Histone deacetylase SIRT6 regulates tryptophan catabolism and prevents metabolite imbalance associated with neurodegeneration

Abstract

In the brain, tryptophan byproducts are involved in the biosynthesis of proteins, energy-rich molecules (e.g., NAD⁺), and neurotransmitters (serotonin and melatonin). Impaired tryptophan catabolism, seen in aging, neurodegeneration and psychiatric diseases, affects mood, learning, and sleep; however, the reasons for those impairments in the elderly and in those suffering from these ailments remain unknown. Our results from cellular, Drosophila melanogaster, and mouse models indicate that Sirtuin 6 (SIRT6) regulates tryptophan catabolism by balancing its usage. Mechanistically, SIRT6 regulates tryptophan and sleep quality through changes in gene expression of key genes (e.g., TDO2, AANAT), which results in elevated concentration of neurotoxic metabolites from the kynurenic pathway at the expense of serotonin and melatonin production. Such neurotoxic metabolites can affect various processes in the brain. However, by redirecting tryptophan through TDO2 inhibition in a SIRT6 knockout D. melanogaster model, the impairments in neuromotor behavior and vacuolar formation - parameters of neurodegeneration - can be significantly reversed.

Tryptophan metabolism is disrupted in aging and neurological disorders. Here, the authors show that histone deacetylase sirtuin 6 regulates tryptophan usage, and its absence results in neurotoxic products and impaired sleep that can be reversed by inhibiting the tryptophan processing enzyme TDO2.

Introduction

A growing body of evidence implicates altered tryptophan (Tryp) metabolism in the aging brain and neurodegeneration^1–4. Byproducts of Tryp catabolism have been associated with Huntington’s (HD), Parkinson’s (PD), and Alzheimer’s disease (AD)^4–6. Moreover, changes in the expression of Tryp metabolism genes have been observed in a broad range of psychiatric disorders^3,5,7. Serotonin (Sero) and melatonin (Mel) production, derived from Tryp, are reduced with age and neurodegeneration, affecting mood and sleep cycles^8–10. However, what originates these changes in the aging brain and further influences the brain to a pathological state is not completely understood.

Tryp is an essential amino acid that has a potential role in aging as a regulator of protein homeostasis⁵. In the cell, it can be directed to four main pathways: protein synthesis, tryptamine production, Sero and Mel production, and the kynurenine (Kyn) pathway, which is used for energy production in the de novo synthesis of NAD⁺. The Kyn pathway generates a variety of metabolites that are both neuroactive and neurotoxic, with significant roles in modulating immune function and neural health^3–5. For instance, quinolinic acid (QA) is an N-methyl-D-aspartate (NMDA) receptor agonist^11,12 that can increase inflammation, lipid peroxidation, and oxidative stress and potentially lead to neuronal lost¹¹. 3-Hydroxykynurenine, also known to be neurotoxic via the generation of free radicals and oxidative stress^13,14. Conversely, kynurenic acid (KA) acts as an antagonist of NMDA and alpha-7 nicotinic receptors^7,15, excessive blockage of glutamate signaling has been linked to psychotic symptoms and cognitive impairment. Indeed, elevated KA levels have been observed in the cerebrospinal fluid (CSF) of patients with mood disorders, including schizophrenia, bipolar disorder, and major depressive disorder^16–18. Similarly, xanthurenic acid (XA) has been associated with schizophrenia due to its influence on dopaminergic activity¹⁹, while anthranilic acid (AA) has been proposed as a potential biomarker for schizophrenia as well²⁰.

Most importantly, the neuroactive and neurotoxic properties of many Kyn pathway metabolites suggest their significant involvement in neurodegeneration. Notably, altered levels of metabolites such as Kyn, QA, KA, and AA in plasma and CSF have been proposed as potential biomarkers for AD^21–23, PD²⁴, HD⁴, and Amyotrophic Lateral Sclerosis (ALS)²⁵. Furthermore, growing body of evidence highlights the importance of Tryp catabolism in indole pathway, which is mediated by the gut microbiota, influencing mood disorders and neurodegeneration through the gut-brain axis^26–28. Therefore, maintaining a balanced Tryp metabolism is crucial for preventing neurodegenerative and psychiatric disorders.

Sirtuin 6 (SIRT6) is an NAD⁺-dependent histone deacetylase that epigenetically regulates many pathways through its histone deacetylase activity^29–36 or ADP-ribosylase activity^29–31. It is particularly important in the brain due to its DNA repair and epigenetic regulator roles^{29,31,37–39}. SIRT6 expression declines dramatically in the brains of the elderly⁴⁰ and patients with AD⁴¹. Our previous work showed that mice with a brain-specific deletion of SIRT6 (brS6KO) have an AD-like phenotype, with behavioral and learning impairments, increased cell death, increased acetylation of histone H3 (H3K9ac), and activation of glycogen synthase kinase 3 (GSK3)⁴¹. Importantly, lack of SIRT6 leads to accumulation of pathological Tau⁴², mitochondrial dysfunction, and energy depletion, all pathological hallmarks of several neurodegenerative diseases such as AD, HD, PD, and ALS⁴³. Moreover, we observed an overlap in gene expression between the brains of patients with AD and PD and the brains of SIRT6 knockout animals⁴⁴. SIRT6 does not bind specific DNA sequences, but rather depends on the association and regulation of other transcription factors, allowing it flexibility. For example, Hif1α for glucose metabolism⁴⁵ and Myc for cell growth⁴⁶. In here, we observe that SIRT6 regulates tryptophan metabolism, circadian rhythm and sleep quality through transcriptional regulation of these pathways.

We discovered an evolutionary conserved role of SIRT6 as a key regulator of Tryp catabolism. SIRT6 deficiency leads to a switch in gene expression, increasing the Kyn pathway and its metabolites and decreasing Sero and Mel production. This switch leads to the accumulation of toxic byproducts, leading to circadian rhythm and sleep disruption, accelerating neurodegeneration. Mechanistically, we showed that SIRT6 localizes to the promoters and regulates many of the disrupted genes directly. We confirmed that SIRT6 regulation of Tryp catabolism is conserved from D. melanogaster to mouse and human cellular models (via gene expression and metabolomics). Moreover, metabolomic changes in SIRT6-deficient cells significantly overlap with the metabolome of human patients’ CSF, particularly in AD, Schizophrenia, and several brain inflammatory diseases. Importantly, we showed in our D. melanogaster model that behavioral defects, brain vacuolization, and gene expression related to neurodegeneration can be rescued by redirecting tryptophan from the Kyn pathway through TDO2 inhibition.

Results

SIRT6 depletion results in a shift in tryptophan metabolism

We took advantage of murine embryonic stem (ES) cells metabolomics data⁴³ of wild-type (WT) vs. SIRT6 KO mice and analyzed it to test if SIRT6-dependent changes associate with metabolite sets reported in CSF from various diseases as a proxy for brain health. The analysis revealed enrichment associated with degenerative, psychiatric, and brain infection diseases (Fig. 1A). Interestingly, among the associated metabolites for AD, dementia, schizophrenia and brain inflammatory diseases are tryptophan and its derivatives- KA and QA. This overlap was interesting, since Tryp and its derivative metabolites are known biomarkers in the CSF for degenerative and psychiatric diseases^{7,16,22,47–49}, and SIRT6 deficiency could recapitulate in a significant manner these metabolic signatures. Focusing on Tryp and its catabolites, we found significant changes in various metabolites, including an increase in Tryp and its derivatives levels such as KA and QA in SIRT6 KO cells. (Fig. 1B–D). These results were validated in two additional human cell lines, neuroblastoma-derived SH-SY5Y and adenocarcinoma-derived HeLa SIRT6 WT and KO cells (Fig. 1D). In all experimental models examined, the absence of SIRT6 resulted in an augmentation of Tryp levels (Fig. 1B, D and Supplementary Fig. 1A). Since Tryp is an essential amino acid, we tested the levels of its transporter Slc75a (the most abundant tryptophan transporter in the brain^50,51) and found an increased expression in the brains of brS6KO mice (Fig. 1E).

Fig. 1. Tryptophan metabolism is affected in SIRT6-deficient models.

A Enrichment plot of diseases associated with metabolite sets reported in cerebrospinal fluid (CSF). The plot summarizes the results of over representation analysis (ORA) conducted on metabolites showing significant abundance changes between SIRT6 KO and WT samples from mouse ES cells. ORA was implemented using the hypergeometric test to evaluate whether a particular metabolite set is represented more than expected by chance within the given compound list. One-tailed P values are provided after adjusting for multiple testing. Created in MetaboAnalyst. B Abundance changes of several tryptophan derivative metabolites between SIRT6 KO and WT samples from mouse ES cells. Purple circles correspond to metabolites decreased in SIRT6 KO and orange circles to metabolites with elevated abundance levels in SIRT6 KO cells. Selected metabolites associated with the tryptophan pathway are marked by circles of larger size (Smirnov et al.; Y axis represents -log₁₀-transformed FDR-adjusted P value calculated using two-sided t-test). C Selected metabolites extracted from ES WT and SIRT6 KO cells. (KA kynurenic acid, Kyn Kynurenine, QA Quinolinic acid, n = 3 replicates; NAD⁺ FDR P value = 1.52 × 10⁻⁵; Acetyl-Coa FDR P value = 1.89 × 10⁻¹; Kynurenic acid FDR P value = 3.86 × 10⁻⁵; Kynurenine FDR P value = 8.57 × 10⁻¹; Quinolinic acid FDR P value = 4.72 × 10⁻²; all P values from two-sided t-test). D Tryptophan levels in SIRT6 KO’s cell lines (ES n = 3 replicates, FDR P value = 0.000842, HeLa n = 5 replicates, P value = 8.83e-06; SH-SY5Y n = 4 replicates, P value = 0.0276; all p values from two-sided t-test; metabolomic data set; Tryp Tryptophan). E RNA-Seq analysis of WT and brS6KO brains indicating significant changes in tryptophan transporter (Slc7a5). (n = 4 mice; FDR P value = 6.46 × 10⁻²²; two-sided Wald test). F Metabolomics analysis of tryptophan derivatives metabolites/precursor ratio change significantly in the SIRT6 KO cells (numbers above the bars represent the P values, AA anthranilic acid, Ser serotonin, KA kynurenic acid, Kyn Kynurenine, n = 4 replicates; two-sided unpaired t-test; In Hela, AA/Kyn P value = 0.0248, Ser/Tryp P value = 0.0071; SH-SY5Y, KA/Kyn P value = 0.0251, AA/Kyn P value = 0.0007, Ser/Tryp P value = 0.0087). G Schematic representation of tryptophan catabolism pathways in SIRT6 KO models. Red boxes represent decreased levels of the metabolite in the SIRT6 KO cells, dark blue represents increased levels in the SIRT6 KO cells. Purple box represents high levels of the metabolite in ES SIRT6 KO cells and low in SH-SY5Y SIRT6 KO. The dashed arrows represent processes involving multiple steps that are not detailed in the schema. Created in BioRender. Toiber, D. (2025) https://BioRender.com/1bn2qc1. Data represent mean ± SEM (*P < 0.05, **P < 0.01, ***P < 0.001). Source data are provided as a Source Data file.

Furthermore, directed metabolomic analysis of the SIRT6 KO in SH-SY5Y and HeLa cell lines showed increase in AA/Kyn with a reduction in serotonin (Sero)/Tryp and KA/Kyn (Fig. 1F and Supplementary Fig. 1B), suggesting impaired metabolic processing and a shift favoring the Kyn pathway (Fig. 1G). However, the levels of NAD⁺, measured in ES, were also reduced, even though its precursor QA was upregulated (similar to the phenomena observed in aging^52,53, Fig. 1C).

We performed untargeted metabolomic analysis of SIRT6 deficiency in SH-SY5Y (representing a similar to neuron line), HeLa (representing cancer line), and ARPE19 (normal retina-pigmented epithelium⁵⁴) cell lines, using principal component analysis. The metabolome differed between cells that did and did not express SIRT6, but the scope of differences depended on the cell type (Supplementary Fig. 1C–G). Importantly, in HeLa cells, Tryp metabolic pathway was one of top 10 categories contributing to differences between control and SIRT6 KO cells (Supplementary Fig. 1D).

Overall, SIRT6 affected Tryp levels and its catabolic byproducts in mouse and human cell lines, suggesting a conserved basic mechanism across animal and cell types that shifts the metabolites toward the Kyn pathway (Fig. 1G), although Tryp metabolism is changed in all the models used, each cell line has a different response based on its own transcriptome and metabolic needs.

SIRT6 regulates transcription of genes involved in tryptophan catabolism

To investigate whether these changes are associated with gene regulation, we conducted a transcriptomic analysis of brains from full-body SIRT6 knockout (SIRT6 KO) and brS6KO mice, using microarray and RNA-seq methods, respectively^43,44. This analysis revealed significant differences between SIRT6 KO and WT mice in the expression of genes involved in Tryp metabolic pathways (Fig. 2A, B). The hypergeometric test indicates correlation in the results of the two separated models. This emphasizes the similarity between the datasets and shows how fundamental SIRT6's role in Tryp metabolism is—even in different mouse strains (Fig. 2B). In addition, we performed a quantitative PCR with additional animals and using probes for genes that were not detected in the transcriptomic experiments. Importantly tryptophan 2,3-dioxygenase 2 (Tdo2; predominantly expressed in the brain⁵⁵) and indoleamine 2,3-dioxygenase (Ido1 and Ido2; predominantly expressed in immune cells^56,57), both encoding rate-limiting enzymes involved in the conversion of Tryp to Kyn^58,59 were upregulated (Fig. 2C). Additionally, the levels of kynureninase (Kynu) mRNA were significantly elevated in the brains of brS6KO mice versus WT controls, whereas the expression of kynurenine 3-monooxygenase (Kmo), which is crucial for energy production, was diminished (Fig. 2C). No significant changes were observed in the expression levels of Kynurenine aminotransferase (Kyat) family including Aadat, which predominantly express in the brain^60,61, Got, Ccbl1, and Ccbl2 (Fig. 2C and Supplementary Fig. 2A) as well as in 3-Hydroxyanthranilate 3,4-Dioxygenase (Haao) and Aromatic-L-Amino-Acid Decarboxylase (Ddc) (Fig. 2C). This results suggest this genes are not regulated by SIRT6. In this context, the accumulation of specific metabolites, such as KA, likely depends on other enzymes, such as the rate-limiting enzymes TDO2, IDO, and KMO. In the Sero-Mel pathway, tryptophan hydroxylase 2 (Tph2) expression, which catalyzes the initial step in Mel production and is predominantly expressed in the central nervous system (CNS)^62,63, did not change significantly in the brains of brS6KO mice versus controls. In contrast, the expression of rate-limiting enzymes aralkylamine N-acetyltransferase (Aant) and N-acetylserotonin O-methyltransferase (Asmt) was downregulated (Fig. 2C-D).

Fig. 2. Tryptophan metabolism transcriptome is impaired in the SIRT6-deficient mouse model.

A Heatmaps indicate significant changes in mice full-body SIRT6 KO’s or RNA-seq results of brains from Nestin-Cre mice (brS6KO). (right dataset nWT = 2, nHet = 1, nSIRT6 KO = 3 mice; left dataset nWT = 4, n brS6KO = 4 mice) B Scatter plot visualizing overlap in tryptophan metabolism changes between mouse models. Hypergeometric test P value = 3.14 × 10⁻²² corresponds to the significance of gene overlap in the upper right quadrant. C qPCR of brains RNA from Nestin-Cre mice. (two-sided unpaired t-test; For Ido1, Kmo, Haao, Qprt, Aanat WT/brS6KO n = 11/10; Tdo2, Ido1 WT/brS6KO n = 12/10; Aadat WT/brS6KO n = 9/9; Kynu WT/brS6KO n = 10/10; Ddc WT/brS6KO n = 10/9; Tph2 WT/brS6KO n = 8/7; n represents number of mice; arbitrary units, arb. units). D Schematic representation of tryptophan metabolism pathways in SIRT6 KO models. Red boxes represent decreased levels of the metabolite in the SIRT6 KO cells, dark blue represents increased levels in the SIRT6 KO cells. Purple box represents increased levels of the metabolite in ES SIRT6 KO cells and decreased levels in SH-SY5Y SIRT6 KO. Red arrows represent down-regulation of the enzyme in the brS6KO brains, blue arrows represent up-regulation of the enzyme in brS6KO brains. The dashed arrows represent processes involving multiple steps that are not detailed in the schema. Created in BioRender. Toiber, D. (2025) https://BioRender.com/h3kety3. Data represent mean ± SEM (*P < 0.05, **P < 0.01, ***P < 0.001). Source data are provided as a Source Data file.

Using ChIP-seq and ChIP-qPCR, we confirmed that SIRT6 binds directly to the promoter regions of TDO2, IDO1, and AANAT (Supplementary Fig. 2B, C). These results suggest that SIRT6 balances the expression levels of the rate-limiting enzymes of the two pathways, and in the absence of SIRT6, tryptophan would be directed to the Kynurenine pathway, at the expense of the serotonin/melatonin pathway.

Serotonin and melatonin production and their corresponding pathways are affected in brS6KO mice

Changes in Tryp-derived metabolites, such as Sero levels, could impair mood behavior, while changes in Mel could impair sleep, both parameters affected in aging and neurodegeneration. We detected decreased serotonin levels in the serum of SIRT6 KO mice (Fig. 3A), which is consistent with the effect of SIRT6 knockout in HeLa and SH-SY5Y cells (Fig. 1F).

Fig. 3. Changes in the production of brain-related metabolites.

A ELISA for Serotonin in serum from WT and SIRT6 KO mice. (WT n = 5, SIRT6 KO n = 4 mice; two-sided unpaired t-test). B Melatonin levels in WT vs. brS6KO mice serum during a 24 h period. (n = 2 mice from each genotype at each time point).C qPCR results of RNA isolated from mouse brains, indicating the oscillations in gene expression (two-sided unpaired t-test; For Asmat, WT vs. brS6KO for ZT18 P value = 0.012; WT n = 5 mice at each time point, except for time point ZT 18 n = 4, brS6KO n = 4 at each time point; arbitrary units, arb. units). Created in BioRender. Toiber, D. (2025) https://BioRender.com/jgmijt9. D Schematic representation of tryptophan metabolism pathways in SIRT6 KO models. Red box represents decreased levels of the metabolite in the SIRT6 KO cells /mice, blue box represents increased levels in the SIRT6 KO cells. Purple box represents increased levels of the metabolite in ES SIRT6 KO cells and decreased levels in SH-SY5Y SIRT6 KO. Red arrows represent down-regulation of the enzyme in the brS6KO brains, blue arrows represent up-regulation of the enzyme in brS6KO brains. The dashed arrows represent processes involving multiple steps that are not detailed in the scheme. Created in BioRender. Toiber, D. (2025) https://BioRender.com/mubi3nh. Data represent mean ± SEM (*P < 0.05, **P < 0.01, ***P < 0.001). Source data are provided as a Source Data file.

Mel hormone oscillates in the serum in accordance with dark- light cycles due to transcriptional activation of AANAT. Once expressed, this enzyme becomes the rate-limiting factor in Mel production and secretion⁶⁴. The low levels of Aanat and Asmt expression in the brain of brS6KO mice (Fig. 2C) suggest an abnormal melatonin secretion cycle. Indeed, WT mice showed a normal oscillation of melatonin levels, including an increase during the dark phase, whereas the amplitude of melatonin oscillation in brS6KO mice was significantly smaller and the levels did not increase during the dark phase (Fig. 3B). Importantly, we measured the oscillation in gene expression in brS6KO mice brains over a 24-h period and the Aanat gene expression patterns correspond with the oscillations in melatonin levels in both WT and brS6KO mice. The Aanat and melatonin patterns of brS6KO mice are in opposite phases to those of WT. The expression of Aanat in brS6KO mice exceeded the expression in controls around the 12-h during light (Fig. 3C). In contrast, Asmt expression was deregulated over the entire 24-h period, while Tph2 did not change significantly in all time points tested (Fig. 3C and Supplementary Fig. 3A). These findings underscore the role of SIRT6 in modulating Sero and Mel secretion through the regulation of rate-limiting enzyme expression.

Aanat mRNA levels in the brain do correlate with melatonin levels, but it needs to be pointed out that C57BL/6 mice, the background strain for our brS6KO model, have a splicing defect in the Aanat⁶⁵, which results in a reduced melatonin production Therefore, we hypothesized that SIRT6 could influence sleep and circadian patterns not only through melatonin but also through the influence of Tryp catabolites⁶⁶.

Lack of SIRT6 impairs the circadian clock in the brain

Tryp catabolites can impair the circadian regulation through various pathways, integrating various environmental signals^66–68. Furthermore, it is known that SIRT6 possesses a role in circadian clock regulation in peripheral tissue^69,70 but it was never demonstrated in the brain, the pacemaker of the whole body circadian machinery. Since sleep quality decays with age and neurodegeneration, this could be affected by the SIRT6-Tryp axis. Therefore, we analyzed the transcriptomic profile of the circadian clock pathway in the brains of brS6KO mice, and they is significantly different from the corresponding profile of WT animals (Fig. 4A, B). We explored the expression of the negative regulators of the circadian machinery, Cry1, Per1, and Per2, in 6-h intervals (total of 24 h with 12:12 light-dark cycle). The mRNA levels of Per1 and Per2 (but not Cry1) were consistently higher in brS6KO brains; this discrepancy was not observed in the liver, where SIRT6 is present (Fig. 4C).

Fig. 4. Lack of SIRT6 impairs the circadian clock in the brain.

A Heatmap indicating significant changes in the circadian rhythm pathway in brS6KO mice (WT n = 4 mice, brS6KO n = 4). B qPCR results of RNA isolated from mouse brains normalized to Actin, ZT = 6 h. (two-sided unpaired t-test; For Per2, Cry1 WT/brS6KO n = 12/10; Per1 WT/brS6KO n = 11/10, n represents number of mice). C Circadian oscillation in RNA expression in mouse brains (two-sided unpaired t-test; For Per1, WT vs. brS6KO for ZT0 P value = 0.004, ZT6 P value = 0.033, ZT12 P value = 0.005, ZT18 P value = 0.009; For Per2, WT vs. brS6KO for ZT6 P value = 0.038.; WT n = 5 mice at each time point except for time point ZT18 n = 4, brS6KO n = 4 at each time point) and Liver (n = 2 mice per genotype at each time point). Created in BioRender. Toiber, D. (2025) https://BioRender.com/jgmijt9D Circadian oscillation in RNA expression in mouse brains (two-sided unpaired t-test; For Tdo2, WT vs. brS6KO for ZT0 P value < 0.0001, ZT6 P value < 0.0001 ZT12 P value < 0.0001, ZT 18 P value < 0.0001; For Ido2, WT vs. brS6KO for ZT0 P value < 0.0001, ZT6 P value < 0.0001, ZT12 P value = 0.004, ZT18 P value = 0.004; WT n = 5 mice at each time point except for time point ZT 18 n = 4, brS6KO n = 4 at each time point). E Western blots of chromatin proteins extracted from mouse brains (two-sided unpaired t-test; n = 4 mice per genotype at ZT = 6 h). F Percentage of activity during dark stress (4 days). (activity in light represents the chronological lighted, period but measure in darkenss; WT n = 4 mice, brS6KO n = 4; two-sided unpaired t-test). G Representative data of activity recorded for 80 days. Days 0–20—dark-light cycles (12 h:12 h), days 20–50 – constant darkness, days 50–80—recovery dark-light cycles (12 h:12 h). Each row represents one day. Black represents high activity. H Mean of slopes (red dashed line), indicating daily change in wakening time under darkness. (WT n = 8 mice, brS6KO n = 8; two-sided unpaired t-test). I Mean of the shift in waking time during constant dark stress. (WT n = 8 mice, brS6KO n = 8; two-sided unpaired t-test). Data represent mean ± SEM (*P < 0.05, **P < 0.01, ***P < 0.001) (arbitrary units, arb. units). Source data are provided as a Source Data file.

Since metabolic genes respond to the circadian clock machinery^71,72, we measured the oscillation in tryptophan-related gene expression in the brain over a 24-h period. We found that the levels of Ido2, Tdo2, and Ddc (Fig. 4D and Supplementary Fig. 4A), but not Ido1 (Supplementary Fig. 4A), were elevated at all time points, showing an impairment in the circadian regulation of gene expression.

Moreover, testing the relative abundance of BMAL1 and CLOCK, the core transcription activators of the circadian clock machinery, revealed no significant differences in the total brain extract of brS6KO mice and controls (Supplementary Fig. 4B). However, these two proteins were significantly less abundant in the chromatin extract of brS6KO mice (Fig. 4E), which suggests a role of SIRT6 in their chromatin recruitment in the brain as well.

A ChIP-qPCR/seq analysis of PER1, CRY1, and CLOCK promoters shows a significant enrichment of SIRT6 binding, compared to the negative control, suggesting that SIRT6 is directly involved in the regulation of their transcription, also in the brain (Supplementary Fig. 4B, C).

These findings highlight the role of SIRT6 in regulating circadian clock machinery in the brain, both through gene expression regulation and independently through changes in metabolite production. This underscores its importance in maintaining proper circadian rhythms, which are crucial for sleep quality—a property that is fundamentally impaired in aging and neurodegeneration.

Brain SIRT6 KO mice (brS6KO) have impaired circadian patterns

Sleep cycle disruption and fragmented sleep are common features of aging and neurodegeneration^73–77 often related to decreased Mel secretion and loss of the robustness of the circadian machinery in aging^9,10,78,79. Given the shifts in gene expression observed in the brains of brS6KO mice, we speculated that these changes might also lead to a behavioral shift.

We monitored the activity of WT and brS6KO mice during the dark-light cycles for 21 days in regular dark-light cycles (12:12). Compared with WT animals, brS6KO exhibited greater spontaneous activity in the resting (light) hours, as well as shorter sleep episodes (Fig. 4F and Supplementary Fig. 4E), suggesting a decline in sleep quality. To test their response to a disruption of the dark-light cycle, we then kept them in constant darkness for 3 weeks (dark stress), which was followed by a 4-week recovery period, spent in the normal dark-light cycle. After switching from the normal cycle to the period of dark stress, WT, and brS6KO mice began progressively shifting the periods of high activity (mice are nocturnal animals) into the hours previously spent resting. However, the effect was more pronounced in brS6KO mice, which underwent a full cycle shift, almost reaching by three weeks the time point 0 (Fig. 4G). This progressive shift in activity throughout the dark stress period was linear; its slope (i.e., duration of additional, high-activity period per day of total darkness, see Supplementary Fig. 4F for calculation explanation) was greater in brS6KO mice than in controls (Fig. 4H), and so was the average shift in waking hours (12 h vs. 7 h; Fig. 4I). Overall, lack of SIRT6 not only affects circadian behavior in mice but also impaired sleep quality, with spontaneous awakening, shorter sleep episodes during sleep hours, and less resistance to circadian stress.

A SIRT6 KO Drosophila Melanogaster model for neurodegeneration

Mouse models tend to have different metabolic behavior than humans, and in many cases, changes observed in mouse models are not conserved in humans; moreover, the mutation in our model regarding melatonin production could bias some of our findings; therefore, we decided to develop a D. melanogaster model for SIRT6-KO which allowed faster interventions and could suggest that if the pathways are conserved from D. melanogaster to mouse, the probability of these changes to be conserved to a human would be higher. Recently, two SIRT6-deficient D. melanogaster strains were developed, each showing reduced longevity and metabolic impairment and showing a viable model^80,81.

First, to validate our own SIRT6-KO D. melanogaster model, we analyzed the mutation sequence using BLAST. A stop codon was generated after 30 amino acids, suggesting a nonsense-mediated mRNA decay or the formation of a truncated protein (data not shown). Since there are no available antibodies for dSIRT6, we cloned this transcript into a mammalian vector with a Flag tag on the N-terminus and transfected SH-SY5Y SIRT6KO cells to test whether the mutated sequence could generate a different SIRT6 isoform. Only cells transfected with full-length SIRT6-Flag showed protein production, while the expression of our mutated D. melanogaster sequence could not be detected. In addition, transfection of SIRT6KO cells with the full-length SIRT6-Flag reduced the acetylation of histone H3 on lysine residues 56 and 9 (known targets of SIRT6 activity), while transfection with the D. melanogaster mutant had no effect (Supplementary Fig. 5A). Importantly, these histone acetylation patterns were also observed in fly protein extracts (Supplementary Fig. 5B). Thus, our novel SIRT6-KO Drosophila melanogaster model follows the expected epigenetic patterns of SIRT6 deficiency.

Behavioral and metabolic impairments in SIRT6 KO D. melanogaster mimic those in brS6KO mice

To test whether this model presents a degenerative phenotype, we performed a behavioral test commonly used to assess aging and neurodegeneration; negative geotaxis ability (climbing), to determine neuronal/motor impairment. SIRT6 KO flies showed an impairment in climbing ability versus control flies as early as seven days post eclosion, and this discrepancy increased with age (Fig. 5A and Supplementary Movie 1). This suggests an accelerated onset of aging, which has also been observed in SIRT6 KO mice⁸². In addition, in the brains of SIRT6 KO D. melanogaster, there was an increase in the relative amounts of H3K9ac (a SIRT6 target, Supplementary Fig. 5B) and the phosphorylated H2AX and 53BP1 (indicators of DNA damage), versus control animals (Fig. 5B). This is similar to brS6KO mice, which also exhibit reduced deacetylation of target proteins and increased DNA damage signaling⁴¹.

Fig. 5. SIRT6 KO D. melanogaster model presents a neurodegenerative phenotype.

A Negative geotaxis ability (climbing test assay) of WT vs. SIRT6 KO at the ages of 7,14, and 21 days. The data present the percentages of flies successfully climbing an 8 cm vertical distance within 10 s in response to a negative geotaxis stimulus. The graphs depict the average results obtained from 10 repeated trials within the same experimental group. Values in the table represent p values. (For WT male 7, 14, and 21 days n = 39,83 and 69; WT female 7, 14, and 21 days n = 44,69 and 52; SIRT6 KO male 7, 14, and 21 days n = 36,88 and 53; SIRT6 KO female 7, 14, and 21 days n = 44, 81 and 59; n represents number of flies; Two-sided unpaired t-test). B Immunofluorescent (IF) of Drosophila brains WT vs. SIRT6KO. γH2AX and 53BP1-p. γH2AX n = 4, 53BP1-p (n = 3 fly brains, few measurements from each brain; two-sided unpaired t-test; Scale bar represents 100 μm). C Representative data of inner planes from WT vs. SIRT6 KO D. melanogaster brains (14 days old) showing vacuolization differences that suggest neurodegeneration. White arrows mark vacuoles. Holes found on the brain surface (superficial planes) were not counted as vacuoles to avoid false positives derived from the dissection process or from any other damage (Scale bar represents 100 μm). D Average vacuole area in brains of 14 days old Drosophila WT vs. SIRT6 KO with 14 days of DMSO/TDO2 inhibitor treatment. (WT DMSO n = 4; WT TDO2 inhibitor n = 6; SIRT6 KO DMSO n = 5; SIRT6 KO TDO2 inhibitor n = 5; n represents number of flies brains; Dunnett’s multiple comparisons test, two-tailed). E qPCR results of RNA isolated from D. melanogaster heads, normalized to Actin. (each replicate contains 10 heads; n WT = 5 replicates; n SIRT6 KO = 7 replicates; two-sided unpaired t-test). F D. melanogaster metabolomics results indicating significant differences in tryptophan and its derivative metabolites, whole body (Each replicate contains 10 flies; WT/ SIRT6 KO n = 12 replicates; F.C fold change, Tryp tryptophan, Kyn kynurenine, KA kynurenic acid, XA xanthurenic acid, Mel melatonin, two-sided unpaired t-test). G Schematic representation of tryptophan metabolism pathways. Red boxes represent decreased levels in the SIRT6 KO Drosophila, dark blue boxes represent increased levels in the SIRT6 KO Drosophilas. Red arrow represents downregulation of the enzyme in the SIRT6 KO brains, blue arrow represents upregulation of the enzyme in SIRT6 KO brains. Created in BioRender. Toiber, D. (2025) https://BioRender.com/cytkhbk. Data represent mean ± SEM (*P < 0.05, **P < 0.01, ***P < 0.001) (arbitrary units, arb. units). Source data are provided as a Source Data file.

Moreover, using the Behnke et al. protocol⁸³, we observed a greater number of vacuoles and a greater average vacuole size in the brains of SIRT6KO D. melanogaster versus control flies, which indicates a higher degree of neurodegeneration (Fig. 5C, D and Supplementary Fig. 5C, D).

Therefore, we developed and validated a new D. melanogaster model for SIRT6 deficiency that shows epigenetic alterations, DNA damage, accelerated neurodegeneration, and behavioral impairment.

Finally, our metabolomic findings indicate that tryptophan metabolism in SIRT6 mutant flies is shifted in ways similar to those observed in brS6KO mice. In D. melanogaster, the Tryp-metabolizing enzymes TDO2 and KMO are encoded by vermillion and cinnabar genes, respectively. The expression of vermillion was significantly higher in the heads of SIRT6 KO D. melanogaster than in control flies, whereas the effect was the opposite for cinnabar (Fig. 5E).

SIRT6 KO flies also have higher levels of Tryp, KA, and Kyn versus controls and lower levels of XA and Mel (Fig. 5F and Supplementary Fig. 5E, F), which is consistent with the changes in expression of rate-limiting enzymes TDO2 and KMO, and similar to the metabolomics findings obtained in SH-SY5Y, HeLa, and ES KO cells (Fig. 1C, D, F and Supplementary Fig. 1A).

SIRT6 KO-related changes in metabolite levels were observed in both sexes, except in the case of dopamine (Dop), for which significantly lower levels in SIRT KO vs control flies were observed in males only (Supplementary Fig. 5F).

Finally, we observed a significant decrease in cry expression and no significant change in per in SIRT6 KO D. melanogaster brains versus control flies (Supplementary Fig. 5G).

Overall, our results suggest a conserved role for SIRT6 in regulating epigenetic changes, DNA repair, tryptophan catabolism, the circadian clock and preventing neurodegeneration.

TDO2 inhibition improves climbing ability in SIRT6KO D. melanogaster flies

The kynurenine pathway has been identified previously as a potential target for the treatment of aging and neurodegeneration^{1–4,6,84,85}. In C. elegans, TDO depletion was shown to extend lifespan and delay age-related decline in protein homeostasis². In D. melanogaster models of AD and PD, pharmacological inhibition of TDO improved disease-specific pathology, as well as the climbing ability⁸⁵.

We speculated that pharmacological inhibition of TDO2 will prevent byproducts accumulation and would improve climbing ability in our SIRT6 KO D. melanogaster. Therefore, we treated our SIRT6 KO D. melanogaster flies with 100 µM 680C91, an inhibitor of TDO2 and vehicle only (DMSO), for 21 days and did the same with WT flies (Supplementary Fig. 6A). Feeding protocol was adapted from Breda et al.⁸⁵, who first highlighted the significant potential of TDO2 inhibition in preventing the accumulation of neurotoxic byproducts from the kynurenine pathway in various neurodegenerative D. melanogaster models, thereby rescuing the reduction in climbing ability. At 14 and 21 days, the climbing ability of SIRT6 KO flies, with or without TDO2 inhibition, remained significantly inferior to the performance of their WT counterparts (Fig. 6A, Source Data). However, we also observed a significant improvement in TDO inhibitor-treated SIRT6 KO flies versus those treated with vehicle only: in males (but not females) at day 14 and in females (but not males) at day 21 (Fig. 6, Source Data).

Fig. 6. TDO2 inhibitor treatment rescues the neurodegenerative phenotype.

A, B The graph illustrates the performance of D. melanogaster in negative geotaxis assays (climbing test assay). The data presents the percentages of flies successfully climbing an 8 cm vertical distance within 10 s in response to a negative geotaxis stimulus. The graphs depict the average results of 10 repeated trials within the same experimental group. A Negative geotaxis ability (climbing test assay) of flies under 100 µM TDO2 inhibitor / DMSO treatment. (Tukey’s multiple comparisons test; n represents number of flies in each cohort; 14 days: WT male DMSO/TDO2i n = 83/88; WT female DMSO/TDO2i n = 69/77; SIRT6 KO male DMSO/TDO2i n = 88/68; SIRT6 KO female DMSO/TDO2i n = 81/61. For 21 days: WT male DMSO/TDO2i n = 69/57; WT female DMSO/TDO2i n = 52/67; SIRT6 KO male DMSO/TDO2i n = 53/38; SIRT6 KO female DMSO/ TDO2i n = 59/51). B Negative geotaxis ability (climbing test assay) of flies under 0.5 mM Melatonin/Ethanol treatment. (Tukey’s multiple comparisons test; n represents number of flies in each cohort; 14 days: WT male Ethanol/Melatonin n = 39/32; WT female Ethanol/Melatonin n = 26/32; SIRT6 KO male Ethanol/Melatonin n = 46/30; SIRT6 KO female Ethanol/Melatonin n = 18/50. 21 days: WT male Ethanol/Melatonin n = 15/33; WT female Ethanol/Melatonin n = 22/22; SIRT6 KO male Ethanol/Melatonin n = 30/34; SIRT6 KO female Ethanol/Melatonin n = 22/20.) C Drosophila single-head RNA-seq enrichment analysis (RummaGEO) of human orthologs differentially expressed between SIRT6 KO and WT under DMSO (control) or TDO2i feeding (n = 3 fly heads per genotype per treatment; Enrichment significance was assessed with one-tailed Fisher’s exact test). Data represent mean ± SEM (*P < 0.05, **P < 0.01, ***P < 0.001). Source data are provided as a Source Data file.

In addition, the average vacuolar size in the brains of SIRT6KO flies was reduced by more than 50% with TDO2 inhibitor treatment (Fig. 5D). However, the average number of vacuoles per brain more than doubled (Supplementary Fig. 5C). This suggests that vacuoles are forming in SIRT6-deficient brains, and that TDO2 inhibition prevents their development into larger structures by enlargement and fusion with each other^86,87.

Mel has emerged as a critical regulator of neuronal health, with its antioxidant and neuroprotective properties playing a pivotal role in maintaining cellular homeostasis^88–90. Studies using D. melanogaster models of diseases such as AD, PD, and HD have demonstrated that melatonin supplementation can alleviate neurotoxic effects, improve locomotor function, and reduce the accumulation of harmful protein aggregates^91–93. These findings and our findings regarding Melatonin reduction in our SIRT6 KO D. melanogaster highlight the potential of Mel treatment in rescuing the phenotype. We used the feeding protocol published at Ortega-Arellano et al.⁹² and fed the files with 0.5 mM Mel for 21 days. However, treatment with Mel did not improve climbing activity in male SIRT6 KO flies and even worsened it in females (Fig. 6B, Source Data), suggesting it is not only the lack of melatonin but rather the metabolic imbalance that causes neurodegenerative phenotype.

TDO2 inhibitor restores neurodegeneration-related transcriptomics changes in SIRT6KO D. melanogaster

To better understand the effect of the TDO2 inhibitor and whether it is preventing neurodegeneration, we developed a single-head RNA-seq protocol on our D. melanogaster model. First, we wanted to confirm whether the RNA-seq shows enrichment for pathways associated with neurodegeneration and aimed to identify reversal by the TDO2 inhibitor (Supplementary Fig. 6B).

Pearson’s correlation heatmap and PCA analysis illustrate the clustering of the samples, showing clear separation by genotype (Supplementary Fig. 6C, D, PCA—blue vs. red), with PC1 explaining 39% of the variance. In contrast, treatment (circles vs. triangles) is captured in PC2, accounting for only 23% of the variance, though following the same trend in WT and SIRT6 KO heads (Supplementary Fig. 6D).

Next, we analyzed the differences between SIRT6 KO and WT flies under control treatment (DMSO). We identified 1,662 differentially expressed genes (DEGs) (Supplementary Fig. 6E, F). A general enrichment analysis of the DEGs revealed categories such as metabolic processes, protein folding, chaperone activity, oxidative stress, and response to stress, similar to other SIRT6-deficient models (see Supplementary Table 5).

Among the downregulated categories, we observed decreased regulation of amino acid activation and tRNA aminoacylation for protein translation pathways, including trpRS (Tryptophanyl-tRNA synthetase), which encodes the enzyme responsible for attaching tryptophan to its cognate tRNA⁹⁴. (Supplementary Fig. 6G).

Upregulated genes were associated with mitochondrial function (foz, drp1)^94–97, synaptic development and function (sec5, faf)^98,99, tubulin assembly (αtub84B), protein folding stress (hsp83)¹⁰⁰, and cell cycle regulation (polo)¹⁰¹ (Supplementary Fig. 6G).

To investigate the excess tryptophan in D. melanogaster, we analyzed amino acid transporters. We found that differently than mouse, there was a significant upregulation of mah, gb, and tadr transporters, while CG33296, CG5585, CG13794, JhL-21, and CG13793 were downregulated (Supplementary Fig. 7A), suggesting a complex regulatory landscape for amino acid transport in D. melanogaster heads (Supplementary Fig. 7A, B).

In the SIRT6 KO vs. WT comparison under TDO2 inhibitor treatment, we identified 1460 DEGs (Supplementary Fig. 6E, F). Enrichment analysis revealed multiple categories associated with the D. melanogaster visual system, as well as pathways linked to amino acid metabolism, including amino acid transmembrane transporters and the aminoacyl-tRNA synthetase multienzyme complex (see Supplementary Table 6). Categories related to protein folding, oxidative stress and cell stress were absent from the GO pathways, suggesting a reversal of stress responses in the SIRT6 KO flies.

Next, we analyzed the changes after TDO2 inhibition, we found downregulated ribosome biogenesis and translation-related pathways such as: peptide biosynthetic process, amino acid activation, amino acid import across plasma membrane (Supplementary Fig. 6H). These results suggest the inhibitor effectively modulates its target pathway, resulting in an alteration of tryptophan availability.

Interestingly, the upregulated DEGs showed a significant enrichment in categories such as response to light stimulus, anatomical structure homeostasis. Among the upregulated genes involved in these processes are circadian clock genes such as Per, nervous system and the eye development (arr1, ninaE, syn, RdgB)^102–106 (Supplementary Fig. 6H). As 60% of D. melanogaster brain devoted to sight^107,108 the appearance of sight-related categories and improvement in response to light stimulus under treatment may suggest an improvement in neuronal health.

Overall, we observed the reversal effect of the treatment in the SIR6KO flies of pathways and genes related to cell stress, oxidative stress, mitochondrial function, protein folding, circadian rhythms, and sight system development. Since the retina and the sight system are part of the CNS, they may reflect the brain condition. Pathways related to translational regulation, such as aminoacyl-tRNA biosynthesis and ribosome biogenesis, were significantly impacted by the TDO2 inhibitor. However, a deeper analysis revealed distinct differences in translational fidelity and protein quality control between WT and KO flies. For instance, while the TDO2 inhibitor reversed several pathways linked to protein misfolding and aggregation, it also induced unique pathways related to ribosomal activity and amino acid metabolism. These findings suggest a nuanced role for the TDO2 inhibitor in modulating translation under conditions of SIRT6 deficiency, therefore, whether TDO2 should be used to prevent neurodegeneration requires further investigation to better understand the gained pathways.

To validate the RNA-seq results, we extracted proteins from the heads of flies treated with either DMSO or a TDO2 inhibitor. We then performed Western blot analysis to assess the differences between SIRT6 knockout (KO) and wild-type (WT) flies. Based on our RNA-seq data, the enriched pathways in the DEGs between WT and SIRT6 KO flies treated with DMSO included unfolded protein binding, chaperone complex, and protein folding. As an increase in ubiquitinated proteins is a known marker of aging and neurodegeneration due to the decline of proteostasis, we decided to check for ubiquitinated proteins as well^109,110.

Similarly, the accumulation of ubiquitinated proteins with age has been observed in Drosophila models of neurodegeneration^111–113, as well as in human neurodegenerative diseases^114–117. Interestingly, we detected different ubiquitination patterns in the SIRT6 knockout (KO) flies, with increased ubiquitination compared to the wild-type (WT) flies treated with DMSO (Supplementary Fig. 8A, B).

This finding suggests that treatment with the TDO2 inhibitor rescues this phenotype. It also correlates nicely with the disappearance of the protein folding and chaperone pathways from the enrichment analysis of DEGs between WT and SIRT6 KO flies following TDO2 inhibitor treatment.

Next, we used cleaved Caspase-3 as a common marker for apoptosis in the brain and neurodegeneration^118–120. In our Drosophila model, we used the caspase-like protein DRONC/DrICE. DrICE is a cysteine protease that is required for the induction of cell death; its truncated form rapidly induces apoptosis^121–123. We observed multiple truncated DrICE bands in the SIRT6 knockout (KO) flies (Supplementary Fig. 8C), whereas the wild-type (WT) flies primarily showed a single, larger band. The unique DrICE truncation pattern in SIRT6 KO flies may indicate increased apoptosis in their heads, a process that does not appear to be reversed by the TDO2 inhibitor. Overall, these results suggest that TDO2i can prevent an earlier stage of damage, such as the accumulation of ubiquitinated proteins, but not later stages like apoptosis.

SIRT6 deficiency in D. melanogaster results in transcription profile similar to neurodegeneration patients

To enable cross-species comparisons and to find similarities with brain expression and neurodegenerative diseases, we identified and converted orthologues of DEGs using the babelgene R package (Dolgalev 2022) [https://cloud.r-project.org/web/packages/babelgene/index.html] (Supplementary Fig. 6B). This process mapped 318 orthologues from the WT vs. KO comparison and 225 from the WT + TDO2 inhibitor vs. KO + TDO2 inhibitor comparison. The mapped genes were analyzed in RummaGEO against human databases for the brain to identify enriched terms, which were categorized to highlight the most significant biologically relevant pathways and processes.

RummaGEO analysis revealed significant overlaps in DEGs of SIRT6 KO D. melanogaster heads, with enriched terms and pathways associated with neurodegenerative diseases such as PD, ALS, and Wide mouth (also known as Pitt-Hopkins syndrome, a rare genetic disorder characterized by developmental delays, intellectual disability, and distinctive facial features)¹²⁴. This was consistent with findings from our mouse model and indicative of a neurodegenerative phenotype⁴¹. Interestingly, pathways related to lactic acidosis and increased CSF lactate were enriched, all known byproducts of using aerobic glycolysis, as it has been shown to occur in several systems lacking SIRT6^125–127. Importantly, elevated lactate levels in the CSF indicate impaired brain metabolism, often due to mitochondrial dysfunction, hypoxia, and neuroinflammation¹²⁸, and are associated with AD¹²⁹ (Fig.6C).

Notably, treatment with a TDO2 inhibitor drastically reduced the enrichment of neurodegenerative disease categories. Interestingly, some pathways, such as ribosome biogenesis, RNA transport, rRNA processing and amino acid metabolism, were uniquely enriched after TDO2 inhibitor treatment. These findings suggest that the TDO2 inhibitor partially reverses the neurodegenerative phenotype while also influences novel pathways, such as translation. This indicates that TDO2 inhibitors have alternative effects that merit further study before their use in neurodegenerative diseases.

Taken together, our findings from the humanized RNA-seq analysis revealed a profile resembling neurodegenerative diseases, which was significantly diminished by TDO2 inhibition (Fig.6 and Supplementary Figs. 6, 7).

Overall, we showed in several models that SIRT6 impairment leads to changes in Tryp catabolism, redirecting most of it to the Kyn pathway, increasing the levels of neurotoxic byproduct and reduction of the levels of Sero and Mel in the organism. In addition, we established a new D. melanogaster model that presents neurodegenerative phenotypes, which can be alleviated by TDO2 inhibition by allowing the proper usage of tryptophan. Overall, SIRT6 is a conserved regulator of tryptophan catabolism; in its absence, Tryp redirection accelerates the neurodegenerative phenotype through the accumulation of neurotoxic metabolites, and interfering with essential signals such as Mel and Sero—which eventually affects sleep and behavior dramatically (see summary Fig. 7).

Fig. 7. Graphical summary. SIRT6’s conserved role as a regulator of tryptophan metabolism.

SIRT6 regulates tryptophan metabolism, from its uptake to genes keeping balance between the kynurenine and serotonin-melatonin pathways. This conserved SIRT6-regulated pathway ensures balanced tryptophan catabolism, controlled sleep cycles, balanced mood, and avoids accumulation of neuroactive byproducts that may lead to inflammation and cell death. In aging and AD patients’ brains, reduction of SIRT6 leads to missregulation of the kynurenine pathway and accumulation of neuroactive byproducts leading to inflammation, cell death, and neurodegeneration. The depletion of tryptophan by its usage in the kynurenine pathway leads to decrease of serotonin and melatonin production, resulting in sleep disruption and imbalanced mood. Importantly, TDO2 inhibition can rescue the phenotype by redirecting tryptophan to other pathways, avoiding accumulation of neuroactive byproducts. Created in BioRender. Toiber, D. (2025) https://BioRender.com/utc1kfs.

Discussion

Tryptophan catabolism and SIRT6 in aging and neurodegeneration

Perturbations in tryptophan catabolism, which are characterized by an overproduction of potentially cytotoxic metabolites and a high activity of enzymes within the Kyn pathway (Fig. 1), have been observed in aging, neurological diseases (AD, HD, PD and ALS)^4–6,130, and psychiatric disorders^5,7,20. Despite this, the chain of events that leads to age-related defects in Tryp catabolism is not known.

Reduced expression of SIRT6, a critical enzyme in epigenetic regulation and DNA repair, has been associated with aging and neurodegenerative conditions^38,40. Here, we demonstrate that SIRT6 depletion leads to significant disruptions in Tryp metabolism, which may underpin neurodegeneration and behavioral impairments. Interestingly, the absence of SIRT6 alters metabolite production, aligning with changes observed in the CSF metabolome of patients with neurodegenerative, psychiatric, and inflammatory diseases.

We demonstrated that, in all models examined, the absence of SIRT6 is associated with an imbalance in the production of Tryp metabolites, as well as in significant changes in the expression of enzymes involved in Tryp transport and catabolism. In addition, we found changes in the expression of genes encoding for IDO, TDO2, KMO and AANAT, all essential rate-limiting enzymes. In the absence of SIRT6 in the brain, Tdo2 and Ido mRNA levels are highly elevated, while Kmo and Aanat gene levels are reduced. These findings align with elevated levels of KA, QA, and Kyn, as well as abnormal Mel oscillation in mouse serum. They underscore the critical role of these rate-limiting enzymes, as their expression levels in each model corresponded directly with the observed changes in Tryp metabolite concentrations. Taken together, these data suggest that, in the absence of SIRT6, the Kyn pathway of Tryp catabolism in the brain is favored over the Sero-Mel pathway, which is consistent with the observation that TDO2 and IDO1/2 overactivation leads to tryptophan depletion and reduction of its availability for other metabolic pathways. This, in turn, leads to a reduction in serotonin synthesis and depression-like behavior^58,131,132. Moreover, it is plausible to speculate that this SIRT6-dependent mechanism is involved in Sero depletion and the development of depression in patients with age-related neurodegenerative conditions. Indeed, we showed a reduction in Sero in mice serum while Tph2, the limiting factor for Sero production, is not significantly altered. Unfortunately, assessing depressive behavior in the brS6KO mouse model is challenging, as these mice exhibit hyperactivity and memory deficits⁴¹, which confound behavioral evaluations.

Of note, some metabolites within the Kyn pathway, which we would expect to be increased (including NAD+, a SIRT6 co-substrate), were depleted in SIRT6-deficient cells, similar to what occurs in aging. Elucidating the mechanisms responsible for these changes would require further investigation and is beyond the scope of this paper.

Melatonin and SIRT6: implications for neurodegeneration

Another defect in brS6KO mice that fits well with the proposed scenario of neurodegeneration acceleration through age-related SIRT6 depletion, and disruption of Tryp metabolism is the underproduction of Mel. Mel production declines with age¹³³, and its supplementation was found effective in arresting neurodegenerative phenomena in experimental models of AD, PD, and ischemic stroke in the short term^134,135. Mel is a free radical scavenger; its antioxidant properties were shown to prevent neuroinflammation¹³⁶. Moreover, Mel inhibits Tau aggregation, a pathology observed in several neurodegenerative conditions^137,138. However, clinical trials using Mel showed promising results only in mild cognitive impairment¹³⁹. Interestingly, in our study, Mel treatment of SIRT6 KO D. melanogaster, unlike TDO2 inhibition, did not improve their climbing behavior. Therefore, it may not be possible to counterbalance an excess in Kyn pathway metabolites by Mel supplementation¹⁴⁰.

Here, we demonstrate that the absence of SIRT6 in mouse brains results in abnormal Mel oscillation in the serum, which correlates with altered Aanat mRNA levels. Notably, we identified that SIRT6 binds directly to the promoter region of the AANAT gene, suggesting that SIRT6 plays a direct role in regulating Mel production and secretion. This function of SIRT6 appears to be evolutionarily conserved, as reduced Mel levels were also observed in the SIRT6 knockout D. melanogaster model. A recent study suggested that SIRT6 and Mel samples from the buccal epithelium can be biomarkers of age in humans¹⁴¹, which could also mean use as predictors of age-related neurodegeneration. In addition, there is evidence that an increase in SIRT6 activity is mediated via the cardiac Mel receptor¹⁴², indicating a possible feedback loop. Our findings highlight a potential mechanistic interplay between melatonin deficiency, circadian disruptions, and neurodegeneration in SIRT6-deficient models.

Circadian dysregulation in SIRT6-deficient models

SIRT6's role in the circadian clock has been investigated in liver⁷⁰; SIRT6 was found to directly regulate PER by deacetylating it⁶⁹, as well as to directly interact with BMAL1 and CLOCK⁷⁰. We observed that SIRT6 depletion in brS6KO mice brains is associated with a reduced recruitment of the clock core transcription activators—BMAL1 and CLOCK—to the chromatin. It is possible that this direct interaction affected their recruitment. Moreover, we observed significant disruption in the oscillation of the expression of circadian genes Per1/2 and Cry in the brains of brS6KO mice and SIRT6 KO D. melanogaster. We also found that SIRT6 binds directly to the promoter regions of CRY and PER. Therefore, SIRT6 may play an essential role in regulating the circadian clock in the brain by at least three mechanisms: transcription regulation, recruitment to chromatin, and shifting of Tryp metabolism. Notably, the Kyn pathway byproducts are natural activators of the aryl hydrocarbon receptor (AhR)^143,144, which have a role in inflammation, circadian clock regulation and neurodegeneration^67,144–150. Importantly, the robustness of the circadian clock machinery declines with age^151–153, and it is plausible to speculate that SIRT6 reduction may be one of the contributing factors to this decline.

Sleep Disruptions and Neurodegeneration in SIRT6 deficiency

Our results indicated abnormal sleeping patterns, increased total activity, and more fragmented sleep, as well as faster loss of sleep rhythm in dark-stress conditions in the absence of SIRT6 in the brain. Various studies have reported a decrease in total sleep time, a decrease in sleep quality, and more fragmented sleep in aging individuals^73–75. Notably, sleep disorders are frequently reported in AD⁷⁶ and PD⁷⁷ patients too. On its own, sleep deprivation has been shown to impair the clearance of toxic misfolded proteins like Tau from the brain¹⁵⁴. Therefore, impairments in melatonin secretion, neurotoxic metabolites and the abnormal circadian clock may account for the accumulation of toxic Tau, and possibly even cognitive decline and memory loss in our brS6KO mice⁴¹.

TDO2/IDO: a potential therapeutic target for neurodegenerative diseases

High expression levels of TDO2 and IDO1, the limiting factors in tryptophan directing to the Kyn pathway⁴, have been associated with neurodegeneration in several studies in a variety of models^{4,55,85,155,156}. Increased levels of these enzymes lead to accumulation of neurotoxic byproducts as well as neuroinflammation^4,157. Importantly, our results highlight a conserved role for SIRT6 in regulating the expression of TDO2 and IDO1/2 genes across human cellular models, mice, and D. melanogaster. In here, we focused on the effects of TDO, the neuronal expressed limiting factor, however, IDO1 was also affected, and this could impair also microglia function.

SIRT6 elimination in D. melanogaster resulted in neurodegenerative changes and behavioral defects that could be partly reversed by pharmacological inhibition of TDO2. TDO2 inhibitor alleviates stress-related pathways (oxidative stress and protein folding), while elevating visual system-related pathways. Importantly, cross-species analysis of DEGs supports the relevance of our model as a neurodegenerative model. Notably, TDO2 inhibition reduced the prevalence of neurodegenerative disease pathways while introducing novel effects on translation, ribosomal activity and amino acid metabolism. This supports the scenario in which neurological and behavioral pathologies are caused by SIRT6 deficiency and mediated via Tryp metabolism, but the additional changes in gene expression should be further investigated before recommending this as a therapeutic option.

Key limitations of our study

Experimental models with silenced or knocked out genes do not completely reflect the gradual decay in aging. Moreover, the interconnectedness of cellular pathways makes it very difficult to account for all the ways in which the removal of an important component of those pathways may elicit a physiological or behavioral change. However, SIRT6 activity is reduced in aging, and even further in neurodegeneration, suggesting this model is could be physiologically relevant. Verifying these observations in other experimental models, using different experimental approaches, would be needed for the extrapolation of our findings to the organisms outside of this study.

In conclusion, our findings from human, murine, and insect models demonstrate that SIRT6 loss disrupts tryptophan metabolism, impairing sleep quality and circadian rhythms, and potentially leading to neurodegeneration and behavioral pathology. The consistency of these effects across species suggests an evolutionarily conserved function, though with species-specific variations. This highlights the critical role of SIRT6 in tryptophan regulation, likely through multiple or even redundant mechanisms (Fig. 8). A deeper understanding of these pathways could pave the way for novel therapeutic approaches for neurological and neurodegenerative diseases.

Fig. 8. Graphical summary of SIRT6’s regulatory role in balancing tryptophan metabolism and its effect on brain health.

SIRT6 regulates tryptophan metabolism, maintaining balance between the kynurenine and serotonin-melatonin pathways. Left panel: Under normal SIRT6 function, balanced tryptophan catabolism ensures proper serotonin and melatonin secretion, maintains circadian regulation, enhances sleep quality, and prevents neurodegeneration. Right panel: SIRT6 deficiency leads to excessive kynurenine pathway activation, producing toxic metabolites (KA kynurenic acid, XA xanthurenic acid, QA quinolinic acid, AA anthranilic acid) that cause cell death. Tryptophan depletion through the kynurenine pathway reduces serotonin and melatonin production, resulting in altered sleeping patterns, reduced neurotransmitter secretion, and neurodegeneration. SIRT6 acts as a gatekeeper by balancing tryptophan catabolism gene expression, restricting the accumulation of kynurenine pathway byproducts, and preserving neuronal integrity. Created in BioRender. Toiber, D. (2025) https://BioRender.com/gq8zznc.

Methods

This research complies with all relevant ethical regulations; All animal handling and experimentation were conducted in accordance with procedures approved by the animal facility of Ben-Gurion University of the Negev, Israel. All surgical and experimental procedures were approved by the Institutional Animal Care and Use Committee of Ben-Gurion University. (Authorization number: IL-55-09-2023C, IL-33-05-2019C).

Resource availability

Plasmids and strains generated in this study (Supplementary information Table 1) are available and can be requested from the lead author with a completed material transfer agreement. This study did not generate new unique reagents. Antibodies, reagents, kits, probs and primers information are provided in Supplementary Information file Supplementary Table 1-4.

Experimental model and subject details

Generation of SIRT6 KO cells

SH-SY5Y, HeLa, and ARPE19 SIRT6 KO cells were generated as described previously in Kaluski et al.⁴¹. Briefly, cells were infected with the lentivirus GeCKO system. We used 2 sgRNA molecules, targeting Sirt6 CRISPR2 (GCTGTCGCCGTACGCGGACA) and CRISPR3 (GCTCCACGGGAACATGTTTG), and a scrambled gRNA as a control. Constructions were kindly donated by the Prof. Aharoni lab (Ben-Gurion University, Israel). Cells were selected by means of 2 µg/mL puromycine for a week, followed by serial dilutions to a single cell colony.

Cell cultures

All cells were cultured in Dulbecco’s Minimal Essential Medium (DMEM with 4.5 g/L glucose)/ DMEM/F-12 (only ARPE19 cell line). Mediums were supplemented with 10% fetal bovine serum (FBS), 1% penicillin and streptomycin cocktail, and 1% L-glutamine, in an atmosphere with 5% CO₂ at 37 °C.

Generation of brS6KO mice

brS6KO mice were generated as described previously described in Kaluski et al.⁴¹. Briefly, Sirt6 conditional vector was inserted via a Neo cassette (flanked by two Frt seq) together with Sirt6 exon2 flanked by two loxP sites. Targeted ES cells (V6.5) were injected into C57BL6/J blastocysts. The Neo cassette was deleted in vivo by crossing the chimeras with a mouse expressing the Flpe endonuclease. Mice were backcrossed for 3 generations with C57BL6/J mice to obtain heterozygous mice with 97% of genetic background originating from the C57BL6/J strain. These mice were bred with C57BL/Nestin-Cre/J mice (Jackson Laboratories).

Circadian behavior analysis

Regular animal handling

Four mice were housed per cage in a room with 12-h light-dark cycles (7:00 AM to 7:00 PM), at 22–24 °C, with ad libitum access to food and water, for 2–4 months. All animal handling and experimentation were conducted in accordance with procedures approved by the animal facility of Ben-Gurion University of the Negev, Israel.

Animal handling for circadian behavior evaluation (dark stress)

Mice were separated into individual cages equipped with 4 sensors above to monitor movement. Initially, animal movement was recorded 2–3 weeks in regular 12-h dark-light cycles. After that, mice were kept in constant darkness for 30 days, which was followed by a recovery period of 3–6 weeks under the normal 12-h dark-light cycle. After the recovery period, the animals were sacrificed in 6-h intervals using Isoflurane anesthesia in order to collect blood and tissue samples.

Circadian behavior analysis

Mouse movement was recorded in 1-min intervals, using 4 movement sensors and averaging their reads. To minimize artifacts from hyperactive animals, the averaged reads for each 1-min point for each mouse were divided by the mean value of all averaged reads for that mouse. Since wake/sleep is a binary state, any movement sensor values indicating movement (ie, those above 0) were set to 1, and all zero values (non-movement) were set to 0. The movement graphs were plotted using a self-developed MATLAB program, and the following quantitative parameters were extracted:

1. For each mouse, the percentage of activity time was calculated as [(time awake)/(recorded time)] × 100 for dark periods (when the mouse should be awake), light periods (when it should be asleep), and for the total recorded time.

2. Under the condition of dark stress (ie, replacement of the light period with dark), animals began increasing the duration of their wake time in this new dark (formerly light) period. To quantify circadian clock disturbance, we calculated the shift in time awake during this new dark period by calculating the difference in hours between the animal’s waking time on the first and last day of dark stress.

Finally, the daily change in time awake during the dark stress period was calculated as the difference in waking hours divided by the number of days. For example, a daily change of 5 min in waking time means that the mouse woke up 5 min earlier on each day of the dark stress, compared to the previous day.

Immunofluorescence

Cells were rinsed with phosphate buffer saline (PBS) and fixed with 4% paraformaldehyde for 10 min at room temperature, followed by two additional washes. Cells were permeabilized (0.1% tri-sodium citrate and 0.1% Triton X-100 in Distilled water, pH 6) for 5 min and rinsed again. After 30 min of blocking (0.5% bovine serum albumin [BSA], 5% goat serum, 0.1% Tween-20 in PBS), cells were incubated with primary antibody diluted in blocking buffer overnight at 4 °C. The next day, cells were washed three times with wash buffer (0.25% BSA, 0.1% Tween-20 in PBS), incubated for 1 h with the secondary antibody (diluted in blocking buffer 1:200) at room temperature and rinsed three more times. Cells were then DAPI-stained for 3 min at room temperature and rinsed with PBS twice before imaging. Nuclear mean intensity was measured and analyzed using ImageJ (https://imagej.net/ij/download.html).

Chromatin Extraction

Cells were collected, rinsed in PBS, and resuspended in 2–5 pellet volumes of lysis buffer (10 mM HEPES pH 7.4, 10 mM KCl, 0.05% NP-40 and protease, deacetylase, and phosphatase inhibitors). Samples were incubated on ice for 20 min and centrifuged at 14,000 × g at 4 °C for 10 min. The supernatants containing cytoplasmic proteins were removed and kept separately. Cell pellets were resuspended with 2–5 volumes of 0.2 M HCl, incubated on ice for 20 min, and centrifuged at 14,000 × g at 4 °C for 10 min. Supernatants were collected and neutralized with an equal volume of 1 M Tris-HCl, pH 8.

Western blots and immunostaining

For each Western blot analysis, 10–40 µg protein samples were loaded onto 4–20% Tris-glycine polyacrylamide gel (BioRad) or 12% gels made in house. Proteins were separated for 1 h at 120 V and then blotted to nitrocellulose membranes at 100 V for 90 min. The blots were blocked with 5% skim milk in TBST (15 mM Tris-HCl, pH 7.5, 200 mM NaCl, and 0.1% Tween 20) for 1 h at room temperature. Membranes were incubated overnight with primary antibodies, diluted per manufacturer recommendation and developed using a chemiluminescence reagent (WesternBright Quantum HRP substrate, K-12042-C20).

Brain RNA preparation

One brain hemisphere was used for RNA preparation. RNA was then extracted using Nucleospin RNA Plus Kit (740984, MACHEREY-NAGEL GmbH & Co. KG, Duren, Germany).

SIRT6 ChIP- qPCR

ChIP experiments were performed as described previously in Nathan P Gomes et al.¹⁵⁸. Briefly, SH-SY5Y cells expressing a degron containing a DD-SIRT6-myc chimera were cultured up to 80% of confluence. SIRT6-myc expression was induced by the addition of 50 nM Shield-1 during 6 h, to avoid protein degradation. In the case of ChIP-negative controls, the induction of SIRT6-Myc protein was avoided. Cells were cross-linked with 1% formaldehyde for 10 min and blocked with 0.125 M Glycine for 5 min. Cross-linked cells were scrapped in RIPA buffer (150 mM NaCl, 1% NP-40, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate [SDS], 50 mM Tris pH 8, 5 mM EDTA, 0.5 mM PMSF, 50 mM NaF or 0.2 mM sodium orthovanadate, 5 µM trichostatin A), and chromatin was sonicated with a needle sonicator (VibraCell VCX130, Sonics, CT, USA) for 25 min, with 30-s on-off cycles at 40% of amplitude. One milligram of the sonicated chromatin was incubated overnight with previously blocked myc-tag magnetic beads (HY-K0206, MedChemExpress, NJ, USA). The beads were then washed 4 times with RIPA buffer, 4 times with LiCl buffer (500 mM LiCl, 100 mM Tris-HCl pH 8.5, 1% NP-40, 1% sodium deoxycholate), and 2 times with TE buffer (10 mM Tris-HCl pH 8, 1 mM EDTA). Chromatin was removed from the beads using elution buffer (70 mM Tris-HCl pH 8, 1 mM EDTA, 1.5% SDS) and cross-linking was reversed by adding 200 mM NaCl, 1 U of proteinase K, and incubating the eluted chromatin at 65 °C for 5 h. The obtained DNA was further purified using the Nucleo Spin Gel and PCR Clean-Up kit (740609, Macherey-Nagel, Duren, Germany). For qPCR analysis of SIRT6-specific binding regions in the genome, we used the set of primers listed in Supplementary Table 3, and the Sso Advanced Universal SYBR green supermix (BioRad, 1725274, CA, USA). We used 1 μL of Input and ChIP samples per qPCR reaction.

For ChIP-seq analysis, Input and ChIP samples were sequenced at 25 million sequencing deep in a NovaSeq 6000 sequencer (Illumina, CA, USA). Adapter sequences were removed from obtained reads using trimmomatic package (PMC4103590), and curated reads were aligned against the hg38 human genome using a bowtie2 aligner (10.1038/nmeth.1923). The obtained BAM files were filtered for unmapped and duplicated reads using SAMtools (PMC7931819). ChIP peaks were identified using the MACS3 package (10.1186/gb-2008-9-9-r137). Visualization of the obtained BED files was done using IGV software.

RNA-seq

RNA preparation and quality control

RNA was extracted from the left brain hemispheres of 12 WT and 10 SIRT6KO mice, using the NucleoSpin RNA Plus kit (MACHEREY-NAGEL GmbH & Co. KG, catalog number 740984.50), according to the manufacturer’s manual.

The purified RNA was then cleaned from residual genomic DNA contamination using the RNeasy MinElute Cleanup Kit (QIAGEN, catalog number 74204), according to the manufacturer’s manual.

We assessed RNA Integrity Number (RIN) via TapeStation and only used samples with RIN > 8.7.

Full-length poly-A RNA sequencing

Library preparation was conducted by The Crown Genomics Institute, Nancy and Stephen Grand Israel National Center for Personalized Medicine, Weizmann Institute of Science, Israel (G-INCPM).

Briefly, we used an in-house INCPM mRNA-seq library kit (G-INCPM, Weizmann Institute of Science) for full-length RNA-seq with polyA-based capturing. Sequencing was done using 2 lanes of NextSeq 500 High Output v2.5 Kit (75 cycles) (Illumina Inc., catalog number 20024906).

RNA-seq data analysis

Read preprocessing, alignment, and differential gene expression analyses of WT and brS6KO mouse transcriptomic profiles (deposited under GEO accession GSE221077) were described previously in Smirnov et al.⁴³. Briefly, the nf-core pipeline was used to filter reads and remove adapters via Trim Galore! tool, align reads to the reference genome using STAR software, and quantify expression levels using Salmon. Differential expression analysis was performed via DESeq2 package with FDR p value < 0.05 threshold used to determine significantly changed genes.

Melatonin and serotonin measurements in mice serum

Melatonin and serotonin in mouse serum were extracted and measured using Melatonin ELISA Kit (Abcam, ab213978) and Serotonin ELISA Kit (Enzo, ADI-900-175), respectively.

Tryptophan and kynurenic acid measurements

Tryptophan and KA in cells and D. melanogaster flies (10 flies per sample) were extracted and measured using L-Tryptophan (Trp) Fluorescence Assay Kit (Mediomics Bridge-It®, 1-1-1002TRP-PS384) and the General Kynurenic Acid Kit (MyBioSource, MBS9136204), respectively.

Metabolomics

Metabolite extraction

ES cells

ES cell metabolomic was performed by Asara lab in Proteomics, Lipidomics, Metabolomics Core at BIDMC, Harvard University, according to the paper by Yuan et al. 2012, Nature Protocol179 and is briefly described here:

Sample preparation

Polar metabolites were extracted from cultured cells, using 80% methanol (−80 °C) to precipitate proteins and extract metabolites. ≥2–3 million cells were lysed with cold methanol, scraped, centrifuged, and supernatants were collected.

LC-MS/MS analysis

Metabolite profiling was performed using hydrophilic interaction liquid chromatography (HILIC) coupled to a 5500 QTRAP hybrid triple quadrupole mass spectrometer (AB/SCIEX), enabling positive/negative ion-switching.

Chromatography

Utilized amide-capped (XBridge) or amino-capped (Luna NH2) HPLC columns under high-pH conditions (pH 9.0) with buffers containing ammonium hydroxide and ammonium acetate.

Mass spectrometry

Operated in selected reaction monitoring (SRM) mode, targeting 258 metabolites across major metabolic pathways. A 15-min LC-MS/MS run was used per sample, with polarity switching (<50 ms) to detect both positive and negative ions.

Data processing and quantification:

Peak integration was performed using MultiQuant 2.0.

Relative quantification was based on peak areas without internal standards.

Statistical analysis and pathway mapping were carried out using MetaboAnalyst, enabling PCA, heat maps, and pathway enrichment.

Replicates: Biological triplicates were used for reproducibility and statistical validation.

Reagents & Equipment

Reagents: LC/MS-grade water, methanol, acetonitrile, ammonium hydroxide, ammonium acetate.

Equipment: 5500 QTRAP mass spectrometer, HILIC HPLC columns, SpeedVac/lyophilizer, and computational tools (MetaboAnalyst, MATLAB, R).

The source and compound settings were as described in Table 1.

Table 1.

LC-MS/MS analysis source and compound settings

Temperature	475°C
Curtain gas	20–25 (nitrogen)
Collision gas	High (nitrogen)
Ion source gas 1	33
Ion source gas 2	33
Declustering potential	+ 93 in positive ion mode/−93 in negative ion mode
Entrance potential	+10 in positive ion mode/−10 in negative ion mode
Collision cell exit potential	+10 in positive ion mode/−10 in negative ion mode

The generic HPLC gradient used in this experiment is detailed in Table 2.

• The data extracted and analyzed according to the above protocol are presented in Fig. 1A–D (the ESC plot). For the full list of metabolites, raw data and statistics, please see the Source Data file: 1A–D; Supplementary 1.1A–1.1C.

Table 2.

Generic HPLC gradient used in LC-MS/MS experiment

HPLC columns	1. Amide XBridge HPLC column (3.5 µm; 4.6 mm inner diameter (i.d.) × 100 mm length; Waters, cat. no. 186004868) 2. Luna NH2 HPLC column (5.0 µm; 4.6 mm i.d. × 50 mm length; Phenomenex, cat. no. 00B-4378-E0)
HPLC buffer A	pH = 9.0: 95% (vol/vol) water, 5% (vol/vol) acetonitrile, 20 mM ammonium hydroxide, 20 mM ammonium acetate
HPLC B buffer B	100% acetonitrile
Flow rate	Approximately 350–400 μl min−1 (back pressure should not exceed ~3000 p.s.i. at 2% (vol/vol) B)
85% (vol/vol) B	0.0 min
85–30% B	3.0 min
30–2% B	12.0 min
2% B	15.0 min
2–85% B	16.0 min
85%	23.0 min

Cell lines (SH-SY5Y, HeLa, ARPE19) and D. melanogaster: Sample collection

SH-SY5Y, HeLa, and ARPE19 (control and SIRT6 KO) cells were seeded in 6-well plates at a density of 3 × 10⁵ cells per well. After 24 h, cells were washed twice with ice-cold saline, snap-frozen in liquid nitrogen, and stored at –80 °C until processing. For Drosophila melanogaster analyses, each sample comprised 10 flies aged 14 days, which were snap-frozen in liquid nitrogen and stored at –80 °C.

Metabolite extraction was performed in a mixture of ice/dry ice, by a cold two-phase methanol–water–chloroform solution78. The samples were resuspended in 800 μl of precooled methanol/water (5/3) (v/v). Afterwards, 500 μl of precooled chloroform was added to each sample. Samples were vortexed for 10 min at 4 °C and then centrifuged (max. speed, 10 min, 4 °C). The methanol–water phase containing polar metabolites was separated and dried using a vacuum concentrator at 4 °C overnight.

For the detection of tryptophan derivatives by LC–MS, an Infinity II 1290 (Agilent Technology) with a thermal autosampler set at 4 °C, coupled to a Q-TOF 6546 (Agilent Technology) was used for the separation of metabolites. Samples were resuspended in 50 µL of water and 20 µL of sample were injected, the separation of metabolites was achieved with a flow rate of 0.25 ml/min, at 30 °C, on a C18 column (Acquity UPLC Premier HSS C18 1.8 μm 2.1 × 100 mm). A gradient was applied for 40 min (solvent A: H2O, 0.1% Formic acid, 15 mM acetic acid—solvent B: Acetonitrile + 0.1% Formic acid) to separate the targeted metabolites (0 min: 8% B, 2 min: 8% B, 14 min:90% B, 16 min: 90% B, 17 min: 8% B, 22 min: 8% B).

LC–MS data, raw files were processed using Profinder (Agilent, version 10.0) and MassHunter Qualitative Analysis (Agilent, version 10.0) for peak extraction, alignment, and metabolite identification. Quantification and normalization to protein or tissue weight were performed using MATLAB (R2023b; MathWorks, Natick, MA, USA).

(Note: in D. Melanogaster analyses, each sample contained material from 10 flies aged 14 days. The melatonin peaks from one WT male sample and one SIRT6KO male sample were removed from the analysis due to a 10-fold and 500-fold increase, respectively. They were considered outliers, which was supported by an outlier test).

Metabolomics Data Analysis

We analyzed metabolomics data from control and SIRT6 KO mouse ES cells as described previously in Smirnov et al.³⁰. Briefly, raw metabolomics measurements were acquired in triplicate via relative quantification using a targeted LC-MS/MS protocol with positive/negative polarity switching, as previously described¹⁵⁹. Chromatographic Q3 peak areas were integrated and quantified via MultiQuant v2.0 with the default settings provided in Asara et al.¹⁵⁹. Next, metabolomics features with more than 50% missing values were removed using the “RemoveMissingPercent()” function in MetaboAnalyst. Missing value imputation was performed using the K-nearest neighbors (KNN) algorithm with “ImputeMissingVar()” function in MetaboAnalyst. Finally, median normalization and log₁₀ transformation were applied to the obtained metabolomics profiles, and differential features were identified via Student’s t test. The acceptable FDR was set as <0.05, with |log₂FC| > 0.58. The complete MetaboAnalyst analysis, including figures and the full R script of the analysis (Rhistory.R), is available on GitHub: https://github.com/SIRT6/Kaluski_et_al_2024/tree/main/Metaboanalyst_4.12.21. HeLa metabolite abundances were filtered with a coefficient of variation (CoV) <30 in each group, and significantly changed metabolites were identified using the Wilcoxon rank sum test (FDR < 0.05). Principal component analysis (PCA) of HeLa samples was performed via prcomp function in R, followed by the calculation of the metabolite contributions to PC1 using factoextra package. Quantified and annotated metabolites in SH-SY5Y cells were normalized in a sample-wise manner with median normalization method and then t-test was applied to retrieve differentially changed features (p < 0.05) between SIRT6 KO and control samples. The abundances of tryptophan derivatives metabolites in HeLa, SH-SY5Y, and ARPE19 cells datasets were normalized to the abundance of the precursor to reduce the variability between samples of the metabolite and compared using t-test.

The metabolite/precursor ratios used in this paper:

• Kynurenine/tryptophan

• Kynurenic acid/kynurenine

• Anthranilic acid/kynurenine

• Serotonin/tryptophan

• Xanthurenic acid/kynurenine

• In D. Melanogaster datasets, the abundances of tryptophan derivatives metabolites were normalized to the WT average of each batch. The data extracted and analyzed according to the above protocol are presented in Fig. 1D (SH-SY5Y and HeLa plots only), Fig. 1F, Supplementary Fig. 1C–F, Fig. 5F and Supplementary Fig 5F. For raw data and statistics, please see the Source Data file: 1D, 1F, 5F and Supplementary 5F.

• Tryp levels measured using a commercial kit (L-Tryptophan [Trp] Fluorescence Assay Kit, Mediomics Bridge-It®, 1-1-1002TRP-PS384) are shown in Supplementary Fig. 1A and 5E.

• KA levels measured using a commercial kit (General Kynurenic Acid ELISA Kit, MyBioSource, MBS9136204) are shown in Supplementary Fig. 5E.

D. melanogaster strains

D. melanogaster flies, strain w1118, were kindly donated by Dr David Ben-Menahem (Ben-Gurion University, Israel). All flies used in the study were backcrossed to w1118 for 8 generations. SIRT6 (#39) D. melanogaster mutant strain was a kind gift from Dr Alena Bruce Krejci from the University of South Bohemia in the Czech Republic. The SIRT6 knockout flies were created by crossing transgenic Sirt6 gRNA flies with UAS-Cas9 stock, selecting the Sirt6-deficient allele and back-crossing the stock five times to the w1118 background. The resulting homozygous stock contained Sirt6 alleles with four nucleotide deletions at position 96–99 after ORF, causing a translational shift after the first 31 amino acids of SIRT6.

D. melanogaster growth conditions

Flies were maintained under non-crowding conditions on standard cornmeal diet 5 g dextrose, 2.5% yeast extract, 8.6% cornmeal, 2% agar, 0.1% ortho-phosphoric acid, and 0.1% propionic acid and grown at 25 °C with 12 h light/dark cycle. Age-matched flies were used for all experiments.

D. melanogaster RNA extraction

RNA from D. melanogaster heads was extracted using the single cell RNA purification kit (NORGEN Cat. 51800). Each sample contained 10 heads.

D. melanogaster negative geotaxis assay (limbing test)

Eight to ten flies aged 7, 14, or 21 days were placed in a cylindrical tube (diameter, 2 cm; height, 33 cm) and left to acclimatize for 5 min. Tubes were tapped 5 times in order to gather the flies at the base. The flies were then allowed to climb for 30 s, and the number of flies that climbed at least 8 cm was determined. The same cohort of flies was tested 10 times, with 1 min rest between each trial.

Administration of TDO2 inhibitor and melatonin to D. melanogaster

TDO inhibitor 680C91 (Tocris Bioscience, Bristol, UK) was dissolved in dimethyl sulfoxide (DMSO), and melatonin (phr1767, Sigma) was dissolved in ethanol. Both were added to the yeast extract in the diet to a final concentration of 100 µM (680C91)⁸⁵ or 0.5 mM (melatonin)⁹². Flies were introduced to the inhibitor-supplemented media upon eclosion; supplemented media were changed every other day.

Whole-brain vacuole assessment in D. melanogaster

Immunofluorescence procedures were performed as described previously by Behnke et al.⁸³.

Briefly, flies were fed with the TDO2 inhibitor or melatonin for 21 days post eclosion, brains were dissected in PBS and fixated with 4% paraformaldehyde, rinsed, and permeabilized with PBST 0.1%. The fixated brains were stained with DAPI and phalloidin. Whole-brain images were obtained using confocal microscopy and used for vacuole analysis.

D. melanogaster single head RNA-seq

RNA preparation

RNA was extracted from single head of 21 days old verging female w1118 / SIRT6 KO D. melanogaster fed with 100 µM TDO2 inhibitor/ DMSO for 21 days. RNA was extracted using BioTri RNA (Bio-Lab, catalog number 959758027100), according to the manufacturer’s manual followed by cleaning step using RNA Clean & Concentrator MagBead (Zymo Research, catalog number R1081).

RNA sequencing by Azenta

RNA sample QC, library preparations, sequencing reactions, and initial bioinformatic analysis were conducted at GENEWIZ, LLC./Azenta US, Inc (South Plainfield, NJ, USA) as follows:

Sample QC

Total RNA samples were quantified using Qubit 4.0 Fluorometer (Life Technologies, Carlsbad, CA, USA), and RNA integrity was checked with 4200 TapeStation (Agilent Technologies, Palo Alto, CA, USA).

Library preparation and sequencing

Samples were initially treated with TURBO DNase (Thermo Fisher Scientific, Waltham, MA, USA) to remove DNA contaminants. ERCC RNA Spike-In Mix (Cat: #4456740) from ThermoFisher Scientific was added to normalized total RNA prior to library preparation following the manufacturer’s protocol. The next steps included performing rRNA depletion using QIAGEN FastSelect rRNA HMR or Fly Kit (Qiagen, Germantown, MD, USA), which was conducted following the manufacturer’s protocol. RNA sequencing libraries were constructed with the NEBNext Ultra II RNA Library Preparation Kit for Illumina by following the manufacturer’s recommendations. Briefly, enriched RNAs are fragmented for 15 min at 94 °C. First strand and second strand cDNA are subsequently synthesized. cDNA fragments are end repaired and adenylated at 3’ends, and universal adapters are ligated to cDNA fragments, followed by index addition and library enrichment with limited cycle PCR. Sequencing libraries were validated using the Agilent Tapestation 4200 (Agilent Technologies, Palo Alto, CA, USA), and quantified using Qubit 4.0 Fluorometer (ThermoFisher Scientific, Waltham, MA, USA) as well as by quantitative PCR (KAPA Biosystems, Wilmington, MA, USA). Azenta US, Inc. 2910 Fortune Circle West, Suite E, Indianapolis, IN 46241 | Tel: (858)527 - 7080.

The sequencing libraries were multiplexed and clustered on the flowcell. After clustering, the flowcell was loaded on the Illumina NovaSeq instrument according to the manufacturer’s instructions. The samples were sequenced using a 2 × 150 Pair-End (PE) configuration.

RNA-seq data analysis

An analysis of Drosophila melanogaster profiles (GSE…) was performed according to the protocol described above and using BDGP6.46 genome assembly. Low-expressed genes were filtered out, requiring a minimum of one count in at least one sample per group and no less than 20 counts in total detected for a gene across all samples. Volcano plots for DE testing results were prepared via the EnhancedVolcano package (v1.22.0). Gene ontology analysis of the up- and down-regulated DE genes was done using the ClusterProfiler (v4.12.6) and org.Dm.eg.db (v3.19.1) R packages. The list of genes associated with amino acid transport was obtained from the FlyBase resource (https://flybase.org/). The list of human orthologs of DE D. melanogaster genes was obtained using “orthologs()” functions from the babelgene package (v22.9).

To enable the comparison of D. melanogaster RNA-seq data with human datasets in Rummageo, we identified and converted orthologues of differentially expressed genes using “orthologs” function from babelgene R package. These mapped genes were analyzed against all available human databases in Rummageo to identify enriched terms. The resulting terms were then categorized to highlight the most common and biologically relevant pathways and processes consistent with our dataset.

Code of the D. melanogaster single-head RNS-seq analysis is available on: https://github.com/SIRT6/Kaluski_et_al_2024/blob/main/Scripts/Drosophila_analysis_paper.ipynb.

D. Melanogaster total protein extraction

Flies were fed with the TDO2 inhibitor/DMSO for 21 days post-eclosion. Flies were snap-frozen and briefly vortexed to allow head separation. Six heads were collected into an Eppendorf tube. Each sample was homogenized using a spatula in 40 μL RIPA buffer (Tris-HCl pH 7.4 0.05 M, NaCl 0.15 M, EDTA 1 mM, NP-40 1%, SDS 0.1%, PMSF 1 mM, sodium deoxycholate 0.5%, phosphatase inhibitor cocktail), followed by 10 min of bath sonication, 30 min of incubation on ice, and centrifugation at 12,000 × g for 10 min at 4 °C. Protein lysates were collected from the supernatant.

Protein was extracted from pools of six heads per sample.

Statistics and reproducibility

Statistical analyses were used as accepted in the field. All replication experiments were included. No statistical method was used to predetermine sample size. The Investigators were not blinded to allocation during experiments and outcome assessment, since the samples must be marked (in the mouse experiments, samples were labeled with mouse numbers and therefore blinded). Between-group comparisons that used t-test, ANOVA (one-way or two-way, with post hoc Dunnet or Tukey tests), and Wilcoxson rank sum test were conducted at the significance level of 0.05 and performed using GraphPad Prism 10.

Mice sex related experimental design

We aimed to balance the distribution of sexes as evenly as possible across groups. The sexes of the individual subjects in each group are detailed in the tables Supplementary Table 2 (Supplementary Data 1). However, since we maintain the colony and perform the breeding ourselves, we cannot fully control the distribution of sex and genotype in each cohort. As a result, we are unable to create perfectly balanced groups in terms of age, genotype, and sex. For each experiment, we have provided tables detailing the sex distribution within each group. Due to the small number of individuals per group (≤2), we are unable to perform statistical tests to assess the effect of sex on the results.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Author contributions

S.K.-K. designed the study, performed experiments, analyzed data and wrote the manuscript. D.S. designed the study, analyzed data and contributed to the writing of the manuscript. A.G.-V., A.M.F.C., M.P., and E.D.A.-B. performed the experiments and analyzed data. D.Sm. performed bioinformatics work (data analysis and coding) and contributed to the writing of the manuscript. B.G., A.Z., E.E., M.Po., and M.E. performed the experiments. A.B.K. and U.A. provided SIRT6 KO flies, provided access to the Drosophila facility and essential materials that supported this research. E.Kh., D.G., S.-M.F., and D.T. supervised the research and designed the study and provided conceptual input. D.T. coordinated the project, wrote the manuscript. All authors discussed the results and commented on the manuscript.

Peer review

Peer review information

Nature Communications thanks the anonymous reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Data availability

The SIRT6 ChIP-seq data created and used in this study have been deposited in the GEO database under accession code GSE295913. The D. Melanogaster Singel-Head RNA-seq data acreated and used in this study have been deposited in the GEO database under accession code GSE309387. The Mouse WT and brS6KO transcriptomic profiles (Smirnov et al.⁴³) data used in this study are available in the GEO database under accession code GSE221077. The Mouse SIRT6 KO, Het and WT brain profiles (Stein et al.⁴⁴) data used in this study are available in the GEO database under accession code GSE178432. Softwares and algorithms used in this paper are listed in Supplamentary Table 1. Metabolomic data is available at the NIH Common Fund’s National Metabolomics Data Repository (NMDR) website, the Metabolomics Workbench, where it has been assigned Project ID PR002694. The data can be accessed directly via its Project [10.21228/M8F26W]. This work is supported by Metabolomics Workbench/National Metabolomics Data Repository (NMDR) (grant# U2C-DK119886), Common Fund Data Ecosystem (CFDE) (grant# 3OT2OD030544) and Metabolomics Consortium Coordinating Center (M3C) (grant# 1U2C-DK119889). Source data are provided with this paper.

Code availability

The code generated and used to analyze data in this study has been deposited in the Zenodo database under [10.5281/zenodo.17250775].

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s41467-025-67021-y.