Microtubule regulators act in the nervous system to modulate fat metabolism and longevity through DAF‐16 in C. elegans

Abstract Microtubule (MT) regulation is involved in both neuronal function and the maintenance of neuronal structure, and MT dysregulation appears to be a general downstream indicator and effector of age‐related neurodegeneration. But the role of MTs in natural aging is largely unknown. Here, we demonstrate a role of MT regulators in regulating longevity. We find that loss of EFA‐6, a modulator of MT dynamics, can delay both neuronal aging and extend the lifespan of C. elegans. Through the use of genetic mutants affecting other MT‐regulating genes in C. elegans, we find that loss of MT stabilizing genes (including ptrn‐1 and ptl‐1) shortens lifespan, while loss of MT destabilizing gene hdac‐6 extends lifespan. Via the use of tissue‐specific transgenes, we further show that these MT regulators can act in the nervous system to modulate lifespan. Through RNA‐seq analyses, we found that genes involved in lipid metabolism were differentially expressed in MT regulator mutants, and via the use of Nile Red and Oil Red O staining, we show that the MT regulator mutants have altered fat storage. We further find that the increased fat storage and extended lifespan of the long‐lived MT regulator mutants are dependent on the DAF‐16/FOXO transcription factor. Our results suggest that neuronal MT status might affect organismal aging through DAF‐16‐regulated changes in fat metabolism, and therefore, MT‐based therapies might represent a novel intervention to promote healthy aging.

and stability can be modulated by a large number of MT-interacting proteins. Some MT-associated proteins (MAPs), such as Tau, bind along the MT structure to promote MT assembly and stability by protecting MTs from severing by destabilizing factors (Kapitein & Hoogenraad, 2015). Due to the larger cytoplasmic volume and polarized morphology, neurons rely more on MT function compared with other cell types. MTs serve as tracks for long-distance transport and play a critical role in the establishment and maintenance of neuronal structure and polarity (Kapitein & Hoogenraad, 2015). Consequently, abnormalities in MT organization and dynamics, as well as MT protein expression or distribution, have been observed in many neurological disorders. Mutations in tubulin genes or MAPs have also been reported to affect neuronal integrity during aging (Chew, Fan, Gotz, & Nicholas, 2013;Pan, Peng, Chen, & McIntire, 2011). The importance of MT regulation in healthy neuronal aging is underscored by the critical role of Tau in MT stabilization and its dysfunction related to neurodegenerative diseases where the impairment of axonal transport is a common factor (Baird & Bennett, 2013).
The nervous system is the central integrator of information regarding the internal milieu of an organism and the external environment, and also plays a key role in regulating aging and longevity (Alcedo, Flatt, & Pasyukova, 2013). It allows animals to process and transmit extrinsic signals to neuronal or non-neuronal endocrine cells that regulate the release of hormones involved in growth and metabolism. In addition, studies have shown that chemosensory neurons regulate animal longevity in C. elegans. The laser ablation of gustatory neurons or olfactory neurons is sufficient to extend lifespan, and these two types of neurons function in parallel in regulating lifespan (Alcedo & Kenyon, 2004). Mutations that disrupt the sensory cilia used by these neurons also lead to defective sensory perception and increased longevity (Apfeld & Kenyon, 1999). Similarly, the loss of olfactory function in Drosophila has also been shown to increase lifespan (Libert et al., 2007). The neuronal influence on lifespan is mediated by several mechanisms including alterations in the daf-2/insulin/insulin-like growth factor-1 signaling (IIS) pathway (Alcedo & Kenyon, 2004;Apfeld & Kenyon, 1999). This pathway plays an evolutionarily conserved role in regulating longevity, stress resistance, and metabolism in various model organisms (Partridge, Alic, Bjedov, & Piper, 2011;Taguchi & White, 2008;Tatar, Bartke, & Antebi, 2003).
Given the critical role of MTs in regulating neuronal function and key role of the nervous system in modulating aging, we tested the hypothesis that MT regulators can affect neuronal function to thereby alter organismal aging. MT regulator mutants have been implicated in the aging process, but they have not been specifically studied and the underlying mechanisms are unknown. For example, loss of PTL-1, the C.elegans ortholog of Tau, has been shown to accelerate neuronal aging and shorten animal lifespan (Chew et al., 2013). Loss of MT regulator Stathmin resulted in reduced levels of stable axonal MTs and shortened lifespan in Drosophila (Duncan, Lytle, Zuniga, & Goldstein, 2013). In this study, we demonstrate that loss of EFA-6, a negative regulator of MT growth, delays neuronal aging and extends organismal lifespan and health span in C. elegans.
Mutations in other MT regulator genes, including hdac-6 and ptrn-1, can also affect lifespan. We show by tissue-specific transgene rescue that these MT regulators function in the nervous system to modulate longevity. Through the use of RNA-seq analysis, we show that genes involved in lipid metabolism are differentially expressed in MT regulator mutants and lead to an increase in fat storage. Lastly, via epistasis experiments we show that the lifespan extension phenotypes of the efa-6 and hdac-6 mutants depend on the DAF-16/FOXO transcription factor. Our results suggest that neuronal MT status can affect organismal longevity through modulating fat metabolism.

| Loss of EFA-6 delays neuronal aging and extends lifespan
We have previously identified EFA-6 as a modulator of neuronal MT dynamics . MTs are crucial cytoskeleton for neuronal integrity and MT defects have been characterized in different neurodegenerative conditions (Dubey, Ratnakaran, & Koushika, 2015). As a result, we tested whether EFA-6 plays a role in maintaining neuronal integrity during aging. Mechanosensory neurons (touch neurons) offer an excellent model to study changes associated with neuronal aging in C. elegans (Toth et al., 2012). We used a mec-7p:: GFP reporter to label mechanosensory neurons and then examined the effect of age on their structure. During aging, these axons of these neurons accumulated branches and blebs, and cell bodies displayed branching as the animals aged. (Figure 1a-d). We found that these age-dependent morphological changes were significantly delayed in efa-6(tm3124) mutants. By Day 9 of adulthood, more than 50% of the wild-type animals showed blebbing/branching phenotype, whereas only around 30% of efa-6 mutants at Day 9 displayed this defect. By Day 12, the proportion of wild-type animals with abnormal touch neuron morphology reached 70%, compared to less than 50% in efa-6 mutants at this stage ( Figure 1d). This delayed neuronal aging in efa-6 mutants is contrary to the previously reported, accelerated neuronal aging phenotype of the ptl-1 mutants, which in contrast to efa-6 mutations results in the destabilization of the MTs (Chew et al., 2013).
MT-based transport is essential for neuronal function, and the impairment of this process is a common factor in several neurodegenerative diseases (Franker & Hoogenraad, 2013). To test whether loss of EFA-6 affects intracellular transport, we examined the distribution of synaptic vesicles that depend on MT-based transport.
Aging wild-type animals displayed an increase in the ectopic accumulation of synaptic vesicles labeled by RAB-3::GFP in the dendritic region of the PLM neuron (Figure 1e-f). This age-dependent change in the dendritic distribution of RAB-3::GFP was partially rescued in efa-6 mutants but exaggerated in ptl-1 mutants (Figure 1e-f). Sensitivity to light touches, a function of mechanosensory neurons, decreases in aged wild-type animals (Jiang et al., 2015). Consistent with the age-associated morphological changes and ectopic RAB-3 distribution, we found that the touch sensitivity was better maintained in aged efa-6 mutants but further diminished in aged ptl-1 mutants ( Figure 1g).
As worms age, their mobility declines, and the correlation between age-dependent decreases in mobility and defects in neuronal morphology has been previously reported (Tank, Rodgers, & Kenyon, 2011). Since loss of EFA-6 delayed neuronal aging, we asked whether it affected age-dependent mobility decline. We measured the speed of the animal bending in liquid during aging. We found that the movement speed of the animals gradually declined with age in the wild-type animals. The efa-6(tm3124) mutants displayed significantly higher movement speed at all stages examined (Supporting information Figure S1a), and they showed a slower rate of reduction in the relative speed (normalized to Day 1) ( Figure 1h).
Overall, these results suggest that the loss of EFA-6, a regulator of MT dynamics, could delay age-dependent changes in neuronal morphology and neuronal function that would lead to improved mobility during aging.
Given the involvement of EFA-6 in neuronal age-related phenotypes, we examined the lifespan of the efa-6 mutants. We found efa-6(tm3124) had a longer lifespan compared to wild-type controls, with (i) Survival curve for wild-type and two independent efa-6 deletion mutants. ***p < 0.0001 for wt vs tm3124, **p = 0.0032 for wt vs ju1200. Log-rank test. (j) Number of progenies per 24 hr from wild-type and efa-6(tm3124) mutants at different stages. N ≥ 12 for each bar. Student's t test a more obvious increase in mean lifespan than that in maximal lifespan ( Figure 1i). When we performed the lifespan assays using NGM plates that do not contain FUDR, we noted that efa-6(tm3124) mutants continued to lay eggs until later in adulthood, when the wild-type animals had already stopped producing progeny. We then evaluated this extended reproduction period by measuring the number of progenies produced by wild-type and efa-6 mutant worms from Days 1 through 6 of adulthood. We found that the efa-6 mutants had slightly fewer progeny on Day 1 and Day 2, but after Day 3, the efa-6 mutants produced more progeny compared to wildtype worms of the same age, with total progeny number similar to wild-type ( Figure 1j, Supporting information Figure S1b-c).

| EFA-6 functions in neurons to regulate organismal longevity
We next sought to confirm that the effects observed in our efa-6 (tm3124) worms were due to the loss of function of efa-6 via the use of tissue-specific transgenes expressing eaf-6. While high level expression of efa-6 in the nervous system can disrupt development , lower level of expression did revert the efa-6 neuronal aging phenotype back to wild-type without causing developmental defects (Figure 2a). We also found that neuronal expression of efa-6 was able to completely rescue the enhanced mobility of the efa-6 mutants during aging (on both Day 10 and Day 14), whereas muscle-specific expression of efa-6 had no effect (Figure 2b). Remarkably, neuronal expression of efa-6 was able to completely rescue the lifespan phenotype of efa-6(tm3124) mutant. In contrast, a transgene expressing efa-6 only in the muscle failed to rescue the lifespan phenotype ( Figure 2c). We further tested tissuespecific EFA-6 expression in intestine (Pges-1), epidermis (Pdpy-7), and germline (Ppie-1) for the effect on lifespan extension at efa-6 mutant background. The germline and epidermis-specific transgenes were not able to rescue the lifespan phenotype of efa-6, while the intestinal transgenic expression of EFA-6 was able to modestly rescue (Supporting information Figure S2). These data suggest that EFA-6 can act in neurons to modulate organismal aging.
The EFA-6 protein is dependent on its N-terminus to regulate cortical MT growth and axon regeneration O'Rourke, Christensen, & Bowerman, 2010). We next tested whether the N-terminus of the EFA-6 protein is also required for its role in longevity. The neuronal expression of a 150 amino acid fragment from the N-terminus, but not the EFA-6 protein lacking the N-terminus (EFA-6ΔN), could reduce the increased lifespan of the efa-6 mutant ( Figure 2d). Together these results suggest that the N-terminus is necessary for the effects of efa-6 on longevity. Neuronal expression (Pregf-1), but not muscular expression (Pmyo-3), of EFA-6 is able to rescue the lifespan extension phenotype in efa-6(tm3124) mutants. (d) Survival curves for indicated genotypes. The N-terminus of EFA-6 is required for its function. Transgenic expression of the N-terminus, but not a truncated EFA-6 lacking N-terminus, rescues lifespan phenotype in efa-6(tm3124) mutants

| Mutations in microtubule regulators affect longevity
We then tested whether efa-6 was unique or whether other MT regulators played similar roles in modulating longevity. We first tested the hdac-6(ok3203) mutant, which affects a histone deacetylase that also targets nonhistone proteins, including α-tubulin (Li, Jiang, Chang, Xie, & Hu, 2011). Specifically, HDAC6 deacetylates α-tubulin to reduce MT stability (Matsuyama et al., 2002). We therefore tested the lifespan of hdac-6 mutant animals containing a deletion mutation.
We observed extended lifespan and enhanced mobility in hdac-6 (ok3203) mutant, similar to efa-6 mutants (Figure 3a,b). To ask whether hdac-6 acts in the nervous system to modulate longevity, we expressed HDAC-6 using pan-neuronal promoter in the hdac-6 (ok3203) background. We found that neuronal expression of HDAC-6 could completely revert the extended lifespan to a wild-type level ( Figure 3c), indicating that hdac-6 also acts within neurons. Inhibiting HDACs is known to extend lifespan in Drosophila and C. elegans through epigenetic effects (Pasyukova & Vaiserman, 2017), although a role for HDAC6 in aging has not been reported previously.
HDAC6 belongs to the class IIb family of HDACs, which is predominantly localized in the cytosol and prefers nonhistone targets (Simoes-Pires et al., 2013). A unique feature of HDAC6 is the presence of two catalytic domains. The second deacetylase domain (DD2) has been demonstrated to deacetylate tubulin specifically (Haggarty, Koeller, Wong, Grozinger, & Schreiber, 2003;Kaluza et al., 2011). Loss of DD2 activity in HDAC6 is sufficient to rescue the MT defects induced by ectopic human tau expression in Drosophila (Xiong et al., 2013). To test whether the tubulin-specific deacetylase activity is responsible for the role of HDAC-6 in regulating longevity, we generated transgenic animals with pan-neuronal expression of HDAC-6 (H146A) or HDAC-6 (H561A), two mutant forms with the conserved catalytic residue in DD1 and DD2 mutated. We found that both mutant HDAC-6 proteins were able to partially rescue the lifespan phenotype in hdac-6 mutant, but DD2 mutant had a weaker rescuing effect ( Figure 3d). These data indicate that tubulin-specific deacetylase activity conferred by DD2 is perhaps more critical for the role of HDAC-6 in regulating lifespan.
We further extended our study to more MT regulator genes including ptrn-1 and ptl-1. The CAMSAP family of proteins are a group of conserved, MT minus end-binding proteins. PTRN-1, the CAMSAP homolog in C. elegans, promotes MT stability in neurons and epidermis (Marcette, Chen, & Nonet, 2014;Richardson et al., 2014;Wang et al., 2015). ptrn-1 mutants showed shortened lifespan and the difference between control and ptrn-1 was moderate but significant. Similar to efa-6 and hdac-6 mutants, the lifespan phenotype in ptrn-1 mutant could be rescued by the transgenic expression of ptrn-1 in neurons ( Figure 3e). PTL-1 has been previously reported to regulate both neuronal and organismal aging (Chew et al., 2013).
Consistent with the previous studies, we found that ptl-1(ok621) null mutants displayed shortened lifespan, which can be rescued by panneuronal expression of PTL-1 (Figure 3f), further supporting that neuronal MT regulation can affect longevity.

| Aberrant expression of genes involved in fat metabolism and altered fat storage in microtubule regulator mutants
To understand how the MT regulator mutants might affect animal aging, we performed RNA-seq to identify genes that are differentially regulated in four of these mutants, including the long-lived efa-6 (tm3124) and hdac-6(ok3203), and the short-lived ptl-1(ok621) and ptrn-1(tm5597) mutants. We compared the transcriptome of each of these mutants with that of wild-type control and identified several hundreds of up-and down-regulated genes with the threshold set at ***p < 0.001 (Figure 4a, b, Supporting information Table S1). To narrow down the potential effector genes, we focused on the overlapped genes between the two long-lived MT mutants (efa-6 and hdac-6), as well as those between the two short-lived MT mutants (ptl-1 and ptrn-1). We then used the DAVID functional annotation program to identify biologic themes within the common up-and down-regulated genes. Within the common up-regulated genes in efa-6 and hdac-6 mutants, we found that genes involved in lipid metabolism are significantly enriched (Figure 4a,b and Supporting information Table S1). We also found that histone genes are enriched in the common down-regulated genes between efa-6 and hdac-6 mutants, as well as in the common up-regulated genes in ptl-1 and ptrn-1 mutants. Multiple histone modification has been linked to aging regulation through epigenetic mechanisms (Tessarz et al., 2014). Elevated histone expression has been shown to extend lifespan in yeast, and RNAi of C. elegans histone genes has been reported to extend the lifespan of daf-2;aak-1;aak-2 dauer larvae (Xie & Roy, 2012), but the effects of the changes in histone gene expression on aging of multicellular organisms are currently unclear.
We next looked into the genes whose expression is oppositely affected between the two long-lived (efa-6 and hdac-6) and the two short-lived (ptl-1 and ptrn-1) mutants ( to extend lifespan and enhance fat storage in wild-type animals (Cheong et al., 2013). acdh-1 encodes a short-chain acyl-CoA dehydrogenase that catalyzes fatty acid beta-oxidation. ACDH-1 protein level is up-regulated in eat-2 mutants, and acdh-1 RNAi significantly decreases the lifespan of eat-2 (Yuan et al., 2012).
Given that genes involved in fat metabolism were differentially expressed in MT mutants, we tested whether fat metabolism was affected in these mutants. We compared fat staining in fixed wildtype and mutant animals using the lipophylic dye Nile Red, which stains neutral lipids in fixed tissues (Brooks & Liang, 2009). We found that the intensity of stained lipid droplets is significantly reduced in ptl-1, but increased in efa-6 mutants (Figure 4c), correlating with the changes in lifespan of these mutants. We also confirmed the fat staining phenotype using Oil Red O staining EFA-6 and PTL-1 act in the nervous system to regulate fat accumulation, we performed transgenic rescue experiments. We found that the fat storage phenotype in efa-6 and ptl-1 mutants could be rescued by neuron-specific transgenic expression of EFA-6 and PTL-1, respectively ( Figure 4c). Emerging studies from both invertebrates and mammals suggest that elevated fat storage can affect longevity (Hansen, Flatt, & Aguilaniu, 2013;Steinbaugh et al., 2015). To test whether lipid metabolism is critical for the role of MT regulators in modulating longevity, we treated efa-6 mutants with acdh-1 RNAi.
acdh-1 RNAi partially suppressed the lifespan extension in efa-6 mutants, but did not affect lifespan at wild-type background ( Ogg & Ruvkun, 1998). We found that daf-18 RNAi significantly shortened the lifespan and displayed additive effect with efa-6(tm3124) (Figure 5h), suggesting that efa-6 might not function directly through IIS pathway.

As FOXO transcription factors play a conserved role in regulat-
ing energy homeostasis and lipid metabolism, we tested whether the fat storage effect in MT regulator mutants was dependent on DAF-16. We found that the increase of fat storage in efa-6 mutants was significantly attenuated in the efa-6;daf-16 mutants (Figure 6ab). During C. elegans aging, touch sensory neurons display morphological changes, and daf-16 is required to maintain touch neuron integrity (Pan et al., 2011;Tank et al., 2011;Toth et al., 2012). We tested whether the delayed neuronal aging in efa-6 mutants was dependent on DAF-16. We found that daf-16 mutants showed mild defects in touch neuron morphology compared to wild-type control

| Microtubule-associated proteins as novel longevity regulators
Microtubule dysfunction is associated with age-related neurodegenerative diseases (Dubey et al., 2015). Microtubule regulators have been implicated, but not been directly linked to lifespan modulation.
Recently, studies on PTL-1, the worm homolog of mammalian Tau protein, have demonstrated that PTL-1 is involved in age-related preservation of neuronal structural integrity and in promoting a normal lifespan (Chew et al., 2013). Similarly, a previous study in Drosophila showed that loss of stathmin, a microtubule regulator, led to a reduction in axonal MT stability and a decrease in lifespan (Duncan et al., 2013). These studies suggested that a disruption in neuronal MTs might be detrimental for longevity. However, whether manipulating MTs can promote longevity remains unknown.
Here, we demonstrate that EFA-6 normally negatively regulates We found that efa-6 and hdac-6 mutants are long-lived, while ptl-1and ptrn-1 mutants are short-lived. The opposing phenotypes in the two groups of mutants suggest a connection between their effects on MTs and the resulting effect on aging. Specifically, PTL-1 and PTRN-1 have been shown to promote MT stability. PTL-1 is highly conserved in both structure and function to Tau, which binds to and stabilizes MTs (Drubin & Kirschner, 1986). PTRN-1 promotes MT stability in C. elegans neurons and epidermis (Marcette et al., 2014;Richardson et al., 2014;Wang et al., 2015). On the other hand, EFA-6 appears to promote MT catastrophe (O'Rourke et al., 2010), and HDAC6 deacetylates alpha-tubulin and negatively regulates MT stability (Matsuyama et al., 2002). Hence, it appears that increased MT stability leads to increased organismal longevity whereas decreased stability shortens lifespan. In people, reduced MT stability is associated with several neurodegenerative diseases such as Alzheimer's disease and Parkinson's disease. However, hyperstable MTs can also lead to neurodegeneration as found in Hereditary Spastic Paraplegia (Dubey et al., 2015). Therefore, the right level of MT stability might be critical for neuronal function. Perturbing the balance in MT dynamics might disrupt neuronal function, while promoting the homeostasis is protective for neurons to preserve function during aging. Further investigation on how those MT regulators impact MT dynamics will be needed to understand their roles influencing the aging process.

| MT regulators act in the nervous system to influence organismal longevity
The nervous system is known to modulate lifespan in diverse species. It detects sensory cues from the environment and internal signals from the animal, and coordinates organismal metabolic homeostasis and energy balance . Previous studies have shown that chemosensory neurons regulate animal longevity in C. elegans (Alcedo & Kenyon, 2004). Mutations that disrupt sensory cilia lead to defective sensory perception and promote longevity (Apfeld & Kenyon, 1999). Similarly, loss of olfactory function in Drosophila has also been shown to increase lifespan (Libert et al., 2007). Later studies have then shown that these sensory neurons influence longevity through multiple mechanisms. First, specific sensory neurons synthesize INS-6 and DAF-28, which act via DAF-2 to inhibit the accumulation of DAF-16 in the nucleus and the subsequent activation of DAF-16 target genes (Kenyon, 2010;Riera, Merkwirth, Magalhaes Filho, & Dillin, 2016). Second, lifespan extension conferred by dietary restriction has been shown to be mediated by SKN-1 that acts in the ASI sensory neurons (Bishop & Guarente, 2007). Finally, the AFD thermosensory neurons promote lifespan at warm temperature by up-regulating the DAF-9 sterol hormone signaling (Lee & Kenyon, 2009), while cool-sensitive IL1 neurons promote lifespan at cool temperature through DAF-16 in the intestine (Xiao et al., 2013;Zhang et al., 2018). Besides these direct effects of sensory neurons on longevity, neurons can also detect diverse cellular stressors and trigger animal-wide stress responses to protect the animal and ensure survival (Durieux, Wolff, & Dillin, 2011;Prahlad, Cornelius, & Morimoto, 2008;Tatum et al., 2015;Taylor & Dillin, 2013). Therefore, different neuron types respond to distinct external and internal cues and activate distinct neuronal signals and signaling pathways to extend or shorten lifespan.
Consistent with the important role identified for the nervous system in aging, we find that the MT regulators appear to act in the nervous system to modulate longevity, as the lifespan phenotypes in MT regulator mutants are rescued by neuron-specific transgenic expression of these MT regulators. However, MT regulators in the intestine might also have an effect on longevity, as intestinal expression of EFA-6 can partially rescue the longevity phenotype in efa-6 mutants. It remains to be determined whether their function is required in all neurons or if a particular subset of neurons is more important for their effect on aging. Since the MT cytoskeleton plays critical structural role in cilia, and many genetic mutants that have impaired ciliary structures or functions display altered lifespan (Apfeld & Kenyon, 1999), it is possible that the MT regulators could regulate cilia function to affect lifespan. However, many of the longlived mutations affecting cilia lead to the disruption of ciliary structure, whereas we find that enhancing MT stability promotes instead of reduces lifespan. This could instead suggest a different mechanism such as alterations in the localization or release of vesicles encoding neurotransmitters or neuropeptides. More work will be needed to understand the mechanism(s) involved.

| The role of MT regulators in regulating aging and metabolism requires DAF-16
Insulin is involved in many cellular processes and the IIS pathway is known to play conserved roles in regulating metabolism, stress resistance, and lifespan (Kenyon, 2005). As one of the most important downstream transcription factors in IIS pathway, DAF-16/FOXO regulates genes are involved in dauer formation, fat storage, stress response, and longevity (Murphy et al., 2003). IIS within the nervous system has been shown to be important for its role in longevity regulation in both C. elegans and mice (Iser et al., 2007;Taguchi et al., 2007;Wolkow et al., 2000). IIS pathway is involved in age-dependent decay of neuron integrity. During aging, touch sensory neurons develop blebs and aberrant branches. These age-dependent morphological changes are delayed in daf-2 mutants, and daf-16 is required for the beneficial effects in daf-2 mutants (Pan et al., 2011;Tank et al., 2011;Toth et al., 2012). We report that loss of EFA-6 delays neuronal aging phenotypes in a DAF-16-dependent manner. We also show that MT regulators act in the nervous system to regulate lifespan, and that the lifespan extension in efa-6 and hdac-6 mutants are dependent on DAF-16. In addition, loss of DAF-16 abolishes the effect on fat storage in efa-6 mutants, suggesting that the role of MT regulators in regulating metabolism and aging requires DAF-16 function. Our RNA-seq data showed that genes involved in metabolic pathways were differentially expressed in the MT regulator mutants. Many of these genes (e.g., fatty acid desaturase genes and acyl-CoA dehydrogenase genes) are targets of DAF-16 (Murphy et al., 2003). Among the genes whose expression is regulated in opposite directions between the two long-lived and short-lived MT regulator mutants, nap-1 encodes a conserved nucleosome assembly protein (NAP) family member that can function as a cofactor of DAF-16 to regulate transcription of longevity genes (Cheong et al., 2013;Ihara et al., 2017). acdh-1 was previously reported as a target of DAF-16 (Murphy et al., 2003). Also, nap-1 RNAi reduces acdh-1p:: GFP expression (MacNeil, Watson, Arda, Zhu, & Walhout, 2013), consistent with the notion that NAP-1 might act as a cofactor of DAF-16 to regulate transcription (Cheong et al., 2013). A previous study reported that FoxO regulates neuronal MT stability and its protein level was negatively regulated by MT disruption in Drosophila neurons (Nechipurenko & Broihier, 2012). In agreement with that, we detect increased DAF-16 nuclear localization in efa-6 mutants.
Together, these results suggest that neuronal MT status might affect DAF-16 activity and subsequently affect fat metabolism and longevity. However, daf-18 RNAi, which increases insulin signaling (Gil et al., 1999;Mihaylova et al., 1999;Ogg & Ruvkun, 1998), displays an additive effect with efa-6(tm3124) on longevity, suggesting that MT regulators might function in parallel with IIS pathway. Future study will be required to determine the precise mechanism by which DAF-16 mediates the role of MT regulators in the context of aging.

| Touch sensitivity assay
Touch assay was performed as previously described (Chalfie & Sulston, 1981). Briefly, each animal was touched in the head and then tail with an eye bow hair for 5 times (10 touches in total). A positive response was determined as acceleration of the animal away from touch. The percentage of positive response was quantified.
Thrashing assays were performed in 12-well plates with 2 ml of M9

| Lifespan assays
Animals were cultured at 20°C and survival was scored at room tem-
When animals reached young adulthood, young adult animals were picked to fresh NGM plates with OP50 and allowed to lay eggs for between each mutant and wild-type groups, with more than 1.5-fold expression change as the cutoff (Love, Huber, & Anders, 2014). The RNA-seq data are available at the Gene Expression Omnibus under the accession number GSE115531 or the token "avqxsgogjxifvwv".

| RT-PCR
Synchronized adult Day 1 worms (four replicates for each strain) were collected for total RNA extraction using Trizol (Invitrogen).
First-strand cDNA was prepared from total RNA using the Super-

| Oil Red O staining
Staining was performed as previously described (Li et al., 2016

| Nile Red staining
Staining was performed as previously described (Li et al., 2016). Similar to Oil Red O staining, well-fed animals were fixed with 1% paraformaldehyde and treated with 3 freeze/thaw cycles and dehydrated in 60% isopropanol. Animals were then stained with 1 μg/ml Nile Red in 60% isopropanol for 30 min. Animals were washed and mounted to 2% agarose padded slides for imaging using Olympus IX83 microscope. Intensity of staining was quantified using Image J.

ACKNOWLEDGMENTS
We thank Dr. Hongning Wang for assistance in the thrashing assay, (Cancer Center at UT Health San Antonio), NIH Shared Instrument grant 1S10OD021805-01 (S10 grant), and CPRIT Core Facility Award (RP160732).

CONFLI CT OF INTEREST
None Declared.

AUTHOR CONTRI BUTIONS
Conceived and designed the experiments: AX, AF, ZL, and LC. Performed the experiments: AX, SK, and LC. Analyzed the data: AX, ZZ, and LC. Wrote the paper: AX, ZZ, ZL, and LC.