Nobiletin: The Citrus Peel Flavone in Laboratory and Human Studies

Table of Contents

1. A Flavone Covered in Methoxy Groups
2. Where Nobiletin Is Found
3. How Much Reaches the Blood
4. The Body-Clock Finding in Mice
5. Blood Fats, Insulin and Arteries in Mice
6. Memory in Rat and Mouse Models
7. The Human Randomized Trials
8. What the Human Trials Can and Cannot Show
9. Safety and Drug-Enzyme Findings
10. What Is Not Known
11. Key Research Papers
12. Connections
13. Featured Videos

A Flavone Covered in Methoxy Groups

Flavonoids are a large family of plant pigments. Most of the familiar ones, such as quercetin in onions or the catechins in tea, carry several hydroxyl groups (an oxygen and a hydrogen) on their ring structure. Those groups make them fairly water-friendly, and the body attaches sugars or sulfates to them quickly and clears them.

Nobiletin belongs to a small sub-group called the polymethoxylated flavones, often shortened in research papers to PMFs. In these molecules almost every hydroxyl group has been replaced by a methoxy group (an oxygen bonded to a small methyl group). Nobiletin carries six of them. Its close relative tangeretin carries five. The result is a molecule that dissolves well in fat and poorly in water, which changes where it goes in the body and how it is broken down.

The heavy methoxy coating is the reason researchers became interested in these compounds: a fat-soluble flavone can cross cell membranes more easily than a typical water-soluble one. It is also the reason the gut and liver handle it differently. Much of nobiletin's breakdown in the body is the stepwise removal of those methyl groups, producing "demethylated" forms that are themselves being studied.

The wider family is described on the Flavonoids hub. The best-studied citrus flavonoid in people is a different molecule, the flavanone hesperidin, covered on the Hesperidin page. Results for hesperidin do not transfer to nobiletin, and the reverse is also true.

Where Nobiletin Is Found

Nobiletin sits in the outer, coloured layer of citrus peel. The amount varies a great deal between citrus species and between cultivars of the same species.

• Shikuwasa. A Japanese study of four cultivation lines of shikuwasa (also spelled shiikuwasha, sikwasa or hirami lemon) measured the polymethoxylated flavones in the peel. Nobiletin made up 56.74–64.77% of them and tangeretin 23.17–34.70%. The same peels held mainly hesperidin among the ordinary citrus flavanones, except one line in which neohesperidin dominated (Asikin 2012).
• Different citrus species compared. A Belgian laboratory profiled seven citrus species and detected 38 polymethoxylated flavonoids in all, 8 of which were present in every species. Wild-type species such as Citrus depressa gave higher yields than common commercial species such as lemon. The broadest variety of these compounds appeared in shikuwasa, in mandarin (Citrus reticulata) and in a mandarin-orange hybrid (Brits 2021).
• Aged tangerine peel. The cultivar-by-cultivar nobiletin content of chen pi, and how much of it reaches a cup of tea, is covered in detail on the Chen Pi Polymethoxyflavones and Metabolism page.

Because nobiletin is fat-soluble, how a peel is prepared matters. The human trials described below did not use plain peel tea. They used concentrated peel extracts, purified nobiletin powders or dried immature fruit powders packed into capsules or tablets.

How Much Reaches the Blood

Very few studies have measured nobiletin in human blood. The clearest human figure comes from an unusual source: a traditional Japanese (Kampo) stomach formula called rikkunshito, which includes citrus peel.

• In people. In a randomized, open-label crossover study of 21 healthy Japanese volunteers, researchers gave rikkunshito at 2.5, 5.0 or 7.5 g a day and measured nine of its ingredients in the blood. Nobiletin was one of the ingredients that was absorbed, but at a very low level: after the 7.5 g dose its average peak concentration was 35.9 picograms per millilitre (95% confidence interval 24.6–52.5) (Kitagawa 2015). A picogram is a millionth of a microgram.
• In rats. A study comparing lean rats and rats made obese on a high-fat diet found that about a fifth of an oral dose of nobiletin reached the circulation: an absolute oral bioavailability of 22.37% in lean and 18.67% in obese rats, a difference that was not statistically significant. The gut bacteria removed methyl groups from nobiletin more actively than the rats' own tissues did, and obese rats had more of these demethylated forms in their faeces and blood (Zhang 2020).
• Peel extract in rats. One hour after rats were given a shikuwasa peel extract at 1 g per kilogram of body weight, nobiletin was found in the blood and in various tissues, and was most abundant in the liver (Li 2022).

These numbers matter when reading the animal studies below. The mouse and rat experiments used doses far larger, relative to body size, than a person gets from food or from the products tested in the human trials.

The Body-Clock Finding in Mice

Almost every cell in the body runs a roughly 24-hour clock built from a loop of genes that switch one another on and off. The broad biology is shown in the Sleep Stages and Your Body Clock animation. Disrupted clocks are linked with weight gain and poor blood-sugar control in both animal and human studies, which led a group in Texas to search for molecules that could strengthen the clock.

• The screen (cells). Using fibroblast cells engineered to glow with each turn of the clock gene PER2, the researchers screened a chemical library and identified nobiletin as a molecule that increases the amplitude of the clock, meaning the daily rise and fall becomes stronger (He 2016).
• Obese mice. In mice made obese on a high-fat diet, nobiletin strongly counteracted the features of metabolic syndrome and increased energy expenditure and physical activity. These effects depended on a working Clock gene. The clock was also required for nobiletin's effects in a second, genetically diabetic mouse strain. The researchers identified a family of nuclear receptors called the retinoid acid receptor-related orphan receptors (RORs) as the direct targets (He 2016).
• Aging mice. A follow-up study by the same group gave nobiletin to naturally aged mice. On a regular diet it improved their metabolic fitness; on a high-fat diet it supported healthier aging. In the skeletal muscle of high-fat-fed mice, nobiletin switched on genes for the mitochondrial respiratory chain and strengthened its structure, leading to more energy production (ATP) and fewer reactive oxygen species (Nohara 2019). One author of that paper reported co-founding a company working on circadian rhythms and metabolism.

All of this is mouse and cell work. No published randomized trial has measured whether nobiletin changes the body clock itself in people, for example by tracking melatonin or core body temperature rhythms. The one human trial built on the clock idea, the nocturia trial described below, measured night-time urination, not the clock.

Blood Fats, Insulin and Arteries in Mice

A Canadian group studied nobiletin in human liver cells grown in a dish and in mice prone to clogged arteries (mice lacking the low-density lipoprotein receptor) fed a Western-style diet (Mulvihill 2011).

• In liver cells. Nobiletin activated a signalling pathway that markedly cut the cells' release of apolipoprotein B100, the protein backbone of the particles that carry fat out of the liver. Unlike insulin, it did this without triggering the insulin receptor and without stimulating fat production.
• In mice. Nobiletin lowered the liver's output of fat-rich very-low-density lipoprotein (VLDL) particles, prevented fat building up in the liver, increased fat burning, increased the sensitivity of the liver and other tissues to insulin, improved glucose tolerance and "dramatically" reduced atherosclerosis at the root of the aorta. These effects did not run through the PPAR receptors that several diabetes and cholesterol drugs act on.

The authors wrote that nobiletin "provides insight into treatments" for these conditions. No randomized trial of nobiletin on its own has measured blood lipids, blood sugar or artery disease in people.

Memory in Rat and Mouse Models

A Japanese research group spent more than a decade on nobiletin and the brain. Two of their early papers set out the idea that the later human memory trials were built on.

• Rats and nerve cells. In cultured rat hippocampal neurons, a fragment of amyloid-beta (the protein that forms plaques in Alzheimer's disease) blunted the switching-on of CREB, a protein involved in forming long-term memories. Pretreatment with nobiletin reversed that effect. In rats given a continuous infusion of amyloid-beta into the brain, nobiletin protected their ability to learn an eight-arm radial maze (Matsuzaki 2006).
• Transgenic mice. In mice carrying two human gene mutations that cause early-onset Alzheimer's disease, nobiletin improved memory deficits. Four months of nobiletin markedly reduced the amount of two forms of amyloid-beta in the brain and reduced plaques in the hippocampus (Onozuka 2008).

These are animal models that reproduce part of Alzheimer's disease, not the disease as it occurs in people. The human condition is described on the Alzheimer's Disease page. A 2022 narrative review by members of the same research group summarises their view of the citrus-peel brain research, including the human studies (Matsuzaki 2022).

The Human Randomized Trials

Four randomized trials published as PubMed abstracts tested nobiletin or a nobiletin-rich citrus product in people. All four were run in Japan, all were small, and three of the four gave nobiletin together with other ingredients. Each is set out below with what the abstract states.

Night-time urination: nobiletin plus tangeretin

• Who: 40 people aged 50 or over with nocturia (getting up to pass urine more than twice a night on a bladder diary); 13 were women and the average age was 73.5 years. 36 completed the study and 4 withdrew.
• Design: randomized, placebo-controlled, double-blind crossover. Each person took both the active product and the placebo in turn, with a washout of at least two weeks between.
• What was given: a mixture of nobiletin and tangeretin from shikuwasa peel, 50 mg once daily for 6 weeks, against a placebo.
• Main outcome: the change in nocturnal bladder capacity (how much the bladder holds at night). Not met: the mixture had little effect on it compared with placebo.
• Secondary outcomes: night-time voids fell by 0.5 per night more than on placebo (p = 0.040). The share of urine made at night (the nocturnal polyuria index) fell by 2.8% from the start to the end of the active period (p = 0.048), a change measured within the active period rather than against placebo; the authors described it as a tendency.
• Safety: no adverse events directly related to the mixture were observed (Ito 2023).

The background to this symptom is on the Frequent Urination and Overactive Bladder pages.

Memory in healthy older adults: a three-ingredient product

• Who: healthy elderly Japanese people at several centres. The abstract reports the per-protocol group analysed, 108 people; it does not give the number randomized.
• Design: multicentre, randomized, double-blind, placebo-controlled.
• What was given: a combination product of high-purity nobiletin powder from shikuwasa peel together with dried root powder of black ginger (Kaempferia parviflora) and dried leaf powder of Peucedanum japonicum, a coastal plant of the carrot family, once a day for 16 weeks. The abstract does not state the nobiletin dose.
• Main outcome: global memory on the Japanese version of the Wechsler Memory Scale-Revised. Reported as met: the "general memory" and "visual memory" index scores were significantly higher with the product than with placebo, and the change in total score was 9.0 (±7.20) points with the product against 5.9 (±7.70) with placebo, a significant difference.
• Subgroups: similar changes were seen in people aged 74 or under. In people with Mini-Mental State Examination scores of 24 to 28, the "figural memory" subscale was higher with the product (Yamada 2021).
• Notes: because black ginger and the second plant were given with the nobiletin, the result cannot be assigned to nobiletin alone. Two of the six authors list an affiliation with a private research company. Black ginger has its own research record, described on the Black Ginger (Kaempferia parviflora) page; its rhizome also contains methoxyflavones.

Memory over a year: perilla seed oil with or without ponkan powder

• Who: 49 healthy elderly Japanese people.
• Design: 12-month randomized, double-blind, parallel-group trial. There was no placebo group: both arms took perilla seed oil, and the comparison was whether ponkan powder was added.
• What was given: one group took capsules of 1.47 mL of perilla seed oil daily (0.88 g of alpha-linolenic acid). The other took the same oil plus 1.12 g of air-dried immature ponkan (a mandarin variety) powder, which supplied 2.91 mg of nobiletin a day.
• Outcome: at the end, the group with ponkan powder had significantly higher cognitive index scores than the oil-only group. That group also showed rises in alpha-linolenic acid and docosahexaenoic acid in red blood cell membranes, in serum brain-derived neurotrophic factor (BDNF) and in a blood antioxidant measure (Hashimoto 2022).
• Notes: the daily nobiletin amount, 2.91 mg, is small, and the authors linked the result partly to the rise in omega-3 fats in red blood cells. The trial shows a difference between two products, not an effect of nobiletin against placebo.

Muscle after stroke: shikuwasa fruit extract

• Who: 40 people in the subacute stage after a stroke, with moderate-to-severe weakness of one side, who were in rehabilitation.
• Design: double-blind, placebo-controlled, randomized pilot trial.
• What was given: a shikuwasa fruit extract or placebo for 12 weeks. The abstract does not state the dose or the nobiletin content.
• Outcome: thigh muscle area measured by computed tomography. Total muscle area was not different between groups. The area of normal (low-fat) muscle in the thigh on the unaffected side was significantly higher with the extract at 12 weeks.
• Notes: the authors called it a pilot with a small sample that needs a large trial to confirm it. Three of the seven authors are employees of a food company (Honaga 2022).

What the Human Trials Can and Cannot Show

Read together, the four trials are a start rather than an answer. Several features limit what they show.

• Size and length. The trials enrolled 40, 49, 40 and (in the analysed group) 108 people, for 6 to 52 weeks. Trials of that size can miss real effects and can also produce chance findings.
• Mixtures. Only the nocturia trial gave polymethoxylated flavones without other plants, and even it combined nobiletin with tangeretin. The two memory trials paired nobiletin with black ginger and a second plant, or with perilla seed oil.
• Doses spread widely. The amounts stated range from 2.91 mg of nobiletin a day in the ponkan trial to 50 mg a day of a nobiletin-tangeretin mixture in the nocturia trial; two trials do not state a dose in the abstract. The mouse and rat studies used far larger amounts relative to body weight.
• Outcomes. In the nocturia trial the main outcome was not met; the positive results were secondary. The memory trial results come from a per-protocol analysis, which counts only people who followed the protocol.
• One country. All four were conducted in Japan, in older adults or stroke patients.
• Gaps. A PubMed search at the time of writing (November 2026) found no randomized trial of purified nobiletin alone measuring blood lipids, blood sugar, body weight or circadian rhythm markers in people.

The animal record and the human record are therefore about different things: in mice, a clock-acting molecule at high doses; in people, small trials of low-dose, mostly mixed citrus products on memory scores, night-time urination and thigh muscle.

Safety and Drug-Enzyme Findings

The human safety record for nobiletin is thin because the trials were small and short.

• In the trials. The nocturia trial reported no adverse events directly related to the nobiletin-tangeretin mixture over 6 weeks in 40 older adults (Ito 2023). The other trial abstracts do not report adverse-event figures.
• Drug-processing enzymes in the laboratory. In liver microsome preparations, nobiletin itself inhibited the activity of several cytochrome P450 (CYP) enzymes, the liver enzymes that break down most medicines, at concentrations of 13.0 to 33.2 micromolar; a shikuwasa peel extract did the same at 18.5 to 54.4 micrograms per millilitre (Li 2022).
• Drug-processing enzymes in living rats. In the same study, rats fed a high-fat diet containing 4% shikuwasa peel for 11 weeks had lower activity of the liver enzymes CYP1A1, CYP1A2, CYP2B and CYP2E1 than rats on the high-fat diet alone, but higher activity of CYP3A, the enzyme family that handles the largest share of prescription drugs. A drug-export transporter (multidrug resistance-associated protein 2) was also raised (Li 2022).

These are laboratory and rat findings. No study in people was found that measured whether nobiletin, or a nobiletin-rich peel product, changes blood levels of any medicine. The concentrations that inhibited enzymes in the test tube are far above the picogram-per-millilitre levels measured in human blood after a citrus-containing herbal formula (Kitagawa 2015), so the size of any real-life effect is unknown. The cautions recorded for whole aged tangerine peel are on the Chen Pi page.

No data were found on nobiletin supplements in pregnancy, breastfeeding or childhood.

What Is Not Known

• Whether nobiletin strengthens the body clock in people, as it does in mice.
• Whether nobiletin taken alone changes blood fats, blood sugar, insulin sensitivity, body weight or artery disease in people.
• Whether the memory results come from nobiletin, from the other ingredients in the same products, or from chance, and whether they would hold in larger trials outside Japan.
• Whether the nocturia finding on night-time voids would be repeated in a larger trial, and how a 0.5-void difference compares with other approaches.
• What dose would be needed in people to reach the blood and tissue levels used in the animal studies.
• How the demethylated breakdown products made by gut bacteria contribute to any effect.
• Whether nobiletin changes the handling of any prescribed medicine in people, and its safety over months to years.
• How much nobiletin people actually take in from eating citrus peel, marmalade or peel-flavoured dishes.

Decisions about supplements and medicines belong with a clinician who knows the person's full list of medicines.

Key Research Papers

1. He B, Nohara K, Park N, Park YS, Guillory B, Zhao Z, Garcia JM, Koike N, Lee CC, Takahashi JS, Yoo SH, Chen Z. The Small Molecule Nobiletin Targets the Molecular Oscillator to Enhance Circadian Rhythms and Protect against Metabolic Syndrome. Cell Metab. 2016;23(4):610-21. PubMed PMID: 27076076
2. Nohara K, Mallampalli V, Nemkov T, Wirianto M, Yang J, Ye Y, Sun Y, Han L, Esser KA, Mileykovskaya E, D'Alessandro A, Green CB, Takahashi JS, Dowhan W, Yoo SH, Chen Z. Nobiletin fortifies mitochondrial respiration in skeletal muscle to promote healthy aging against metabolic challenge. Nat Commun. 2019;10(1):3923. PubMed PMID: 31462679
3. Mulvihill EE, Assini JM, Lee JK, Allister EM, Sutherland BG, Koppes JB, Sawyez CG, Edwards JY, Telford DE, Charbonneau A, St-Pierre P, Marette A, Huff MW. Nobiletin attenuates VLDL overproduction, dyslipidemia, and atherosclerosis in mice with diet-induced insulin resistance. Diabetes. 2011;60(5):1446-57. PubMed PMID: 21471511
4. Matsuzaki K, Yamakuni T, Hashimoto M, Haque AM, Shido O, Mimaki Y, Sashida Y, Ohizumi Y. Nobiletin restoring beta-amyloid-impaired CREB phosphorylation rescues memory deterioration in Alzheimer's disease model rats. Neurosci Lett. 2006;400(3):230-4. PubMed PMID: 16581185
5. Onozuka H, Nakajima A, Matsuzaki K, Shin RW, Ogino K, Saigusa D, Tetsu N, Yokosuka A, Sashida Y, Mimaki Y, Yamakuni T, Ohizumi Y. Nobiletin, a citrus flavonoid, improves memory impairment and Abeta pathology in a transgenic mouse model of Alzheimer's disease. J Pharmacol Exp Ther. 2008;326(3):739-44. PubMed PMID: 18544674
6. Ito H, Negoro H, Kono J, Hayata N, Miura T, Manabe Y, Miyazaki Y, Mishina M, Woo JT, Sakane N, Okuno H. Effectiveness and Safety of a Mixture of Nobiletin and Tangeretin in Nocturia Patients: A Randomized, Placebo-Controlled, Double-Blind, Crossover Study. J Clin Med. 2023;12(8):2757. PubMed PMID: 37109094
7. Yamada S, Shirai M, Ono K, Teruya T, Yamano A, Tae Woo J. Beneficial effects of a nobiletin-rich formulated supplement of Sikwasa (C. depressa) peel on cognitive function in elderly Japanese subjects; A multicenter, randomized, double-blind, placebo-controlled study. Food Sci Nutr. 2021;9(12):6844-6853. PubMed PMID: 34925812
8. Hashimoto M, Matsuzaki K, Maruyama K, Hossain S, Sumiyoshi E, Wakatsuki H, Kato S, Ohno M, Tanabe Y, Kuroda Y, Yamaguchi S, Kajima K, Ohizumi Y, Shido O. Perilla seed oil in combination with nobiletin-rich ponkan powder enhances cognitive function in healthy elderly Japanese individuals: a possible supplement for brain health in the elderly. Food Funct. 2022;13(5):2768-2781. PubMed PMID: 35171190
9. Honaga K, Otaka Y, Kaji K, Sakata Y, Hamano H, Nakamura H, Kondo K. Effects of Citrus depressa Hayata Fruit Extract on Thigh Muscles Mass and Composition in Subacute Stroke Patients: A Double-Blind, Randomized, Controlled Pilot Trial. J Integr Complement Med. 2022;28(5):391-398. PubMed PMID: 35100035
10. Kitagawa H, Munekage M, Matsumoto T, Sadakane C, Fukutake M, Aoki K, Watanabe J, Maemura K, Hattori T, Kase Y, Uezono Y, Inui A, Hanazaki K. Pharmacokinetic Profiles of Active Ingredients and Its Metabolites Derived from Rikkunshito, a Ghrelin Enhancer, in Healthy Japanese Volunteers: A Cross-Over, Randomized Study. PLoS One. 2015;10(7):e0133159. PubMed PMID: 26186592
11. Zhang M, Xin Y, Feng K, Yin B, Kan Q, Xiao J, Cao Y, Ho CT, Huang Q. Comparative Analyses of Bioavailability, Biotransformation, and Excretion of Nobiletin in Lean and Obese Rats. J Agric Food Chem. 2020;68(39):10709-10718. PubMed PMID: 32880448
12. Li ML, Huang CY, Chen SH, Lee CN, Yao HT. Citrus depressa Hayata peel ameliorates nonalcoholic fatty liver and modulates the activity of hepatic drug-metabolizing enzymes and transporters in rats fed a high-fat diet. Food Funct. 2022;13(6):3353-3367. PubMed PMID: 35212695
13. Asikin Y, Taira I, Inafuku-Teramoto S, Sumi H, Ohta H, Takara K, Wada K. The composition of volatile aroma components, flavanones, and polymethoxylated flavones in Shiikuwasha (Citrus depressa Hayata) peels of different cultivation lines. J Agric Food Chem. 2012;60(32):7973-80. PubMed PMID: 22804782
14. Brits M, Naessens T, Theunis M, Taktak O, Allouche N, Pieters L, Foubert K. Identification and Quantification of Polymethoxylated Flavonoids in Different Citrus Species Using UPLC-QTOF-MS/MS and HPLC-DAD. Planta Med. 2021;87(12-13):1080-1088. PubMed PMID: 34412145
15. Matsuzaki K, Nakajima A, Guo Y, Ohizumi Y. A Narrative Review of the Effects of Citrus Peels and Extracts on Human Brain Health and Metabolism. Nutrients. 2022;14(9):1847. PubMed PMID: 35565814

PubMed Topic Searches

1. Nobiletin randomized controlled trials
2. Nobiletin and circadian rhythm
3. Nobiletin and memory
4. Nobiletin pharmacokinetics
5. Polymethoxylated flavones in citrus peel

Back to Table of Contents

Connections

• Flavonoids — the family nobiletin belongs to
• Chen Pi (Aged Tangerine Peel) — the traditional peel richest in nobiletin
• Chen Pi Polymethoxyflavones and Metabolism — peel content and the dose gap
• Hesperidin — the citrus flavanone with the larger human trial record
• Black Ginger (Kaempferia parviflora) — a co-ingredient in one memory trial
• Circadian Rhythm Sleep-Wake Disorders — the body clock in people
• Alzheimer's Disease — the condition the animal memory models imitate
• Frequent Urination — the symptom studied in the nocturia trial

Back to Table of Contents

Featured Videos