ACE-22-0212 Proof Hi

Aging Cell
Riboflavin Depletion Promotes Longevity and Metabolic

Hormesis in Caenorhabditis elegans
Journal: Aging Cell
Manuscript ID ACE-22-0212
Wiley - Manuscript type: Research Article

Fo
Date Submitted by the
29-Mar-2022
Author:
Complete List of Authors: Yerevanian, Armen; Massachusetts General Hospital, Center for Genomic
rP
Medicine; Harvard Medical School

Murphy, Luke; Massachusetts General Hospital, Center for Genomic
Medicine
Emans, Sinclair; Massachusetts General Hospital, Center for Genomic
ee
Medicine
Zhou, Yifei; Massachusetts General Hospital, Center for Genomic
Medicine
rR
Ahsan, Fasih; Massachusetts General Hospital, Center for Genomic

Medicine
Baker, Daniel; Massachusetts General Hospital, Center for Genomic
Medicine
ev
Li, Sainan; Massachusetts General Hospital, Center for Genomic Medicine

Adedoja, Adebanjo; Massachusetts General Hospital, Center for Genomic
Medicine
Cedillo, Lucydalila; Massachusetts General Hospital, Center for Genomic
iew
Medicine
Gnanatheepam, Einstein; Tufts University School of Engineering,
Department of Biomedical Engineering
Dao, Khoi; University of California San Diego, Medicine and
Pharmacology
Jain, Mohit; University of California San Diego, Medicine and
Pharmacology
Georgakoudi, Irene; Tufts University School of Engineering, Department
of Biomedical Engineering
Soukas, Alexander; Massachusetts General Hospital, Center for Genomic
Medicine; Harvard Medical School; Broad Institute
Riboflavin, longevity, rft-1, AMPK, FOXO, dietary restriction, C. elegans,

Keywords:
FAD
ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901 Support: (434) 964 4100
Page 1 of 35 Aging Cell
Riboflavin Depletion Promotes Longevity and Metabolic Hormesis in Caenorhabditis elegans
Armen Yerevanian1,2, Luke Murphy1,, Sinclair Emans1, Yifei Zhou1, Fasih Ahsan1 Daniel Baker1, Sainan Li1,
Adebanjo Adedoja1, Lucydalila Cedillo1, Einstein Gnanatheepam4, Khoi Dao5, Mohit Jain5, Irene Georgakoudi4,
Alexander Soukas1,2,3
1Department of Medicine, Diabetes Unit and Center for Genomic Medicine, Massachusetts General Hospital,
Boston, MA 02114, USA

2Department of Medicine, Harvard Medical School, Boston, MA 02115, USA
3Broad Institute of Harvard and MIT, Cambridge, MA 02142, USA
4Department of Biomedical Engineering, Tufts University School of Engineering, Medford, MA
5Department of Medicine and Pharmacology, University of California San Diego, San Diego CA 92023
Correspondence: asoukas@mgh.harvard.edu
Keywords:
Riboflavin, FAD, dietary restriction, longevity, rft-1, AMPK, FOXO, C. elegans
Fo
rP
ee
rR
ev
iew
1
Aging Cell Page 2 of 35
AGING CELL AUTHOR CHECKLIST

Title: Riboflavin Depletion Promotes Longevity and Metabolic Hormesis in Caenorhabditis elegans
Authors: Armen Yerevanian, Luke Murphy, Sinclair Emans, Yifei Zhou, Fasih Ahsan, Daniel Baker, Sainan Li,
Adebanjo Adedoja, Lucydalila Cedillo, Einstein Gnanatheepam, Khoi Dao, Mohit Jain, Irene Georgakoudi,
Alexander Soukas
Manuscript Type: Research Article
Total Character Count (including spaces): 49,969
Word count of Summary: 122
Number of papers cited in the References: 50
Listing of all Tables (Table1, Table 2 etc): N/A
Figure specifications:
Fo
Figure No Color Single Double Size of Figure Smallest Font
Column Column
1 Yes Yes 21 x 12 6
rP
2 Yes Yes 16 x 15.5 6

3 Yes Yes 27 x 33 6
4 Yes Yes 22.5 x 16 6
ee
5 Yes Yes 37 x 22.5 6

6 Yes Yes 28 x 21.5 6
rR
ev
iew
2
1 Summary
2 Riboflavin is an essential cofactor in many enzymatic processes and in the production of flavin adenine
3 dinucleotide (FAD). Here we report that the partial depletion of riboflavin through knockdown of the C. elegans
4 riboflavin transporter 1 (rft-1) promotes metabolic health by reducing intracellular flavin concentrations.
5 Knockdown of rft-1 significantly increases lifespan in a manner dependent on FOXO/daf-16, AMP-activated
6 protein kinase (AMPK)/aak-2, the mitochondrial unfolded protein response, and mTOR complex 2 (mTORC2).
7 Riboflavin depletion promotes altered energetic and redox states and increases adiposity, independent of
8 lifespan genetic dependencies. Riboflavin depleted animals also exhibit activation of caloric restriction reporters
9 without any reduction in caloric intake. Our findings indicate that riboflavin depletion activates an integrated,
10 hormetic response that promotes lifespan and healthspan in C. elegans.
11
12
13
14
Fo
rP
ee
rR
ev
iew
3
15 Introduction
16 Healthy mitochondrial function requires the coordination of multiple cellular inputs including sufficient energetic
17 substrates, amino acids and micronutrients. Vitamin cofactors are key to metabolic processes such as the citric
18 acid cycle, electron transport chain physiology, and energy shuttling into the cytosol. One family of vitamins that
19 participate in mitochondrial physiology are the flavins (Mansoorabadi, Thibodeaux et al. 2007). The flavin co-
20 factors include flavin mononucleotide (FMN) and flavin adenine dinucleotide (FAD) and are essential for redox
21 chemistry and electron shuttling (Massey 1995). FAD is classically known to serve as an electron acceptor in
22 the conversion of succinate to fumarate by succinate dehydrogenase in the citric acid cycle, as well as an
23 electron donor to complex II of the electron transport chain. The flavins are also cofactors for multiple enzyme
24 classes including the oxidoreductases and the fatty acid dehydrogenases (Lienhart, Gudipati et al. 2013). They
25 are derived from riboflavin, a water soluble ribitol derivative also known as vitamin B2. Riboflavin is an essential
26 nutrient for all animals that must be acquired either from food sources or from commensal gut flora (Powers
27 2003).
28 The animal kingdom has evolved specific transporters to import riboflavin from the gut lumen and to transport
29 them intracellularly. (Moriyama 2011) In humans and mice, three isoforms of these transporters are expressed
30 by three distinct genes: Slc52A1, Slc52A2 and Slc52A3 (Yonezawa, Masuda et al. 2008, Yamamoto, Inoue et
31 al. 2009, Subramanian, Subramanya et al. 2011). Disruption in the function of these transporters is known to
32 induce pathology through cellular riboflavin depletion (Nabokina, Subramanian et al. 2012). Congenital
Fo
33 deficiency in these transporters is associated with clinical syndromes in humans including Brown-Vialetto-Van
34 Laere syndrome, where patients experience progressive neurologic deficits, and is treated with extremely large
35 doses of riboflavin (Dipti, Childs et al. 2005, Spagnoli and De Sousa 2012). Nutritional riboflavin deficiency
rP
36 (ariboflavinosis) is a very rare but also described clinical disorder which is associated with dermatologic and
37 hematologic manifestations (Sydenstricker 1941).
ee
38 Despite the clinical and epidemiologic data describing syndromes of riboflavin deficiency, the metabolic impacts
39 of riboflavin depletion at a cellular level are not well described. Flavin deficiency is thought to alter fatty acid
40 metabolism through disrupted beta-oxidation as well as reduced flux through the citric acid cycle due to a lack
rR
41 of FAD. We hypothesized that these disruptions may lead to the activation of signaling pathways that are
42 associated with mitochondrial and energetic stress. We further hypothesized that partial depletion of riboflavin
43 will have beneficial metabolic effects by activating cellular responses associated with caloric restriction and
ev
44 longevity.
45 In the present study, we determine that physiologic riboflavin depletion alters cellular energetics and activates
iew
46 key longevity factors such as AMPK, FOXO, and the mitochondrial unfolded protein response (UPRmt). We report
47 that riboflavin depletion via riboflavin transporter knockdown extends lifespan and promotes healthspan in C.
48 elegans, suggesting that riboflavin depletion provides metabolic benefits and mimics favorable features of caloric
49 restriction without actual reductions in caloric intake.
50
4
51 Results
52 Knockdown of rft-1 Promotes Longevity via Riboflavin Depletion
53 We elected to approach the hypothesis that partial riboflavin depletion may provide metabolic benefits by
54 performing functional genomics in C. elegans. The nematode genome encodes two riboflavin transporter
55 orthologs, rft-1 and rft-2 (Biswas, Elmatari et al. 2013). rft-1 exhibits strong intestinal expression, so we first
56 examined consequences of its knockdown (Gandhimathi, Karthi et al. 2015). rft-1 depletion via RNA interference
57 (RNAi) prompts a significant, 25% increase in lifespan (Figure 1a). The RNAi knockdown of rft-1 reduces the
58 transporter’s mRNA by approximately 70%, suggesting that partial riboflavin transporter deficiency rather than
59 complete knockdown induces this phenotype, as rft-1 mutants are not viable (Figure 1b). mRNA encoding the
60 paralogous riboflavin transporter rft-2 trends non-significantly upwards with rft-1 RNAi, which indicates that there
61 is not significant compensation for rft-1 knockdown by rft-2, and confirms the specificity of the RNAi-based
62 knockdown of rft-1 (Figure 1b). To confirm that the longevity phenotype of the rft-1 knockdown animals is due
63 to reduced riboflavin uptake rather than a non-canonical effect of the transporter, we administered high dose
64 riboflavin supplementation to attempt to overcome the deficit in transport. As expected, high doses of riboflavin
65 abrogate the lifespan increase attributable to rft-1 RNAi (Figure 1a). This strongly suggests that riboflavin is the
66 etiologic factor in the rft-1 RNAi phenotype, and further that depletion of riboflavin due to transporter deficiency
67 is the source of lifespan extension. Worm lysates of young adult worms treated with rft-1 RNAi exhibit marked
68 reductions in riboflavin, FMN and FAD levels as assessed by quantitative liquid chromatography/mass
69 spectrometry (LC/MS) (Figure 1c). In parallel, we evaluated the expression of enzymes key to flavin co-enzyme
Fo
70 synthesis including riboflavin kinase (rfk-1) and FAD synthetase (flad-1). rft-1 RNAi was not associated with
71 reductions in expression of flad-1 and rfk-1 which produce FAD and FMN respectively, suggesting stoichiometric
72 depletion of FMN/FAD rather than reductions in the enzymes that govern production (Figure S1a).
rP
73
74 Previous descriptions of brood size deficits in rft-1 knockdown animals suggests that the germline may play a
75 role in the phenotype(Biswas, Elmatari et al. 2013, Qi, Kniazeva et al. 2017, Yen, Ruter et al. 2020). We utilized
ee
76 rrf-1 (somatic RNAi blunted, germline RNAi competent) and ppw-1 (somatic RNAi competent, germline RNAi
77 incompetent) mutants and examined brood size on rft-1 RNAi. The loss of brood size is dependent on somatic
78 action of rft-1 (i.e. normalized in rrf-1 mutants which do not efficiently conduct somatic RNAi). This suggests that
rR
79 a global process is altering metabolic and reproductive capacity (Figure 1d). The reduction in brood size raised
80 the question of whether germ line stem cells and oocyte production is altered by riboflavin depletion. We
81 examined the presence of the germline stem cells and oocytes via DAPI staining, which revealed an intact
ev
82 germline and oocyte production (Figure S1b). Riboflavin deficiency is associated with neurologic sequelae in
83 mammals, and we wanted to verify that feeding behavior is unchanged compared to control animals(Qi, Kniazeva
84 et al. 2017). Knockdown of rft-1 does not affect eating behavior as measured by the pharyngeal pumping rate
iew
85 (Figure 1e). We also sought to evaluate whether there was a change in developmental rate as a potential source
86 for this phenotype. RNAi to rft-1 does not significantly change the rate of development to adulthood. RNAi of rft-1
87 in the RNA interference sensitive strain eri-1 also does not alter developmental rate (Figure 1f). These findings
88 suggest that animals treated with rft-1 knockdown exhibit normal development and behavior, and that nutritional
89 depletion of riboflavin does not preclude them from reaching adulthood.
90
91 Riboflavin Depletion Promotes Longevity Through FOXO
92 FOXO is known to act genetically downstream of nutrient-sending manipulations that extend lifespan (Lee,
93 Kennedy et al. 2003, Greer, Dowlatshahi et al. 2007) and we hypothesized that it may be activated in the context
94 of riboflavin depletion. Indeed, daf-16 loss of function is epistatic to lifespan extension with rft-1 RNAi, with or
95 without the presence of supplemental riboflavin (Figure 2a). A DAF-16::GFP translational reporter demonstrates
96 greater nuclear localization at adult day 1 and day 3 with riboflavin depletion, suggesting that rft-1 RNAi activates
97 DAF-16 (Figure 2b and S2a). Transgenic DAF-16::GFP worms treated with empty vector and rft-1 RNAi as
98 synchronous L1 were subsequently transferred at adult day 1 to plates with and without riboflavin. By Adult Day
99 3, additional riboflavin completely abrogates the DAF-16 nuclear localization evident in in rft-1 RNAi treated
100 animals (Figure S2b). This indicates that the consequences of riboflavin depletion on daf-16 nuclear localization
101 are reversible in adulthood. Confirming increased transcriptional activity of DAF-16 in the setting of riboflavin
102 depletion, RNAi of rft-1 led to the upregulation of a sod-3p::GFP reporter (Figure 2c). This activation was present
103 through adult day 7, indicating consistent FOXO activation post-developmentally (Figure S2c).
104
5
105 The activation of FOXO suggests either suppression of insulin-like/PI-3 kinase signaling (canonical FOXO
106 activation) or activation via another mechanism. We examined whether insulin signaling was playing a role by
107 examining the effect of rft-1 RNAi on akt-1 and pdk-1 gain of function mutants (Paradis and Ruvkun 1998). These
108 animals are short lived due to constitutive inhibition of FOXO. The akt-1 gain-of-function mutation abrogates
109 lifespan extension attributable to rft-1 RNAi (Figure 2d). The pdk-1 gain-of-function mutant on rft-1 RNAi,
110 conversely, still exhibits lifespan extension (Figure S2d). This indicates that suppression of DAF-16 via Akt
111 abrogates lifespan extension prompted by riboflavin deficiency.
112
113 Riboflavin Depletion Alters Cellular Redox Ratio and Energetics
114 We suspected that AMPK may be similarly activated by energetic stress with riboflavin deficiency, and further
115 that AMPK activation may be mechanistically linked to lifespan extension with rft-1 RNAi. Knockdown of rft-1
116 fails to promote longevity with loss of function in the AMPK catalytic subunit aak-2 (Figure 3a), and this pattern
117 is not altered by addition of riboflavin (Figure 3b). Consistent with AMPK activation by riboflavin depletion,
118 phospho-AMPKT172 levels are increased in rft-1 RNAi treated animals, and this effect is abrogated by the addition
119 of riboflavin (Figure 3c). In order to determine whether AMPK is required for activation of DAF-16 under riboflavin
120 depletion, we examined DAF-16::GFP nuclear localization with and without functional aak-2. Under rft-1 RNAi
121 conditions, DAF-16 nuclear localization still increases in the absence of aak-2 (Figure S3a). This could suggest
122 that AMPK and FOXO are each required in parallel to promote lifespan downstream of riboflavin deficiency.
123
124
Fo
Activation of AMPK suggests that modulation of cellular energetics might play a role in the longevity phenotype
125 seen with the rft-1 knockdown. We hypothesized that reductions in flavin cofactor (FAD, FMN) concentrations
126 induce mitochondrial stress responses due to changes in organellar energetics by altering redox state. We
127 examined the impact of riboflavin depletion on the redox ratio utilizing label-free multiphoton microscopy and
rP
128 fluorescence lifetime imaging (FLIM) of intestinal cells in control and rft-1 RNAi treated animals (Figure Sb-c).
129 Animals treated with rft-1 RNAi versus empty vector control-treated animals exhibit a significant decrease in the
130 optical redox ratio, defined as the ratio of FAD/(NAD(P)H + FAD) calculated based on the autofluorescence
ee
131 signatures of the corresponding co-enzymes. There is also an increase in mitochondrial clustering, suggesting
132 altered mitochondrial energetics in an oxidized state and morphologic changes to the mitochondrial network
133 (Figure 3d)(Liu, Pouli et al. 2018). A significant decrease in the NAD(P)H protein bound fraction suggests
rR
134 decreased levels of glutaminolysis and enhanced utilization of the glutathione pathway (Alonzo, Karaliota et al.
135 2016). An increase in the FAD bound fraction suggested that overall FAD depletion was causing aggressive
136 capture of flavin co-factors by enzymatic machinery (Figure 3d). LC/MS metabolomics of rft-1 RNAi treated
137 animals indicates significant changes in multiple metabolic pathways. Increases in purine catabolism metabolites
ev
138 were present, including xanthine, hypoxanthine, and guanosine (Figure 3e). Pathway enrichment analysis
139 reveals that other than riboflavin metabolism, riboflavin deficiency leads to significant impact on metabolite
140 changes in the glutathione and purine metabolism pathways (Figure S3d). Components of the citric acid cycle
iew
141 including citrate, isocitrate, and α-ketoglutarate, as well as ATP, were reduced by riboflavin depletion. Glutamate
142 and glutamine levels were also reduced suggestive of disruptions in glutamine synthesis (Figure 3f).
143
144 Riboflavin Depletion Activates the Mitochondrial Unfolded Protein Response
145 The frank changes in energetic status, altered redox ratio, and presence of mitochondrial clustering all suggested
146 that mitochondrial stress responses may also be contributing to the longevity response to riboflavin deficiency.
147 Indeed, the UPRmt is activated by rft-1 knockdown, as evidenced by induction of hsp-6p::GFP (Labbadia,
148 Brielmann et al. 2017) induction on days 1 and 3 of adulthood, and this effect is abrogated by the addition of
149 riboflavin (Figure 4a). Full activation of the UPRmt is known to require the transcription factor ATFS-1, which
150 translocates to the nucleus to activate stress response pathways (Wu, Senchuk et al. 2018). Established target
151 genes of ATFS-1, including cdr-2, hrg-9, and C07G1.7 (Soo and Van Raamsdonk 2021) are upregulated with
152 rft-1 knockdown, with the previously undescribed gene C07G1.7 exhibiting a 2000-fold increase (Figure 4b).
153 The UPRmt activation is necessary for lifespan extension, as atfs-1 loss of function animals, which have lower
154 lifespans compared to wild-type at baseline, do not exhibit lifespan extension with rft-1 knockdown (Figure 4c).
155 In the setting of gain-of-function mutations in atfs-1, which lead to shortened lifespan (Bennett, Vander Wende
156 et al. 2014), rft-1 knockdown still promotes significant extension of lifespan (Figure 4d). This indicates that the
157 UPRmt response is necessary but not sufficient for the riboflavin depletion longevity phenotype.
158
159 Riboflavin Depletion Alters Somatic Lipid Stores
6
160 The long-lived phenotype of riboflavin depletion and the role of flavin cofactors in beta-oxidation suggests that
161 changes in lipid composition may be manifest following rft-1 knockdown. We hypothesized that changes in lipid
162 metabolism occur upstream of or in parallel to FOXO activation, due to changes in enzymatic function (such as
163 reduced activity of lipid dehydrogenases). RNAi of rft-1 induces significant increases in fat mass in both the
164 intestine and the germline in adult day 1 worms, as exhibited by fixative oil-red-O and Nile red staining (Figure
165 5a, S5a) Epistasis analysis indicates that rft-1 RNAi increases fat mass in daf-16, aak-2 and raga-1 animals but
166 not in atfs-1 or rict-1 animals by fixative-based Nile red staining (Figure 5a). Post-developmental rft-1 RNAi
167 beginning at young adult stage in enhanced RNAi eri-1 animals also increases fat mass, indicating that riboflavin
168 deficiency does not impact lipid metabolism exclusively through a developmental pleiotropy (Figure 5a).
169 Confirming these observations and further delineating the nature of the lipids increased in abundance following
170 rft-1 RNAi, stimulated Raman scattering (SRS) analysis of live adult day 1 animals indicates increased total
171 signal of unsaturated fatty acids and the unsaturated to total lipid ratio in riboflavin deficiency (Figure S5b)
172 (Nieva, Marro et al. 2012, Potcoava, Futia et al. 2014). By gas chromatography/mass spectroscopy (GC/MS) of
173 triglyceride and phospholipid fractions separated by solid phase extraction, global triglyceride stores increase by
174 40% in both young adult and adult day 1 rft-1 RNAi-treated animals, consistent with the spectroscopic imaging
175 and fixative-based lipid staining (Figure 5b). While only small changes are evident by young adulthood (Figure
176 S5c), by adult day 1 animals exhibit significant differences in their lipid composition, with increases in unsaturated
177 and branched chain fatty acids, and reductions in cyclopropyl fatty acids in both triglyceride and phospholipid
178 fractions (Figure 5c, S5c). Due to increases in branched chain fatty acid synthesis, we examined the expression
Fo
179 of acdh-1, which is a known branched chain dehydrogenase in elegans and that has been previously reported
180 as a dietary sensor (Watson, MacNeil et al. 2013). An acdh-1 promoter GFP reporter is significantly increased
181 ~70% with rft-1 RNAi at adult day 1 (Figure 5d). In order to begin to determine whether unsaturated fatty acids
rP
182 are elevated in riboflavin deficiency owing to increased production vs utilization, we examined expression of the
183 fatty acid desaturase fat-7 (Han, Schroeder et al. 2017). Riboflavin depletion does not promote changes in fat-7
184 expression early in life but preserves it with aging (Figure S5d).
ee
185
186 Riboflavin Depletion Activates Dietary Restriction Pathways
187 The long-lived phenotype of riboflavin depletion, concomitant with decreases in energetics, AMPK activation,
rR
188 and impairment in lipid beta-oxidation, suggested to us that riboflavin depletion resembles a dietary restriction-
189 like phenotype. The acs-2 and bigr-1 genes have been well established to be transcriptionally upregulated during
190 periods of caloric restriction in C. elegans (Van Gilst, Hadjivassiliou et al. 2005, Wu, Zhou et al. 2016). rft-1 RNAi
ev
191 induced bigr-1::RFP and acs-2p::GFP expression with age (Figure 6a). We sought to assess whether other
192 canonical caloric restriction factors were involved in riboflavin depletion, particularly the C. elegans FoxA
193 homolog pha-4 (Panowski, Wolff et al. 2007). Consistent with our hypothesis, lifespan extension with rft-1 RNAi
iew
194 is dependent on pha-4/FoxA (Figure 6b). Inhibition of target of rapamycin (TOR) signaling is also important in
195 the response to dietary restriction. Thus, we examined whether mutants in the TOR complex 1 (TORC1) and
196 TOR complex 2 (TORC2) pathways exhibit longevity with riboflavin depletion. RAGA/raga-1 mutants experience
197 lifespan extension on rft-1 RNAi (Figure 6c). To further determine whether altered TORC1 activity is required
198 for the hormetic effects of riboflavin depletion, we used a strain of elegans that contains a knock-in, humanized
199 S6K, which permits immunoblotting for phospho-S6K to determine the activity of TORC1. No difference is evident
200 in p-S6K between rft-1 RNAi and empty vector control, suggesting that altered TORC1 signaling is not essential
201 to the biological response to riboflavin depletion (Figure 6d). In contrast, loss of function mutations in the
202 essential TORC2 subunit rict-1 experience no lifespan extension with rft-1 RNAi, indicating that riboflavin
203 depletion requires TORC2 activity to exact its favorable effects (Figure 6e).
204
205 Discussion
206 Vitamins, as essential co-factors for life, have traditionally been viewed as highly beneficial entities independent
207 of their concentrations. This is particularly true of the B-class vitamins which are water soluble and do not exhibit
208 significant toxicities at moderate supraphysiologic doses. Our work in C. elegans counters the notion that more
209 is always better, as depletion of key enzymatic co-factor riboflavin can trigger metabolic and physiologic stress
210 responses that are hormetic in nature and extend lifespan.
211
212 We were initially concerned that the reduction of the flavin cofactors such as FAD and FMN would be frankly
213 toxic to the organism. This was particularly true when LC/MS revealed that FAD levels in the rft-1 treated animals
7
214 were 80-90% below normal. Alternatively, it is clear from data presented here that this level of reduction promotes
215 energetic perturbations most consistent with disruptions in mitochondrial respiration. Unexpectedly, however,
216 the riboflavin depletion phenotype is different than that of classic electron transport chain (ETC) disruption, such
217 as in cco-1 and frh-1 (frataxin) knockdowns (Durieux, Wolff et al. 2011, Schiavi, Torgovnick et al. 2013). Previous
218 examinations of cco-1 and frh-1 RNAi have shown that lifespan extension with ETC disruption is AMPK and
219 FOXO independent (Durieux, Wolff et al. 2011, Schiavi, Torgovnick et al. 2013). Unlike traditional ETC disruption,
220 riboflavin depletion’s lifespan extending properties depend on these key energetic sensing pathways. We
221 anticipated that intestinal FOXO activation would depend on increased AMPK activity, however nuclear
222 localization of DAF-16 still occurs in AMPK mutants, indicating AMPK-independent governance of FOXO activity.
223 Alternatively, FOXO activation may be dependent upon inhibition of akt and/or insulin-like signaling as gain-of-
224 function mutations in akt-1 also suppress lifespan prolongation by riboflavin deficiency. Why this phenotype is
225 not shared with pdk-1 gain-of-function mutations could be because of the relative strength of the latter, or
226 complexities of riboflavin deficiency on cellular signaling. Determining between these possibilities will requires
227 additional testing. Finally, additional requirements of the UPRmt and potentially TORC2 for lifespan extension
228 suggest that these pathways function as parallel, critical effectors to achieve a concerted cellular response to
229 the reductions of FAD and FMN (Figure 6f).
230
231 Riboflavin depleted animals exhibit normal food consumption (pharyngeal pumping rate), normal TORC1
232 activation, and elevated triglyceride stores, suggesting that the phenotype is not due to a reduction in
Fo
233 macronutrient availability, but may be due to defective nutrient mobilization and utilization. The dependencies
234 on FOXO/daf-16, FOXA/pha-4, and AMPK/aak-2, as well as activation of canonical “starvation” reporters acs-2
235 and bigr-1, suggest that riboflavin depleted animals experience a caloric restriction-like phenotype with
rP
236 micronutrient depletion only. Amino acid sensing and dietary restriction via essential nutrients such as
237 methionine have been previously described to extend lifespan (Cabreiro, Au et al. 2013). Depletion of canonical
238 vitamin co-factors however, has proven deleterious in previous investigations. Depletion of biotin, B12
ee
239 derivatives, and folate have previously shown to shorten lifespan (Austin, Liau et al. 2010, Bito, Matsunaga et
240 al. 2013). To our knowledge, our work is the first to show that depletion of a vitamin co-factor can mimic features
241 of dietary restriction and extend lifespan using shared molecular mechanisms.
rR
242
243 Elevated triglyceride stores following riboflavin depletion is independent of canonical lifespan regulating
244 pathways such as FOXO, AMPK and TORC1. This decoupling of fat mass and lifespan suggest that the lipid
245 phenotype may be regulated by enzymatic processing of lipids upstream of the energetic stress axes. The
ev
246 exceptions to this were the atfs-1 and rict-1 mutants. Phenotypes associated with UPRmt activation are known to
247 induce lipid accumulation (Kim, Grant et al. 2016, Yang, Li et al. 2022). Recent work has identified NHR-80 as a
iew
248 key regulator of citrate sensing and lipid accumulation in the UPRmt phenotype (Yang, Li et al. 2022). The
249 relevance of atfs-1 activity to dve-1 and ubl-5 function suggests that the UPRmt may be instrumental in the
250 communication of flavin depletion and related citric acid cycle disruptions on organismal energetics. The lack of
251 fat mass increase in rict-1 mutants, which at baseline exhibit higher lipid content, suggests either a dependency
252 or inability for riboflavin depletion to overcome the excess lipid accumulation associated with TORC2 knockout.
253 TORC2 has been well described as a nutritional sensor that regulates lipid biogenesis (Soukas, Kane et al.
254 2009), and it is entirely plausible that there are distinct inputs in mitochondrial energetics, mito-stress and TORC2
255 activation that are governed by flavin biology. Further investigation would be beneficial to identify whether
256 TORC2 can directly sense changes in flavin levels, as this would have significant implications for the nutritional
257 regulation of anabolic signals in senescence and cancer.
258
259 rft-1 RNAi does not impact developmental rate and metabolic phenotypes manifest most impressively at the
260 young adult to adult day 1 transition. This is in parallel to the growth and development of the germline and the
261 oocytes. This pattern suggests either that the larval stages are relatively resistant to riboflavin depletion, or, more
262 likely, contain and accumulate sufficient flavin cofactors at time of egg lay and during early larval development
263 (prior to rft-1 knockdown) to proceed through development normally. We surmise that somatic growth dilutes
264 endogenous flavin cofactors, subsequently inducing the favorable effects of riboflavin deficiency uniquely in
265 adulthood. Further, development of the germline and riboflavin shunting into oocytes in late larval and early adult
266 stages may prompt further, rapid riboflavin depletion, inducing the metabolic stress required to induce the
267 phenotype identified in this work. It is worthy of mention that riboflavin deficiency leads to sterility, but this sterility
8
268 is not accompanied by a decrease in germline stem cell numbers. Thus, effects on the germline are unlikely to
269 be responsible for the shifts in lifespan and fat mass evident in riboflavin deficiency.
270
271 The presence of a post-developmental fat increase with depletion of riboflavin suggests that acute depletion in
272 adulthood has important impacts that are likely different from depletion starting at larval stages. The most likely
273 etiology for these post-developmental changes are alterations in enzymatic activity due to loss of flavin co-
274 factors. The flavin co-factors are important for a wide variety of enzymes particularly those associated with oxido-
275 reductase functions including the fatty acid dehydrogenases. The ‘flavoproteome’ is an established set of
276 enzymes requiring FAD, FMN or riboflavin to function (Lienhart, Gudipati et al. 2013). The impact of riboflavin
277 depletion globally on the proteome is likely to produce stoichiometric shifts in key metabolites that will alter global
278 physiology. Differential utilization of different dehydrogenases (branched chain vs long chain) may also explain
279 the unique lipid phenotype that is produced with riboflavin depletion. Using metabolomics, we identified other
280 examples of likely enzymatic effects, with evidence of reductions in purine catabolism likely due to loss-of-
281 function in xanthine dehydrogenase. Alterations in xanthine metabolism have been previously described as
282 beneficial and lifespan extending (Gioran, Piazzesi et al. 2019). The impact of riboflavin depletion on enzymatic
283 processes is complex and there are likely to be both hormetic and harmful impacts of this process. Identifying
284 the enzymatic pathways where riboflavin depletion provides beneficial versus detrimental impacts will provide
285 new insights into mechanistic targets promoting longevity. We suggest that further investigations into the
286 functions of the flavoproteome and flavin biology will serve to identify new therapeutic and investigational targets
Fo
287 for the metabolism of aging and aging associated diseases.
288
289
rP
ee
rR
ev
iew
9
290 Experimental Methods

291 C. elegans genetics
292 Strains were maintained at 20oC on nematode growth medium (NGM) plates seeded with E. coli OP50. All
293 experiments were conducted at 20oC unless otherwise specified. The following strains were utilized: wild type
294 (N2 Bristol ancestral), NL3511 ppw-1(pk1425), NL2098 rrf-1(pk1417), daf-16(mgDF47), TJ356 zls356[daf-
295 16p::daf-16a/bGFP+rol-6(su1006), CF1553 muls84[(pAD76)sod-3p::GFP+rol-6(su1006)], GR1318 pdk-
296 1(mg142gf), GR1310 akt-1(mg144gf), RB754 aak-2(ok524), VC3201 atfs-1(gk3094), QC118 atfs-1(et18),
297 SJ4100 zcls13[hsp-6p::GFP+lin-15(+)], DMS303 nls590[fat-7p::fat-7::GFP +lin15(+)], VL749 wwls24[acdh-
298 1p::GFP +unc-119(+)] MGH266 rict-1(mg451), VC222 raga-1 (ok386), MGH559 aak-2(ok754);zls356[daf-
299 16p::daf-16a/b::GFP+rol-6(su1006)], MGH249 alxls19 [bigr-1::bigr-1::mRFP3-HA;myo-2p::GFP], WBM392
300 Is(Pacs-2::GFP+rol-6(su1006)), MGH430 rsks-1(alx48 humanized S6K hydrophobic motif).
301
302 E. coli Strains
303 Non-RNAi experiments were all conducted on NGM plates containing E. coli OP50-1 (CGC) as the food source
304 and used 3-7 days after seeding. Cultures of E. coli OP50 were grown in Luria Broth (LB) for 15 hrs. at 37oC
305 without shaking and seeded directly onto NGM plates. RNA interference experiments were conducted using E.
306 coli HT115(DE3) bacteria (Ahringer library) as the food source. Clones were isolated from the primary RNAi
307 library and plated on ampicillin/tetracycline plates. Individual clones were grown in LB broth for 15 hours with
Fo
308 shaking. Cultures were concentrated 1:5 and seeded directly onto NGM plates containing. 5 mM isopropyl-B-D-
309 thiogalactopyranoside and 200 mg/ml carbenicillin. Plates were used 1-5 days after seeding. All RNAi clones
310 were sequence verified prior to utilization. Riboflavin solution or vehicle was applied to the plate and allowed to
rP
311 dry for at least 30 minutes prior to seeding with animals.
312
313 Riboflavin Treatment
314 Culture grade riboflavin (Sigma-Aldrich) was dissolved in 50mM NaOH to a concentration of 13.3 mM (5 mg/ml).
ee
315 Fully seeded plates were treated with 500ul riboflavin solution (final concentration 665 µM) and allowed to dry
316 on the plate prior to worm placement. Vehicle plates were treated with 500ul 50mM NaOH solution.
317
rR
318 Longevity assays

319 Lifespan analysis was conducted at 20oC except where indicated. Gravid adults were grown on NGM plates and
320 isolated eggs were incubated overnight in M9 solution for hatching. Synchronized L1 animals were seeded unto
ev
321 RNAi plates and allowed to grow until the adult stage. Adult animals were subsequently transferred to fresh RNAi
322 plates every other day until post-reproductive stage where they were maintained on a single plate. Dead worms
323 were counted daily or every other day. Statistical analysis for survival curves was performed using OASIS2
iew
324 software (Han, Lee et al. 2016).

325
326 Development Assays
327 Synchronized L1s were prepared via bleach prep and plated on RNAi plates containing empty vector or rft-1
328 RNAi grown at 20oC. Larvae were examined every 2 hours starting 41 hours after drop and scored for their
329 transition to adulthood by the appearance of the vulvar slit.
330
331 Brood Size
332 Fifty synchronized L1 animals of each strain were dropped on EV and rft-1 RNAi treated plates. Two days later,
333 2 young adult animals from each condition were dropped onto new EV and rft-1 RNAi plates respectively. These
334 animals were transferred every 2 days until the two adults from each condition became post-reproductive. All
335 animals on residual plates were counted once they reached L4/YA to calculate brood size.
336
337 Pharyngeal Pumping
338 Pumping rate was determined using a Sony camera attached to a Nikon SMZ1500 microscope that recorded 10
339 well fed Day 1 adult animals per genotype. Pharyngeal contractions in 15 second time periods for 4 technical
340 replicates were counted (by slowing video playback speed by 4x) for each animal using OpenShot and pumping
341 rates per minute were calculated.
342
343
10
344
345 Oil-Red-O and Nile Red Staining
346 Lipid staining protocol was adapted from Escorcia et al (Escorcia, Ruter et al. 2018). Adult day 1 animals were
347 collected via washing and washed twice with M9 solution. Animals were then fixed with 40% isopropanol for 3
348 minutes with shaking. For oil-red-O staining, working solution of oil-red-O was generated from stock solution and
349 fixed animals were stained in working solution for two hours. Animals were subsequently placed in M9 solution
350 to remove excess stain and were imaged on a Leica Thunder multichannel microscope to generate composite
351 images. For Nile red staining, Nile red working solution was generated by mixing 6 ul/Nile red stock solution in 1
352 ml isopropanol. Animals were stained in working solution for two hours followed by 30 minutes of wash in M9
353 solution. Nile red imaging was performed on the Leica Thunder GFP setting with 10ms exposure at 5X
354 magnification.
355
356 Western Blotting
357 Worms were isolated by washing with M9 buffer and centrifuged into a pellet. Worm lysates were prepared by
358 adding RIPA buffer and proteinase inhibitor cocktail (Roche) followed by water bath sonication in a Diagenode
359 Bioruptor XL 4 at maximum strength for 15 minutes. Lysates were cleared of debris via centrifugation at 21,000g
360 for 15 minutes at 4oC and supernatant was collected. Protein concentration as measured using the Pierce BCA
361 Assay (Thermo Fisher). Lysate was subsequently mixed with 4X Laemmli buffer (Bio-Rad) and boiled for 10
362 minutes. Samples were run on SDS-PAGE protocol (Bio-Rad) and transferred to nitrocellulose membrane via
Fo
363 wet transfer at 100V for 1 hour. Immunoblotting was performed according to primary antibody manufacturer’s
364 protocols. Secondary antibody treatment utilized goat -anti-rabbit HRP conjugate or goat-anti-mouse-HRP
365 conjugate (GE Healthcare) at 1:10,000 and 1:5000 dilutions, respectively. HRP chemiluminescence was
rP
366 detected via West-Pico substrate (Thermo Pierce). The western blot results shown are representative of at least
367 two experiments. Primary antibodies used were the following:
368 Rabbit monoclonal anti-Phospho-AMPKα (Thr172), Cell Signaling Technology
ee
369 Rabbit monoclonal anti-p70 Phospho-S6 Kinase (Thr389), Cell Signaling Technology
370 Mouse monoclonal anti-Actin, Abcam
371
rR
372 Quantitative RT-PCR

373 Worms samples were flash frozen in liquid nitrogen and kept in -80°C until RNA preparation. Samples were lysed
374 through the use of metal beads and the Tissuelyser (Qiagen) Total RNA was extracted using RNAzol RT
375 (Molecular Research Center) according to manufacturer instructions. RNA was treated with RNAse free DNAse
ev
376 prior to reverse transcription with the Quantitect reverse transcription kit (Qiagen). Quantitative RT-PCR was
377 conducted in triplicate using a Quantitect SYBR Green PCR reagent (Qiagen) following manufacturer instructions
iew
378 on a Bio-Rad CFX96 Real-Time PCR system (Bio-Rad) Expression levels of tested genes were presented as
379 normalized fold changes to the mRNA abundance of control genes indicated in the figures by the δδCt method.
380 The primers used for the qPCR are as follows:
381 rft-1 forward: GCTATTGTTCAGATCGCGTGC
382 rft-1 reverse: CAGAGACCCAATTGACAAATACATGC
383 rft-2 forward: CGGGAGTTGTTCAGATCGCT
384 rft-2 reverse: GAGTCCCAGTTGACAACAGCA
385 rfk forward: TGTTGGAAAAAGAAACGAAAGAA
386 rfk reverse: TCGATTAAAATTCGGTAACAACG
387 flad-1 forward: TGCCTGGAGTTCCAAAATTC
388 flad-1 reverse: GAAGGGCTGGGTGTTTTACA
389 C07G1.7 forward: GCTGAAGAAGCTTCAACCGTAG
390 C07G1.7 reverse: TCTCGTGTCAATTCCGGTCT
391 hrg-9 forward: TGGAATATTGAGTGGCGTTG
392 hrg-9 reverse: CCTCCTCTACTTGGTGCATGT
393 cdr-2 forward: CGAGCCTCATTTGGAAAGAA
394 cdr-2 reverse: GCATCTGCCGCTGTAACTTT
395
396 GC/MS Lipid Analysis
397 Lipid extraction and GC/MS of extracted, acid-methanol derivatized lipids was performed as described previously
398 (Pino and Soukas 2020). Briefly, 10,000 synchronous mid-L4 animals were sonicated with a probe sonicator on
399 high intensity in a microfuge tube in 100-250 microliters total volume. Following sonication, lipids were extracted
11
400 in 3:1 methanol: methylene chloride following the addition of acetyl chloride in sealed borosilicate glass tubes,
401 which were then incubated in a 75°C water bath for 1 hour. Derivatized fatty acids and fatty alcohols were
402 neutralized with 7% potassium carbonate, extracted with hexane, and washed with acetonitrile prior to
403 evaporation under nitrogen. Lipids were resuspended in 200 microliters of hexane and analyzed on an Agilent
404 GC/MS equipped with a Supelcowax-10 column as previously described (Pino, Webster et al. 2013). Fatty acids
405 were indicated as the normalized peak area of the total of derivatized fatty acids detected in the sample,
406 normalized by recovery of spiked-in, standard phospholipid and triglyceride.
407
408
409 LC/MS Metabolite Analysis
410 Four biologic replicates of adult day 1 wild-type worms treated either with empty vector or rft-1 RNAi were
411 collected with M9 wash and frozen by liquid nitrogen into a worm pellet. Polar metabolites of homogenized worms
412 were analyzed using a Thermo QExactive orbitrap mass spectrometer coupled to a Thermo Vanquish UPLC
413 system, as previously described.(Garratt, Lagerborg et al. 2018) Bioactive lipids metabolites were profiled on the
414 same system, as previous described.(Lagerborg, Watrous et al. 2019) Collected data were imported into the
415 mzMine 2 software suite for analysis (version 2.53). Metabolites were annotated by using an in-house library of
416 commercially available standards. Please see supplemental methods for detailed methods. All mass integration
417 values, normalized abundance values, significance testing scores, and pathway enrichment scores are included
418 in this manuscript as Supplementary Table 2.
Fo
419
420 Quantification and statistical analysis
421 All western blotting quantifications were conducted on Bio-Rad Image Lab. Intensity analysis for fluorescent
rP
422 images was performed on ImageJ. Statistical analyses were performed using Prism (GraphPad Software). The
423 statistical differences between control and experimental groups were determined by two-tailed students t-test
424 (two groups), one-way ANOVA (more than two groups), two-way ANOVA (two independent experimental
ee
425 variables), or log-rank (survival analyses) as indicated in each figure legend, with numbers of samples indicated
426 and corrected P values < 0.05 considered significant.
427
rR
428 Fluorescent Microscopy

429 DIC, brightfield and fluorescence Imaging of animals was performed utilizing the Leica Thunder microscopy
430 system. Animals were picked onto a slide containing agar and 2.5mM levimasole solution. Imaging was
431 performed within 5 minutes of slide placement. GFP and RFP Images were taken at 10ms exposure at 30% FIM
ev
432 and at 5X magnification, unless otherwise specified. Fluorescence intensity for quantification was calculated
433 utilizing ImageJ software. For signal intensity experiments, quantification was performed on 20 worms (10 worms
iew
434 of two biological replicates).

435
436 Two-Photon and Fluorescence Lifetime Imaging
437 Wild type and mutant C. elegans were immobilized for fluorescence imaging using a previously proposed
438 protocol (Kim, Sun et al. 2013). Endogenous two-photon excited fluorescence (TPEF) images of C. elegans were
439 acquired using a laser scanning microscope (Leica TCS SP8, Wetzlar, Germany) equipped with a femtosecond
440 laser (Insight Deep See, Spectra Physics, Mountain View, CA). Fluorescence lifetime images (512 × 512 pixels)
441 of C. elegans were acquired using the same excitation and emission settings as for intensity NAD(P)H and FAD
442 images and a PicoHarp 300 time-correlated single photon counting unit (PicoQuant, Berlin, Germany) integrated
443 in the Leica SP8 system. Please see Supplemental methods for further details on methods and analysis.
444
445 Stimulated Raman Scattering imaging
446 Stimulated Raman scattering (SRS) images of C. elegans were acquired using a laser scanning confocal
447 microscope (Leica TCS SP8, Wetzlar, Germany) equipped with a picosecond NIR laser (picoEmerald, APE,
448 Berlin, Germany). SRS images were acquired in the wavenumber range of 2798 cm-1 to 3103 cm-1 with an
449 interval of 6 cm-1. The Nd:VAN 1064.2 nm output was used as the SRS Stokes beam and the OPO beam tuned
450 from 800 nm to 820 nm with step size of 0.4 nm was used as the pump laser. SRS images (620 × 620 microns
451 x 51 wavenumbers, 512 × 512 pixels x 51 wavenumbers, 0.75 zoom) were acquired using a water immersion
452 objective HCX IRAPO L 25x/0.95 NA with pixel dwell time of 4.9 μs. The pixels corresponding to regions occupied
12
453 by C elegans were identified by implementing a global threshold of 300 (intensities ranged from 0 to 800). The
454 SRS spectrum of each remaining pixel was normalized by the maximum SRS value of the entire field spectral
455 image. To estimate the relative unsaturation level of the lipids in C. elegans, a ratio metric approach was adapted
456 (Verma and Wallach 1977, Freudiger, Min et al. 2011, Nieva, Marro et al. 2012, Potcoava, Futia et al. 2014).
457 Specifically, the ratio of the area under the SRS spectrum for wavenumbers spanning 2991 and 3022 cm-1 and
458 wavenumbers spanning 2830 and 2870 cm-1 ,was estimated to represent the relative unsaturation levels (Verma
459 and Wallach 1977, Freudiger, Min et al. 2011, Nieva, Marro et al. 2012, Potcoava, Futia et al. 2014). Both
460 fluorescence and SRS images were calibrated for laser power before analysis.
461
462
463
464
465 Acknowledgements:
466 We thank Dr. Yuyao Zhang and Talia Hart for their creative input. This work was funded by the NIH/NIA Grants
467 R01AG058259 and R01AG69677 and the Weissman Family MGH Research Scholar Award (to Alexander
468 Soukas). This work was also funded by the Foundation for Women’s Wellness, the Human Growth Foundation
469 and NIH/NIDDK F32DK124948 (to Armen Yerevanian). Thanks to the Nutrition Obesity Research Center
470 (NORC) at Harvard (P30DK040561) for core services. Some strains were provided by the CGC, which is funded
Fo
471 by the NIH Office of Research Infrastructure Programs (P40OD010440). Figure 3f and 6f were generated by
472 BioRender.com. SRS and two photon images were acquired using microscopes acquired through NIH S10
473 OD021624 and NSF Major Research Instrumentation 1531683 grants.
474
rP
475 Conflict of Interest Statement: The authors report no competing interests.

476
477 Author Contributions: Conceptualization: AY, LM, DB, AAS; Methodology: AY, LM, DB, SE; Validation: AY,
ee
478 LM, DB, SE; Investigation: AY, LM, SE, DB, SL, YZ, AA, LC, AAS; Formal Analyses: AY, EG, KD, MJ, IG, AAS;
479 Writing: AY, AAS; Review and Editing: AY, FA, SE, LM, YZ, SL, LC, AA, DB, EG, MJ, IG, AAS; Funding: AY,IG
480 AAS.
rR
481
482 Data Availability Statement: The data that support the findings of this study are available from the
483 corresponding author upon reasonable request.
ev
484
485
iew
486
487
488
489
490
13
491 Figure Legends:

492
493 FIGURE 1 rft-1 RNAi depletes flavins and extends lifespan. (a) Knockdown of the riboflavin transporter rft-1 via
494 RNA interference significantly extends lifespan in wild type (wt) animals versus empty vector RNAi (EV). Addition
495 of 665 µM riboflavin abrogates this lifespan extension (see table S1 for tabular data and replicates). (b)
496 Knockdown of rft-1 produces 70% reduction in transcript levels by quantitative RT-PCR and does not induce
497 corresponding reductions in orthologous transporter rft-2 mRNA. (c) LC-MS/MS analysis of worm lysates
498 collected at adult day 1 treated with rft-1 RNAi reveals significant reductions in organismal riboflavin, FMN, and
499 FAD concentrations. (d) Brood size is diminished in rft-1 RNAi treated animals and remains suppressed in ppw-1
500 animals (somatic RNAi competent, germline RNAi incompetent) but is rescued in rrf-1 animals (somatic RNAi
501 blunted, germline RNAi competent), suggesting a somatic site of rft-1 action. (e) Pharyngeal pumping assay on
502 adult day 1 animals reveals no difference between EV and rft-1 RNAi. (f) Developmental rate from the first larval
503 stage to adulthood is unchanged in animals treated with rft-1 RNAi, with or without riboflavin. For a-f, results are
504 representative of three biological replicates. * indicates P < 0.05, **, P < 0.001 by log-rank analysis (a and f),
505 two-way ANOVA of ΔCt values (b), standard two-way ANOVA (c-d) and two-tailed Student’s t-test (e). Bars
506 represent means  SEM. Ns presents non-significant.
507
508 FIGURE 2 Riboflavin depletion promotes longevity by activating FOXO/daf-16. (a) Lifespan extension with
509 knockdown of the rft-1 transporter is abrogated in daf-16 mutants, with and without supplemental riboflavin. (b)
Fo
510 Nuclear localization of DAF-16 was increased following rft-1 RNAi in a riboflavin-dependent manner as assessed
511 by a DAF-16::GFP translational reporter. Animals were binned into no nuclear localization, low levels of
512 localization and high levels of localization (see Fig. S2 for representative images). N = 60 animals per condition,
rP
513 representative of two biological replicates. (c) A sod-3p::GFP transcriptional reporter indicates increased activity
514 of DAF-16 significantly over the lifespan of rft-1 RNAi treated animals. (d) A chromosomally-located akt-1 gain-
515 of-function mutation blunts the longevity response to riboflavin depletion vs. wt. For a-d, results are
ee
516 representative of at least two biological replicates. * indicates P < 0.05 by log-rank analysis (a and d), and by
517 two-way ANOVA followed by Sidak’s multiple comparisons post-hoc test (c). See table S1 for tabular data and
rR
518 replicates of survival analyses. Bars represent means  SEM.

519
520 FIGURE 3 Riboflavin depletion alters cellular energetics. (a) Lifespan extension with rft-1 RNAi is abrogated in
ev
521 AMPK/aak-2 mutants. (b) Addition of riboflavin has a deleterious effect on lifespan in aak-2 mutants. (c) Western
522 blotting of phospho-AMPKThr172 in lysates collected from young adult animals indicates activation following RNAi
523 to rft-1, an effect abrogated by the addition of riboflavin. (d) Box plots of results from two-photon and fluorescence
iew
524 lifetime imaging, including organismal redox ratio, mitochondrial clustering, NAD(P)H bound fraction and FAD
525 bound fraction for EV and rft-1 RNAi treated animals. Riboflavin depletion decreases the redox ratio and
526 increases intestinal mitochondrial clustering and the FAD bound fraction. (e) Volcano plot and heatmap of
527 differentially abundant metabolites quantified by LC-MS reveals reductions in citric acid metabolites including
528 citric acid, isocitric acid and α-ketoglutarate following riboflavin depletion. Purine metabolites including xanthine,
529 hypoxanthine and guanosine are enriched with rft-1 RNAi. (f) Representation of citric acid metabolites impacted
530 by riboflavin depletion. See table S1 for tabular data and replicates of survival analyses. For (a-c) results are
531 representative of two biological replicates. For (d) results are from 8 worms of two biological replicates. * indicates
532 P < 0.01 by log-rank analysis (a-b) and by two tailed t-test (d). Box and whisker plots (d) indicate median and
533 5/95th percentile respectively.
14
534
535 FIGURE 4 Activation of the mitochondrial unfolded protein response (UPRmt) is required for riboflavin depletion
536 to promote longevity. (a) RNAi to rft-1 promotes activation of hsp-6p::GFP progressively on adult days 1 and 3,
537 an effect reversed by riboflavin supplementation (for binning, N = 60 worms per condition, representative of two
538 biological replicates) Images above at 40X, binning performed at 10X magnification. (b) Quantitative RT-PCR of
539 established atfs-1 target genes indicates marked upregulation with riboflavin depletion. (c) atfs-1 loss of function
540 mutants do not experience lifespan extension with riboflavin depletion. (d) Gain of function mutants in atfs-1 are
541 short lived but preserve responsiveness to rft-1 RNAi. For (b), results are representative of at least three
542 biological replicates unless otherwise specified. See table S1 for tabular data and replicates of survival analyses.
543 * indicates P < 0.05 by two-way ANOVA of ΔCt values (b), and by log-rank analysis (c and d). Bars represent
544 means  SEM.
545
546 FIGURE 5 Riboflavin depletion alters lipid metabolism. (a) Fixative-based Nile red staining and quantification of
547 images reveal significant increases in fat mass with post developmental rft-1 RNAi in eri-1 enhanced RNAi
548 mutants and with larval exposure to rft-1 RNAi in multiple mutants including daf-16, aak-2, atfs-1(GOF), and
549 raga-1. atfs-1 and rict-1 loss-of-function mutants do not exhibit increased fat on rft-1 RNAi. (b) Lipid analysis via
550 GC/MS reveals an increase in overall fat stores (triglyceride/phospholipid ratio). (c) Shifts towards unsaturated
551 fatty acid side chains and branched chain lipids in phospholipid and triglyceride fractions in aggregate (d) acdh-
Fo
552 1p::GFP expression is significantly upregulated in intestine with rft-1 RNAi. * indicates P < 0.05 by students 2-
553 tailed t-test (b,d) and by two-way ANOVA (c). Bars (b-d) represent means  SEM. Dots in (a) represent individual
554 biological replicates.
rP
555
556 FIGURE 6 Riboflavin depletion mimics features of dietary restriction. (a) Imaging of reporters known to be
557 activated with dietary restriction (bigr-1::RFP, right, and acs-2p::GFP, left) indicates activation under rft-1 RNAi
ee
558 that progresses with age. (b) Lifespan extension with rft-1 RNAi is abrogated in pha-4 loss-of function mutants
559 versus temperature sensitive smg-1 mutant controls. (c) TORC1 mutant raga-1 exhibits lifespan extension with
560 rft-1 RNAi. (d) Western blot of adult day 1 animals containing humanized S6K (permitting western blotting for
rR
561 TORC1-mediated phosphorylation of S6KT389) reveals no change in phospho-S6K concentrations between

562 control and riboflavin depleted worms. (e) TORC2 mutant rict-1 does not exhibit lifespan extension with riboflavin
563 depletion. (f) Model of riboflavin depletion impact on metabolism and longevity. Knockdown of the riboflavin
ev
564 transporter leads to depletion of flavin co-factors, influencing the energetic status of the animal and affecting
565 global enzymatic activities dependent on FAD and FMN. This creates an integrated metabolic signaling response
566 that promotes increased triglyceride stores and extends lifespan. See table S1 for tabular data and replicates of
iew
567 survival analyses. * indicates P < 0.05 by two-way ANOVA (a), log-rank test (b,c and e). Bars represent means
568  SEM.
15
569 References
570 Alonzo, C. A., S. Karaliota, D. Pouli, Z. Liu, K. P. Karalis and I. Georgakoudi (2016). "Two-photon excited fluorescence of
571 intrinsic fluorophores enables label-free assessment of adipose tissue function." Sci Rep 6: 31012.
572 Austin, M. U., W. S. Liau, K. Balamurugan, B. Ashokkumar, H. M. Said and C. W. LaMunyon (2010). "Knockout of the
573 folate transporter folt-1 causes germline and somatic defects in C. elegans." BMC Dev Biol 10: 46.
574 Bennett, C. F., H. Vander Wende, M. Simko, S. Klum, S. Barfield, H. Choi, V. V. Pineda and M. Kaeberlein (2014).
575 "Activation of the mitochondrial unfolded protein response does not predict longevity in Caenorhabditis elegans." Nat
576 Commun 5: 3483.
577 Biswas, A., D. Elmatari, J. Rothman, C. W. LaMunyon and H. M. Said (2013). "Identification and functional
578 characterization of the Caenorhabditis elegans riboflavin transporters rft-1 and rft-2." PLoS One 8(3): e58190.
579 Bito, T., Y. Matsunaga, Y. Yabuta, T. Kawano and F. Watanabe (2013). "Vitamin B12 deficiency in Caenorhabditis elegans
580 results in loss of fertility, extended life cycle, and reduced lifespan." FEBS Open Bio 3: 112-117.
581 Cabreiro, F., C. Au, K. Y. Leung, N. Vergara-Irigaray, H. M. Cocheme, T. Noori, D. Weinkove, E. Schuster, N. D. Greene and
582 D. Gems (2013). "Metformin retards aging in C. elegans by altering microbial folate and methionine metabolism." Cell
583 153(1): 228-239.
584 Dipti, S., A. M. Childs, J. H. Livingston, A. K. Aggarwal, M. Miller, C. Williams and Y. J. Crow (2005). "Brown-Vialetto-Van
585 Laere syndrome; variability in age at onset and disease progression highlighting the phenotypic overlap with Fazio-Londe
Fo
586 disease." Brain Dev 27(6): 443-446.
587 Durieux, J., S. Wolff and A. Dillin (2011). "The cell-non-autonomous nature of electron transport chain-mediated
588 longevity." Cell 144(1): 79-91.
rP
589 Escorcia, W., D. L. Ruter, J. Nhan and S. P. Curran (2018). "Quantification of Lipid Abundance and Evaluation of Lipid
590 Distribution in Caenorhabditis elegans by Nile Red and Oil Red O Staining." J Vis Exp(133).
591 Freudiger, C. W., W. Min, G. R. Holtom, B. Xu, M. Dantus and X. S. Xie (2011). "Highly specific label-free molecular
ee
592 imaging with spectrally tailored excitation stimulated Raman scattering (STE-SRS) microscopy." Nat Photonics 5(2): 103-
593 109.
594 Gandhimathi, K., S. Karthi, P. Manimaran, P. Varalakshmi and B. Ashokkumar (2015). "Riboflavin transporter-2 (rft-2) of
rR
595 Caenorhabditis elegans: Adaptive and developmental regulation." J Biosci 40(2): 257-268.
596 Garratt, M., K. A. Lagerborg, Y. M. Tsai, A. Galecki, M. Jain and R. A. Miller (2018). "Male lifespan extension with 17-alpha
597 estradiol is linked to a sex-specific metabolomic response modulated by gonadal hormones in mice." Aging Cell 17(4):
ev
598 e12786.
599 Gioran, A., A. Piazzesi, F. Bertan, J. Schroer, L. Wischhof, P. Nicotera and D. Bano (2019). "Multi-omics identify xanthine
600 as a pro-survival metabolite for nematodes with mitochondrial dysfunction." EMBO J 38(6).
iew
601 Greer, E. L., D. Dowlatshahi, M. R. Banko, J. Villen, K. Hoang, D. Blanchard, S. P. Gygi and A. Brunet (2007). "An AMPK-
602 FOXO pathway mediates longevity induced by a novel method of dietary restriction in C. elegans." Curr Biol 17(19):
603 1646-1656.
604 Han, S., E. A. Schroeder, C. G. Silva-Garcia, K. Hebestreit, W. B. Mair and A. Brunet (2017). "Mono-unsaturated fatty acids
605 link H3K4me3 modifiers to C. elegans lifespan." Nature 544(7649): 185-190.
606 Han, S. K., D. Lee, H. Lee, D. Kim, H. G. Son, J. S. Yang, S. V. Lee and S. Kim (2016). "OASIS 2: online application for survival
607 analysis 2 with features for the analysis of maximal lifespan and healthspan in aging research." Oncotarget 7(35): 56147-
608 56152.
609 Kim, E., L. Sun, C. V. Gabel and C. Fang-Yen (2013). "Long-term imaging of Caenorhabditis elegans using nanoparticle-
610 mediated immobilization." PLoS One 8(1): e53419.
611 Kim, H. E., A. R. Grant, M. S. Simic, R. A. Kohnz, D. K. Nomura, J. Durieux, C. E. Riera, M. Sanchez, E. Kapernick, S. Wolff
612 and A. Dillin (2016). "Lipid Biosynthesis Coordinates a Mitochondrial-to-Cytosolic Stress Response." Cell 166(6): 1539-
613 1552 e1516.
614 Labbadia, J., R. M. Brielmann, M. F. Neto, Y. F. Lin, C. M. Haynes and R. I. Morimoto (2017). "Mitochondrial Stress
615 Restores the Heat Shock Response and Prevents Proteostasis Collapse during Aging." Cell Rep 21(6): 1481-1494.
616 Lagerborg, K. A., J. D. Watrous, S. Cheng and M. Jain (2019). "High-Throughput Measure of Bioactive Lipids Using Non-
617 targeted Mass Spectrometry." Methods Mol Biol 1862: 17-35.
618 Lee, S. S., S. Kennedy, A. C. Tolonen and G. Ruvkun (2003). "DAF-16 target genes that control C. elegans life-span and
619 metabolism." Science 300(5619): 644-647.
16
620 Lienhart, W. D., V. Gudipati and P. Macheroux (2013). "The human flavoproteome." Arch Biochem Biophys 535(2): 150-
621 162.
622 Liu, Z., D. Pouli, C. A. Alonzo, A. Varone, S. Karaliota, K. P. Quinn, K. Munger, K. P. Karalis and I. Georgakoudi (2018).
623 "Mapping metabolic changes by noninvasive, multiparametric, high-resolution imaging using endogenous contrast." Sci
624 Adv 4(3): eaap9302.
625 Mansoorabadi, S. O., C. J. Thibodeaux and H. W. Liu (2007). "The diverse roles of flavin coenzymes--nature's most
626 versatile thespians." J Org Chem 72(17): 6329-6342.
627 Massey, V. (1995). "Introduction: flavoprotein structure and mechanism." FASEB J 9(7): 473-475.
628 Moriyama, Y. (2011). "Riboflavin transporter is finally identified." J Biochem 150(4): 341-343.
629 Nabokina, S. M., V. S. Subramanian and H. M. Said (2012). "Effect of clinical mutations on functionality of the human
630 riboflavin transporter-2 (hRFT-2)." Mol Genet Metab 105(4): 652-657.
631 Nieva, C., M. Marro, N. Santana-Codina, S. Rao, D. Petrov and A. Sierra (2012). "The lipid phenotype of breast cancer
632 cells characterized by Raman microspectroscopy: towards a stratification of malignancy." PLoS One 7(10): e46456.
633 Panowski, S. H., S. Wolff, H. Aguilaniu, J. Durieux and A. Dillin (2007). "PHA-4/Foxa mediates diet-restriction-induced
634 longevity of C. elegans." Nature 447(7144): 550-555.
635 Paradis, S. and G. Ruvkun (1998). "Caenorhabditis elegans Akt/PKB transduces insulin receptor-like signals from AGE-1
636 PI3 kinase to the DAF-16 transcription factor." Genes Dev 12(16): 2488-2498.
637 Pino, E. C. and A. A. Soukas (2020). "Quantitative Profiling of Lipid Species in Caenorhabditis elegans with Gas
Fo
638 Chromatography-Mass Spectrometry." Methods Mol Biol 2144: 111-123.
639 Pino, E. C., C. M. Webster, C. E. Carr and A. A. Soukas (2013). "Biochemical and high throughput microscopic assessment
640 of fat mass in Caenorhabditis elegans." J Vis Exp(73).
rP
641 Potcoava, M. C., G. L. Futia, J. Aughenbaugh, I. R. Schlaepfer and E. A. Gibson (2014). "Raman and coherent anti-Stokes
642 Raman scattering microscopy studies of changes in lipid content and composition in hormone-treated breast and
643 prostate cancer cells." J Biomed Opt 19(11): 111605.
ee
644 Powers, H. J. (2003). "Riboflavin (vitamin B-2) and health." Am J Clin Nutr 77(6): 1352-1360.
645 Qi, B., M. Kniazeva and M. Han (2017). "A vitamin-B2-sensing mechanism that regulates gut protease activity to impact
646 animal's food behavior and growth." Elife 6.
rR
647 Schiavi, A., A. Torgovnick, A. Kell, E. Megalou, N. Castelein, I. Guccini, L. Marzocchella, S. Gelino, M. Hansen, F. Malisan, I.
648 Condo, R. Bei, S. L. Rea, B. P. Braeckman, N. Tavernarakis, R. Testi and N. Ventura (2013). "Autophagy induction extends
649 lifespan and reduces lipid content in response to frataxin silencing in C. elegans." Exp Gerontol 48(2): 191-201.
ev
650 Soo, S. K. and J. M. Van Raamsdonk (2021). "High confidence ATFS-1 target genes for quantifying activation of the
651 mitochondrial unfolded protein response." MicroPubl Biol 2021.
iew
652 Soukas, A. A., E. A. Kane, C. E. Carr, J. A. Melo and G. Ruvkun (2009). "Rictor/TORC2 regulates fat metabolism, feeding,
653 growth, and life span in Caenorhabditis elegans." Genes Dev 23(4): 496-511.
654 Spagnoli, C. and C. De Sousa (2012). "Brown-Vialetto-Van Laere syndrome and Fazio-Londe disease - treatable motor
655 neuron diseases of childhood." Dev Med Child Neurol 54(4): 292-293.
656 Subramanian, V. S., S. B. Subramanya, L. Rapp, J. S. Marchant, T. Y. Ma and H. M. Said (2011). "Differential expression of
657 human riboflavin transporters -1, -2, and -3 in polarized epithelia: a key role for hRFT-2 in intestinal riboflavin uptake."
658 Biochim Biophys Acta 1808(12): 3016-3021.
659 Sydenstricker, V. P. (1941). "Clinical Manifestations of Ariboflavinosis." Am J Public Health Nations Health 31(4): 344-
660 350.
661 Van Gilst, M. R., H. Hadjivassiliou and K. R. Yamamoto (2005). "A Caenorhabditis elegans nutrient response system
662 partially dependent on nuclear receptor NHR-49." Proc Natl Acad Sci U S A 102(38): 13496-13501.
663 Verma, S. P. and D. F. Wallach (1977). "Raman spectra of some saturated, unsaturated and deuterated C18 fatty acids in
664 the HCH-deformation and CH-stretching regions." Biochim Biophys Acta 486(2): 217-227.
665 Watson, E., L. T. MacNeil, H. E. Arda, L. J. Zhu and A. J. M. Walhout (2013). "Integration of metabolic and gene regulatory
666 networks modulates the C. elegans dietary response." Cell 153(1): 253-266.
667 Wu, L., B. Zhou, N. Oshiro-Rapley, M. Li, J. A. Paulo, C. M. Webster, F. Mou, M. C. Kacergis, M. E. Talkowski, C. E. Carr, S.
668 P. Gygi, B. Zheng and A. A. Soukas (2016). "An Ancient, Unified Mechanism for Metformin Growth Inhibition in C. elegans
669 and Cancer." Cell 167(7): 1705-1718 e1713.
17
670 Wu, Z., M. M. Senchuk, D. J. Dues, B. K. Johnson, J. F. Cooper, L. Lew, E. Machiela, C. E. Schaar, H. DeJonge, T. K.
671 Blackwell and J. M. Van Raamsdonk (2018). "Mitochondrial unfolded protein response transcription factor ATFS-1
672 promotes longevity in a long-lived mitochondrial mutant through activation of stress response pathways." BMC Biol
673 16(1): 147.
674 Yamamoto, S., K. Inoue, K. Y. Ohta, R. Fukatsu, J. Y. Maeda, Y. Yoshida and H. Yuasa (2009). "Identification and functional
675 characterization of rat riboflavin transporter 2." J Biochem 145(4): 437-443.
676 Yang, R., Y. Li, Y. Wang, J. Zhang, Q. Fan, J. Tan, W. Li, X. Zou and B. Liang (2022). "NHR-80 senses the mitochondrial UPR
677 to rewire citrate metabolism for lipid accumulation in Caenorhabditis elegans." Cell Rep 38(2): 110206.
678 Yen, C. A., D. L. Ruter, C. D. Turner, S. Pang and S. P. Curran (2020). "Loss of flavin adenine dinucleotide (FAD) impairs
679 sperm function and male reproductive advantage in C. elegans." Elife 9.
680 Yonezawa, A., S. Masuda, T. Katsura and K. Inui (2008). "Identification and functional characterization of a novel human
681 and rat riboflavin transporter, RFT1." Am J Physiol Cell Physiol 295(3): C632-641.
682
683
Fo
rP
ee
rR
ev
iew
18
684 Supporting Information

685 Additional Supporting Information may be found in the online versions of this article at the publisher’s
686 website.
687
688 Figure S1: RT-PCR of riboflavin metabolism genes, DAPI staining of germline
689 Figure S2: DAF-16::GFP images, sod-3p::GFP images
690 Figure S3: FLIM Methods, Metabolite pathway analysis
691 Figure S4: Example images of hsp-6p::GFP activation
692 Figure S5: GC/MS lipid profiles of EV and rft-1 RNAi treated animals
693 Appendix S1: Supplemental Methods
694 Supplemental References
695
Fo
rP
ee
rR
ev
iew
19
AGING CELL AUTHOR CHECKLIST

Title: Riboflavin Depletion Promotes Longevity and Metabolic Hormesis in Caenorhabditis elegans
Authors: Armen Yerevanian, Luke Murphy, Sinclair Emans, Yifei Zhou, Fasih Ahsan, Daniel Baker, Sainan Li,
Adebanjo Adedoja, Lucydalila Cedillo, Einstein Gnanatheepam, Khoi Dao, Mohit Jain, Irene Georgakoudi,
Alexander Soukas
Manuscript Type: Research Article
Total Character Count (including spaces): 49,969
Word count of Summary: 122
Number of papers cited in the References: 50
Listing of all Tables (Table1, Table 2 etc): N/A
Figure specifications:
Fo
Figure No Color Single Double Size of Figure Smallest Font
Column Column
1 Yes Yes 21 x 12 6
rP
2 Yes Yes 16 x 15.5 6

3 Yes Yes 27 x 33 6
4 Yes Yes 22.5 x 16 6
ee
5 Yes Yes 37 x 22.5 6

6 Yes Yes 28 x 21.5 6
rR
ev
iew
2
(a) wt (EV RNAi) wt (EV RNAi) + Riboflavin (b) rft-1 rft-2

(c) EV RNAi rft-1 RNAi
wt (rft-1 RNAi) wt (rft-1 RNAi) + Riboflavin
✱ 300000 ✱ ✱ ✱
Normalized Concentration (A.U)

2.5 5
1.00
vs **
2.0 4
vs ns
Fold Change
Fold Change
vs ns ns
1.5 3 200000
0.75 vs **
vs ns 1.0 2
Fraction Alive
0.5 1
0.50
100000
0.0 0
e e in in
cl cl v v
n
e
e
i i
fla fla
cl
cl
vi
vi
h h
ve ve
hi
hi
fla
fla
0.25 o o
ib ib
ve
ve
bo
bo
i+ i+ R R
Fo
i+
+
Ri
Ri
A A
i+ i+ 0
Ai
N N
+
R R A A
RN
RN
Ai
Ai
rP
-1 N N
EV Riboflavin FMN FAD
RN
rft R R
RN
-1
EV
EV -1
rft
rft
-1
eer
EV
0.00
rft
0 10 20 30
Day
Rens wt (EV RNAi) + vehicle wt (rft-1 RNAi) + vehicle
(d) ✱ ✱ (e) vie (f)

wt (EV RNAi) + riboflavin wt (rft-1 RNAi) + riboflavin
400
EV RNAi w
Pumping Rate (pumps/min)

300
Brood Size (# progeny)
1.0
rft-1 RNAi
Fraction Reaching Adulthood

300 0.8
vs ns
200 vs ns
vs ns
200 0.6
vs ns
100 0.4
100
0.2
0
0 i i
wt rrf-1 ppw-1 N A N A 0.0
R R 35 37 39 41 43 45 47 49 51 53 59 61 63
EV rf t-1 Hour Since Hatching
(a)
wt (EV RNAi) daf-16 (EV RNAi) wt (EV RNAi) daf-16 (EV RNAi)
wt (rft-1 RNAi) daf-16 (rft-1 RNAi) wt (rft-1 RNAi) daf-16 (rft-1 RNAi)
1.00
1.00
vs * vs ns
vs ns vs ns
vs * vs
0.75 vs *
* 0.75 vs ns
Fraction Alive
Fraction Alive
0.50 0.50
No Riboflavin Suppl. Riboflavin
0.25 0.25
0.00 0.00
0 10 20 30 0 10 20 30
Day Day
EV RNAi + vehicle rft-1 RNAi + vehicle rft-1 RNAi + riboflavin EV RNAi + riboflavin
(b) 120
None
120
None
120
None
120
None
100
Low
High
100
Fo Low
High
100
Low
High
100
Low
High
80 80 rP 80 80
ee
Individuals
Individuals
Individuals
Individuals
60 60 60 60
rR
ev
40 40 40 40
20 20 20
ie 20
0
lt 1 3
0 0 w lt 1 3
0
du ay ay du ay ay
3
1
1
lt
lt
du
du
ay
ay
ay
ay
A D D A D D
t t t t
A
A
g g
D
D
D
l l l l
un du du un du du
ng
ng
lt
lt
lt
lt
du
du
du
du
o A A o A A
u
u
Y Y
Yo
Yo
A
A
A
sod-3p::GFP
(c) ns ✱ ✱ ✱
(d) wt (EV RNAi)
wt (rft-1 RNAi)
akt-1GOF (EV RNAi)
akt-1GOF (rft-1 RNAi)
60
EV RNAi
1.0
rft-1 RNAi vs *
Signal Intensity (Arbitrary Units)
50 vs *
vs *
vs *
40
Fraction Alive
30 0.5
20
10
0 0.0
Adult Day 1 Adult Day 3 Adult Day 5 Adult Day 7 0 10 20 30 40
Fold Change 1.16 2.02 2.06 2.05 Day
(a) (e)
wt (EV RNAi) aak-2 (EV RNAi)

Flavins Purine Metabolism
1.00
wt (rft-1 RNAi) aak-2 (rft-1 RNAi) 4 Miscellaneous Citric Acid Cycle
vs *
vs ns FMN
vs *
0.75 vs * 3
-log10(FDR)
Fraction Alive
RIBOFLAVIN
FAD GLUTATHIONE REDUCED
GUANINE
OXOGLUTARATE
0.50
No Riboflavin 2 ISOCITRATE GUANOSINE MONOPHOSPHATE
GUANOSINE
DEOXYGUANOSINE HYPOXANTHINE
ADENOSINE TRIPHOSPHATE XANTHINE

0.25
1
0.00
0 10 20 30 40
(b) 0
Day
wt (EV RNAi) aak-2 (EV RNAi) -6 -4 -2 0 2
wt (rft-1 RNAi) aak-2 (rft-1 RNAi)
1.00 log2(rft-1 RNAi / EV)
vs ns
vs *
vs ns
0.75 vs *
Fraction Alive
0.50
Suppl. Riboflavin
0.25
0.00
0 10 20 30
(c) Riboflavin
Day
- - + +
Fo
rP
rft-1 RNAi - + - +
ee
rR
p-AMPK
ev
iew
Actin
(d) Redox Ratio Mito Clustering

0.010 2.0
✱ ✱
0.008
1.8
0.006
Beta
1.6
0.004 rft-1 RNAi EV RNAi
(f)
1.4
0.002
0.000 1.2
EV RNAi rft-1 RNAi EV RNAi rft-1 RNAi
NAD(P)H Bound Fraction FAD Bound Fraction

0.65 0.60 ✱
✱
0.55
0.60
0.50
Fraction
Fraction
0.55
0.45
0.50
0.40
0.45 0.35
(a) (c)
EV RNAi with Vehicle rft-1 RNAi with Vehicle rft-1 RNAi with Riboflavin
wt (EV RNAi) atfs-1 (EV RNAi)
hsp-6p::GFP AD1
wt (rft-1 RNAi) atfs-1 (rft-1 RNAi)
1.0 vs *
vs ns
vs *
0.8 vs *
Fraction Alive
0.6
Young Adult Adult Day 1 Adult Day 3 0.4

120 120 120
None None None
Low Low Low
100 100 100 0.2
High High High
Fo
80 80 80
Individuals
Individuals
Individuals
60 60 60
rP 0.0
ee 0 5 10 15 20 25 30 35 40 45
40 40 40
rR Day
20 20 20
ev (d)
0 0 0
iew wt (EV RNAi) atfs-1GOF (EV RNAi)
wt (rft-1 RNAi) atfs-1GOF (rft-1 RNAi)
i
i
i
i
A
A
A
A
N
N
N
N
R
1R
R
1.0 vs
-1
-1
*
EV
EV
EV
-
rft
rft
rft
hsp-6p::GFP Intensity hsp-6p::GFP Intensity hsp-6p::GFP Intensity
vs *
(b) ✱
C07G1.7 hrg-9 ✱ cdr-2 0.8
vs
vs
*
*
2500 15 ✱ 25
Fraction Alive
2000 20 0.6
Fold Change
Fold Change
Fold Change
10
1500 15
0.4
1000 10
5
500 5
0.2
0 0 0
n
n
Ai
Ai
n
n
Ai
Ai
n
n
Ai
Ai
vi
vi
vi
vi
vi
vi
RN
RN
RN
RN
RN
RN
fla
fla
fla
fla
fla
fla
0.0
bo
bo
-1
bo
EV
bo
bo
-1
bo
1
EV
EV
rft
rft
rft
Ri
RI
Ri
Ri
Ri
RI
0 5 10 15 20 25 30 35 40 45
+
+
+
+
+
Ai
Ai
Ai
Ai
Ai
Ai
RN
RN
RN
RN
RN
RN
Day
-1
EV
1
EV
1
EV
-
-
rft
rft
rft
(a) (b) Adult Day 1 Lipid Content (d) acdh-1p ::GFP

Triglyceride/Phospholipid 150
2.5 ✱
✱

125
2.0
aak-2 (EV RNAi) aak-2 (rft-1 RNAi)
Arbitrary Units (A.U)

100
1.5
75
1.0
50
daf-16 (EV RNAi) daf-16 (rft-1 RNAi)
0.5
25
0.0
0
EV RNAi rft-1 RNAi
FC: 1.42
raga-1 (rft-1 RNAi) FC: 1.77
raga-1 (EV RNAi)
Fo
1.8
(c) rP
ee Triglycerides Phospholipids
rR
Fold Change Intensity (rft-1 vs EV RNAi)
60
FC: 0.79
e vie
FC: 1.10 FC: 1.17 FC: 0.89
80
FC: 0.84 FC: 1.03 FC: 1.42 FC: 0.83
1.6
✱ ✱ w ✱ ✱
✱ ✱ ✱ ✱
50
1.4 60
40
% Abdundance
% Abundance
1.2
30 40
1.0 20
20
10
0.8
0 0
ds
0.6
ds
s
ds
yl
n
yl
n
d
i
op
ci
ci
ci
ha
ci
op
ha
wt atfs-1
A
A
A
A
pr
C
pr
C
tty
tty
tty
tty
lo
d
lo
d
he
he
Fa
Fa
yc
Fa
Fa
yc
eri-1 Post Dev RNAi atfs-1GOF
nc
nc
C
C
ed
ed
ed
e d
ra
ra
at
at
at
at
B
ur
ur
ur
ur
daf-16 raga-1
at
t
t
at
rft-1 RNAi
Sa
EV RNAi rft-1 RNAi
Sa
EV RNAi
ns
ns
U
U
aak-2 rict-1 ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901 Support: (434) 964 4100
(a)
bigr-1::RFP acs-2p::GFP
300 ✱ ✱ ✱ 80 ✱
70
EV EV

250
RNAi RNAi 60
200
50
150 40
30
rft-1 100 rft-1
RNAi RNAi 20
50
10
0 0
rft-1 RNAi
Young Adult Adult Day EV
1 RNAi
Adult Day 3 Young Adult Adult Day 1 Adult Day 3
(b) smg-1 (EV RNAi) smg-1; pha-4 (EV RNAi) (c) wt (EV RNAi) raga-1(EV RNAi) (d)
smg-1 (rft-1 RNAi) smg-1; pha-4 (rft-1 RNAi) wt (rft-1 RNAi) raga-1(rft-1 RNAi) Protein Concentration
1.00
vs
1.0
Fo vs *
ns
rP
* 5
vs ns vs
- - -
*
+ + +
ee
Arbitrary Units (A.U)

vs * vs * EV RNAi 4
0.8
vs vs ns
rR - - - + + +
0.75 * rft-1 RNAi 3
ev
Fraction Alive
Fraction Alive
0.6 2
0.50 iew p-S6K

1
0.4 0
Actin A
i
A
i
N N
R R
0.25 EV rft
-1
0.2
0.00 0.0
0 10 20 30 40 50 0 5 10 15 20 25 30 35 40 45 50 55
Day Day
(e) wt (EV RNAi)

wt (rft-1 RNAi)
rict-1 (EV RNAi)
rict-1 (rft-1 RNAi) (f)
1.0
vs *
vs *
0.8 vs ns
vs *
Fraction Alive
0.6
0.4
0.2
0.0
0 5 10 15 20 25 30 35 40 45
Day ScholarOne, 375 Greenbrier Drive, Charlottesville, VA, 22901 Support: (434) 964 4100
Supporting Information
Figure S1: RT-PCR of riboflavin metabolism genes, DAPI staining of germline
Figure S2: DAF-16::GFP images, sod-3p::GFP images
Figure S3: FLIM Methods, Metabolite pathway analysis
Figure S4: Example images of hsp-6p::GFP activation
Figure S5: GC/MS lipid profiles of EV and rft-1 RNAi treated animals
Appendix S1: Supplemental Methods
Supplemental References Fo
rP
ee
rR
ev
iew
Fo
rP
ee
rR
FIGURE S1 (a) Quantitative RT-PCR of downstream flavin processing genes reveals no change in riboflavin
ev
kinase (rfk) and FAD synthetase (flad-1) levels with rft-1 knockdown. Addition of riboflavin upregulates flad-1
expression under both EV and rft-1 knockdown. Results represent three biologic replicates. Bars represent
means  SEM. (b) Differential interference contrast (DIC) and fluorescence imaging of DAPI stained EV and rft-1
iew
RNAi treated adult day 1 animals reveals similar germ line stem cell morphology and the presence of oocytes.
Fo
rP
ee
rR
ev
iew
FIGURE S2 Activation of DAF-16 by riboflavin deficiency. (a) Representative images of DAF-16::GFP high, low, and no (none) nuclear localization at
10X magnification (b) Transfer of rft-1 RNAi treated animals at adult day 1 to empty vector (EV) RNAi plates or riboflavin supplemented plates shows
that nuclear localization of DAF-16 is reversed by removal from rft-1 RNAi and the addition of riboflavin. (c) Representative images of sod-3p::GFP
animals over the lifespan, with significant increases evident from adult day 3 onward with rft-1 RNAi. Persistent activation is seen in intestine, pharynx,
and vulva. (d) Riboflavin depletion promotes lifespan extension in pdk-1 gain of function animals. * indicates P < 0.05 by log rank analysis.
Fo
rP
ee
rR
ev
iew
FIGURE S3 Metabolite two-photon imaging and LC/MS quantification in riboflavin depletion. (a) Representative
images of aak-2;DAF-16::GFP animals treated with EV and rft-1 RNAi reveals nuclear localization with riboflavin
depletion. (b) Representative phasors (top panels) and corresponding LLIF coded images (bottom panels) for
NAD(P)H and FAD Lifetime Images. NAD(P)H images were acquired using 755 nm excitation/ 460 nm detection
and FAD images were acquired using 860nm excitation/525 nm detection. (c) Representative images of total
autofluorescence of NAD(P)H, FAD, and redox ratio map of C. elegans (1-4 respectively). Images of NAD(P)H,
FAD, and redox map are shown only in masked region. (d) Summary plot of quantitative enrichment analysis
(QEA) of metabolites between EV treated and rft-1 RNAi treated animals reveals enrichment of pathways
involved in riboflavin, glutathione and purine metabolism (n=4, P<0.01).
FIGURE S4 Representative images of hsp-6p::GFP expression categories in Figure 4a.

Fo
rP
ee
rR
ev
iew
Fo
rP
ee
rR
ev
iew
FIGURE S5 Lipid analyses with riboflavin depletion indicate increased fat mass. (a) Oil-red-O staining of adult day 1 animals treated with EV and rft-
1. (b) Pseudo color map of (1, 4) total unsaturated fatty acid (area under curve between wavenumbers 2991 and 3022 cm-1), (2, 5) total fatty acid
(area under curve between wavenumbers 2830 and 2870 cm-1) and (3, 6) ratio of TUFA and TFA for EV treated (4-6) and rft-1 RNAi treated (1-3) C.
elegans acquired using SRS imaging. (c) GC/MS analysis of triglyceride and phospholipid fractions in adult day 1 and young adult worms. Ratios
represent fold change of worms treated with rft-1 RNAi compared to EV treated animals (d) Imaging of fat-7p::GFP animals reveals unchanged
expression of fat-7 until adult day 5, where rft-1 RNAi preserves an aging-related decrease in desaturase expression. * indicates P < 0.05 by two-
tailed Student’s t-test (b) and by two-way ANOVA (c). Bars represent means  SE.
Supplemental Methods
DAPI Staining
Adult day 1 animals were collected via washing and washed twice with M9 solution. Animals were then fixed with
40% isopropanol for 3 minutes with shaking. For DAPI staining, a working solution was generated by mixing 2ul
of 1 mg/ml DAPI solution in 1ml of 40% isopropanol. Animals were stained in working solution for 45 minutes.
Imaging was performed on the Leica Thunder A4 setting with 50ms exposure at 20X magnification.
LC/MS Metabolite Analysis
Each worm pellet contained 4000 worms. Worm pellets were transferred to 2 mL impact resistant
homogenization tubes containing 300 mg of 1 mm zirconium beads and 1 mL of 80:20 ethanol:water containing
internal standards(Lagerborg, Watrous et al. 2019). Using an Omni Bead Ruptor Elite, samples were
homogenized in three 10 second cycles at speed 8 m/s with 10 seconds pause between cycles to prevent
overheating. All samples were then transferred to a clean 1.5 mL microcentrifuge tubes and placed at -20 °C for
20 minutes to facilitate protein precipitation. Samples were then centrifuged at 14,000 g for 10 minutes at 4 °C.
After which 150 uL of supernatant was processed through solid-phase extraction to isolate lipid metabolites using
a Phenomenex Stata-X polymeric 10 mg/mL 96-well SPE plate, as previously described. For polar metabolites,
50 uL of supernatant was dried in vacuo and resuspended in 50 uL of 40:40:20 acetonitrile:methanol:water
containing 1 ng/uL of L-Phenylalanine-13C9,15N.
Fo
All visualization and significance testing of metabolomics was conducted using the MetaboAnalyst 5.0 package
(Pang, Chong et al. 2021). Mass integration values for 121 identified metabolites were extracted from full-scan
rP
LC-MS/MS measurements of L4 wildtype (Bristol N2) C. elegans treated with L4440 vector (EV) or rft-1 RNAi
throughout larval development (n = 4 independent biological replicates). Abundance values were subsequently
log10 transformed and normalized by mean centering and division by the standard deviation of each variable
ee
(auto-scaling). Normalized abundance values were then assessed for statistical significance via independent
two-sample t-tests followed by false discovery rate (FDR) control using the Benjamini-Hochberg (BH) method.
Volcano plot visualizations were plotted using the -log10(FDR) in comparison to the log2(fold change) between
rR
normalized rft-1 RNAi/EV using Prism 9 (GraphPad Software). Metabolites were considered differentially
abundant with an FDR controlled P value < 0.05. The top 30 metabolites across treatment (ranked by t statistic
and FDR value) were visualized using a heatmap, with hierarchical clustering of samples and normalized
ev
compound abundances included. Quantitative enrichment analysis (QEA) was conducted on differentially
abundant metabolites using internal R package globaltest, calculating an average Q statistic across all
iew
compounds belonging to a given metabolic pathway as defined by the KEGG database (Goeman, van de Geer
et al. 2004, Kanehisa, Sato et al. 2016). Dotplots were generated by comparing the -log10(FDR) with the Q
enrichment ratio of the top 25 most enriched metabolite sets.
Two-Photon and Lifetime Imaging
C. elegans were mounted over 5–10% agarose pads on glass slides and 0.25–2 µL of a suspension of
polystyrene beads (Polysciences, 2.5% by volume, 0.1 µm diameter) were applied around the sample and on
the coverslip. The coverslip was placed upside down and worms were immobilized in gel-microbead matrix
between glass slide and coverslip. The cover glass was sealed with nail polish to prevent dehydration.
As per our previous studies (Quinn, Sridharan et al. 2013, Varone, Xylas et al. 2014, Liu, Pouli et al. 2018),
NAD(P)H images were acquired with an excitation wavelength of 755 nm and recorded at 460 ± 25 nm using a
non-descanned detector. Fluorescence spectra of NADH and NADPH are indistinguishable, and their combined
fluorescence is represented as NAD(P)H. FAD images were acquired with an excitation wavelength of 860 nm
wavelength and recorded at 525 ± 25 nm using a non-descanned detector. Laser light was focused on the
sample using a water immersion 40x objective with 1.1 numerical aperture and average laser powers of 24 mW
for both wavelengths. The TPEF images (290 × 290 μm, 1024 × 1024 pixels, 1 zoom) were acquired with pixel
dwell time of 0.7 μs and 8-line average.
For analysis of the intensity and lifetime TPEF images, the intestine region of the C. elegans was selected
manually relying on the integrated intensity of the NAD(P)H lifetime images, since the autofluorescence signal
was significantly higher than in the surrounding tissues. Pixel-wise lifetime data was analyzed using the phasor
approach, as described previously (Liu, Pouli et al. 2018). Briefly, after Fourier decomposition at the repetition
frequency of the laser, the normalized cos and sin components were represented along the x and y axis,
respectively, so that the fluorescence lifetime spectrum from each pixel was represented by a single point in
phasor space. The phasors of spectra corresponding to a single exponential decay were represented by points
that lay on the universal semi-circle, while biexponential decays like that of NAD(P)H and FAD, were represented
by ellipsoid shapes within the semicircle. A line fit to the major axis of such an ellipsoid crossed the universal
semi-circle at points that corresponded to the short and long lifetime components. The relative distance of the
ellipsoid centroid on the line represented the long lifetime intensity fraction (LLIF) of the fluorophore, which is
correlated with the mean lifetime (Alonzo, Karaliota et al. 2016, Liu, Meng et al. 2021). The original phasors of
C elegans presented more complex shapes indicating the presence of additional fluorophores. The
corresponding fluorescence lifetime images coded according to the LLIF value revealed the presence of
numerous granule-like structures with blue hues corresponding to very low LLIF values. Strong fluorescence
from such granules has been previously reported and attributed to tryptophan metabolite anthranilic acid glucosyl
esters or advanced glycation end products (Coburn and Gems 2013, Teuscher and Ewald 2018, Komura,
Yamanaka et al. 2021). To remove contributions from these granules, all pixels yielding NAD(P)H LLIF values
lower than 0.4 were identified and removed. Lipid droplets have also been identified in C. elegans intestinal cells.
Since lipids can autofluoresce (Datta, Alfonso-García et al. 2015, Alonzo, Karaliota et al. 2016), we implemented
Fo
an additional thresholding procedure established in our previous study to identify and remove such contributions
(Alonzo, Karaliota et al. 2016). This method relied on the observation that lipid droplets within white and brown
adipose tissues were characterized by high LLIF in the 755/460nm channel (Alonzo, Karaliota et al. 2016). Three
rP
level Otsu thresholding was applied to the images in the 755/460nm and 860/525 nm channel images to remove
weakly autofluorescent nuclei and intercellular regions. Then the regions with LLIF of more than 0.7 were
identified as lipid droplets and removed from the image. When contributions from the pixels removed based on
ee
low and high LLIF thresholding were also removed from the phasors associated with the fluorescence associated
with 860 nm excitation/525 nm detection, there was limited impact on the shape of the phasor, indicating that
rR
lipid droplet fluorescence was not prominent in that wavelength regime, consistent with our previous findings
(Alonzo, Karaliota et al. 2016). The regions that contained pixels with LLIF values below the low threshold and
above the high LLIF threshold were dilated by applying a 5 × 5 pixel averaging window on the corresponding
ev
binary masks to ensure that borders from these organelles were excluded from the cytoplasmic region used to
assess NAD(P)H and FAD contributions. The optical redox ratio, was calculated as the pixel-wise ratio of
FAD/(NAD(P)H+FAD) intensity images. The intensity fluctuations with the NAD(P)H images were analyzed used
iew
approaches described in detail previously to assess the levels of mitochondrial clustering/fragmentation (Pouli,
Balu et al. 2016, Liu, Pouli et al. 2018)
References
Alonzo, C. A., S. Karaliota, D. Pouli, Z. Liu, K. P. Karalis and I. Georgakoudi (2016). "Two-photon excited fluorescence of
intrinsic fluorophores enables label-free assessment of adipose tissue function." Sci Rep 6: 31012.
Coburn, C. and D. Gems (2013). "The mysterious case of the C. elegans gut granule: death fluorescence, anthranilic acid
and the kynurenine pathway." Front Genet 4: 151.
Datta, R., A. Alfonso-García, R. Cinco and E. Gratton (2015). "Fluorescence lifetime imaging of endogenous biomarker of
oxidative stress." Sci Rep 5: 9848.
Goeman, J. J., S. A. van de Geer, F. de Kort and H. C. van Houwelingen (2004). "A global test for groups of genes: testing
association with a clinical outcome." Bioinformatics 20(1): 93-99.
Kanehisa, M., Y. Sato, M. Kawashima, M. Furumichi and M. Tanabe (2016). "KEGG as a reference resource for gene and
protein annotation." Nucleic Acids Res 44(D1): D457-462.
Komura, T., M. Yamanaka, K. Nishimura, K. Hara and Y. Nishikawa (2021). "Autofluorescence as a noninvasive biomarker
of senescence and advanced glycation end products in Caenorhabditis elegans." NPJ Aging Mech Dis 7(1): 12.
Lagerborg, K. A., J. D. Watrous, S. Cheng and M. Jain (2019). "High-Throughput Measure of Bioactive Lipids Using Non-
targeted Mass Spectrometry." Methods Mol Biol 1862: 17-35.
Liu, Z., J. Meng, K. P. Quinn and I. Georgakoudi (2021). "Tissue Imaging and Quantification Relying on Endogenous
Contrast." Adv Exp Med Biol 3233: 257-288.
Fo
Liu, Z., D. Pouli, C. A. Alonzo, A. Varone, S. Karaliota, K. P. Quinn, K. Munger, K. P. Karalis and I. Georgakoudi (2018).
"Mapping metabolic changes by noninvasive, multiparametric, high-resolution imaging using endogenous contrast." Sci
Adv 4(3): eaap9302.
rP
Liu, Z., D. Pouli, C. A. Alonzo, A. Varone, S. Karaliota, K. P. Quinn, K. Münger, K. P. Karalis and I. Georgakoudi (2018).
"Mapping metabolic changes by noninvasive, multiparametric, high-resolution imaging using endogenous contrast." Sci
Adv 4(3): eaap9302.
ee
Pang, Z., J. Chong, G. Zhou, D. A. de Lima Morais, L. Chang, M. Barrette, C. Gauthier, P. E. Jacques, S. Li and J. Xia (2021).
"MetaboAnalyst 5.0: narrowing the gap between raw spectra and functional insights." Nucleic Acids Res 49(W1): W388-
W396.
rR
Pouli, D., M. Balu, C. A. Alonzo, Z. Liu, K. P. Quinn, F. Rius-Diaz, R. M. Harris, K. M. Kelly, B. J. Tromberg and I.
Georgakoudi (2016). "Imaging mitochondrial dynamics in human skin reveals depth-dependent hypoxia and malignant
potential for diagnosis." Sci Transl Med 8(367): 367ra169.
ev
Quinn, K. P., G. V. Sridharan, R. S. Hayden, D. L. Kaplan, K. Lee and I. Georgakoudi (2013). "Quantitative metabolic
imaging using endogenous fluorescence to detect stem cell differentiation." Sci Rep 3: 3432.
Teuscher, A. C. and C. Y. Ewald (2018). "Overcoming Autofluorescence to Assess GFP Expression During Normal
iew
Physiology and Aging in Caenorhabditis elegans." Bio Protoc 8(14).

Varone, A., J. Xylas, K. P. Quinn, D. Pouli, G. Sridharan, M. E. McLaughlin-Drubin, C. Alonzo, K. Lee, K. Munger and I.
Georgakoudi (2014). "Endogenous two-photon fluorescence imaging elucidates metabolic changes related to enhanced
glycolysis and glutamine consumption in precancerous epithelial tissues." Cancer Res 74(11): 3067-3075.

ACE-22-0212 Proof Hi

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

ACE-22-0212 Proof Hi

Uploaded by

Copyright:

Available Formats

Aging Cell

Riboflavin Depletion Promotes Longevity and Metabolic

Journal: Aging Cell

Wiley - Manuscript type: Research Article

Medicine; Harvard Medical School

Ahsan, Fasih; Massachusetts General Hospital, Center for Genomic

Li, Sainan; Massachusetts General Hospital, Center for Genomic Medicine

Riboflavin, longevity, rft-1, AMPK, FOXO, dietary restriction, C. elegans,

Riboflavin Depletion Promotes Longevity and Metabolic Hormesis in Caenorhabditis elegans

Boston, MA 02114, USA

AGING CELL AUTHOR CHECKLIST

2 Yes Yes 16 x 15.5 6

5 Yes Yes 37 x 22.5 6

290 Experimental Methods

318 Longevity assays

324 software (Han, Lee et al. 2016).

372 Quantitative RT-PCR

428 Fluorescent Microscopy

434 of two biological replicates).

475 Conflict of Interest Statement: The authors report no competing interests.

491 Figure Legends:

518 replicates of survival analyses. Bars represent means  SEM.

561 TORC1-mediated phosphorylation of S6KT389) reveals no change in phospho-S6K concentrations between

684 Supporting Information

AGING CELL AUTHOR CHECKLIST

2 Yes Yes 16 x 15.5 6

5 Yes Yes 37 x 22.5 6

(a) wt (EV RNAi) wt (EV RNAi) + Riboflavin (b) rft-1 rft-2

Normalized Concentration (A.U)

(d) ✱ ✱ (e) vie (f)

Pumping Rate (pumps/min)

Fraction Reaching Adulthood

No Riboflavin Suppl. Riboflavin

wt (EV RNAi) aak-2 (EV RNAi)

ADENOSINE TRIPHOSPHATE XANTHINE

(d) Redox Ratio Mito Clustering

NAD(P)H Bound Fraction FAD Bound Fraction

Young Adult Adult Day 1 Adult Day 3 0.4

(a) (b) Adult Day 1 Lipid Content (d) acdh-1p ::GFP

Signal Intensity (Arbitrary Units)

Arbitrary Units (A.U)

EV RNAi rft-1 RNAi

Signal Intensity (Arbitrary Units)

Arbitrary Units (A.U)

0.50 iew p-S6K

(e) wt (EV RNAi)

FIGURE S4 Representative images of hsp-6p::GFP expression categories in Figure 4a.

Physiology and Aging in Caenorhabditis elegans." Bio Protoc 8(14).

You might also like