Bioactive compounds
Caffeine, chlorogenic acids, trigonelline, diterpenes, melanoidins and minerals in a cup of coffee – how much a cup holds, where the body takes them up, and what has been measured in people.
Across the chain: Science
Contents
In brief
- A cup of coffee holds caffeine, chlorogenic acids, trigonelline, the diterpenes cafestol and kahweol, melanoidins and minerals; amounts vary widely with the coffee and the brew.1,10
- Most chlorogenic acids pass the small intestine; most of what reaches the blood forms after gut bacteria in the colon have changed them.5,1
- The clearest measured effect in people is the rise in serum cholesterol from cafestol and kahweol in unfiltered coffee; a paper filter retains them.7,8
- Assessing a company’s claim, the European Food Safety Authority (EFSA) found no established cause-and-effect link between a coffee standardised for chlorogenic acids and trigonelline and DNA protection.6
- Umbrella reviews see plausible mechanisms behind observational associations, but these remain hypotheses; the evidence is mostly observational and of low quality.2
What “bioactive” means#
Coffearium uses “bioactive compounds” for the substances in coffee that can act on the body beyond supplying energy or nutrients. A 2017 umbrella review describes coffee as a complex mixture of such compounds, with plausible biological mechanisms for benefiting health.2 “Plausible” is the key word. A compound that acts on cells in a dish may do nothing in a person.
The body changes most of these compounds before they reach the blood. Very little of the chlorogenic acid that is absorbed keeps the structure it had in the drink.1 The authors of a 2014 review therefore call many claims of benefit “gross over-extrapolations from in vitro studies”.1 They warn against uncritical extrapolation from animal studies that ignores species differences, metabolism and dose.1
This entry follows each compound group from the cup to the body. It names what has been measured in people and what has not. Long-term health outcomes, such as disease risk, are covered in Coffee and long-term health.
The main groups in a cup#
Six groups stand out in the reviews opened for this entry:1,10
- Caffeine. A typical cup holds 50–100 mg, according to a 2014 review.1 Café cups vary far more. The Caffeine entry covers amounts, metabolism and intake advice.
- Chlorogenic acids. These phenolic acids fall sharply during roasting. In a 2012 survey of 20 espressos bought in Glasgow, one serving held 24–422 mg of caffeoylquinic acids.11 Regular drinkers may take in 1 g or even 2 g a day.1 Their chemistry is covered in Chlorogenic acids.
- Trigonelline and its roast products. Trigonelline is a nitrogen compound. Roasting breaks part of it down to nicotinic acid (a form of niacin) and to pyridine compounds such as N-methylpyridinium (NMP).1 A serving holds about 40–110 mg of trigonelline and 1–3 mg of nicotinic acid.1
- Diterpenes. Cafestol and kahweol are fat-like compounds in the coffee oil. Boiled, French press and Turkish coffee carry 6–12 mg per cup; filter and instant coffee less than 0.6 mg.1 See Coffee oil.
- Melanoidins. These brown polymers form during roasting.9 A 2011 estimate put them at about 99–433 mg per serving, depending on the brew.10 See Melanoidins.
- Minerals. One cup provides 7–24 mg of magnesium and 34–116 mg of potassium, according to the 2014 review.1
Unwanted compounds that form in roasting, such as acrylamide and furan, are covered in Roasting contaminants.
Data table
| mg per cup or serving | Source | |
|---|---|---|
| Caffeine, typical cup | 50–100 mg | 1 |
| Caffeine, specialty coffees, US coffee shops | 58–259 mg | 1 |
| Caffeine, espresso, 20 Glasgow outlets | 51–322 mg | 1, 11 |
| Caffeoylquinic acids, espresso, 20 Glasgow outlets | 24–422 mg | 11 |
| Trigonelline, per serving | 40–110 mg | 1 |
| Potassium | 34–116 mg | 1 |
| Magnesium | 7–24 mg | 1 |
| Cafestol and kahweol, boiled, French press, Turkish | 6–12 mg | 1 |
| Cafestol and kahweol, espresso | ≈4 mg | 1 |
| Cafestol and kahweol, filter, instant | < 0.6 mg | 1 |
| Nicotinic acid | 1–3 mg | 1 |
Where the body takes them up#
The compounds do not all enter the body in the same place.
- Caffeine is absorbed rapidly and almost completely in the stomach and small intestine. Blood levels peak within 60 minutes.1
- Chlorogenic acids mostly pass the small intestine unabsorbed. In a 2010 study, people without a colon recovered about 71% of the intake in the fluid leaving the small intestine.5 In the colon, gut bacteria turn them into smaller molecules that are then absorbed.1,5
- Trigonelline appears in the blood within 15 minutes of drinking coffee. About half of it leaves in the urine within 8 hours.1
- Melanoidins are not digested in the upper gut and are mainly recovered in the faeces.1 Gut bacteria in the colon can ferment them, much as they ferment dietary fibre.10
- Diterpenes and minerals. No source opened for this entry measured how cafestol, kahweol, potassium or magnesium from coffee are absorbed.
What has been measured in people#
Only two effects of single compound groups are clear from the sources opened for this entry.
Cafestol and kahweol raise cholesterol. Some brewing methods raise total and LDL (low-density lipoprotein) cholesterol in the blood, and these two diterpenes are responsible, according to a 1997 review.7 A paper filter retains them, so filtered coffee does not have this effect.7 A 2012 meta-analysis of randomised trials confirmed a larger rise with unfiltered coffee.8 The rise reverses when people stop.1,2
Caffeine raises blood pressure. In meta-analyses of randomised trials, caffeine was associated with a rise in blood pressure.3 The European Food Safety Authority (EFSA) considers habitual intakes up to 400 mg a day of no safety concern for healthy non-pregnant adults.4 The Caffeine entry gives the full advice.
Chlorogenic acids, trigonelline and NMP: not established. In 2020 EFSA assessed a coffee standardised for these compounds. The claimed effect was protection of DNA from strand breaks. EFSA concluded that a cause-and-effect relationship had not been established.6
Melanoidins and minerals: no measured effect in the sources opened. A 2012 review lists antioxidant, anti-inflammatory and other activities that have been attributed to coffee melanoidins.9 No study in people and no health authority assessment of them was opened for this entry.
What umbrella reviews suggest#
Umbrella reviews pool many meta-analyses. A 2017 umbrella review in the BMJ found 201 meta-analyses of observational studies and 17 of trials.2 Its summary estimates showed the largest risk reduction for various outcomes at three to four cups a day.2 Most of that evidence is observational: it shows associations, not causes.2
The authors name possible mechanisms. These include antioxidant effects of chlorogenic acids and caffeine, and detoxifying enzymes induced by cafestol and kahweol.2 They are hypotheses, not findings in people. The authors call for robust randomised trials to show whether the associations are causal.2 Coffee and long-term health covers the outcomes; How to read coffee health news explains the study types.
Reference values#
Values as reported in the sources. Per-cup values come from reviews that collected studies of different coffees, roasts and cup sizes; they show ranges, not targets.
| Parameter | Value reported | Source |
|---|---|---|
| Caffeine, typical cup | 50–100 mg | Ludwig et al.1 |
| Caffeine, 20 specialty coffees, US coffee shops | 58–259 mg per cup | Ludwig et al. (review)1 |
| Caffeine, espresso, 20 Glasgow outlets | 51–322 mg per cup | Crozier et al.11; Ludwig et al. (review)1 |
| Caffeoylquinic acids, espresso, 20 Glasgow outlets | 24–422 mg per serving | Crozier et al.11 |
| Caffeoylquinic acids, regular drinkers | 1 g or even 2 g per day | Ludwig et al.1 |
| Trigonelline | 40–110 mg per serving | Ludwig et al.1 |
| Nicotinic acid | 1–3 mg per cup | Ludwig et al.1 |
| Cafestol and kahweol, boiled, French press, Turkish | 6–12 mg per cup | Ludwig et al.1 |
| Cafestol and kahweol, espresso | ≈4 mg per cup | Ludwig et al.1 |
| Cafestol and kahweol, filter and instant | < 0.6 mg per cup | Ludwig et al.1 |
| Melanoidins, espresso (50 mL) | 111.6 mg (estimate) | Fogliano & Morales10 |
| Melanoidins, filter (130 mL) | 233.9 mg (estimate) | Fogliano & Morales10 |
| Melanoidins, “Italian” brew, as labelled in the source (60 mL) | 99.3 mg (estimate) | Fogliano & Morales10 |
| Melanoidins, soluble coffee (100 mL) | 433.2 mg (estimate) | Fogliano & Morales10 |
| Melanoidins, daily intake | 0.5–2.0 g per day (moderate to heavy drinkers) | Fogliano & Morales10 |
| Magnesium | 7–24 mg per cup | Ludwig et al.1 |
| Potassium | 34–116 mg per cup | Ludwig et al.1 |
| Chlorogenic acids recovered after the small intestine (ileostomy) | 71 ± 7% of intake, 0–24 h | Stalmach et al.5 |
| Chlorogenic acid metabolites in urine | 8 ± 1% (ileostomy); 29 ± 4% (intact colon) | Stalmach et al.5 |
| Trigonelline, after one cup (808 µmol) | Peak ≈6 µM at ≈3 h; half-life ≈5 h | Ludwig et al.1 |
| Trigonelline in urine, 0–8 h | 57% (men); 46% (women) | Ludwig et al.1 |
| NMP, after one cup (172 µmol) | Peak ≈0.8 µM; 69% in urine | Ludwig et al.1 |
| Serum lipids, 12 randomised trials (1,017 participants) | Total cholesterol +8.1 mg/dL; LDL +5.4 mg/dL | Cai et al.8 |
The melanoidin values per serving are the authors’ own estimates: a mean melanoidin content of the coffee multiplied by the amount of coffee used for each brew.10 The espresso ranges come from a 2012 survey of single espressos from 20 outlets in Glasgow.11 The 2014 review reproduces it but gives the upper chlorogenic acid value as 423 mg;1 this entry uses the original 422 mg.
Chlorogenic acids in the gut#
The clearest picture of chlorogenic acid uptake comes from people with an ileostomy, an opening that drains the small intestine and bypasses the colon. In a 2010 study, they drank 200 mL of instant coffee with about 385 µmol of chlorogenic acids.5 About 71% of the intake reached the ileal fluid within 24 h, so it had passed the small intestine.5 Only 8% appeared in their urine, against 29% in people with an intact colon after a similar coffee.5 The authors conclude that this highlights the importance of metabolism in the colon.5
The timing tells the same story. Metabolites absorbed in the small intestine, such as caffeic acid and ferulic acid sulphates, peak about 1 h after drinking.1 More substantial amounts of metabolites from the colon, among them dihydrocaffeic and dihydroferulic acid, peak after 4.3–5.2 h.1 Dihydroferulic acid reached a peak plasma concentration of 385 nM about 4.7 h after a cup of coffee.1 Intact caffeoylquinic acids reach only very low, short-lived concentrations.1
This matters for every laboratory claim. The compounds that circulate in the blood are mostly metabolites, at nanomolar concentrations. The 2014 review stresses that studies should use the compounds that actually appear in the circulation, at real-world concentrations.1
Trigonelline and N-methylpyridinium#
Trigonelline makes up about 1% of green coffee dry weight.1 Roasting degrades part of it to nicotinic acid and several pyridine derivatives, including 1-methylpyridinium (NMP) and 1,2-dimethylpyridinium.1 According to the review, the 1–3 mg of nicotinic acid per cup equals 6–18% of the recommended daily allowance of 16 mg for adult men.1
In a human study summarised in the 2014 review, one cup provided 808 µmol of trigonelline and 172 µmol of NMP.1 Trigonelline peaked in plasma at about 6 µM after about 3 h; NMP at about 0.8 µM.1 Urinary excretion was high: 57% (men) and 46% (women) for trigonelline over 8 h, and 69% for NMP.1 Detection 15 min after drinking suggests that uptake begins in the stomach.1 No source opened for this entry measured a health effect of trigonelline or NMP in people.
Diterpenes and serum lipids#
Cafestol and kahweol make up about 1% of coffee beans by weight. Hot water extracts them, but a paper filter retains them.7 This explains why filtered coffee does not affect cholesterol, while Scandinavian boiled, cafetière (French press) and Turkish coffee do.7 Espresso has high concentrations, but its small servings make it an intermediate source, at about 4 mg per cup.1 The Coffee oil entry covers diterpene levels measured by brew method.
A 2012 meta-analysis pooled 12 randomised controlled trials with 1,017 participants, all in Western countries.8 In these trials, drinking coffee for a median of 45 days raised total cholesterol by 8.1 mg/dL and LDL cholesterol by 5.4 mg/dL (about 0.21 and 0.14 mmol/L; our conversion).8 The rise was greater in trials with unfiltered and caffeinated coffee, and grew with the amount drunk.8 A 2017 umbrella review found that in meta-analyses of trials coffee was associated with a rise in serum lipids, though the results varied widely between studies.3 According to the 2017 BMJ umbrella review, filtered coffee gave only a marginal rise.2
Melanoidins as fibre-like material#
Coffee brew is one of the main sources of melanoidins in the diet.9 A 2011 study estimated intake at 0.5–2.0 g per day for moderate and heavy coffee drinkers.10 Melanoidins escape digestion in the upper gut and can be fermented by bacteria in the colon; the authors note that they share some properties of dietary fibre.10 A 2012 review lists antioxidant, antimicrobial, anticariogenic, anti-inflammatory, antihypertensive and antiglycative activities attributed to them.9 Their exact structures are still unknown, so it is not clear which part of them would carry any effect.9
Reading the evidence#
The evidence for each compound group sits at a different level. From weakest to strongest for an effect in people:
| Level | What it shows | Example in this entry |
|---|---|---|
| Laboratory (in vitro) | Activity in cells or test tubes, often at concentrations far above blood levels | Activities attributed to melanoidins9; one caffeine study used concentrations more than 40-fold above blood levels after coffee1 |
| Uptake studies in people | Whether and where a compound or its metabolites reach the blood | Chlorogenic acids, trigonelline, NMP1,5 |
| Randomised trials on a marker | A measured change in a blood value or similar marker | Serum cholesterol from unfiltered coffee8; blood pressure from caffeine3 |
| Authority assessment | A formal review of a specific claim | EFSA on caffeine safety4; EFSA on Coffee C216 |
| Observational studies, pooled | Associations with disease outcomes; not proof of cause | Umbrella reviews2,3 |
Observational evidence can involve far more people than trials, but it ranks low for causation. The 2017 BMJ review graded most outcomes as low or very low quality, and even its meta-analyses of randomised trials as low quality because of risk of bias.2 Residual confounding can distort associations: the authors suggest that smoking, despite some adjustment, explains some apparent harmful associations.2
Decaffeinated coffee as a natural test. Decaf has a composition similar to regular coffee, but little or no caffeine.2 The BMJ review found no convincing harmful associations for decaf.2 The 2014 review notes that beneficial associations are often seen with decaf too.1 This points to compounds other than caffeine, but the BMJ authors caution that decaf drinkers may differ from other coffee drinkers.2
The Coffee C21 claim#
EFSA’s 2020 opinion is the only authority assessment of a non-caffeine coffee compound group opened for this entry. Coffee C21 was standardised by its content of caffeoylquinic acids, trigonelline and NMP.6 The applicant was a German coffee company, Tchibo GmbH, and the studies EFSA weighed were those it submitted.6 The opinion is an assessment of that evidence, not evidence of a benefit. The main trial was randomised, placebo-controlled and single-blind, in healthy non-smoking men who drank 750 mL a day in three portions for 4 weeks.6 EFSA found some evidence of an effect in it, but a similar study at the same centre did not replicate it.6 The panel concluded that a cause-and-effect relationship had not been established.6
Open questions#
- Which compound explains which association. The umbrella reviews propose mechanisms for chlorogenic acids, caffeine and diterpenes2, but no source opened for this entry tested one compound against a disease outcome in people.
- Diterpene uptake. No source opened for this entry measured how much cafestol and kahweol the body absorbs.
- Melanoidins in people. Intake has been estimated10, but no human study of their effects was opened for this entry.
- Minerals. The per-cup ranges come from one review1; no source opened here compared mineral content across brew methods.
Key studies#
Stalmach et al., 2010 – Chlorogenic acids in people with an ileostomy. Volunteers without a colon drank instant coffee, and their ileal fluid and urine were analysed. Most of the chlorogenic acids passed the small intestine, and far less appeared in urine than in people with an intact colon, which points to the colon as the main site of uptake after microbial metabolism.
Cai et al., 2012 – Coffee and serum lipids. A meta-analysis of 12 randomised controlled trials with 1,017 participants in Western countries. Coffee raised total cholesterol, LDL cholesterol and triglycerides, more with unfiltered and caffeinated coffee and more at higher intakes.
Poole et al., 2017 – Umbrella review of coffee and health. Collected 201 meta-analyses of observational studies and 17 of trials. Coffee was generally associated with lower risks at usual intakes, with pregnancy as a main exception; the authors grade most evidence as low quality and call for randomised trials. The review received no specific funding.
EFSA NDA Panel, 2020 – Coffee C21 and DNA strand breaks. Assessed a company’s claim for a coffee standardised by caffeoylquinic acids, trigonelline and NMP. The panel did not find a cause-and-effect relationship established, partly because a similar study did not replicate the main result.
Fogliano & Morales, 2011 – Melanoidin intake from coffee and bread. Combined published melanoidin contents with serving data to estimate intake. Coffee drinkers take in roughly 0.5–2.0 g of coffee melanoidins a day, material that escapes digestion and is fermented in the colon.
General information, not medical advice.
- Clinical weight of the cholesterol rise: a 2012 meta-analysis of 12 randomised trials found that coffee raised total and LDL cholesterol, more so with unfiltered coffee, and with a dose–response relation.8 The authors of a 2017 umbrella review describe the clinical relevance of such small increases with unfiltered coffee as hard to judge, and note that the changes reversed when people stopped drinking coffee.2 Both agree on the direction of the effect.
Coffearium (2026). Bioactive compounds. Coffearium – The open coffee compendium. Retrieved 22 October 2026, from https://coffearium.org/entries/bioactive-compounds/
@misc{coffearium2026-bioactive-compounds,
author = {{Coffearium}},
title = {{Bioactive compounds}},
year = {2026},
howpublished = {Coffearium -- The open coffee compendium},
url = {https://coffearium.org/entries/bioactive-compounds/},
urldate = {2026-10-22},
note = {Retrieved 22 October 2026}
}Entries are updated when the evidence changes; the retrieval date tells your readers which version you used. Changes are listed in the Change log.
Change log
- – Espresso caffeoylquinic acid range corrected from 24–423 to 24–422 mg and cited to the original 2012 survey (new source 11). Reason: The 2014 review gives 423 mg, but the survey it reproduces reports 24–422 mg per serving.
Sources
- ALudwig, I. A.; Clifford, M. N.; Lean, M. E. J.; Ashihara, H.; Crozier, A. (2014). Coffee: biochemistry and potential impact on health. Food & Function 5(8), 1695–1717. doi:10.1039/C4FO00042K · other copies
- APoole, R.; Kennedy, O. J.; Roderick, P.; Fallowfield, J. A.; Hayes, P. C.; Parkes, J. (2017). Coffee consumption and health: umbrella review of meta-analyses of multiple health outcomes. BMJ 359, j5024. doi:10.1136/bmj.j5024 · other copies
- AGrosso, G.; Godos, J.; Galvano, F.; Giovannucci, E. L. (2017). Coffee, Caffeine, and Health Outcomes: An Umbrella Review. Annual Review of Nutrition 37, 131–156. doi:10.1146/annurev-nutr-071816-064941 · other copies
- AEFSA Panel on Dietetic Products, Nutrition and Allergies (NDA) (2015). Scientific Opinion on the safety of caffeine. EFSA Journal 13(5), 4102. doi:10.2903/j.efsa.2015.4102 · other copies
- AStalmach, A.; Steiling, H.; Williamson, G.; Crozier, A. (2010). Bioavailability of chlorogenic acids following acute ingestion of coffee by humans with an ileostomy. Archives of Biochemistry and Biophysics 501(1), 98–105. doi:10.1016/j.abb.2010.03.005 · other copies
- AEFSA Panel on Nutrition, Novel Foods and Food Allergens (NDA) (2020). Coffee C21 and protection of DNA from strand breaks: evaluation of a health claim pursuant to Article 13(5) of Regulation (EC) No 1924/2006. EFSA Journal 18(3), e06055. doi:10.2903/j.efsa.2020.6055 · other copies
- AUrgert, R.; Katan, M. B. (1997). The cholesterol-raising factor from coffee beans. Annual Review of Nutrition 17, 305–324. doi:10.1146/annurev.nutr.17.1.305 · other copies
- ACai, L.; Ma, D.; Zhang, Y.; et al. (2012). The effect of coffee consumption on serum lipids: a meta-analysis of randomized controlled trials. European Journal of Clinical Nutrition 66, 872–877. doi:10.1038/ejcn.2012.68 · other copies
- AMoreira, A. S. P.; Nunes, F. M.; Domingues, M. R.; Coimbra, M. A. (2012). Coffee melanoidins: structures, mechanisms of formation and potential health impacts. Food & Function 3(9), 903–915. doi:10.1039/c2fo30048f · other copies
- AFogliano, V.; Morales, F. J. (2011). Estimation of dietary intake of melanoidins from coffee and bread. Food & Function 2(2), 117–123. doi:10.1039/c0fo00156b · other copies
- ACrozier, T. W. M.; Stalmach, A.; Lean, M. E. J.; Crozier, A. (2012). Espresso coffees, caffeine and chlorogenic acid intake: potential health implications. Food & Function 3(1), 30–33. doi:10.1039/C1FO10240K · other copies
Tier A · authoritative · B · trade and expert · C · general. Supporting quotes for every source are in the data version of this entry. If a publisher’s site refuses access, try the doi link or “other copies”.
Discussion
A space for readers to share experience from the farm, the roastery or the bar, ask each other and offer other views. The Coffearium team does not reply here. Please keep to the community rules; to change the entry itself, use Suggest a change.