Figures and data
Overexpression of Ssd1 extends lifespan and forms transient cytoplasmic foci during replicative aging.
A. Levels of GFP tagged Ssd1 in the indicated strains, for 10000 cells per strain. B RLS of strains shown in panel A for 90 ssd1Δ cells, 63 SSD1-GFP cells and 63 PGPD1-SSD1-GFP cells, p- values are as follows: for control strain versus ssd1Δ strain is 0, for control versus PGPD1-SSD1-GFP is 8 x 10-8 C. Transient formation of Ssd1-GFP foci in PGPD1-SSD1-GFP strain during aging at the indicated cell division number of the RLS for one cell. Arrows indicate Ssd1-GFP foci. D. Single cell trajectories showing total cell Ssd1-GFP intensity and number of Ssd1-GFP foci formation for the indicated strains throughout the RLS. Each rectangle represents one cell cycle.
Ssd1 foci formed during aging are cell cycle regulated and distinct from Hsp104 foci, P-bodies and stress granules.
A. Example showing Ssd1-GFP foci in G2 phase during aging. The images are taken of the same cell every 20 minutes and the cartoons indicate the bud size. Below are shown single cell trajectories of cell cycle stage at which Ssd1 foci are present during aging (left) and whether the foci dissolved at the following cytokinesis (shown in blue) or were persistent through cytokinesis (shown in red) (right). B. Colocalization of Ssd1 foci with P-bodies and Hsp104 foci in cells treated with 10 mM Sodium azide. Quantitation is shown below for over 100 cells per condition. C. Colocalization of Ssd1 foci with P-bodies (marked by Edc3) and stress granules (marked by Pab1) under nutrient deprivation and quantitation. D. Failure of Ssd1 foci to colocalize with age-induced Hsp104 foci or P-bodies.
CR and overexpression of Ssd1 are epistatic for lifespan extension, both influencing iron metabolism
A. Calorie restriction and overexpression of Ssd1 are epistatic for extension of RLS, for 54 control cells, 48 PGPD-SSD1 cells, 48 control cells undergoing CR, and 44 PGPD-SSD1 cells + CR. p-values determined by student’s T-test for control versus PGPD-SSD1, CR and PGPD-SSD1 + CR are 1 x 106, 3 x 10-4 and 5 x 10“* respectively. There is no significant difference between PGPD-SSD1, CR and PGPD-SSD1 + CR. B. CR and overexpression of Ssd1 protect cells from lifespan reduction due to iron supplementation with 10 µM Fe+3, for 55 control cells, 53 control cells + Fe+3, 47 PGPD-SSD1 cells, 41 PGPD-SSD1 + Fe+3 cells, 37 CR cells, and 50 CR + Fe+3 cells. p-value for control versus Fe+3 is 6 × 10-7. There is no significant difference between PGPD-SSD1 with and without Fe+3 or for CR with and without Fe’3. C. CR and Ssd1 overexpression protect cells from the lifespan reduction caused by iron depletion with 200 µM BPS, for 52 control cells, 40 control cells + BPS, 60 PGPD-SSD1 cells, 34 PGPD-SSD1 cells + BPS, 44 control cells under CR, and 53 cells under CR + BPS. p-values for the control + BPS, versus PGPD-SSD1 + BPS and CR +BPS are 0, while there is no significant difference between the control with and without BPS treated and untreated control are not significantly different, control compared to PGPD-SSD1 and CR are 0.
Age-dependent induction of the iron regulon predicts the end of lifespan, and is blocked by Ssd1 overexpression and CR.
A. Schematic of iron regulon activation. B Single cell trajectories of nuclear localization of GFP-Aft1 during the RLS of control or CR yeast. C. Earlier nuclear localization of Aft1 in the lifespan correlates with a shorter total lifespan. Analysis of the data shown in panel B. D. Single cell trajectories of FIT2-mCherry expression during aging and Peak log fold change (FC) in mcherry intensity throughout the lifespan. E. Single cell trajectories of ARN1-mCherry expression during aging and fold change intensity throughout the lifespan. Statistical difference is indicated where n.s. is not significant change with p > 0.05, * p < 0.05, ** p ≤ 0.01, *** p ≤ 0.001 and **** p ≤ 0.0001.
CR and Ssd1 overexpression extend lifespan by preventing the accumulation of iron resulting from age-dependent activation of the iron regulon.
A. Pre-induction of iron regulon by BPS treatment measured by mCherry-tagged expression of Fit2 and Arn1, for 10000 cells per condition. B. RLS of cells that had the iron regulon induced or uninduced from panel A measured by Fit2-mcherry expression, for 63 (a), 41 (b), 63 (c), 50 (d), 63 (e), and 31 (f) cells, p-values of control vs induced: 0.002, PGPOI-SSD1 vs induced: 0 and CR vs induced: 0. C. RLS of yeast overexpressing Ssd1 as well as under CR are protected from sensitivity to Artemisinin. Left plot: for 62 control, 64 20 µM Art, 37 50 µM Art, 86 PGPDI-SSD1, 80 PGPDI-SSD1 + 20 µM Art and 107 PGPDI-SSD1 + 50 µM Art cells, p-values are control vs 20 µM Art: 4 x10-6, control vs 50 µM Art: 0, PGPOI-SSD1 vs + PGPDI-SSD1 20 µM Art: 0.98 and PGPOI-SSDI vs + PGPDI-SSD1 + 50 µM Art: 0.01. Right plot: 38 control, 34 20 µM Art, 30 50 µM Art, 61 CR, 61 CR + 20 µM Art, and 50 CR + 50 µM Art cells, p-values are control vs 20 µM Art: 0.01, control vs 50 µM Art: 1 x 10-7, CR vs CR 20 µM Art: 0.95 and CR vs + CR + 50 µM Art: 0.005. D. Relative levels of Fe2+ at indicated cell division of the RLS measured by Phen Green. E. Relative amount of Fe2+ in all cells of the indicated strains from panel D during the RLS, regardless of age. F. RLS of the indicated strains / conditions with and without aft1Δ, for 39 control, 71 aft1Δ, 37 PGPDI-SSD1, 72 PGPDI-SSD1/aft1Δ, 42 CR, and 74 CR/aft1Δ cells, p-values of control compared to all other strains are ≤ 1 x 10-4 while there is no significant difference between the strains / conditions other than the control.
Model for molecular mechanism of RLS extension by Ssd1 overexpression and CR.
During normal aging, cells have limited intracellular iron, potentially due to mitochondrial dysfunction reducing the assembly of iron-sulfur clusters, which leads to nuclear translocation of Aft1 and activation of the iron regulon. This subsequent deleterious increase in intracellular iron limits lifespan. CR also extends lifespan in a manner dependent on failure to activate the iron regulon; mechanistically CR indirectly prevents nuclear localization of Aft1, preventing production of the mRNAs from the iron regulon and the subsequent age-dependent increase in deleterious iron accumulation. In aging cells overexpressing Ssd1, transient cytoplasmic Ssd1 condensates appear that correlate with lifespan extension, where the lifespan extension is due to failure to activate the iron regulon preventing deleterious iron accumulation. We speculate that Ssd1 dephosphorylation by Sit4 after G phase likely leads to formation of Ssd1 condensates in aging cells overexpressing Ssd1 including mRNAs of the iron regulon genes FIT2 and ARN1, while dissolution of Ssd1 condensates occurs prior to cytokinesis upon Cbk1 -mediated phosphorylation of Ssd1.
Single cell replicative aging of one cell, where images were taken for every cell cycle in its lifespan in the bright field and GFP channel for SSD1-GFP and in the bright field for PGPD1- SSD1-GFP and corresponds to image taken in the GFP channel shown in Fig-1C.
Relative SSD1-GFP protein amount per cell throughout the lifespan, based on total cellular fluorescence intensity of strains SSD1-GFP and PGPDI-SSD1-GFP during the cell division in which 100% of the population (mothers) are actively dividing for the data shown in Fig. 1D.
Average and standard deviation for 63 cells per strain is shown. The box indicates interquartile range (IQR), whiskers extend to the smallest and largest non-outlier values and outliers are plotted as individual points outside Quartile(Q)1-1.5×IQR or Q3+1.5×IQR, where IQR = Q3-Q1. Data are shown for cell divisions in which 100% of the mothers are still dividing.
A. Ssd1-GFP foci formed during aging are cytoplasmic. mCherry-NLS indicates the nucleus. B. 1-6hexanediol dissolves Ssd1-GFP foci, images taken before, or at 10 or 20 mins after, addition of 1-6hexanediol.
Frequency of Ssd1 foci formation in exponentially growing cells, treated with sodium azide and under glucose starvation in strain PGPD1-SSD1-GFP.
Average and standard deviation are shown for analysis of biological triplicate experiments for >60 cells per condition per experiment, p-values determined by Student’s t-test are indicated by ** for 0.01 and ***** for 0.0001.
A. Example of old cell with (top) and without (bottom) nuclear localization of GFP-Aft1 on the left, profile after threshold (see materials and methods) on the right. B. Localization of Aft1 to the nucleus occurs on average 5 divisions before the end of lifespan. The analysis is of the data shown in Fig. 4B. C. CR does not affect GFP-Aft1 expression as determined by fluorescence intensity. 50 cells were analyzed per conditions in each plot. n.s. indicates a not significant change of p > 0.05 and **** is p ≤ 0.0001.
Replicative lifespan for experiments shown in Figure 4D and 4E.
p-values for both graphs of control versus PGPD-SSD1 and CR is 0, while for PGPD-SSD1 versus CR is not significant, 50 cells were analyzed for each condition.
Single cell examples of experiments shown in Figure 4D and 4E, to show variability between different cells.
The number indicates the cell division number in the RLS. Top is imaged under bright light and the bottom under mCherry fluorescence detection.
RLS of the indicated strains / conditions with and without cth2Δ, the number of cells counted per condition are control: 91, cth2Δ: 150, PGPD1-SSD1: 39, PGPD1- SSD1/cth2Δ: 91, CR: 101, CR/cth2Δ: 98, p-values: control compared to cth2Δ is 6 x 10-3, cth2Δ compared to CR, PGPD1- SSD1/cth2Δ, cth2Δ under CR and PGPD1-SSD1 are 5 x 10-6, 1×10-5, 3 x 10-8 and 2 x 10-4, respectively.
