Isocitrate Dehydrogenase Alpha-1 Modulates Lifespan and Oxidative Stress Tolerance in Caenorhabditis elegans

Altered metabolism is a hallmark of aging. The tricarboxylic acid cycle (TCA cycle) is an essential metabolic pathway and plays an important role in lifespan regulation. Supplementation of α-ketoglutarate, a metabolite converted by isocitrate dehydrogenase alpha-1 (idha-1) in the TCA cycle, increases lifespan in C. elegans. However, whether idha-1 can regulate lifespan in C. elegans remains unknown. Here, we reported that the expression of idha-1 modulates lifespan and oxidative stress tolerance in C. elegans. Transgenic overexpression of idha-1 extends lifespan, increases the levels of NADPH/NADP+ ratio, and elevates the tolerance to oxidative stress. Conversely, RNAi knockdown of idha-1 exhibits the opposite effects. In addition, the longevity of eat-2 (ad1116) mutant via dietary restriction (DR) was reduced by idha-1 knockdown, indicating that idha-1 may play a role in DR-mediated longevity. Furthermore, idha-1 mediated lifespan may depend on the target of rapamycin (TOR) signaling. Moreover, the phosphorylation levels of S6 kinase (p-S6K) inversely correlate with idha-1 expression, supporting that the idha-1-mediated lifespan regulation may involve the TOR signaling pathway. Together, our data provide new insights into the understanding of idha-1 new function in lifespan regulation probably via DR and TOR signaling and in oxidative stress tolerance in C. elegans.


Introduction
Aging is an irreversible progression accompanied by physiologic function decline progressively and eventually leads to death [1]. Aging and metabolism are highly intertwined [2]. Several mechanisms have been reported to compose a complex regulation of lifespan, and one of the evolutionarily conserved mechanisms is dietary restriction (DR) [3,4]. DR prolongs the lifespan of yeast, C. elegans, Drosophila, and mammals [5]. DR not only affects metabolic rate but also promotes longevity and stress resistance via several signaling pathways [5,6].
Multiple signaling pathways are involved in the mechanism of dietary restriction, such as insulin/IGF-1-like signaling, sirtuin, and target of rapamycin (TOR) signaling [7]. Among those pathways, the TOR pathway is closely linked to DR [8,9]. TOR acts as a nutrientsensing signaling pathway in which its activity increases when food is abundant and vice versa [7,8,10]. TOR controls protein synthesis by phosphorylating ribosomal protein S6 kinase (S6K) and restrains the activity of transcription factor PHA-4 and FOXO/DAF-16, as well as inhibits autophagy and the expression of stress-related genes [7,9]. Repressing TOR activity by rapamycin is sufficient to extend the lifespan in C. elegans, Drosophila, and mice [11]. Several studies indicated that RNAi silencing TOR or its downstream target S6K can increase lifespan in yeast, Drosophila, and C. elegans [12][13][14].
Metabolic alternation plays an important role during the aging process [15]. The manipulation of genes and metabolites in the tricarboxylic acid (TCA) cycle has been reported to regulate lifespan [16]. Lifespan extension occurs when adding either α-ketoglutarate, succinate, pyruvate, malate, fumarate, or citrate which are all metabolites generated by the TCA cycle, in C. elegans or in Drosophila [17][18][19][20][21][22]. Among these metabolites, α-ketoglutarate supplementation not only prolongs the lifespan in worms and flies but also enhances longevity in mice [23] and may also help in promoting human health [24]. Besides metabolites supplementation, the manipulation of the enzymes in the TCA cycle also can modulate lifespan. The inhibition of gdh-1 and dld-1, which both increase the α-ketoglutarate level, can extend the lifespan in C. elegans [17,25]. Isocitrate dehydrogenase (IDH) is a rate-limiting enzyme which catalyzes isocitrate to become α-ketoglutarate in the TCA cycle [26]. IDH composes three isoenzymes, IDH1, IDH2, and IDH3, which are conserved in eukaryotes. IDH1 and IDH2 are NADP + dependent, whereas IDH3 is NAD + dependent. In humans, IDH3 is a heterotetramer containing two alpha subunits (IDH3A), one beat subunit (IDH3B) and one gamma subunit (IDH3G). However, the role of isocitrate dehydrogenase in lifespan regulation remains unknown.
Here, we report that expression of isocitrate dehydrogenase alpha-1 (idha-1), the ortholog of human IHD3A, regulates lifespan and modulates oxidative stress tolerance in C. elegans. Overexpression of idha-1 prolongs lifespan, increases oxidative stress resistance, and elevates the levels of the NADPH/NADP + ratio. Oppositely, RNAi knockdown of idha-1 displays reduced lifespan, lowered tolerance to oxidative stress, and diminished levels of the NADPH/NADP + ratio. Mechanistically, the knockdown of idha-1 partially abolishes the longevity in eat-2 (ad1116) dietary restriction (DR) mutant, suggesting idha-1 participates in DR-mediated longevity. Furthermore, the knockdown of idha-1 does not significantly block the enhanced lifespan of the rsks-1 (ok1255) mutant, implying the involvement of TOR signaling. Moreover, the phosphorylation levels of S6 kinase (p-S6K) inversely correlate with the idha-1 expression, supporting that the idha-1-mediated lifespan regulation may involve the TOR signaling pathway. Our study sheds new light on the function of idha-1 in lifespan regulation and in oxidative stress response.

Expression of idha-1 Manages the Tolerance to Oxidative Stress and Levels of NADPH/NADP + Ratio
Longevity organism is usually associated with better stress resistance [27][28][29][30]. To examine whether the expression of idha-1 which regulates lifespan is also associated with the tolerance to oxidative stress, we performed an oxidative stress assay with the idha-1 overexpression and knockdown worms and their controls under paraquat-induced oxidative stress. The idha-1 overexpression strain exhibited significantly increased survival under 48 and 72 h of paraquat treatment compared with the control (Figure 3a). On the other hand, the knockdown of idha-1 in N2 significantly reduced the tolerance to oxidative stress compared with the control at 48 and 72 h of the paraquat treatment ( Figure 3b). Thus, these data suggest that the expression of idha-1 is associated with the tolerance to oxidative stress by paraquat.
NADPH can reduce glutathione disulfide (GSSG) to glutathione (GSH) to lower free radical levels. The ratio of NADPH/NADP + level is considered an index for oxidative stress tolerance. The higher the NADPH level, the better the oxidative stress tolerance [28]. Therefore, we measured the NADPH/NADP + ratio in both idha-1 overexpression and knockdown worms. Overexpression of idha-1 displayed a significant increase in NADPH/NADP + ratio compared with that in the control (Figure 3c). On the other side, the knockdown of idha-1 showed significantly decreased levels of NADPH/NADP + ratio compared with the control (Figure 3d). Together, these results reveal that the expression of idha-1, which regulates lifespan, also manages the tolerance to oxidative stress and the levels of NADPH/NADP + ratio in C. elegans.

idha-1 Plays A Role in Dietary Restriction Induced Longevity
Reduced fecundity is associated with dietary restriction (DR)-mediated longevity. The long-lived eat-2 (ad1116) mutant, which exhibits reduced pharyngeal pumping rate for lower food intake and is used as a DR model, exhibits a reduction in progeny production. The supplementation of α-ketoglutarate was reported to be unable to further prolong the extended lifespan in the eat-2 (ad1116) mutant, suggesting the longevity by α-ketoglutarate is similar to that by DR. Therefore, we asked whether the long-lived idha-1 overexpression transgenic strain which shows an elevated α-ketoglutarate level displays reduced fecundity. We measured the brood size and found a significant decrease in the number of progenies from day 2 to day 6 in the idha-1 overexpression transgenic line compared with the control (Figure 4a). The average of the total progeny number significantly declines by about 50% in the idha-1 overexpression line compared with the control (Figure 4b). Interestingly, we found the idha-1 mRNA levels are elevated in both the DR-treated worms by fasting and the eat-2 (ad1116) mutant DR model compared with their controls in the GEO data ( Figure S2a,b) as well as in the verification by quantitative RT-PCR analysis ( Figure S2c). The expression of many other TCA cycle genes is also upregulated in the DR-treated worms by fasting and the eat-2 (ad1116) mutant DR model compared with their controls in the GEO data ( Figure S2d,e). The data imply the longevity of idha-1 overexpression may also be involved in DR-induced longevity. Moreover, the knockdown of idha-1 significantly diminished the extended lifespan in eat-2 (ad1116) (Figure 4c, Table S1), which indicates the longevity in eat-2 (ad1116) mutant partially depends on idha-1. Overall, the results support our hypothesis that idha-1 plays a role in dietary restriction-mediated longevity in C. elegans. idha-1 (RNAi) 3 , n = 158) significantly reduces the tolerance to oxidative stress compared with the control (black dot, N2 EV, n = 157) after 24-, 48-and 72-h treatment (** p < 0.01, **** p < 0.0001, one-way ANOVA test). Each dot represents mean ± SD, n = 3. (c) Overexpression of idha-1 (red bar) significantly increases the NADPH/NADP + ratio levels compared with the control (grey bar). (** p < 0.01, t-test). (d) Knockdown of idha-1 (blue bar, N2 idha-1 (RNAi) 5 and yellow bar, N2 idha-1 (RNAi) 3 ) significantly decreases the NADPH/NADP + ratio levels compared with the control (black bar, N2 EV). (*** p < 0.001 and * p < 0.05, one-way ANOVA test). Each bar represents mean ± SD, n = 3. (a) Overexpression of idha-1 (red dot) significantly decreases progeny number between day 2 to day 6 compared with the control (black dot). Each dot plot represents mean ± SD, n = 3 (* p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001, t-test) (b) Total brood size is significantly diminished in idha-1 overexpression strain (red bar) compared with the control worm (grey bar). Each dot plot represents mean ± SD, n = 5 (* p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001, t-test).

Expression of idha-1 May Regulate Lifespan via Inversely Modulating TOR Signaling
Dietary restriction is known to interact through multiple signaling pathways, such as the insulin/IGF-1 signaling (IIS) pathway, AMP-activated protein kinase (AMPK) pathway and Target of Rapamycin (TOR) pathway [7]. Lifespan extension caused by reducing the IIS pathway and TOR pathway signaling or activating the AMPK pathway is conserved from yeast to mammals [4]. Our finding suggests that idha-1 may be involved in DR-induced longevity. To further study the underlying molecular mechanism of idha-1-mediated lifespan regulation, we examined the lifespan by reducing idha-1 expression in the daf-16 (mu86), aak-2 (gt33), and rsks-1 (ok1255) mutant strains related to IIS, AMPK, and TOR pathways, respectively. Knockdown of idha-1 decreased the lifespan in short-lived daf-16 (mu86) (Figure 5a, Table S1), suggesting that idha-1-mediated lifespan is independent of daf-16. Similarly, the knockdown of idha-1 also reduced the lifespan in the short-lived aak-2 (gt33) mutant (Figure 5b, Table S1), indicating that idha-1-mediated lifespan is independent of aak-2. On the other hand, the knockdown of idha-1 did not significantly shorten the lifespan in long-lived rsks-1 (ok1255) mutant (Figure 5c, Table S1), implying that idha-1mediated lifespan may depend on reducing TOR signaling. To strengthen our hypothesis that idha-1-mediated lifespan may depend on lowered TOR signaling, we examined the levels of p-S6K, which is a downstream effector of TOR signaling in both idha-1 overexpression and knockdown strains. The long-lived idha-1 overexpression strain displayed significantly lower p-S6K levels compared with the control (Figure 6a,b), whereas the short-lived idha-1 knockdown lines showed the opposite results with the elevated p-S6K levels compared with the control (Figure 6c,d). Together, the data support the finding that the idha-1-mediated lifespan depends on TOR signaling in C. elegans. Figure 6. Expression of idha-1 negatively modulates the phosphorylated S6K levels. The level of phosphorylated S6K (p-S6K), a TOR activity indicator, in each strain was determined and normalized with β-actin level. (a,b) Long-lived idha-1 overexpression strain (red bar, idha-1 o/e) displays significantly reduced p-S6K levels compared with the control (grey bar) (** p < 0.01, t test). Each bar represents mean ± SD, n = 4. (c,d) Short-lived idha-1 knockdown strains (blue bar, N2 idha-1 (RNAi) 5 ; and yellow bar, N2 idha-1 (RNAi) 3 ) exhibit significantly increased p-S6K levels compared with the control (black bar, N2 EV) (* p < 0.05, t test). Each bar represents mean ± SD, n = 3.

Discussion
Despite as an enzyme in the TCA cycle, in this study, we uncover a new function of idha-1 in lifespan regulation and oxidative stress response in C. elegans. Those phenotypes are associated with the levels of NADPH/DADP+ ratio. This is in accordance with the report that Sirt3 can activate mitochondrial isocitrate dehydrogenase 2 to increase NADPH levels in response to caloric restriction in mice [31]. Our previous study also reported that the longevity and oxidative stress resistance by ribose-5-phosphate isomerase knockdown are associated with the levels of NADPH/DADP + ratio in Drosophila [28]. In addition, we disclose that the mechanism of lifespan regulation by idha-1 could be mediated by dietary restriction and TOR signaling. This is consistent with the previous findings that αketoglutarate supplementation prolongs lifespan via inhibiting TOR signaling in C. elegans and Drosophila [17,20]. Moreover, those idha-1-mediated phenotypes not only occur in C. elegans but also are associated with an allele related to isocitrate dehydrogenase in Drosophila by selected breeding [32], suggesting that lifespan regulation and the oxidative stress response by idha-1 may be evolutionary conserved. IDHA-1 is conserved among the Caenorhabditis genus ( Figure S3).
The TCA cycle is an energy-producing and metabolism process. Recent studies revealed that the supplementation of TCA cycle metabolites, such as α-ketoglutarate, malate, and fumarate, benefits longevity [17,22]. Not only can the supplementation of certain metabolites in the TCA cycle extend lifespan, but also genetic manipulation of some genes in the TCA cycle can regulate longevity. The knockdown of ogdh-1, sdha-1, and dld-1 can increase α-ketoglutarate levels and also extend the lifespan in C. elegans [17,22,25]. Apart from manipulating the TCA cycle directly, a previous study showed that long-lived rodents tend to upregulate TCA cycle genes or show no decline in TCA cycle function [33,34]. However, among all these studies, no report yet shows single TCA cycle gene overexpression can prolong lifespan. Here, we provide the first new strategy of lifespan extension in nematode by overexpressing the TCA cycle gene idha-1 in C. elegans.
Better oxidative stress tolerance is usually accompanied by a higher antioxidant capacity. Since the elevation of NADPH levels shows better resistance against ROS and free radicals, the NADPH/NADP + ratio can reflect the tolerance against oxidative stress [28,35]. Here, we found idha-1 overexpression strain exhibits higher NADPH levels, and the knockdown of idha-1 reduced NADPH levels and vice versa. Although the supplementation of α-ketoglutarate has been shown to extend lifespan, it is unable to elevate the oxidative stress tolerance across the species [17,20,23]. The discrepancy could be the additional genetic effect of idha-1 expression, which can modulate NADPH levels. The other possibility is that idha-1 expression may regulate some antioxidant gene expression. Our study provided another new piece of evidence that idha-1 expression modulates oxidative stress tolerance in C. elegans.
Moreover, we discovered that idha-1 overexpression may be linked to the DR-induced longevity phenotype and the reduction in brood size [36]. To investigate the direct relationship between idha-1 and DR, we perform a lifespan assay under DR model conditions and find out whether DR-induced longevity may partially rely on idha-1 expression. DRinduced longevity includes several nutrient-signaling pathways, such as the IIS pathway, AMPK pathway, and TOR pathway [37]. Unraveling the potential mechanism idha-1mediated lifespan regulation may depend on, we knockdown idha-1 in the mutants of these pathways. The result points out that only the TOR signaling pathway, but not the IIS or AMPK, is responsible for idha-1-mediated lifespan regulation. Together, these results support our hypothesis that idha-1 plays a role in DR-induced longevity through the TOR signaling pathway.
Besides the IIS, AMPK, and TOR pathways, autophagy and sirtuins are also involved in DR-induced longevity [38]. The relationship between the TOR pathway and autophagy among species is well documented. The inhibition of the TOR signaling pathway may cause the activation of autophagy. To further validate the relation between autophagy and idha-1 mediation, we may investigate the molecular changes of some autophagy-related genes, such as bec-1, lgg-1, and sqst-1 [39], or even measure the autophagy flux in the future study.
Aside from autophagy, sirtuins may also involve in idha-1-mediated lifespan regulation. Sirtuins are a family of NAD + -dependent protein deacetylases, which use NAD + to remove acetyl moieties on histones and proteins [40]. Sirt1 is one of the sirtuins in humans which participates in DR-induced longevity [41]. The activity of Sirt1 accompanies the elevation of NAD + levels during DR. The orthologue of Sirt1, sir-2.1, can partially regulate DR-induced longevity by activating DAF-16/FOXO [42]. Based on the function of idha-1, which converts NAD + to NADH, we assumed overexpression of idha-1 may elevate the NADH level. However, we did not observe increased NADH levels when overexpressing idha-1. Surprisingly, we found an elevation of the NAD + level, which is associated with DR-mediated longevity, upon idha-1 overexpression ( Figure S4). A previous study demonstrated that the elevation of NAD + level accompanies better mitochondrial efficiency and also proposes that the elevation of the NAD + level is a consequence of DR [43]. As our data show that idha-1 overexpression elevates NAD + levels and participates in DR-induced longevity, it may imply the involvement of sir-2.1. We may investigate the role of sir-2.1 in idha-1-mediated lifespan regulation in future work.
In summary (Figure 7), we disclose the new role of idha-1 in lifespan regulation and oxidative stress tolerance. Our study demonstrates that idha-1 regulates lifespan by participating in DR-induced longevity inversely through the TOR signaling pathway and modulates oxidative stress tolerance via the levels of NADPH/NADP + ratio. Increasing amounts of evidence demonstrate that TCA metabolites and enzymes control human physiology and diseases [44,45]. The information from this study will help in the development of α-ketoglutarate as dietary supplementation in promoting human health [24]. Figure 7. The mechanism of idha-1-mediated lifespan regulation and oxidative stress response in C. elegans. DR may induce idha-1 upregulation. Overexpression of idha-1 results in increased levels of α-ketoglutarate and NAD + . In addition, idha-1 overexpression also leads to elevated levels of NADPH/NADH + ratio which increases oxidative stress tolerance. Moreover, idha-1 overexpression may negatively regulate TOR signaling to extend lifespan in C. elegans.

Generation of idha-1 Overexpression Transgenic Worms
To establish the idha-1 overexpression transgenic construct, we amplified the idha-1 genomic DNA containing 1 kb upstream promoter region and the whole idha-1 genomic DNA to be cloned into L3691 vector by using the primers p1-F idha-1-pstI (5 -TTTCTGCAGTCGA AGTTGTCAAAATCCACGAGA-3 ) and p2-R-idha-1-kpnI (5 -TTTGGTACCTCTGAACCC TGGAAACAAAAATATTT-3 ). The resultant idha-1 overexpression transgenic construct was named Pidha-1::idha-1. For the control transgenic construct, we cloned the 1 kb idha-1 upstream promoter region into L3691 to express GFP and named the construct Pidha-1::gfp. To generate the idha-1 overexpression transgenic and the control transgenic worms, N2 worms were prepared and synchronized via timed eggs prepared as described previously [30]. The plasmid construct was diluted in ddH 2 O and injected into young adult worms (4 to 8 eggs) with co-injection vector Psur-5::rfp at a total concentration of 20 ng/µL. The progenies of the injected worms were then screened for RFP expression and with GFP as well for the control and established as the transgenic lines. Worms expressing RFP were picked and transferred onto new plates every 5 days to maintain a stable line and verified by PCR. The control transgenic worm Pidha-1::gfp was used to examine the tissue expression patterns of the GFP driven by the idha-1 promoter.

Lifespan Analysis and Oxidative Stress Assay
Lifespan assay was performed at 20 • C as described previously [30]. Parent worms were grown to the day-1-adult stage on NGM plates seeded with E. coli strain OP50. Nearly 20 adult worms were transferred to RNAi-containing plates or normal NGM plates for 6 h and picked off. The offspring were synchronized on the plates by timed egg lay assay. After the offspring grew to the L4 stage, 30 worms were picked onto each plate, and at least 120 worms existed in each group per trial. Worms were transferred to new plates and the number of deaths was counted every 2 days until all worms died. For oxidative stress assay, 50 worms were picked onto a new plate containing 10 mM paraquat (1, 1 -Dimethyl-4, 4 -bipyridinium dichloride, Sigma-Aldrich). The number of dead worms was calculated every 24 h. The statistical analyses of lifespan and oxidative stress assay were performed by OASIS 2 and p-values were calculated by log-rank test for lifespan and t-test or one-way ANOVA test for oxidative stress assay.

RNA Extraction and qRT-PCR
The worms were washed with an M9 buffer three times and collected in a 1.5-mL micro-centrifuge tube. The RNA extraction and RT-QPCR procedures were described previously [30]. The RNA pellet was dissolved by 10 µL DEPC-treated ddH 2 O and measured the concentration by NanoDrop 2000 (Thermo). The qRT-PCR was performed by using the Step One Plus Real-Time PCR system (ABI). The relative expression levels of genes between different samples were calculated on the basis of ∆∆Ct (threshold cycles) values and act-1 was used as the internal control for normalization. Statistical analysis was calculated by t-test or one-way ANOVA test. The primers for idha-1 are the forward primer 5 -GCACGCGAACAAAGTTGGAC-3 and the reverse primer 5 -CTTCAAGGGAACGGCATGGA-3 . The primers for act-1 are the forward primer 5 -CTCTTGCCCCATCAACCATGA-3 and the reverse primer 5 -TTGCGGTGAACGATGG ATGG-3 .

Measurement of α-Ketoglutarate Level
The α-ketoglutarate levels were measured by a colorimetric quantification kit (BioVision, Cat# K677-100) for Figure 3c and by a bioluminescent assay-based kit (NADP/NADPH-Glo™ assay; Promega, Cat#G9081) for Figure 3d according to the manufacturer's protocols. Worms were collected into 1.5-mL centrifuge tubes and homogenized in the extraction buffer provided with the kit. The samples were centrifuged at 13,000 rpm for 5 min and the supernatants were transferred into new labeled tubes. To detect the α-ketoglutarate level, 50 µL samples were transferred into a 96-well ELISA plate. 2 µL of cycling enzyme mix and 48 µL of buffer mix were added in each well and incubated at 37 • C for 30 min. The plate was read under OD 570 nm with an ELISA reader, and the α-ketoglutarate level was calculated according to the standard curve and normalized by protein concentration.

NADPH/NADP + and NAD + /NADH Quantification
The NADPH/NADP + and NAD + /NADH quantifications were measured by using the quantification colorimetric kits (BioVision, Cat#: K337-100 and K347-100). The worms were collected into a 1.5-mL eppendorf tube and homogenized by the extraction buffer. 50 µL of each sample was transferred into a 96-well ELISA plate according to the manufacturer's protocols. Calculated the ratio according to the standard curve.