Paraxanthine

Paraxanthine is 1,7-dimethylxanthine, the compound caffeine mostly becomes in the body. In people taking caffeine regularly, about 80% of it was converted to paraxanthine, against about 11% to theobromine and 4% to theophylline [1]. Tea and coffee drinkers carry about 2 mg/L of it in their blood across the day [5], and with regular caffeine use, serum paraxanthine averages two thirds of the caffeine level [2]. Anyone who drinks coffee is already exposed to it. It is now also made synthetically and sold as a supplement ingredient, enfinity [6], which its sponsor's trials describe as a nootropic alternative to caffeine [7].

Until 2021, only two human studies had given paraxanthine on its own [3]: a pharmacokinetic study in six men [5] and an NIH-funded crossover study in 12 coffee drinkers that measured its stimulant effects [2]. Since then the ingredient's sponsor has funded a run of small crossover trials, and its principals, who hold patent applications on paraxanthine, are co-authors [3][4][7][8].

Cognition. In 13 healthy adults, a single 200 mg dose was followed by fewer card-sorting errors at six hours, faster short-term memory responses and better-held vigilance than placebo [3]. A dose comparison in 12 adults (50, 100 and 200 mg, each then taken daily for a week) reported benefits at every dose, most consistently at 100 and 200 mg [4]. In both trials the main repeated-measures test, a treatment-by-time interaction, was not significant for any cognitive test; the reported benefits come from comparing each condition's change from baseline [3][4]. In 12 trained runners given 200 mg before a 10-km treadmill run, paraxanthine did not beat placebo on card sorting. It did give faster reaction times than placebo after the run, and fewer perseverative errors than 200 mg of caffeine. Adding caffeine to paraxanthine brought no further benefit [7].

Exercise. Paraxanthine alone has not improved performance in people. Run times did not differ between paraxanthine, caffeine, both or placebo [7]. In 14 male university rowers, 200 mg before an evening 2000-m ergometer trial did not clearly improve time or power; only paraxanthine plus 200 mg of caffeine beat placebo, by about 3 seconds [9]. The muscle results come from mice, where four weeks of dosing increased grip strength by 17%, treadmill distance by 39% and calf muscle mass [10].

Metabolism. In 21 adults, 200 mg raised resting energy expenditure above placebo at every time point over three hours, and it blunted the rise in hunger seen after placebo. Free fatty acids rose only after 300 mg, and glycerol, a marker of fat breakdown, did not change at any dose [8]. A 1990 study pointed the same way. It gave 10 men intravenous caffeine (4 mg/kg of lean mass), with no placebo arm. Over three hours, as caffeine fell and paraxanthine rose, free fatty acids rose with it (r = 0.93 against paraxanthine), while glycerol again did not change. The authors read this as a possible lipolytic role for paraxanthine and noted that a correlation does not show cause [11].

Paraxanthine is an adenosine receptor antagonist, like caffeine. In rat brain it displaced A1 receptor binding at 40 to 65 µM, against 90 to 110 µM for caffeine, and blocked A2 receptor signaling in the striatum at about half caffeine's concentration [12]. Its stimulant effect in rats counteracted an A1 agonist more efficiently than an A2A agonist, as caffeine's does [13].

In people the effects look like caffeine's. Twelve regular coffee drinkers took capsules of caffeine and of paraxanthine, each after 3 days off caffeine. At 4 mg/kg (seven people), paraxanthine raised diastolic blood pressure, plasma adrenaline and free fatty acids about as much as the same dose of caffeine. At 2 mg/kg (five people) caffeine had the larger effect: paraxanthine raised diastolic pressure but did not significantly raise adrenaline or free fatty acids. Neither drug significantly changed heart rate or cortisol, and the authors report the systolic rises as not statistically significant. Placebo was always given the day before the active dose. The authors suggest that caffeine, being less polar, reaches its sites of action more readily at low doses [2].

It also does something caffeine does not. In rats injected with 30 mg/kg, paraxanthine produced stronger locomotor activation than caffeine, theophylline or theobromine. It raised cGMP and extracellular dopamine (by about 50%) in the striatum, which caffeine at the same dose did not. Blocking nitric oxide synthesis reduced both its locomotor and its dopamine effects, and rats trained to recognize caffeine generalized poorly to paraxanthine. The authors attribute the difference to inhibition of a cGMP-preferring phosphodiesterase (PDE9), which strengthens nitric oxide signaling [13].

Other effects come from cells and rodents:

• In cultures of embryonic rat midbrain, 800 µM paraxanthine increased the number of surviving dopamine neurons by 169%, against 40% for caffeine at the same concentration, and the rescued neurons still released dopamine. The effect ran through ryanodine receptor calcium channels (the blocker dantrolene abolished it), not adenosine receptors or cAMP. The authors note that this concentration is probably severalfold above what can be reached in the brain [14].
• In mice, 10 or 30 mg/kg given before the toxin MPTP reduced the loss of striatal dopamine, as caffeine did [15].
• In mouse hippocampal slices and cultured cells it increased cysteine uptake through the EAAC1 transporter and raised glutathione, which the other caffeine metabolites tested did not [16].
• In young and aged rats, 10 days of oral dosing improved water-maze learning and raised brain BDNF more than the same dose of caffeine [17].
• In narcoleptic and normal mice, oral paraxanthine (6.25 to 100 mg/kg) promoted wakefulness in step with the dose. Given in the resting (light) period it was stronger and longer lasting than equimolar caffeine, and 100 mg/kg roughly matched 200 mg/kg of Modafinil. Given in the active (dark) period, caffeine's effect was the more robust, lasting over 12 hours. Both methylxanthines, unlike modafinil, reduced the narcolepsy-specific jumps from waking straight into REM sleep. The top caffeine dose dropped body temperature by more than 5 °C for over 6 hours; paraxanthine did not. Caffeine raised anxiety-like behavior on two tests, while paraxanthine had no effect on one and only a marginal one (p = 0.06) on the other [18].
• In rats with blocked bile ducts, paraxanthine given by mouth reduced liver injury, collagen deposition and fibrosis signaling [19].

The neuron and narcolepsy studies were funded in part by the drug company Pierre Fabre, which supplied the paraxanthine and whose researchers co-authored both; the narcolepsy paper discloses a patent application on paraxanthine's therapeutic uses [14][18].

Human safety data come from the short trials. Single doses of 50 to 300 mg, and 50 to 200 mg a day for a week, did not change blood counts or metabolic panels in any clinically meaningful way, and side effects were no more common than with placebo [4][7][8]. After 300 mg a few blood values (glucose, calcium, bilirubin, basophils) shifted significantly but stayed within normal ranges, and one participant reported dizziness, a racing heart, lightheadedness, nervousness and shortness of breath, all minor and gone within a day [8]. In the 10-km run trial, caffeine raised ratings of racing heart, shortness of breath and nervousness, and paraxanthine did not [7]. It is still a stimulant: at 4 mg/kg it raised diastolic blood pressure and adrenaline as much as caffeine [2].

Two industry-sponsored toxicology batteries in rats found no mutagenicity or genotoxicity. The oral LD50 was 1,601 mg/kg for the enfinity ingredient [6] and 829 mg/kg for another maker's product [23]. In 90-day studies no adverse effects were seen at the highest doses tested, 300 mg/kg a day [6] and 185 mg/kg a day [23]. In the second study, two rats given 185 mg/kg of caffeine died and none on paraxanthine did, so its no-adverse-effect level was set at 185 mg/kg against 150 mg/kg for caffeine. Rats on paraxanthine showed brief hyperactivity and hyperreflexia after each dose [23].

Paraxanthine is cleared mainly by CYP1A2, the same liver enzyme that clears caffeine, with CYP2A6 handling a second route [21][22]. The CYP1A2 inhibitor fluvoxamine stretches caffeine's half-life from about 5 hours to 56 [24]. A similar slowing of paraxanthine is an inference from the shared enzyme, not a measured result.

Taking it with caffeine adds to the dose, because caffeine itself is mostly converted to paraxanthine [1]. In the rowing trial only the combination improved performance, and only the combination significantly worsened that night's sleep and the next day's sleepiness [9]. In the running trial, adding caffeine brought no cognitive benefit over paraxanthine alone, and reaction times after the run were slower with the combination [7].