Optogenetic rejuvenation of mitochondrial membrane potential extends C. elegans lifespan

Abstract

Mitochondrial dysfunction plays a central role in aging but the exact biological causes are still being determined. Here, we show that optogenetically increasing mitochondrial membrane potential during adulthood using a light-activated proton pump improves age-associated phenotypes and extends lifespan in C. elegans. Our findings provide direct causal evidence that rescuing the age-related decline in mitochondrial membrane potential is sufficient to slow the rate of aging and extend healthspan and lifespan.

The causal role of mitochondrial dysfunction and metabolic decline are central questions of aging research [1, 2]. The voltage potential across the inner membrane of mitochondria (membrane potential: Δψ_m) decreases with age in many model systems [3–6]. Δψ_m is a fundamental driver of diverse mitochondrial functions, including ATP production, immune signaling, and genetic and epigenetic regulation [7]. Decreased Δψ_m is an attractive explanation for the complex dysfunctions of aging, however, it is unclear whether decreased Δψ_m is a cause or a consequence of cellular aging.

To test these questions in a metazoan, we used optogenetics to harness light energy using a mitochondria-targeted light-activated proton pump to increase Δψ_m. Using a mitochondrial targeting sequence, we previously expressed a rhodopsin-related proton-specific pump [8] in the inner membrane of mitochondria [9]. We called this tool “mitochondria-ON” or mtON (Figure 1A) and previously characterized its optogenetic function [9]. mtON isolates Δψ_m as a single experimental variable in vivo, and requires both light activation and a cofactor, all-trans retinal (ATR), for proton pumping activity [9]. C. elegans do not produce ATR, allowing for control conditions of light exposure alone (which can be damaging [10]), mtON protein expression alone, and ATR supplementation alone (which does not affect lifespan or physiology in this context, Supplementary Table 1 and [11]). Only animals supplemented with ATR and illuminated will have mtON activity and increased Δψ_m (Supplementary Figure 1) [9].

Figure 1.

Mitochondria-ON (mtON) increased Δψ_m in vivo. A) The mitochondrial inner membrane (IM) contains the electron transport chain (ETC) which pumps protons to generate mitochondrial membrane potential (Δψ_m). mtON is an engineered light-activated proton pump, and in response to light and all trans-retinal (ATR) supplementation, pumps protons across the IM to generate Δψ_m. B) Adult worms stained with the Δψ_m indicator, TMRE. Scale bar is 250 μm. C) Quantification of relative TMRE fluorescence. One-way ANOVA with Tukey’s multiple comparisons test, day 1 vs day 4 p = 0.0001, day 1 vs day 10 p = 0.0002, n = 44, 18, 30. Data are medians ± quartiles (dotted lines). D) Pharynx TMRE fluorescence normalized to mitochondrial mass in day 4 worms. One-way ANOVA with Tukey’s test, all significant differences p = 0.0001, n = 33, 34, 26, 35 animals for each bar from left to right. Data are medians ± quartiles. E) Basal Oxygen consumption of day 1 (left) and day 4 (right) animals, no ATR n = 5 populations, ATR n = 6 populations. Data are means ± SEM, dots are individual populations. F) Maximal oxygen consumption (induced by FCCP) of the same populations (no ATR n = 5 populations, ATR n = 6 populations) in panel H. Data are means ± SEM, dots are individual populations.

We found that Δψ_m naturally declines with age in C. elegans (Figure 1B&C), as expected [2, 3, 5]. mtON activation reversed that decline in two different genetic backgrounds (Figure 1D & Supplementary Figure 2&3). mtON activation did not impact mitochondrial mass observed by mitochondrial staining and quantitative proteomics (Supplementary Figure 3D & Supplementary Figure 4A). Respiration rates were similar across conditions (Figure 1E&F) as expected, given mtON’s specificity for Δψm alone [9].

We activated mtON throughout lifespan beginning in adulthood and found an increase in lifespan compared to controls (Figure 2A&B, Supplementary Figure 1). Lifespan extension was replicated independently across three different strains, different light intensities, and in different laboratories (Supplementary Table 1), and was sensitive to the mitochondrial uncoupler, FCCP, (Figure 2C) which dissipates Δψ_m. FCCP had no effect on lifespan on its own (Figure 2C), and ATR and light did not affect lifespan in wildtype animals (Supplementary Table 1). When exposed to light below the threshold to maximally activate mtON [9] lifespan was not extended (Supplementary Figure 6A). These data together indicate that increasing Δψ_m causes increased lifespan.

Figure 2.

mtON extended lifespan and healthspan. Light treatment began at day 1 of adulthood for all experiments. A) Survival curves of mtON-expressing animals (extrachromosomal array). Only mtON activation (+ATR +light) significantly extended lifespan, log-rank (Mantel-Cox) test, *p = 0.019. Detailed statistical information for all lifespans is presented in Supplementary Table 1. B) Survival curves of mtON-expressing animals (CRISPR insertion). mtON activation significantly extended lifespan compared to the light control by log-rank (Mantel-Cox) test, *p = 0.0001, gray and light green curves. C) Increased lifespan by mtON activation was sensitive to FCCP, log-rank (Mantel-Cox) test, p = 0.0001. D) mtON activation did not affect locomotion on solid media. One-way ANOVA with Tukey’s test n = 30 animals for day 4 conditions and 40 animals for day 10. Data are medians ± quartiles (dotted lines). E) mtON activation improved mobility in liquid with age. One-way ANOVA with Tukey’s test, Day 10 no ATR light vs day 10 ATR light p = 0.002, all other significant comparisons p = 0.0001. n = 32, 32, 40, 40 animals for each violin from left to right. Data are medians ± quartiles. All statistical comparisons are presented in Supplementary Table 2. F) Model showing effects of mtON activation in vivo. The dotted arrow represents the molecular mechanisms to be investigated that link Δψ_m to aged physiology.

Mild inhibition of mitochondrial function during development (but not during adulthood) extends C. elegans lifespan [12, 13]; conversely, here we show that attenuating the age-associated decrease in Δψ_m in adults can extend lifespan. Accordingly, targets of the mitochondrial unfolded protein response did not change after mtON activation (Supplementary Figure 4B). These differences may reflect a lifespan-extending hormetic response from mitochondrial perturbation during development [14–16] versus beneficial effects from directly sustaining mitochondrial function during adulthood [2].

Organisms including humans and C. elegans have trouble moving as they age due to physiologic decline [17–19]. This functional decline was mitigated by mtON activation in worms thrashing in liquid, but not on solid media (Figure 2D&E and Supplementary Figure 5A–C). These results show that age-associated physiologic dysfunction can be improved by reversing the loss of Δψ_m that occurs with age. How improving Δψ_m may influence redox metabolites, including NAD⁺/NADH, which are known to impact biological aging, should be further assessed. To probe a potential signaling pathway, we tested the effect of mtON in long-lived worms with constitutively active AMPK signaling [20]. mtON further increased lifespan in this model (Supplementary Figure 6B and Supplementary Table 1), indicating that increased Δψ_m can additionally contribute to longevity in parallel of a canonical signaling pathway.

In summary, we used a technology that harnesses the light energy to generate Δψ_m to test the hypothesis that Δψ_m causally determines longevity in C. elegans (Figure 2F). A limitation of this study relates to the concept of optogenetic control over mitochondria; it is unclear whether this mechanism of lifespan extension is involved in other longevity paradigms. Prior studies reported that inhibition of mitochondrial function during development can increase lifespan, and our results extend the role of mitochondrial function in aging to adult intervention. Despite limitations, we show that preserved Δψ_m during adulthood is sufficient to slow normative aging and improve at least some functional measures of health. This work provides important context for understanding the role of mitochondrial function during aging and suggests the potential of novel approaches to delay aging by targeting Δψ_m specifically.

Methods

All research was approved by the University of Rochester Institutional Biosafety Committee.

C. elegans Strains and Maintenance

Nematode growth medium (NGM) was used for C. elegans culture, and all maintenance and experiments were carried out at 20 °C. OP50 E. coli was used as a food source for all experiments. Where indicated, all-trans retinal (ATR) and/or FCCP was added to the food for a final concentration of 100 μM and 10 μM respectively, accounting for the volume of the NGM. Egg-lay synchronized day 1 adult hermaphrodite animals were used for all experiments unless otherwise noted. APW32 (genotype: pha-1(e2123ts) III; jbmEx11 [pBJB20(Peft-3∷Mitofilin(N′ 187 aa)∷Mac∷GFP), pC1 (pha-1(+))]) expresses mtON as an extrachromosomal array [9]. APW273 (genotype: jbmSi10[eft-3p∷Mitofilin(187N′aa)∷Mac:mKate∷unc-54 3′UTR *cxTi10816] IV) was created using Mos1 Element-Mediated CRISPR Integration approach [21]. Briefly, PCR fragments (Supplementary Table 3) were amplified and incorporated into a Mos1 element on chromosome IV using CRISPR/Cas9 Homology-directed repair [22]. C. elegans were injected with a mix containing 25 mM KCl, 7.5 mM HEPES, 4 μg/μL tracrRNA, 0.8 μg/μL Mos1 crRNA2 (target sequence: GTCCGCGTTTGCTCTTTATT), 0.8 μg/μL dpy-10 crRNA, 50 ng/μL dpy-10 ssODN, 2.5 μg/μL purified Cas9, and 300–400 pmol/μL of each PCR repair template fragment. APW312 (genotype: jbmSi10 IV; uthIs248) expresses a constitutively active AAK-2 with mtON and was generated by crossing WBM60 (genotype: uthIs248 [Paak-2∷aak-2 genomic(aa 1–321 with T181D)∷GFP∷unc-54 3′UTR, Pmyo-2∷tdTomato] with APW273.

In Vivo Mitochondrial Membrane Potential Measurement

Animals were stained for 24 hours with both 100 nM TMRE and 12 μM Mito Tracker Green FM. TMRE was dissolved in ethanol and placed onto seeded plates, and Mito Tracker Green FM was dissolved in DMSO and added to the OP50 food. Final concentrations accounted for the entire plate NGM volume. TMRE and Mito Tracker Green FM were used to measure mitochondrial membrane potential and mitochondrial mass, respectively. Staining began at day 4 of adulthood, and animals were transferred to plates without dye for 1 hour prior to imaging to clear the gut of residual dye. Animals were mounted on 2% agarose pads under tetramisole (0.1% w/v) anesthesia. Texas Red and GFP filter sets were used to record images on an epifluorescence microscope (Nikon MVX10). Images were recorded with a Lumenera camera and associated software (Infinity Analyze). Fluorescence intensity was quantified using ImageJ by drawing regions of interest around individual animals, around the head region alone, or around individual pharynxes where indicated to determine their fluorescence intensity. Head region analysis served to quantify pharynx fluorescence and to specifically exclude intestinal staining as previously described [23]. Background signal was averaged and manually subtracted using ImageJ. Data are from 3 different experimental days. Confocal images were acquired using a Leica SP8X DMI6000 confocal microscope using a 63x oil immersion objective and a tunable white light laser (470–670 nn). Confocal images were single optical slices and not from z-stacks or maximum projections. Images were analyzed and prepared using Leica LASX Expert software and ImageJ.

mtON Activation

Illumination was carried out with a 590 nm LED array with STOmk-II stimulator by Amuza placed 2 cm away from the surface of NGM plates. Intensity was measured using a calibrated optical power meter (1916-R, Newport Corporation). Animals were exposed to 1 Hz, 0.01 – 2.1 mW/mm² light in order to help maintain temperature across experiments. Note that under all lighting conditions mtON is maximally activated (lower limit of 0.01 mW/mm² [9]) unless otherwise noted. Lifespans were carried out starting at day 1 of adulthood until death or until they were removed to measure mitochondrial parameters. A digital temperature probe was used to report temperature variability across different incubators and different light sources to ensure data comparability across lifespan experiments. This was necessary due to light sources causing small local increases in temperature. All incubators were always set to 20 °C.

Whole organism respiration

Oxygen consumption rate was measured using a Clark-type oxygen electrode (S1 electrode disc, DW2/2 electrode chamber, and Oxy-Lab control unit, Hansatech Instruments, Norfolk UK). Around 1000 animals per condition, per replicate were collected in M9 allowed to settle by gravity, rinsed in M9 buffer, settled again, and finally added to the electrode chamber in 0.5 mL of continuously stirred M9 buffer. FCCP was added at 160 μM final concentration in the chamber to induce maximal respiration. Respiration rates were measured for 10 minutes or until stable. Animals were then collected in M9 buffer for protein quantification using the Folinphenol method.

Lifespan Analysis

Animals were transferred to new plates every 2 days until reproduction ceased, and as necessary to replenish food. Where indicated, 50 μM FUDR was used in the NGM to prevent progeny from developing. Animals that did not move in response to a light touch to the head with a platinum wire were scored as dead and removed from assay plates. 1 – 3 plates for each condition were scored concurrently with ~15–70 animals per plate. All experiments were performed to maintain 20° C which sometimes required adjustment of light intensity (noted in Supplementary Table 1). Animals were illuminated only during adulthood. FCCP was added to plates for 10 μM final concentration. This dose did not affect lifespan on its own (see supplementary table 1, N2 FCCP lifespan. All lifespans comprise 3 biological replicates pooled for each experiment.

Locomotion Assays

Synchronized animals were observed and locomotion was scored by counting body bends according to previous methods [24]. Locomotion was scored in the presence of food on solid media. Thrashing was similarly analyzed with animals placed in M9 buffer to move freely in liquid. Body bends were counted for 30 seconds for each animal in all cases, and multiplied by 2 to represent body bends per minute in accordance with previously used protocols [9, 25]. Data are from at least 3 different experimental days.

Protein extraction from C. elegans

C. elegans were washed and stored in water after 4 days of illumination. Worms were centrifuged for 5 minutes at 200x g at 4°C, supernatant was discarded. The worm pellet was then resuspended in 100μl Lyse (iST Sample Preparation Kit, Preomics, Planegg/Martinsried), incubated at 95°C for 10 minutes with 500 rpm shaking and sonicated 10x 10 seconds at 30% amplitude. The sample was then centrifuged at 8000x g for 15 minutes at 4°C and the supernatant was transferred to a new vial. The protein concentration was measured with the NanoDrop 2000.

Proteomics sample preparation

100 μg protein in 50μl Lyse was recommended as starting material for the sample preparation with the Preomics iST Sample Preparation Kit. In the case the protein concentration was higher than 2μg/μl, the sample was diluted with Lyse. The preparation was performed according to the supplier guidelines. (Kit: iST Sample Preparation Kit, Preomics, Order nr. P.O.00001).

LC-MS/MS data acquisition

LC-MS/MS measurements were performed on an Ultimate 3000 RSLCnano system coupled to a Q-Exactive HF-X mass spectrometer (Thermo Fisher Scientific). Peptides were delivered to a trap column (ReproSil-pur C18-AQ, 5 μm, Dr. Maisch, 20 mm × 75 μm, self-packed) at a flow rate of 5 μL/min in 100% solvent A (0.1% formic acid in HPLC grade water). After 10 minutes of loading, peptides were transferred to an analytical column (ReproSil Gold C18-AQ, 3 μm, Dr. Maisch, 450 mm × 75 μm, self-packed) and separated using a 110 min gradient from 4% to 32% of solvent B (0.1% formic acid in acetonitrile and 5% (v/v) DMSO) at 300 nL/min flow rate. The Q-Exactive HF-X mass spectrometer was operated in data dependent acquisition (DDA) and positive ionization mode. MS1 spectra (360–1300 m/z) were recorded at a resolution of 60,000 using an automatic gain control (AGC) target value of 3e6 and maximum injection time (maxIT) of 45 msec. Up to 18 peptide precursors were selected for fragmentation in case of the full proteome analyses. Only precursors with charge state 2 to 6 were selected and dynamic exclusion of 30 sec was enabled. Peptide fragmentation was performed using higher energy collision induced dissociation (HCD) and a normalized collision energy (NCE) of 26%. The precursor isolation window width was set to 1.3 m/z. MS2 Resolution was 15.000 with an automatic gain control (AGC) target value of 1e5 and maximum injection time (maxIT) of 25 msec.

LC-MS/MS data analysis

Peptide identification and quantification was performed using MaxQuant (version 1.6.3.4). MS2 spectra were searched against the Uniprot C. elegans proteome database (UP000001940, 26672 protein entries, downloaded 21.12.2020) supplemented with the mKate-tagged proton pump protein plus common contaminants. Trypsin/P was specified as proteolytic enzyme. Precursor tolerance was set to 4.5 ppm, and fragment ion tolerance to 20 ppm. Results were adjusted to 1 % false discovery rate (FDR) on peptide spectrum match (PSM) level and protein level employing a target-decoy approach using reversed protein sequences. The minimal peptide length was defined as 7 amino acids, the “match-between-run” function was disabled. Carbamidomethylated cysteine was set as fixed modification and oxidation of methionine and N-terminal protein acetylation as variable modifications. The label free quantification (LFQ) [26] from MaxQuant was used to represent the relative abundance of proteins across samples. The ATP synthase, HSP6, HSP 60, TOM70, VDAC, mitochondrial complex proteins were manually selected. The different expression of these protein between the ATR positive and negative samples were performed using student’s t test.

Statistics and Reproducibility

Statistics were performed in GraphPad PRISM (9.3.0). No data were excluded from the analysis. No statistical method was used to predetermine sample size. Within experimental groups, animals were randomized for each experimental replicate. The Investigators were not blinded to allocation during experiments and outcome assessment. Data distribution was assumed to be normal but this was not formally tested, therefore data distributions are visualized in each figure.

Data availability

All other data supporting the findings of this study are available from the corresponding author upon reasonable request. The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium (http://proteomecentral.proteomexchange.org;) via the PRIDE [27–29] partner repository with the dataset identifier PXD033901 (http://www.ebi.ac.uk/pride/archive/projects/PXD033901).