Elovanoids are neural resiliency epigenomic regulators targeting histone modifications, DNA methylation, tau phosphorylation, telomere integrity, senescence programming, and dendrite integrity

Cellular identity, developmental reorganization, genomic structure modulation, and susceptibility to diseases are determined by epigenomic regulation by multiple signaling interplay. Here we demonstrate that elovanoids (ELVs), mediators derived from very-long-chain polyunsaturated fatty acids (VLC-PUFAs, n-3, C > 28), and their precursors in neurons in culture overcome the damage triggered by oligomeric amyloid-beta (OAβ), erastin (ferroptosis-dependent cell death), or other insults that target epigenomic signaling. We uncover that ELVs counteract damage targeting histones H3K9 and H3K27 methylation and acetylation; tau hyperphosphorylation (pThr181, pThr217, pThr231, and pSer202/pThr205 (AT8)); senescence gene programming (p16INK4a, p27KIP, p21CIP1, and p53); DNA methylation (DNAm) modifying enzymes: TET (DNA hydroxymethylase), DNA methyltransferase, DNA demethylase, and DNAm (5mC) phenotype. Moreover, ELVs revert OAβ-triggered telomere length (TL) attrition as well as upregulation of telomerase reverse transcriptase (TERT) expression fostering dendrite protection and neuronal survival. Thus, ELVs modulate epigenomic resiliency by pleiotropic interrelated signaling.


Introduction
Neurons, along with other brain cells, implement thought, memory, and behavior. However, they are often exposed to injury, disease, and various insults, such as uncompensated oxidative stress (UOS). Thus, it can be predicted that neurons utilize adaptive responses to preserve function and homeostasis. Histone modi cations, DNA methylation (DNAm), tau phosphorylation, senescence gene programming, and sustainment of telomere length (TL) are involved in healthy aging, unsuccessful aging, and pathologies, including age-related epigenomics associated with neurodegenerative diseases, including Alzheimer's disease (AD) 1,2 .
DNAm, an epigenomics event in AD onset and progression 3,4 , contributes to modulating synaptic plasticity and homeostasis 4 . There is a void in our understanding of how dysregulated responses can be controlled, including which mediators may be engaged-issues critically important to sustaining cognitive decline associated with age and AD.
In previous studies, we identi ed elovanoids (ELVs), low-abundance, high-potency pro-homeostatic mediators from the omega-3 fatty acid family 5,6 . ELVs are biosynthesized from precursors made by ELOVL4 (elongation of very long chain fatty acids-4), an enzyme selectively expressed in neurons and enriched in the hippocampus 7 . Mutations in the encoding gene of this enzyme are causative of neurological disorders, including mental retardation 7,8 . The ELOVL4 pathway products, very-long-chain polyunsaturated fatty acids (VLC-PUFAs) 32-carbon-n3 (32:6n-3) and 34-carbon-n3 (34:6n-3) yield biologically active stereospeci c di-hydroxylated ELV-N32 and ELV-N34, respectively 5,6 . ELV-mediated signaling elicits neuroprotection when cells are exposed to oligomeric amyloid-beta peptide (OAβ) 9 and other forms of damage, including in vivo experimental ischemic stroke 5 . ELVs also counteract OAβmediated cytotoxicity, senescence gene programs, and SASP expression in the retina 9 . Moreover, the abundance of sirtuin 1, recognized as a participant in epitranscriptomics 10,11 , is upregulated by ELVs in UOS conditions in neural cells 5 . These observations indicate the importance of ELVs in preserving neuronal integrity. Critically, it remains unclear whether and how the interplay of ELVs would sustain homeostasis, neuronal survival, and epigenomics.
In this article, we report the discovery that in neurons challenged with stressors, ELVs restore histone modi cations, DNAm, tau phosphorylation, telomere integrity, senescence programming, and dendrite integrity. It is of interest that dendrites (dendron; tree) are targeted since these branches from the neuronal soma conforming to a tree-like structure contain ribosomes, endoplasmic reticulum, Golgi apparatus, and active protein-synthesizing activity. Thus, dendrites modulate protein density in response to neuronal inputs and are active participants in synaptic transmission and memory formation by sustaining and sorting out numerous signals arriving from neurons and transferring them to proper circuits 12,13 .
Our ndings reveal a novel layer of regulation and the mediators involved that play a role in the pathogenic mechanisms of neurodegeneration. The stressors used are known to be engaged in acute or chronic cell damage that triggers responses that aim to endure, adapt, or restore homeostasis comprising hallmarks of resilience.
Similarly, HNG cells challenged with other stressors-UOS, OGD, or NMDA receptor-induced excitotoxicity -also show pronounced damage caused by the stressors (Supplementary Fig. 1a) (vehicle) in comparison to untreated (control), and 32:6 or 34:6 protected dendritic morphology. Supplementary  Fig. 1b-g depicts the quanti cation of cell survival by LDH and MTT assays.
ELVs or NPD1 counteract injury-induced tau hyperphosphorylation and neuronal damage Hyperphosphorylation of neuronal tau, a hallmark of neuronal tauopathies, occurs at key threonine, serine, or tyrosine residues. A key master site is pThr181, which triggers multi-site phosphorylation, thereby fostering assembles of neuro brillary tangles 14 . Moreover, plasma pThr181 is an early biomarker of MCI and AD 15,16 . Similarly, cerebrospinal uid pThr217 is a biomarker of AD development 17 . So, we investigated using western blotting key tau phosphorylation residues pThr181, pSer202/pThr205 (AT8), pThr217, pThr231, and total tau (HT7) in primary rat hippocampal neurons stressed with OAβ (10 µM) (Fig. 3a,b). Consistent with previous results, we see hyperphosphorylation of tau at all the residues induced by OAβ and remarkable downregulation of tau phosphorylation by either NPD1 or ELV-N34 Me at 250 nM. There were no changes in total tau HT7 across treatment groups: control, OAβ stressed, NPD1, or ELV. Supplementary Figs. 2-6 show western blots for all ve residues: pThr181, pSer202/pThr205 (AT8), pThr217, pThr231, and total tau (HT7). Also, we used confocal microscopy and unbiassed image analysis to investigate tau phosphorylation at residues pSer202/pThr205 (AT8), Y18 (pTyr18), and pThr231 ( Fig. 2c-o). OAβ-induced tau phosphorylation was determined by an increase in the mean signal intensity of those residues, and ELV-N34 Me (250 nM) reduced the mean signal intensity.
ELVs reduce OAβ-induced perturbations in DNAm (5-mC), DNA hydroxymethylation (5-hmC) TET activity, DNMT activity, demethylase activity, TL protection, and transcriptional regulation of TERT in HNG cells Figure 6a shows the experimental layout of the treatment of primary HNG in culture stressed with OAβ.  Fig. 6b, there was inhibition of TET activity in HNG cells when stressed by OAβ resulting in a decrease in DNA hydroxymethylation (5-hmC), which was counteracted by only ELV-N34 Me. Next, we investigated DNA methyltransferase (DNMT) and demethylase activity in HNG cells stressed with OAβ (Fig. 6c,d). There was marked DNA methyltransferase inhibition upon OAβ challenge, which was restored by ELVs. Similarly, the increase in demethylase activity was also counteracted by ELVs, but NPD1 did not alter the TET, DNMT, or demethylase activity. Next, we investigated how the canonical and noncanonical pro-survival functions of telomerase were impacted in HNG cells upon exposure to OAβ. Cells were stressed, lipid mediators were added (200 nM) and incubated for 24 hours, and then DNA was extracted to assess TL per diploid genome copy using qPCR. TL was measured using a quantitative realtime polymerase chain reaction (qPCR) by using an oligomer standard (TTAGGG) 14 . For the samples stressed with OAβ, there is a decrease in the length of telomeres compared to controls, which is restored by treatment with ELVs (200 nM) (Fig. 6h). Telomerase reverse transcriptase (TERT, the catalytic subunit of the telomerase holoenzyme, is the limiting step for telomerase activation and protection of TL. qPCR reveals the downregulation of TERT transcription by OAβ (10µM), which is reversed by ELVs (200 nM) Me and Me-A (Fig. 6i).

NPD1 or ELVs differentially target H3K27 and H3K9 methylation and acetylation upon challenging HNG cells with OAβ or erastin
In ammatory responses and other events associated with longevity involve post-transcriptional histone tail modi cations, such as acetylation and methylation of lysine residues, that enable chromatin modi cations to activate or repress the transcription of genes. After stressing HNG with OAβ or erastin +/-NPD1 or ELVs, histone modi cations were quanti ed using ELISA assays for methylation and acetylation at H3K9 and H3K27 residues, which speci cally are repressor sites for human TERT. Both OAβ or erastin induces hypermethylation of histone 3 at lysine residues 27 and 9 ( Fig. 7a,b,e,f), which are counteracted by ELVs or NPD1. However, for H3K27 methylation, NPD1 had no effect. Upon checking for H3K27 and H3K9 acetylation after treatment with OAβ or erastin, we found that OAβ induces hypoacetylation of histone H3 at lysine residues 27 and 9, whereas erastin induces hyperacetylation at H3K27 and H3K9 residues (Fig. 7c,d,g,h). ELVs reversed the changes in hypo-and hyper-acetylation at these residues.

Discussion
A transcriptomic imbalance during aging and certain pathologies lead to loss of cellular homeostasis and proteostasis, accumulation of protein aggregates, increased genomic instability, mitochondrial dysfunction, telomere attrition, cellular senescence, altered intercellular communication, deregulated nutrient sensing, and a decline in tissue functions 18,19 . These uncontrolled events are linked to epigenomic dysregulation 20,21 related to histone modi cations and DNAm, which regulates gene expression 22,23 and plays an important role in AD, which presents an accumulation of tau tangles and OAβ plaques 24-29 . Also, several tauopathies and α-synucleinopathies arise, leading to transmission of tau and α-synuclein, which is a mechanism for the progression of neurodegenerative disorders 30 . Interrelated pathways transduce epigenome modulatory signals that include DNAm that modi es cytosine base by DNA methyltransferase enzymes to form 5-methyl-cytosine (5mC). This, in turn, can be modi ed by the TET proteins (TET1, TET2, TET3) to produce the oxidation products 5hydroxymethylcytosine (5hmC), 5-formylcytosine (5fC), and 5-carboxylcytosine (5caC). The dysregulation of TET enzymes and 5hmC exhibit differential hydroxymethylation at genes associated with synaptic plasticity, neurogenesis, and neurodevelopment 31-33 , including genes linked to several susceptible AD loci compared to healthy controls 32,34 . Consistent with this, loss-of-function mutations in TET2 have been identi ed with EOAD patients 33 , and selective genome-wide reduction of 5hmC in neurons leads to hyperphosphorylation of tau and amyloid beta accumulation in the 3xTg-AD mouse model 35 . As demonstrated in Fig. 6, ELVs reduce OAβ-induced perturbations in DNAm (5-mC); in activities of DNA hydroxymethylation (5-hmC) TET, DNMT and demethylase, as well as transcriptional regulation of TERT (catalytic subunit of the telomerase holoenzyme) the limiting step for telomerase activation and protection of telomere length.
In this study, we used different cell injury triggers to explore whether ELVs regulate epigenomic signaling.
The stressors used, OAβ 42, a precursor to AD plaques that damages neurons by itself 36 , and erastin to set ferroptosis in motion, limiting the uptake of cystine, consuming GSH, and inhibiting the cystine/glutamate antiporter system (System Xc − ) and glutathione peroxidase 4 (GPX4) 37 . Ferroptosis is implicated in the amyloidogenic build-up of OAβ in AD due to lipid peroxidation, abnormal iron dynamics, and accumulation in amyloid precursor protein 38-40 . Moreover, UOS, OGD, or NMDA induce cellular excitotoxicity which are also attenuated by ELVs.
In summary, our results show protection of dendritic integrity and cell survival of HNG cells by precursors of ELVs (32:6 and 34:6) or ELV-N32 or ELV-N34 upon being challenged by either OAβ, erastin or other stressors like OGD, NMDA or UOS by precursors 32:6 and 34:6. The results using confocal microscopy, followed by unbiassed image analysis and western blotting demonstrated that ELVs reverse hyperphosphorylation of tau at several key residues important biomarkers of AD-pThr181, pSer202/pThr205 (AT8), pThr217, pThr231-while total tau (HT7) does not change with any of the treatments. Our results on tau hyperphosphorylation-pThr181 41-45 , pThr217 41,46-48 , pThr231 41 , and pSer202/pThr205 (AT8) 41 -are consistent with previous ndings that there is hyperphosphorylation of tau at these residues as the cells are challenged by OAβ. Our results show a reversal of hyperphosphorylation by treating the cells with ELVs. This is a novel and critical nding as all these tau residues are correlated with Braak stages II/II and III/IV 41 and are able to reverse these changes with ELVs. Previous research from our lab has shown that in an Alzheimer's disease mouse model, Moreover, we show that ELVs counteract SASP 51,52 and senescence gene programming 53-55 induced by OAβ or erastin: p16INK4a 54 , p27KIP, p21CIP1, and p53 55 . ELVs also modulate DNAm modifying enzymes: TET (DNA hydroxymethylase), DNA methyltransferase, DNA demethylase, and DNAm (5mC) phenotype. Histone modi cations H3K9 and H3K27 methylation and acetylation play an important role in epigenomic signaling. Hence, we studied these histone modi cations in HNG cells stressed with OAβ or erastin and found reversal by ELVs. Upon assessment of TL attrition induced by OAβ, there is protection by ELVs, along with upregulation of TERT expression by treatment with ELVs, which was reduced when cells were stressed with OAβ.
Our study, for the rst time, identi es ELVs modulating epigenomic signaling by pleiotropic interrelated mechanisms modulating tau hyperphosphorylation after DNAm, histone modi cations, and telomerase activity. We demonstrate that ELVs also counteract senescence gene programming and SASP, suggesting that ELVs can restore epigenomic landscape perturbations that consolidate intrinsic and environmental cues for genome expression and organ functions. With single cell and special transcriptomics/epigenomics 56-61 , we are investigating DNAm 62 and histone modi cations (H3K27me3, H3K27ac or H3K4me3) 63 , to de ne the interplay of OAβ and tau at synapses 64 and discern how ELVs may act in resetting epigenomic clock to provide resiliency and conserve the epigenomic signals 65 . Epigenomic perturbations are associated with in ammatory responses, cancer, metabolic disorders (e.g., diabetes, obesity), cardiovascular disorders, aging, and age-related neurodegenerative diseases. The identi cation of epigenomic regulators would facilitate the understanding and unraveling of functions and dysfunctions of the molecular mechanisms involved. Epigenomic tunings protect or hasten declined cell functions during aging and in age-related disorders. Tau aggregation depends on several post-translational modi cations including phosphorylation 41 , participant also of epigenomic modi cations.
However, the identi cation of the ELV targeted receptors involved is needed.
Overall, we provide here a characterization of critical targets of ELVs that open insight into regulatory epigenomic landscapes in neurons. We also offer evidence that ELVs target dendrites and neuronal survival. Speci cally, we found strong pleiotropic links between histone modi cations, DNAm, and other events that are part of epigenomic mechanisms that enable neurons to adjust their expression patterns temporarily or permanently. This study opens opportunities, including developing experimental and clinical interventions that speci cally assess the potential for ELVs to enhance resilience for neurodegenerative diseases and/or stroke.

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

Data Availability
All relevant data are included in the paper. This study did not generate any datasets that were deposited in external repositories. All raw data that support the ndings, tools, and reagents will be shared on an unrestricted basis. Concerning this, reasonable requests should be directed to the corresponding author.

CODE Availability
No custom code was used in this study.

Statistics
Data are expressed as mean ± SEM of three or more independent experiments. The data were analyzed by one-way ANOVA followed by Holm-Sidak's multiple comparisons post hoc test at a 95% con dence level to compare the different groups and considered signi cant with a P < 0.05. Statistical analyses were performed using GraphPad Prism software (9.5.1, San Diego, CA, United States). Figure 1 Dendrites protection by precursors of ELVs (32:6 and 34:6). a, Representative bright eld images (10x mag) of primary human neuronal-glial (HNG) cells in culture challenged with oligomeric amyloid-beta (OAβ) (10 µM) or erastin (10 µM), +/-ELV precursors 32:6 or 34:6 (500 nM). b-e, Quanti cation of neuroprotection by LDH or MTT assay. Data are mean ± SEM of n = 8. P values were determined by oneway analysis of variance (ANOVA) at 95% con dence level, followed by Holms Sidak's post hoc test. ****P ≤ 0.0001. f, Mature neurons stained with β-III tubulin showed dendritic damage upon being stressed with either OAβ or erastin +/-32:6 or 34:6. g,h, Quanti cation of surface area of dendrites with unbiassed image analyses of confocal images of mature neurons stained with β-III tubulin (green). Data are mean ± SEM of n = 6. Pvalues determined by one-way analysis of variance (ANOVA) at 95% con dence level, followed by Holms Sidak's post hoc test. ****P ≤ 0.0001. Damage quanti cation of dendritic surface area induced by either OAβ or erastin (10 µM) and the protection with ELV-N32 and ELV-N34 as determined by unbiassed image analysis with confocal microscopy and Imaris microscopy image analysis software. s, Dendritic Protection of HNG stressed with OAβ (10 µM) plus 32:6 and 34:6 (500 nM) each. t,Quanti cation of neuroprotection, as determined by change in dendritic volume between different treatment conditions. Data are mean ± SD. P values were determined by one-way ANOVA, followed by Holms Sidak's post hoc test. ****P≤ 0.0001.

Figure 3
ELVs or NPD1 restore OAβ -induced tau hyperphosphorylation at several phosphorylation sites. a, Western blots displaying changes in phosphorylated tau at different phosphorylation residues normalized to GAPDH. There are no changes in total tau (HT7) abundance. b, Quanti cation of phosphorylated tau abundance as determined by densitometry analysis showing downregulation of phosphorylation by NPD1 or ELV-N34 at all phosphorylation sites. c-e, Representative bright eld images of primary rat hippocampal neurons stressed with OAβ and treated with ELV-N34 (250 nM). f-n,Representative confocal micrographs of neuronal cultures stressed with OAβ, treated with ELV-N34 (250 nM) and stained with antibodies for β-III tubulin (red) and phosphorylated tau residues pSer202/pThr205 (AT8), Y18 (pTyr18), or pThr231 (green). o, Quanti cation of change in phosphorylation between unstressed (control), stressed (OAβ), or treated (OAβ stressed + ELV-N34) cells as determined by unbiassed Imaris analysis of green signal intensity. Data are mean ± SEM. P values were determined by one-way analysis of variance (ANOVA), followed by Holm Sidak's post hoc test. ***P ≤ 0.001, ****P≤ 0.0001.

Figure 4
OAβ or erastin-mediated activation of SASP in HNG cells in primary culture counteracted by ELVs. a, Experimental design of treatment of HNG co-cultures with lipid mediators. Human neuro-progenitor cells were differentiated into neuronal-glial co-cultures, grown 21 days for maturity, and then stressed with either OAβ or erastin (10 µM). 30 minutes later, lipid mediators were added (200 nM) and incubated for 48 hours, after which cells were xed and stained for senescence-associated β-galactosidase (SA-β-Gal) activity. b-h, Representative images of HNG cells treated with OAβ (10µM) ± ELVs. j-p, HNG cells treated with erastin (10µM) ± ELVs. i,q, Bar graphs of quanti cation of the % of β-galactosidase positive cells (senescence-associated secretory phenotype; SASP) and the degree of protection of HNG cells by ELVs treated with either OAβ or erastin, respectively. ELVs decreased positive senescent cells. Data are mean ± SD of n = 3. P values were determined by one-way ANOVA, followed by Holm Sidak's post hoc test. ***P ≤ 0.001, ****P ≤ 0.0001.

Figure 6
ELVs counteract Oaβ-induced DNA methylation (5-mC), global DNA hydroxymethylation (5-hmC) TET inhibition, DNMT activity, demethylase activity, TL protection, and restoration of TERT expression. We have used primary cultures of HNG cells (a,i). a,Experimental design of treatment of HNG co-cultures with lipid mediators. Human neuro-progenitor cells were differentiated into neuronal-glial co-cultures, grown 21