Oxaloacetate supplementation increases lifespan in Caenorhabditis elegans through an AMPK/FOXO-dependent pathway

Summary

Reduced dietary intake increases lifespan in a wide variety of organisms. It also retards disease progression. We tested whether dietary supplementation of citric acid cycle metabolites could mimic this lifespan effect. We report that oxaloacetate supplementation increased lifespan in Caenorhabditis elegans. The increase was dependent on the transcription factor, FOXO/DAF-16, and the energy sensor, AMP-activated protein kinase, indicating involvement of a pathway that is also required for lifespan extension through dietary restriction. These results demonstrate that supplementation of the citric acid cycle metabolite, oxaloacetate, influences a longevity pathway, and suggest a tractable means of introducing the health-related benefits of dietary restriction.

Introduction

Dietary restriction (DR) is generally defined as a diet in which dietary intake is reduced from that obtained through ad libitum feeding, but does not result in malnutrition. It has been demonstrated to increase lifespan in organisms, ranging from yeast to mammals (Koubova & Guarente, 2003). Of particular significance, DR also retards the progression of many disorders. In mammals, DR has been reported to delay the incidence of kidney disease, autoimmune disease, and diabetes, reduce age-related neuron loss in mouse models of Parkinson's disease and Alzheimer's disease, and substantially lower cancer risk (Koubova & Guarente, 2003; Mai et al., 2003).

One molecular pathway that has been shown to be activated by DR of the nematode worm, Caenorhabditis elegans, is through the energy sensor, AMP-activated protein kinase (AMPK), in concert with the FOXO transcription factor, DAF-16. Both AMPK and DAF-16 are required for the mediation of longevity and delay in age-dependent decline observed in C. elegans as a result of DR, administrated by food reduction on agar plates (Greer et al., 2007). AMPK has also been shown to be necessary for the lifespan extension that results from reducing glucose availability (Schulz et al., 2007). While reduced glucose levels can activate AMPK, the enzyme is also especially sensitive to NAD⁺ and NADH. High levels of NAD⁺ activate AMPK, whereas NADH inhibits the activity of AMPK (Rafaeloff-Phail et al., 2004). Increases in the NAD⁺/NADH ratio have been shown to increase replicative lifespan in yeast (Lin et al., 2000).

Based on these reports, we hypothesized that alterations in the NAD⁺/NADH ratio, through dietary supplementation of metabolites of the citric acid cycle, might mimic the lifespan effect of caloric restriction. The conversion of oxaloacetate (3-carboxy-3-oxopropanoic acid) to malate is an energy favorable reaction in cells that promotes the conversion of NADH to NAD⁺. We report that the addition of oxaloacetate to agar plates supporting C. elegans increased lifespan in these worms in a FOXO/DAF-16 and AMPK-dependent manner.

Results and discussion

Supplementation with 8 mM oxaloacetate resulted in a significant shift in survival curves, with a mean increase of 25% in median lifespan and a mean increase of 13% in maximal lifespan (Fig. 1A,E; Table 1; Supporting Table S1). In preliminary experiments, we did not detect any difference in lifespan as a result of supplementation with 2 mM oxaloacetate or with the breakdown product of oxaloacetate, pyruvate (data not shown).

Fig. 1.

Survival of Caenorhabditis elegans on agar cultures with and without oxaloacetate supplementation. (A–D) Survival curves of control (N2) and mutant [sir-2.1 (ok434), daf-16 (mgDf50), aak-2 (ok524)] worms. The curves represent the means of the survival curves from the individual experiments. (A) Increased lifespan due to 8 mm oxaloacetate in control worms. (B) Increased lifespan due to 8 mm oxaloacetate in sir-2.1 mutant worms (data from N2 worms tested in the same experiments are also shown). (C) Increased lifespan due to 8 mm oxaloacetate is blocked in FOXO mutant daf-16 worms (data from N2 worms tested in the same experiments are also shown). (D) Increased lifespan due to 8 mm oxaloacetate is blocked in AMP-activated protein kinase mutant aak-2 worms (data from N2 worms tested in the same experiments are also shown). (E) Bar graph showing the change in median lifespan in response to 8 mm oxaloacetate (relative to the same strain grown without oxaloacetate). The change shown by each bar is the mean of the medians measured from all repeats of each experiment ± SEM (shown as error bars), where n = number of repeats for a given experiment. This graph illustrates some of the data shown in Table 1, where data on maximal lifespan and the results of statistical analyses are also given. **P < 0.001, *P < 0.01, by log rank test.

Table 1.

Results of different experiments on the effect of oxaloacetate on lifespan^†

Strain	Median lifespan (days ± SEM)		Difference (%)	n	P-value‡	Max lifespan (days ± SEM)		Difference (%)
	Crtl	Exptl				Crtl	Exptl
N2 control	14.4 ± 0.9	18.0 ± 2.0	+25	5	≪ 0.001	23.6 ± 3.2	26.4 ± 3.2	+13
N2 control§	12.5 ± 0.7	14.4 ± 1.1	+15	3	< 0.05	29.0 ± 1.0	31.5 ± 0.5	+5
SIR-2.1 mutant	13.9 ± 3.6	16.2 ± 4.0	+16	2	< 0.01	19.5 ± 4.5	25.0 ± 7.0	+27
DAF-16 mutant	15.0 ± 0.3	13.4 ± 1.0	−11	3	< 0.05	22.0 ± 3.0	22.5 ± 2.5	+3
AAK-2 mutant	11.9 ± 0.2	12.1 ± 0.3	+1	3	1.00	16.7 ± 0.7	14.7 ± 0.3	−5

^†Comparisons were made between groups treated with 0 (Crtl) or 8 mm (Exptl) oxaloacetate.

^‡Probability (P-value) of no significant difference in median lifespan between control and experimental groups; the P-value given is the higher one of that determined by a single anova test or by Fisher's combined probability test of log rank data.

^§Dead bacteria were used instead of live bacteria. n, number of repeats for a given experiment.

Oxaloacetate did not appear to affect the amount of food consumption, as the rate of pharyngeal pumping was not significantly different between control (320 pumps min⁻¹ ± 29 SD) and experimental groups (328 pumps min⁻¹ ± 35 SD) of age-synchronized worms (P > 0.2, one-way anova). In experiments where the worms fed on dead bacteria instead of live bacteria, the median lifespan was also increased (Table 1), indicating that at least a component of the increase is not due to an effect of oxaloacetate on the bacteria.

An increase in NAD⁺ levels should increase the activity of Sir-2, an NAD⁺-dependent histone deacetylase, reported to modulate aging and lifespan in C. elegans and other organisms (Lin et al., 2000; Tissenbaum & Guarente, 2001). However, we found that oxaloacetate still increased median lifespan in Sir-2.1 mutants (Fig. 1B,E; Table 1), indicating the involvement of a pathway(s) that is independent of Sir-2.1. The addition of 50 μm splitomycin, a Sir-2 inhibitor, produced a similar result, also failing to block the oxaloacetate-induced increase in lifespan (data not shown).

To test the role of FOXO and AMPK in mediating the oxaloacetate-induced extension of lifespan, we experimented with DAF-16 and AAK-2 mutants. We used daf-16 (mgDf50) worms, which are null for DAF-16, and aak-2 (ok524) worms that contain a partial deletion of the AAK-2 subunit of AMPK, rendering the kinase inactive (Apfeld et al., 2004). Both the DAF-16 and AAK-2 mutants were unaffected by the addition of oxaloacetate (Fig. 1C–E; Table 1), indicating the requirement of these genes for the oxaloacetate-induced increase in lifespan.

Our results demonstrate that oxaloacetate supplementation increases the lifespan of C. elegans maintained on agar plates. The lifespan increase is dependent on the transcription factor, FOXO/DAF-16, and the energy sensor, AMPK, but shows independence from the NAD⁺-dependent histone deacetylase, Sir-2.1. An increased dosage of the Sir-2.1 gene increases lifespan in C. elegans through the insulin-like signaling pathway (Tissenbaum & Guarente, 2001). Therefore, while an effect on insulin-like signaling may occur in response to oxaloacetate, another pathway appears to be involved. DAF-16 is required for lifespan extension through DR as well as insulin signaling in C. elegans raised on agar plates (Kenyon et al., 1993; Lin et al., 1997, 2001; Ogg et al., 1997; Greer et al., 2007). Its requirement, together with that of AMPK activity for the oxaloacetate-induced lifespan extension suggests that the oxaloacetate effect involves a pathway that also mediates lifespan extension through DR (cf. Greer et al., 2007). It is noteworthy, however, that the requirement of DAF-16 and AMPK for lifespan extension through DR varies according to conditions for maintaining the worms. Both are required when worms are reared on agar plates, but not when maintained in liquid medium (Bishop & Guarente, 2007; Greer et al., 2007; Mair et al., 2009). DAF-16, but not AMPK, is required for life extension effected by intermittent feeding (Honjoh et al., 2009) or by the eat-2 mutation (Curtis et al., 2006). Therefore, the effect of oxaloacetate may vary according toconditions under which animals are maintained.

Stimulation of a DR pathway by oxaloacetate may be initiated by its exergonic conversion to malate, and the resulting conversion of NADH to NAD⁺. An increase in the NAD⁺/NADH ratio has been linked to DR and increased lifespan in yeast (Lin & Guarente, 2003; Easlon et al., 2008), and has been measured in mitochondria exposed to external oxaloacetate (Haslam & Krebs, 1968). Increased mitochondrial NAD⁺ has also been demonstrated during DR (Yang et al., 2007). Energy levels and lifespan appear to be linked by AMPK activity, which is stimulated by increases in the NAD⁺/NADH ratio (Rafaeloff-Phail et al., 2004). Our finding that AMPK activity is required for the oxaloacetate-induced increase in lifespan is therefore consistent with the initiation of this effect through an increase in NAD⁺/NADH.

In conclusion, we report that supplemental oxaloacetate increases lifespan in C. elegans, most likely by stimulating a pathway that is also stimulated by at least one type of DR, beginning with an increase in NAD⁺/NADH, and requiring AMPK activity. Oxaloacetate is a human metabolite, with high tolerance levels [diabetic patients have been treated with up to 1 g per day without any adverse effects (Yoshikawa, 1968)], so that our results suggest it may provide a feasible means to access the health benefits of DR.

Materials and methods

The C. elegans strains used were N2 (wild-type), sir-2.1 (ok434), daf-16 (mgDf50), and aak-2 (ok524). Strains were provided by the Caenorhabditis Genetics Center at the University of Minnesota, with the exception of Sir-2.1 and DAF-16 deletion mutants, which were provided from the Dillin laboratory, Salk Institute. Strain aak-2 (ok524) of C. elegans has part of the AAK-2 subunit of AMPK deleted, rendering the kinase inactive (Apfeld et al., 2004). All strains were cultivated at 20 °C on NGM (nematode growth medium) agar plates. The plates had been seeded with 200 μL of E. coli (strain OP50) in Luria broth, and left for 1–2 days to allow the bacteria to fill the plates. Agar plates were prepared following standard protocols, with the exception that 1 m KPO₄ buffer (pH 6.0) was increased to 36.25 mL 500 mL⁻¹ agar. Both the control and treated agar were adjusted to pH 5.5. Each plate (Petri dish, 10 cm in diameter) contained 15 mL agar. Oxaloacetic acid was obtained as a solid powder from Sigma (#07753; Sigma, St. Louis, MO, USA) and stirred into the agar.

Adult worms were placed on the agar plates (without oxaloacetate) for 6 h to lay eggs, and then the adults were removed. After 3–5 days, on the same plates, the recent hatchlings were transferred to new agar plates (containing oxaloacetate in case of experimental plates), and maintained by the addition of 9.8 mg 500 mL⁻¹ fluorodeoxyuridine (FudR) (Sigma #F0503) to the plates to prevent egg hatching. Plates were changed every 5–7 days. Worms were counted every day by individuals who were blind to plate identification. Worms were scored as dead when they no longer responded to prodding. Lifespan was defined as the time elapsed from when the worms hatched until they were found dead. Worms that crawled off the plates or were otherwise lost during the assay were excluded from calculations.

To test the effect of using dead bacteria, bacteria were killed by heat prior to use, as described (Wood et al., 2004). Pharyngeal pumping rates were obtained from movies of age-synchronized 3-day-old worms on either control or oxaloacetate-supplemented FudR agar plates (oxaloacetate had been present for 48 h). The movies were obtained using a Zeiss Axiophot microscope, and played in slow motion to count the pharyngeal pumps.

In each experiment, each condition was performed at least in triplicate, with a minimum of 24 worms in each set.