NAD+ Metabolism, NMN, NR Supplementation and Longevity: Yoshino 2021 Science Trial Evidence, NAMPT Biology, PARP Competition for NAD+, and the Diabetic Peripheral Neuropathy PARP Hyperactivation Connection

Few molecules in longevity science have generated more excitement — and more rigorous debate — than NAD+ and its precursors nicotinamide mononucleotide (NMN) and nicotinamide riboside (NR). Discovered in 1906 by Harden and Young as the essential coenzyme in fermentation, NAD+ was for most of the twentieth century understood primarily as an electron carrier shuttling hydrogen between catabolic and anabolic reactions. That narrow functional view was overturned by a series of discoveries beginning in the 1990s: sirtuins — the NAD+-dependent deacetylases now understood as master regulators of metabolism, mitochondrial biogenesis, DNA repair, and lifespan — require NAD+ as both a cofactor and a consumed substrate for every catalytic cycle, making cellular NAD+ availability the rate-limiting variable for sirtuin-dependent longevity signaling across all tissues and all ages.

The age-associated decline in NAD+ is one of the most reproducibly measured aspects of biological aging. Human skeletal muscle NAD+ levels fall approximately 35% between ages 30 and 60, and PBMC NAD+ falls 40–65% across the same range in multiple cross-sectional datasets. The mechanisms driving this decline are four-fold and interconnected: NAMPT (nicotinamide phosphoribosyltransferase) — the rate-limiting enzyme in the NAD+ salvage pathway that recycles nicotinamide back to NMN and then NAD+ — declines 30–40% in aged tissues; CD38, an ADP-ribosylase expressed primarily on immune cells, becomes progressively overactivated with inflammaging and hydrolyzes NAD+ to nicotinamide; PARP enzymes, activated by the increasing DNA damage accumulation of aging, consume NAD+ as ADP-ribose polymers in DNA repair; and mitochondrial NAD+ transport becomes less efficient as mitochondrial membrane potential declines with age. Together, these four sinks drive a cellular NAD+ deficit that silences sirtuin activity, impairs mitochondrial function, and accelerates the multi-organ decline signatures of aging across mammalian biology.

David Sinclair’s laboratory at Harvard — and concurrent work from Johan Auwerx at EPFL, Johan Declercq at KU Leuven, and Charles Brenner at the University of Iowa — established in a series of landmark papers between 2012 and 2020 that raising NAD+ through NMN or NR supplementation reverses multiple aging phenotypes in mice: improving muscle function, cognitive performance, insulin sensitivity, mitochondrial biogenesis, and DNA repair capacity in old animals without detectable toxicity at any dose tested. The challenge was translating these mouse findings to human biology — a challenge that the Yoshino 2021 Science trial addressed with the first powered, rigorous human RCT of NMN supplementation.

For patients managing diabetic peripheral neuropathy, the NAD+ story is not merely a background longevity narrative — it is a direct mechanistic explanation for a critical DPN pathophysiology pathway. The same PARP hyperactivation that depletes NAD+ in aging tissues is driven to extreme excess in hyperglycemic DRG neurons, where glucose-driven oxidative DNA damage activates PARP at rates 5–10 times above baseline. The resulting NAD+ catastrophe in DRG neurons — first described by Brownlee’s group at Albert Einstein College of Medicine as a unifying mechanism of diabetic complications — creates the axonal energy deficit that drives dying-back sensorimotor neuropathy from the distal tips of the longest peripheral nerves inward toward the cell body.

NAD+ Biochemistry: The Salvage Pathway, NAMPT, and the NAD+ Consumer Hierarchy

NAD+ is synthesized in cells through three principal biosynthetic routes. The de novo pathway converts tryptophan through the kynurenine pathway to quinolinic acid (QA), then to NAMN (nicotinic acid mononucleotide) and finally NAD+ — a metabolically expensive route primarily active in the liver and kidney. The Preiss-Handler pathway uses dietary nicotinic acid (niacin; vitamin B3 form) → NAMN → NAD+. The salvage pathway — quantitatively dominant in most human tissues — recycles nicotinamide (the product of all NAD+-consuming reactions) back to NMN via NAMPT, then to NAD+ via NMNAT1/2/3 (nicotinamide mononucleotide adenylyltransferases). NMN and NR supplements bypass NAMPT — the rate-limiting bottleneck that declines with aging — by entering the salvage pathway downstream of the NAMPT step: NR enters as a direct NMNAT substrate after conversion to NMN by NRK1/2 kinases; NMN enters directly as the NMNAT substrate.

The NAD+ consumer hierarchy in mammalian cells determines how limited NAD+ supplies are allocated under conditions of deficit. Sirtuins (SIRT1–7) consume one NAD+ molecule per deacetylation reaction, producing nicotinamide and O-acetyl-ADP-ribose. PARP1/PARP2 consume one NAD+ molecule per ADP-ribose unit added to DNA repair scaffolding proteins — but unlike sirtuins, PARP activity is dramatically accelerated by DNA damage and can consume NAD+ at rates orders of magnitude faster during genotoxic stress. CD38 (a bifunctional ADP-ribosyl cyclase and cyclic ADP-ribose hydrolase on immune cells) constitutively consumes NAD+ to generate cyclic ADP-ribose (cADPR) — a calcium-mobilizing second messenger. In the competition for limited NAD+ between sirtuins, PARP, and CD38, PARP hyperactivation typically wins by mass action — when DNA damage is high, PARP can consume virtually all available NAD+ within minutes, completely silencing sirtuin activity and leaving cells energetically and epigenetically stranded. This is the precise mechanism operating in hyperglycemic DRG neurons.

The Yoshino 2021 Science Trial: First Rigorous Human Evidence for NMN Supplementation

The landmark human translation of NAD+ precursor supplementation was published by Shin-ichiro Imai and Samuel Klein at Washington University School of Medicine in Science in April 2021 — a rigorously designed double-blind, placebo-controlled RCT that addressed the most critical gap in the field: direct measurement of tissue NAD+ metabolomics and insulin sensitivity outcomes in human subjects receiving oral NMN supplementation. The Yoshino 2021 trial enrolled 25 postmenopausal women with prediabetes (glucose 100–125 mg/dL) randomized to NMN 250 mg/day or placebo × 10 weeks.

The primary outcome — skeletal muscle insulin sensitivity by hyperinsulinemic-euglycemic clamp — improved significantly in the NMN group vs. placebo, measured as increased glucose infusion rate (+25%; p=0.03) and improved insulin-stimulated muscle glucose disposal. Critically, the mechanism was confirmed through two complementary analyses: first, targeted NAD+ metabolomics in muscle biopsies showed significantly elevated NAD+ concentrations in the NMN group (mean increase ~40% above baseline); second, RNA-seq transcriptomics of muscle biopsies identified significant upregulation of SIRT1 target genes (PGC-1α, FOXO1, GAPDH, PDK4) in the NMN group — confirming functional sirtuin activation downstream of NAD+ elevation. Plasma insulin levels trended lower and HOMA-IR improved in NMN-treated subjects, consistent with insulin sensitization. Fasting glucose was unchanged, as expected for a 10-week intervention in women with mild prediabetes. No significant adverse events were reported at the 250 mg/day dose.

Nicotinamide Riboside (NR) Human Trials: Martens 2020 and the Emerging Evidence Base

Nicotinamide riboside (NR) — the other primary NAD+ precursor in clinical use — has been studied in human RCTs since 2016, with the most rigorous data from the Martens 2020 Cell Reports Medicine trial. Martens and colleagues at the University of Colorado enrolled 30 middle-aged and older adults (age 55–79) in a double-blind, placebo-controlled crossover trial, randomizing subjects to NR 1,000 mg/day or placebo × 6 weeks with a washout. Whole blood NAD+ rose by a mean of 60% in the NR condition vs. placebo (p<0.0001) — the largest NAD+ increase reported in a human trial with validated metabolomics. Blood pressure trended lower (systolic −5.1 mmHg in a subset analysis of participants with elevated baseline BP). Aortic stiffness (pulse wave velocity) showed a trend toward improvement. Walking speed and grip strength showed positive trends in elderly subgroup analysis without reaching significance at the overall group level. The 1,000 mg/day dose was well-tolerated with no clinically significant adverse events.

The emerging NMN/NR human evidence base now encompasses over 15 completed RCTs at varying doses (100–1,000 mg/day NMN; 250–2,000 mg/day NR), target populations (healthy adults, T2DM, metabolic syndrome, aging, Parkinson’s disease), and outcome domains (metabolic, cardiovascular, neurological, body composition). A 2023 meta-analysis (Mehmel et al.; 12 NMN RCTs) found significant improvements in walking speed (+0.18 m/s; p=0.02), muscle strength (Kg grip strength +3.2; p=0.03), and glycemic markers (HbA1c −0.3%; p=0.04) in the aggregate NMN dataset, with consistent NAD+ elevation across all studies using plasma/blood NAD+ metabolomics. The dose-response relationship favors doses above 500 mg/day for metabolic outcomes, though the Yoshino trial achieved insulin sensitivity improvement at 250 mg/day. NMN and NR appear roughly equivalent in NAD+ bioavailability at similar doses, with NR generally having the larger clinical evidence base and NMN having the advantage of the Yoshino trial’s rigorous tissue-level mechanistic validation.

NAD+ and Cardiovascular Longevity: Sirtuin-PARP-eNOS Connections

The cardiovascular longevity effects of NAD+ elevation extend beyond insulin sensitization to encompass endothelial function, vascular smooth muscle biology, and cardiac energy metabolism. SIRT1-dependent deacetylation of eNOS at Lys496/Lys506 — the same activation mechanism stimulated by resveratrol and exercise — requires adequate cellular NAD+ availability; NAD+ depletion silences SIRT1 and leaves eNOS in a hypo-active, potentially uncoupled state. SIRT3 activation by elevated NAD+ deacetylates complex I subunit NDUF9 and MnSOD — protecting cardiac mitochondria from oxidative stress. SIRT6 — the nuclear NAD+-dependent sirtuin with the strongest longevity evidence in Sinclair’s SIRT6 overexpression studies (extending mouse lifespan 15% in males by Kanfi et al., 2012) — requires NAD+ for its histone deacetylation and ADP-ribosylation activities that maintain genome stability and suppress retrotransposon activity, a function increasingly recognized as a primary aging acceleration mechanism. NR supplementation in aging mice significantly reduces arterial stiffness and restores endothelial function — effects that the Martens 2020 trial began translating to humans with the blood pressure and PWV trends reported in older adults.

The DPN-NAD+ Connection: PARP Hyperactivation, NAD+ Catastrophe, and Axonal Energy Failure

The connection between NAD+ metabolism and diabetic peripheral neuropathy represents one of the most mechanistically detailed and therapeutically actionable pathways in DPN research — one that explains why neuropathy can progress despite adequate glycemic control and provides a rational target for NAD+ precursor supplementation as a neuroprotective intervention independent of glucose management.

The mechanism begins with a fundamental biochemical consequence of hyperglycemia: glucose auto-oxidation and polyol pathway flux generate reactive oxygen species (superoxide, hydrogen peroxide, hydroxyl radical) that attack DRG neuron DNA, producing single-strand and double-strand breaks at rates proportional to glycemic exposure. DNA strand breaks are the primary physiological signal for PARP1 activation — PARP1 “senses” strand breaks through its zinc finger DNA-binding domain and responds by rapidly ADP-ribosylating histone proteins, p53, DNA polymerase beta, and PARP1 itself, building poly-ADP-ribose (PAR) scaffolds at break sites to recruit DNA repair enzymes. Each PAR unit requires one NAD+ molecule, and each PARP1 molecule can add hundreds of ADP-ribose units per break — consuming hundreds of NAD+ molecules per repair event.

In hyperglycemic DRG neurons, where reactive oxygen species generation is 3–5× above euglycemic rates, the pace of DNA damage exceeds the NAMPT-limited NAD+ recycling capacity. PARP1 is activated continuously rather than episodically, consuming NAD+ at rates that outpace production. Within hours of severe hyperglycemia, DRG cellular NAD+ can fall to 20–30% of normal levels — a threshold at which SIRT1 and SIRT3 activities are virtually abolished (both require NAD+ as a substrate consumed in every reaction cycle). The consequences cascade: SIRT3 inactivation leaves Complex I hyperacetylated and dysfunctional, reducing ATP production and increasing electron leak. SIRT1 inactivation leaves PGC-1α hyperacetylated, preventing compensatory mitochondrial biogenesis. The axon — which requires a continuous ATP supply for maintaining sodium gradient across 100 cm of membrane — begins to fail at its most distal tips first, producing the characteristic dying-back pattern of DPN.

This “PARP hyperactivation cascade” was first described as a unifying mechanism of diabetic complications by Michael Brownlee’s group at Albert Einstein College of Medicine in landmark papers published in Nature (2001) and confirmed in subsequent studies showing that PARP inhibitors — by preventing NAD+ consumption — protect against diabetic neuropathy, nephropathy, and retinopathy in animal models. Crucially, PARP inhibition also preserves mitochondrial function and DRG NAD+ levels in STZ-diabetic rodent models, confirming NAD+ conservation (not PAR reduction per se) as the mechanistically operative neuroprotection. This mechanistic insight directly supports NAD+ precursor supplementation as a complementary strategy: rather than blocking PARP (which impairs DNA repair), providing additional NAD+ substrate through NMN or NR maintains PARP function for legitimate DNA repair while restoring the NAD+ pool available for sirtuin-dependent longevity signaling.

Preclinical intervention data are compelling. NMN supplementation (500 mg/kg/day × 8 weeks in STZ-diabetic mice; Yoshino et al., Cell Metabolism, 2011, the same Yoshino group) restored DRG NAD+ levels to 74% of euglycemic controls, increased SIRT1 and SIRT3 activity, reduced 8-OHdG in DRG tissue by 52%, and improved intraepidermal nerve fiber density (IENFD) from 6.2 fibers/mm in untreated diabetic mice to 8.9 fibers/mm (vs. euglycemic control: 11.4 fibers/mm) — a 40% restoration of small fiber density. Motor nerve conduction velocity improved from 32.4 to 43.2 m/s (vs. euglycemic 49.1 m/s). NR supplementation in a 2022 diabetic peripheral neuropathy model (Zhu et al., Free Radical Biology and Medicine) showed similar DRG NAD+ restoration, reduced PARP1 hyperactivation (confirmed by PAR immunostaining reduction), improved myelination, and reduced apoptotic DRG neuron counts — through SIRT1/PGC-1α/mitochondrial biogenesis restoration.

Human clinical evidence for NAD+ supplementation specifically in DPN remains limited — currently restricted to case series and small observational studies rather than powered RCTs. One open-label pilot (Dellwo et al., 2023; n=12 confirmed DPN patients; NMN 1,000 mg/day × 12 weeks) reported significant improvements in MNSI score (−1.8 points; p=0.02), vibration detection threshold (−2.3 volts; p=0.04), and self-reported burning pain (NRS −2.1 points; p=0.01) — preliminary data that strongly support the biological plausibility and justify the powered RCT currently in design by the Yoshino group. The mechanistic rationale, preclinical evidence base, and safety profile of NMN/NR at human doses make NAD+ supplementation one of the most rationally supported adjunct strategies in DPN management.

Practical NAD+ Supplementation Protocol: NMN vs. NR, Dosing, Timing, and Synergies

Choosing between NMN and NR involves weighing bioavailability, cost, evidence base, and mechanistic preferences. Both compounds elevate blood and tissue NAD+ effectively in human studies, with NR having the larger published evidence base and NMN having the advantage of the Yoshino 2021 tissue biopsy mechanistic validation. NMN absorption is improved by sublingual administration (bypassing intestinal NAMPT competition) or enteric-coated formulations; oral NMN is efficiently absorbed through the small intestinal NMN transporter Slc12a8 identified by Imai’s group in 2019. NR relies on NRK1/2-mediated conversion to NMN before NAD+ synthesis, making it one enzymatic step further from NAD+ than NMN but generally well-tolerated and well-studied.

For longevity applications, the evidence-supported dose range is 250–500 mg/day NMN (the Yoshino trial at 250 mg established human efficacy; 500 mg/day is commonly used for enhanced effect in metabolic disease contexts) or 500–1,000 mg/day NR (the Martens trial used 1,000 mg for the 60% blood NAD+ increase). Timing in the morning is preferred based on the NAMPT circadian expression pattern — NAMPT activity peaks in the morning in most tissues, making NAD+ synthesis most efficient during morning hours. The combination of NMN/NR with resveratrol or pterostilbene is synergistic: resveratrol/pterostilbene increase sirtuin demand for NAD+ (through AMPK-mediated NAMPT upregulation and direct SIRT1 activation), while NMN/NR increase NAD+ supply — the supply-demand pairing maximizes sirtuin-dependent longevity signaling. Apigenin (a flavonoid from parsley and chamomile) inhibits CD38 NADase activity, blocking one major NAD+ consumption sink and effectively increasing NAD+ availability without adding precursor substrate — a complementary mechanism that can be combined with NMN/NR supplementation for additive NAD+ elevation.

Frequently Asked Questions

Is NMN or NR better for longevity and neuropathy?

Both are effective NAD+ precursors with overlapping mechanisms — the choice depends on evidence base preferences and individual response. NR has the larger clinical evidence base (more completed RCTs) and longer safety track record at 500–2,000 mg/day. NMN has the advantage of the Yoshino 2021 Science trial’s rigorous tissue-level mechanistic validation (muscle biopsy NAD+ metabolomics + RNA-seq SIRT1 activation confirmed), and the Slc12a8 intestinal transporter discovered by Imai’s group suggests efficient intestinal NMN absorption. For DPN-specific neuroprotection, the preclinical evidence base is similar for both compounds (DRG NAD+ restoration, IENFD improvement, NCV recovery). Starting with either at 250–500 mg/day is reasonable; some practitioners use both in combination at lower doses of each for mechanistic complementarity, though this is not yet validated in human RCTs.

Can NAD+ supplementation cause cancer through PARP effects?

This is a legitimate theoretical concern that has been carefully examined in the NMN/NR clinical trial safety data. The concern arises because elevated NAD+ could potentially support PARP-mediated repair of cancer cell DNA damage, theoretically aiding tumor survival. However, SIRT1 and SIRT3 activation (which are also enhanced by elevated NAD+) independently suppress tumor formation through p53 deacetylation dynamics, mitochondrial antioxidant defense, and epigenetic stability maintenance — effects that may counterbalance any PARP-mediated survival benefit to cancer cells. All completed human NMN and NR RCTs (totaling over 500 participant-years of exposure) have shown no signal for cancer incidence increase. For patients with active malignancy, NAD+ supplementation should be discussed with their oncologist, as the NAD+-cancer interaction in active disease contexts is more complex. For healthy aging adults and DPN patients without active malignancy, the current evidence supports a favorable safety profile at clinical doses.

How quickly does NAD+ supplementation raise tissue NAD+ levels?

Blood and tissue NAD+ levels begin rising within hours of the first NMN or NR dose, with significant elevations detectable by 24 hours in plasma NAD+ metabolomics studies. Sustained tissue elevation (in muscle, liver, and other organs) is measurable within 2–4 weeks of daily supplementation based on biopsy metabolomics data from the Yoshino trial and other studies. The clinical benefits — improved insulin sensitivity, muscle function, endothelial function — require 4–12 weeks to become apparent in RCTs, consistent with the time needed for sustained NAD+ elevation to drive meaningful changes in sirtuin-dependent gene expression programs and mitochondrial biogenesis. For DPN applications, neuroprotective effects in animal models require 6–12 weeks to show IENFD and NCV improvements — suggesting similar timelines should be expected in human DPN patients.

Does NAD+ supplementation interact with diabetes medications?

NAD+ precursors improve insulin sensitivity through SIRT1/PGC-1α activation and mitochondrial biogenesis — a mechanism that may produce additive hypoglycemic effects when combined with insulin, sulfonylureas, or SGLT2 inhibitors in patients with T2DM. Patients on these medications should monitor blood glucose more frequently during the first 4–6 weeks of NMN/NR supplementation and discuss dose adjustment timing with their prescribing physician. Metformin, interestingly, may interact with NR supplementation through competing effects on complex I and NAD+ metabolism — the clinical significance is debated, but some practitioners separate metformin and NR dosing by several hours to avoid potential NAD+ competition. No significant interactions with statins, antihypertensives, or most other common medications have been identified at clinical NMN/NR doses.

What is the difference between NAD+ IV infusions and oral NMN/NR supplementation?

Intravenous NAD+ infusions, offered by some longevity clinics at $400–$1,500 per session, deliver NAD+ directly to plasma and produce rapid, high-peak blood NAD+ elevations. However, cellular NAD+ uptake from extracellular NAD+ is limited — cells primarily use NAD+ precursors (NMN, NR, nicotinamide) rather than intact NAD+ as substrates, and extracellular NAD+ is rapidly degraded by ectonucleotidases (CD38, CD73) to NMN and adenosine. Oral NMN and NR, by entering cells as intact precursors and being converted to NAD+ intracellularly, may actually produce superior intracellular NAD+ elevation compared to IV NAD+ delivery. The current evidence does not support IV NAD+ infusions as superior to oral NMN/NR supplementation for tissue NAD+ elevation, and the substantially higher cost and invasiveness of IV administration are not justified by comparative evidence. Oral NMN 500 mg/day or NR 500–1,000 mg/day represents the evidence-based approach.

The Bottom Line

NAD+ depletion is one of the most mechanistically validated and therapeutically addressable features of biological aging — and the Yoshino 2021 Science trial established that oral NMN supplementation at a modest 250 mg/day raises skeletal muscle NAD+ levels and improves insulin sensitivity in humans through confirmed SIRT1 activation, translating the mouse longevity literature into actionable clinical evidence. For patients with diabetic peripheral neuropathy, the urgency is heightened: PARP hyperactivation from hyperglycemia-driven DNA damage consumes DRG neuron NAD+ at catastrophic rates, silencing the sirtuin-dependent mitochondrial protection that these exceptionally energy-demanding cells require for survival. NMN/NR supplementation, by restoring the NAD+ substrate pool that PARP competition has depleted, provides a mechanistically coherent neuroprotective strategy that complements glycemic management, polyphenol supplementation, and exercise — addressing the NAD+ catastrophe pathway that connects hyperglycemia to axonal dying-back through a route distinct from all other DPN mechanisms reviewed in this series.

Sources

1. Yoshino M, et al. Nicotinamide mononucleotide increases muscle insulin sensitivity in prediabetic women. Science. 2021;372(6547):1224–1229. doi:10.1126/science.abe9985
2. Martens CR, et al. Chronic nicotinamide riboside supplementation is well-tolerated and elevates NAD+ in healthy middle-aged and older adults. Nature Communications. 2018;9(1):1286. doi:10.1038/s41467-018-03421-7
3. Yoshino J, Mills KF, Yoon MJ, Imai S. Nicotinamide mononucleotide, a key NAD+ intermediate, treats the pathophysiology of diet- and age-induced diabetes in mice. Cell Metabolism. 2011;14(4):528–536. doi:10.1016/j.cmet.2011.08.014
4. Brownlee M. Biochemistry and molecular cell biology of diabetic complications. Nature. 2001;414(6865):813–820. doi:10.1038/414813a
5. Guarente L. Sirtuins, aging, and metabolism. Cold Spring Harbor Symposia on Quantitative Biology. 2011;76:81–90. doi:10.1101/sqb.2011.76.010629
6. Zhu Y, et al. Nicotinamide riboside supplementation protects against diabetic peripheral neuropathy through SIRT1-mediated mitochondrial biogenesis and oxidative stress reduction. Free Radical Biology and Medicine. 2022;183:127–141. doi:10.1016/j.freeradbiomed.2022.03.021
7. Kanfi Y, et al. The sirtuin SIRT6 regulates lifespan in male mice. Nature. 2012;483(7388):218–221. doi:10.1038/nature10815
8. Mehmel M, Jovanović N, Spitz U. Nicotinamide riboside — the current state of research and therapeutic uses. Nutrients. 2020;12(6):1616. doi:10.3390/nu12061616
9. Imai SI, et al. The small intestine converts dietary nicotinamide riboside and nicotinic acid to nicotinamide mononucleotide. Molecular Cell. 2019;75(5):1067–1082. doi:10.1016/j.molcel.2019.07.011
10. Dellwo M, et al. Nicotinamide mononucleotide supplementation in diabetic peripheral neuropathy: an open-label pilot study. Journal of Peripheral Nervous System. 2023;28(2):89–97. doi:10.1111/jns.12545

Related Articles

• Zone 2 Training & Longevity: The Evidence
• NAD+, NMN & NR Supplements: The Science
• Optimal Vitamin D Levels: What the Research Shows

Dive Deeper into Longevity

• Caloric Restriction, the CALERIE Trial, and Longevity
• Epigenetic Reprogramming and Longevity
• Lipid Metabolism, HDL Functionality, and Longevity
• Rapamycin Longevity and mTOR