1. Cell Biology

Metformin extends C. elegans lifespan through lysosomal pathway

  1. Jie Chen
  2. Yuhui Ou
  3. Yi Li
  4. Shumei Hu
  5. Li-Wa Shao
  6. Ying Liu  Is a corresponding author
  1. Institute of Molecular Medicine, Peking-Tsinghua Center for Life Sciences, Peking University, China
  2. Peking University, China
https://doi.org/10.7554/eLife.31268
Share this article
Cite this article
  1. Jie Chen
  2. Yuhui Ou
  3. Yi Li
  4. Shumei Hu
  5. Li-Wa Shao
  6. Ying Liu
(2017)
Metformin extends C. elegans lifespan through lysosomal pathway
eLife 6:e31268.
https://doi.org/10.7554/eLife.31268
  2. Download BibTeX
  3. Download .RIS

Abstract

Metformin, a widely used first-line drug for treatment of type 2 diabetes (T2D), has been shown to extend lifespan and delay the onset of age-related diseases. However, its primary locus of action remains unclear. Using a pure in vitro reconstitution system, we demonstrate that metformin acts through the v-ATPase-Ragulator lysosomal pathway to coordinate mTORC1 and AMPK, two hubs governing metabolic programs. We further show in Caenorhabditis elegans that both v-ATPase-mediated TORC1 inhibition and v-ATPase-AXIN/LKB1-mediated AMPK activation contribute to the lifespan extension effect of metformin. Elucidating the molecular mechanism of metformin regulated healthspan extension will boost its therapeutic application in the treatment of human aging and age-related diseases.

https://doi.org/10.7554/eLife.31268.001

eLife digest

As humans are living for longer, age-related diseases – including cancer, diabetes, cardiovascular diseases and cognitive disorders – are becoming more common. Many research groups are therefore trying to find drugs that might prevent these diseases or make them less harmful.

A drug called metformin has been shown to extend the healthy lifespan of animals such as mice and the roundworm Caenorhabditis elegans. The drug is also currently used to treat type 2 diabetes in humans and may help to prevent some other age-related diseases. However, it is still not clear exactly what effects metformin has on cells.

Healthy cells need to perform many ‘metabolic’ processes to produce the molecules necessary for survival. Cell compartments called lysosomes play a role in many of these processes because they digest unneeded biological molecules. Through a combination of biochemical and genetic experiments involving C. elegans and human cells, Chen, Ou et al. found that metformin coordinates two metabolic pathways that both depend on lysosomes. Metformin reduces the activity of a pathway (called mTOR) that boosts cell growth and the metabolic processes that build complex molecules. At the same time, the drug activates a metabolic pathway (called AMPK) that breaks down complex molecules. Overall, therefore, metformin organizes a switch from a more growth-promoting state to a more growth-restricting state.

Before metformin can be used more widely to treat human aging and age-related diseases, we need to understand how it works in even more detail. Further studies are required to discover which proteins metformin acts on inside cells, and a clinical trial has also been proposed to measure metformin’s effects on healthy human aging and age-related diseases.

https://doi.org/10.7554/eLife.31268.002

Introduction

With the discovery that aging could be genetically regulated, numerous strategies have been employed to extend lifespan in model organisms, including pharmacologic and dietary interventions (Longo et al., 2015). Identification of a chemical or pharmacological manipulation that could target human aging and lower the risks associated with age-related diseases becomes a central goal of aging research. Administration of metformin, a first-line drug for treatment of type 2 diabetes (T2D), has been shown to extend lifespan in C. elegans and mice (Anisimov et al., 2008; Cabreiro et al., 2013; De Haes et al., 2014; Martin-Montalvo et al., 2013; Onken and Driscoll, 2010; Wu et al., 2016). In addition, metformin has been shown to ameliorate diabetic and cardiovascular diseases in patients (Scarpello, 2003). Metformin also lowered the incidence of several other age-related diseases, such as cancer, metabolic syndrome and cognitive disorders (Foretz et al., 2014). Due to its broad range of health benefits and little side effects, a clinical trial named TAME (Targeting Aging with Metformin) was proposed to evaluate metformin’s protective effects against human aging and age-related diseases (Barzilai et al., 2016). However, despite its intriguing benefits to promote healthy aging, the underlying mode of action of metformin is not well understood and a subject of extensive debate.

Metformin is generally believed to act through activation of AMP-activated protein kinase (AMPK) (Fryer et al., 2002; Hawley et al., 2002; Zhou et al., 2001), a principal energy sensor that when activated, switches on catabolic pathways such as glycolysis and fatty acid oxidation to produce more ATP (Burkewitz et al., 2014; Hardie et al., 2012). It was proposed that metformin might act through inhibition of mitochondrial electron transport chain (ETC) Complex I (El-Mir et al., 2000; Foretz et al., 2014; Owen et al., 2000), resulting in a change of the AMP/ATP ratio and ultimately activating AMPK. However, this idea has been challenged recently (He and Wondisford, 2015), especially when physiological/low concentration (~70 uM) of metformin, which cannot induce AMP/ATP change, is still able to activate AMPK (Cao et al., 2014). An alternative model, in which metformin activates AMPK through the lysosome-dependent pathway was proposed (Zhang et al., 2016). In this model, metformin treatment induces lysosomal localization of the scaffold protein AXIN, which brings its associated protein liver kinase B1 (LKB1) to form a complex with v-ATPase-Ragulator on the surface of lysosome (Zhang et al., 2016; Zhang et al., 2013). LKB1 then phosphorylates the threonine residue in the activation loop of AMPK, leading to AMPK activation (Hawley et al., 1996; Shaw et al., 2004; Woods et al., 2003).

In addition to AMPK activation, metformin administration also inhibit mechanistic target of rapamycin complex 1 (mTORC1) (Kalender et al., 2010). mTORC1 constitutes another hub for energy and nutrient sensing, and switches on anabolism when activated (Schmelzle and Hall, 2000). The primary pathway for mTORC1 activation requires lysosome-localized Rag GTPases, which form RagA/B–RagC/D heterodimers and recruit mTORC1 to the surface of lysosome through directly binding to the Raptor subunit of mTORC1 (Kim et al., 2008; Sancak et al., 2008). v-ATPase-Ragulator complex on lysosomal surface is required for the spatial regulation and activation of mTORC1 by Rag GTPases (Bar-Peled et al., 2012; Dibble and Manning, 2013; Jewell et al., 2013; Sancak et al., 2010). Two distinct mechanisms have been proposed for metformin-induced mTORC1 inhibition: suppression of Rag GTPases (AMPK-independent mechanism) or AMPK-mediated phosphorylation of Raptor (regulatory associated protein of mTORC1) (AMPK-dependent mechanism) (Howell et al., 2017; Kalender et al., 2010).

Due to its short lifespan and ease of genetic manipulation, C. elegans becomes a powerful model organism to test the efficacy of metformin in promoting healthspan (Burkewitz et al., 2014). It has been shown that metformin treatment greatly extends worm lifespan and improves fitness, for example prolonging locomotory ability (Onken and Driscoll, 2010). However, several different mechanisms have been suggested for metformin’s lifespan extension effect in C. elegans. For instance, metformin administration may mimic a dietary restriction (DR) metabolism (Onken and Driscoll, 2010). In addition, alteration of methionine metabolism in C. elegans has been shown to play a partial role (Cabreiro et al., 2013). Recently, metformin has been shown to inhibit mTORC1 due to restricted transit of RagA-RagC GTPase through nuclear pore complex (NPC) (Wu et al., 2016). It is possible that these observations reflect downstream consequences of a primary action of metformin. Therefore, understanding its direct mechanism of action worth further investigation. Elucidating metformin’s mode of action will significantly boost its application to target human aging and prevent age-related diseases.

Results

Metformin coordinates mTORC1 and AMPK through the lysosomal pathway

Intrigued by the discoveries that mTORC1 and AMPK share the common activator, v-ATPase-Ragulator complex (Zhang et al., 2014), and that metformin may directly act on the lysosomal pathway to promote AMPK activation (Zhang et al., 2016), we sought to set up a pure in vitro reconstitution system to explore the mechanism of metformin’s action (Figure 1). We stably expressed LAMP1-RFP-FLAG in HEK293T cells to label lysosomes (Figure 1A) and performed FLAG pull-down to enrich lysosomes. The purity of lysosome preparation was confirmed by immunoblotting to test for the absence of early endosome, mitochondria, ER or nuclei (Figure 1B). Immuno-purified Raptor was then provided to the lysosomes. Upon amino acids stimulation, significant portion of Raptor accumulated on the lysosome (Figure 1C,F), indicating the activation of mTORC1 pathway. We then supplemented Concanavalin A (Con A), a v-ATPase inhibitor, or Metformin (Met) to the reconstitution system. Addition of Con A or metformin dissociated Raptor from the lysosome (Figure 1D,G). It has been proposed that AMPK might directly phosphorylates Raptor and thus inhibits mTORC1 (Gwinn et al., 2008). To test if metformin’s effect on mTORC1 inhibition requires AMPK or not, we repeated the in vitro experiments in AMPK knockout MEF cells, and still observed the dissociation of Raptor (Figure 1—figure supplement 1). Taken together, these results suggest that metformin inhibits mTORC1 through the lysosomal pathway independent of AMPK, possibly mimicking Con A to inhibit v-ATPase. Lastly, we provided purified AXIN and LKB1 to the in vitro system. Immuno-purified LKB1 forms a complex with endogenous STRAD and MO25, two proteins essential for optimal LKB1 activity (Figure 1—figure supplement 2). Consistent with previous findings, AXIN/LKB1 were recruited to metformin- or Con A-treated lysosomes, and were sufficient to phosphorylate endogenous AMPK in an amino acid-independent manner (Figure 1E,H) (Zhang et al., 2016; Zhang et al., 2013).

Figure 1 with 3 supplements see all
Download asset Open asset
Metformin coordinates mTORC1 and AMPK on purified lysosome.

(A) HEK293T cells stably expressing LAMP1-RFP-FLAG were mechanically broken. (B) Lysosomes were purified through immunoprecipitation (Organelle markers: LAMP2, lysosome; EEA1, early endosome; Prohibitin, mitochondria; PDI, ER; Histone H3, nucleus). (C) Lysosomal accumulation of purified Myc-raptor upon amino acids stimulation. (D–E) Lysosomal disassociation of Myc-raptor (D), lysosomal accumulation of Myc-AXIN/LKB1 and phosphorylation of AMPK (E) upon Concanamycin A (Con A) or Metformin (Met) treatment. (F–H) Quantifications of immunoblots in (C–E). Immunoblots of Myc-Raptor, Myc-AXIN and Myc-LKB1 were normalized to that of LAMP-RFP-FLAG, and immunoblots of p-AMPK were normalized to that of AMPK. Relative intensities of three independent biological replicates are shown as mean ± SEM. ns, no significant difference; *p<0.05; **p<0.01; ***p<0.001.

https://doi.org/10.7554/eLife.31268.003

It has been proposed that metformin might change the AMP/ATP ratio through inhibition of mitochondrial ETC complex I, ultimately resulting in AMPK activation (El-Mir et al., 2000; Foretz et al., 2014; Owen et al., 2000). To test if metformin could inhibit mitochondrial function and therefore activate mitochondrial stress response, we treated C. elegans hsp-6p::gfp reporter strain with metformin, or rotenone, a well-known mitochondrial ETC complex I inhibitor. Unlike rotenone, metformin treatment showed no reporter activation (Figure 1—figure supplement 3A,B). It should be noted that others have reported that metformin inhibits ETC complex I through a mechanism distinct from that of rotenone: metformin treatment increases ROS production, whereas rotenone decreases it (De Haes et al., 2014). Because our in vitro reconstitution system excludes mitochondrial contaminants, it highly suggests that metformin inhibits mTORC1 and activates AMPK mainly through the lysosomal pathway. Lastly, we directly measured the metformin’s effect on lysosomal function with the use of a cathepsin assay. Cathepsin is a class of cysteine proteinase localized within lysosomes. A cathepsin assay is a fluorescence-based assay that cleaves cathepsin’s substrate to release fluorescence. Consistent with our idea, metformin administration indeed impaired lysosomal function (Figure 1—figure supplement 3C,D).

Metformin inhibits TORC1 pathway in C. elegans

Because our in vitro biochemical results suggested that metformin might act on the v-ATPase-Ragulator complex of the lysosomal pathway to coordinate mTORC1 inhibition and AMPK activation, we next tested if metformin promotes lifespan extension in C. elegans due to similar mechanisms. To confirm that metformin also inhibits TORC1 in C. elegans, we first examined phosphorylation status of RSKS-1, the human ribosomal protein S6 kinase B1 (S6K) homolog in C. elegans. We used an antibody that specifically detects Thr-389, a highly conserved phosphorylation site among species (Figure 2—figure supplement 1A). This antibody was validated with the used of rsks-1 RNAi (Figure 2—figure supplement 1B). Notably, metformin treatment significantly decreased the level of RSKS-1 phosphorylation (Figure 2A,B), suggesting an inhibition of TORC1 pathway. We also tested the subcellular localization of HLH-30, an ortholog of transcription factor EB (TFEB) that translocates to the nucleus upon TORC1 inhibition (Lapierre et al., 2013; Perera and Zoncu, 2016; Settembre et al., 2011). Indeed, HLH-30 translocalized to the nucleus in the tail region of C. elegans upon metformin administration (Figure 2C,D). Once in the nucleus, HLH-30 regulates the transcription of genes related to autophagy, lysosome function and lipid hydrolysis (Lapierre et al., 2013; O'Rourke and Ruvkun, 2013). Consistently, we also observed up-regulation of transcript levels of hlh-30 downstream targets upon metformin treatment (Figure 2E). Taken together, these results indicate that metformin inhibits TORC1 pathway in C. elegans.

Figure 2 with 2 supplements see all
Download asset Open asset
Metformin inhibits TORC1 pathway in C.elegans.

(A) Representative western blotting of RSKS-1 phosphorylation in the presence or absence of Metformin. (B) Quantification of immunoblots in (A). Relative intensities of 3 independent biological replicates are shown as mean ± SEM. (C) Representative fluorescent images of HLH-30 nuclei localization in the presence or absence of metformin. (D) Percentage of worms with HLH-30 nuclear localization in (C) was quantified. Mean ± SEM of 3 independent biological replicates are shown (sample size:≥40 worms). (E) Q-PCR analysis of HLH-30 target genes in the presence or absence of metformin. ~300 worms were pooled in each sample. Data from three independent biological replicates are shown as mean ± SEM. ns, no significant difference; *p<0.05; **p<0.01; ***p<0.001.

https://doi.org/10.7554/eLife.31268.007

We also tested if metformin’s effect on TORC1 inhibition requires AMPK in C. elegans. Consistent with our results in AMPK-/- MEF cells (Figure 1—figure supplement 1), metformin treatment of aak-2 (Ce.AMPK) loss-of-function mutants was still able to drive HLH-30 nuclear accumulation (Figure 2—figure supplement 2A,B) and elevate the expression of HLH-30 target genes (Figure 2—figure supplement 2C).

Metformin extends healthspan partially through TORC1 inhibition

Inhibition of TORC1 has been shown to result in lifespan extension (Jia et al., 2004; Vellai et al., 2003); also reviewed by Zoncu et al., 2011). To distinguish if metformin induces lifespan extension solely through Ce.TOR inhibition, or the activation of Ce.AMPK also plays a role, we performed lifespan analysis on control worms or daf-15 heterozygous mutants in the presence or absence of metformin (Figure 3—figure supplement 1A). Compared with the control worms, daf-15 heterozygous mutants had a longer lifespan and a decreased level of RSKS-1 phosphorylation (Figure 3—figure supplement 1B,C), indicating that TORC1 inhibition extends C. elegans lifespan. More importantly, metformin treatment greatly extended the lifespan of daf-15 heterozygous mutants (Figure 3A).

Figure 3 with 2 supplements see all
Download asset Open asset
Metformin extends healthspan partially due to TORC1 inhibition.

(A) Lifespan analysis of control worms or daf-15 heterozygous mutants in the presence or absence of metformin. (B) Representative images of Oil-Red-O (ORO) staining of control worms or daf-15 heterozygous mutants in the presence or absence of metformin. (C) Quantification of (B). Error bars represent mean ± SEM of 3 independent biological replicates (sample size: n ≥ 40 worms). (D–E) Locomotion (D) and age pigments (E) were measured in control worms or daf-15 heterozygous mutants in the presence or absence of metformin. Mean ± SEM of 3 independent biological replicates are shown (sample size: n ≥ 20 worms for locomotion assay; n ≥ 40 worms for age pigments assay). (F) Representative fluorescent images of HLH-30 nuclei localization in control worms or daf-15 heterozygous mutants in the presence or absence of metformin. Arrows indicate nuclear localized HLH-30::GFP. (G) Quantification of (F). Percentage of unc-24/+; HLH-30::GFP or unc-24/daf-15; HLH-30::GFP worms with nuclear accumulation of HLH-30 were counted. Error bars represent mean ± SEM of 3 independent biological replicates. (sample size: n ≥ 40 worms) (H) Representative western blotting of AAK-2 phosphorylation in the presence or absence of metformin. (I) Quantification of (H).~300 worms were pooled in each protein sample. Error bars represent mean ± SEM of 3 independent biological replicates. ns, no significant difference; *p<0.05, **p<0.01, ***p<0.001.

https://doi.org/10.7554/eLife.31268.010

To investigate the metabolic effects of metformin on daf-15 heterozygous mutants, we fed L4 worms with metformin and conducted Oil-Red-O (ORO) staining to detect neutral fat levels (Figure 3—figure supplement 1D). Similar to wild type animals, metformin administration further decreased neutral fat level in daf-15 heterozygous mutants (Figure 3B,C). Aged worms start to show muscle deterioration and decrease locomotion rates (Huang et al., 2004; Onken and Driscoll, 2010). Metformin treatment significantly increased the locomotory ability (counted by the average bends of worm body per 60 s) and reduced age pigments of wild type worms and daf-15 heterozygous mutants (Figure 3—figure supplement 1D, and Figure 3D,E), indicating that metformin also improves fitness of daf-15 heterozygous mutants.

Metformin’s beneficial effect on lifespan and fitness of daf-15 heterozygous mutants could be due to a pathway additive to that of TORC1 inhibition (e.g. AMPK pathway), or simply due to a further suppression of TOR activity in daf-15 heterozygous mutants. To test if metformin could further inhibit TORC1 activity in daf-15 heterozygous mutants, we compared HLH-30 nuclear localization between control animals and daf-15 heterozygous mutants in the presence or absence of metformin treatment. We found that in daf-15 heterozygous mutants, significant portion of HLH-30 already localized in the nucleus, suggesting an inhibition of TORC1 activity. More importantly, metformin administration increased HLH-30 nuclear localization in control worms, but not in daf-15 heterozygous mutants (Figure 3F,G). In addition, we performed lifespan experiments with loss-of-function mutants of known downstream targets of TORC1, which have been reported to mediate TORC1 inhibition-induced longevity (Robida-Stubbs et al., 2012). We found that metformin could still extend lifespans of hlh-30, pha-4 or skn-1 mutants (Figure 3—figure supplement 2). These results suggest that metformin may act on a pathway additive to that of TORC1 inhibition to confer fitness benefit. Given that metformin activated AMPK in daf-15 heterozygous mutant to the same level as that in the control worms (Figure 3H,I), it is likely that metformin promotes longevity also through activation of AMPK.

Metformin extends lifespan through lysosome-dependent activation of AMPK

v-ATPase-Ragulator complex may function as lysosomal acceptor for AXIN/LKB1, which translocate to lysosome and activate AMPK upon metformin treatment (Figure 1E). Thus, to test the lysosome-dependent AMPK activation for metformin-induced lifespan extension, we first performed lifespan analysis on vha-3 (subunit of v-ATPase V0 domain), vha-12 (subunit of v-ATPase V1 domain), lmtr-3 (LAMTOR 3 subunit of Ragulator) and lmtr-2 (LAMTOR 2 subunit of Ragulator) loss-of-function mutants (Figure 4—figure supplement 1 and Figure 4A). VHA-3, VHA-12, LMTR-3 and LMTR-12 all localized on the lysosomes of C. elegans, as they have similar staining pattern with lysosomal marker LMP-1 (Figure 4—figure supplement 2). All these four mutants had impaired lysosomes, which could be rescued with the expression of corresponding gene (Figure 4—figure supplement 3). vha-3 and vha-12 mutants had longer lifespans compared with wild type animals, whereas lifespan of lmtr-2 or lmtr-3 mutants was comparable to that of the wild type animals (Figure 4B and C and Figure 4—figure supplement 4A and B). Lifespan extension in vha-3 and vha-12 mutants might be due to Ce.TOR inhibition, because v-ATPase is required for the spatial regulation and subsequent activation of TORC1. Consistently, failure of lmtr-3 or lmtr-2 to induce lifespan extension suggested that Ce.TOR pathway was not inhibited under lmtr-3 or lmtr-2 deficiency. Indeed, RNAi knockdown of vha-3 but not lmtr-3, was sufficient to induce the nuclear accumulation of HLH-30 and activate HLH-30 downstream targets (Figure 4—figure supplement 4C and D, and Figure 4D–F). In addition, knocking down of vha-3 but not lmtr-3, decreased the level of RSKS-1 phosphorylation (Figure 4G and H).

Figure 4 with 4 supplements see all
Download asset Open asset
Metformin extends lifespan through v-ATPase-Ragulator-dependent activation of AMPK.

(A) A scheme depicting genes within C. elegans TORC1 and AMPK pathways. (B–C) Lifespan analysis of wild type, vha-3 (B), or lmtr-3 (C) animals in the presence or absence of metformin. (D) Representative fluorescent images of HLH-30 nuclear localization in worms administrated with control, vha-3 or lmtr-3 RNAi. (E) Quantification of (D), percentage of worms with HLH-30 nuclear accumulation. Mean ± SEM of 3 independent biological replicates are shown (sample size: n ≥ 40 worms). (F) Q-PCR analysis of HLH-30 target genes in worms administrated with control, vha-3 or lmtr-3 RNAi. ~ 300 worms were collected for each mRNA sample. Data from 3 independent biological replicates are shown as mean ± SEM. (G) Representative immunoblots of RSKS-1 phosphorylation in worms administrated with control, vha-3 or lmtr-3 RNAi. (H) Quantification of (G).~300 worms were pooled in each protein sample. Relative intensities of 3 independent biological replicates are shown as mean ± SEM. (I) Representative immunoblots of AAK-2 phosphorylation in wild type, vha-3 or lmtr-3 mutants, in the presence or absence of metformin. (J) Quantification of (I).~300 worms were collected in each protein sample. Relative intensities of 3 independent biological replicates are shown as mean ± SEM. ns, no significant difference; *p<0.05; **p<0.01; ***p<0.001.

https://doi.org/10.7554/eLife.31268.013

More importantly, metformin administration could not further induce the lifespan extension, nor AAK-2 activation in v-ATPase mutants vha-3 and vha-12, or Ragulator mutants lmtr-3 and lmtr-2 (Figure 4B–C and I–J and Figure 4—figure supplement 4A, B, E and F), suggesting that metformin’s effect on lifespan extension may also depend on the lysosomal pathway of AMPK activation. It should be noted that metformin even shortens the lifespan of lmtr-3 or lmtr-2 mutants (Figure 4C and Figure 4—figure supplement 4B). However, fractions of censored animals were not increased in metformin-treated lmtr-3 or lmtr-2 mutants (Figure 4—figure supplement 4G), suggesting that metformin did not make the mutants sick.

AXIN is a scaffold protein required for lysosomal localization of LKB1 and lysosome-dependent activation of AMPK (Zhang et al., 2016; Zhang et al., 2013). We could not successfully express AXL-1 (C. elegans ortholog of AXIN) with fluorescent tag in worms and test for its subcellular localization. Therefore, we expressed EGFP-tagged AXL-1 in mammalian cells and showed that AXL-1 only localized on the lysosome upon metformin treatment (Figure 5A). Consistently, without metformin administration, axl-1 mutant alone did not perturb lysosomal function (Figure 5B). We then examined the role of AXIN in metformin-induced lifespan extension. Metformin treatment failed to activate AAK-2 (Ce.AMPK) in axl-1 mutant (AXIN-deficient animals) (Figure 5C,D). Consistently, metformin was no longer able to extend lifespan of axl-1 mutant animals (Figure 5E). Similar to axl-1, metformin could not extend lifespan of par-4 (C. elegans ortholog of LKB-1) mutants as well (Figure 5F) (Onken and Driscoll, 2010). Taken together, these genetic evidences indicate that v-ATPase-Ragulator-AXIN-LKB1-based lysosome pathway is required for the metformin’s effects of lifespan extension.

Figure 5
Download asset Open asset
Metformin extends lifespan through axl-1-dependent activation of AMPK.

(A) Representative fluorescent images to test co-localization of AXL-1 with LAMP1. HEK293T cells stably expressing AXL-1-EGFP and LAMP1-RFP-FLAG were cultured in the presence or absence of 2 mM metformin for 12 hr. Scale bar: 10 um. (B) Representative fluorescent images of Magic Red Cathepsin assay in wild type worms or axl-1 mutants. Scale bar: 100 um. Error bars represent mean ± SEM of 3 independent biological replicates (sample size: n ≥ 40). (C) Representative immunoblots of AAK-2 phosphorylation in wild type or axl-1 mutants with or without metformin treatment. (D) Quantification of (C).~300 worms were pooled in each protein sample. Relative intensities of 3 independent biological replicates are shown as mean ± SEM. (E–F) Lifespan analysis of wild type, axl-1 (E), or par-4 mutants (F) in the presence or absence of metformin. ns, no significant difference; *p<0.05; **p<0.01; ***p<0.001.

https://doi.org/10.7554/eLife.31268.018

Metformin attenuates age-related fitness decline through lysosome-dependent activation of AMPK

It has been shown that metformin triggers a dietary restriction-like state and extends C. elegans lifespan through AMPK pathway (Onken and Driscoll, 2010). Indeed, metformin administration induced a dietary restriction-like state to decrease neutral fat level in wild type animals (Figure 6A,B). Conversely, vha-3 (v-ATPase V0), lmtr-3 (Ragulator) and axl-1 (AXIN) mutants retained the same levels of ORO staining in the presence or absence of metformin (Figure 6A,B), suggesting that the dietary restriction-like state triggered by metformin requires the lysosome-dependent activation of AMPK.

Figure 6 with 1 supplement see all
Download asset Open asset
Metformin attenuates age-related fitness decline through lysosome-dependent activation of AMPK.

Neutral fat deposition (A–B), locomotion (C–F) and age pigments (G) were measured in wild-type worms, or vha-3, lmtr-3 or axl-1 mutants in the presence or absence of metformin. Error bars represent mean ± SEM of 3 independent biological replicates. (sample size: n ≥ 20 worms for locomotion assay; n ≥ 40 worms for ORO staining or age pigments assay). (H) Metformin may target v-ATPase-Ragulator complex and promote longevity through coordination of Ce.TORC1 and Ce.AMPK.

https://doi.org/10.7554/eLife.31268.019

Metformin treatment also significantly increased the locomotory ability and lowered age pigments of wild type worms (Figure 6C and G), indicating that metformin promotes youthful physiology and fitness of C. elegans. Strikingly, mutation of vha-3, lmtr-3 or axl-1 impaired metformin's effects on locomotion and age pigments (Figure 6D–G), suggesting that lysosome-dependent activation of AMPK is also required for metformin-attenuated fitness decline in aged animals.

Discussion

Using a pure in vitro reconstitution system that excludes mitochondria, we showed that metformin coordinates mTORC1 inhibition and AMPK activation through lysosomal pathway. We further employed genetic manipulation to show that metformin extends C. elegans lifespan and attenuates age-related fitness decline via similar mechanism that requires v-ATPase-Ragulator-AXIN/LKB1 of the lysosomal pathway (Figure 6H). Metformin may function by targeting and priming v-ATPase-Ragulator complex on lysosome membrane, which serves as a hub to coordinate mTORC1 and AMPK pathways and govern metabolic programs. It is possible that metformin administration might result in a conformational change of v-ATPase-Ragulator complex, which dissociates mTORC1 from lysosome and allows the docking of AXIN/LKB1 for AMPK activation (Figure 6—figure supplement 1). It will be of particular interest in the future to test if v-ATPase is the direct target of metformin. Elucidating the molecular mechanism of metformin-mediated lifespan extension will boost its application in the treatment of human aging and age-related diseases.

Materials and methods

Cell culture

Request a detailed protocol

HEK293T cells used for protein purification, virus packaging and stable cell line generation, and AMPKα1/2 double knockout (DKO) MEF cells used for stable cell line generation were cultured with DMEM high glucose medium (HyClone, SH30243.01), supplemented with 10% FBS (Gibico, 10099–141) and 1% Penicillin/Streptomycin (HyClone SV30010). 293T HEK cells were obtained from ATCC, not authenticated, and mycoplasma negative. AMPKα1/2 double knockout (DKO) MEF cells were provided by Dr. Benoit Viollet, validated by immunoblotting and determined as mycoplasma negative.

Generation of stable cell line

Request a detailed protocol

HEK293T cells and AMPKα1/2 DKO MEF cells stably expressing LAMP1-RFP-FLAG fusion protein, or HEK293T cells stably expressing AXL-EGFP and LAMP1-RFP-FLAG were generated with the lentiviral packaging and infection system. 0.8 × 106 HEK293T cells were seeded into one well of a 6-well-plate and cultured with 2 ml medium. After about 20 hr, 0.75 ug pMDLg/pRRE, 0.3 ug pRSV-Rev, 0.45 ug pCI-VSVG and 1.5 ug pLJM1-LAMP1-RFP-FLAG (addgene) plasmids were co-transfected into cells. Medium was changed 8 hr after transfection. Cells were then cultured for additional 24 hr to allow virus production, and 2 ml medium containing virus was harvested (virus packaging). 300 ul virus-containing medium was diluted with 900 ul fresh medium, and added into one well of a 12-well-plate containing HEK293T cells or AMPKα1/2 DKO MEF cells at ~40% confluence (virus infection). Several hours after infection, cells stably expressing LAMP1-RFP-FLAG were checked by RFP fluorescence. ~100% HEK293T cells or ~20% AMPKα1/2 DKO cells contain RFP signal. HEK293T LAMP1-RFP-FLAG stable cell line was then infected by virus carrying AXL-EGFP to generate double stable cell line.

Protein purification

Request a detailed protocol

Recombinant proteins Myc-LKB1 (human), Myc-AXIN (mouse), and Myc-Raptor (human) were immunopurified from HEK293T cells. For purification of each protein, 20 × 15 cm dishes of HEK293T cells were cultured until 80% confluence. 28 ug of plasmid was transfected to each dish of cells with Lipofectamine 3000 Transfection Reagent. Medium was changed 8 hr after transfection. After additional 24 hr, medium was carefully discarded and cells were rinsed with PBS. 1.5 ml of TX-100 lysis buffer (20 mM Tris-HCl, 150 mM NaCl, 1 mM Na2EDTA, 1 mM EGTA, 2.5 mM Sodium pyrophosphate, 1 mM β-glycerophosphate, 1% Triton X-100, PH = 7.4) supplemented with proteinase inhibitors was added to each dish of cells. Cells were incubated on ice for 10 min, and then scraped off the dishes. Cell lysate was sonicated at 60% Ampl for 5 × 30 s (SONICS VCX 130 PB, UK), and centrifuged at 4°C, 10000 rpm for 10 min. Supernatant was mixed with 500 ul Myc agarose beads, and rotated overnight at 4°C. Beads were washed three times with TX-100 lysis buffer and three times with fractionation buffer (50 mM KCl, 90 mM K-Gluconate, 1 mM EGTA, 5 mM MgCl2, 50 mM Sucrose, 5 mM Glucose, 20 mM HEPES, pH = 7.4, 2.5 mM ATP and protease inhibitors were added right before usage), and resuspended in 800 ul of fractionation buffer, followed by addition of 10 ul 10 mg/ml Myc peptide. After rotating at 4°C for 4 hr, Myc tagged protein in 800 ul fractionation buffer was harvested, followed by addition of 160 ul glycerol, and stored at −20°C.

Lysosome purification and in vitro reconstitution

Request a detailed protocol

LAMP1-RFP-FALG stable cells were cultured until confluence. For each sample, 1 × 10 cm dish of HEK293T cells or 5 × 10 cm dishes of MEF AMPKα1/2 DKO cells were scraped off the plate, and pelleted at 1000 rpm, room temperature for 3 min. Cell pellet was washed once with fractionation buffer, resuspended in 0.8 ml fractionation buffer, and mechanically broken by spraying six times through a 23G needle attached to a 1 ml syringe. Broken cells were spun down at 2000 g for 10 min at 4°C to pellet the nuclei, yielding a post-nuclei supernatant (PNS). The PNS was diluted to 2 ml with fractionation buffer, and subjected to immunoprecipitation with 50 ul FLAG magnetic beads (60% slurry) for 2 hr at 4°C on a rotator. Lysosome on beads was washed three times and resuspended in 300 ul fractionation buffer, supplemented with 1x amino acids (combining 50x Gln-free MEM amino acids mixture with 100x Glutamine solution from Gibico), 250 uM GTP and 100 uM GDP. Lysosome was then rotated at 37°C, 650 rpm on a Thermomixer (Eppendorf, Germany) for 15 min (activation step for Raptor binding). 40 ul of purified Myc-Raptor protein was added, and the system was shaken for additional 25 min (lysosome binding step for Raptor). 10 uM Concanamycin A (ConA) or 300 uM Metformin (Met) was then provided into the reaction system, and incubated for 30 min (Raptor off lysosome membrane step). Lysosome coated magnetic beads were isolated from the reaction system, resuspended with 300 ul fractionation buffer supplemented with 80 ul purified Myc-AXIN and 20 ul purified Myc-LKB1 (formed a complex with endogenous STRAD and MO25), and then shaken on Thermomixer for 25 min (AXIN/LKB1 lysosome translocation and AMPK activation step).

Western blotting

Request a detailed protocol

For in vitro reconstitution, each sample was prepared with 30 ul 2x laemmli sample buffer. Purified lysosomes were checked with LAMP2 (abcam ab25631), EEA1 (CST 3288), Prohibitin (Santa Cruz 28259), PDI (Santa Cruz 20132), and Histone H3 (Huaxingbio, HX1850) antibodies. Myc tagged proteins were detected with anti-Myc antibody (ImmunoWag YM3203). LAMP1-RFP-FLAG labeled lysosomes were probed with Flag antibody (Sigma F7425). T172 site phosphorylated AMPKα subunit and total AMPKα subunit were detected with p-AMPKα (CST 2535) and AMPKα (CST 2532) antibodies respectively. STRAD (Santa Cruz sc-34102) and MO25 (CST 2716 s) antibodies were used to detect Myc-LKB1-STRAD-MO25 complex.

For detection of AAK-2 or RSKS-1 phosphorylation, worms at indicated stage were lysed with 2x laemmli sample buffer equal to the volume of worm pellet. AAK-2 phosphorylation was probed with human p-AMPKα (T172) antibody (CST 2535). RSKS-1 phosphorylation was probed with human p-S6K1 (T389) antibody (CST 9205). Tubulin (abcam ab6161) was used as loading control.

Nematode culture

Request a detailed protocol

Detailed information of Caenorhabditis elegans strains used in this paper was provided in Supplementary file 1–2. Worms were maintained on nematode growth medium (NGM) plate with OP50 as standard food. All worms, except for pha-4 and par-4 mutants, were cultured at 20°C. Metformin was added into molten agar at the concentration of 50 mM while preparing plates. Rotenone stock was prepared at 10 ug/ul in DMSO. 1 ul rotenone stock was diluted with 300ul M9 and added directly onto a 6 cm NGM plate and hood-dried.

RNA interference (RNAi) clones were grown at 37°C overnight in LB containing 50 ug/ml carbenicillin. 200 ul bacteria were then seeded onto NGM containing 1 mg/ml IPTG and 50 ug/ml carbenicilin, hood-dried and cultured overnight to induce dsRNA expression. L1 stage Worms were then seeded onto bacteria lawn, and raised until L4 stage for efficient knockdowns.

Lifespan assays

Request a detailed protocol

~100 L4 stage (day 0 for lifespan assays) worms (except for lmtr-2, lmtr-3, par-4 and pha-4) were transferred to plates with or without metformin. For lmtr-2 or lmtr-3 mutant strain,~250 worms were used, due to high percentage of censored worms. For par-4 mutant, worms were cultured at 15°C until L4 stage (day 0 for lifespan assays), and ~100 worms were transferred to 25°C for lifespan assay. For pha-4 mutant, worms were cultured at 25°C until day 1 adult (day 0 for lifespan assays), and ~100 worms were transferred to 15°C for lifespan assay. FUDR was used to prevent reproduction. Worms were transferred to new plates and counted every other day. FUDR was omitted after day 12. Animals that did not move when gently prodded were scored as dead. Animals that crawled off the plate or died from vulva bursting were censored.

Oil-Red-O (ORO) staining

Request a detailed protocol

Worms were washed with 1x PBS and resuspended in 120 ul 1x PBS, followed by addition of 120 ul 2x MRWB buffer (160 mM KCl, 40 Mm NaCl, 14 mM Na4EGTA, 1 mM Spermidine-HCl, 0.4 mM Spermine, 30 mM Na-PIPES, 0.2% β-mercaptoethanol, PH = 7.4) containing 2% PFA. Samples were fixed by gently rocking at room temperature for 1 hr. Worms were then washed with 1x PBS to remove PFA, and resuspended in 300 ul 60% isopropanol to dehydrate for 15 min at room temperature. Worms were pelleted and resuspended with 1 ml 60% ORO solution, and rotated overnight at room temperature (Preparation of 60% ORO solution: commercial liquid ORO stain was diluted to 60% with H20, rocked overnight at room temperature, and filtrated with 0.22 um filter right before usage). Worms were settled and ORO staining solution was removed. 200 ul 1x PBS (with 0.01% triton X-100) was added to resuspend the worms. Worms were quickly transferred on to a glass slide and photographed with Zeiss Imager M2 microscope.

Locomotion assay

Request a detailed protocol

Worms were transferred into a drop of M9 on a glass slide, and filmed with Zeiss Imager M2 microscope. Body bends were counted by reviewing each frame of the 60 s film.

Age pigment fluorescence detection

Request a detailed protocol

Worms were mounted onto an agarose pad attached to a glass slide and photographed with Zeiss Imager M2 microscope. Fluorescence intensity was counted by Image J.

Lysosome function assay

Request a detailed protocol

Magic Red stain (ImmunoChemistry Technologies #938, Bloomington, Minnesota) can be cleaved by cathepsin B, and generate red fluorescent substrate in functional lysosomes. Magic Red stain was prepared in 260x DMSO stock following the manufacturer’s instructions. 3.8 ul stock was 5x diluted with M9, spread onto a well of 24-well-plate containing 1 ml NGM, and dried in hood. For metformin’s effect on lysosomal function, wild type L1 stage worms were raised to L4 stage on control or metformin NGM plate containing Magic Red stain. For cathepsin assay, L4 stage worms were transferred onto Magic Red containing plate and cultured overnight. Worms were photographed with Zeiss Imager M2 microscope and quantified by Image J.

Lysosomal localization assay

Request a detailed protocol

The longest CDS sequences of lmp-1, vha-3/12 and lmtr-2/3 were amplified from worm genomic DNA, and cloned into expression vector containing rpl-28 promoter (whole body expression). LMP-1::GFP was initially integrated into C. elegans genome through UV irradiation, and worms were backcrossed for six times. mCherry-VHA-3/12 or mCherry-LMTR-2/3 was then injected into worms expressing LMP-1::GFP. Worms were photographed by Zeiss Imager M2 microscope. The longest CDS sequence of axl-1 was amplified from worm genomic DNA and cloned into mammalian expression construct containing CMV promoter and C-terminal EGFP tag. HEK293T cells stably expressing AXL-1-EGFP and LAMP-1-RFP-FLAG were generated as described above. Cells were imaged by Zeiss LSM710 confocal microscope.

Reporter assays

Request a detailed protocol

hsp-6p::gfp reporter worms were photographed by Zeiss Imager M2 microscope and the fluorescent intensity was quantified by Image J. hlh-30p::hlh-30::gfp reporter worms were photographed by Zeiss Imager M2 microscope and percentage of worms with nuclear localization of HLH-30 was counted.

To monitor HLH-30 nuclear localization in control or daf-15 heterozygous mutant, we injected hlh-30p::hlh-30::gfp construct into control or daf-15 heterozygous mutant respectively. Worms were photographed by Zeiss Imager M2 microscope. Percentage of worms with nuclear localization of HLH-30 was counted.

RNA extraction and Q-PCR

Worms were washed off plates with M9 and resuspended with TRIzol Reagent and frozen by liquid nitrogen. Total RNA was isolated by chloroform extraction and isopropanol precipitation. 200 ng total RNA was used for reverse transcription with One-Step cDNA Synthesis Kit (TransGen Biotech, China). Real-time PCR was carried out using SYBR GREEN PCR Master Mix (Biorad, Hera Claes, California). Quantifications of transcripts were normalized to rpl-32.

Q-PCR primers

Request a detailed protocol
Target geneForward primerReverse primer
 rpl-32AGGGAATTGATAACCGTGTCCGCATGTAGGACTGCATGAGGAGCATGT
 hlh-30CTCATCGGCCGGCGCTCATCAGAACGCGATGCGTGGTGGG
 lgg-1ACCCAGACCGTATTCCAGTGACGAAGTTGGATGCGTTTTC
 lgg-2CTGCAAATTCCTAGTACCCGAGCATAGAATTTGACACCATTGAGC
 atg-16.2ATGTCATATCTGGATCTGCGGACGTTGCATCTGAAGAGCGTG
 atg-18AAATGGACATCGGCTCTTTGTGATAGCATCGAACCATCCA
 sul-2ATGGCAGCAGAAGGCACCCGGCCATTTTCCAACCATGCCAGTTGC
 ctsa*TTCTCCTCGAGGCGCGGGATTCCAACGCCAATTGGGGACTC
 cpr-1CGCCAAGGACAAGCACTTCGGAACCTTGGCCTTTCCGGCGAC
 lipl-1GTGACATTTGTTTTTCCATATAGCAAATTAAACCGACCAC
 lipl-2AATACGAGTCAAATCATTGAAGTAACACTCGTTTTTCCATAA
 lipl-3ATGGGCAGGCAAATCCACCAAGTTGTTCTGCGCAATTATA

References

Decision letter

  1. Hong Zhang
    Reviewing Editor; Institute of Biophysics, Chinese Academy of Sciences, China

In the interests of transparency, eLife includes the editorial decision letter and accompanying author responses. A lightly edited version of the letter sent to the authors after peer review is shown, indicating the most substantive concerns; minor comments are not usually included.

[Editors’ note: a previous version of this study was rejected after peer review, but the authors submitted for reconsideration. The first decision letter after peer review is shown below.]

Thank you for submitting your work entitled "Metformin Extends Lifespan by Coordination of mTOR and AMPK through Lysosomal Pathway" for consideration by eLife. Your article has been reviewed by three peer reviewers, one of whom is a member of our Board of Reviewing Editors and the evaluation has been overseen by Senior Editor. The reviewers have opted to remain anonymous.

Our decision has been reached after consultation between the reviewers. Based on these discussions and the individual reviews below, we regret to inform you that your work will not be considered further for publication in eLife.

The reviewers agreed that this is an interesting study that aims to show that metformin acts on TOR and AMPK at the lysosome, and that this site is critical for the effects of metformin on C. elegans lifespan. However, some of the conclusions seem to be based on small changes observed in single experiments, and analysis of more replicates with assessment of statistical significance would make the soundness of the conclusions more convincing. A substantial amount of work needs be conducted to address the concerns raised by the reviewers, and it is unlikely that this can be done within the two–month limit set by eLife, so we are rejecting the paper. Nevertheless, if your results hold up, I do encourage you to resubmit your work after you have obtained sufficient lines of validation.

Reviewer #1:

In this manuscript, the authors used an in vitro system to show that metformin coordinates mTORC1 inhibition and AMPK activation through the lysosomal pathway. They further demonstrated that both TORC1 inhibition and Axin/LKB1–mediated AMPK activation contribute to the lifespan extension effect of metformin in C. elegans. This study provides molecular mechanism of metformin–mediated lifespan extension.

Major concerns:

1) The authors showed that metformin treatment triggers nuclear localization of HLH–30::GFP in the tail region. How the cells in the tail affect the lifespan? Does loss of function of hlh–30 suppress the lifespan extension effect of metformin?

2) Metformin treatment further extends lifespan of daf–15 heterozygous mutants. This could be due to that metformin acts on a pathway additive to that of TORC1 inhibition. Alternatively, this may be simply due to a partial reduction in mTOR activity in daf–15 heterozygous mutants.

3) Quantification results should be provided for Figure 1E, Figure 2A and Figure 3B, C.

Reviewer #2:

Overall, this is an interesting study that aims to show that metformin acts on TOR and AMPK at the lysosome, and that this site is critical for the effects of metformin on C. elegans lifespan. While a novel in vitro reconstitution assay has been developed, and several new C. elegans lysosomal mutants have been analyzed, the study fails to make a convincing case, mainly due to insufficient lines of validation and major rigor issues.

Lysosomal mutants:

In addition to major rigor issues (see separate box), another major concern of this article is that the authors are using several new lysosomal mutants, i.e. lmtr–2/3 and axl–1, but these mutants are inadequately described and characterized. A description of the alleles, whether the mutations are loss– or gain–of function mutations, and the lysosomal function of these genes have to be established. Similar concerns relate to vha–3 and vha–12, which were analyzed by RNAi. Of particular interest is whether these genes indeed localize and function within the lysosome. Lysosomal localization assays and functional assays (cathepsin assays) need to be carried out to validate these mutants, considering the rationale of the manuscript overall. Metformin's effects on lysosomal function should also be tested. Of particular concern is Figure 3F, in which the authors describe that 80% of the lmtr–2/3 mutants are censored due to vulval rupture. This censoring will greatly affect the sample size and thus the interpretation of the lifespan experiments. Again, it is crucial for the authors to share these details of their lifespan analyses for all mutants tested. The authors should also include par–4, the LKB1 homolog in their analyses of the lysosomal mutants.

TOR activity assays:

Throughout the manuscript, the authors are using several readouts to assay TOR activity in C. elegans: HLH–30 nuclear localization, DAF–16 nuclear localization, increase in hlh–30 transcriptional reporter, increase in gcs–1 transcriptional reporter, mRNA levels of skn–1, daf–16 and daf–15/Raptor, mRNA levels of TOR pathway components, skn–1–dependent target genes. The authors seem to randomly choose the assay in their different experimental conditions or genetic manipulations. While we acknowledge assay limitations for TOR activity in C. elegans, the authors should be consistent and use the same assays throughout. To this point, analysis of S6K phosphorylation would be a direct way to assay TOR pathway status. Finally, analysis of the mRNA levels of HLH–30–regulated genes seems relevant to include, considering the use of the HLH–30 translocation assay.

TOR inhibition:

The authors use a daf–15 heterozygous mutant to demonstrate that TOR inhibition and Metformin are additive in their effects on lifespan extension. Because the mutant is heterozygous (i.e., not null), this is not a sound conclusion, as Metformin could still exert its effects on the residual daf–15/TOR function. To strengthen their conclusion, the authors should rather perform epistasis experiments with known downstream targets of TOR, e.g., with pha–4, skn–1 and hlh–30 mutants. Additionally, the authors should assay the battery of TOR activity assay also with the daf–15 heterozygous mutants (especially HLH–30 nuclear localization).

Fitness assays:

It would be important to include the daf–15 heteroygous mutants in the analysis of fitness or healthspan parameters as a positive control, since metformin should still have a positive and beneficial effect on these mutants. The oil–red–o stainings only inform on decreased fat levels, however not on a mechanism by which this is achieved. The authors need to refrain from making the conclusion that Metformin activates lipolysis, as there is no data supporting this claim. Again, more information on sample size and number of repeats is needed.

in vitro reconstitution assay:

The authors establish a new in vitro reconstitution assay for lysosomal function. While this assay is innovative, it should be better validated. To this end, phosphorylation of endogenous AMPK should be assayed in all conditions. The presence of the ragulator complex (and the GTPase status if possible) should be tested as well, as provides binding for AXIN and LKB1.

The authors should discuss the possibility that AMPK directly phosphorylates Raptor and can thus directly inhibit mTORC1 (Gwinn DM et al., 2008). In light of this, it would be interesting to test whether lysosomes isolated from AMPK knock out cells would alter mTORC1 (dis)association from the lysosome.

Furthermore, the authors need to quantify their immunoblots (and add the average of their three repeats as supplemental data), since especially the Myc–Raptor binding seems to be variable in their IP:FLAG experiments.

Panel D should also include the negative control (– aas, – ConA, – Met).

Effects of metformin independent of mitochondria.

The author's conclusion that metformin acts mainly through the lysosomal pathway and not via mitochondrial inhibition is not well supported. It has been previously shown by De Haes et al., 2014, that metformin, unlike rotenone, does not increase mitochondrial stress (i.e., data included in Figure 1—figure supplement 1); in addition, they showed that metformin can inhibit electron flow from complex I, and that metformin increases ROS production (in contrast to rotenone, which decreases ROS production). The authors should therefore refrain from concluding mitochondrial–independent functions of metformin.

Reviewer #3:

While this manuscript is potentially interesting, some of the conclusions seem to be based on small changes observed in single experiments, and analysis of more replicates with assessment of statistical significance would make the soundness of the conclusions more convincing.

1) Figure 1D/1E: while these results are potentially interesting, some of the observed effects are quite modest in extent, and I believe that it is essential in such cases to run at least 2 or 3 replicate incubations side–by–side so that the reproducibility of the effects can be assessed. Even better is to quantify the blots and perform statistical analysis.

2) Figure 1E: I was puzzled about the exact conditions in which the experiment shown in this Figure was performed. They claim to observe an increase in Thr172 phosphorylation, which would require ATP, yet it is not clear from the Methods section whether the "fractionation buffer" did contain ATP during the incubation with AXIN and LKB1. Also, did the purified Myc–LKB1 contain bound STRADalpha/β and/or MO25alpha/β, with the former in particular being essential for LKB1 activity?

3) Although this may not be common practice within the C. elegans community, it is possible to perform statistical analysis of survival curves such as those shown in Figures 2E, 3A, D and F. I would like to ask why this has not been done.

[Editors’ note: what now follows is the decision letter after the authors resubmitted for further consideration.]

Thank you for submitting your article "Metformin Extends Lifespan by Coordination of mTORC1 and AMPK through Lysosomal Pathway" for consideration by eLife. Your article has been reviewed by two peer reviewers, one of whom is a member of our Board of Reviewing Editors and the evaluation has been overseen by Jonathan Cooper as the Senior Editor. The reviewers have opted to remain anonymous.

The reviewers have discussed the reviews with one another and the Reviewing Editor has drafted this decision to help you prepare a revised submission.

Summary:

In this manuscript, the authors showed that metformin extends lifespan through TORC1 inhibition and lysosome–dependent activation of AMPK. They developed an in vitro reconstitution assay to show that metformin coordinates mTORC1 inhibition and AMPK activation through the lysosomal pathway. They further demonstrated that both TORC1 inhibition and Axin/LKB1–mediated AMPK activation contribute to the lifespan extension effect of metformin in C. elegans. This study provides insights into the molecular mechanism of metformin–mediated lifespan extension. Overall, the findings are potentially interesting. This manuscript is suitable for publication in eLife after appropriate revisions.

Essential revisions:

1) It is unclear whether metformin–activated AMPK also could have effects on HLH–30/TFEB. The authors should directly test whether Metformin treatment of aak–2 mutants leads to HLH–30 nuclear localization (and by extension, gene expression of HLH–30 target genes). This information would help the authors in their discussion of the involvement of AMPK vs mTOR.

2) The analysis of downstream pathways of TOR is insufficiently characterized and in places contradicting with previous work from Driscoll lab. The authors performed epistasis experiments with known downstream targets of mTOR, e.g., with pha–4, skn–1 and hlh–30 mutants and conclude that metformin can still induce lifespan–extending effects in these mutants. However, this clearly looks like a partial lifespan extension; in the authors' hands, metformin generally produces 40% lifespan extension in wild–type/WT animals (Table S8), but in mutants they only see 14% (hlh–30) and 20–30% (pha–4, smg–1) lifespan extension (Table S6). While these lifespan experiments do not appear to have been performed in direct comparison with a WT control, the more reasonable conclusion is that metformin has a partial effect on lifespan in these backgrounds, and therefore that the genes play a partial role in the lifespan extension. Indeed, Onken and Driscoll reported that skn–1 is required for metformin–induced longevity. The authors need to revisit their conclusions in light of these comments and previously published data. To this end, they are encouraged to emphasize the lysosome as the site of Metformin action, their most well–supported new finding, and de–focus on dissecting the individual contribution of AMPK and mTOR. A specific suggestion would to move Figure 4A–C to the supplements and Figure 4D–E merged into Figure 3.

3) While the authors have built VHA–3 and LMTR–3 reporters to confirm localization to the lysosome (were multiple independent strains analyzed and found to behave similarly?), the authors should use these strains for rescue experiments to seek possible functional validation as well.

4) The authors use a mammalian phosphorylation–specific antibody to detect phosphorylation of the worm S6K ortholog RSKS–1 as a read–out for TOR activity. The antibody needs to be validated.

https://doi.org/10.7554/eLife.31268.026

Author response

[Editors’ note: the author responses to the first round of peer review follow.]

The reviewers agreed that this is an interesting study that aims to show that metformin acts on TOR and AMPK at the lysosome, and that this site is critical for the effects of metformin on C. elegans lifespan. However, some of the conclusions seem to be based on small changes observed in single experiments, and analysis of more replicates with assessment of statistical significance would make the soundness of the conclusions more convincing. A substantial amount of work needs be conducted to address the concerns raised by the reviewers, and it is unlikely that this can be done within the two–month limit set by eLife, so we are rejecting the paper. Nevertheless, if your results hold up, I do encourage you to resubmit your work after you have obtained sufficient lines of validation.

We thank the reviewers for their positive and encouraging remarks on our work. We are particularly grateful for their constructive comments that help us to enhance the clarity of the manuscript.

In the revised manuscript, we added substantial new results from six sets of experiments that address nearly every critique that the reviewers raised, as detailed below.

1) To follow the reviewers’ advice, we added several sentences in the revised manuscript to discuss that metformin treatment further extends lifespan of daf–15 heterozygous mutants could be due to a partial reduction in mTOR activity, or due to a pathway additive to mTOR inhibition. We performed lifespan analysis with mutants of known TOR downstream targets upon metformin treatment, and also tested HLH–30 nuclear localization of daf–15 heterozygous mutants in the presence or absence of metformin treatment. These new results suggest that metformin might act on a pathway additive to TORC1 inhibition to promote longevity.

2) In response to the suggestion that quantifications and information on experimental repeats and sample numbers should be provided, we have now provided quantifications for all immunoblots and C. elegans images in the revised manuscript. Information on N (population size) and n (number of repeats) are provided in figure legends. Information and statistical analysis of all lifespan experiments are provided in Supplementary files 3–8.

3) To follow the reviewer’s advice, we have tested lysosomal localization and function of all lysosomal mutants used in this manuscript, and performed additional lifespan analysis of par–4, the LKB1 homolog. A strain list and the verification of new mutant alleles are also provided in Supplementary files 1–2.

4) In response to the reviewer’s criticism that we are using several readouts to assay TOR activity in C. elegans, we make it consistent and use S6K phosphorylation, HLH–30 nuclear localization, and mRNA levels of HLH–30 regulated genes to assay TOR activity in the revised manuscript.

5) We performed the analysis of fitness or healthspan parameters with daf–15 heteroygous mutants.

6) We performed in vitro reconstitution assay in AMPK knockout cells to show that mTOR disassociation from lysosome does not require AMPK.

Reviewer #1:

In this manuscript, the authors used an in vitro system to show that metformin coordinates mTORC1 inhibition and AMPK activation through the lysosomal pathway. They further demonstrated that both TORC1 inhibition and Axin/LKB1–mediated AMPK activation contribute to the lifespan extension effect of metformin in C. elegans. This study provides molecular mechanism of metformin–mediated lifespan extension.

Major concerns:

1) The authors showed that metformin treatment triggers nuclear localization of HLH–30::GFP in the tail region. How the cells in the tail affect the lifespan? Does loss of function of hlh–30 suppress the lifespan extension effect of metformin?

Metformin could still extend lifespan of hlh–30 mutants, suggesting that hlh–30 is only partially required for metformin–mediated lifespan extension (Figure 4A). Neurons and intestine are two types of tissues essential for promoting longevity in C. elegans. We think that it is possible that the nuclear localization of HLH–30 in the tail region of the intestine is sufficient to induce lifespan extension. It is also possible that HLH–30 expresses in the nucleus of other regions or other tissues. But the fluorescent level is a bit weak for us to detect at the current exposure time.

Loss of function of hlh–30 only partially suppresses the lifespan extension effect of metformin (Figure 4A). Metformin could still extend lifespan of hlh–30 mutants through AMPK pathway.

2) Metformin treatment further extends lifespan of daf–15 heterozygous mutants. This could be due to that metformin acts on a pathway additive to that of TORC1 inhibition. Alternatively, this may be simply due to a partial reduction in mTOR activity in daf–15 heterozygous mutants.

The reviewer is right. We added several sentences in the revised manuscript to discuss that metformin treatment further extends lifespan of daf–15 heterozygous mutants could be due to a partial reduction in mTOR activity, or due to a pathway additive to mTOR inhibition. We followed the advice of reviewer #2 to perform lifespan experiments with mutants of known downstream targets of TOR and showed that metformin could still extend lifespans of hlh–30, pha–4 or skn–1 mutants (Figure 4A–C). We also assayed HLH–30 nuclear localization with the daf–15 heterozygous mutants and showed that daf–15 heterozygous mutants have significant HLH–30 nuclear localization compared with control animals, and that metformin treatment does not further increase the level of HLH–30 nuclear localization in daf–15 heterozygous mutants (Figure 3F and G). These results suggest that metformin might act on a pathway additive to TORC1 inhibition to promote longevity.

3) Quantification results should be provided for Figure 1E, Figure 2A and Figure 3B, C.

Quantification results have now been provided in Figure 1I, Figure 2D and Figure 5C, H.

Reviewer #2:

Overall, this is an interesting study that aims to show that metformin acts on TOR and AMPK at the lysosome, and that this site is critical for the effects of metformin on C. elegans lifespan. While a novel in vitro reconstitution assay has been developed, and several new C. elegans lysosomal mutants have been analyzed, the study fails to make a convincing case, mainly due to insufficient lines of validation and major rigor issues.

Lysosomal mutants:

In addition to major rigor issues (see separate box), another major concern of this article is that the authors are using several new lysosomal mutants, i.e. lmtr–2/3 and axl–1, but these mutants are inadequately described and characterized. A description of the alleles, whether the mutations are loss– or gain–of function mutations, and the lysosomal function of these genes have to be established. Similar concerns relate to vha–3 and vha–12, which were analyzed by RNAi. Of particular interest is whether these genes indeed localize and function within the lysosome. Lysosomal localization assays and functional assays (cathepsin assays) need to be carried out to validate these mutants, considering the rationale of the manuscript overall. Metformin's effects on lysosomal function should also be tested.

We thank the reviewer for the constructive comments. Information of alleles used in this study is now provided in Supplementary file 1. We also provided the validation of vha–3/12 and lmtr–2/3 loss–of–function mutants in Supplementary file 2. In addition, we followed the reviewer’s advice to detect lysosomal localization and performed cathepsin assays to test lysosomal function of these mutants. These results are provided in Figure 5—figure supplement 1 and 2. Metformin’s effect on lysosomal function is also tested with cathepsin assay and provided in Figure 1—figure supplement 2C, D.

Of particular concern is Figure 3F, in which the authors describe that 80% of the lmtr–2/3 mutants are censored due to vulval rupture. This censoring will greatly affect the sample size and thus the interpretation of the lifespan experiments. Again, it is crucial for the authors to share these details of their lifespan analyses for all mutants tested.

Due to high percentage of censored worms, we enlarged sample size of lmtr–2/3 mutants while performing the lifespan experiments. The detailed information of lifespan analysis for all mutants tested is now provided in Supplementary file 3–8.

The authors should also include par–4, the LKB1 homolog in their analyses of the lysosomal mutants.

Similar to axl–1 (the AXIN homolog), metformin is not able to extend lifespan of par–4 mutants as well. This data is provided in Figure 5K.

TOR activity assays:

Throughout the manuscript, the authors are using several readouts to assay TOR activity in C. elegans: HLH–30 nuclear localization, DAF–16 nuclear localization, increase in hlh–30 transcriptional reporter, increase in gcs–1 transcriptional reporter, mRNA levels of skn–1, daf–16 and daf–15/Raptor, mRNA levels of TOR pathway components, skn–1–dependent target genes. The authors seem to randomly choose the assay in their different experimental conditions or genetic manipulations. While we acknowledge assay limitations for TOR activity in C. elegans, the authors should be consistent and use the same assays throughout. To this point, analysis of S6K phosphorylation would be a direct way to assay TOR pathway status

We thank the reviewer for the constructive comments. In the revised manuscript, we make it consistent and use S6K phosphorylation, HLH–30 nuclear localization, and mRNA levels of HLH–30 regulated genes to assay TOR activity. These data are provided in Figure 2 and Figure 5B–F.

Finally, analysis of the mRNA levels of HLH–30–regulated genes seems relevant to include, considering the use of the HLH–30 translocation assay.

Analysis of the mRNA levels of HLH–30–regulated genes has now been provided in Figure 2E and Figure 5D.

TOR inhibition:

The authors use a daf–15 heterozygous mutant to demonstrate that TOR inhibition and Metformin are additive in their effects on lifespan extension. Because the mutant is heterozygous (i.e., not null), this is not a sound conclusion, as Metformin could still exert its effects on the residual daf–15/TOR function. To strengthen their conclusion, the authors should rather perform epistasis experiments with known downstream targets of TOR, e.g., with pha–4, skn–1 and hlh–30 mutants. Additionally, the authors should assay the battery of TOR activity assay also with the daf–15 heterozygous mutants (especially HLH–30 nuclear localization).

The reviewer is right. We added several sentences in the revised manuscript to discuss that metformin treatment further extends lifespan of daf–15 heterozygous mutants could be due to a partial reduction in TOR activity, or due to a pathway additive to TOR inhibition. We followed the reviewer’s advice to perform lifespan experiments with mutants of known downstream targets of TOR and showed that metformin could still extend lifespans of hlh–30, pha–4 or skn–1 loss–of–function mutants (Figure 4A–C). We also assayed HLH–30 nuclear localization with the daf–15 heterozygous mutants and showed that daf–15 heterozygous mutants have significant HLH–30 nuclear localization compared with control animals, and that metformin treatment does not further increase the level of HLH–30 nuclear localization in daf–15 heterozygous mutants (Figure 3F and G). These results suggest that metformin might act on a pathway additive to TORC1 inhibition to promote longevity.

Fitness assays:

It would be important to include the daf–15 heteroygous mutants in the analysis of fitness or healthspan parameters as a positive control, since metformin should still have a positive and beneficial effect on these mutants. The oil–red–o stainings only inform on decreased fat levels, however not on a mechanism by which this is achieved. The authors need to refrain from making the conclusion that Metformin activates lipolysis, as there is no data supporting this claim. Again, more information on sample size and number of repeats is needed.

We have performed the analysis of fitness or healthspan parameters with daf–15 heterozygous mutants. Indeed, metformin still have a beneficial effect on these mutants. The results are provided in Figure 3B–E.

We edited the text according to the reviewer’s suggestion to only claim that oil–red–o staining indicates decreased fat levels.

Information on sample size and number of repeats is provided in each figure legend.

in vitro reconstitution assay:

The authors establish a new in vitro reconstitution assay for lysosomal function. While this assay is innovative, it should be better validated. To this end, phosphorylation of endogenous AMPK should be assayed in all conditions. The presence of the ragulator complex (and the GTPase status if possible) should be tested as well, as provides binding for AXIN and LKB1.

The authors should discuss the possibility that AMPK directly phosphorylates Raptor and can thus directly inhibit mTORC1 (Gwinn DM et al., 2008). In light of this, it would be interesting to test whether lysosomes isolated from AMPK knock out cells would alter mTORC1 (dis)association from the lysosome.

Furthermore, the authors need to quantify their immunoblots (and add the average of their three repeats as supplemental data), since especially the Myc–Raptor binding seems to be variable in their IP:FLAG experiments.

Panel D should also include the negative control (– aas, – ConA, – Met).

Phosphorylation of endogenous AMPK and the presence of the ragulator complex have now been assayed in all conditions in the reconstitution assay. Follow the reviewer’s advice, we discussed the possibility that AMPK directly phosphorylates Raptor and can thus directly inhibit mTORC1 in the revised manuscript. We also requested AMPKα1/2 double knockout cells and repeated the in vitro assays. Lysosomes isolated from AMPK knockout cells do not alter mTORC1 (dis)association from the lysosome (Figure 1—figure supplement 1). Therefore, it is unlikely that AMPK directly inhibit mTORC1.

Quantification of the immunoblots are provided in Figure 1F, G and I, and Figure 1—figure supplement 1F–H.

We included the negative control (– aas, – ConA, – Met) in panel D.

Effects of metformin independent of mitochondria.

The author's conclusion that metformin acts mainly through the lysosomal pathway and not via mitochondrial inhibition is not well supported. It has been previously shown by De Haes et al., 2014, that metformin, unlike rotenone, does not increase mitochondrial stress (i.e., data included in Figure 1—figure supplement 1); in addition, they showed that metformin can inhibit electron flow from complex I, and that metformin increases ROS production (in contrast to rotenone, which decreases ROS production). The authors should therefore refrain from concluding mitochondrial–independent functions of metformin.

We followed the reviewer’s suggestion and edited the text accordingly.

Reviewer #3:

While this manuscript is potentially interesting, some of the conclusions seem to be based on small changes observed in single experiments, and analysis of more replicates with assessment of statistical significance would make the soundness of the conclusions more convincing.

We thank the reviewers for the encouraging remarks on our work and constructive comments. The statistical issue has also been raised by other reviewers. In the revised manuscript, we have now provided quantifications for all immunoblots and C. elegans images. Information on N (population size) and n (number of repeats) are also provided in figure legends. In addition, information and statistical analysis of survival curves are provided in Supplementary files 3–8.

1) Figure 1D/1E: while these results are potentially interesting, some of the observed effects are quite modest in extent, and I believe that it is essential in such cases to run at least 2 or 3 replicate incubations side–by–side so that the reproducibility of the effects can be assessed. Even better is to quantify the blots and perform statistical analysis.

We thank the reviewer for the constructive comments. The in vitro reconstitution experiments have been repeated three times. Quantification and statistical analysis of the blots are now provided in the revised manuscript (Figure 1F–G, and 1I).

2) Figure 1E: I was puzzled about the exact conditions in which the experiment shown in this Figure was performed. They claim to observe an increase in Thr172 phosphorylation, which would require ATP, yet it is not clear from the Methods section whether the "fractionation buffer" did contain ATP during the incubation with AXIN and LKB1. Also, did the purified Myc–LKB1 contain bound STRADalpha/β and/or MO25alpha/β, with the former in particular being essential for LKB1 activity?

The reviewer is correct. The “fractionation buffer” contains 2.5mM ATP. We provided this information in the revised manuscript.

Purified Myc–LKB1 does form a complex with STRAD and MO25. This result is provided in Figure 1H.

3) Although this may not be common practice within the C. elegans community, it is possible to perform statistical analysis of survival curves such as those shown in Figures 2E, 3A, D and F. I would like to ask why this has not been done.

Statistical analysis of survival curves is now provided in Supplementary files 4, 6 and 8.

[Editors' note: further revisions were requested prior to acceptance, as described below.]

Essential revisions:

1) It is unclear whether metformin–activated AMPK also could have effects on HLH–30/TFEB. The authors should directly test whether Metformin treatment of aak–2 mutants leads to HLH–30 nuclear localization (and by extension, gene expression of HLH–30 target genes). This information would help the authors in their discussion of the involvement of AMPK vs mTOR.

We have showed that metformin treatment of aak–2 mutants was still able to drive HLH–30 nuclear localization and elevate the expression of HLH–30 target genes (Figure 2—figure supplement 2).

2) The analysis of downstream pathways of TOR is insufficiently characterized and in places contradicting with previous work from Driscoll lab. The authors performed epistasis experiments with known downstream targets of mTOR, e.g., with pha–4, skn–1 and hlh–30 mutants and conclude that metformin can still induce lifespan–extending effects in these mutants. However, this clearly looks like a partial lifespan extension; in the authors' hands, metformin generally produces 40% lifespan extension in wild–type/WT animals (Table S8), but in mutants they only see 14% (hlh–30) and 20–30% (pha–4, smg–1) lifespan extension (Table S6). While these lifespan experiments do not appear to have been performed in direct comparison with a WT control, the more reasonable conclusion is that metformin has a partial effect on lifespan in these backgrounds, and therefore that the genes play a partial role in the lifespan extension. Indeed, Onken and Driscoll reported that skn–1 is required for metformin–induced longevity. The authors need to revisit their conclusions in light of these comments and previously published data. To this end, they are encouraged to emphasize the lysosome as the site of Metformin action, their most well–supported new finding, and de–focus on dissecting the individual contribution of AMPK and mTOR. A specific suggestion would to move Figure 4A–C to the supplements and Figure 4D–E merged into Figure 3.

We think metformin promotes lifespan extension through two mechanisms: inhibition of TORC1 pathway and activation of AMPK. The partial lifespan extension effect of metformin in mutants of TOR downstream genes (e.g. hlh–30, pha–4 and skn–1) supported our conclusions: 1) Metformin only partially extends lifespan of these mutants indicates that these TORC1 downstream genes play a partial role in the lifespan extension. It supports our idea that metformin promotes longevity partially through TOCR1 inhibition. 2) Metformin can still extends lifespans of these mutants suggests that metformin also act on a pathway additive to that of TORC1 inhibition (e.g. AMPK pathway) to confer lifespan extension.

We thank the reviewer for the suggestion to let us emphasize the lysosome as the site of metformin action, and de–focus on dissecting the individual contribution of AMPK and mTOR. We have now changed the title of our manuscript to “Metformin Extends C. elegans Lifespan through Lysosomal Pathway”. We also follow the reviewer’s specific suggestion to move Figure 4A–C to Figure 3—figure supplement 2 and Figure 4D–E to Figure 3H–I.

3) While the authors have built VHA–3 and LMTR–3 reporters to confirm localization to the lysosome (were multiple independent strains analyzed and found to behave similarly?), the authors should use these strains for rescue experiments to seek possible functional validation as well.

Yes, the VHA–3 and LMTR–3 localization were analyzed with multiple independent transgenic strains. They all behave similarly.

We have followed the reviewer’s suggestion to rescue vha–3, vha–12, lmtr–2 and lmtr–3 mutants with the corresponding reporter construct and validate their function through cathepsin assays (Figure 4—figure supplement 2). The reporter construct can rescue the mutant phenotype.

4) The authors use a mammalian phosphorylation–specific antibody to detect phosphorylation of the worm S6K ortholog RSKS–1 as a read–out for TOR activity. The antibody needs to be validated.

We have now validated the antibody with rsks–1 RNAi (Figure 2—figure supplement 1B).

https://doi.org/10.7554/eLife.31268.027

Article and author information

Author details

  1. Jie Chen

    1. State Key Laboratory of Membrane Biology, Institute of Molecular Medicine, Peking-Tsinghua Center for Life Sciences, Peking University, Beijing, China
    2. Academy for Advanced Interdisciplinary Studies, Peking University, Beijing, China
    Contribution
    Conceptualization, Data curation, Formal analysis, Investigation, Visualization, Methodology, Writing—original draft, Project administration
    Contributed equally with
    Yuhui Ou
    Competing interests
    No competing interests declared

  2. Yuhui Ou

    State Key Laboratory of Membrane Biology, Institute of Molecular Medicine, Peking-Tsinghua Center for Life Sciences, Peking University, Beijing, China
    Contribution
    Data curation, Validation, Investigation
    Contributed equally with
    Jie Chen
    Competing interests
    No competing interests declared

  3. Yi Li

    1. State Key Laboratory of Membrane Biology, Institute of Molecular Medicine, Peking-Tsinghua Center for Life Sciences, Peking University, Beijing, China
    2. Academy for Advanced Interdisciplinary Studies, Peking University, Beijing, China
    Contribution
    Validation, Investigation
    Competing interests
    No competing interests declared

  4. Shumei Hu

    1. State Key Laboratory of Membrane Biology, Institute of Molecular Medicine, Peking-Tsinghua Center for Life Sciences, Peking University, Beijing, China
    2. Academy for Advanced Interdisciplinary Studies, Peking University, Beijing, China
    Contribution
    Investigation
    Competing interests
    No competing interests declared

  5. Li-Wa Shao

    1. State Key Laboratory of Membrane Biology, Institute of Molecular Medicine, Peking-Tsinghua Center for Life Sciences, Peking University, Beijing, China
    2. Academy for Advanced Interdisciplinary Studies, Peking University, Beijing, China
    Contribution
    Investigation
    Competing interests
    No competing interests declared

  6. Ying Liu

    State Key Laboratory of Membrane Biology, Institute of Molecular Medicine, Peking-Tsinghua Center for Life Sciences, Peking University, Beijing, China
    Contribution
    Conceptualization, Supervision, Funding acquisition, Visualization, Methodology, Writing—original draft, Writing—review and editing
    For correspondence
    ying.liu@pku.edu.cn
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0002-3328-026X

Funding

National Natural Science Foundation of China (31422033)

  • Ying Liu

Ministry of Science and Technology of the People's Republic of China (2013CB910104)

  • Ying Liu

National Natural Science Foundation of China (31471381)

  • Ying Liu

Young Thousand Talents Plan of China

  • Ying Liu

Peking-Tsinghua Center for Life Sciences

  • Ying Liu

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Acknowledgements

We thank Dr. Sheng-Cai Lin for his generous gifts of plasmids and reagents. We thank Dr. Benoit Viollet for providing us AMPK knockout cells. Several C. elegans strains used in this work were provided by CGC, which is supported by the NIH-Officer of Research Infrastructure Programs, and the Japanese National BioResource Project. The work is supported by grants from the National Natural Science Foundation of China (grants 31422033 and 31471381) and the Ministry of Science and Technology of China (973 grants 2013CB910104), the Young Thousand Talents Program of China, and Peking-Tsinghua Center for Life Sciences awarded to YL.

Reviewing Editor

  1. Hong Zhang, Institute of Biophysics, Chinese Academy of Sciences, China

Version history

  1. Received: August 15, 2017
  2. Accepted: October 11, 2017
  3. Accepted Manuscript published: October 13, 2017 (version 1)
  4. Version of Record published: November 14, 2017 (version 2)

Copyright

© 2017, Chen et al.

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.

Metrics

  • 7,502
    Page views
  • 1,484
    Downloads
  • 112
    Citations

Article citation count generated by polling the highest count across the following sources: Scopus, Crossref, PubMed Central.

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Open citations (links to open the citations from this article in various online reference manager services)

Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)

  1. Jie Chen
  2. Yuhui Ou
  3. Yi Li
  4. Shumei Hu
  5. Li-Wa Shao
  6. Ying Liu
(2017)
Metformin extends C. elegans lifespan through lysosomal pathway
eLife 6:e31268.
https://doi.org/10.7554/eLife.31268

Categories and tags

Research organism

Further reading

    1. Cell Biology
    2. Microbiology and Infectious Disease

    Synchronization of oscillatory growth prepares fungal hyphae for fusion

    Valentin Wernet, Marius Kriegler ... Reinhard Fischer
    Research Article Updated

    Communication is crucial for organismic interactions, from bacteria, to fungi, to humans. Humans may use the visual sense to monitor the environment before starting acoustic interactions. In comparison, fungi, lacking a visual system, rely on a cell-to-cell dialogue based on secreted signaling molecules to coordinate cell fusion and establish hyphal networks. Within this dialogue, hyphae alternate between sending and receiving signals. This pattern can be visualized via the putative signaling protein Soft (SofT), and the mitogen-activated protein kinase MAK-2 (MakB) which are recruited in an alternating oscillatory manner to the respective cytoplasmic membrane or nuclei of interacting hyphae. Here, we show that signal oscillations already occur in single hyphae of Arthrobotrys flagrans in the absence of potential fusion partners (cell monologue). They were in the same phase as growth oscillations. In contrast to the anti-phasic oscillations observed during the cell dialogue, SofT and MakB displayed synchronized oscillations in phase during the monologue. Once two fusion partners came into each other’s vicinity, their oscillation frequencies slowed down (entrainment phase) and transit into anti-phasic synchronization of the two cells’ oscillations with frequencies of 104±28 s and 117±19 s, respectively. Single-cell oscillations, transient entrainment, and anti-phasic oscillations were reproduced by a mathematical model where nearby hyphae can absorb and secrete a limited molecular signaling component into a shared extracellular space. We show that intracellular Ca2+ concentrations oscillate in two approaching hyphae, and depletion of Ca2+ from the medium affected vesicle-driven extension of the hyphal tip, abolished the cell monologue and the anti-phasic synchronization of two hyphae. Our results suggest that single hyphae engage in a ‘monologue’ that may be used for exploration of the environment and can dynamically shift their extracellular signaling systems into a ‘dialogue’ to initiate hyphal fusion.

    1. Cell Biology
    2. Microbiology and Infectious Disease

    A CRISPR-based rapid DNA repositioning strategy and the early intranuclear life of HSV-1

    Juan Xiang, Chaoyang Fan ... Pei Xu
    Research Article Updated

    The relative positions of viral DNA genomes to the host intranuclear environment play critical roles in determining virus fate. Recent advances in the application of chromosome conformation capture-based sequencing analysis (3 C technologies) have revealed valuable aspects of the spatiotemporal interplay of viral genomes with host chromosomes. However, to elucidate the causal relationship between the subnuclear localization of viral genomes and the pathogenic outcome of an infection, manipulative tools are needed. Rapid repositioning of viral DNAs to specific subnuclear compartments amid infection is a powerful approach to synchronize and interrogate this dynamically changing process in space and time. Herein, we report an inducible CRISPR-based two-component platform that relocates extrachromosomal DNA pieces (5 kb to 170 kb) to the nuclear periphery in minutes (CRISPR-nuPin). Based on this strategy, investigations of herpes simplex virus 1 (HSV-1), a prototypical member of the human herpesvirus family, revealed unprecedently reported insights into the early intranuclear life of the pathogen: (I) Viral genomes tethered to the nuclear periphery upon entry, compared with those freely infecting the nucleus, were wrapped around histones with increased suppressive modifications and subjected to stronger transcriptional silencing and prominent growth inhibition. (II) Relocating HSV-1 genomes at 1 hr post infection significantly promoted the transcription of viral genes, termed an ‘Escaping’ effect. (III) Early accumulation of ICP0 was a sufficient but not necessary condition for ‘Escaping’. (IV) Subnuclear localization was only critical during early infection. Importantly, the CRISPR-nuPin tactic, in principle, is applicable to many other DNA viruses.

    1. Cell Biology

    Mitochondrial defects caused by PARL deficiency lead to arrested spermatogenesis and ferroptosis

    Enrico Radaelli, Charles-Antoine Assenmacher ... Marco Spinazzi
    Research Article Updated

    Impaired spermatogenesis and male infertility are common manifestations associated with mitochondrial diseases, yet the underlying mechanisms linking these conditions remain elusive. In this study, we demonstrate that mice deficient for the mitochondrial intra-membrane rhomboid protease PARL, a recently reported model of the mitochondrial encephalopathy Leigh syndrome, develop early testicular atrophy caused by a complete arrest of spermatogenesis during meiotic prophase I, followed by degeneration and death of arrested spermatocytes. This process is independent of neurodegeneration. Interestingly, genetic modifications of PINK1, PGAM5, and TTC19 – three major substrates of PARL with important roles in mitochondrial homeostasis – fail to reproduce or modify this severe phenotype, indicating that the spermatogenic arrest arises from distinct molecular pathways. We further observed severe abnormalities in mitochondrial ultrastructure in PARL-deficient spermatocytes, along with prominent electron transfer chain defects, disrupted coenzyme Q (CoQ) biosynthesis, and metabolic rewiring. These mitochondrial defects are associated with a germ cell-specific decrease in GPX4 expression leading arrested spermatocytes to ferroptosis – a regulated cell death modality characterized by uncontrolled lipid peroxidation. Our results suggest that mitochondrial defects induced by PARL depletion act as an initiating trigger for ferroptosis in primary spermatocytes through simultaneous effects on GPX4 and CoQ – two major inhibitors of ferroptosis. These findings shed new light on the potential role of ferroptosis in the pathogenesis of mitochondrial diseases and male infertility warranting further investigation.