Intermediary metabolism of fructose3€¦ · reaching consequences to carbohydrate and lipid...

12
754S Am J C/in Nutr 1993;58(suppl):754S-65S. Printed in USA. © 1993 American Society for Clinical Nutrition Intermediary metabolism of fructose3 Peter A Mayes ABSTRACT Most of the metabolic effects of fructose arc due to its rapid utilization by the liver and it by-passing the phos- phofructokinase regulatory step in glycolysis, leading to far reaching consequences to carbohydrate and lipid metabolism. These consequences include immediate hepatic increases in py- ruvate and lactate production, activation of pyruvate dehydro- genase, and a shift in balance from oxidation to esterification of nonesterified fatty acids, resulting in increased secretion of very- low-density-lipoprotein (VLDL). These effects are augmented by long-term absorption of fructose, which causes enzyme adapta- tions that increase lipogenesis and VLDL secretion, leading to triglyceridemia, decreased glucose tolerance, and hyperinsuli- nemia. Acute loading of the liver with fructose causes sequestra- tion of inorganic phosphate in fructose-l-phosphate and dimin- ished ATP synthesis. Consequently, the inhibition by Al? of the enzymes of adenine nucleotide degradation is removed and uric acid formation accelerates with consequent hyperuricemia. These effects are of particular significance to potentially hypertrigly- ceridemic or hyperuricemic individuals. Am J Clin Nutr 1993;58(suppl):754S-765S. KEY WORDS Fructose, intermediary metabolism, sucrose, immediate effects, long-term effects, liver, carbohydrate metab- olism, lipid metabolism, enzyme adaptation, lipogenesis, very- low-density lipoproteins, insulin, nonesterified fatty acids, hy- pertriglyceridemia, hyperuricemia Introduction Virtually all the unique metabolic properties of fructose are due to two primary factors: its rapid uptake by the liver, and its entry to the pathway of glycolysis or gluconeogenesis at the tn- ose phosphate level after bypassing the phosphofructokinase reg- ulatory step. The metabolic consequence is the provision of in- creased substrate in all the metabolic pathways leading from tn- ose phosphate (Fig 1). Current interest in fructose metabolism has arisen because of its increased use as a sweetener in the food industry in the form of high-fructose corn syrup (HFS). Also, whether based on sound science or not, fructose has been promoted as an appetite de- pressant, as a food for non-insulin-dependent diabetics, for par- enteral feeding, and as a food supplement for endurance athletes. Representing half of the sucrose molecule, fructose has been red- ognized for many years as being largely responsible for the met- abolic effects of high-sucrose diets. Concern has arisen because of the realization that fructose, at elevated concentrations, can promote metabolic changes that are actually or potentially dele- terious, eg, hyperlipidemia, hyperuricemia, nonenzymic fructo- sylation of proteins, lactacidemia, and disturbances in copper me- tabolism. The paper concentrates on the general metabolism of fructose, particularly concerning its impact on carbohydrate, lipid and punine metabolism. Other aspects referred to are the subjects of other reviews in this volume. Of further consideration is whether augmented insulin secretion is responsible for some of the metabolic effects of fructose intake. Compared with glucose, fructose feeding does not directly cause an increase in plasma insulin (1) whereas sucrose feeding does (2). Thus whenever intake of fructose is accompanied by glucose, the added effect of insulin secretion must be taken into account. Not only is this the case with sucrose itself but some HFSs are equiv- alent to hydrolyzed sucrose because they are - 50% glucose and 50% fructose (3). Similar considerations apply to supplements for athletes that contain both fructose and glucose. General metabolism Utilization of blood fructose and uptake into tissues The consequence of the digestion of sucrose and other fructose- containing foods such as honey, fruits, and some vegetables, is absorption and transport of fructose by the intestinal epithelium into the hepatic portal vein. Therefore, all fructose absorbed flows through the liver initially. Because of the presence of an active hepatic enzyme system for metabolizing fructose, fructose readily passes into the liver, accounting for a fractional uptake of 55% and 71%, respectively, of the fructose presented to the liver after tube feeding fed or starved rats (4). In humans it was shown that the liver metabolized at least half of the fructose injected intrave- nously (5). In the perfused rat liver we found a value of 40% for the fractional extraction of fructose (6). As a consequence of the high rate of extraction of fructose by the liver, correspondingly low concentrations of fructose are found in the systemic blood vessels after meals containing fructose or sucrose are consumed (4, 7, 8). Some 20% of fructose administered intravenously is taken up by the kidney (9). Thus, somewhat less than this amount would be expected to be taken up by this organ after oral feeding where the liver takes up some 50% of the initial influx. A considerably smaller fraction would be available for adipose tissue (10) and I From the Division of Biochemistry, Department of Veterinary Basic Sciences, Royal Veterinary College, University of London. 2 Supported by the British Heart Foundation. 3 Address reprint requests to PA Mayes, Division of Biochemistry, Department of Veterinary Basic Sciences, Royal Veterinary College, University of London, London NW1 0Th, UK. by guest on July 14, 2014 ajcn.nutrition.org Downloaded from

Transcript of Intermediary metabolism of fructose3€¦ · reaching consequences to carbohydrate and lipid...

754S Am J C/in Nutr 1993;58(suppl):754S-65S. Printed in USA. © 1993 American Society for Clinical Nutrition

Intermediary metabolism of fructose�3

Peter A Mayes

ABSTRACT Most of the metabolic effects of fructose arc

due to its rapid utilization by the liver and it by-passing the phos-

phofructokinase regulatory step in glycolysis, leading to far

reaching consequences to carbohydrate and lipid metabolism.

These consequences include immediate hepatic increases in py-

ruvate and lactate production, activation of pyruvate dehydro-

genase, and a shift in balance from oxidation to esterification of

nonesterified fatty acids, resulting in increased secretion of very-

low-density-lipoprotein (VLDL). These effects are augmented by

long-term absorption of fructose, which causes enzyme adapta-

tions that increase lipogenesis and VLDL secretion, leading to

triglyceridemia, decreased glucose tolerance, and hyperinsuli-

nemia. Acute loading of the liver with fructose causes sequestra-

tion of inorganic phosphate in fructose-l-phosphate and dimin-

ished ATP synthesis. Consequently, the inhibition by Al? of the

enzymes of adenine nucleotide degradation is removed and uric

acid formation accelerates with consequent hyperuricemia. These

effects are of particular significance to potentially hypertrigly-

ceridemic or hyperuricemic individuals. Am J Clin Nutr1993;58(suppl):754S-765S.

KEY WORDS Fructose, intermediary metabolism, sucrose,

immediate effects, long-term effects, liver, carbohydrate metab-

olism, lipid metabolism, enzyme adaptation, lipogenesis, very-

low-density lipoproteins, insulin, nonesterified fatty acids, hy-

pertriglyceridemia, hyperuricemia

Introduction

Virtually all the unique metabolic properties of fructose are

due to two primary factors: its rapid uptake by the liver, and its

entry to the pathway of glycolysis or gluconeogenesis at the tn-

ose phosphate level after bypassing the phosphofructokinase reg-

ulatory step. The metabolic consequence is the provision of in-

creased substrate in all the metabolic pathways leading from tn-

ose phosphate (Fig 1).Current interest in fructose metabolism has arisen because of

its increased use as a sweetener in the food industry in the form

of high-fructose corn syrup (HFS). Also, whether based on sound

science or not, fructose has been promoted as an appetite de-

pressant, as a food for non-insulin-dependent diabetics, for par-

enteral feeding, and as a food supplement for endurance athletes.

Representing half of the sucrose molecule, fructose has been red-

ognized for many years as being largely responsible for the met-

abolic effects of high-sucrose diets. Concern has arisen because

of the realization that fructose, at elevated concentrations, can

promote metabolic changes that are actually or potentially dele-

terious, eg, hyperlipidemia, hyperuricemia, nonenzymic fructo-

sylation of proteins, lactacidemia, and disturbances in copper me-

tabolism. The paper concentrates on the general metabolism of

fructose, particularly concerning its impact on carbohydrate, lipid

and punine metabolism. Other aspects referred to are the subjects

of other reviews in this volume.

Of further consideration is whether augmented insulin secretion

is responsible for some of the metabolic effects of fructose intake.

Compared with glucose, fructose feeding does not directly cause

an increase in plasma insulin (1) whereas sucrose feeding does (2).

Thus whenever intake of fructose is accompanied by glucose, the

added effect of insulin secretion must be taken into account. Not

only is this the case with sucrose itself but some HFSs are equiv-

alent to hydrolyzed sucrose because they are - 50% glucose and50% fructose (3). Similar considerations apply to supplements for

athletes that contain both fructose and glucose.

General metabolism

Utilization of blood fructose and uptake into tissues

The consequence of the digestion of sucrose and other fructose-

containing foods such as honey, fruits, and some vegetables, is

absorption and transport of fructose by the intestinal epithelium

into the hepatic portal vein. Therefore, all fructose absorbed flows

through the liver initially. Because of the presence of an active

hepatic enzyme system for metabolizing fructose, fructose readily

passes into the liver, accounting for a fractional uptake of 55%

and 71%, respectively, of the fructose presented to the liver after

tube feeding fed or starved rats (4). In humans it was shown that

the liver metabolized at least half of the fructose injected intrave-

nously (5). In the perfused rat liver we found a value of 40% for

the fractional extraction of fructose (6). As a consequence of the

high rate of extraction of fructose by the liver, correspondingly

low concentrations of fructose are found in the systemic blood

vessels after meals containing fructose or sucrose are consumed

(4, 7, 8). Some 20% of fructose administered intravenously is taken

up by the kidney (9). Thus, somewhat less than this amount would

be expected to be taken up by this organ after oral feeding where

the liver takes up some 50% of the initial influx. A considerably

smaller fraction would be available for adipose tissue (10) and

I From the Division of Biochemistry, Department of Veterinary BasicSciences, Royal Veterinary College, University of London.

2 Supported by the British Heart Foundation.

3 Address reprint requests to PA Mayes, Division of Biochemistry,Department of Veterinary Basic Sciences, Royal Veterinary College,

University of London, London NW1 0Th, UK.

by guest on July 14, 2014ajcn.nutrition.org

Dow

nloaded from

Co2

METABOLISM OF FRUCTOSE 755S

Hexokinase will catalyse the phosphorylation of most hexose

sugars, including fructose. However, when fructose is present

FRUCT�SE

FRUCTOSE 1-P

Glyceraldehyde

I ciucose 6.pase_�jcr3I G1uc::::L,.4��

GLUCOSE6-P’��..�..�.#{216}. Glycogen

FRUCTOSE 6-P

[ FRUCTOSE 1.6- bis.P�e �( ) I �4O6PHOFRUCTOKIP4ASE

FRUCTOSE 1 ,6-bis-P

FIG 1. Fructose utilization in the liver showing its interrelationship with glucose and fatty acid metabolism. Pase,

phosphatase; P, phosphate.

skeletal muscle (11). A recent study (12) showed that when fruc-

tose is infused into exercising subjects to maintain a concentration

of 5.5 mmol/L, which is above physiological concentrations and

above the glucose concentration, there is considerably more fruc-

tose oxidation by exercising and resting muscles.

In a previous review (13) we pointed out that many investiga-

tions both in vivo and in vitro had been carried out using unphy-

siological concentrations of fructose and that deductions made

from such observations could be misleading, particularly when

applied to humans consuming normal quantities of sucrose and

fructose. Of particular relevance to hepatic metabolism arc the con-

centrations of fructose likely to be attained in the hepatic portal

vein. In humans (14) and baboons (8) maximal concentrations of

2.2 mmol/L have been recorded after a fructose or sucrose meal.

We found values within the range 1.1-2.2 mmolfL, when fed or

starved rats were given a large fructose meal by gastric intubation

(4). In humans a maximum concentration of 1.0 mmol fructoscfL

was recorded in peripheral blood (14). Because the normal blood

fructose concentration is zero when no fructose is being absorbed,

fructose concentrations in blood will vary from zero up to the

maximum recorded above, according to the quantity in the diet.

Specific metabolic pathways offructose utilization in individual

tissues

with glucose at physiological concentrations its phosphorylation

is largely inhibited by glucose (10). Nevertheless, it is probably

via this route that fructose, after uptake from the blood, enters

the metabolic pathways in adipose tissue and muscle. Of far

greater significance is the pathway discovered by Hers (15),

which involves three particular enzymes (fructokinase, aldolase

B, and triokinase), two of which arc specific for fructose metab-olism (Fig 1). These enzymes arc present in liver and kidney and

probably in the small intestine of some species-such as the

golden hamster (16), guinea pig (17), and dog (18)-that are able

to convert some absorbed fructose into glucose. However, this is

not possible in the intestine of either humans (19) or rats (17),

which do not contain glucose-6-phosphatasc, the enzyme nec-

essary for releasing glucose. Thus in these two species there is

no conversion of fructose to glucose during absorption. In this

respect the rat is a good model of human fructose metabolism.

Fructose is rapidly phosphorylated by AlP in the liver to form

fructose 1-phosphate, catalyzed by the first enzyme of the fructose

pathway-fructokinasc (20). This enzyme is virtually specific for

fructose because it is a kctohexokinase, and fructose is the only

ketohexose of physiological significance in the diet. The high ac-

tivity of fructokinase underlies the ability of the liver to extract so

much of the fructose passing through it. Fructose-i-phosphate is

split by liver aldolase (aldolase B) into glyceraldehyde and dihy-

droxyacetonc phosphate, a member of the glycolysis sequence of

intermediates. Aldolase B also functions in the liver in glycolysis

by guest on July 14, 2014ajcn.nutrition.org

Dow

nloaded from

756S MAYES

to split fructose-1,6-bisphosphate to glyceraldehyde-3-phosphate

and dihydroxyacetone phosphate. The third enzyme in the fructose

pathway is tniokinase (21), which catalyzes the phosphorylation of

glyceraldehyde by Al? to form glyceraldehyde-3-phosphate, an-

other intermediate of the glycolytic pathway.

Thus, the pathways of glucose and fructose metabolism in the

liver converge at the triose phosphate stage of metabolism and

from this point on, their metabolism is qualitatively similar. Fruc-

tose has arrived at this stage in metabolism, without passing

through the main rate-controlling step in glycolysis catalyzed by

phosphofructokinase (22). In this way fructose, on presentation

to the liver, is rapidly phosphorylated, providing increased sub-

strate to the metabolic pathways leading from triose phosphate

(ie, glycolysis, glycogenesis, gluconeogenesis, lipogenesis, and

possibly fatty acid estenification). Thus, major products of fruc-

tose metabolism in the liver are glucose, glycogen, and lactate

(23). Smaller quantities arc oxidized to carbon dioxide, ketonc

bodies or converted to tniacylglyccrol (13). The general dispo-

sition of fructose carbon between its major end products is altered

by changes in nutritional and endocrine status, eg, gluconeogen-

esis from fructose is increased during starvation, diabetes, or ad-

ministration of ethanol or glucagon (24).

Biosynthesis offructose in mammalian tissues

Although the primary purpose of this paper is to review the

effects of exogenously derived fructose, discussion of the metab-

olism of fructose is not complete without a very brief reference

to its synthesis in a few specialized tissues.

Free fructose is found in the lens, seminal fluid, and the fetal

circulation of ungulates and whales. It is formed by the sorbitol

(polyol) pathway (25), which is responsible for fructose forma-

tion from glucose. This pathway is present in the lens, seminal

vesicles, and placenta of the groups mentioned above. It increases

in activity as glucose concentrations rise in diabetics in those

tissues that are not insulin sensitive, such as the lens. Glucose

undergoes reduction by NADPH to sorbitol, catalyzed by aldose

reductase, followed by oxidation of sorbitol to fructose by son-

bitol (polyol) dehydrogenase in the presence of NAD. Both son-

bitol and fructose accumulate in the human lens in diabetics,

causing osmotic damage, which is probably involved in the path-

ogenesis of diabetic cataract. Sorbitol dehydnogenase, but not

aldose reductase, is found in liver and is responsible for the con-

version of any exogenously derived sorbitol to fructose.

Effects of fructose on carbohydrate metabolism

Effects on the concentrations of important intermediates

Experiments with both the perfused liven (23, 26, 27) and cath-

etenized human subjects (9) indicate that in the starved state,

� 66% of a fructose dose is converted to glucose and up to 25%

is released as lactate. Up to a further 8% may form glycogen

(23). After administration of fructose either orally or by intra-

venous loading there is a rapid and marked increase in glucose

1-phosphate in those tissues containing the fructose pathway, ie,

liver (28-31), kidney (30, 32), and small intestine (33, 34). Re-

cent investigations using �‘ P magnetic-resonance spectroscopy

have confirmed the accumulation of fructose 1-phosphate in the

human liver after intravenous administration of fructose (35). Al-

though most of these effects were obtained by using unphysio-

logical concentrations of fructose, experiments with isolated he-

patocytes using graded concentrations of fructose (31) demon-

strated that significant elevations of fructose 1-phosphate

occurred at concentrations of 0.5 and 1 .0 mmol fructosc/L, which

are available in the portal vein in vivo after a fructose meal.

Other important intermediates whose concentrations may alter

as a result of fructose administration include glycerol 3-phos-

phate, fructose 2,6-bisphosphate, pyruvate, lactate, ATP, and in-

organic phosphate (Pi). These will be discussed in following 5cc-

tions.

Effects on glycolysis, glycogenesis, and gluconeogenesis

As a result of the loading of the initial pathways of fructose

metabolism there is a tendency for intermediates of glycolysis to

increase in concentration, resulting in an increased flux through

the pathway, evidenced by increased lactate formation and raised

blood lactate concentrations (36, 37). In the span of glycolytic

reactions from glyceraldehyde-3-phosphate to pyruvate and lac-

tate, the rate-controlling step is catalyzed by pyruvate kinase.

This enzyme is normally under feed-forward control because of

allostenic activation by fructosc-1,6-bisphosphatc. Although this

metabolite may double in concentration when fructose is added

to hepatocytes (31), of more significance are the large increases

in fructose-1-phosphate concentrations, which extend a similar

but more enhanced activation of pyruvate kinase (38), thus fa-

cilitating passage of fructose carbon to pyruvate and lactate.

Glycogen synthesis and breakdown is controlled by a complex

series of reactions involving covalent modification by protein

phosphorylation and dephosphorylation. Briefly, regulation cen-

ters around two rate-controlling enzymes-glycogcn synthase

and glycogen phosphorylase. (See ref 25 for a general review.)

The active form of glycogen synthase (synthase a) is the de-

phosphoenzyme, whereas the inactive synthase b is phosphoryl-

ated. On the other hand active glycogen phosphorylase a is the

phosphoenzymc, whereas the inactive b form is dephosphory-

lated. Protein kinases carry out phosphorylations and protein

phosphatases carry out dephosphorylations of these enzymes.

Both processes are controlled by hormonal and allostenic modi-

fiers. Control of glycogen metabolism in liver differs in some

detail from that of glycogen metabolism in muscle and has been

reviewed by Hers (39).

A study of the literature reveals a disparity in results on whether

fructose promotes liven glycogen deposition, with the balance of

studies in vivo in the fed condition indicating that fructose is a

better promoter of glycogenesis than is glucose (31). The net dep-

osition of glycogen appears to result from both activation of gly-

cogen synthase (40, 41) and inhibition of glycogen phosphorylase

(40, 42). This appears to be brought about by several mechanisms.

Phosphorylase a is inhibited by fructosc-1-phosphate (42-44),

which accumulates after administration of fructose. Also, glucose-

6-phosphate increases in concentration and activates glycogen syn-

thase and inhibits phosphorylase (45). We have carried out per-

fusions with the liver of fed rats, using whole blood containing

physiological concentrations of glucose, amino acids, and free fatty

acids into which was infused glucose, fructose, or both sugars at

physiological rates (46). When either sugar was infused alone.

there was a net output of glucose from the liver, with no change

in glycogen concentration. However, when glucose and fructose

were infused together, there was a marked uptake of glucose and

an increase in glycogen. Fructose uptake was the same with or

without a concomitant glucose infusion.

by guest on July 14, 2014ajcn.nutrition.org

Dow

nloaded from

METABOLISM OF FRUCTOSE 757S

Thus, fructose on its own does not seem to be glycogenic but

in the presence of extra glucose, as would be the case after a

sucrose meal, fructose ‘ ‘opens the door’ ‘ for hepatic glucose up-

take and synthesis of glycogen. This is achieved, presumably, by

activation of glycogen synthase and inhibition of phosphorylase,

but fructose itself is not able to make use of the facility because

an increase in fructose-1,6-bisphosphate inhibits fructose-i,6-bis-

phosphatase (22). It was significant that lactate concentrations in-

creased only in the presence of a fructose infusion, the uptake of

which into the liver could account for the increased lactate pro-

duction. In the presence of a simultaneous glucose infusion, there

was an additional small increment in lactate production. Thus in

the fed state, fructose would appear to be predominantly glyco-

lyzed to lactate rather than converted to glycogen or glucose. An

oral load of fructose or sucrose, but not glucose, also leads to

increases in blood lactate concentration in humans (47, 48).

In the starved condition gluconcogenesis and glucose produc-

tion from fructose is an active process. even exceeding the rates

of gluconeogenesis from lactate (23, 49). Unfortunately, many

studies on gluconcogenesis from fructose, particularly in vitro,

have used unphysiologically high concentrations of fructose (ic

� 5 mmol/L) and will not be considered in the present discus-

sion. The glycolysis and gluconeogenic pathways make use of

many common intermediates and enzymes but arc controlled by

several nonequilibrium enzymes unique to each pathway (25).

Two pivotal reactions, for which the activities arc reciprocal and

coordinated, dominate these pathways. These arc catalyzed by

phosphofructokinase in glycolysis and fructose-i,6-bisphospha-

tase in gluconeogenesis (Fig 1). Their activities arc both induced

by hormones and also by allostenic modifiers often acting on the

key regulatory molecule-fructosc-2,6-bisphosphatc-which ac-

tivates phosphofructokinase and inhibits fructosc-1,6-bisphos-

phatase (50). The concentration of fructose-2,6-bisphosphatc is

elevated in liver from fed rats but falls during starvation, thus

activating fructose-i ,6-bisphosphatase and the gluconeogenic

pathway and inhibiting phosphofructokinasc and glycolysis. Un-

der these conditions, fructose- 1,6-bisphosphate does not accu-

mulate even when the flux from fructose metabolites at the triose

phosphate level increases. Therefore, there is no inhibition of

gluconcogenesis by fructose. In isolated hepatocytes there is

some increase in fructose-2,6-bisphosphate under physiological

concentrations of fructose, which leads to increased substrate cy-

cling at the locus of phosphofructokinasc and fructosc-i,6-bis-

phosphatase (51). However, there is no evidence that this inhibits

gluconeogenesis from fructose.

Long-term effects of a high fructose intake

Experimental data have been obtained from humans or animals

consuming between 7.5% and 70% of their energy-intake re-

quirement as fructose. Long-term feeding of fructose-containing

diets leads to adaptive increases in the activity of many enzymes

in the pathways of fructose metabolism. These enzymes can be

grouped into those involved initially with fructose uptake and

metabolism and those involved in lipogenesis and triglyceride

synthesis (13). Long-term effects of feeding high-fructose diets

may differ from those of feeding sucrose, because the glucose

moiety of sucrose may cause increased insulin secretion. This in

turn would be expected to cause adaptive effects. The majority

of opinion, however, would regard the long-term effects of a

sucrose diet as being due mainly to its fructose content (52).

Generally, it has been the case that glucose tolerance has been

worse under a long-term sucrose regime compared with a starch-

based diet, both in animals (53) and humans (54). The reason for

the decreased glucose tolerance does not appear to be an insuf-

ficiency in insulin secretion, rather the reverse. Increased plasma

insulin concentrations have been reported after sucrose feeding

(55-58) and insulin sensitivity, as measured by the hypoglycemic

response to administered insulin, is less (59). Thus decreased

glucose tolerance is due to increased insulin resistance, brought

about most likely by increased nonestenified fatty acid (NEFA)

concentrations and utilization (60, 61).

The decreased utilization of glucose, as found in the whole or-

ganism when placed on a high-sucrose or -fructose diet, is also

exhibited in studies of individual tissues. However, this is usually

accompanied by an increased ability to metabolize fructose. Thus,

in the liver there is decreased utilization and oxidation of glucose

(62, 63). This is no doubt due to depressed glucokinase and in-

creased activity of glucose-6-phosphatasc in the liver after fruc-

tose-containing diets are consumed (64). As opposed to the im-

mediate effect of fructose in facilitating conversion of glucose to

glycogen in the liver (46), long-term feeding of fructosc-containing

diets reduces the conversion of glucose to liver glycogen (65, 66).

However, conversion of fructose to liver glycogen is increased

because of enzyme adaptation (64). Consumption of a sucrose diet

before consumption of a sucrose load did not elicit any extra in-

crease in blood lactate compared with a starch diet (67).

Muscle tissue also shows decreased ability to metabolize glu-

cose and increased ability to oxidize fatty acids after animals

have been fed high-fructose diets (68). Likewise, the utilization

of glucose in adipose tissue is impaired with respect to its uptake,

oxidation, and conversion to glycogen (62, 69).

Effects of fructose on lipid metabolism

Immediate effects on the initial pathways of lipid metabolism

and on lipogenesis

Fructose has both immediate and long-term effects on lipid

metabolism. Short-term or acute effects arc those that occur as a

result of fructose metabolism by existing enzyme capacity,

whereas long-term effects result mainly from enzyme adaptation

to diets containing high concentrations of sucrose or fructose.

Because of the importance of the liver in fructose uptake from

the blood, many of its effects on lipid metabolism arc found in

this organ. They involve the major pathways of fatty acid oxi-

dation and estcnification, and lipogenesis. In this respect, dangers

of applying results in experimental animals to humans must be

appreciated. For example, it is likely that lipogenesis is a more

important pathway in rats than in humans, where lipogenesis may

be confined to the liver and where even there the key enzymes

of lipogenesis may be low in activity (70).

In the liver, lipid metabolism is affected by fructose at the

dihydroxyacetone phosphate and pyruvate locations. Dihydroxy-

acetone phosphate is in equilibrium with glycerol-3-phosphatc,

cosubstrate for esterification of long-chain acyl-CoA in the syn-

thesis of triglyceride and phospholipids. Triglyceride is the major

precursor and determinant of very-low-density lipoprotcins

(VLDLs) secreted by the liver and which constitute the bulk of

endogenously derived plasma triglyceride (71). Fructose also

generates pyruvate, which, besides forming lactate, enters the

mitochondnion to form acetyl-CoA as a result of pyruvate de-

by guest on July 14, 2014ajcn.nutrition.org

Dow

nloaded from

758S MAYES

hydrogenase (PDH) activity. Here it can act as a carbon source

for three products: carbon dioxide, after oxidation in the citric

acid cycle; long-chain fatty acids, after entering the lipogenesis

sequence of reactions; and, ketone bodies (72, 73). Acetyl-CoA

is the major carbon source for lipogenesis. However, as lipogen-

esis occurs in the cytosol, acetyl-CoA must be transported

through the mitochondnial membrane as citrate, reforming acetyl-

CoA in the cytosol by the action of the lipogenic enzyme, AlT-citrate lyase. Acctyl-CoA is converted to long-chain fatty acid

via the important cytosolic intermediate malonyl-CoA. By these

pathways, fructose provides carbon atoms for both the glycerol

and the acyl portions of the acylglycerol molecule.

The activity of PDH is a key factor in determining the fate of

pyruvate. It is activated by a decreased ratio of [AlT] to [ADP]

and by an increase in pyruvate (74); it is inhibited by increased

concentrations of NEFA (74, 75). As discussed in reference 76,

fructose administration, particularly at high amounts, causes de-

pletion not only of AlT but of all adenine nucleotides; therefore,

the ratio of [ATP] to [ADP] may not change appreciably. Oral

administration of large amounts of fructose to humans produces

hypcruriccmia, which is known to follow depletion of Al? and

other adcninc nucleotides (77, 78). However, no decrease in he-

patic [ATP] was found in fructose-fed rats (79). Adenine nude-

otide concentrations and PDH activity did not change when fruc-

tose was added to the perfused rat liver at physiological concen-

trations of 1 .3 mmol/L. However, at 1 1 .0 mmol/L both AlP and

total adenine nucleotides were decreased and PDH was activated

(80). At a fructose concentration of 1 .5 mmol/L there was a per-

ceptible increase in PDH activity, which was not accompanied

by any change in adenine nucleotide concentrations (13). Of

more significance was the fact that at this physiological concen-

tration of fructose, the inhibitory effect of increased oleate (a

NEFA) on PDH activity was reversed. Generally, these results

support the view that fructose causes increases in PDH activity

by increasing pyruvate concentrations, rather than by decreasing

the ratio of [Al?] to [ADP] (81). In summary, it would appear

that some activation of PDH occurs with high physiological con-

centrations of fructose, especially in antagonism to the depressant

effect of increased concentrations of NEFAs.

When fructose was injected intravenously into rats there was

an immediate though transient increase in glycerol-3-phosphate

and pyruvate concentrations (82). In contrast, after glucose ad-

ministration there was no immediate increase. When fructose was

administered intrapenitoneally there was an increase in hepatic

dihydroxyacetone phosphate, glycerol-3-phosphate, pyruvate,

and lactate (83). Similar results have been obtained in the per-

fused liver (27, 84, 85). Acetyl-CoA concentrations also increase

(72). It is also likely that malonyl-CoA concentrations rise when

fructose is administered because an increase has been recorded

in hepatocytes in the presence of lactate and pyruvate (86), both

of which increase in the presence of fructose. Thus, fructose

causes increased concentrations of its metabolites in both the

glycerol-3-phosphatc and glycolytic and acetyl-CoA branches of

its metabolism. Although many lipogenic intermediates increase

in concentration after administration of fructose, there is little

evidence that fructose on its own has an immediate effect on

lipogenesis. Thus, fructose stimulated lipogenesis from acetate

in chick liver slices, but not when fructose was added alone (87).

Using the tnitiated water technique we were unable to demon-

strate any increase in lipogenesis in perfused livers from fed rats

in the presence of physiological concentrations of fructose (13).

It is likely that if fructose were administered with an equal

amount of glucose, there might well be increased lipogenesis,

particularly if this was accompanied by a rise in insulin concen-

tration.

Immediate effects on fatty acid oxidation and esterification and

on lipoprotein formation and utilization

Although fructose does not seem to cause any marked imme-

diate increase in lipogenesis by itself, it does have marked and

immediate effects on the fate of NEFAs. These fatty acids arise

as a result of lipid mobilization in adipose tissue or from hydro-

lysis of triglyceride-rich lipoprotcins by lipoprotcin lipase. We

established in the perfused rat liver that NEFAs, which are taken

up by the liver to the extent of 30-40% per pass, are either oxi-

dized or esterified (88, 89). An inverse relationship was dem-

onstrated between the quantity of fatty acids oxidized to carbon

dioxide and kctonc bodies on the one hand, and those esterified

in liver acylglyccrols and in VLDLS on the other. Regulation of

this balance controlled both ketogenesis and VLDL secretion in

a reciprocal manner. For a given load of fatty acids taken up by

the liver, livers from fed animals oxidized less fatty acids but

esterified more than did livers from starved animals, demonstrat-

ing that a mechanism exists, which is affected by the nutritional

state, for regulating the partition of fatty acids between these two

major pathways. We showed (89) that livers from fasted animals

maintained a constant fractional rate of cstcrification irrespective

of the NEFA load, demonstrating that glycerol-3-phosphate

availability could not have been rate limiting in fatty acid ester-

ification. It has now been firmly established that the regulatory

site for controlling the balance between oxidation and esterifi-

cation lies in the oxidative pathway, where long-chain acyl

groups enter the mitochondrion (90). The rate-limiting step in

mitochondnial oxidation of long-chain fatty acids is catalyzed by

palmitoyltransferasc I. Fatty acids failing to enter the mitochon-

dna for oxidation do not accumulate but are esterified immedi-

ately. Palmitoyltransfcrasc I is inhibited by malonyl-CoA (90),

whose concentration increases in the fed condition when there is

active lipogcnesis. This can explain the change in balance be-

twecn esterification and oxidation of fatty acids between livers

of fed and starved animals.

To test whether fructose and insulin had direct and immediateeffects on VLDL secretion by the liver, we perfused livers from

fed rats with whole blood into which [‘4C]oleate NEFA was in-

fused to maintain a constant physiological concentration (46, 91).

Increased concentrations of insulin or an infusion of a physio-

logical quantity of fructose decreased oxidation and increased

estenification of fatty acids and secretion of VLDL triglyceride

from the liver. When insulin was added together with fructose,

the separate effects were additive with a further decrease in ox-

idation and enhancement of estenification and VLDL secretion.

Malonyl-CoA is the product of acctyl-CoA carboxylase activityand it is known that this enzyme is activated by covalent modi-

fication by insulin and inhibited by glucagon (92). Therefore,

insulin and fructose can independently increase the concentration

of malonyl-CoA. Fructose may also enhance estenification byraising the concentration of glyccrol-3-phosphate (93, 94) al-

though this seems unlikely if physiological concentrations of

fructose do not in fact influence the concentration of this inter-

mediate (82). Also, we have shown that mitochondnial glycerol

phosphate acyltransferase activity is enhanced by insulin (95),

by guest on July 14, 2014ajcn.nutrition.org

Dow

nloaded from

METABOLISM OF FRUCTOSE 759S

and this could also account for a direct effect of insulin in pro-

moting esterification and VLDL secretion.

These effects of fructose on NEFA metabolism are unique to

this sugar. In a direct comparison with glucose, the effect of an

infusion providing physiological concentrations of fructose in the

perfusate of the perfused liver were compared with a comparable

infusion of glucose. Only the fructose infusion boosted

[‘4C]NEFA estenification and [‘4C]VLDL secretion (46). When

similar amounts of the two sugars were then infused simultane-

ously to simulate sucrose digestion and absorption, there were

no further increases in NEFA esterification or secretion as VLDL.

Perfusions were also carried out (73) to test the effect of a fruc-

tose concentration above the physiological range (8.9 mmollL)

against a concentration within the physiological range (1.5 mmol/

L). At the higher concentration, VLDL production was cut toone-third of the amount found under the physiological concen-

tration. Because lipoprotein production is a complex energy-re-

quining process, the lower ATP availability associated with

high-fructose concentrations (76) was probably the reason for the

reduced VLDL output under these conditions.

When a 20% fructose solution was given intravenously, serum

triglycerides decreased in female baboons but increased in males,

whereas a comparable solution of glucose caused a decrease in

both sexes (96). Although the increase in triglycerides is most

likely the result of a direct effect on the liver, as described above,

the fall in triglyceride concentration is probably due to a decrease

in NEFA concentration, which in turn provides less NEFA sub-

strate for the liver and therefore less VLDL production. In hu-

mans, fructose causes a decrease in NEFAs with little change in

insulin concentrations (97), indicating that any reduced secretion

of VLDL is due to decreased NEFA availability and any in-

creased secretion is most likely due to the direct hepatic action

of fructose. The net output of VLDL by the liver is therefore a

balance between these two opposing effects of fructose.

Fructose is both antiketogenic and ketogenic depending on the

circumstances. In vivo, in the starved state, a physiological intake

of fructose is antiketogenic. This is nearly always due to its in-

hibition of NEFA mobilization from adipose tissue (97), which

results in reduced uptake of the main ketogenic substrate by the

liver. It appears that there is also a direct antiketogenic effect of

fructose on the liver, as demonstrated in vivo in the presence of

a constant plasma NEFA concentration (98). Similar experiments

have been carried out in the perfused liver and isolated hepato-

cytes in the presence of added NEFA (72, 84, 91, 99, 100). Most

of these experiments were carried out in the starved state. It is

unclear whether or not malonyl-CoA inhibition of carnitine pal-

mitoyltransferase I can explain these results. It would require the

immediate activation of acctyl-CoA carboxylase, which has not

been reported under these conditions. Although glyccrol-3-phos-

phate availability does not appear to be rate-limiting on esterifi-

cation of NEFAS, it is possible that boosting its concentration

above the normal concentration as a result of fructose metabolism

could cause a greater fractional estenification of NEFAs and less

oxidation to ketonc bodies (98).

That fructose can also be ketogenic was first shown in perfused

livers from starved rats infused with the sugar at 20 mmollL (72).

The phenomenon was also demonstrated with livers from fed

animals by using lower but still unphysiological concentrations

of fructose (73, 101). To test whether this effect could be ob-

tamed with fructose within physiological concentrations we stud-

ied ketogenesis in perfused livers from fed animals and showed

that ketogenesis progressively increased as the concentration of

fructose was raised (13). At 1.5 mmol fructose/L, there was a

significant increase compared with control perfusions in the ab-

sence of fructose. Some of these studies also used [‘4C]fructosc.

At 20 mmol/L, all of the carbon atoms found in ketone bodiescame from fructose (72) whereas at 8.9 mmol/L, half came from

fructose and at the physiological concentration of i .5 mmol fruc-

tose/L, 15% of the carbon atoms originated from fructose (73).

Radiolabeled fructose has also been used to follow the course

taken by the sugar in the various metabolic pathways. Initially,

there is no dilution of fructose label because there is no endog-

enous pool of free fructose, but as fructose is converted into

intermediates common with glucose and fatty acid metabolism,

the label is extensively diluted. Therefore, although it is not pos-

sible to quantify the fate of fructose in a whole animal or tissue

without knowledge of the specific radioactivity of intermediary

pools, some idea may be obtained of the fate of administered

fructose. Thus, when the metabolism of starved rats given either

[ U-’4C]glucosc or [U-’4C]fructose by gastric intubation was com-

pared, it was shown that although � 2.5 times as much newly

synthesized triglyceride came from total body glucose as from

administered fructose, twice as much of the administered fructose

formed triglyceride compared with a similar quantity of admin-

istered glucose (102). These data reflect the fact that fructose is

taken up predominantly by the liver whereas glucose is utilized

mainly by the extrahepatic tissues. In liver slices, radiolabeled

fructose is converted to lactate, pyruvate, carbon dioxide, and

triglyceride more rapidly than is glucose (103). Similar results

have been reported for incorporation into plasma lipids in

guinea pigs (104) and baboons (105). After administration of

[‘4C}fructosc, more radioactivity is found in glyceride glycerol

than in the fatty acid portions of hepatic triglyceride (102, 103),

due in part to greater dilution and exchange of label in the path-

way from fructose to long-chain fatty acids than in that to glyc-

erol-3-phosphatc (Fig 1). When [U-’4Clfructose was infused to

maintain a concentration of 1.5 mmol/L in the perfusate of livers

prepared from fed rats, 12% was oxidized to carbon dioxide and

2.4% was converted to ketone bodies; 4.1% was in liver lipids,

1.6% was incorporated into VLDLS, and only 0.4% was in total

cholesterol (13). As the fructose load was increased, proportion-

ally less of the labeled fructose entered lipid products except for

ketone bodies, which remained the same. Thus, ketonc bodies

and lactate act as overflows of excess carbon as fructose saturates

the metabolic pathways. Hence, ketone bodies fulfill a similar

role in the liver with respect to carbohydrate as they do when

fatty acids are present in excess.

Long-term effects of a high fructose intake

Enzyme adaptation is also responsible for many of the long-

term effects of a high-fructose diet on lipid metabolism. as well

as for the long-term effects on carbohydrate metabolism. In the

liver, fructokinase activity increased in rats on a fructose-rich diet

(106). Fructose or sucrose feeding also increased the activity of

glycerol-3-phosphate dehydrogenase, which is necessary for the

conversion of fructose to glycerol-3-phosphate via the glycolytic

intermediate dihydroxyacetone phosphate (107, 108). After 50 d

on a fructose diet, but not on a sucrose diet, pyruvate kinase in

rats increased substantially (109). Similarly fructose diets in-

creased liver PDH activity in rats (1 10). In the pathway of lipo-

genesis and triglyceride formation, ATP-citratc lyase (108, 111,

112), acetyl-CoA carboxylase (111-113), fatty acid synthase

by guest on July 14, 2014ajcn.nutrition.org

Dow

nloaded from

760S MAYES

(1 1 i-i 14), and phosphatidate phosphohydrolase (1 15) are re-

ported to be increased in activity in animals on fructose-contain-

ing diets. Also, the enzymes responsible for generation of reduc-

ing power for lipogencsis-glucose-6-phosphatc dehydrogenase

(107-109, iii, 112, 116), 6-phosphogluconate dehydrogcnase

(107, 108, 1 1 1, 1 12), and NADP malate dehydrogenase (‘malic

enzyme’) (107, 1 1 1, 1 12, 1 16)-are all increased in activity.

Many of these adaptive changes in liver enzyme activity arc not

unique for fructose because glucose feeding will also increase

their activities (107), illustrating that this is a more general effect

of soluble carbohydrates. Of considerable interest is that sucrose

or fructose feeding reduces the activities of hexokinase, pyruvate

kinase, PDH, ATP-citrate lyase, acetyl-CoA carboxylase, fatty

acid synthase, glucose-6-phosphatc dehydrogenase, and NADP

malate dehydrogenase in adipose tissue, whereas feeding glucose

or starch enhances their activities (52). It would appear from a

review of these enzyme activities that lipogenesis is elevated in

the livers but depressed in adipose tissue of sucrose- or fructose-

fed animals. In keeping with this conclusion the concentrations

of the following intermediates in the liver are elevated: pyruvate,

malate, acctyl-CoA (1 17, 1 18), acetoacetyl-CoA, and long-chain

acyl-CoA (1 17). However, glycerol-3-phosphate concentrations

apparently do not increase (119).

Biosynthesis of long-chain fatty acids (lipogenesis) has been

shown to increase in liver slices from rats fed either fructose

(120) or sucrose (121, 122), as measured directly by using la-

beled acetate or tritiated water. Increased incorporation of labeled

fructose has also been shown (118, i23). In the whole animal,

incorporation of tritiated water into liver fatty acids was elevated

in rats fed fructose compared with those fed glucose (124). Also,

F‘4C]fructosc was incorporated more rapidly into fatty acids andacylglycerol glycerol in plasma and liver in rats on a fructose

diet than in rats on a glucose diet (125).

In isolated livers perfused with whole blood, we studied li-

pogenesis in rats that had been fed the standard laboratory diet

or a sucrose-supplemented diet (13, 126). Half the perfusions

were infused with fructose to maintain a physiological perfusate

concentration of 1 .4 mmolfL. Incorporation of 3H2O into liver

acylglycerol fatty acids increased significantly only in the group

of perfusions in which fructose was infused into sucrose-fed rats.

Thus, lipogenesis is increased on sucrose diets but only when

fructose specifically acts as the substrate. This finding is similar

to those found in liver slices from rats on a glucose diet, which

enhanced selectively the conversion of [‘4C]glucose to fatty ac-

ids, but not � ‘4Cjfructose, and where slices from rats on a fructose

diet selectively incorporated [‘4Cjfructose into fatty acids but not

[‘4C]glucose. They demonstrate the highly specific and selective

nature of the lipogenic adaptations to different sugars in the diet,

clearly based on the enzyme adaptations previously discussed.

Plasma triglycerides increase in both humans and rats when

diets are enriched with carbohydrate, especially sucrose and fruc-

tose (127). This has been, and still is, of considerable interest in

view of the fact that raised concentrations of plasma triglyceride

may be an independent factor associated with coronary heart dis-

case (128). The adaptive changes in enzyme activity, evidence

of increase in lipogenic potential, plus the direct actions of fruc-

tose previously described, all lead to increased output of VLDL

triglyceride from the liver, increasing the amount in the circula-

tion. The resultant concentration of triglyceride also depends on

the rate of hydrolysis by lipoprotein lipase and any impairment

of this enzyme would also lead to increased concentrations of

plasma triglyceride (129). However, long-term fructose or su-

crose feeding results in hyperinsulinemia (55-58) and increases

post hepanin lipolytic activity, indicating an enhancement of li-

poprotcin lipase activity (52). Therefore, raised plasma concen-

trations of triglyceride in fructose- or sucrose-fed animals, in-

cluding humans, reflects an increased rate of VLDL secretion by

the liver together with an increased rate of VLDL catabolism.

Fructose feeding enhances the release of NEFAs from adipose

tissue (60), increasing its plasma concentration (60, 61), because

of a decreased rate of re-esterification of NEFAs within adipose

tissue (69). Esterification of NEFAs in adipose tissue depends on

glucose utilization, which is depressed under conditions of fruc-

tose feeding. NEFAs are the major precursors of VLDL triglyc-

eride in the liver (71). Therefore, they will augment VLDL for-

mation in animals on high-fructose or -sucrose diets. Mindful of

the fact that NEFAs arc always present in plasma, we studied the

effects on VLDL secretion of a physiological infusion of fruc-

tose, and of sucrose supplementation of the diet, in isolated per-

fused livers infused with [1-’4C]oleate NEFA (126). VLDL tn-

glyceride production increased in both the livers infused with

fructose alone and in those derived from sucrose-fed rats. When

livers from sucrose-fed animals were infused with fructose, the

rate of secretion more than doubled. Because shifts in balance

between oxidation and estenification of NEFAs in the direction

of estenification have been shown to increase secretion of VLDL

triglyceride (89), we also measured these indexes. Oxidation of

[ 4C]oleatc to 4C0 was highest and estenification was lowest in

control perfusions when livers from normal animals were used.

Fructose infusion or sucrose feeding shifted the balance in favor

of estenification and increased output of [‘4C]VLDL. The corn-

bination of fructose infusion plus sucrose feeding showed the

lowest rate of oxidation and the highest rate of estenification and

output of [‘4C]VLDL. The marked extra increase in VLDL output

in the combined group was probably due to increased lipogenesis

from fructose in this group (13, 126).

The potential adverse effects of fructose on lipid metabolism

in humans has been reviewed (130, 131). It was concluded from

a review of a large number of studies that normal individuals

consuming diets containing average quantities of fructose have

normal fasting triglyceride concentrations. However, this is only

a measure of triglyceride clearance and does not give information

on prandial and postprandial concentrations of triglycerides,

which might be raised on these diets. The condition was exac-

erbated in individuals having defects in carbohydrate metabolism

leading to hypertniglycenidemia, even with low intakes of fruc-

tose. Similar conclusions apply to plasma cholesterol, which may

increase if higher than normal amounts of VLDLS arc present. In

addition, the combination of saturated fat and fructose in the diet

would appear to favor elevated cholesterol concentrations.

Effects of fructose on purine metabolism

The hyperuricemic effect of fructose

It was first reported (132) that when fructose was administered

intravenously to both normal children and those with hereditary

fructose intolerance, there was an increase in serum and urinary

uric acid. The hyperunicemic effect appears to be dose dependent

and the fructose infusion needs to be > 0.5 gkg body wt ‘

to cause detectible hyperunicemia. Also, the effect is fructose

specific because comparable infusions of either glucose or galac-

by guest on July 14, 2014ajcn.nutrition.org

Dow

nloaded from

METABOLISM OF FRUCTOSE

FRUCTOSEI�CT��1

FRUCTOSE 1�Ptose do not raise the plasma uric acid concentration (133). Pa-

tients suffering from gout arc more sensitive to fructose admin-

istration than are normal subjects, as are diabetics (130). These

results apply to fructose administered parenterally, but is it pos-

sible to increase blood uric acid concentrations when fructose is

consumed orally? The effect of feeding fructose at a dose of 1 g!

kg body wt was studied in healthy subjects, in patients with gout,

and in children of parents with gout (78). Hyperunicemia oc-

curred in all groups but it was more marked and prolonged in

those subjects, or offspring of parents, with gout. Subjects con-

suming higher than � 18% of their energy intake as sucrose (ic,

9% as fructose) showed significant increases in serum uric acid

(130). These results suggest that the average intake of fructose

or sucrose in a mixed diet is on the borderline for provoking

hyperunicemia in normal individuals. Note that gouty patients do

have lower serum uric acid conccntrations when fed lower-fruc-

tose diets; even healthy individuals may be at risk of hyperuni-

cemia when consuming a diet high in fructose.

Pathways of adenine nucleotide catabolism and biosynthesis

To understand the mechanism of the hyperunicemic effect of

fructose a knowledge of puninc nucleotide metabolism and its

regulation is necessary. Adenylate kinase maintains equilibrium

between the adenine nucleotides ATP, adenosinc diphosphate

(ADP), and adenosine monophosphate (AMP) (Fig 2). Thus, all

three adenine nuclcotides are able to be degraded and formed via

AMP. However, as with many other common metabolites, the

pathway of degradation of adenine nucleotides is not the simple

reverse of the pathway of biosynthesis. In the degradative path-

way, inosine monophosphate (IMP) is formed from AMP by the

action of AMP deaminase. Phosphate is then removed by 5’-

nucleotidase to form inosine. Alternatively, inosine may be

formed from AMP via adenosinc but experiments in isolated he-

patocytes (31) have indicated that this is not as important a route

for AMP degradation as that via IMP. In addition, the reaction

catalyzed by AMP deaminase appears to be the rate-limiting step

in the breakdown of hepatic adeninc nucleotides (134-136), and

Pi at normal intracellular concentrations is an important inhibitor

of this enzyme. Inosinc is broken down to uric acid via hypo-

xanthine and xanthinc. In humans and other primates this is the

end product of purinc metabolism and is excreted as such, but

other mammals possess the enzyme unicase, which allows further

oxidation to the much more soluble allantoin. A consequence of

the absence of unicase in humans is the occurrence of gout, which

is associated with hypcruniccmia.

IMP is also the first formed punine nucleotide in the pathway

of de novo synthesis from nibose 5-phosphate. This pathway in-

volves no less than 1 1 separate steps (137). Of these, only the

first two need to be mentioned because they catalyze the rate-

controlling reactions, both of which are held in check by a feed-

back regulation because of high concentrations of puninc nude-

otides. In addition, 5-phosphonibosyl-1-pyrophosphatc (PRPP),

the product of the first reaction in the pathway is an allostenic

activator of the second reaction in the pathway catalyzed by

PRPP glutamyl amidotransferase (Fig 2).

Mechanism underlying the hyperuricemic action of fructose

The essential finding that led to the explanation as to why

fructose administration caused increased uric acid formation was

the observation that it was accompanied by a sharp fall not only

76lS

e 3.f�

AMP+ATP 1 � 2ADP

Adenosine Adenylo- � IMP4� � 9 steps

Inossne � 5-Phospho-p ribosylamine

ATPMJdIEOS1O#{128} AMP. GMP

� I4

PRPP GL.UTAMYI.

Hypoxanthine M4IDOTR4j�G:ER*.SE

A

XANTHINE �

(O)�DASE) PRPP

AMP. ADP. AlP

Xanthine S

_____________ 4XANTHINE I PRPP

o&{�DmGB� �SYN�1ETASE

(OXIDASE)

Uric acid Ribose 5-P

FIG 2. The relationship between fructose metabolism and uric acidformation in the liver. P. phosphate; AMP, adenosine monophosphate:

ADP, adenosine diphosphate; ATP, adenosine triphosphate; Pi, organic

phosphate; GMP, guanosine monophosphate; PRPP, 5-phosphonibosyl-

1-pyrophosphate; IMP, inosine monophosphate.

in hepatic AlT concentration but also in total adeninc nucleotides

(76). The reason for this became apparent when it was shown

that fructosc-1-phosphatc accumulated and Pi concentrations fell

(27). Thus, the sequence of events after fructose administration

involves rapid phosphorylation of the sugar to fructose-1-phos-

phate because of the high activity of fructokinase. This leads to

depletion of ATP due to inhibition of oxidative phosphorylation

of ADP because of a shortage of Pi sequestered in fructose-i-

phosphate. The lowering of AlP concentration is also assisted

by the activity of tniokinase in utilizing Al? in the phosphory-

lation of glyceraldehyde to glyccraldehyde-3-phosphatc. The de-

pletion of Pi and ATP leads to the removal of the allostenic in-

hibition on the enzymes that degrade AMP, respectively, AMP

deaminase and 5’-nuclcotidase. There is a rise in inosinc con-

centration that leads to increased formation of uric acid with de-

pletion of the total adenine nucleotide pool.

The depletion of adenine nucleotides also leads to stimulation

of the pathway of punine nucleotide synthesis as a feedback re-

action (Fig 2). The lowering of the concentration of adenine nu-

cleotides removes the allostenic inhibition of the first two steps

in the pathway catalyzed by PRPP synthetase and PRPP glutamyl

amidotransferase. The production of PRPP acts as a further ac-

tivator of PRPP glutamyl amidotransferase, leading to IMP syn-

thesis. If 5’-nucleotidase is still active, IMP will be degraded to

uric acid rather than converted to AMP. In this way the initial

production of uric acid from the adenine nuclcotidc pool is aug-

mented by dc novo synthesis.

by guest on July 14, 2014ajcn.nutrition.org

Dow

nloaded from

762S MAYES

quantities of fructose may be at increased risk.

The reduction in concentration of ATP and other high energy

phosphates has other profound effects on metabolism and other

functions dependent on a continual supply of these vital sources

of free energy, eg, inhibition of protein (76) and nucleic acid

synthesis (138).

Summary and conclusions

Many investigations both in vivo and in vitro have been carried

out by using unphysiological concentrations of fructose. Deduc-

tions made from such observations can be misleading, particu-

larly when applied to humans consuming normal quantities of

sucrose and fructose. This aspect has been taken into account in

compiling this review.

Specific enzymes in the liver-fructokinase, aldolase B, and

tniokinase-allow fructose ready access to the tniose phosphate

pool and all pathways leading from it, after bypassing the phos-

phofructokinase regulatory step in glycolysis. In the fed state,

this allows a greater saturation of the glycolysis pathway with

consequent lactate production, activation of PDH, and domi-

nance of the oxidative pathways leading to carbon dioxide for-

mation and ketone body production when under fructose load.

Glucose production, glycogenesis, and lipogenesis are not en-

hanced but there is an immediate shift in the balance between

oxidation and estenification of NEFAs in favor of estenification.

This leads to augmented output of VLDLS from the liver. In the

liver of starved animals, the enzymes of gluconeogenesis are ac-

tive and fructose will form much more glucose under these con-

ditions.

As a result of enzyme adaptation to the long-term feeding of

fructose or sucrose diets, the pattern of fructose metabolism is

changed. Enhanced activity of fructose-1,6-bisphosphatase, gly-

cogen synthase, and glucose-6-phosphatase allow more fructose

to form glycogen and glucose. Enhanced activity of lipogenic

enzymes in the liver, but not in adipose tissue, stimulates long-

chain fatty acid synthesis, which augments the immediate hepatic

actions of fructose in promoting VLDL triglyceride output. This

leads to hypertniglycenidemia. Fructose metabolism in adipose

tissue also causes impaired glucose utilization and impaired es-

tenification of fatty acids, which promotes release of NEFAs with

consequent raised plasma NEFA concentrations and increased

VLDL production. The increased triglyceride and NEFA con-

centrations impair utilization of glucose in muscle, decreasing

glucose tolerance and increasing insulin resistance with conse-

quent hypeninsulinemia. This, in turn, will stimulate the already

increased VLDL production by the liver. Because VLDLS con-

tam � 20% cholesterol, there is a corresponding rise in plasma

cholesterol, with little change in LDL cholesterol.

Acute loading of the liver with fructose is also a cause of

hyperunicemia. This is due to utilization of Al? in the phospho-

rylation of fructose and sequestration of Pi in fructose-1-phos-

phate, preventing oxidative regeneration of AlT from ADP. Be-

cause the enzymes of adeninc nucleotide degradation are inhib-

ited by AlT and Pi, removal of the inhibition leads to destruction

of the total adeninc nucleotide pool and generation of uric acid.

Both the hyperlipidemic and hyperuricemic effects of fructose

can be demonstrated in humans. Although these effects seem

minimal in healthy individuals on normal diets, individuals on

very-high-fructose diets and individuals who are actually or p0-

tentially hyperlipidemic or hyperunicemic who take in average

It will be apparent that although there have been noteworthy

contributions on fructose metabolism in humans, most investi-

gations have been carried out in laboratory animals, particularly

rats. This is inevitable when precise information about intracel-

lular processes is required, involving invasive and critical pro-

cedures. It is clear that systematic investigations in humans are

needed to ascertain the precise amounts, both of fructose con-

sumption and of its concentration in the blood, at which dde-

tenious effects such as hyperlipidemia and hyperunicemia occur.

Nuclear magnetic resonance spectroscopy is clearly of value as

a noninvasive technique in monitoring intracellular changes in

the concentration of key metabolites. U

References

1. Bruckdorfer KR, Kang 55, Yudkin J. Plasma concentrations of

insulin, corticosterone, lipids and sugars in rats fed on meals withglucose and fructose. Proc Nutr Soc 1973;32: 12A-B.

2. Reiser 5, Michaelis 0, Putney J, Hallfrisch J. Effect of sucrosefeeding on the intestinal transport of sugars in two strains of rats.J Nutr 1975;105:894-905.

3. Silliman K, Coulston AM. Sugars in the diet. In: Kretchmer N,Hollenbeck CB, eds. Sugars and sweetners. London: CRC Press,

1991: 17-35.4. Topping DL, Mayes PA. The concentrations of fructose, glucose

and lactate in the splanchnic blood vessels of rats absorbing fruc-

tose. Nutr Metab 1971;13:331-8.5. Mendeloff Al, Weichselbaum TE. Role of the human liver in the

assimilation of intravenously administered fructose. Metabolism1953;2:450-8.

6. Topping DL, Mayes PA. Comparative effects of fructose and glu-cose on the lipid and carbohydrate metabolism of perfused rat liver.

Br J Nutr 1976;36:113-26.7. Macdonald I, Turner U. Serum fructose levels after sucrose or its

constituent monosaccharides. Lancet 1968; 1:841-3.8. Crossley JN, Macdonald I. The influence in male baboons of a high

sucrose diet on the portal and arterial levels of glucose and fructose

following a sucrose meal. Nutr Metab 1970;12:171-8.9. BjOrkman 0, Felig P. Role of the kidney in the metabolism of

fructose in 60-hour fasted humans. Diabetes 1982:31:516-20.10. Froesch ER, Ginsberg JL. Fructose metabolism of adipose tissue.

I. Comparison of fructose and glucose metabolism in epididymaladipose tissue of normal rats. J Biol Chem 1962;237:3317-24.

1 1 . Bergstrom J, Hultman E. Synthesis of muscle glycogen in man afterglucose and fructose infusion. Acta Med Scand 1967;182:93-107.

12. Ahlborg G, BjOrkman 0. Splanchnic and muscle fructose metab-olism during and after exercise. J Appl Physiol 1990;69:1244-51.

13. Mayes PA, Laker ME. Effects of acute and long-term fructose ad-ministration on liver lipid metabolism. In: Macdonald I, Vrana A,

eds. Metabolic effects of dietary carbohydrates. Prog BiochemPharmacol 1986;21 :33-58.

14. Holdsworth CD, Dawson AM. Absorption of fructose in man. Proc

Soc Exp Biol Med 1965;118:142-5.

15. Hers HG. Fructose metabolism. (La m#{233}tabolisme du fructose.)Brussels: Arscia, 1957 (in French).

16. Wilson TH, Vincent TN. Absorption of sugars in vitro by the in-

testine of the golden hamster. J Biol Chem 1955:216:851-66.17. Ginsburg V, Hers HG. On the conversion of fructose to glucose by

guinea pig intestine. Biochim Biophys Acta 1960:38:427-34.18. Shoemaker WC, Yanof HM, Turk LN, Wilson TH. Glucose and

fructose absorption in the unanesthetized dog. Gastroenterology

l963;44:654-63.19. Cook GC, Jacobson J. Individual variation in fructose metabolism

in man. Br J Nutr 1971;26:187-95.

20. Hers HG. Fructokinase of the liver. (La fructokinase du foie.)Biochim Biophys Acta 1952;8:416-23.

by guest on July 14, 2014ajcn.nutrition.org

Dow

nloaded from

METABOLISM OF FRUCTOSE 763S

21. Hers HG. Tniokinase. In: Colowick SP, Kaplan NO, eds. Methods

in enzymology. Academic Press, New York: l962;5:362-4.

22. Underwood AH, Newsholme EA. Properties of phosphofructoki-nase from rat liver and their relation to the control of glycolysis

and gluconeogenesis. Biochem J 1965;95:868-75.

23. Exton JH, Park CR. Control of gluconeogenesis in liver. I. General

features of gluconeogenesis in the perfused livers of rats. J Biol

Chem 1967;242:2622-36.24. Van den Berghe G. Metabolic effects of fructose in the liver. Curr

Top Cell Regul 1978;13:97-135.

25. Murray RK, Granner DK, Mayes PA, Rodwell VW. Harper’s

biochemistry. 23rd ed. Norwalk, CT: Appleton and Lange,1993:206-11.

26. Sestoff L, Fleron P. Determination of the kinetic constants of fruc-tose transport and phosphorylation in the perfused rat liver.

Biochim Biophys Acta 1974:345:27-38.

27. Woods HF, Eggleston LV, Krebs HA. The cause of hepatic accu-mulation of fructose-i-phosphate on fructose loading. Biochem J

l970;1 19:501-10.

28. Kjerulf-Jensen K. The phosphate esters formed in the liver tissueof rats and rabbits during assimilation of hexoses and glycerol. Acta

Physiol Scand 1942;4:249-58.

29. Gunther MA, Sillero A, Sols A. Fructokinase assay with a specificspectrophotometnic method using l-phosphofructokinase. Enzymol

Biol Clin (Basel) 1967:8:341-52.30. Bunch HB, Lowry OH, Meinhardt L, Max P, Chyu KJ. Effect of

fructose, dihydroxyacetone, glycerol and glucose on metabolites

and related compounds in liver and kidney. J Biol Chem1970;245:2092-102.

3 1 . Van den Berghe G. Fructose: metabolism and short-term effects oncarbohydrate and purine metabolic pathways. In: Macdonald I,Vrana A, eds. Metabolic effects of dietary carbohydrates. Prog

Biochem Pharmacol 1986;21:i-32.32. Burch HB, Choi 5, Dence CN, Alvey TR, Cole BR, Lowry OH.

Metabolic effects of large fructose loads in different parts of the

rat nephron. J Biol Chem 1980;255:8239-44.

33. Kjerulf-Jensen K. The hexosemonophosphonic acids formed withinthe intestinal mucosa during absorption of fructose, glucose and

galactose. Acta Physiol Scand 1942;4:225-48.

34. Lamers JMJ, HUlsmann WC. The effect of fructose on the stores

of energy-rich phosphate in rat jejunum in vivo. Biochim BiophysActa 1973;3l3:l-8.

35. Segebarth C, Gnivegn#{233}eAR, Longo R, Luyten PR, den HollanderJA. In vivo monitoring of fructose metabolism in the human liverby means of �‘P magnetic resonance spectroscopy. Biochimie

1991 ;73: 105-8.

36. Kayc R, Williams ML, Barbero G. Comparative study of glucose

and fructose metabolism in infants with reference to utilization and

to the accumulation of glycolytic intermediates. J Clin Invest1958:37:752-62.

37. Sahebjami H, Scalettar R. Effects of fructose infusion on lactate

and uric acid metabolism. Lancet l971;1:366-9.

38. Eggleston LV, Woods HF. Activation of liver pyruvate kinase by

fructose-i-phosphate. FEBS Lett 1970;6:43-5.

39. Hers HG. The control of glycogen metabolism in the liver. Annu

Rev Biochem 1976;45:167-89.40. Hue L, Stalmans W, Van den Berghe G, Hers HG. Effect of fruc-

tose injection on the activity of glycogen phosphorylase and syn-

thetase. Stockholm: Royal Academy of Sciences, 1973.

41. Whitton PD, Hems DA. Glycogen synthesis in the perfused liver

of streptozotocin-diabetic rats. Biochem J 1975;150:153-65.

42. Thurston JH, Jones EM, Hauhart RE. Decrease and inhibition of

liver glycogen phosphorylase after fructose. An experimental

model for the study of hereditary fructose intolerance. Diabetes1974;23:597-604.

43. Kaufmann U, Froesch ER. Inhibition of phosphorylase-a by fruc-

tose-1-phosphate, a-glycerophosphate and fructose- 1 ,6-diphos-

phate. Explanation for fructose-induced hypoglycemia in heredi-

tary fructose intolerance and fructose- 1,6-diphosphatase defi-

ciency. Eur J Clin Invest 1973;3:407-13.44. Van den Berghe G, Hue L, Hers HG. Effect of the administration

of fructose, on the glycogenolytic action of glucagon. An investi-gation of the pathogeny of hereditary fructose intolerance. Biochem

J 1973;l34:637-45.45. Hers HG, Stalmans W, de Wuif H, Laloux M, Hue L. The control

of glycogen metabolism in the liver. In: Lundquist F, Tygstrup N,eds. Regulation of hepatic metabolism. Copenhagen, Munksgaard:

1974:237-49.

46. Topping DL, Mayes PA. Comparative effects of fructose and glu-cose on the lipid and carbohydrate metabolism of perfused rat liver.

BrJ Nutr i976;36:113.-26.47. Cook GC. Absorption and metabolism of D(-) fructose in man.

Am J Clin Nutr 1971;24:i302-7.48. Macdonald I, Keyser A, Pacy D. Some effects in man of varying

the load of glucose, sucrose, fructose, or sorbitol on various me-

tabolites in blood. Am J Clin Nutr 1978;31:1305-1 1.

49. Ross BD, Hems R, Krebs HA. The rate of gluconeogenesis fromvarious precursors in the perfused rat liver. Biochem J 1967;

102:942-51.

50. Hers HG, Van Schaftingen E. Fructose 2,6-bisphosphate 2 yearsafter its discovery. Biochem J 1982;206:l-i2.

51. Clark DG, Filsell OH, Topping DL. Effects of fructose concentra-tion on carbohydrate metabolism, heat production and substrate

cycling in isolated rat hepatocytes. Biochem J 1979:184:501-7.

52. Vr#{225}naA, Fabry P. Metabolic effects of high sucrose or fructoseintake. World Res Nutr Diet 1983;42:56-lOi.

53. Uram J, Friedman L, Kline 0. Influence of diet on glucose toler-

ance. Am J Physiol 1958:192:521-4.54. Cohen A, Teitelbaum A, Balogh M, Green J. Effect of interchang-

ing bread and sucrose as main source of carbohydrate in a low fat

diet on the glucose tolerance curve of healthy volunteer subjects.

Am J Clin Nutr 1966:19:59-62.55. Shafrir E, Teitelbaum A, Cohen AM. Hyperlipidemia and impaired

glucose tolerance in Acomys cahirinus maintained on synthetic car-

bohydrate diets. Isr J Med Sci l972;8:990-2.56. Bruckdorfer KR, Yudkin J. A comparison of dietary starch and

sucrose in the pig. Nutr Metab 1975;i9:225-32.

57. Reaven GM, Risser TR, Chen YDI, Reaven EP. Characterizationof a model of dietary-induced hypertriglycenidemia in young, non-

obese rats. J Lipid Res i979;20:371-8.58. Sleder J, Chen YDI, Cully MD, Reaven GM. Hypeninsulinemia in

fructose induced hypertniglycenidemia in the rat. Metabolism

1980;29:303-5.

59. Vr#{225}naA, Slabochov#{225} Z, Fabry P, Kazdov#{225} L. Influence of diet with

a high starch or sucrose content on glucose tolerance, serum insulin

level and insulin sensitivity in rats. Physiol Bohemoslov 1974;

23:305-10.

60. Vr#{225}naA, Fabry P. Effect of dietary fructose on free fatty acidrelease from adipose tissue and serum free fatty acid concentrationin the rat. Nutr Metab 1974;17:74-83.

61. Merkens IS, Tepperman HM, Tepperman J. Effects of short-termdietary glucose and fructose on rat serum triglyceride concentra-

tion. J Nutr 1980;110:982-8.

62. Bender AE, Thadini PV. Some metabolic effects of dietary sucrose.Nutr Metab i970;i2:22-39.

63. Tuovinen CGR, Bender AE. Some metabolic effects of prolongedfeeding of starch, sucrose, fructose and carbohydrate-free diet in

the rat. Nutr Metab 1975;19:i6i-72.

64. Freedland RA, Harper AE. Metabolic adaptations in higher ani-mals. I. Dietary effects on liver glucose-6-phosphatase. J BiolChem 1957;228:743-5 1.

65. Vr#{225}naA, Fabry P. Kazdov#{225}L. Liver glycogen synthesis and glu-

cose tolerance in rats adapted to diets with a high proportion of

fructose or glucose. Nutr Metab 1978;22:262-8.

by guest on July 14, 2014ajcn.nutrition.org

Dow

nloaded from

764S MAYES

66. Vr#{225}naA, Fabry P. Kazdov#{225} L, Zvolankov#{225} K. Effect of the type

and proportion of dietary carbohydrate on serum glucose levels andliver and muscle glycogen synthesis in the rat. Nutr Metab

1978;22:313-20.

67. Kelsay JL, Behall KM, Moser PB, Prather ES. The effect of kind

of carbohydrate in the diet and use of oral contraceptives on me-

tabolism of young women. I. Blood and urinary lactate, uric acid,

and phosphorus. Am J Clin Nutr i977;30:2016-22.

68. Vr#{225}naA, Poledne R, Fabry P. Kazdov#{225} L. Palmitate and glucose

oxidation by diaphragm of rats with fructose induced hypertrigly-

cenidemia. Metabolism i978;27:885-8.

69. Vr#{225}naA, Fabry P. Slabochov#{225} Z, Kazdov#{225}L. Effect of dietaryfructose on free fatty acid release from adipose tissue and serum

free fatty acid concentration in the rat. Nutr Metab l974;i7:74-

83.

70. Bjorntorp P, Sj#{246}strOm L. Carbohydrate storage in man: specu-

lations and some quantitative considerations. Metabolism 1978;

27:1853-65.

7i. Havel Ri, Felts JM, Van Duyne CM. Formation and fate of endog-

cnous triglycerides in blood plasma of rabbits. J Lipid Res 1962;

3:297-308.

72. Soling HD, Willms B, Janson G. Ketogenic action of fructose in

the isolated perfused rat and guinea pig liver. FEBS Lett 1970;

11:324-7.

73. Laker ME, Mayes PA. [U-’4C] Fructose and lipid metabolism

in the perfused rat liver. Biochem Soc Trans 1979:7:1279-

81.

74. Wieland OH, Siess EA, Weiss L, et al. Regulation of the mam-

malian pyruvate dehydrogenase complex by covalent modification.

In: Davies DD, ed. Rate control of biological processes. Cam-

bridge: Cambridge University Press, 1973:371-400.

75. Laker ME, Mayes PA. Investigations into the direct effects of in-

sulin on hepatic ketogenesis, lipoprotein secretion and pyruvate

dehydrogenase activity. Biochim Biophys Acta i984;795:427-30.

76. Maenpaa PH, Raivio KO, Kekom#{228}ki MP. Liver adenine nucleo-

tides: fructose-induced depletion and its effect on protein synthesis.

Science i968;i6i: 1253-4.

77. Perheentupa J, Raivio K. Fructose-induced hyperuricaemia. Lancet

1967;2:528-31.

78. Stirpe F, Corte ED, Bonetti E, Abbondanza A, Abbati A,

Dc Stefano F. Fructose-induced hyperunicaemia. Lancet 1970;2:1310-i.

79. Romsos DR, Leveille GA. Effect of dietary fructose on in vitro and

in vivo fatty acid synthesis in the rat. Biochim Biophys Acta

1974;360: i-i 1.

80. Topping DL, Mayes PA. Effects of fructose concentration on ad-

enine nucleotide concentrations and pyruvate dehydrogenase activ-

ity of perfused rat liver. Biochem Soc Trans 1977;5:1001-2.81. Patzelt C, Loffler G, Wieland OH. Interconversion of pyruvate de-

hydrogenase in the isolated perfused rat liver. Eur J Biochem

i973;33: 117-22.

82. Zakim D, Herman RH. The effect of intravenous fructose and glu-

cose on the hepatic a-glycerophosphate concentration in the rat.

Biochim Biophys Acta 1968;165:374-9.

83. Bunch HB, Max P, Chyu K, Lowry OH. Metabolic intermediates

in liver of rats given large amounts of fructose or dihydroxyace-

tone. Biochem Biophys Res Commun i969;34:6i9-26.

84. Wieland 0, Matschinsky F. On the antiketogenic effect of glycerol

and fructose. (Zu den antiketogenen Wirkungen von glycerin und

fructose.) Life Sci 1962;2:49-54.

85. Exton JH, Park CR. Control of gluconeogenesis in liver III. Effects

of L-lactate, pyruvate, fructose, glucagon, epinephnine and adeno-

sine-3’5’-monophosphate on gluconeogenic intermediates in per-

fused rat liver. J Biol Chem 1969;244:1424-33.

86. McGarry JD, Takabayashi Y, Foster DW. The role of malonyl-CoA

in the coordination of fatty acid synthesis and oxidation in isolated

rat hepatocytes. J Biol Chem i978;253:8294-300.

87. Goodnidge AG. Regulation oflipogenesis. Stimulation of fatty acid

synthesis in vivo and in vitro in the liver of the newly hatched

chick. Biochem J 1970;i18:259-63.

88. Mayes PA, Felts JM. Liven function studied by liver perfusion. Proc

Eur Soc Study Drug Tox 1966;7:16-29.89. Mayes PA, Felts JM. Regulation of fat metabolism in the liver.

Nature 1967;215:716-8.

90. McGarny JD, Foster DW. Regulation of hepatic fatty acid oxidation

and ketone body production. Annu Rev Biochem 1980:49:395-420.

91. Topping DL, Mayes PA. The immediate effects of insulin and fruc-tose on the metabolism of the perfused liver. Changes in lipoprotein

secretion, fatty acid oxidation and estenification, lipogenesis and

carbohydrate metabolism. Biochem J 1972:126:295-311.

92. Witters LA, Monianity D, Martin DB. Regulation of hepatic acetyl

coenzyme A carboxylase by insulin and glucagon. J Biol Chem

1979;254:6644-9.

93. Borrebach R, Chnistiansen R, Chnistopherson BO, Bremer J. The

role ofacyltransferases in fatty acid utilization. Cinc Res i976;38:I-

16 to 21.

94. Declerecq PE, Debeen Ii, Mannaerts GP. Role of glycerol-3-phos-

phate and glycerophosphate acyltransferase in the nutritional con-

trol of hepatic tniacylglycerol synthesis. Biochem J 1982:204:247-

56.

95. Bates EJ, Topping DL, Sooranna SP, Saggerson D, Mayes PA.

Acute effects of insulin on glycerol phosphate acyltransferase ac-

tivity, ketogenesis and serum free fatty acid concentration in per-

fused liver. FEBS Lett 1977;84:225-8.

96. Jourdan MH. A difference between male and female baboons in

the serum glyceride response to rapid intravenous injections of

fructose. J Physiol 1969;20l:27P-8P.

97. Rozen P, Shafnir E. Comparison of changes in plasma free fatty

acids, glycerol and insulin following glucose and fructose loads.

Isr J Med Sci 1972:8:838-40.98. Rawat AK, Menahan LA. Antiketogenic action of fructose, glyc-

eraldehyde, and sorbitol in the rat in vivo. Diabetes 1975:24:926-

32.

99. Exton JH, Edson NL. The antiketogenic action of sorbitol. Biochem

J 1964;91:478-83.100. Ontko JA. Metabolism of free fatty acids in isolated liver cells. J

Biol Chem 1972:247:1788-800.

101. Sestoft L. Regulation of fructose metabolism in the perfused rat

liver. Interrelation with onthophosphate, glucose, ketone body and

ethanol metabolism. Biochim Biophys Acta 1974:343:1-16.

102. Mahurama Y, Macdonald I. Incorporation of orally administered

glucose-U’4C and fructose-U’4C into the triglyceride of the liver,

plasma and adipose tissue of rats. Metabolism 1973;22: 1205-15.103. Pereira JN, Jangaard NO. Different rates of glucose and fructose

metabolism in rat liver tissue in vitro. Metabolism i97i;20:392-400.

104. Gale MM, Crawford MA. Different rate of incorporation of glucose

and fructose into plasma and liver lipids in guinea pig. Metabolism

1969;i8: 1021-5.

105. Jourdan MH. The incorporation of fructose and glucose, given in-

travcnously, into the serum lipids of baboons. Nutr Metab 1972:

14:339-48.

106. Heinz F. The enzymes of carbohydrate degradation. Prog BiochemPharmacol 1973;8:1-56.

107. Fitch WH, Chaikoff IL. Extent and patterns of adaptation of en-

zyme activities in livers of normal rats fed diets high in glucose

and fructose. J Biol Chem i960;235:554-7.

108. Moser PB, Berdamier CD. Effect of early sucrose feeding on themetabolic patterns of mature rats. J Nutr i974;104:687-94.

109. Naismith DJ. Differences in the metabolism of dietary carbohy-

drates studied in the rat. Proc Nutn Soc 1971;30:259-65.

1 10. Vr#{225}naA, Roulin J, Loriette C, Kazdov#{225}L. Basal pyruvate de-

hydrogenase activity in the liver, adipose tissue and brain of rats

by guest on July 14, 2014ajcn.nutrition.org

Dow

nloaded from

METABOLISM OF FRUCTOSE 765S

with fructose-induced hypentniglycenidemia. Nutn Rep Int 1983;

28: i437-44.

1 1 1 . Shafnir E, Benchimol A, Orevi M. Hyperlipidemia and hepatic hy-

perlipogenesis in spiny mice �Acomys cahirinus) on carbohydratediets; relationship to peripheral unresponsiveness and low insulin

release. Isr J Med Sci 1975; 1 i :738-52.

1 12. Winder WW, Booth FW, Fitts RH, Holloszy JO. Effect of exercise

on response of liver lipogenic enzymes to a high fructose diet. Proc

Soc Exp Biol Med 1975:148:1150-i 154.I 13. Waterman RA, Romsos DR. Tsai AC, Miller ER, Leveille GA.

Effects of dietary carbohydrate source on growth, plasma metab-

olites and lipogenesis in rats, pigs and chicks. Proc Soc Exp Biol

Med 1975:150:220-5.1 i4. Bruckdorfer KR, Khan IH, Yudkin J. Fatty acid synthetase activity

in the liver and adipose tissue of rats fed with various carbohy-

drates. Biochem J 1972; 129:439-46.

1 15. Sturton G, Pritchard PH, Han LY, Bnindley DN. The involvement

of phosphatidate phosphohydrolase and phospholipase A activities

in the control of hepatic glycerolipid synthesis. Biochem J

1978; 174:667-70.

1 16. Sagawa-Katayama Y, Monita N. Effects of a high fructose diet on

lipogenic enzyme activities in some organs of rats fed ad libitum.

J Nutr 1975;105:1377-83.

117. Griffaton G, Rozen R, Ardouin B, Lowy R. Effects of lipogenic

precursors in rat liver, according to the nature of the dietary car-

bohydrate. Enzyme 1974;i7:319-32.

1 i8. Zakim D. The effect of fructose on hepatic synthesis of fatty acids.

Acta Med Scand 1972:542:205-14.

i 19. Zakim D, Pardini RS, Herman RH, Sauberlich HE. Mechanism for

the differential effects of high carbohydrate diets on lipogenesis in

rat liver. Biochim Biophys Acta 1967;i44:242-51.

120. Chnistophe J, Mayer J. Influence of diet on utilization of glucose

and incorporation of acetate-i-C’4 into liven fatty acid and choles-

terol in rats. Am J Physiol 1959;197:55-9.

121. Fabry P, Poledne R, Kazdov#{225} L, Braun T. The effect of feeding

frequency and type of dietary carbohydrate on hepatic lipogenesis

in the albino rat. Nutr Dicta 1968;i0:81-90.

122. Teppenman J, Tepperman HM. Gluconeogenesis, lipogenesis and

the Sherningtonian metaphor. Fed Proc 1970;29:1284-93.

123. Chevalier MM, Wiley JH, Leveille GA. Effect of dietary fructose

on fatty acid synthesis in adipose tissue and liven of the rat. J Nutr1972; 102:337-42.

124. Romsos DR, Leveille GA. Effect of dietary fructose on in vitro and

in vivo fatty acid synthesis in the rat. Biochim Biophys Acta

1974;360: 1-1 i.i25. Maruhama Y, Macdonald I. Some changes in the triglyceride me-

tabolism of rats on high fructose or glucose diets. Metabolism

1972:21:835-42.

126. Gardner RS, Mayes PA. The singular and combined effects of acute

fructose loading and adaptation to a high sucrose diet on hepatic

very-low-density lipoprotein secretion. Biochem Soc Trans

i974;2:1 127-8.

127. Macdonald I. Effects of dietary carbohydrate on serum lipids. Prog

Biochem Pharmacol 1973;8:216-41.

128. Carlson LA, Bottiger LE. Ischaemic heart disease in relation to

fasting values of plasma triglycerides and cholesterol. Lancet

i972;i :865-8.

129. Cnyer A, Riley SE, Williams ER, Robinson DS. Effects of fructose,

sucrose and glucose feeding on plasma insulin concentrations and

on adipose-tissue cleaning factor lipase activity in the rat. Biochem

J i974;i40:56i-3.130. Reiser 5, Hallfnisch J. Metabolic effects of dietary fructose. Boca

Raton, FL: CRC Press, 1987.131. Henry RR, Crapo PA. Current issues in fructose metabolism. Annu

Rev Nutr 199i;1i:21-39.132. Perheentupa J, Raivio K. Fructose-induced hyperunicaemia. Lancet

i967;2:528-31.

i33. Narins RG, Weisberg JS, Meyers AR. Effects of carbohydrates on

uric acid metabolism. Metabolism 1974:23:455-65.

134. Van den Berghe G, Bnonfman M, Vanneste R, Hers HG. The mech-

anism of adenosine tniphosphate depletion in the liver after a load

of fructose. A kinetic study of liver adenylate deaminase. Biochem

J 1977;162:601-9.

135. Van den Berghe G, Bontemps F, Hers HG. Punine catabolism in

isolated rat hepatocytes. Influence of coformycin. Biochem J

1980; 188:913-20.

136. Vincent MF, Van den Berghe G, Hers HG. The pathway of adenine

nucleotide catabolism and its control in isolated rat hepatocytes

subjected to anoxia. Biochem J 1982:202:117-23.

137. Murray RK, Granner DK, Mayes PA, Rodwell VW. Harper’s bio-

chemistry. 23rd ed. Norwalk, CT: Appleton and Lange 1993:363-77.138. Bode JC, Zelder 0, Rumpelt HJ, Wittkamp U. Depletion of liver

adenosine phosphates and metabolic effects of intravenous infusion

of fructose or sorbitol in man and in the rat. Eur J Clin Investi973;3:436-41.

by guest on July 14, 2014ajcn.nutrition.org

Dow

nloaded from