WHO FOOD ADDITIVES SERIES: 52
First draft prepared by
Dr D.J. Benford
Food Standards Agency, London, United Kingdom
D-Tagatose is a ketohexose, an epimer of D-fructose inverted at C-4. It is obtained from D-galactose by isomerization under alkaline conditions in the presence of calcium. Its properties permit its use as a bulk sweetener, humectant, texturizer and stabilizer.
D-Tagatose was evaluated by the Committee at its fifty-fifth and fifty-seventh meetings (Annex 1, references 149and 154). At its fifty-fifth meeting, the Committee concluded that D-Tagatose was not genotoxic, embryotoxic or teratogenic. It also concluded that an ADI could not be allocated for D-Tagatose because of concern about its potential to induce glycogen deposition and hypertrophy in the liver and to increase the concentrations of uric acid in serum. At its fifty-seventh meeting, the Committee evaluated the results of four studies in experimental animals, the results of a study in volunteers and some publications concerning the increased uric acid concentrations in serum after intake of D-Tagatose, other sugars, and other food components.
The Committee decided to base its evaluation on the human data reviewed in the course of the two meetings. A NOEL of 0.75 g/kg bw per day was identified from a 28-day study in which no effects were observed in humans receiving three doses of 15 g of D-Tagatose per day. An ADI of 0–80 mg/kg bw for D-Tagatose was established on the basis of this NOEL and a safety factor of 10.
At its present meeting, the Committee reviewed the results of two new studies of toxicity in rats, and of two new studies of plasma concentrations of uric acid in human volunteers.
A study was conducted to compare the effect of D-Tagatose on liver weights in six different strains of rat (Lewis, Fischer, Brown Norway, Lister Hooded, Sprague-Dawley and Wistar). Groups of 15 male rats of each strain were fed ad libitum for 28 days with diets modified by replacing 20% barley with either 20% D-Tagatose or 20% pregelatinized potato starch (control group). These concentrations were equal to an average dose of D-Tagatose of 11–13 g/kg bw per day. Body weights, food consumption and food conversion efficiency were recorded twice weekly throughout the study. Ten animals from each group were fasted overnight before necropsy. At necropsy, the livers were weighed, but no samples were taken for histological analysis. Consumption of D-Tagatose resulted in significantly increased relative liver weights in fasted rats of all strains (6–20% increase compared with the control group), and of the non-fasted rats of all strains except Wistar (8–25% increase compared with the control group). There was no statistical analysis of the effects of fasting, but percentage increases in fasted and un-fasted animals appeared to be similar. The authors assumed that increased liver weight was a result of the accumulation of glycogen, but data on glycogen content were not presented. The study was conducted essentially in accordance with the principles of good laboratory practice (GLP), but was not subject to quality assurance (QA) audit (Appel, 2002).
A modified study of carcinogenicity was conducted to investigate the long-term effects of D-Tagatose on the liver of Wistar rats. Diets were modified by replacing 20% barley with test substance and/or pregelatinized potato starch to produce concentrations of 2.5%, 5%, or 10% D-Tagatose, 20% fructose or 10% D-Tagatose plus 10% fructose. The control diet contained 20% pregelatinized potato starch. Groups of 50 male and 50 female rats were fed ad libitum with these diets for 24 months. The average doses to male rats were: D-Tagatose, 1.0, 2.0 and 4.0 g/kg bw per day; fructose, 7.6 g/kg bw per day; and D-Tagatose plus fructose, each at a dose of 4.0 g/kg bw per day. The average doses to female rats were: D-Tagatose, 1.2, 2.5 and 4.9 g/kg bw per day; fructose, 9.6 g/kg bw per day; and D-Tagatose plus fructose, 5.0 g and 4.9 g/kg bw per day, respectively. Animals were observed daily, body weights and food consumption were recorded weekly for the first 13 weeks and monthly thereafter. Ophthalmoscopic examinations were performed at 12 and 24 months. Blood was collected from the orbital sinuses for determination of haematology and clinical chemistry parameters at 6 months and 12 months, and from the abdominal aorta at final necropsy. All animals were fasted overnight before necropsy, which included macroscopic examination and weighing of the adrenals, liver, kidneys, testes and caecum. Livers were sectioned, stained with haematoxylin and eosin and subjected to histopathological examination.
About 70% of the animals survived to scheduled necropsy, and the rate of mortality was not affected by treatment. Mean body weights were decreased in males and females fed 10% D-Tagatose, with or without 10% fructose, throughout the study and in males and females fed 5% D-Tagatose for most of the study. Mean body weights of treated animals were generally >90% of those of the control group; less difference in terminal body weights was recorded after overnight fasting. Food intake was slightly decreased during the first 4 weeks of the study, after which it did not differ significantly from that of the control group. Food conversion efficiency was not affected by treatment. Haemoglobin concentration and erythrocyte volume fraction were decreased in females fed 10% D-Tagatose, with or without 10% fructose. In females, absolute and relative adrenal weights were increased at all doses of D-Tagatose, but not in animals receiving fructose only. The effect was not clearly dose-related, with increases in relative adrenal weights of 28%, 72%, 39% and 33% in rats fed diets containing 2.5%, 5%, 10% D-Tagatose, and 10% D-Tagatose plus 10% fructose, respectively. In general, changes in organ weights are considered to be possibly of toxicological relevance if >10%, even if a dose-response relationship is not evident. In addition, absolute and relative kidney weights were increased in animals receiving 10% D-Tagatose with or without fructose, and in animals receiving 10% fructose. Relative kidney weights were also increased in rats fed diets containing 5% D-Tagatose. There were also increases in relative (and in most instances, absolute) liver weights and full and empty caecal weights in rats fed diets containing 10% D-Tagatose, with or without fructose. In males, relative, but not absolute, adrenal weights were increased in rats receiving 5 and 10% D-Tagatose, with or without 10% fructose. Relative and absolute full and empty caecal weights were increased in rats fed 5 and 10% D-Tagatose. Relative, but not absolute, testes weights were increased in rats receiving 10% D-Tagatose, with or without 10% fructose and relative, but not absolute, liver weights were increased in rats fed a combination of 10% D-Tagatose and 10% fructose. Organ weights were not increased in male rats receiving fructose only. Macroscopic examination revealed a relatively high incidence of enlargement of the adrenals in some treatment groups, which was generally consistent with the increased adrenal weights. The number of females with uterine nodules appeared to be higher in the group receiving 10% D-Tagatose plus 10% fructose. There were no histopathological changes related to treatment with D-Tagatose and the incidence of liver tumours in all groups was low. Adrenals were not examined histologically. The authors concluded that administration of D-Tagatose at up to 10% in the diet (equal to approximately 4 and 5 g/kg bw per day for males and females respectively) did not have any adverse long-term effects on the liver of Wistar rats. A QA statement was included (Lina et al., 2002).
Six healthy male volunteers (aged 22–24 years) were given a light breakfast consisting of unsweetened herbal tea, bread and marmalade containing 30 g D-Tagatose. Blood samples were taken over a period of 4 h, a 24-h urine sample was collected and the subjects were asked if they experienced any side-effects. There was no control group of volunteers receiving the same meal without D-Tagatose. Plasma glucose and uric acid concentrations increased after the meal, and plasma phosphate concentrations decreased slightly. All concentrations were cited to be within the normal range. Urinary excretion of uric acid was also within the normal range. Gastrointestinal side-effects were not reported during the study or the 24-h post-treatment observation period (Diamantis & Bär, 2001).
In a follow-up study, 12 hyperuricaemic male volunteers (aged 57–64 years) were given a light breakfast consisting of unsweetened herbal tea, bread and marmalade containing 15 g D-Tagatose. Blood samples were taken over a period of 4 h, urine was collected for 4 h and a further 24 h. The men were asked if they experienced any side-effects and a clinical examination was performed. There was no control group of volunteers receiving the same meal without D-Tagatose. Plasma concentrations of glucose and uric acid increased after the meal, the maximum increase in uric acid being about 2.5% of the baseline value obtained before the meal was consumed. The plasma concentration of phosphate was significantly increased after 4 h, but not at other sampling times. Plasma concentration of lactate did not vary. Urinary concentrations of uric acid were reported to be above the normal range for the institution for 3/12 men, but excretion of uric acid, normalized to creatinine excretion, did not vary between the post-prandial 4-h period and the subsequent 24-h period. Gastrointestinal side-effects were not reported during the study or the 24-h post-treatment observation period (Diamantis & Bär, 2002).
Studies of D-Tagatose administered to rats in the diet, reviewed previously by the Committee, focused on the hepatic effects of D-Tagatose, in particular, increased liver weight and hypertrophy. These studies indicated that these effects were caused, at least in part, by glycogen accumulation, and that Sprague-Dawley rats were more sensitive to these effects than Wistar rats. The new 28-day study investigating the effects of 20% D-Tagatose in the diet has shown that, of six rat strains, the largest increase in liver weight occurred in Sprague-Dawley rats, and the smallest increase occurred in Wistar rats, confirming the previous observation of strain differences. The role of glycogen, however, was not specifically investigated.
In a 2-year study in Wistar rats, the administration of diets containing 2.5, 5 or 10% D-Tagatose, 20% fructose, or 10% D-Tagatose plus 10% fructose did not result in histological changes in the liver, although increased liver weights were reported in male and female rats fed 10% D-Tagatose. Increased absolute and relative adrenal weights were observed in female rats at all doses of D-Tagatose, but not in those receiving fructose alone. Increased adrenal weights were also reported in male rats fed with 5% and 10% D-Tagatose. The weights of the kidneys in females, the testes in males, and the caecum in each sex were also increased in animals fed 10% D-Tagatose, and in some cases, 5% D-Tagatose. In the absence of histopathological confirmation of the nature of the changes induced by D-Tagatose in the adrenals, kidneys and testes, it is not possible to assess their toxicological significance to humans.
Two new studies in humans have shown that a single dose of 30 g D-Tagatose to small numbers of healthy volunteers, or 15 g D-Tagatose to hyperuricaemic individuals, had no biologically significant effect on uric acid production or excretion, and no recorded gastrointestinal effects. At its forty-eighth meeting, the Committee noted that D-fructose increases uric acid production by accelerating the degradation of purine nucleotides, probably by hepatocellular depletion of inorganic phosphate resulting from accumulation of ketohexose-1-phosphate. The degradation of D-Tagatose-1-phosphate is slower than that of D-fructose-1-phosphate, and therefore the hyperuricaemic effect of D-Tagatose may be greater than that of D-fructose; hyperuricaemic individuals are therefore potentially vulnerable to the adverse effects of D-Tagatose. The new study demonstrated no increase in serum concentrations of uric acid within 4 h of consumption of 15 g of D-Tagatose by this vulnerable group. In studies reviewed previously by the Committee, the maximum increases in serum uric acid and D-Tagatose and the maximum decrease in serum ATP were seen within 1 h of ingesting D-Tagatose. It is therefore anticipated that no effect would be observed in hyperuricaemic individuals after repeated consumption of 15 g of D-Tagatose at subsequent meals.
The Committee concluded that the results of the 2-year study in rats established that the previously-reported liver glycogen deposition and hypertrophy observed after long-term administration of D-Tagatose did not result in histopathological changes, and thus addressed concerns expressed at the fifty-fifth meeting. This study, however, also identified new effects, namely increased adrenal, kidney and testes weights. The Committee considered that these changes might have been caused by high osmotic load resulting from the high dietary doses administered, but this could not be confirmed in the absence of histopathological examination of these tissues. Pending provision of the histopathology data, the Committee confirmed that the human data provided the most relevant basis for assessing the acceptable intake of D-Tagatose.
At its fifty-seventh meeting, the Committee identified a NOEL for healthy individuals of 45 g of D-Tagatose per day in three divided doses. The study on hyperuricaemic individuals discussed at the current meeting indicated that the NOEL is also applicable to this vulnerable group. The Committee considered that a safety factor of 3 would be appropriate to allow for interindividual variation. In view of the additional uncertainty regarding the nature of the effects observed in the adrenals, kidneys and testes in the 2-year study in rats, the Committee concluded that the ADI should be temporary and applied an additional safety factor of 2. The previous ADI was thus removed, and on the basis of the NOEL of 0.75 g/kg bw per day, and a safety factor of 6, the Committee allocated a temporary ADI for D-Tagatose of 0–125 mg/kg bw.
The temporary ADI does not apply to individuals with hereditary fructose intolerance caused by deficiency in 1-phosphofructoaldolase (aldolase B) or fructose 1,6-diphosphatase. The Committee requested information on the histological examination of the adrenals, kidneys and testes of the rats from the 2-year study, by 2006.
The intake assessment prepared by the Committee at its fifty-seventh meeting is still valid.
Appel, M.J. (2002) A 28-day comparative study with D-Tagatose in male rats of 6 different strains. Unpublished report V4252 from TNO Nutrition and Food Research, Zeist, the Netherlands. Submitted to WHO by Bioresco Ltd., Basel, Switzerland.
Diamantis, I. & Bär, A. (2001) Effect of an oral 30g dose of D-Tagatose on the plasma uric acid levels of healthy male volunteers. Unpublished study report from University of Crete, Heraklion, Greece. Submitted to WHO by Bioresco Ltd, Basel, Switzerland.
Diamantis, I. & Bär, A. (2002) Effect of an oral 15 g dose of D-Tagatose on the plasma uric acid levels of hyperuricemic male volunteers. Unpublished study report from University of Crete, Heraklion, Greece. Submitted to WHO by Bioresco Ltd, Basel, Switzerland.
Lina, B.A.R. & Kuper, C.F. (2002) Chronic toxicity and carcinogenicity study with D-Tagatose and fructose in Wistar rats. Unpublished report V4533 from TNO Nutrition and Food Research, Zeist, Netherlands. Submitted to WHO by Bioresco Ltd, Basel, Switzerland.
See Also: Toxicological Abbreviations D-TAGATOSE (JECFA Evaluation)