Urolithin A augments angiogenic pathways in skeletal muscle by bolstering NAD+ and SIRT1

Urolithin A (UA) is a natural compound that is known to improve muscle function. In this work we sought to evaluate the effect of UA on muscle angiogenesis and identify the underlying molecular mechanisms. C57BL/6 mice were administered with UA (10 mg/body weight) for 12–16 weeks. ATP levels and NAD+ levels were measured using in vivo 31P NMR and HPLC, respectively. UA significantly increased ATP and NAD+ levels in mice skeletal muscle. Unbiased transcriptomics analysis followed by Ingenuity Pathway Analysis (IPA) revealed upregulation of angiogenic pathways upon UA supplementation in murine muscle. The expression of the differentially regulated genes were validated using quantitative real-time polymerase chain reaction (qRT-PCR) and immunohistochemistry (IHC). Angiogenic markers such as VEGFA and CDH5 which were blunted in skeletal muscles of 28 week old mice were found to be upregulated upon UA supplementation. Such augmentation of skeletal muscle vascularization was found to be bolstered via Silent information regulator 1 (SIRT1) and peroxisome proliferator-activated receptor-gamma coactivator-1-alpha (PGC-1α) pathway. Inhibition of SIRT1 by selisistat EX527 blunted UA-induced angiogenic markers in C2C12 cells. Thus this work provides maiden evidence demonstrating that UA supplementation bolsters skeletal muscle ATP and NAD+ levels causing upregulated angiogenic pathways via a SIRT1-PGC-1α pathway.

www.nature.com/scientificreports/ supporting skeletal muscle health and function is compelling 2 . Both geriatric and sedentary populations in the United States heavily rely on nutritional supplements for maintenance of health and fitness 3,4 . Maintenance of and development of skeletal muscle health is of specific interest to the aging population who face the threat of sarcopenia 5 . Age-related decline in skeletal muscle vascular health is of significance concern in this regard 6 . Nutritional supplements are known to improve muscle macro and microcirculation 7 . Ellagitannins (ETs) are natural products that are known to preserve muscle health 8,9 . Upon hydrolysis, ETs release ellagic acid (EA) which undergoes metabolism by the gut microflora into urolithins 10 . Urolithins (also known as Dibenzo-α-pyrones or DBPs) are natural metabolites obtained from the transformation of ellagitannins (ETs) by the gut bacteria 11 . In addition, urolithins are abundant in Shilajit, a herbomineral used in traditional Ayurvedic medicine 19 . Urolithin A (UA), urolithin B (UB), urolithin C (UC) and urolithin D (UD) are the metabolites of ETs and EA that are found in humans 10,12 . UA possess antioxidant, anti-inflammatory and anti-proliferative properties 13,14 . Urolithins improve muscle function 15 . In this work we sought to understand the mechanism of action of orally supplemented UA on limb skeletal muscles.

Results
Delayed onset of muscular aging in response to UA supplementation in 28-week old mice. To evaluate the effect of UA on onset of skeletal muscle aging in vivo, C57BL/6 mice (12 weeks old) were orally supplemented with UA (10 mg/kg) for 16 weeks. UA supplementation was safely tolerated (Supplementary Table 1). Skeletal muscles (vastus lateralis and gastrocnemius) were collected from the adult mice (28 weeks old; equivalent to ~ 35-40 years old human) (Fig. S1). At that age, murine skeletal muscle tissues (vastus lateralis and gastrocnemius) supplemented with placebo showed significantly elevated aging markers P16, 8-hydroxy-2′-deoxyguanosine  and Ataxia-Telangiectasia Mutated (ATM) compared to skeletal muscle tissues of young C57BL/6 mice (8-weeks old). In mice supplemented with UA, induction of these age-related markers were significantly blunted (Fig. 1).
UA supplementation increased ATP and NAD + levels in murine skeletal muscle. Adenine nucleotide pools in either the cytosolic or mitochondrial compartment may serve as indicator of the energy status of the cell in terms of phosphate potential. Because UA supplementation blunted induction of age-related markers, we sought to investigate the skeletal muscle ATP levels. In vivo 31 P NMR spectroscopy revealed elevated total ATP in the skeletal muscle in response to UA supplementation ( Fig. 2A-C). Specifically, α-ATP and γ-ATP levels were increased (Fig. S2A). γ-ATP is the primary phosphate group on the ATP molecules that has a higher energy of hydrolysis than either α or β phosphate. Tissue NAD + depletion is also a hallmark of aging 16 . The relationship between ATP and NAD + is linear 17 . Because UA supplementation resulted in delayed induction of markers of muscular aging and bolstered ATP levels, the NAD + and NADH levels of the UA supplemented mice skeletal muscle were analyzed. UA supplementation increased skeletal muscle NAD + levels (Fig. 2D,G) and NAD + / NADH ratio (Fig. 2F,I) significantly in 28-week old mice. Such increased levels of NAD + upon UA supplementation was comparable to the effect obtained by supplementing nicotinamide riboside, precursor of NAD + , at a five-fold higher dose ( Fig. S2B-D). The ability of UA to elevate NAD + levels and augment NAD + /NADH ratio was reproduced in C2C12 murine skeletal muscle cells (Fig. 2J,L). UA did not influence NADH levels both in the muscle of supplemented mice as well as in C2C12 cells (Fig. 2E,H,K).
Transcriptome profiling of murine vastus muscle post 12 week of oral UA supplementation. To look into the molecular mechanisms of action of UA, unbiased transcriptome profiling of murine vastus lateralis tissue was performed on UA supplemented mice. GeneChip data analyses was performed using Affymetrix. GeneChip.Mouse430_2 following RNA extraction and target labeling to determine the alterations in the transcriptome of vastus muscle in response to oral UA supplementation. A total of ~ 2200 annotated probe sets were differentially (p < 0.05) regulated following 12 week supplementation as compared to placebo (Fig. 3A,B). The expression data have been submitted to Gene Expression Omnibus (GEO) at NCBI (GSE136552).

Discussion
Skeletal muscle vascularization is a key determinant of its function 22 . Structural and functional decline of the skeletal muscle occurs with aging 23,24 . Aging compromises skeletal muscle circulation as well as blunts the angiogenic properties of this tissue 25,26 . Impaired microvascular functions perturb myogenic cell homeostasis, limit nitric oxide (NO) activity, increase production of vasoconstrictor factors, and is associated with inflammationrelated oxidative stress 27,28 . In humans, age-related impairments in skeletal muscle start in the fourth decade of life 23,29 . In this work, 12-week old mice were orally gavaged with UA for 12-16 weeks. After 6 months of age, the maturational rate of mice is 25-fold increased than humans 30 . Seven month old mice, as utilized in this work, represents human age of 34-42 years. Consistent with this assessment, tissue markers of aging were elevated in the skeletal muscle of 28 week-old mice that received the placebo. The observation that UA supplementation significantly blunted such markers of aging point towards a beneficial effect of the natural supplement on skeletal muscles. This is consistent with prior reports demonstrating that UA increases the life-span of C.elegans 15 . In aging rodents, UA is known to benefit muscle function by increasing mitophagy 15 . NAD + plays a central role in supporting skeletal muscle development and health 31 . Aging depletes skeletal muscle reserves of NAD +16 . Depleted muscle NAD + is a major threat to muscle health 31 . Strategies to boost NAD + under such conditions www.nature.com/scientificreports/ have produced encouraging results 31 . This work presents first evidence demonstrating that long-term oral supplementation of UA is successful in bolstering skeletal muscle NAD + of sedentary middle-aged mice. Elevated tissue NAD + is known to be associated with higher ATP levels. Cytosolic NAD + participates in the anaerobic/ glycolytic metabolism of glucose into ATP. Additionally, mitochondrial ATP production and membrane potential requires NAD + which gains two electrons and a proton from substrates at multiple tricarboxylic acid cycle steps and gets reduced to NADH 32 . Consistent with this observation, we also observed that UA supplementation improved skeletal muscle ATP content as measured from live animals. The observations that urolithin enriched natural product (UENP) improves ATP production is in agreement with our finding [33][34][35] . Transcriptome-based approach to understand molecular mechanism offers the power of unbiased interrogation such that structured data mining may underscore the primary pathways affected. Screening of over 45,000 probes in the skeletal muscle identified induction of 2.3% of all transcripts. Over three-fourth of these candidate transcripts were annotated and were therefore utilized for data mining. Employing standardized IPA analysis these candidates were mined for pathway analysis. Such unbiased data mining identified two major pathways that were induced by UA, endothelial development and proliferation of endothelial cells. Collectively, these are present as angiogenic pathways of the skeletal muscle. Upregulation of these angiogenic pathways in the vastus lateralis muscle, as tested using GeneChip, proved to be also true for gastrocnemius muscle of UA supplemented mice.
Expression of VEGF, a known angiogenic factor, is reported to be downregulated in aged animals 36 causing impaired VEGF-induced angiogenesis in the ischemic limb of old mice 37,38 . UA supplementation upregulated VEGFA and VEGFR2 expression in murine skeletal muscle. UA supplementation also upregulated Tenascin-C (Tnc), Pecam1, CD105 and vWF expression. TNC is known to promote angiogenesis 39 and PECAM1, CD105 and vWF are known markers of vascular endothelium 40 .
SIRT1 is (NAD + )-dependent histone deacetylase that supports angiogenic signaling 41 . Pharmacological activation of SIRT1 significantly improved endothelial function in aged mice 42 . Nicotinamide mononucleotide (NMN), a precursor of NAD + , activated SIRT1 and improved endothelial function in the aged vasculature 43 . Observation in this work that UA supplementation turns on angiogenic signaling pathways in skeletal muscle was backed by the finding that under the same conditions NAD + levels and SIRT1 in the muscle were also elevated. In support of the contention that SIRT1 is directly implicated in angiogenic signaling, it is reported that SIRT1 inhibitor EX527 blunted angiogenic pathways 44 . Such angiogenic markers, VEGFA and VEGFR2 have been implicated in skeletal muscle angiogenesis 45,46 . Though VEGFA binds toVEGFR1 with a higher affinity than VEGFR2, the latter is considered to be the main mediator of angiogenesis, since the kinase activity of VEGFR1 is weak 47 . We observed that UA induced angiogenic markers in murine skeletal muscle via a SIRT1-PGC-1α pathway. SIRT1 deacetylates and activates the PPAR gamma coactivator 1 (PGC-1), a transcription co-regulator of PPARs 48 . In skeletal myofibers, PGC-1α is strongly induced by exercise and β-adrenergic signaling 49,50 . Induced PGC-1α potently stimulates mitochondrial biogenesis and the release of angiogenic factors 51,52 . In elderly mice, the NAD + precursor NMN is also known to improve blood flow and increase endurance via a SIRT1-PGC-1α pathway 53 . Urolithins modulate aryl hydrocarbon receptor (AHR) activity 54 . AHR transcriptionally regulates inflammatory mediators, including cytokines IL-6, IL-1β, chemokines CXCL5, CCL20, and prostaglandin-endoperoxide synthase PTGS2 [55][56][57][58] . Multiple pathways contribute to the anti-inflammatory activities of urolithins, including, c-Jun 59 , NF-κB/AP1 60 and MAPK 61 . While transcription factor NF-κB has been implicated in aging 62 , the role of AP-1 in angiogenesis is well-documented 63 . Thus, it is plausible that in addition to SIRT1, such targets of Urolithin may be responsible for the large scale change in gene expression.
GeneChip probe array analyses and IPA. GeneChip analysis was done using Affymetrix Clariom D Assay on vastus lateralis of UA-supplemented (for 12 weeks) animals as described previously 19,67,74,75 to identify sets of genes differentially expressed in the skeletal muscle upon UA supplementation. Briefly, total RNA was isolated using the miRVana Isolation Kit as per the manufacturer's protocol (Thermo Fisher Scientific, MA). RNA integrity was assessed using the Agilent 2100 Bioanalyzer (Agilent, CA). The isolated RNA was used to generate ss-cDNA using the GeneChip WT PLUS reagent kit. Biotin-labeled ss-cDNA was hybridized, washed and stained on the Affymetrix Fluidics Station 450 according to the manufacturer's protocol and scanned with the Affymetrix GeneChip Scanner 3000 7G (Affymetrix, CA) 19,67,74,75 . GCOS (Gene Chip Operating Software, Affymetrix) was used for acquisition of data and processing of image. Genespring GX (Agilent, CA) was employed to analyze raw data. Additional data processing was performed using dChip software (Harvard University) 19,66,67,74,75 . Differentially expressed genes were identified using a two-class t-test where significance level was set at p < 0.05 with Benjamin-Hochberg correction for false discovery rate 19,66,67,74,75 . Data were analyzed through the use of IPA (QIAGEN Inc., https ://www.qiage nbioi nform atics .com/produ cts/ingen uityp athwa y-analy sis).
Validation of microarray results using quantitative real-time PCR. Total RNA extraction from murine vastus lateralis and gastrocnemius muscle was performed with mirVana RNA Isolation Kit Thermo Fisher Scientific, MA) according to the manufacturer's instructions. For gene expression, total cDNA was achieved using the SuperScript III First Strand Synthesis System or Vilo (Life Technologies, Carlsbad, CA) 18,76 . Candidate genes were verified by RT-PCR using SYBR green-I and primers (Supplementary Table 4) as previously described using GAPDH as housekeeping gene.
NAD + /NADH measurement in murine muscle tissue (HPLC). NAD + and NADH were extracted from the mice vastus and gastrocnemius muscle tissue using perchloroacetic acid and 0.5 M KOH respectively as previously reported 77 . Tissue extracts were separated using high performance liquid chromatography (HPLC, Dionex Ultimate 3000, RS Diode Array Detector, Thermo Scientific, MA) equipped with a Column RP-18 Endcapped (LiChrospher 100, 5 μm, Millipore Sigma, MA). The following gradient over 10 min was carried out at a flow rate of 1 ml/min with a mixture of mobile phase A (0.1 M sodium phosphate buffer, pH 6.50) and mobile