V-ATPase Disassembly at the Yeast Lysosome-Like Vacuole Is a Phenotypic Driver of Lysosome Dysfunction in Replicative Aging

Aging, lysosomes, proton pumps, Saccharomyces cerevisiae, caloric restriction 20 21 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.23.604825doi: bioRxiv preprint 3

22 It is well-documented that lysosomal acidification is compromised with age across many 23 organisms (Nixon, 2020). Reduced lysosomal acidification has multiple downstream functional 24 consequences. Lysosomal hydrolases represent a major arm of the cellular proteostatic 25 machinery and operate optimally at the acidic pH, so cargo degradation is compromised at higher 26 pH (Vilchez, Saez, & Dillin, 2014). Lysosomes are the terminal compartment for multiple 27 autophagy pathways, so clearance of autophagic cargoes and recycling of nutrients, both critical 28 in aging cells, is slowed (Hansen, Rubinsztein, & Walker, 2018; Kaushik et al., 2021). Iron and 29 other heavy metals are sequestered and buffered in the acidic lysosomes; loss of sequestration 30 can induce both oxidative stress (Diab & Kane, 2013; Kurz, Terman, Gustafsson, & Brunk, 31 2008) and deficiency in mitochondrial iron-sulfur proteins (Chen et al., 2020). Reduced 32 lysosomal storage can also create toxic imbalances in amino acids such as cysteine that 33 contribute to loss of mitochondrial function (C. E. Hughes et al., 2020). Recent work has 34 highlighted the central role of the lysosome in nutritional signaling and many aspects of this 35 signaling are linked to acidification (Perera & Zoncu, 2016). It is clear that loss of lysosomal 36 acidification can impact many processes associated with age-related functional decline, but the 37 mechanisms behind increases in lysosomal pH are not fully understood. 38 The highly conserved vacuolar H+- ATPase (V-ATPase) acidifies the lumen of lysosomes 39 and lysosome-like vacuoles, as well as endosomes and the late Golgi apparatus, in all eukaryotes 40 (Collins & Forgac, 2020). V-ATPases are multi-subunit protein complexes that couple ATP 41 hydrolysis to proton pumping into organelle lumens. The V-ATPase consists of two 42 subcomplexes: a peripheral V1 subcomplex oriented toward the cytosol that is responsible for 43 ATP hydrolysis connected to a membrane-embedded V0 subcomplex containing the proton pore. 44 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.23.604825doi: bioRxiv preprint 4 V-ATPase subunit sequences are conserved across eukaryotes and recent V-ATPase structures 45 indicate very strong structural similarity between yeast and mammalian V-ATPases (Oot & 46 Wilkens, 2020). 47 V-ATPase activity is highly regulated and responsive to multiple environmental 48 conditions. Reversible disassembly is a versatile mechanism of V-ATPase regulation that fine-49 tunes the activity of the proton pump to meet cellular demands (Collins & Forgac, 2020; 50 Jaskolka, Winkley, & Kane, 2021; Wilkens, Khan, Knight, & Oot, 2023). In reversible 51 disassembly, the V1 subcomplex is released from the V0 subcomplex inhibiting both ATP 52 hydrolysis and proton pumping (Kane, 1995; Sumner et al., 1995). V1 subunit C is dissociated 53 from both subcomplexes and also becomes cytosolic during disassembly (Kane, 1995). 54 Disassembly is post-translational and rapidly reversible (Kane, 1995). It was first observed in the 55 yeast S. cerevisiae and the tobacco hornworm M. sexta upon acute glucose deprivation and was 56 reversed by glucose replenishment (Kane, 1995; Sumner et al., 1995). Since that time, it has 57 become clear that reversible disassembly occurs in many different settings and in response to 58 diverse signals. For example, unlike yeast cells, most mammalian cells appear to promote V-59 ATPase reassembly under conditions of nutrient deprivation and mTOR inhibition, possibly as a 60 means of promoting lysosomal proteolysis and nutrient recycling (Ratto et al., 2022; Stransky & 61 Forgac, 2015). In neurons, V-ATPases are reversibly disassembled as part of each synaptic 62 vesicle cycle (Bodzeta, Kahms, & Klingauf, 2017). In cardiomyocytes, lipid overload can 63 promote V-ATPase disassembly in endosomes, ultimately contributing to the long-term insulin 64 resistance (Liu et al., 2017). Reversible disassembly of V-ATPases can also be manipulated by 65 both host cells and pathogens to prevent or facilitate infection (Kohio & Adamson, 2013). 66 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.23.604825doi: bioRxiv preprint 5 Several mechanisms for loss of lysosomal acidification with age have been proposed. In 67 yeast, the long-lived plasma membrane proton pump, Pma1, accumulates in mother cells, and it 68 has been proposed that increased proton export through Pma1 disrupts the balance of Pma1 and 69 V-ATPase activities and compromises organelle acidification (Henderson, Hughes, & 70 Gottschling, 2014). Early loss of vacuolar acidification in yeast has been correlated with 71 defective mitochondrial morphology and function. Both the deacidification and mitochondrial 72 morphology phenotypes are suppressed by overexpression of the V-ATPase catalytic subunit, 73 VMA1, or an ER-localized assembly factor, VPH2, suggesting a possible deficiency in these 74 factors with age (A. L. Hughes & Gottschling, 2012). In C. elegans, stability of the VMA1 75 transcript is controlled by a microRNA, miR-1, which can globally control lysosomal 76 acidification (Schiffer et al., 2021). In several systems, mRNA expression for one or more V-77 ATPase subunit genes has been reported to decrease with age (Ghavidel et al., 2018), but in most 78 cases protein levels have not been assessed. All of these mechanisms could contribute to loss of 79 vacuolar/lysosomal acidification. Wilms et al. (Wilms et al., 2017) demonstrated that the mTOR 80 effector Sch9 promotes V-ATPase assembly and extends yeast chronological lifespan. 81 However, despite the importance of reversible disassembly in regulation of V-ATPase 82 activity, V-ATPase assembly state during aging has not been explored extensively. Here we show 83 that V-ATPase assembly does change with age in a yeast replicative aging model. Increased V-84 ATPase disassembly in older cells is accompanied by decreased vacuolar acidification that does 85 not appear to stem from reduced V-ATPase subunit levels. Instead, we provide evidence that 86 reduced activity of the RA VE (Regulator of H+-ATPase of Vacuolar and Endosomal membranes) 87 assembly complex may give rise to net V-ATPase disassembly and increased lysosomal pH with 88 age. 89 90 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.23.604825doi: bioRxiv preprint 6

91 V-ATPases are more disassembled and vacuoles more alkaline after ~5 cell divisions. 92 Given the evidence that vacuoles and lysosomes are less acidic in older cells, we 93 hypothesized that V-ATPase assembly and activity might also be changing with age. Replicative 94 aging in the yeast S. cerevisiae is a widely accepted model for aging (He, Zhou, & Kennedy, 95 2018). Briefly, yeast cells divide asymmetrically with each cell division giving rise to a new, 96 "rejuvenated" daughter cell from an established mother cell. Each cell division leaves a bud scar 97 on the mother, allowing visual assessment of age. The yeast V1C subunit (Vma5) is released 98 from both V1 and V0 during V-ATPase disassembly, so we first visualized Vma5-GFP 99 localization in a mixed age population of cells (Figure 1). (Note that vacuoles are visible as 100 indentations under DIC (differential interference contrast) optics on the left of the images.) In 101 parallel, we monitored replicative age of each cell by staining with calcofluor white, which labels 102 the bud scars on mother cells. As shown in Figure 1a, Vma5-GFP is tightly localized to the 103 vacuolar membrane in cells with few or no bud scars. In contrast, in older cells with more bud 104 scars, Vma5-GFP exhibited a marked decrease in fluorescence at the vacuolar membrane and a 105 notable increase in cytosolic fluorescence. In order to compare Vma5-GFP localization between 106 cells, we quantitated maximum fluorescence from a line scan across each cell. As shown in 107 Figure 1b, the line scan from a young cell has prominent peaks corresponding to the edges of the 108 vacuole with a high maximum fluorescence signal, while the older cells have less prominent 109 peaks. We conducted the same analysis across populations of cells of mixed age, normalized to 110 the maximum fluorescence signal of young cells and binned the results by the number of bud 111 scars. As shown at the left of Figure 1c, young cells, defined as having less than 5 bud scars, 112 displayed Vma5-GFP localization at the vacuolar membrane. However, the normalized 113 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.23.604825doi: bioRxiv preprint 7 maximum fluorescence signal, representative of vacuole localization, decreases significantly in 114 cells with five bud scars or more. This early onset of V-ATPase disassembly aligns with previous 115 reports indicating changes in lysosomal pH early in replicative aging (A. L. Hughes & 116 Gottschling, 2012). 117 Vma5 is a V1 subunit that bridges the V1 and V0 subcomplexes of the V-ATPase. The V1 118 subcomplex also contains three copies of Vma2, and the V0 subcomplex contains the largest 119 subunit, Vph1, which comprises part of the proton pore (Figure 2a). Although Vma5 appears to 120 be the subunit that is released most completely from the membrane by reversible disassembly 121 (Tabke et al., 2014), the rest of the V1 sector also dissociates from the vacuolar membrane. We 122 assessed the cellular distribution of Vma2-GFP, a V1 subunit, and Vph1-GFP, a V0 subunit 123 (Figure 2b). As shown in Figure 2b, there was less membrane-bound Vma2-GFP in older cells 124 relative to younger cells as assessed by line scans as in Figure 1. However, vacuolar Vph1-GFP 125 levels were the same between old and young cells (Figure 2c). These results are consistent with 126 disassembly of the V-ATPase as cells age. 127 Although comparable levels of Vph1 at the vacuole suggests that expression of V0 128 subunits and V0 assembly are intact, V1 subunits could become cytosolic because of reduced V1 129 subunit levels in older cells. To address this question, we isolated populations old and young 130 yeast cells by biotinylating the cell walls in a mixed age population, allowing growth to continue 131 for several generations, and then obtaining "old" cells by biotin-streptavidin affinity 132 chromatography (Jin, Cao, & Liu, 2021). Daughter cells that emerged after biotinylation cannot 133 bind to magnetic streptavidin beads and represent the "young" population. The age distribution 134 was determined by counting bud scars in each population and binning by the number of buds per 135 cell (Figure 2d). When prepared by this method, the population of old cells peaks at 20-24 bud 136 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.23.604825doi: bioRxiv preprint 8 scars, while there was a median value of 0-4 bud scars in the young population. Cell lysates were 137 prepared from each population and analyzed by SDS-PAGE and immunoblotting. As shown in 138 Figures 2e and f, there is no significant difference in the cellular levels of V1 subunits Vma1, 139 Vma2, and Vma5 between young and old populations. These results indicate that the cytosolic 140 populations of V1 subunits arise from disassembly of the V-ATPase, rather than inability to 141 assemble because of lack of V-ATPase subunits. 142 Reversible disassembly of V-ATPases is employed in a number of contexts to provide 143 dynamic regulation of the complex in response to changing cellular conditions. Disassembled V1 144 and V0 subcomplexes lack ATPase and proton transport activity and ATP-driven proton pumping 145 is restored upon reassembly. In order to test whether the lower levels of assembled V-ATPases in 146 old cells result in reduced capacity for acidification, we measured the response of young and old 147 cells to an acute glucose deprivation (an abrupt shift to 0% glucose), which promotes 148 disassembly of the yeast V-ATPase, followed by readdition of glucose to a 2% final 149 concentration, which promotes reassembly and reactivation of the complex (Kane, 1995). 150 Young and old yeast cells obtained as described above were loaded with the ratiometric 151 pH sensor BCECF-AM (2',7'-Bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein acetoxy methyl 152 ester) which localizes to the vacuole in yeast cells (Diakov, Tarsio, & Kane, 2013). Both young 153 and old cells were shifted to medium with no glucose for ~30 minutes. Fluorescence of cell 154 suspensions was then monitored continuously (Figure 3a), and glucose was added at the 155 indicated time. This assay revealed clear age-dependent differences in the vacuolar pH response 156 to glucose stimulation. Young cells exhibit a rapid drop in vacuolar pH upon glucose addition. 157 This drop was previously shown to be V-ATPase-dependent and to correlate with V-ATPase 158 assembly (Martinez-Munoz & Kane, 2008). In old cells, however, there was a smaller pH 159 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.23.604825doi: bioRxiv preprint 9 decrease after glucose addition, indicating a more alkaline vacuole and consistent with the lower 160 levels of V-ATPase assembly observed by microscopy during growth in glucose-replete 161 conditions in Figures 1 and 2. Quantitative analysis of pH at defined time points across 162 biological replicates (Figure 3b) indicates that vacuoles in old cells are significantly more 163 alkaline than their younger counterparts at each of the indicated time points. These results 164 suggest an age-related alteration in vacuolar pH regulation through the inability of V-ATPase to 165 reassemble. As a result, vacuoles in old cells display a more alkaline pH than vacuoles in young 166 cells. 167 Caloric restriction restores V-ATPase assembly and vacuolar acidification in older cells. 168 Caloric restriction (CR) is defined as a reduction in caloric intake in the presence of 169 adequate nutrition (Longo & Anderson, 2022). CR promotes both cellular health and longevity. 170 Extensive research, conducted in diverse model organisms ranging from yeast to worms, flies, 171 and rodents, suggests that CR can have significant anti-aging effects and promote overall health 172 (Longo & Anderson, 2022). In the context of the yeast S. cerevisiae, adjusting the concentration 173 of glucose in the growth medium from the 2% used above to 0.5% is a common way to induce 174 CR. This treatment does not significantly reduce growth rate over several cell divisions 175 (Supporting information, Figure 1). 176 As shown in Figure 4, CR reverses the V-ATPase disassembly in older cells. Notably, the 177 V1 subunits Vma5-GFP (Figure 4a) and Vma2-GFP (Figure 4b) are recruited to the vacuolar 178 membrane in both young and old cells, and there is no significant difference in fluorescence at 179 the vacuolar membrane with age. (Fluorescence intensities of Vph1-GFP continue to be similar 180 between old and young cells (Figure 4c).) This reversal suggests that CR extends V-ATPase 181 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.23.604825doi: bioRxiv preprint 10 assembly beyond the replicative age of >5 bud scars when cells grown in higher glucose begin to 182 show disassembly. 183 We hypothesized that given the improvement in V-ATPase assembly, vacuolar pH in old 184 cells might also be restored. We grew cells under CR conditions, loaded the vacuoles with 185 BCECF-AM and monitored vacuolar pH before and after addition of glucose as described above. 186 Figure 4d demonstrates that the glucose-stimulated decrease in vacuolar pH, which was 187 compromised in older cells grown in 2% glucose (Figure 3), was restored to the level of young 188 cells in cells after growth under CR conditions. This observation indicates that CR has a direct 189 impact on both V-ATPase assembly and vacuolar acidification in aging cells. In addition, it 190 further highlights the potential connection between V-ATPase assembly, vacuolar acidification 191 and aging. 192 193 Regulators of V-ATPase assembly state affect replicative lifespan. 194 If reversible disassembly of the V-ATPase plays a central role in aging of yeast cells, we 195 hypothesized that the cellular factors that regulate V-ATPase assembly might also affect 196 replicative lifespan. The RA VE complex plays a crucial role in regulation of the V-ATPase by 197 reversible disassembly (Jaskolka, Winkley, et al., 2021; Seol, Shevchenko, & Deshaies, 2001). It 198 consists of three subunits: Rav1, Rav2, and Skp1. The RA VE complex is required for reassembly 199 of V-ATPase complexes disassembled by glucose deprivation. In mutants lacking Rav1 or Rav2, 200 V-ATPases are predominantly disassembled into V1 and V0 subcomplexes and vacuolar 201 acidification is lost (Seol et al., 2001; Smardon, Tarsio, & Kane, 2002). The RA VE complex 202 associates with V1 subcomplexes in the cytosol. Although V1 subunit C is also released from the 203 vacuolar membrane, very little is associated with cytosolic RA VE-V1 (Jaskolka, Tarsio, 204 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.23.604825doi: bioRxiv preprint 11 Smardon, Khan, & Kane, 2021). In contrast, recent studies suggest that Oxr1, a protein originally 205 associated with resistance to oxidative stress, promotes disassembly of V-ATPases (Khan et al., 206 2022; Khan & Wilkens, 2024; Klossel et al., 2024) suggesting that RA VE and Oxr1 have 207 opposing effects on the assembly state of V-ATPases as diagrammed in Figure 5a. 208 To further explore the functional significance of these assembly regulators in aging cells, 209 we examined the replicative lifespans of deletion mutants lacking Rav1 or Oxr1 (Figure 5b). 210 Replicative lifespan was measured on YEPD, pH 5 plates, conditions that are optimal for growth 211 of rav1∆ strains. Deletion of Rav1 shortened replicative lifespan (median 17 cell divisions, n= 212 30) by 26% relative to wild-type cells (median 23 cell divisions, n= 30). In contrast, the deletion 213 of Oxr1 extended replicative lifespan by 47.8% over wild-type (median 34 cell divisions, n=30). 214 These results reinforce the significance of V-ATPase assembly in replicative aging and suggest 215 that RA VE pro-assembly activity (disrupted in the rav1∆ mutant) and Oxr1 anti-assembly 216 activity (lost in oxr1∆) may be central determinants of lifespan. 217 Given these results, we asked whether there were differences in levels of RA VE subunits 218 or Oxr1 between young and old cells. We isolated young and old populations of cells containing 219 myc13-tagged Rav1 or Rav2 by biotinylation and streptavidin magnetic separation as described 220 above (Figure 2), then assessed the levels of the tagged proteins in cell lysates. Although there is 221 no significant difference in Rav1-myc13 levels between young and old cells (Figure 6a), we 222 consistently observed a significant decrease in protein levels of Rav2-myc13 in older cells 223 (Figure 6b). We also assessed expression of RAV2 in young and old cells by quantitative PCR but 224 observed no significant difference in mRNA levels (Figure 6c). Because Rav2 is required for 225 RA VE complex function in promoting V-ATPase assembly (Seol et al., 2001; Smardon et al., 226 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.23.604825doi: bioRxiv preprint 12 2002), these results suggest that partial loss of RA VE function in old cells could contribute to 227 reduced V-ATPase assembly and replicative aging. 228 To further explore the basis of the extended replicative lifespan in oxr1∆ cells, we 229 observed Vma5-GFP localization in a population of oxr1∆ cells of mixed age. As shown in 230 Figure 6d, oxr1∆ mutants localize Vma5-GFP to the vacuole even in older cells with >5 bud 231 scars. In order to quantitate this effect, we again binned cells by the number of bud scars as 232 described above, and compared Vma5-GFP localization to localization in young daughter cells 233 (Figure 6e). In contrast to wild-type cells, there is no significant difference in Vma5-GFP 234 localization in the oxr1∆ mutant until cells have divided 25 or more times. We also isolated 235 young and old populations from an HA-tagged Oxr1 strain. As shown in Figure 6f, the levels of 236 Oxr1 do not change with age. However, loss of RA VE function could ultimately favor Oxr1-237 induced disassembly. 238 Intriguingly, these results suggest that the oxr1∆ mutation mirrors the effects of CR on V-239 ATPase assembly and longevity. The results also indicate that direct manipulation of V-ATPase 240 assembly by mutation of critical assembly factors can affect replicative lifespan. Specifically, 241 promoting V-ATPase assembly through the RA VE complex appears to be important for 242 preserving replicative lifespan, while the opposing effects of Oxr1 on the V-ATPase tend to 243 shorten lifespan. We hypothesized that restoration of V-ATPase assembly in older cells grown 244 under CR conditions (Figure 4) might be supported by restoration of Rav2 levels. To test this, we 245 isolated young and old populations from Rav2-myc13 tagged cells grown under CR conditions. 246 Under these conditions, V-ATPase assembly and function restored in older cells (Figure 4) and as 247 shown in Figure 6g, Rav2 levels are also restored. 248 249 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.23.604825doi: bioRxiv preprint 13

250 The results described here establish V-ATPase disassembly as an important factor in the 251 reduced vacuolar acidification observed in aging yeast cells. We demonstrate an increase in V-252 ATPase disassembly at relatively early replicative ages (5-9 cell divisions) similar to the age at 253 which Hughes and Gottschling first observed compromised vacuolar acidification (A. L. Hughes 254 & Gottschling, 2012). We show that promoting assembly of the V-ATPase through CR can 255 restore V-ATPase assembly and vacuolar pH in older cells and that deletion of Oxr1, a negative 256 regulator of V-ATPase assembly, can extend replicative lifespan. Taken together, these data 257 support V-ATPase disassembly as a significant age-related factor behind reduced vacuolar and 258 lysosomal acidification and the associated declines in function. This mechanism does not 259 necessarily conflict with those proposed previously. If V-ATPases are more disassembled in older 260 cells, cells may be even less able to tolerate an imbalance between Pma1 activity at the plasma 261 membrane and V-ATPase activity at the vacuole (Henderson et al., 2014). We see no difference 262 in protein levels of the core V1 subunits Vma1 and Vma2, or Vma5, between young and old 263 cells. The presence of similar levels of Vph1 in vacuoles of young and old cells suggests that V0 264 subunits are also expressed at similar levels, since V0 assembly occurs in the ER and reduced 265 levels of any V0 subunit reduces V0 subcomplex levels at the vacuole (Kane, Kuehn, Howald-266 Stevenson, & Stevens, 1992). The data presented here suggest a post-transcriptional regulatory 267 mechanism, but increased levels of a V0 assembly factor like Vph2 (A. L. Hughes & Gottschling, 268 2012) might still promote assembly and improve acidification. 269 Many questions remain to be investigated in the future. The reduction in Rav2 protein 270 levels with aging appears to be post-transcriptional but we do not yet know whether this arises 271 from reduced translation or increased degradation of the protein. Perhaps more importantly, even 272 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.23.604825doi: bioRxiv preprint 14 though reduced levels of Rav2 levels could help explain increased V-ATPase disassembly, we do 273 not know whether the Rav2 reduction occurs as a result of some general vulnerability in aging 274 cells or is deliberately programmed (Gladyshev et al., 2021). The significant increase in lifespan 275 and prolonged V-ATPase assembly in oxr1∆ cells seems to argue against any growth advantage 276 from increased V-ATPase disassembly in older cells, at least at the single cell level. 277 Many aging pathways are tightly linked to nutritional signaling, and reversible 278 disassembly is often driven by nutritional signaling pathways. Given the rapid disassembly of the 279 V-ATPase upon acute glucose deprivation (Kane, 1995), it was initially surprising that growth 280 under CR conditions suppresses V-ATPase disassembly with age. However, we previously 281 showed that the acute disassembly response required glucose concentrations well below 0.5% 282 (Parra & Kane, 1998). In mammalian cells, V-ATPase assembly generally increases in response 283 to nutrient deprivation (Ratto et al., 2022; Stransky & Forgac, 2015). CR could mimic this effect 284 in aging cells, preserving lysosomal acidification and function. Signals involved in reversible 285 disassembly of the V-ATPase are incompletely understood. However, it is intriguing that in yeast, 286 RA VE appears to play a central role in glucose signaling during acute glucose deprivation. 287 RA VE is released from the vacuolar membrane upon acute glucose deprivation and recruited 288 back to the membrane upon glucose restoration, even in the absence of V1 subunit C and the V1 289 subcomplex (Jaskolka, Winkley, et al., 2021). The RA VE complex appears to be a major 290 determinant of V-ATPase assembly state in multiple situations, including aging. 291 Reduced V-ATPase assembly could certainly be a factor in the age-related decline in 292 lysosomal acidification and function in higher eukaryotes. Reversible disassembly actively 293 occurs in higher eukaryotes including mammalian cells, so the apparatus for age-induced 294 assembly regulation is available. Rabconnectin-3 complexes of higher eukaryotes are the 295 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.23.604825doi: bioRxiv preprint 15 functional homologues of the yeast RA VE complex (Ratto et al., 2022; Yan, Denef, & 296 Schupbach, 2009). Oxr1 belongs to a family of TLDc proteins that are also found in mammals, 297 and several of these proteins have been shown to bind to mammalian V-ATPases (Eaton, Brown, 298 & Merkulova, 2021; Wilkens et al., 2023). These data suggest that the core elements for 299 controlled V-ATPase disassembly during aging are present in other cells. Here, we observed V-300 ATPase disassembly in a yeast replicative aging model, which is most comparable to mammalian 301 cell types that continue to divide, such as adult stem cells (He et al., 2018). However, V-ATPase 302 activity is also critical in the yeast chronological aging model, which is more analogous to long-303 lived, non-dividing mammalian cells. In this model increased V-ATPase assembly has been 304 associated with longevity (Wilms et al., 2017). Taken together, these data suggest that V-ATPase 305 assembly state is linked to multiple aging models and could easily play a role in aging in higher 306 eukaryotes. 307 308

309 Yeast strains and plasmids 310 All strains analyzed were in the BY4741 (MATa his3Δ1 leu2Δ0 met15Δ0 ura3Δ0) or BY4742 311 (MATα his3Δ1 leu2Δ0 lys1∆ ura3Δ0) background. BY4742 strains containing Vma5-GFP::HIS3 312 and Vph1-GFP::HIS3 were constructed as part of the genome-wide GFP-tagging project (Huh et 313 al., 2003) and purchased from Thermo Fisher. VMA2 was C-terminally tagged with GFP by PCR 314 amplification from pFA6a-GFP-KanMX6 and genomic integration (Longtine et al., 1998). The 315 strain containing oxr1∆::kanMX and Vma5-GFP::HIS3 was obtained by crossing BY4741 316 oxr1∆::kanMX from the haploid deletion collection and BY4742 Vma5-GFP::HIS3, sporulating 317 the diploid, and obtaining spores with the desired genotype by tetrad dissection. The pRS316 318 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.23.604825doi: bioRxiv preprint 16 Oxr1-HA plasmid (Khan & Wilkens, 2024) was transformed into oxr1Δ and transformants were 319 selected for on SC medium (fully supplemented minimal medium) lacking uracil. Rav1-320 myc13::kanMX and Rav2-myc13::kanMX strains were described previously (Smardon et al., 321 2002). The tagged alleles were PCR amplified from the original strains and integrated into 322 BY4741. 323 Fluorescence microscopy 324 Strains expressing GFP-tagged Vma5, Vma2, or Vph1 were grown to log phase in SC 325 medium containing either 2% or 0.5% glucose overnight, pelleted by centrifugation, then 326 suspended in fresh medium containing 2% or 0.5% glucose and grown for an additional 2 hours. 327 Cells were stained with calcofluor white (CW) by diluting a 1 mg/ml CW stock to a final 328 concentration of 10 µg/ml with cells, 5 min prior to imaging. Cells were visualized with a 100x 329 oil (NA 1.4) objective on a Zeiss Imager.Z1 fluorescence microscope with a Hamamastu CCD 330 camera and AxioVision software. Cells were viewed through differential interference contrast 331 (DIC) optics or fluorescence was visualized using a DAPI filter set for CW and a GFP filter set 332 for GFP-tagged subunits. Bud scars were counted from CW staining; in mixed age populations, 333 cells with 5 or more bud scars were designated as old and those with less than 5 as young. To 334 obtain a more precise count of bud scars for binning, cells were visualized on multiple focal 335 planes. GFP fluorescence was then determined for each of the binned ranges. Images of GFP-336 tagged proteins were captured then processed in FIJI. To assess vacuolar localization, a line was 337 drawn across a cell and through the vacuolar membrane. From this line scan, a plot illustrating 338 fluorescence intensity along the line was generated, with peaks indicating areas of elevated 339 fluorescence at the vacuolar membrane. By quantifying the peak intensity (maximum 340 fluorescence), we quantitated the vacuolar localization for each GFP-tagged subunit. Each 341 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.23.604825doi: bioRxiv preprint 17 biological replicate corresponds to a distinct culture of yeast cells. Maximum fluorescence was 342 quantitated for at least 20 young and 20 old cells per biological replicate. When cells were 343 binned by age, at least 20 cells per biological replicate were counted for each bin. To normalize 344 maximum fluorescence across replicates, the maximum fluorescence from line scans of young 345 and old cells in each biological replicate was averaged and then divided by the average of the 346 young cells in that replicate. The normalized fluorescence for each biological replicate was 347 plotted, along with average intensities across replicates +/- s.e.m. (standard error of the mean). 348 Statistical significance was determined by t-test for the young and old cell comparisons and by 349 ANOV A for the binned samples. In order to show the range of values for the young cells across 350 experiments, the values for the young cells in each biological replicate were averaged and the 351 values for the individual replicates were divided by the average and shown as points on each 352 graph. 353 Age Enrichment 354 Age enrichment was performed as described by Jin et al. (Jin et al., 2021). Cells were collected 355 from a 50 mL fresh overnight culture in YEP (yeast extract-peptone medium) supplemented with 356 2% or 0.5% glucose to an OD600 of 1.0 and washed twice with cold phosphate-buffered saline 357 (PBS), pH 7.4. Cells were pelleted by centrifugation and washed three times in cold sterile PBS, 358 then labeled with 1.6 mg/ml EZ-Link Sulfo-NHS-LC-Biotin (Pierce) at room temperature for 30 359 min with gentle agitation. After labeling, the cells were washed three times with cold PBS, pH 360 8.0, to remove free biotin, and resuspended in YEP supplemented with 2% or 0.5% glucose for 361 growth overnight. After 16 hours, cells were pelleted by centrifugation and resuspended in 35 ml 362 of cold PBS, pH 7.4, mixed with 250 μl of a magnetic streptavidin bead suspension (Pierce), and 363 incubated for 60 min at 4°C. The mixture, in a 50 mL conical tube, was loaded onto a magnetic 364 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.23.604825doi: bioRxiv preprint 18 separation column (Permagen) at 4°C to separate biotinylated cells and allow unbound cells to 365 settle, and supernatant was removed gently by pipetting. Magnetically separated cells were 366 subsequently washed three times by resuspending in 35 mL PBS, pH 7.4 supplemented with 2% 367 or 0.5% glucose, repeating magnetic separation, and discarding the supernatant. After washing, 368 cells were resuspended in 200 mL of YEP or SC supplemented with 2% or 0.5% glucose and 369 allowed to grow for an additional 4 hours before obtaining the final "old" mother cells and 370 "young" daughter cells. Cells were loaded on the magnetic separation column as described above 371 and daughter cells obtained from the supernatant were pelleted by centrifugation to obtain a 372 concentrated population of young cells. After the final wash, the magnetic beads were 373 resuspended in 1 ml PBS 7.4 and centrifuged at 4000g to concentrate the old population. The two 374 populations were then stored at -80°C for further biochemical analysis or used immediately for 375 pH measurements (see below). 376 Whole Cell Lysates and Immunoblots 377 Cell pellets from age-enriched young and old cell populations were resuspended in hot 378 cracking buffer (8 M urea, 5% SDS, 1 mM EDTA, 50 mM Tris-HCl, pH 6.8) and glass beads. 379 The mixture was vortexed for 10 sec and incubated at 95 °C for 30 sec repeatedly for a total of 5 380 min. Cellular debris was pelleted by centrifugation at 16,000 × g for 2 min, and supernatants 381 containing whole-cell extracts were used immediately or stored at −80 °C until use. 382 After determination of protein concentrations by Bradford assay, equal concentrations of 383 protein for each sample were separated by SDS-PAGE and transferred to a nitrocellulose 384 membrane. Blots were blocked for 1 hr in TBST (20 mM Tris-HCl pH 7.5, 150 mM NaCl, 385 0.05% Tween-20) plus 5% nonfat milk before incubating overnight with primary antibodies at 4 386 °C with agitation. Primary antibodies (all used at a 1:500 dilution) included mouse monoclonal 387 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.23.604825doi: bioRxiv preprint 19 antibodies: 7A2 against Vma5, 10D7 against Vph1, 8B1 against Vma1, and 13D11 against Vma2 388 (Kane et al., 1992). In addition, anti-myc monoclonal 9E10 (Santa Cruz Biotechnology), anti-HA 389 monoclonal (BioLegend), and anti-GAPDH (Proteintech) antibodies were purchased and used at 390 1:500, 1:500, and 1:10000, respectively. After washing three times with TBST buffer, HRP-391 conjugated anti-mouse secondary antibody (Bio-Rad) was added at a final dilution of 1:2000 and 392 incubated for 60 min at room temperature. The blot was washed again, incubated with Bio-Rad 393 Clarity Western ECL substrate and imaged in an Azure Sapphire FL Biomolecular Imager. 394 Images were quantified using FIJI. Molecular mass markers were included on every blot. In 395 images of blots, the mass of the marker nearest in size to the protein is indicated. 396 Vacuolar pH Measurements 397 Vacuolar pH was measured using the ratiometric fluorescent dye BCECF-AM (Invitrogen) as 398 described previously(Diakov et al., 2013). Age-enriched populations of cells were loaded with 399 BCECF-AM in YEP supplemented with 2% or 0.5% glucose. After washing with YEP media to 400 remove dye, cells were resuspended in YEP, deprived of glucose, and incubated on rotator for 30 401 min. For fluorescence measurement, 20 µl of cell suspension was diluted into 3 ml 1 mm MES 402 pH 5.0 buffer, and fluorescence intensity at excitation wavelengths 450 and 495 nm and emission 403 wavelength 535 nm and was monitored continuously in a Horiba Jovin Yvon Spectrafluor Max 404 fluorometer with temperature maintained at 30°C. The fluorescence ratio for each sample was 405 calibrated for each strain in every experiment by clamping the pH to a range of values from 5.0 406 to 7.0 as described, and the resulting calibration curve was used to convert the experimental 407 fluorescence ratios to vacuolar pH. 408 Replicative Life Span 409 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.23.604825doi: bioRxiv preprint 20 Replicative life span assays were performed on YEP, 2% glucose plates buffered to pH 5. 410 Daughter cells were sequentially removed by micromanipulation (Steffen, Kennedy, & 411 Kaeberlein, 2009). Survival curves are pooled data from experiment-matched controls. The 412 number of divisions for 30 mother cells were scored for each curve. p values for replicative life 413 span survival curve comparisons were calculated with a Cox regression model. Kaplan-Meier 414 survival curves were plotted with GraphPad Prism. 415 RT-PCR of RAV2 from young and old cells 416 RNA was extracted from young and old cells, obtained by age enrichment as described above, 417 using the NEB Monarch Total RNA Miniprep Kit . RT-PCR was conducted using the NEB Luna 418 Universal One-Step RT-qPCR Kit and performed on a Bio-Rad CFX384 Touch System. Data 419 analysis was conducted using CFX Maestro Software to determine expression levels. 420 421

This work was supported by NIH R35 GM145256 to P.M.K. The authors 422 thank M. Murad Khan and Dr. Stephan Wilkens for sharing the oxr1∆ strain and Oxr1-HA 423 plasmid, and Dr. Xin Jie Chen for helpful discussions and a critical reading of the manuscript. 424 Author contributions: F.H. performed experiments, analyzed data, prepared figures, and wrote 425 the first draft of the manuscript; P.M.K. obtained funding for the project, analyzed data, prepared 426 figures, and contributed to writing of the manuscript. 427 Conflict of interest: The authors declare that they have no conflicts of interest with the contents 428 of this article. 429 Data availability: The data that support the findings of this study are openly available in 430 Upstate.figshare.com at https://upstate.figshare.com. DOI: 10.58120/upstate.26023660 431 432 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.23.604825doi: bioRxiv preprint 25

523 524 Bodzeta, A., Kahms, M., & Klingauf, J. (2017). The Presynaptic v -ATPase Reversibly 525 Disassembles and Thereby Modulates Exocytosis but Is Not Part of the Fusion Machinery. 526 Cell Rep, 20(6), 1348-1359. doi:10.1016/j.celrep.2017.07.040 527 Chen, K. L., Ven, T. N., Crane, M. M., Brunner, M. L. C., Pun, A. K., Helget, K. L., . . . Wasko, B. 528 M. (2020). Loss of vacuolar acidity results in iron -sulfur cluster defects and divergent 529 homeostatic responses during aging in Saccharomyces cerevisiae. Geroscience, 42(2), 749-530 764. doi:10.1007/s11357-020-00159-3 531 Collins, M. P., & Forgac, M. (2020). Regulation and function of V -ATPases in physiology and 532 disease. Biochim Biophys Acta Biomembr, 1862 (12), 183341. 533 doi:10.1016/j.bbamem.2020.183341 534 Diab, H. I., & Kane, P. M. (2013). Loss of vacuolar H+ -ATPase (V-ATPase) activity in yeast 535 generates an iron deprivation signal that is moderated by induction of the peroxiredoxin 536 TSA2. Journal of Biological Chemistry, 288(16), 11366-11377. doi:M112.419259 [pii] 537 10.1074/jbc.M112.419259 538 Diakov, T. T., Tarsio, M., & Kane, P. M. (2013). Measurement of vacuolar and cytosolic pH in 539 vivo in yeast cell suspensions. J Vis Exp, Apr 19(74). doi:10.3791/50261 540 Eaton, A. F., Brown, D., & Merkulova, M. (2021). The evolutionary conserved TLDc domain 541 defines a new class of (H(+))V -ATPase interacting proteins. Sci Rep, 11 (1), 22654. 542 doi:10.1038/s41598-021-01809-y 543 Ghavidel, A., Baxi, K., Prusinkiewicz, M., Swan, C., Belak, Z. R., Eskiw, C. H., . . . Harkness, T. 544 A. (2018). Rapid Nuclear Exclusion of Hcm1 in Aging Saccharomyces cerevisiae Leads to 545 Vacuolar Alkalization and Replicative Senescence. G3 (Bethesda), 8 (5), 1579 -1592. 546 doi:10.1534/g3.118.200161 547 Gladyshev, V . N., Kritchevsky, S. B., Clarke, S. G., Cuervo, A. M., Fiehn, O., de Magalhaes, J. P., 548 . . . Cummings, S. R. (2021). Molecular Damage in Aging. Nat Aging, 1(12), 1096-1106. 549 doi:10.1038/s43587-021-00150-3 550 Hansen, M., Rubinsztein, D. C., & Walker, D. W. (2018). Autophagy as a promoter of longevity: 551 insights from model organisms. Nat Rev Mol Cell Biol, 19 (9), 579 -593. 552 doi:10.1038/s41580-018-0033-y 553 He, C., Zhou, C., & Kennedy, B. K. (2018). The yeast replicative aging model. Biochim Biophys 554 Acta Mol Basis Dis, 1864(9 Pt A), 2690-2696. doi:10.1016/j.bbadis.2018.02.023 555 Henderson, K. A., Hughes, A. L., & Gottschling, D. E. (2014). Mother-daughter asymmetry of pH 556 underlies aging and rejuvenation in yeast. Elife, 3, e03504. doi:10.7554/eLife.03504 557 Hughes, A. L., & Gottschling, D. E. (2012). An early age increase in vacuolar pH limits 558 mitochondrial function and lifespan in yeast. Nature, 492(7428), 261-265. doi:nature11654 559 [pii] 560 10.1038/nature11654 561 Hughes, C. E., Coody, T. K., Jeong, M. Y ., Berg, J. A., Winge, D. R., & Hughes, A. L. (2020). 562 Cysteine Toxicity Drives Age -Related Mitochondrial Decline by Altering Iron 563 Homeostasis. Cell, 180(2), 296-310 e218. doi:10.1016/j.cell.2019.12.035 564 Huh, W. K., Falvo, J. V ., Gerke, L. C., Carroll, A. S., Howson, R. W., Weissman, J. S., & O'Shea, 565 E. K. (2003). Global analysis of protein localization in budding yeast. Nature, 425(6959), 566 686-691. Retrieved from 567 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.23.604825doi: bioRxiv preprint 26 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Citati568 on&list_uids=14562095 569 Jaskolka, M. C., Tarsio, M., Smardon, A. M., Khan, M. M., & Kane, P. M. (2021). Defining steps 570 in RA VE -catalyzed V -ATPase assembly using purified RA VE and V -ATPase 571 subcomplexes. J Biol Chem, 296, 100703. doi:10.1016/j.jbc.2021.100703 572 Jaskolka, M. C., Winkley, S. R., & Kane, P. M. (2021). RA VE and Rabconnectin-3 Complexes as 573 Signal Dependent Regulators of Organelle Acidification. Front Cell Dev Biol, 9, 698190. 574 doi:10.3389/fcell.2021.698190 575 Jin, X., Cao, X., & Liu, B. (2021). Isolation of Aged Yeast Cells Using Biotin-Streptavidin Affinity 576 Purification. Methods Mol Biol, 2196, 223-228. doi:10.1007/978-1-0716-0868-5_17 577 Kane, P. M. (1995). Disassembly and reassembly of the yeast vacuolar H(+) -ATPase in vivo. J 578 Biol Chem, 270(28), 17025-17032. 579 Kane, P. M., Kuehn, M. C., Howald-Stevenson, I., & Stevens, T. H. (1992). Assembly and targeting 580 of peripheral and integral membrane subunits of the yeast vacuolar H(+) -ATPase. J Biol 581 Chem, 267(1), 447-454. 582 Kaushik, S., Tasset, I., Arias, E., Pampliega, O., Wong, E., Martinez-Vicente, M., & Cuervo, A. M. 583 (2021). Autophagy and the hallmarks of aging. Ageing Res Rev, 72 , 101468. 584 doi:10.1016/j.arr.2021.101468 585 Khan, M. M., Lee, S., Couoh -Cardel, S., Oot, R. A., Kim, H., Wilkens, S., & Roh, S. H. (2022). 586 Oxidative stress protein Oxr1 promotes V -ATPase holoenzyme disassembly in catalytic 587 activity-independent manner. EMBO J, 41(3), e109360. doi:10.15252/embj.2021109360 588 Khan, M. M., & Wilkens, S. (2024). Molecular mechanism of Oxr1p mediated disassembly of 589 yeast V-ATPase. EMBO Rep. doi:10.1038/s44319-024-00126-5 590 Klossel, S., Zhu, Y ., Amado, L., Bisinski, D. D., Ruta, J., Liu, F., & Gonzalez Montoro, A. (2024). 591 Yeast TLDc domain proteins regulate assembly state and subcellular localization of the V-592 ATPase. EMBO J. doi:10.1038/s44318-024-00097-2 593 Kohio, H. P., & Adamson, A. L. (2013). Glycolytic control of vacuolar -type ATPase activity: a 594 mechanism to regulate influenza viral infection. Virology, 444(1-2), 301-309. doi:S0042-595 6822(13)00396-6 [pii] 596 10.1016/j.virol.2013.06.026 597 Kurz, T., Terman, A., Gustafsson, B., & Brunk, U. T. (2008). Lysosomes in iron metabolism, 598 ageing and apoptosis. Histochem Cell Biol, 129 (4), 389 -406. Retrieved from 599 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Citati600 on&list_uids=18259769 601 Liu, Y ., Steinbusch, L. K. M., Nabben, M., Kapsokalyvas, D., van Zandvoort, M., Schonleitner, P., 602 . . . Luiken, J. (2017). Palmitate -Induced Vacuolar-Type H(+)-ATPase Inhibition Feeds 603 Forward Into Insulin Resistance and Contractile Dysfunction. Diabetes, 66(6), 1521-1534. 604 doi:10.2337/db16-0727 605 Longo, V . D., & Anderson, R. M. (2022). Nutrition, longevity and disease: From molecular 606 mechanisms to interventions. Cell, 185(9), 1455-1470. doi:10.1016/j.cell.2022.04.002 607 Longtine, M. S., McKenzie, A., 3rd, Demarini, D. J., Shah, N. G., Wach, A., Brachat, A., . . . 608 Pringle, J. R. (1998). Additional modules for versatile and economical PCR -based gene 609 deletion and modification in Saccharomyces cerevisiae. Yeast, 14(10), 953-961. 610 Martinez-Munoz, G. A., & Kane, P. (2008). Vacuolar and plasma membrane proton pumps 611 collaborate to achieve cytosolic pH homeostasis in yeast. J Biol Chem, 283 (29), 20309-612 20319. doi:M710470200 [pii] 613 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.23.604825doi: bioRxiv preprint 27 10.1074/jbc.M710470200 614 Nixon, R. A. (2020). The aging lysosome: An essential catalyst for late -onset neurodegenerative 615 diseases. Biochim Biophys Acta Proteins Proteom, 1868 (9), 140443. 616 doi:10.1016/j.bbapap.2020.140443 617 Oot, R. A., & Wilkens, S. (2020). A "Sugar -Coated" Proton Pump Comes into Focus: High -618 Resolution Structure of a Human V -ATPase. Mol Cell, 80 (3), 379 -380. 619 doi:10.1016/j.molcel.2020.10.020 620 Parra, K. J., & Kane, P. M. (1998). Reversible association between the V1 and V0 domains of yeast 621 vacuolar H+-ATPase is an unconventional glucose-induced effect. Mol Cell Biol, 18 (12), 622 7064-7074. Retrieved from http://mcb.asm.org/cgi/content/full/18/12/7064 623 Perera, R. M., & Zoncu, R. (2016). The Lysosome as a Regulatory Hub. Annu Rev Cell Dev Biol, 624 32, 223-253. doi:10.1146/annurev-cellbio-111315-125125 625 Ratto, E., Chowdhury, S. R., Siefert, N. S., Schneider, M., Wittmann, M., Helm, D., & Palm, W. 626 (2022). Direct control of lysosomal catabolic activity by mTORC1 through regulation of 627 V-ATPase assembly. Nat Commun, 13(1), 4848. doi:10.1038/s41467-022-32515-6 628 Schiffer, I., Gerisch, B., Kawamura, K., Laboy, R., Hewitt, J., Denzel, M. S., . . . Antebi, A. (2021). 629 miR-1 coordinately regulates lysosomal v-ATPase and biogenesis to impact proteotoxicity 630 and muscle function during aging. Elife, 10. doi:10.7554/eLife.66768 631 Seol, J. H., Shevchenko, A., & Deshaies, R. J. (2001). Skp1 forms multiple protein complexes, 632 including RA VE, a regulator of V-ATPase assembly. Nat Cell Biol, 3(4), 384-391. 633 Smardon, A. M., Tarsio, M., & Kane, P. M. (2002). The RA VE complex is essential for stable 634 assembly of the yeast V-ATPase. J Biol Chem, 277(16), 13831-13839. 635 Steffen, K. K., Kennedy, B. K., & Kaeberlein, M. (2009). Measuring replicative life span in the 636 budding yeast. J Vis Exp(28). doi:10.3791/1209 637 Stransky, L. A., & Forgac, M. (2015). Amino Acid Availability Modulates Vacuolar H+ -ATPase 638 Assembly. Journal of Biological Chemistry, 290(45), 27360-27369. doi:M115.659128 [pii] 639 10.1074/jbc.M115.659128 640 Sumner, J. P., Dow, J. A., Earley, F. G., Klein, U., Jager, D., & Wieczorek, H. (1995). Regulation 641 of plasma membrane V -ATPase activity by dissociation of peripheral subunits. J Biol 642 Chem, 270(10), 5649-5653. 643 Tabke, K., Albertmelcher, A., Vitavska, O., Huss, M., Schmitz, H. P., & Wieczorek, H. (2014). 644 Reversible disassembly of the yeast V -ATPase revisited under in vivo conditions. 645 Biochemical Journal, 462(1), 185-197. doi:BJ20131293 [pii] 646 10.1042/BJ20131293 647 Vilchez, D., Saez, I., & Dillin, A. (2014). The role of protein clearance mechanisms in organismal 648 ageing and age-related diseases. Nat Commun, 5, 5659. doi:10.1038/ncomms6659 649 Wilkens, S., Khan, M. M., Knight, K., & Oot, R. A. (2023). Tender love and disassembly: How a 650 TLDc domain protein breaks the V -ATPase. Bioessays, 45 (7), e2200251. 651 doi:10.1002/bies.202200251 652 Wilms, T., Swinnen, E., Eskes, E., Dolz-Edo, L., Uwineza, A., Van Essche, R., . . . Winderickx, J. 653 (2017). The yeast protein kinase Sch9 adjusts V-ATPase assembly/disassembly to control 654 pH homeostasis and longevity in response to glucose availability. PLoS Genet, 13 (6), 655 e1006835. doi:10.1371/journal.pgen.1006835 656 Yan, Y ., Denef, N., & Schupbach, T. (2009). The vacuolar proton pump, V-ATPase, is required for 657 notch signaling and endosomal trafficking in Drosophila. Dev Cell, 17 (3), 387 -402. 658 doi:S1534-5807(09)00260-3 [pii] 659 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.23.604825doi: bioRxiv preprint 28 10.1016/j.devcel.2009.07.001 660 661 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.23.604825doi: bioRxiv preprint 21 FIGURE LEGENDS 433 Figure 1: V-ATPases are more disassembled in yeast cells of older replicative age. (a) 434 BY4742 cells expressing Vma5-GFP grown in SC containing 2% glucose. Differential 435 interference contrast microscopy (DIC) was used to visualize vacuoles. Bud scars are stained 436 with calcofluor white (CW) to determine replicative age. (b) Left panel provides representative 437 example of quantitative measurements using line scans. The young cell is from the dashed white 438 box in the Vma5-GFP image of Figure 1a and the old cell is from the dashed red box. Plot 439 profiles are superimposed for the young cell (black) and old cell (red). (c) Quantitation of 440 maximum fluorescent intensity across five biological replicates, after normalization to the 441 average intensity in the youngest bin for each replicate. Each biological replicate (dot) represents 442 at least 20 cells and bars represent the mean +/- s.e.m. Significance was calculated by ordinary 443 one-way ANOV A. “Old” for subsequent experiments (without age enrichment) is categorized at 444 ≥5 bud scars. **** represents a p-value <0.0001. 445 446 Figure 2: Age-enriched populations of cells have comparable levels of V-ATPase subunits. 447 (a) Diagram of V-ATPase showing the relative positions of Vma5, Vma2, and Vph1 (Image was 448 prepared with Biorender.com). (b) BY4742 cells expressing Vma2-GFP (V1B) grow in SC with 449 2% glucose. CW was used to visualize bud scars. The CW images for young cells were 450 overexposed relative to those for the old cells in order to visualize the low level staining in cells 451 with few bud scars. Normalized maximum fluorescence was obtained through line scan 452 quantitation using FIJI as in Figure 1. Means +/- s.e.m. of three biological replicates are shown; 453 each replicate is represented by a dot. Significance was calculated by unpaired Student’s t test, 454 *** p value=0.0009 . (c) Cells expressing Vph1-GFP were visualized and analyzed as described 455 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.23.604825doi: bioRxiv preprint 22 in 2b. (d) After biotin-streptavidin age enrichment, bud scars were counted for 100 cells in 456 young and old populations, binned by bud count, and the age distribution plotted. (e) Lysates 457 were prepared from age-enriched populations defined as in 1c and equal protein concentrations 458 were separated by SDS-PAGE and examined by immunoblot for V-ATPase protein levels in 459 young (Y) vs. old (O) cell populations defined as in 1c. (f) Band intensities were quantified using 460 FIJI, ratios of V-ATPase subunit levels to the GAPDH internal control were calculated, and ratios 461 were normalized to the young population for each biological replicate. Significance calculated by 462 unpaired Student’s t test. Data are presented as mean (horizontal bars) ± s.e.m. (whiskers) of 463 three biological replicates. n.s.= not significant. 464 465 Figure 3: Vacuolar pH is more alkaline in old cells. (a) Vacuolar pH responses were measured 466 for wild-type BY4742 age-enriched young and old populations as described in Methods. 467 Glucose-deprived cultures were loaded with BCECF-AM. Fluorescence intensity values were 468 collected every 10 sec at excitation wavelengths 450 and 490 nm and emission wavelength 535 469 nm, and glucose was added to a final concentration of 2% after 3 min. The ratio of fluorescence 470 signals from the two excitation wavelengths was calculated and converted to pH via a calibration 471 curve. (b) Fluorescence measurements at 1 min. (before glucose addition), 5 min (2 min. after 472 glucose addition), and 8 min. (5 min after glucose addition). Calculated pH measurements are 473 presented as mean (horizontal bars) ± s.e.m. (whiskers) of three biological replicates. * indicates 474 a p value of 0.02, ** represents a p value of 0.001. 475 476 Figure 4: Caloric restriction (0.5% glucose) restores V-ATPase assembly and vacuolar pH 477 in old cells. (a), (b), (c) Strains used in Figures 1 and 2 show recruitment of Vma5-GFP 4a and 478 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.23.604825doi: bioRxiv preprint 23 Vma2-GFP 4b to the vacuolar membrane after growth in SC with 0.5% glucose while Vph1-GFP 479 4c remains unchanged. Maximum fluorescence was measured and normalized as described in 480

and Figure 1. CW was visualized in young and old cells as described in Figure 2b. No 481 significant differences between young and old cells was observed as calculated by unpaired 482 Student’s t test. (d)Young and old cell populations were obtained by biotin-streptavidin age-483 enrichment from cells grown in YEP supplemented with 0.5% glucose, and vacuolar pH 484 responses were measured. Data were collected and analyzed as in Figure 3b. Quantitated data 485 are presented as mean (horizontal bars) ± s.e.m. (whiskers) of three biological replicates. 486 487 Figure 5: Effects of rav1∆ and oxr1∆ on replicative lifespan. (a) Schematic of reversible 488 disassembly highlighting the roles of the RA VE complex and Oxr1 (Image was prepared with 489 Biorender.com). (b) Kaplan-Meier curves comparing the replicative lifespan (RLS) of rav1∆ 490 (green), oxr1∆ (red), and wild-type cells (black). Median number of replicative generations is 491 shown in parentheses for each strain, and the difference between rav1∆ and oxr1∆ median values 492 and wild-type are expressed as %. Deletion of OXR1 significantly increases yeast RLS 493 (p<0.001), and deletion of RA VE component RAV1 shortens RLS (p < 0.01). 494 495 Figure 6: Rav2 level is reduced during replicative aging, but restored by CR. (a) Analysis of 496 Rav1 levels in young and old cells. Age-enriched populations were obtained from BY4741 cells 497 containing Rav1-myc13 (young (Y) versus old (O) cells), and cell lysates were prepared, 498 separated, and quantitated as described in Methods and Figure 2. Normalized data are presented 499 as mean (horizontal bars) ± s.e.m. (whiskers) of three biological replicates. ns=not significant (b) 500 Immunoblot analysis of BY4741 Rav2-myc13 from young and old cells prepared as in 6a. 501 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.23.604825doi: bioRxiv preprint 24 Normalized data are presented as mean (horizontal bars) ± s.e.m. (whiskers) of three biological 502 replicates. P-value 0.0094. (c) Quantitative RT-PCR comparing expression of RAV2 mRNA 503 between young and old cells. (d) Wild-type and oxr1Δ cells containing Vma5-GFP were grown 504 in SC containing 2% glucose. DIC used to visualize vacuoles and CW used to visualize bud scars 505 as described in Figure 2b. (e) BY4741 oxr1∆ cells were binned by the number of bud scars and 506 normalized maximal fluorescence quantitated as described in Figure 1. *** indicates a P value 507 of 0.001 in comparison to the youngest bin; other bins are not significantly different from the 508 youngest bin. (f) Immunoblot analysis of a BY4741 oxr1Δ strain expressing Oxr1-HA from a 509 low copy plasmid. Young and old cell populations were isolated as in 6a. Data are presented as 510 mean (horizontal bars) ± s.e.m. (whiskers) of three biological replicates. (g) Immunoblot analysis 511 comparing Rav2-myc13 levels in cells grown in YEP supplemented with 2% glucose and 0.5% 512 glucose (CR conditions). Samples were prepared and analyzed as in 6b. Quantification of data 513 are presented as mean (horizontal bars) ± s.e.m. (whiskers) of three biological replicates. *** 514 indicates a P value of 0.0001, n.s.= not significant. 515 516 Supporting information, Figure 1: Cells show similar growth rates over 12 hours in 2% 517 and 0.5% glucose. BY4742 cells at log phase were diluted into YEP containing either 2% or 518 0.5% glucose as indicated. OD600 was measured every hour for 12 hours. 519 520 521 522 .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.23.604825doi: bioRxiv preprint .CC-BY-NC-ND 4.0 International licenseavailable under a was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprint (whichthis version posted July 25, 2024. ; https://doi.org/10.1101/2024.07.23.604825doi: bioRxiv preprint CWVph1-GFPCWVph1-GFP YoungOld 0.0 0.5 1.0 1.5Vma2-GFP Normalized Max. Fluor. ✱✱✱ YoungOld 0.0 0.5 1.0 1.5Vph1-GFP Normalized Max. Fluor. ns c.YoungOld d. Young Old 0.0 0.5 1.0 1.5 Vma1 Relative Protein Level ns Young Old 0.0 0.5 1.0 1.5 Vma2 Relative Protein Level ns Young Old 0.0 0.5 1.0 1.5 Vma5 Relative Protein Level ns Vma2 Vma1 Vma5 GAPDH 72- 55-40- 35- YO 0-45-910-1415-1920-2425+0-45-910-1415-1920-2425+ 0 20 40 60 80 Age Distribution Post-Enrichment Number of Bud Scars Number of Cellls Young Old e. a. f. Figure 2 CWVma2-GFPYoungOld CWVma2-GFPb. a. 0 5 10 5.2 5.4 5.6 5.8 6.0 Time (min) Vacuolar pH Young Old Glucose b. 1 min 5 min 8 min 5.0 5.5 6.0 6.5 Vacuolar pH Young Old ✱ ✱ ✱✱ Figure 3 0 5 10 5.4 5.6 5.8 6.0 6.2 Time (min) Vacuolar pH Young Old Vma5-GFP Glucose YoungOld 0.0 0.5 1.0 1.5 CR Vma5-GFP Normalized Max. Fluor. ns YoungOld 0.0 0.5 1.0 1.5 CR Vph1-GFP Normalized Max. Fluor. ns YoungOld 0.0 0.5 1.0 1.5 CR Vma2-GFP Normalized Max. Fluor. ns a. d. 1 min 5 min 8 min 5.0 5.5 6.0 6.5Vacuolar pH Young Old ns ns ns Figure 4 b. a. Figure 5 0 12 24 36 48 0 50 100 Generation % Survival rav1Δ (17, -26.1%) WT (23) oxr1Δ (34, +47.8%) 0-45-910-1415-2021-2425+ 0.0 0.5 1.0 1.5 oxr1Δ Normalized Max. Fluor. ✱✱✱ c. f. Oxr1-HAGAPDH 40-35-YO Young Old0.0 0.5 1.0 1.5Oxr1-HA Relative Protein Level ns YoungOld0.0 0.2 0.4 0.6 0.8 1.0 RAV2 mRNA Normalized Fold Change (log2) Expression ns .a. Rav1-myc GAPDH YO 150-35- Rav2-myc GAPDH YO55-35-Young Old0.0 0.5 1.0 1.5 Rav1-myc Relative Protein Level ns YoungOld0.0 0.5 1.0 1.5 Rav2-myc Relative Protein Level ✱✱b. Rav2-myc GAPDH 55-35-YOYO2% Glucose0.5% Glucose YoungOldYoung Old0.0 0.5 1.0 1.5 CR Rav2-myc Relative Protein Levels ✱✱✱ns 2% 0.5% g. oxr1Δ WT DIC CWVma5-GFP Figure 6 d. e. [glucose] Rav2-myc