L-Sorbose but Not D-Tagatose Induces Hemolysis of Dog Erythrocytes in Vitro A. Ba ¨r* ,1 and W. R. Leeman† *Bioresco AG, Hauptstrasse 63, 4102 Binningen, Switzerland; and TNO–Nutrition and Food Research Institute, P.O. Box 360, 3700 AJ Zeist, The Netherlands Received September 3, 1998 Previous investigations have demonstrated that L- sorbose induces hemolysis of dog erythrocytes. This effect is probably the consequence of an ATP depletion of the red blood cells subsequent to inhibition of hexokinase, and thus the glycolytic pathway, by sor- bose 1-phosphate. In the present study, the suscepti- bility of dog erythrocytes to D-tagatose, a stereoisomer of L-sorbose, was examined. Washed dog erythrocytes were suspended in Hanks’ balanced salt solution (HBSS, containing 5.6 mM glucose) with or without the addition of 0.6, 6, and 60 mM L-sorbose or D-tagatose, or in HBSS with total glucose concentrations of 5.6, 6 and 60 mM D-glucose. After incubation for 24 h at 34°C, the suspensions were centrifuged, and the percentage of hemolysis was determined by measuring the hemoglo- bin in the sediment and the supernatant. The amount of hemoglobin released in the medium did not differ significantly between the control (HBSS) and the test incubations with glucose or D-tagatose supplementa- tion. In contrast, the addition of 6 and 60 mM L-sorbose resulted in significant hemolysis. At the low dose (0.6 mM), L-sorbose did not have an adverse effect. It is concluded that D-tagatose, unlike L-sorbose, does not have a hemolytic effect on canine erythrocytes. © 1999 Academic Press INTRODUCTION L-Sorbose and D-tagatose are stereoisomers of D-fruc- tose. L-Sorbose was proposed for use as an energy- reduced and noncariogenic sugar substitute more than 20 years ago (Mu ¨ hlemann, 1976). However, in the course of the toxicological examination of L-sorbose, this sugar was found to cause lysis of dog erythrocytes in vitro and in vivo (Kistler and Keller, 1977; Keller and Kistler, 1978; Zu ¨ rrer et al., 1978). Erythrocytes from other species, including humans, appeared not to be affected (Kistler and Keller, 1977; Zu ¨ rrer et al., 1978). Subsequent studies on the potential biochemical mechanism of this effect revealed that L-sorbose en- tered the red blood cells, where it was phosphorylated by fructokinase to L-sorbose 1-phosphate (Ba ¨ r, 1978). This metabolite is a known inhibitor of hexokinase, while sorbose 6-phosphate has no inhibitory effect (Lardy et al., 1950). Since L-sorbose 1-phosphate is cleaved only slowly by aldolase, its intracellular con- centration apparently reached levels at which the glycolytic pathway was blocked. Indeed, the lactate formation by dog erythrocytes was significantly sup- pressed when L-sorbose was present (Kistler and Keller, 1978; Ba ¨ r, 1978). Inhibition of lactate formation from glucose and hemolysis of dog erythrocytes was also obtained when L-glyceraldehyde was added (Ba ¨r, 1978). This triose is condensed in a reverse aldolase reaction with dihydroxyacetone phosphate to yield sor- bose 1-phosphate (Lardy et al., 1950). The hemolysis of dog erythrocytes by L-sorbose was apparently the result of an ATP depletion following the shutdown of the glycolytic pathway. This hypothesis was supported by the fact that ATP depletion due to unavailability of substrate also induced hemolysis of dog erythrocytes, and the fact that sorbose did not cause hemolysis when galactose (rather than glucose) was used as the energy-delivering substrate (hexoki- nase is not involved in the metabolism of galactose) (Ba ¨ r, 1978). It is interesting that ATP depletion of human erythrocytes by starvation did not result in cell lysis, but in a release of spectrin-free vesicles. This difference may explain the lack of an adverse effect of L-sorbose on human erythrocytes (Lutz et al., 1977; Feo and Mohandas, 1977). Another between-species differ- ence may lie in different activities of fructokinase in the erythrocytes (Wolf, 1963). Yet, while sorbose did not seem to impair the functionality of human eryth- rocytes, the possibility that the release of spectrin-free vesicles may entail a decreased deformability and a reduced survival time of the cells remains (Marikovsky et al., 1977). D-Tagatose has been proposed more recently for use as a sugar substitute (Levin et al., 1995). The metabo- lism of absorbed D-tagatose resembles that of L-sorbose. 1 To whom correspondence should be addressed. 0273-2300/99 $30.00 S43 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved. Regulatory Toxicology and Pharmacology 29, S43–S45 (1999) Article ID rtph.1998.1254, available online at http://www.idealibrary.com on