Erythritol is approximately 70% as sweet as sucrose and has a mild cooling effect in the mouth with no aftertaste [16]. Erythritol is a naturally occurring sugar alcohol (or polyol) that is found a variety of fruits such as melon, watermelon, pears, grapes; and in fermented foods such as cheese, soy sauce [17,18,19]. Erythritol is also detected in plasma and urine in human subjects and animals [20,21]. It was detected in the plasma and urine of a child with an inborn error of the pentose phosphate pathway (PPP) [21]. Later Hootman et al. demonstrated that erythritol is endogenously produced in healthy human erythrocytes from glucose via PPP [22]. The PPP is a branch of glucose metabolism, present in all organisms, that synthesizes building block for nucleic acid and DNA; generates nicotinamide adenine dinucleotide phosphate hydrogen (NADPH), which is an essential co-factor in many anabolic reactions such as synthesis of fatty acids and non-essential amino acids; and regenerates the antioxidant, glutathione [23]. Schlicker et al. confirmed production of erythritol via PPP in human lung cancer cells and characterized two NADPH-dependent enzymes that catalyze the reduction of erythrose to erythritol, alcohol dehydrogenase 1 (ADH1) and sorbitol dehydrogenase (SORD) [24]. As these enzymes are highly expressed in liver and kidney, the authors [24] further proposed these metabolically active tissues as main contributors to endogenous synthesis of erythritol in mammals. However, factors influencing endogenous erythritol production require investigation.

Erythritol occurs in fruits at levels too low to allow it to be extracted economically. It can be produced chemically, however, this method is also not cost-efficient [25]. In the 1950s, traces of erythritol were found in residue of blackstrap molasses fermented by yeast [26]. This led to the discovery that erythritol can be produced via fermentation by yeast and yeast-like fungi via PPP [25]. Erythritol can also be produced by some lactic acid bacteria from glucose via the phosphoketolase pathway [27]. Fermentation by yeast and yeast-like fungi is currently used as a cost-effective method for commercial large-scale erythritol production utilizing substrates such as glucose, fructose, xylose, sucrose, cellulose, and glycerol [25,27]. Following fermentation, the fermented broth is heated, filtered to remove microorganism and other impurities before it is dried into crystals. The erythritol yields have been increased by optimizing the fermentation parameters and/or by gene-targeting biotechnologies to produce strains with higher activity of enzymes involved in synthesis pathways and/or lower activity of enzymes that enable the organism to utilize erythritol. Methods directed towards improving the cost-efficiency and bio-sustainability of production continue to be investigated, including utilizing readily available byproducts such as molasses or employing bacteria capable of generating the erythritol from wheat straw [26].

Because erythritol occurs in nature, the FDA considers microbial-produced erythritol to be a natural sweetener [3]. Possibly many consumers do not agree and this explains the low health perception ratings that erythritol received in the surveys already discussed [1,9]. However, preliminary results of a 2020 survey showed that 77% of 278 respondents had a positive or very positive attitude towards microbial applications in food production [28]. Interestingly, consumers who considered themselves “environmentally concerned” were more positive towards microbial applications in food compared to those who considered themselves “health concerned”. This perspective of the latter consumers may change, however, as the potential for reprogramed microbes to meet the planet’s increasing demands for the production of environmentally friendly biomolecules related to nutrition, pharmaceuticals, and even biodegradable plastic, continues to grow [6,29].

Safety reviews for erythritol were conducted by multiple regulatory entities. In 2000, The Joint Expert Commission on Food Additives of the World Health Organization and the Food and Agriculture Organization (JECFA) established the Acceptable Daily Intake (ADI) for erythritol as “not specified” [30]. In 2001, the FDA classified erythritol as “generally recognized as safe” (GRAS) substance for use by the general population as a sweetener and flavor-enhancer in food and beverage [10]. In the following years, the FDA approved other GRAS notices for erythritol and erythritol ingredients for uses including as non-nutritive sweetener, flavor enhancer, stabilizer, and thickener in a variety of foods such as bakery fillings, cakes and cookies, frozen dairy desserts, puddings, yogurt, chewing gum, candies, and reduced and low-calorie beverages [15,31,32,33]. In the European Union, The European Food Safety Authority (EFSA) approved erythritol in 2003 as safe for use as a non-sweetening food additive [11]. Later, in 2006, it was permitted as sweetener for use in all food applications as with other polyols [34]. The EFSA authorized erythritol use in beverages as non-sweetening additive in 2010 [35], and as sweetener in 2015 [36].

The safety of erythritol is based on extensive evidence from animal and human studies on its absorption, distribution, metabolism, and excretion; along with short and long-term toxicological studies to examine potential reproductive, developmental and genotoxicity; as well as any mutagenic and carcinogenic effect, as reviewed in detail [37]. In one long-term rat study (104–107 weeks) that examined the toxicity and carcinogenicity of diets containing 0, 2, 5, or 10% erythritol (equivalent to 1.0, 2.6, and 5.4 gm/kg/day in females and 0.9, 2.2, and 4.6 gm/kg/day in males, respectively), erythritol consumption did not affect the survival of the animals, and showed no signs of nephrotoxicity, tumor-inducing or tumor-promoting changes [38].

In general, excessive intake is of sugar alcohols is associated with undesirable gastrointestinal effects, including nausea, abdominal bloating, and diarrhea. These side effects are attributed to the fact that sugar alcohols are poorly absorbed, thus they induce an osmotic effect and water retention in the intestine [25]. In addition, unabsorbed polyols can undergo fermentation by intestinal microbiota resulting in gas formation. However, most of an erythritol load is absorbed with relatively minimal amount reaching the colon [37]. Consequently, erythritol is better tolerated and is associated with less gastrointestinal side effects than sorbitol and xylitol at comparable doses [37]. The tolerance upper limits for erythritol are higher than for other polyols (0.66 gm/kg/day in men and 0.80 gm/kg/day in women) [39]. However, larger doses (1 gm/kg/day) are reported to be well-tolerated [40].

Consisting of four carbons, erythritol is smaller and with lower molecular weight compared to the other commonly consumed sugar alcohols: xylitol (five carbons), sorbitol and mannitol (six carbons). The sugar alcohols are absorbed from the small intestine by passive diffusion in a size-dependent manner. Thus, erythritol is absorbed into the blood at a higher and faster rate than the larger sugar alcohols. Once in the blood, a major proportion of erythritol is un-metabolized and excreted unchanged in the urine. Studies in humans have shown that approximately 90% of a 20 gm-dose is recovered in the urine within 24 h, whereas 80% is recovered within 24 h when a 1 gm/kg dose is utilized [37]. The fate of the 10–20% erythritol not recovered in the urine is not clear. Hootman et al. have suggested that 5–10% of erythritol in blood may be oxidized to erythrose and further to erythronate [22].

It has mainly been assumed that the relatively small amount of unabsorbed erythritol passes to the colon. Data from 24 h in vitro studies utilizing human fecal samples indicate no evidence of erythritol being metabolized by gut microbiota [41]. This corroborated earlier evidence from humans consuming radio-labelled erythritol (25 gm). The study demonstrated that erythritol was almost completely recovered in urine, radiolabeled CO₂ was not detected, and H₂ gas did not increase in breath samples collected for 6 h after dosing [42]. This indicated that the nearly all of the absorbed erythritol was not metabolized systemically and the unabsorbed portion that transitioned to large intestine was not metabolized by gut microbiota.

Evidences from animals are not in agreement with data from human studies, and indicate 6–10% of consumed erythritol is metabolized by colonic microbiota with only about 1% excreted as erythritol in feces [37]. Consistent with erythritol fermentation by intestinal microbiota, mice that were fed a high-fat diet and administered water containing 5% erythritol had increased levels of short-chain fatty acids in their serum and feces [43]. Animal studies also suggest that the rates of erythritol metabolism by microbiota are higher in animals that have been pre-adapted to high erythritol diets [37]. As recently reviewed, the long-term in vivo studies that are needed to understand the metabolism of erythritol by human gut microbiota, as well as the effects of erythritol on human gut microbiota, have not been conducted [44]. Despite the lack of certainty regarding the metabolic processing of erythritol in the colon, and due to the uncertain fate of 10% of a 0.3 gm/kg dose, erythritol’s nutritive value has been estimated to be less than 0.4 kcal/gm [45]. However, for the purposes of nutritional labelling, erythritol it is assumed to contain 0 kcal/gm as compared to 2.4 kcal/gm for the other sugar alcohols [46].