J. Inher. Metab. Dis. 13 (1990) 487-500 © SSIEM and Kluwer AcademicPublishers. Printed in the Netherlands
Regulation of Galactose Metabolism: Implications for Therapy S. SEGAL and S. ROGERS Division of Biochemical Development and Molecular Diseases, Children's Hospital of Philadelphia and the Department of Pediatrics and Medicine, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania, USA
Summary: In view of evidence that dietary therapy of galactose-l-phosphate uridyltransferase deficiency has failed to prevent complications of the disorder, there is a need for new strategies in treatment. The enhancement of residual enzyme activity in tissues of galactosaemic patients should provide such an approach. This possibility is derived from knowledge of the regulation of transferase activity in normal animal tissues. The pertinent observations summarized herein are: (1) that hepatic transferase activity is modulated by various cellular metabolites, uridine nucleotides being of particular significance; (2) that transferase activity in the young rat liver is subject to developmental programming with a several-fold increase after birth; (3) that transferase activity in pregnant rat liver is significantly increased which may be related to hormonal effects of progesterone; and (4) that pharmacological doses of folic acid may increase transferase activity. The basis of such regulation can give insight into sufficient augmentation of the residual activity to increase gatactose utilization and thereby better the long-term outcome.
Galactose restriction has been considered a satisfactory basis for therapy ever since Mason and Turner (1935) described how removing galactose from the diet eliminated the acute galactose toxicity syndrome in a patient with galactose-l-phosphate uridyltransferase deficient galactosaemia. Recent evidence, however, has indicated that there are 'clouds over galactosaemia' (Lancet, Editorial, 1982). Despite early diagnosis and the institution of a galactose-free diet there appears to be an inability to prevent some degree of mental retardation (Fishler et al., 1980; Komrower, 1982) or ovarian failure (Kaufman et al., 1981; Steinmann et aI., 1981), and there have been reports of neurological ataxia syndromes appearing in well-treated older patients (Lo et al., 1984; Friedman et al., 1989). A recent survey of over 300 patients indicates that there is developmental delay, speech abnormality, ovarian dysfunction and growth retardation in a large number of patients, independent of the time after birth of the institution of dietary restriction (Buist et aI., 1988). The reason for the inefficacy of galactose restriction is at present obscure. 487
488
Segal and Rogers
There are two prevalent theories for the poor outcome in this disease. The first of these, espoused by Gitzelmann (Gitzelmann, 1969; Gitzelmann and Hansen, 1974; Gitzelmann et al., 1975) is that there is continuous self-intoxication due to the endogenous production of galactose-1-phosphate from uridine diphosphate galactose. The second one, recently proposed by Ng and colleagues (1987), is based on the observation (Shin et aI., 1985) that there is a depletion of cellular UDPgalactose and postulates a resulting impairment of the synthesis of critical complex substances containing galactose. Figure 1 shows the Leloir pathway where ordinarily exogenous galactose is phosphorylated by ATP, the resulting galactose-l-phosphate (gal-l-P) interacting with UDPglucose by the activity of transferase to form UDPgalactose, which undergoes epimerization to UDPglucose. The sugar in the latter then enters the glucose pathway by a cleavage to glucose-1-phosphate. With a block in transferase, ingestion of galactose is associated with accumulation of gal-l-P and, via alternate paths, galactitol and galactonate. With elimination of dietary galactose one would expect no gal-l-P accumulation but this is not the case, with high cellular levels being observed in the best treated patients. Gitzelmann (1969) proposed that gal-1P is formed via pyrophosphorylitic cleavage of UDPgalactose (indicated by PP in Figure i) which is derived by epimerization of ever present UDPglucose readily formed from UTP and glucose-l-phosphate. The gal-l-P is thought to be related to the continuing toxicity (Komrower, 1982; Gitzelmann and Steinmann, 1984). The low level of UDPgalactose in galactose-1-phosphate uridyltransferase deficient tissues has led Ng and colleagues (Kaufman et aI., 1989) to believe that UDPgalactose depletion results in decreased glycoprotein and galactolipid synthesis, which is the basis for long-term complications. Why cell UDPgalactose is low is not clear since it can be formed from UDPglucose. With failure of dietary therapy it appears there is a need for new strategies in the treatment of the galactose-l-phosphate uridyltransferase deficient disorder. One approach is possible enhancement of residual enzyme activity. This would result in
GALACTOSE METABOLIC PATHWAY Galactonate ~
Galactose
~ Galactitol
It
Galactose-l-P UDPgalactose..~ Gylc°lipids ~t
~'~Glycoproteins
UDPglucose
:A T: Glucose
Figure 1 Reactions of the pathway of galactose metabolism responsible ~r the galactoseglucose interconversion J. lnher. Metab. Dis. 13 (1990)
Regulation o f Galactose Metabolism
489
an increased flux through the pathway with a decrease in gal-l-P and an increase in UDPgalactose, thus satisfying the basis of toxicity of both prevalent theories. From our own studies of 14C-galactose oxidation by galactosaemic patients, we know that there is residual metabolic capacity (Segal et al., 1965; Segal and Cuatrecasas, 1968). Most patients produce galactose-l-phosphate uridyltransferase protein (Tedesco and Mellman, 197t) and some have small amounts of red cell galactose-l-phosphate uridyltransferase activity (Kaufman et al., 1988). Recent studies in our laboratory have shown detectable galactose-1-phosphate uridyltransferase in Percoll-fractionated reticulocytes when whole blood has no assayable activity (Kelley et al., 1989). Galactosaemic fibroblasts possess considerable galactose-l-phosphate uridyltransferase activity soon after subculture which diminishes with confluency (Russell and DeMars, 1967). REGULATION OF GALACTOSE-I-PHOSPHATE URIDYLTRANSFERASE Described below are our studies on the regulation of normal transferase activity in rat liver with the idea that knowledge of how transferase is regulated can give us an insight into the possible augmentation of the residual enzyme activity in the deficient patient. There are four aspects of the regulation of galactose metabolism which should be emphasized: the metabolic regulation of enzyme activity; developmental programming; pregnancy and progesterone effects; and pharmacological effects, notably with folic acid.
The metabolic regulation of enzyme activity That there is metabolic regulation of transferase activity became evident when we observed that normal suckling rat liver perfused with galactose resulted in an increase in the specific activity of transferase within 30 rain (Rogers and Segal, 1981). This and the previous findings that diet could regulate galactose metabolizing enzymes in intestine (Stifel et aI., 1968) led us to study the effect of a high galactose diet on the Leloir pathway enzymes in young adult rat livers (Rogers et at., 1989a). As shown in Figure 2, there is a marked and persistent increase in transferase, a transient increase in epimerase and little alteration in galactokinase after six days of feeding a 40% galactose diet. A physiological counterpart to this study is shown in Table 1. In hepatocytes isolated from livers of rats on the experimental diet there is an increased uptake of galactose, a higher conversion to glucose and an enhanced conversion to lactate. This suggests that transferase is indeed capable of manipulation by metabolic influence. Table 2 summarizes the regulation of galactose-metabolizing enzymes. Galactokinase is inhibited by both its substrate and its product (Cuatrecasas and Segal, 1965), which would indicate that there is a decrease in galactose phosphorylation when a block exists in transferase or epimerase activity. More importantly, transferase is also inhibited by high levels of its substrates, gal-l-P and UDPglucose (Bertoli and Segal, 1966). Glucose-l-phosphate, a product, is perhaps one of the most potent metabolic inhibitors. Of particular significance is the fact that uridine nucleotides, J. Inher. Metab. Dis. 13 (1990)
Segal and Rogers
490 Effect of 40% Galactose Diet on Female Rat Liver
8
%
4
E
x
-# 2 2
~C ~o ,,18 -6 E
5
3
"15
Regulation of Galactose Metabolism: Implications for Therapy S. SEGAL and S. ROGERS Division of Biochemical Development and Molecular Diseases, Children's Hospital of Philadelphia and the Department of Pediatrics and Medicine, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania, USA
Summary: In view of evidence that dietary therapy of galactose-l-phosphate uridyltransferase deficiency has failed to prevent complications of the disorder, there is a need for new strategies in treatment. The enhancement of residual enzyme activity in tissues of galactosaemic patients should provide such an approach. This possibility is derived from knowledge of the regulation of transferase activity in normal animal tissues. The pertinent observations summarized herein are: (1) that hepatic transferase activity is modulated by various cellular metabolites, uridine nucleotides being of particular significance; (2) that transferase activity in the young rat liver is subject to developmental programming with a several-fold increase after birth; (3) that transferase activity in pregnant rat liver is significantly increased which may be related to hormonal effects of progesterone; and (4) that pharmacological doses of folic acid may increase transferase activity. The basis of such regulation can give insight into sufficient augmentation of the residual activity to increase gatactose utilization and thereby better the long-term outcome.
Galactose restriction has been considered a satisfactory basis for therapy ever since Mason and Turner (1935) described how removing galactose from the diet eliminated the acute galactose toxicity syndrome in a patient with galactose-l-phosphate uridyltransferase deficient galactosaemia. Recent evidence, however, has indicated that there are 'clouds over galactosaemia' (Lancet, Editorial, 1982). Despite early diagnosis and the institution of a galactose-free diet there appears to be an inability to prevent some degree of mental retardation (Fishler et al., 1980; Komrower, 1982) or ovarian failure (Kaufman et al., 1981; Steinmann et aI., 1981), and there have been reports of neurological ataxia syndromes appearing in well-treated older patients (Lo et al., 1984; Friedman et al., 1989). A recent survey of over 300 patients indicates that there is developmental delay, speech abnormality, ovarian dysfunction and growth retardation in a large number of patients, independent of the time after birth of the institution of dietary restriction (Buist et aI., 1988). The reason for the inefficacy of galactose restriction is at present obscure. 487
488
Segal and Rogers
There are two prevalent theories for the poor outcome in this disease. The first of these, espoused by Gitzelmann (Gitzelmann, 1969; Gitzelmann and Hansen, 1974; Gitzelmann et al., 1975) is that there is continuous self-intoxication due to the endogenous production of galactose-1-phosphate from uridine diphosphate galactose. The second one, recently proposed by Ng and colleagues (1987), is based on the observation (Shin et aI., 1985) that there is a depletion of cellular UDPgalactose and postulates a resulting impairment of the synthesis of critical complex substances containing galactose. Figure 1 shows the Leloir pathway where ordinarily exogenous galactose is phosphorylated by ATP, the resulting galactose-l-phosphate (gal-l-P) interacting with UDPglucose by the activity of transferase to form UDPgalactose, which undergoes epimerization to UDPglucose. The sugar in the latter then enters the glucose pathway by a cleavage to glucose-1-phosphate. With a block in transferase, ingestion of galactose is associated with accumulation of gal-l-P and, via alternate paths, galactitol and galactonate. With elimination of dietary galactose one would expect no gal-l-P accumulation but this is not the case, with high cellular levels being observed in the best treated patients. Gitzelmann (1969) proposed that gal-1P is formed via pyrophosphorylitic cleavage of UDPgalactose (indicated by PP in Figure i) which is derived by epimerization of ever present UDPglucose readily formed from UTP and glucose-l-phosphate. The gal-l-P is thought to be related to the continuing toxicity (Komrower, 1982; Gitzelmann and Steinmann, 1984). The low level of UDPgalactose in galactose-1-phosphate uridyltransferase deficient tissues has led Ng and colleagues (Kaufman et aI., 1989) to believe that UDPgalactose depletion results in decreased glycoprotein and galactolipid synthesis, which is the basis for long-term complications. Why cell UDPgalactose is low is not clear since it can be formed from UDPglucose. With failure of dietary therapy it appears there is a need for new strategies in the treatment of the galactose-l-phosphate uridyltransferase deficient disorder. One approach is possible enhancement of residual enzyme activity. This would result in
GALACTOSE METABOLIC PATHWAY Galactonate ~
Galactose
~ Galactitol
It
Galactose-l-P UDPgalactose..~ Gylc°lipids ~t
~'~Glycoproteins
UDPglucose
:A T: Glucose
Figure 1 Reactions of the pathway of galactose metabolism responsible ~r the galactoseglucose interconversion J. lnher. Metab. Dis. 13 (1990)
Regulation o f Galactose Metabolism
489
an increased flux through the pathway with a decrease in gal-l-P and an increase in UDPgalactose, thus satisfying the basis of toxicity of both prevalent theories. From our own studies of 14C-galactose oxidation by galactosaemic patients, we know that there is residual metabolic capacity (Segal et al., 1965; Segal and Cuatrecasas, 1968). Most patients produce galactose-l-phosphate uridyltransferase protein (Tedesco and Mellman, 197t) and some have small amounts of red cell galactose-l-phosphate uridyltransferase activity (Kaufman et al., 1988). Recent studies in our laboratory have shown detectable galactose-1-phosphate uridyltransferase in Percoll-fractionated reticulocytes when whole blood has no assayable activity (Kelley et al., 1989). Galactosaemic fibroblasts possess considerable galactose-l-phosphate uridyltransferase activity soon after subculture which diminishes with confluency (Russell and DeMars, 1967). REGULATION OF GALACTOSE-I-PHOSPHATE URIDYLTRANSFERASE Described below are our studies on the regulation of normal transferase activity in rat liver with the idea that knowledge of how transferase is regulated can give us an insight into the possible augmentation of the residual enzyme activity in the deficient patient. There are four aspects of the regulation of galactose metabolism which should be emphasized: the metabolic regulation of enzyme activity; developmental programming; pregnancy and progesterone effects; and pharmacological effects, notably with folic acid.
The metabolic regulation of enzyme activity That there is metabolic regulation of transferase activity became evident when we observed that normal suckling rat liver perfused with galactose resulted in an increase in the specific activity of transferase within 30 rain (Rogers and Segal, 1981). This and the previous findings that diet could regulate galactose metabolizing enzymes in intestine (Stifel et aI., 1968) led us to study the effect of a high galactose diet on the Leloir pathway enzymes in young adult rat livers (Rogers et at., 1989a). As shown in Figure 2, there is a marked and persistent increase in transferase, a transient increase in epimerase and little alteration in galactokinase after six days of feeding a 40% galactose diet. A physiological counterpart to this study is shown in Table 1. In hepatocytes isolated from livers of rats on the experimental diet there is an increased uptake of galactose, a higher conversion to glucose and an enhanced conversion to lactate. This suggests that transferase is indeed capable of manipulation by metabolic influence. Table 2 summarizes the regulation of galactose-metabolizing enzymes. Galactokinase is inhibited by both its substrate and its product (Cuatrecasas and Segal, 1965), which would indicate that there is a decrease in galactose phosphorylation when a block exists in transferase or epimerase activity. More importantly, transferase is also inhibited by high levels of its substrates, gal-l-P and UDPglucose (Bertoli and Segal, 1966). Glucose-l-phosphate, a product, is perhaps one of the most potent metabolic inhibitors. Of particular significance is the fact that uridine nucleotides, J. Inher. Metab. Dis. 13 (1990)
Segal and Rogers
490 Effect of 40% Galactose Diet on Female Rat Liver
8
%
4
E
x
-# 2 2
~C ~o ,,18 -6 E
5
3
"15
