to eliminate experimental bias (Schaffer et al., 2011). The numbers of alive and dead worms were scored. The surviving worms were transferred to fresh plates every 2 days until the post egg-laying period. Worms that failed to move in response to mechanical prodding were scored as dead. Worms that died due to crawling off the plates were censored.

Locomotion activity of the worms were scored following the previously reported scoring framework (Herndon et al., 2002). The locomotion activity of the worms was classified into three classes: Class A, B, and C. Class A animals were healthy and moved spontaneously, class B animals moved in respond to prodding, produced non-sinusoidal tracts while class C animals moved their head and/or tail only.

The relative distance traveled of 10–20 worms of each treatment condition was measured and analyzed under blinded conditions. A mark was made just at the border of the bacterial lawn to indicate the starting point of distance to be traveled. In general, individual worms were transferred onto the marking on the plate and allowed to travel for 15 min. Worms were removed and a single photograph of the whole NGM agar plates was captured using a calibrated Leica MZ10F microscope (Leica, Singapore). The line tool provided by the Leica Application Suite Software (v2.6.0 R1) was utilized to measure the distance traveled by the worms.

MtDNA was extracted as described elsewhere (Yasuda et al., 2006) with several modifications. About 10,000 worms were homogenized in isolation buffer. Debris and nuclei were removed via differential centrifugation at 600 g for 10 min at 4°C. The pellet was discarded and the supernatant was centrifuged at 7200 g for 10 min at 4°C to obtain mitochondria pellet. DNA from crude mitochondria was purified using Prepman Ultra Sample Preparation Reagent (Applied Biosystems) according to manufacturer’s protocol.

MtDNA copy number of a reference sample was quantified using mtDNA copy number method as described in Gruber et al. (2011) and Poovathingal et al. (2012). Briefly, mtDNA copy number of individual nematodes was determined by quantitative real-time PCR (qRT-PCR) amplifying a 71 bp region of the C. elegans mitochondrial genome using the worm lysate as the source of DNA template and the mtDNA (short fragment) primers and probe [Forward primer: 5′-GAG CGT CAT TTA TTG GGA AGA AGA-3′ (nucleotides 1838–1861 mtDNA); Reverse primer: 5′-TGT GCT AAT CCC ATA AAT GTA ACC TT-3′ (nucleotides 1883–1908 mtDNA) and Probe: 5′-FAM-AAA ATC GTC TAG GGC CCA C-3′ (nucleotides 1863–1881 mtDNA)]. The probe was labeled with a specific reporter (FAM-labeled) and has non-fluorescent quencher (MGB Probes). This assay was performed using a reference sample of known copy number (quantified by serial dilution generating actual rather than relative copy numbers).

After determining the mtDNA copy number of the reference sample, Sequence-specific mtDNA damage quantification was performed as described elsewhere (Meyer et al., 2007; Gruber et al., 2011) using GeneAMP XL PCR kit (applied Biosystems). Sequence-specific mtDNA damage was quantified in a 6.3 kbp region of the mitochondria genome using sybr green dye with previously reported primers sequence (Hulbert et al., 2007) (Forward primer: 5′-TCG CTT TTA TTA CTC TAT ATG AGC G-3′ (nucleotides 1818–1842 mtDNA), Reverse primer: (5′-TCA GTT ACC AAA ACC ACC GAT T-3′ (nucleotides 8111–8090 mtDNA) and 71 bp region using Taqman probe with forward primer and reverse primer. The amplification factor was determined for each primer sets used for mtDNA damage assay (Gruber et al., 2011). Under ideal conditions, each DNA molecule would double in each PCR cycle. However, this is not necessarily the case, especially for long extension PCR. Different samples, experimental set-ups, reagents and fluorescence dyes can affect PCR performance, resulted in different amplification factors.

The aliquots from the same DNA sample was amplified using XL primers and SYBR Green I dye or using S primers and Taqman probe. DNA lesion frequency was then calculated using the Poisson equation as described in Govan et al. (1990), Britt (1995), and Hunter et al. (2010).

C. elegans embryonic cells were prepared from eggs isolated from young gravid adults as described in Bianchi and Driscoll (2006), Strange et al. (2007), and Sobkowiak and Lesicki (2009). Briefly, large quantities of gravid adult worms were lysed using egg isolation solution. The lysis reactions were stopped by adding egg buffer solution. The egg pellet was separated from worms’ debris by flotation on a sucrose solution, eggs were then washed to remove sucrose. 1 U/ml Chitinase was added to egg buffer to digest egg shells and to obtain single cell suspension. The dissociated cell suspension was then plated onto Lab-Tek II-CC2 chamber slide (Nunc, Thermo Fisher Scientific Inc.) and kept in a humidified incubator for cell differentiation. After 24 h, the degree of DNA damage was determined using the alkaline comet assay according to Olive and Banath (2006) and Sobkowiak and Lesicki (2009). Briefly, 1% low melting point agarose was added on top of the first layer of agarose contained cells and was allowed to solidify. The microscope slide was then immersed in ice-cold alkaline lysing solution for at least 1 h at 4°C in the dark. After lysis, the microscope slide was placed in a horizontal gel electrophoresis chamber filled with electrophoretic buffer. Electrophoresis was conducted at 25 V and 300 mA for 10 min in a chamber cooled on ice. The slide was then rinse with neutralization solution, dried at room temperature and stained with SYBR Green I. The microscope slide was examined using confocal fluorescence microscope (LSM 510 Carl Zeiss, Jena, Germany) and comets were analyzed using image analysis software, CometScore (TriTek Corp, United States).

GraphPad Prism version 5.02 for Microsoft Windows (GraphPad Software, San Diego, CA, United States) was used for statistical analysis. Lifespan and healthspan curves were analyzed by plotting Kaplan–Meier survival curves and by conducting Log-rank tests. Mean lifespan data was compared using Log-rank test with appropriate correction for multiple comparisons OASIS 2. All other data were plotted as means ± SEM, analyzed using ANOVA and Bonferroni’s multiple comparisons post-test unless otherwise stated. Differences with P < 0.05 were considered as statistically significant. In the figures, p values > 0.05 are summarized as ns, p values ≤ 0.05 are summarized as one asterisk, p values ≤ 0.01 are summarized as two asterisks, p values ≤ 0.001 are summarized as three asterisks, and p values ≤ 0.0001 are summarized as four asterisks.

Our S-XL-qRT-PCR assay is based on amplification of a short (S:71 bp) fragment relative to an extra-long (XL: 6.3 kbp) fragment, both within the same mtDNA sequence (Diagram 1). This extra-long target sequence was selected because this 6.3 kbp of mtDNA fragment covers almost half of the 13.7 kbp C. elegans mitochondrial genome. Under normal physiological conditions, the short fragment is too short for its amplification to be significantly affected by DNA damage and it thus serves as a reference for internal copy number normalization. Assuming that DNA lesions are randomly distributed within the mtDNA molecule, lesions frequency in a DNA sample can be calculated based on real-time amplification of the extra-long fragment relative to the short fragment (Meyer et al., 2007; Hunter et al., 2010) in the C. elegans mitochondrial genome using the Poisson equation (Govan et al., 1990; Britt, 1995; Hunter et al., 2010).

Because UV-mediated DNA damage is persistent and of a type that should be reliably detectable using our assay, we used UV-irradiation for initial test of reliability and sensitivity of our assay. In human, UVC exposure below 200 J/m² typically does not cause significant erythemal response (sunburn) (Hemminki et al., 2001; D’Orazio et al., 2013; Diffey and Farr, 2017). In C. elegans, Meyer et al. (2007) have shown that DNA damage in mtDNA can be detected with exposure as low as 100 J/m² UV-radiation. Our own preliminary experiments showed no immediate detrimental effects at this level in C. elegans (data not shown). We therefore chose 100 J/m² as the lower end of UVC challenge studies. Meyer et al. (2007) found that nematodes exposed to 400 J/m² UVC resulted in up to 10-fold elevation in mtDNA damage, suggesting that 400 J/m² causes significant damage. To compare the sensitivity and reproducibility of our assay, we exposed glp-1 C. elegans to different doses of UVC-radiation (254 nm wavelength) ranging from 100 to 1000 J/m². We analyzed samples at 100, 200, 400, 600, and 1000 J/m² UV-radiation to determine the levels of damaged mtDNA relative to controls using our S-XL-qRT-PCR DNA damage assay. As expected, mtDNA damage increased significantly as a function of increasing UV-irradiation dose (Figure 1A, p < 0.0001, One-way ANOVA). We were unable to show a statistically significant difference in the mtDNA damage at the lowest levels in nematodes exposed to 100 and 200 J/m² UV-radiation individually (Figure 1A, p > 0.05, One-way ANOVA with Bonferroni’s post-test). However, pooling data from the two lowest levels of UV-radiation (100 and 200 J/m²), we found significantly elevated mtDNA damage level compared to the non-irradiated control animals (Figure 1A, p < 0.01, One-way ANOVA with Bonferroni’s post-test). Levels above 200 J/m² resulted in robust elevation of mtDNA damage level, with 7-, 10-, and 15-fold increases in the mtDNA damage relative to untreated control animals in nematodes exposed to 400, 600, and 1000 J/m² UV-radiation, respectively (Figure 1A, p < 0.0001, One-way ANOVA with Bonferroni’s post-test). These UV sensitivity experiment demonstrates that our S-XL-qRT-PCR DNA damage assay is able to sensitively and reproducibly detect UV-induced lesions over a relatively wide range, with significant elevation even at low dose (100 and 200 J/m²) of UV-radiation.

We next tested our assay against DNA damage induced by γ-radiation. Exposure to γ-radiation damages DNA both directly (generating single-and double-strand breaks) (Henner et al., 1982; Min et al., 2003; Sudprasert et al., 2006) and indirectly by the generation of free radicals and ROS, e.g., through radio-lysis of water (Dizdaroglu, 1992; Borek, 2004; Halliwell and Gutteridge, 2015).

In C. elegans, γ-radiation doses of less than 1 kRad have previously been reported to have no immediate effect on survival or reproduction rate (Buisset-Goussen et al., 2014), indicating that γ-radiation at this dose is insufficient to kill even the rapidly dividing cells of the germline. The C. elegans soma consists mainly of non-dividing cells, known to be more resistant to DNA damage than dividing cells. This means C. elegans can typically tolerate higher levels of damaged DNA and is more resistance to radiation than most animals, including humans, that are dependent for their survival on active cell division (Flemming et al., 2000; Bailly and Gartner, 2011). Consistent with this expectation, Johnson and Hartman (1988) have previously reported that radiation doses of more than 100 kRad are required to observe significant lifespan reducing effect in young adult C. elegans.

We initially performed pilot experiments to characterize phenotypes other than survival of nematodes exposed to γ-radiation. We observed that control nematodes that never experienced γ-irradiation appeared healthier, had more offspring and explored more of their plates than γ-irradiated nematodes. Animals exposed to 40 kRad of γ-radiation explored significantly less and has substantially reduced fecundity while those exposed to 20 kRad were less severely affected (Figure 2A,B). We therefore exposed young adult nematodes to either 20 or 40 kRad of γ-radiation, significantly below the level that causes rapid death but well above the level sufficient to kill germline cells and resulting in detectable detriments and therefore expected to induce significant DNA damage to both nDNA and mtDNA.

As shown in Figure 1B, our S-XL-qRT-PCR DNA damage assay was able to detect the elevation of mtDNA damage level in animals challenged with γ-radiation. There was a dose-dependent increase in DNA damage with increasing amount of γ-radiation (p < 0.001, One-way Anova). Post-test analysis revealed that while we were able to detect a trend toward higher mtDNA damage in the nematodes challenged with 20 kRad of γ-radiation, this increase in the mtDNA damage was not statistically significant (Figure 1B). However, nematodes exposed to 40 kRad of γ-radiation showed significant elevation of mtDNA damage levels by approximately twofold compared to untreated control animals (p < 0.01, One-way ANOVA with Bonferroni’s post-test). These data indicate that the S-XL-qRT-PCR DNA damage assay is also able to detect DNA lesions induced by γ-radiation, although with lower sensitivity compared to UV-radiation.

To further evaluate the sensitivity of the S-XL-qRT-PCR DNA damage assay, we next compared its ability to detect DNA lesions caused by γ-radiation to a well-established assay that is commonly used in γ-irradiated animals, the comet assay.

The comet assay is a single cell electrophoresis method, used in radiation biology to quantify nDNA damage (Dusinska and Collins, 2008; Meyer, 2010). Briefly, cells are embedded in agar and exposed to an electrical field. DNA is drawn toward the anode, forming a comet-like image when viewed under a fluorescence microscope (Gedik et al., 1992). DNA containing single or double strand DNA breaks has a higher mobility in agar and during electrophoresis move faster and further (Collins et al., 2008). The amount of DNA within the comet tail therefore correlates to the extent of DNA damage (Collins et al., 2008). While the comet assay is designed to detect γ-induced DNA strand breaks (Collins et al., 1996) in nDNA of single cells, our PCR assay is designed to detect mtDNA damage. However, γ-radiation will penetrate cells and organelles, inducing DNA damage to both nDNA and mtDNA by direct interaction with DNA and through radiolysis of water in the path of radiation. Increased dose-dependent DNA damage is therefore expected to result in both the mitochondrial and nuclear compartments. We exposed young adult nematodes to either 20 or 40 kRad of γ-radiation and compared the resulting γ-induced DNA damage in nDNA and mtDNA as measured by the comet assay and our S-XL-qRT-PCR, respectively.

Figure 3Ai–iii are typical comet images of wild type N2 C. elegans embryonic cells obtained from untreated control compared to cells from nematodes irradiated with 20 or 40 kRad γ-radiation, respectively. As expected, cells from γ-irradiated nematodes on average have higher comet tail intensity (more damaged DNA) compared to less-damaged untreated controls. The amount of migrating DNA in the comet tail increases in a dose-dependence manner following exposure to γ-radiation (Figure 3B, p < 0.01, One-way ANOVA). However, similar to our PCR-based mtDNA damage assay, post-test reveals that the percentage of DNA in the comet tail of the nematodes challenged with 20 kRad γ-irradiation showed a trend to increased damage but this increase individually is not significant relative to the non-irradiated control nematodes (Figure 3B, p = 0.08, Student’s t-test). However, the percentage of increased of damaged DNA in the comet tail following 40 kRad γ-irradiation was 45% higher, and this change was statistically significant relative to non-irradiated control nematodes (Figure 3B, p < 0.001, One-way ANOVA with Bonferroni post-test).

Comparison of mtDNA damage frequency as detected using our S-XL-qRT-PCR DNA damage assay in mtDNA (Figure 1B) to the increase in nDNA strand breaks as detected using the comet assay (Figure 3B), shows a high degree of correlation between the increases in DNA damage in nDNA and mtDNA damage (Best fit R-squared = 0.96) (Figure 4). These data show that both assays are able to detect the increased in damaged DNA at the same level of γ-irradiation. Our S-XL-qRT-PCR DNA damage assay is therefore sensitive toward both UV- and γ-radiation induced mtDNA damage at physiologically relevant levels and its sensitivity for γ-radiation is comparable to that of the comet assay. Next, we therefore applied this tool to explore the role of mtDNA damage in the aging process of C. elegans.

We first asked if mtDNA damage burden, as measured by our assay, increases with age in C. elegans. We thus compared mtDNA damage between young (day 4) and aged (day 14) glp-1 nematodes. Under the conditions used, the glp-1 strain had a mean lifespan of about 13 days in our laboratory, meaning that beyond day 14, worms starting to die rapidly (Figure 8A,B). We indeed observed a clear age-dependent increase in mtDNA lesions with age (Figure 5, p = 0.04, Student’s t-test). There was a 2.3-fold increase in mtDNA damage in old animals (1.09 DNA lesions per 10 kbp) compared to young animals (0.48 DNA lesions per 10 kbp). This is consistent with previous observations, by us and others, reporting age-dependent increases in DNA damage to both nDNA and mtDNA in aging animals (Halliwell and Aruoma, 1991; Ames et al., 1993; Mecocci et al., 1994; de la Asuncion et al., 1996; Chaubey et al., 2001; Hamilton et al., 2001; Dhawan et al., 2009; Gruber et al., 2011). That older animals carry higher mtDNA damage burden than younger animals is expected and is a perquisite for such damage to play a causative role in aging. However, it is also possible that mtDNA damage accumulation may be merely a consequence or even a symptom of aging. If damage burden is causatively linked to aging, we would expect that long-lived strains consistently show less mtDNA damage and that, conversely, mutations that increase damage to mtDNA should shorten lifespan. To test these hypotheses, we next determined mtDNA damage levels and lifespan in a few C. elegans mutant strains that are either known to be long-lived [insulin-like growth receptor (IGF) pathway mutant strain, daf-2] or in strains carrying mutations likely resulting in increased ROS-mediated damage (antioxidant system/ROS mutants).

We first investigated mpst-1 mutant animals. The mpst-1 gene codes for a mitochondrial enzyme responsible for synthesizing hydrogen sulfide (H₂S) (Miller and Roth, 2007; Qabazard et al., 2013). H₂S has been shown to act as antioxidant and modulator of ROS production (Qabazard et al., 2013). We have previously shown that loss of mpst-1 gene in C. elegans reduces H₂S production, significantly elevates ROS production and reduces both lifespan and healthspan (Qabazard et al., 2013; Ng et al., 2018). Using our new assay and in agreement with our previous findings, we found that in short-lived mpst-1 mutants, mtDNA damage was threefold higher than in wild type N2 animals and this increase was statistically significant (Figure 6, p < 0.003, Student’s t-test). In mpst-1, our data are therefore consistent with the assumption that high mtDNA damage contributes to shorter lifespan in mutants with defects in their control of ROS.

We then applied our assay to another mutant strain often considered to suffer from ROS related lifespan detriment. The, mev-1 mutant strain was originally identified as a methyl-viologen (paraquat)-sensitive mutant strain (Ishii et al., 1990). However, later it was found that the gene mev-1 encodes the enzyme succinate dehydrogenase cytochrome b, which is a key component of complex II in the mitochondrial electron transport chain (Ishii et al., 1998). The mev-1 strain has decreased complex II activity (Ishii et al., 2011), is generally hypersensitive to oxidative stress, produces more ROS and suffers from increased oxidative damage and short lifespan (Adachi et al., 1998; Fong et al., 2017). However, despite having repeatedly been shown by us and others to produce more mitochondrial ROS and suffer from elevated oxidative damage, at least as assessed by protein carbonyl levels (Adachi et al., 1998; Fong et al., 2017), we did not detect any increase in mtDNA damage in short-lived mev-1 mutants compared to N2 wild type nematodes (Figure 6, p = 0.96, Student’s t-test).

We next used our mtDNA damage assay to evaluate mtDNA damage levels in a mutant strain defective in a key mitochondrial ROS detoxification system. Superoxide dismutase (SOD) is an antioxidant enzyme that catalyzes dismutation of superoxide anions to hydrogen peroxide and oxygen (Van Raamsdonk and Hekimi, 2009; Gruber et al., 2011; Halliwell and Gutteridge, 2015). C. elegans has six isoforms of SOD, two of which (sod-2 and sod-3) are mitochondrial MnSODs. GA480 is a sod-2/sod-3 double knockout mutant strain lacking both of these mitochondrial forms of SOD. However, surprisingly, despite completely lacking mitochondrial SOD and being hyper-sensitive toward oxidative stress, this strain does not experience lifespan shortening (Doonan et al., 2008), probably due to compensatory suppression of mitochondrial metabolism (Gruber et al., 2011). We have previously shown that sod-2/sod-3 double knockout mutant have significantly lower ROS production rate as evaluated by the DCF-DA assay (Gruber et al., 2011). Applying an earlier version of our mtDNA damage assay, we have previously found a trend toward elevated mtDNA damage in this strain but this trend did not reach significance (Gruber et al., 2011). Whether or not the lifespan of the GA480 strain is normal despite significantly elevated oxidative damage is therefore an important open question (Doonan et al., 2008; Honda et al., 2008; Gruber et al., 2011). We applied our new method to address this question and found a threefold increase in mtDNA damage relative to N2 wild type animals (Figure 6, p < 0.008, Student’s t-test). It therefore appears that animals of the SOD double mutant strain have normal lifespans, despite significantly higher damage to their mtDNA.

In summary, we found that only one mutant strain (mpst-1) shows both high mtDNA damage and short lifespan while one (mev-1) is short-lived despite normal mtDNA damage levels and lastly, one strain (sod-2/sod-3 double mutants) has a normal lifespan despite significantly increased damage levels. These data therefore do not consistently support the notion that mtDNA damage level is mechanically linked to longevity.

We next explored a nematode mutant strain that is known to be long-lived (Kimura et al., 1997). Mutation in the IGF receptor, daf-2 is one of the best studied and most efficacious single gene aging mutations in C. elegans. The daf-2 mutants lacks a functional IGF receptor and lives much longer than the wild type N2 nematodes (Kenyon et al., 1993; Bansal et al., 2015). Studies have shown that the daf-2/IGF pathway regulates aging, and that mutation in daf-2 decreases the formation of free-radicals and thus reduces protein oxidation (Yasuda et al., 1999; Mabon et al., 2009; Brys et al., 2010). Loss of daf-2 also increases resistance to oxidative stress, most likely by activating expression of antioxidant and stress genes downstream of daf-16, including SODs (Brys et al., 2010). However, when we quantified the levels of mtDNA damage in daf-2 mutants, we found that mtDNA damage levels were not lower than wild type N2 controls but were in fact threefold higher in daf-2 mutants relative to wild type N2 controls and this elevation was statistically highly significant (Figure 6, p < 0.0001, Student’s t-test).

The above example shows that genetic perturbations that significantly and reproducibly extend lifespan are not necessarily associated with low or even normal mtDNA damage levels. While in some cases such as the mpst-1 strain, mutation that cause increased ROS production may result in significantly increased damage levels and shortened lifespan, mtDNA damage levels were equally elevated in long-lived daf-2 mutants as in mpst-1 but without any apparent ill effect (Table 1). The lack of consistent impact of mtDNA damage level on aging rate (and vice versa) suggests that changes in mtDNA damage levels do not correlate with aging and lifespan in C. elegans.

In our mutant strain comparison, we found that there was no consistent relationship between differences in mtDNA damage and changes in lifespan. However, these effects were all measured in mutant strains, meaning that differences in mtDNA damage levels were the result of perturbations in endogenous processes that control signaling, ROS production/detoxification, DNA damage rate as well as DNA repair and turnover. In response to such perturbations, mutant strains may activate compensatory mechanisms potentially making them physiology significantly different from wild type animals. To more directly test the effect of mtDNA damage on lifespan, we therefore next asked the question whether exogenous damage to mtDNA shortens lifespan in adult C. elegans.

To evaluate if increased mtDNA damage as induced by UV exposure and at the levels detected by our assay, has consequences on aging trajectories, we exposed glp-1 C. elegans to between 0 to 400 J/m² of UV-radiation early in life. As mentioned earlier, compared to non-irradiated control animals, C. elegans exposed to both 100 and 200 J/m² UV-radiation had fourfold higher mtDNA damage level and animals exposed to 400 J/m² UV-radiation has sevenfold higher damaged mtDNA level. Comparing the survival of the untreated control animals to the nematodes exposed to 100, 200, and 400 J/m² UV-radiation, there was no statistically significant decrease in lifespan in nematodes exposed to 100 J/m² UV-radiation (Figure 7A,B, p > 0.05, Log-rank (Mantel-Cox) Test). In contrast, the lifespan of the nematodes irradiated with 400 J/m² UV-radiation was significantly shorter than non-irradiated control animals (Figure 7A,B, p < 0.05, Log-rank (Mantel-Cox) Test). Surprisingly, in nematodes exposed to 200 J/m² of UV-radiation, the mean survival was increased by 22% compared to non-irradiated control nematodes, suggesting a hormetic lifespan benefit at this level (Figure 7A,B, p < 0.05, Log-rank (Mantel-Cox) Test for survival curve and Log-rank test with Bonferroni multiple comparisons test, OASIS 2 for mean lifespan analysis). Whereas the healthspan of animals exposed to 100 J/m² UV, measuring the motility as an indicator of health, was similar to the non-irradiated control animals [Figure 7C,D, p > 0.05, Log-rank (Mantel-Cox) test]. Exposure to 400 J/m² UV-radiation, as expected, reduced overall health status [Figure 7C,D, p < 0.0001, Log-rank (Mantel-Cox) test]. Again, 200 J/m² UV-radiation resulted in animals that were healthier on average and had extended mean healthspan compared to controls [Figure 7D, p < 0.05, Log-rank (Mantel-Cox) Test for survival curve and Log-rank test with Bonferroni multiple comparisons test, OASIS 2 for mean healthspan analysis]. Furthermore, animals irradiated with 200 J/m² UV-irradiation had increased ability to sustain a high level of activity compared to controls or animals exposed to higher levels of UV (Figure 7E). In summary, animals irradiated with 200 J/m² UV were healthier for a longer period of time, had a longer lifespan and traveled significantly more distances than animals from all other groups, including non-irradiated controls (Figure 7E, p < 0.0001, One-way ANOVA with Bonferroni’s post-test).

We next evaluated if increased mtDNA damage as induced by γ-radiation had a more detrimental effect on aging trajectories than UV-induced damage. Given the preliminary results, we chose exposure to 40 kRad γ-radiation at day 4 of age. At this dosage γ-irradiation significantly elevated mtDNA damage levels to twofold higher (p < 0.01) in nematodes exposed to γ-radiation compared to untreated control animals (Figure 1B) and this amount of γ-radiation caused an observable increase in nDNA as measured by the Comet assay and also rendered WT animals sterile, confirming significant damage to both mtDNA and nDNA (Figure 4). To exclude the possibility that mtDNA damage levels due to γ-radiation were rapidly repaired, we have recently studied the mtDNA repair activity and found no significant repair of mtDNA damage within 24 h (Lakshmanan et al., 2018). Surprisingly, despite the fact that the animals had significantly and persistently higher mtDNA damage levels (Figure 1B), C. elegans exposed to 40 kRad of γ-radiation showed no significant lifespan shortening relative to untreated control animals (Figure 8A). The mean lifespan of the nematodes treated with 40 kRad γ-radiation was not significantly different relative to control animals (Figure 8B, p > 0.05, Log-rank test with Bonferroni multiple comparisons test, OASIS 2).

We then examined motility trajectory and speed of movement as indicators of overall health status of the muscular and nervous system. The health status of the γ-irradiated animals was similar to control animals, with similar mean healthspan at 13.8 days [Figure 8C,D, p > 0.05, Log-rank (Mantel-Cox) Test for survival curve and Log-rank test with Bonferroni multiple comparisons test, OASIS 2 for mean healthspan]. We then compared the average distance traveled within 15 min on day 4 of age between nematodes with or without pre-exposure to 40 kRad of γ-radiation. The typical distance traveled by γ-irradiated nematodes was markedly reduced relative to non-irradiated control animals (Figure 8E, p < 0.0001, Student’s t-test). Non-irradiated control nematodes traveled at 6400 μm per minute, while γ-irradiated nematodes traveled at about 2400 μm per minute (Figure 8E), suggesting some detrimental effects on performance at this level of radiation exposure, which, however, did not translate into shorter lifespan or accelerated aging. Despite significantly higher mtDNA damage in γ-irradiated animals, we observed no lifespan or healthspan shortening effects in γ-irradiated animals. This is surprising as we observed a clear fitness impairment (sterility and significantly reduced motility, Figure 2, 8C–E) in γ-irradiated animals.