Spermidine inhibits neurodegeneration and delays aging via the PINK1-PDR1-dependent mitophagy pathway in C. elegans

Spermidine prolongs lifespan and improves healthspan in the WS C. elegans

First, in order to evaluate potential of spermidine as a treatment for premature aging, we evaluated the impact on a nematode model of WS namely, wrn-1 (gk99). As expected, the wrn-1 (gk99)C. elegans exhibited significantly shorter lifespan (median = 11 days) compared to N₂ wild type (WT) controls (median = 19 days) as previously reported [27]. Spermidine treatment from eggs resulted in an extension of median lifespan, by approximately 36% at the maximal effective dose of 5 mM, in the wrn-1 (gk99) worms in a dose-dependent manner compared to vehicle controls. However, doses exceeding 5 mM were no longer beneficial in lifespan extension (Figure 1A). Table 1 summarizes the effects of different concentrations of spermidine on the median survival time of wrn-1(gk99).

Subsequently, we evaluated the deposition of intestinal lipofuscin, a characteristic feature that accumulates and increases in an age-dependent manner, which reflect health and the rate of aging in worms [31]. Here we demonstrate, that spermidine (5 mM) prevented the accumulation of intestinal lipofuscin that becomes apparent at the age of adult day 7 in wrn-1 (gk99) worms (Figure 1B).

Following observation of spermidine induced increase in lifespan accompanied by reduced age-related accumulation of lipofuscin, in the wrn-1(gk99) worms; we evaluated the impact on the locomotor capacity rate of N2 and wrn-1 (gk99) C. elegans hermaphrodites moving on an agar surface of a petri plate by counting the number of bends in the anterior body region during a 30 s interval [32]. Supplementation with 5 mM spermidine resulted in increase of locomotor capacity in the wrn-1 (gk99) C. elegans at the age of adult day3 and day 5 (Figure 1C).

Spermidine has been shown to stimulate macroautophagy/autophagy [32]. Thus, in order to understand the molecular mechanism underlying the lifespan prolonging and improved healthspan induced by spermidine we evaluated changes in the autophagy machinery. We evaluated the expression levels of autophagy-related genes bec-1 (homologous to Beclin-1), sqst-1 (homologous to SQSTM) and lgg-1 (homologous to LC3). Relative to the vehicle controls, the expression level of bec-1 was significantly increased (P<0.01), meanwhile, the expression of sqst-1 was significantly reduced (P <0.05) following 5 mM spermidine-treatment (Figure 1D). The expression of lgg-1 was not significantly influenced by spermidine-treatment (Supplementary Figure 1). Gene expression indicated that spermidine induced autophagy may underlie the life-prolonging and improved health in the premature WS nematode model, wrn-1 (gk99).

Spermidine protects against behavioral deficits and pathological features of PD in C. elegans model

Next, we evaluated the potential of spermidine treatment in a model of neurodegenerative disease, in particular PD. We utilized a well-studied nematode model of PD, the NL5901 strain, in which human α- synuclein fused to YFP is under control of the muscular unc-54 promoter, transgene pkIs2386 [p_unc-54::αSYN::YFP] [33]. Muscle expression has been used successfully to model protein-misfolding diseases and to identify modifier genes without considering neuronal effects [33, 34]. The NL5901 strain exhibited significantly shorter lifespan (median = 13 days) to that of the N2 controls (median = 19 days). In order to evaluate the effect of spermidine, we supplemented the NL5901 and N2 with various doses spermidine from eggs. Our data shows that spermidine extends median lifespan of NL5901 model of PD, by approximately 15%, compared to vehicle controls from a dose as low as 1mM (Figure 2A). Table 2 summarizes the effects of spermidine on the median survival time of NL5901.

In NL5901 worms, since the fluorescence intensity of α-synuclein is greater than lipofuscin fluorescence, we detected changes of α-synuclein expression. Here we demonstrate, that spermidine (5 mM) prevented the accumulation of α-synuclein that becomes apparent at the age of adult day 7 in NL5901 worms (Figure 2B).

Following observation of spermidine induced increase in lifespan accompanied by reduced accumulation of α-synuclein, in the NL5901 worms; we evaluated the impact on healthspan. First, we quantified the locomotor capacity rate of N2 and NL5901 C. elegans. Supplementation with 5 mM spermidine resulted in increase of locomotor capacity in the NL5901 C. elegans at the age of adult day 3, day5 and day7 (Figure 2C).

Then, we examined the effect of spermidine supplementation on the cognitive ability of the NL5901 using a chemotaxis-based memory assay. This assay is designed to evaluate the ability of nematodes to learn the association between sodium chloride (NaCl) and food. Under starvation conditions, the nematodes will associate the high NaCl concentration with the starvation signal. After 4h starvation conditioning, in order to escape the A area (this area contains 100Mm NaCl) under starvation, the nematodes would display preference for the B area (this area contains 20mM NaCl). The chemotaxis index (CI) of the vehicle control group was positive, while the CI of the 5 mM spermidine-treated group was negative (Figure 2D). There is a statistical difference between the two groups (P<0.05), indicating that NL5901 nematodes are more inclined to move toward the B area with low NaCl concentration after 5mM spermidine treatment, that is, it can better link the starvation signal and the NaCl concentration signal and move more to the B region in order to avoid the high NaCl concentration under starvation.

Similarly, we evaluated the expression levels of autophagy-related genes. Relative to the vehicle controls, the expression levels of bec-1 was significantly increased (P<0.01), meanwhile, the expression of sqst-1 was reduced (P<0.05) following 5 mM spermidine-treatment (Figure 2E). The results showed that spermidine induced autophagy may underlie the life-prolonging and improved health in the PD nematode model, NL5901.

Spermidine prolongs lifespan and protects against memory loss in the AD C. elegans

A nematode model of AD namely, UM0001, in which can co-express human Aβ and tau proteins was used to evaluate the impact of spermidine. As expected, the UM0001 C. elegans exhibited significantly shorter lifespan (median = 8 days) compared to N2 wild type (WT) controls (median = 13 days) at 25°C. Spermidine treatment from eggs resulted in an extension of median lifespan, by approximately 37.5% at the maximal effective dose of 5 mM, in the UM0001 worms in a dose-dependent manner compared to vehicle controls. However, doses exceeding 5 mM were no longer beneficial in lifespan extension (Figure 3A). Table 3 summarizes the effects of different concentrations of spermidine on the median survival time of UM0001.

In addition, we evaluated the deposition of intestinal lipofuscin. Here we demonstrate, that spermidine (5 mM) prevented the accumulation of intestinal lipofuscin that becomes apparent at the age of adult day 7 in UM0001 worms (Figure 2B).

We quantified the locomotor capacity rate of N2 and UM0001 C. elegans. Supplementation with 5 mM spermidine resulted in increase of locomotor capacity in the UM0001 C. elegans at the age of adult day5 (Figure 3C).

For chemotaxis memory ability, the chemotaxis index (CI) of the vehicle control group was positive, while the CI of the 5 mM spermidine-treated group was negative (Figure 3D). There is a statistical difference between the two groups (P<0.01), indicating that UM0001 nematodes are more inclined to move toward the area with low NaCl concentration after 5mM spermidine treatment, that is, it can better link the starvation signal and the NaCl concentration signal.

Again, with respect to the expression levels of autophagy-related genes, relative to the vehicle controls, the expression levels of bec-1 was significantly increased (P<0.01), meanwhile, the expression of sqst-1 was reduced (P<0.05) following 5 mM spermidine-treatment (Figure 3E). Therefore, indicating that spermidine induced autophagy change may underlie the life-prolonging and improved health in the AD nematode model, UM0001.

Spermidine extends lifespan and improves healthspan via the PINK1-PDR1 pathway

As spermidine has been shown to promote life extension and improve healthspan in models of neurodegeneration and premature aging, we extended our study in order to understand the underlying mechanisms. In particular, due to the association of spermidine with macroautophagy/autophagy [8], which coincides with behavioral and pathological benefits highlighted in the present study; we pursued to explore further in this avenue. Coupled with macroautophagy, recent evidence for contribution of defective mitophagy, a sub-form of autophagy for selective removal of damaged/superfluous mitochondria, to development and progression of AD led this study to shed light on the effect of spermidine on the PINK1-PDR1 axis [21]. For this purpose, we selectively knocked-down the expression of pink1 and pdr-1 using RNAi in the models of premature aging (wrn-1 (gk99)), Supplementary Figures 2, 3 and neurodegenerative disease (NL5901 and UM0001), and evaluated the effect of spermidine.

Our results showed that the promoted life extension effects of spermidine disappeared after knocking down the expression of pink1 and pdr-1 in WS, PD, AD model nematodes (Figure 4A–4C). Then, we evaluated the deposition of intestinal lipofuscin of wrn-1 (gk99) and UM0001, as well as the expression level of α- synuclein of NL5901. Here, we demonstrated that spermidine no longer prevented the accumulation of intestinal lipofuscin in the wrn-1 (gk99) and UM0001 nematodes (Figure 4A, 4C) and the expression level of the α- synuclein in NL5901 nematode (Figure 4B). Moreover, we found that the improved locomotor capacity of spermidine treatment disappeared in the three nematodes (Figure 4A–4C).

C. elegans strains and genetics

Standard C. elegans strain maintenance procedures were followed [37]. The following strains were used in this study. N₂: wild-type. The nematode model of WS wrn-1 (gk99): it has a deletion mutation of 196 bp, resulting in complete deletion of WRN-1 protein [30, 31]. NL5901 (Punc-54 :: human α-synuclein :: YFP + unc-119):The nematode model of PD. The NL5901 strains were generously provided by Wei Zou (Translational Medicine Research Institute of Zhejiang University). UM0001:The nematode model of AD, obtained by crossing two lines of Aβ1-42 (CL2355) and TAU (BR5270), which can co-express human Aβ and tau proteins [28].

Culture and spermidine treatment of C. elegans

The N₂, wrn-1 (gk99), NL5901 nematodes were kept on Nematode Growth Medium (NGM) plates at 20° C, the UM0001 was kept at 16° C to L4, then transferred to 25° C. Spermidine was purchased from Aladdin (Shanghai, China). The subsequent treatment of synchronized worm eggs with spermidine (stock, 200 mM in M9 buffer solution) and M9 buffer solution as control. Transferring the nematodes onto a new plate containing spermidine every two days in order to maintain spermidine concentration.

C. elegans lifespan assay

Lifespan analyses were performed on the synchronized eggs. The wrn-1(gk99) and NL5901 worms treated with various concentrations of spermidine extracts at 20°C. The survival and dead worms were counted every two day until all nematodes died by determining their movement and pharyngeal pumping. Approximately 100 worms were scored in each experiment.

C. elegans intestinal lipofuscin deposition analysis

The three kinds of nematodes were cultured and treated as described above. The intestinal lipofuscin deposition was measured as described previously with modifications [38]. The autofluorescence of intestinal lipofuscin was measured at day-5 and day-7 adulthood using an epifluorescence microscope (Leica, Wetzlar, Germany) with excitation/emission wavelength of 350 ± 25/460 ± 25 nm. The relative fluorescence intensity was quantified by Image-Pro Plus software (Media Cybernetics, MD, USA). Approximately 20 worms were examined in each experiment.

Analyses of α-synuclein aggregation

The α-synuclein expressing worms, NL5901, was cultured and treated as described above. The YFP intensity of α-synuclein in the control and treated worms at day-5 and day-7 adulthood were measured by a fluorescence microscope (BX53; Olympus Corp., Tokyo, Japan) and quantified using the Image-Pro Plus software. Approximately 20 worms were examined in each experiment.

C. elegans locomotion assay

Locomotion was measured on solid media in this study. Briefly, to measure body bending, day-3 and day-5 adulthood nematodes were transferred into 1mL M9 buffer, which was used for washing and maintaining the worms for a short time. Then worms were transferred to NGM plates without bacteria and adjusted for 1 min. Subsequently, body bends were counted for 30 s. A single body bend is defined as one sinusoidal movement of a worm. In the locomotion assay, only live worms were counted for the assays. Approximately 20 worms were examined in each experiment.

C. elegans memory assays

Chemotaxis to water-soluble compounds was performed at 20° C, on 9 cm agar plates as described previously [39–42]. The chemotaxis index was calculated by subtracting the number of animals found at the trap from the number of animals at the source of the chemical, divided by the total number of animals subjected to the assay [39]. The resulting values were expressed as percentiles. Three distinct populations of 100 adults (for each strain) were scored during the assay period. For all experiments, spermidine (5 mM) nematodes were treated from eggs. Both naive and conditioned animals were challenged for 1h with gradients of 100mM NaCl. For conditioning, animals were exposed to 20mM NaCl on agar plates devoid of bacterial food, for 4h, prior to assaying chemotaxis to the respective compound. Approximately 40 worms were examined in each experiment, three technical replicates were performed for all nematode experiments.

Quantitative RT-PCR

Quantitative RT-PCR was performed on nematodes which were cultured and treated as described above. Total RNA was isolated using TRIzol (Invitrogen, USA) following the manufacture’s protocol. Then, total RNA was converted into complementary DNA (cDNA) using PrimeScript^TM RT Reagent Kit (Takara, Japan) for RT–qPCR (Bio-Rad, CA, USA). cDNA was diluted to a ratio of 1:10 with TB Green® Premix Ex Taq™ (Takara, Japan) with ROX qRT-PCR (Bio-Rad, CA, USA), forward and reverse primers of specific genes. The qPCR primers for bec-1 were 5’- ATG GAT GCT CAA GTG GCG ACA C-3’ (forward primer, F) and 5’- TCC AGC TCC TTC TCC TCA TCG G-3’ (reverse primer, R), for sqst-1 were 5’-TCC CTC CAT CTC CAA CCA TCG G-3’(F) and 5’- GCT GCA CGG GTC TCC TCA TTG-3’®. for pink-1 were 5’- ACG CTT CCT GCC GAG AAT ATT TCC -3’(F) and 5’-CGA CCG TGG CGA GTT ACA AGG-3’®, for pdr-1 were 5’-CGG CGG TCT GCG AAG AAT GC-3’(F) and 5’- TCC TTC CCA TCA CAA ACA CAG CAC-3’®The actin (act-1) with primers 5'-CGC CAT CCT CCG TCT TGA CTT G-3' (F) and 5'- GCT CAG CGG TGG TGG TGA AAG-3' ®) was used as an internal control.

RNA interference knockdown

RNAi experiments were carried out by feeding worms with bacteria expressing RNA against the gene of interest (or control bacteria carrying the empty L4440 vector). RNAi clones (transformed bacteria) were grown at 37° C overnight and then seeded onto NGM plates containing ampicillin (100 μg/ml). Expression of RNA was induced by the addition of 1 mM IPTG before eggs were added to the bacterial lawns. RNAi clones were obtained from Ahringer’s RNAi library [40].

Statistical analysis

All values are expressed as mean ± SD. A two-tailed Student’s t-test was used to compare means between two groups. Statistical analysis was performed by using GraphPad Prism 8.0 (GraphPadPrism Software, La Jolla, CA).