European Journal of Clinical Nutrition (2000) 54, Suppl 1, S69±S74

ß 2000 Macmillan Publishers Ltd All rights reserved 0954±3007/00 $15.00

www.nature.com/ejcn

Drugs±nutrient interactions: a potential problem during

adolescence

E Alonso-Aperte and G Varela-Moreiras*

Facultad de CC. Experimentales y TeÂcnicas, Universidad San Pablo-CEU, 28668 Boadilla del Monte, Madrid, Spain

The concept of drug ± nutrient interactions is not new, but it has only recently gained currency in medicine.
Although the elderly are normally considered to be at particular risk, other groups may also be at risk: infants,
adolescents, pregnant women, alcohol and tobacco users, etc.
In infants and adolescents there are several factors that may in¯uence the possible interactions: ®rstly, nutrient
needs are usually higher, mainly micronutrients; systems for detoxi®cation of anutrients are not complete; the
tendency to restricted diets (especially girls) that are unable to cover the actual recommended intakes for a
number of micronutients (i.e. vitamins); and the dangerous increase in alcohol consumption either in males or
females.
Administration of drugs in population with adequate vitamin intake is usually not a problem, but adminis-
tration of drugs in those with borderline intake of vitamins or in patients with low nutritional status can result in
symptomatic vitamin de®ciency states. The groups at risk of poor vitamin status are smokers (a high proportion
of adolescents are active smokers); dieters (skipping meals and dieting to lose weight frequently compromise
micronutrient intake, and it should be considered that it is extremely dif®cult to meet all the requirements at
intakes of less than 1200 calories per day), oral contraceptive users, and pregnant and lactating women, excessive
alcohol users, etc. The chapter also focuses on the case of folate: rapidly dividing tissues during the adolescent
growth spurt increase requirements for folate. Because of this increased need, folate status appears to be of
concern during the age of this rapid growth. A variety of drugs are known to interfere with vitamin utilization by
blocking or altering transformation of the vitamin to its metabolically active form. Serum folate levels are known
to be low in a high percentage of patients with rheumatoid arthritis, suggesting that aspirin alters the transport of
folate by competition for binding sites on serum proteins. Methotrexate, a drug commonly used at low doses for
the treatment of psoriasis, rheumatoid arthritis and certain liver disorders, limits the availability of methyl groups
derived from one-carbon metabolism by inhibiting competitively a key enzyme in the intracellular folate
metabolism. In humans, the antiepileptic drug valproic acid (VPA) is associated with two major adverse effects:
teratogenicity and folate de®ciency. The mechanisms by which VPA exerts the teratogenic or antifolate effect
remain unclear, but an alteration in the methionine cycle is the strongest hypothesis proposed.
Descriptors: drugs-nutrient interactions; adolescents; xenobiotics; anutrients; vitamins; de®ciency
European Journal of Clinical Nutrition (2000) 54, Suppl 1, S69±S74

Introduction environment in which we live contains a large number of

non-nutrient chemicals of different nature and origin.
In the world at large and in the Western World in particular,
Therefore, a number of these compounds may be intro-
the average age of the population is increasing, which
duced into the body either by design or by accident that are
results from an increase in lifespan as a consequence of
not normal constituents of the body. These non-nutrients
the advances in preventive medicine and the clinical
include food additives, such as preservatives, antioxidants,
sciences. At the same time, there has been a remarkable
sweeteners, ¯avors, and colors, therapeutic drugs, cos-
rise in the modern pharmaceutical and chemical industries.
metics, detergents, pesticides, and industrial chemicals.
In fact, during the past two to three decades, awareness has
These substances are `foreign' to the major metabolic
grown about factors, including food and nutrients, that can
pathways of the body concerned with the metabolism of
affect pharmacological activity. The disposition of half-life
nutrients, and have variously been called `foreign com-
drugs can also be affected by malnutrition, and the compo-
pounds', `xenobiotics' or `anutrients'. In addition to these
sition of the diet itself may contribute to the incidence of
synthetic foreign compounds, a wide variety of plant food-
nutrient ±drug interactions (Knapp, 1996).
stuffs contain anutrients belonging to various chemical
The concept of drug ±nutrient interactions is not new,
groups, such as alkaloids, ¯avonoids, and others. The
but it has only recently gained currency in medicine. The
variety of these natural anutrients that enter the body is
much greater than the number of drugs, food additives or
*Correspondence: G Varela-Moreiras, Facultad de CC. Experimentales y insecticides. However, both the natural and arti®cially
TeÂcnicas, Universidad San Pablo-CEU, Urb. MonteprõÂncipe, Crta. Boadilla occurring compounds in food are potentially toxic if they
del Monte km 5, 3, 28668 Boadilla del Monte, Madrid, Spain. are consumed or if they accumulate in the body (Roe,
E-mail: gvarela@ceu.es 1990).
Guarantor: G. Varela-Moreiras.
Contributors: Dr G. Varela-Moreiras initiated the study, collected the data,
Drug ±nutrient interactions can be de®ned as events that
designed the draft and wrote the ®rst and ®nal version. Dr E. Alonso- result from chemical, physical, physiological or patho-
Aperte contributed to the writing and revision of the manuscript. physiological relationships between nutrients and drugs.
 Drug ± nutrient interaction
E Alonso-Aperte and G Varela-Moreiras
S70
Although normally interactions are related to adverse
effects such as impaired drug absorption by foods, altered
drug metabolism, or drug-induced nutrition de®ciencies,
several bene®cial effects may also occur: drug absorption
enhanced by foods, `appetite stimulation', improved nutri-
tion with antibiotics in infection, or the use of vitamins as
drug toxicity antidotes. These events are important when
they diminish the intended response to a therapeutic drug.
when the nutritional status of an individual is impaired, or
when it causes an acute or chronic drug toxicity (Roe,
1985).
Drugs and nutrients share a number of characteristics,
such as physicochemical properties, the capability to affect
certain biochemical events, and dose-related toxicity.
Nutrient and drug interactions can occur in the gastro-
intestinal tract, in the blood, and at cellular drug receptors.
Some individuals are at greater risk of nutrient ±drug
interactions. Although the elderly are normally considered
to be at particular risk, other groups may also be at risk:
infants, adolescents, pregnant women, alcohol and tobacco
users, etc. (Hamilton Smith, 1995). In infants and adoles-
cents there are several factors that may in¯uence the
possible interactions: ®rstly, nutrient needs are usually
higher, mainly of micronutrients; systems for detoxi®cation
of anutrients are not complete; the tendency to restricted
diets (especially girls) that are unable to cover the actual
recommended intakes for a number of micronutients (i.e.
vitamins); the dangerous increase in alcohol consumption
either in males or females.
Therefore, instruction in nutritional science can no
longer be restricted to a description of the chemistry of
major dietary constituents, diseases associated with a de®-
ciency, and the amounts of nutrients required to prevent
them. Nutrition today must address the interrelationships of
long-term nutritional habits and chronic diseases of the
cardiovascular system, of cancer, and of osteoporosis,
among others. There is also the role of nutrition as a tool
in the treatment of postoperative and other patients in the
clinical setting. It is at these interfaces that drugs and
nutrients interact in signi®cant ways. The advances and
application of modern drug therapy and of nutrition are not
unrelated. A well-nourished individual is less likely to
contract disease, and when speci®c diseases are encoun-
tered, they tend to be less severe, and response to drug
therapy more favorable. Moreover, response to drug ther-
apy may differ in the same individual at different time
points. A signi®cant proportion of these changes appears to
be a re¯ection of the nutritional status of the individual and
the food that is being consumed at the time of the drug
therapy.
Pharmacological agents (drugs)
In addition to the food-borne chemicals mentioned above,
drugs are another class (the most important but not the
largest) of a group of substances that are chemically foreign
to the body and are generally of no nutritive value. During
the last 50 y medicine has experienced a profound trans-
formation, with the ability to cure most diseases. Unfortu-
nately, however, the interactions between what man eats
and the chemicals with which he comes into contact are
ever increasing, and yet are poorly understood. It has
become increasingly recognized that nutritional status is
one of the many factors capable of modifying the pharma-
cological effect of drugs. On the other hand, drug treatment
European Journal of Clinical Nutrition
may impair the nutritional status of a patient, with con-
sequent profound effects upon the body's response to
drugs. Interactions occur not only with prescribed drugs
but also to over-the-counter drugs such as aspirin, para-
cetamol or antacids. Therefore, the problem involves both
healthy individuals and those suffering from disease (Basu,
1983a, b).
The fate of anutrients in the body
Mammals have developed defence mechanisms that render
them capable of metabolizing and subsequently excreting a
wide range of anutrients. The metabolism of a drug
involves different steps: absorption from the gastrointest-
inal tract, if orally administered; transport in the blood,
usually bound to plasma proteins; deactivation, ®rstly
through oxidation by microsomal enzyme systems
NADPH and cytochrome P450 in liver, lung, and small
intestine, and secondly by conjugation with glucuronic
acid, sulphate or glycine; excretion of the conjugate in
urine or bile (Guengerich, 1995).
The breakdown of drugs and other environmental
chemicals takes place in two different stages: ®rstly, the
biotransformation, which is asynthetic and results in the
formation of a more polar compound by oxidation, reduc-
tion, hydroxylation, etc. The second phase is of a synthetic
nature, involving conjugation of the polar compound with
endogenous molecules such as glucuronic acid, sulphate,
glutathione, glycine and acetate, giving a more hydrophilic
character and making the compound more suitable for
excretion in the bile and urine. The enzyme system respon-
sible for the metabolism of anutrients is the mixed function
oxidases (MFOs), principally located in the endoplasmic
reticulum of the hepatocytes, and is somewhat unique in its
ability to utilize a wide range of substrates.
Dietary and nutritional factors affecting the metabolism
of xenobiotics
Dietary factors affecting drug disposition and metabolism
include: dietary protein intake and its effects on plasma
albumin levels, drug-metabolizing enzymes in gut and
liver, and intestinal micro¯oral; meal composition can
alter intestinal activity and blood ¯ow to accelerate or
retard drug absorption=metabolism; excess body fat may
increase the volume of distribution of lipophilic drugs, etc.
The rate at which the metabolic reactions proceed and
their relative importance may be affected by a number of
factors: genetics, age, hormones, disease, and others, but
probably the most important factor of all is nutritional
status.
As an example (Basu, 1983a, b), protein-energy malnu-
trition due to restricted feeding (i.e. extreme slimming
diets) and starvation is prevalent in much of the world's
population. This nutritional problem can affect the meta-
bolism of drugs and other xenobiotic chemicals in at least
two ways: ®rstly, tissue protein is catabolized and used as a
source of energy, reducing the availability of amino acids
for protein synthesis which, in turn, reduces the amount of
various enzymes in the tissues, including those participat-
ing in the metabolism of foreign compounds. Second,
endogenous substrates such as glucuronide, sulphate, and
glycine are derived from carbohydrate and protein, and
there could be competition between the metabolism of
xenobiotics and the needs of the tissues for these nutrients.

Nutritional consequences of drug therapy
As publicised, there has been a growing interest in how
treatments with drugs of widely differing chemistry and
pharmacological action have the potential to cause nutri-
tional de®ciencies. The actual incidence of drug-induced
nutrient de®ciency is, indeed, dif®cult to identify, since
cause and effect relationships between the drug and the
nutrient de®ciency are dif®cult to prove. However, what is
clear is that certain speci®ed drugs induce de®ciencies of
speci®c nutrients following long-term, high-dose adminis-
tration in susceptible groups of the population. Many of
these drugs may alter taste perception, reduce absorption,
increase excretion or interfere with the utilization of nutri-
ents. The widespread use of certain drugs, normally self-
prescribed, makes it impossible to give advice for each
individual, and therefore for several nutrients possible
interactions with drugs should be considered in relation to
needs on a community basis (Thomas and Tschanz, 1994).
Drug effects on food intake, body weight, nutrient
requirements and growth
Drugs can reduce food intake because they cause a loss of
appetite, induce sedation, or evoke adverse responses when
food is taken. Drugs affecting appetite may do so by central
or peripheral effect. (Welling, 1996):
A. Primary Ð appetite suppression
1. Centrally acting:
 cathecolaminergic, e.g. dextroampethamine;
 dopaminergic, e.g. levodopa;
 seratoninergic, e.g. fen¯uramine;
 endorphin modulators, e.g. naloxone.
2. Peripherally acting:
 agents that inhibit gastric emptying, e.g. levodopa;
 bulking agents, e.g. methyl cellulose.
B. Secondary Ð adverse response to food Ð loss of
appetite
 drugs causing nausea and vomiting, e.g. digoxin-toxin
dose,
 drugs causing loss of taste, e.g. penicillamine;
 drugs causing stomatitis, e.g. ¯uorouracil;
 hepatotoxic agents, e.g. alcohol.
Drugs affecting absorption
The majority of clinically signi®cant nutrient ±drug inter-
actions seem to involve the absorption process. A variety of
drugs may interfere with nutritional status by impairing the
absorption and therefore bioavailability of essential nutri-
ents by several mechanisms:
Drug ± nutrient interaction
E Alonso-Aperte and G Varela-Moreiras
S71
Indirect mechanisms
 Drug-induced changes in gastrointestinal motility (e.g.
anticholinergics).
 Drug-induced malabsorption syndromes (e.g. neomy-
cin).
Direct mechanisms
 Drug-induced pH changes in gastrointestinal tract
(antacids).
 Drug-induced changes in bioavailability.
 Drug-induced retardation of absorption.
Bioavailability and metabolism
A drug bioavailability is de®ned as the degree to which the
drug can be utilized metabolically unchanged once it
reaches the systemic circulation. Several factors that can
affect drug disposition are: diet; nutritional status; genetic
traits; age; sex; pregnancy; renal and cardiovascular func-
tion; pharmacologic variables (dose, route, duration);
stress; diseases states, etc. (Skaar, 1991).
Drug metabolism involves two basic reactions: phase I and
phase II reactions. Phase I includes oxidation, hydroxyl-
ation, reduction, or hydrolysis, resulting in changes in a
functional group of the drug molecule. The mixed function
oxidase system (MFOS) in an inducible enzyme system
catalyzes the oxidation of a wide variety of drugs. The
MFOS is found primarily in the liver. Phase II reactions
include conjugation to glucuronate or glutathione, and
acetylation or sulphonation to functional groups on the
drug's molecule. The metabolism of drugs by phase I and
phase II reactions is catalyzed by different enzymes, and
the formation of metabolites requires other substances
provided by the body through nutrition. A number of
nutrients and micronutrients (vitamins) play signi®cant
roles in phase I oxidation reactions (Table 1).
On the other hand, phase II reactions involving conjuga-
tion depend of the supply of carbohydrates, amino acids,
fats, and proteins.
Interactions of drugs and vitamins
Daily requirements of the vitamins are often expressed as
the minimum amount essential for health. The minimum
requirements, however, may be insuf®cient when one is
under physiological and pathological stresses. This is the
case of pregnancy and lactation or when training for
athletic competition. In pathological states, requirements
for some or all the vitamins may be increased. There is also
a growing interest in knowing how drugs may potentially
cause vitamin de®ciency, ranging from overt depletion with
Table 1 Nutrients in Phase I (Oxidation) reactions

 altering the environment of the intestinal lumen Ð Nutrients
mineral oil, cholestyramine, antacids; Nicotinic acid
 morphology of the mucosa Ð neomycin, methotrexate; Ribo¯avin
 inhibiting the digestive enzymes Ð sulphasalazine; Glycine
 interference with the metabolism of one nutrient, which Iron
Protein
in turn leads to malabsorption of another nutrient Ð Calcium
phenitoyn, phenobarbital, prednisone, etc. Zinc
Magnesium
We may also differentiate the type of interaction on
absorption by indirect and direct mechanisms. Source: Hoyumpa and Schenker (1982).
Component of reaction requiring nutrient
NADPH
FAM and FAD
Heme (cytochrome P-450)
Heme
Apoenzymes
Maintenance of membranes
Maintenance of membranes
Maintenance of membranes

European Journal of Clinical Nutrition

 Drug ± nutrient interaction
E Alonso-Aperte and G Varela-Moreiras
S72
clinical manifestation to subclinical de®ciency. This will Table 4 Effects of drugs on Vitamin assessment
depend on the nutritional status before and during the
Nutrient Drug Alteration
treatment, the type, dosage and duration of drug use, and
the age of the individuals. Therefore, administration of Thiamin Oral contraceptives Increased transketolase stimulation
drugs in population with adequate vitamin intake is usually Vitamin B6 Oral contraceptives Decreased plasma levels
not a problem, but administration of these drugs in those Folic acid Oral contraceptives Reduced serum folate
Folic acid Anticonvulsants Reduced serum folate
with borderline intake of vitamins or in patients with low Folic acid Sulfasalzine Reduced serum folate
nutritional status can result in symptomatic vitamin de®- Cobalamin Oral contraceptives Decreased serum B12
ciency states. Roe (1985) attempted to make several recco- Vitamin C Aspirin Decreased plasma vitamin C
mendations for different vitamin intakes according to drugs
treatments (Table 2). However, cause and effect relation-
ships between drug and vitamin de®ciency are dif®cult to negative impacts on micronutrient status. In consequence,
prove. As a consequence, the drug-induced vitamin de®- diets that eliminate all animal foods provide almost no
ciencies may, in turn, have an enormous effect on the vitamin B12, and have also been associated with cases of
metabolism of drugs and therefore the body's response to rickets and other vitamin de®ciencies. Skipping meals and
the agents (Table 3). dieting to lose weight frequently compromise micronutrient
Drugs can affect vitamin status either directly or intake, and it should be considered that it is extremely
indirectly: alterations in absorption, metabolism, and excre- dif®cult to meet all the requirements at intakes of less than
tion are examples of a drug's direct effect on vitamin status. 1200 calories per day. As an example, when an analysis of
A drug can also induce a vitamin de®ciency indirectly by 11 published reducing diets was done, it was shown that
altering appetite and taste, gastrointestinal ¯ora, and the none provided 100% of the US RDA for representative
rate of stomach emptying. In addition, drugs may interfere vitamins.
with vitamin assessment, leading to a more cumbersome
task to identify the potential interaction (Table 4).
3. Oral contraceptive users. The use of oral contracep-
tives is associated with decreased levels of some vitamins:
Groups at risk of poor vitamin status plasma carotene and pyridoxal phosphate levels, and red
1. Smokers. Although there is an increasing advice by the blood cell folate. Oral contraceptive users also have lower
health authorities concerning the negative effects of smoking, vitamin C levels.
the number of smokers is still large, mainly in the adolescent
age group. Besides the well-known consequences of heavy
smoking, our concern is related to micronutrient status, since 4. Adolescents. During periods of rapid growth, particu-
it has been shown that smokers tend to have a poorer status. larly during adolescence, high nutrient demands may not be
Smokers tend to have lower blood levels of B-carotene and met with a typical diet. Red blood cell folate was at a
pyridoxal-50 -phosphate. People who smoke also appear to de®cient level in 47% of a group of apparently healthy
have an increased requirement for vitamin C, and were also teenage girls in Alabama (Clark et al, 1987). Moreover,
found to have signi®cantly lower levels of vitamin E in lung mean dietary vitamin A intake of Black teenage males in
alveolar ¯uid vs non smokers. Kentucky was only 36% of the US RDA, and White
teenage females consumed about 51% of the US RDA of
that nutrient. There are, of course, a high number of
2. Dieters. Eating behavior, particularly diets that omit or examples showing that in many groups of adolescents
severely restrict whole categories of foods, can have there are subclinical de®ciencies, mainly for vitamins.
While nutrient demands are particularly high during pub-
Table 2 Vitamin intakes for individuals receiving drugs which cause erty, eating habits may be especially poor, which may be
malabsorption due in part to reducing diets.
Drug Recommended total daily vitamin intake

Phenytoin Vitamin A: 800 ± 1200 UI 5. Pregnant and lactating women. Micronutrient needs
Vitamin K: 2 ± 5 mg during pregnancy and lactation are higher and frequently
Folic acid: 0.8 ± 1.2 mg diets do not cover these demands.
Sulfasalazine Folic acid: 0.8 ± 1.2 mg
Cholestyramine Vitamin A: 5000 ± 10,000 UI
Colestipol Vitamin D: 800 ± 1200 UI
6. Premature infants. Premature infants have been found
Source: Roe (1985). to have low plasma vitamin C levels, low vitamin A status,

Table 3 Effect of Vitamin de®ciency on some drug metabolizing enzyme activities in animals

Vitamin de®ciency Affected drug metabolism indices

A Cytochrome P450, oxidations of aminopyrine, ethylmorphine, aniline, benzo(a)pyrene.

B1 Oxidations of N-nitrosodimethylamine, acetaminophen, aniline.
B2 NADPH-P450 reductase, oxidations of aniline, acetanilide, benzo(a)pyrene, aminopyrine.
B3 Oxidation of anaesthesics.
C Cytochrome P450, NADPH-P450 reductase and several mono-oxygenation activities.
E Oxidations of codeine, ethylmorphine, benzo(a)pyrene.
Folate Decrease in induction of cytochrome P450 by barbiturates.

Source: adapted from Guengerich (1995).

European Journal of Clinical Nutrition


and a tendency to vitamin E de®ciency. The nutrients of
particular concern are the fat soluble vitamins, folic acid,
vitamin C, and vitamin B6.
The case of folate
Accelerated growth and increased lean body mass are the
characteristics of adolescents. Rapidly dividing tissues
during the adolescent growth spurt increases requirements
of most nutrients and especially for folate. Because of this
increased need, folate status appears to be of concern
during the age of this rapid growth, mainly in low-
income families, whose diets are likely to be poor in quality
(Sauberlich, 1990). A variety of drugs are known to
interfere with vitamin utilization by blocking or altering
transformation of the vitamin to its metabolically active
form.
Some drugs have been associated with subnormal serum
folate levels, probably due to the speci®c but weakly folate
bound to serum proteins and, therefore, its vulnerability to
be displaced by drugs (Lambie & Johnson, 1985); although
the mechanisms remain to be elucidated, the common
denominator seems to be a decrease in serum binding of
methyltetrahydrofolate, caused by the drug or one of the
metabolites (Labadarios, 1993).
Serum folate levels are known to be low in a high
percentage of patients with rheumatoid arthritis. In fact,
decreased folate levels have been observed in 70% of the
rheumatoid patients, suggesting that aspirin alters the
transport of folate by competition for binding sites on
serum proteins. Acetaminophen (ACAP) is an analgesic
agent employed in many self-prescribed formulations for
pain relief and, as aspirin, has the potential for chronic use
at all age groups. The two major disposition pathways of
ACAP involve glucuronidation or sulphation, and subse-
quent urinary excretion of polar metabolites. The sulpha-
tion pathway becomes saturated as the dose of ACAP
increases: as known, methionine is required for different
physiological functions other than as a sulphur source for
cysteine. The reactive metabolite of methionine, S-adeno-
sylmethionine, provides the methyl groups found in DNA,
RNA, etc. In consequence, high ACAP intakes may make
methionine less available for protein synthesis and methyl-
ation reactions. We have demonstrated in rats that long-
term ACAP administration leads to a signi®cantly lower
dietary protein metabolism when compared to the control
group (Varela-Moreiras et al, 1991; Ragel Prudencio et al,
1994). In addition, when we fed rats for 4 weeks with
aspirin or acetaminophen the treatments were not asso-
ciated with important changes in total hepatic folate content
or folate derivatives distribution. DNA methylation or the
so-called `methylation ratio' was also unaffected, which
implies that in an animal model, long-term administration
of these very popular analgesics seem to be without effect
on folate metabolism and on several markers of the
methionine cycle for which folate pendes a critical regula-
tory standpoint (Varela-Moreiras et al, 1993).
Methotrexate (MTX), a drug commonly used at low
doses for the treatment of psoriasis, rheumatoid arthritis
and certain liver disorders, limits the availability of methyl
groups derived from one-carbon metabolism by inhibiting
competitively a key enzyme in the intracellular folate
metabolism. MTX acts to trap folate and is considered an
antifolate agent (Selhub J et al, 1991). In a recent study we
demonstrated that treatment with MTX to rats at low doses
Drug ± nutrient interaction
E Alonso-Aperte and G Varela-Moreiras
S73
for 2 weeks resulted in a signi®cantly lower brain total
folate content vs control group (0.25 ‡ 0.06 vs 0.69 ‡ 0.5,
P < 0.01; Varela-Moreiras et al, 1995). This quantitative
change was associated with elongations of the glutamate
chains of the folate molecules, mainly an increase in the
proportion of hepta- and octaglutamyl folates in the MTX-
treated animals. However, MTX could not be detected in
brain tissue, in contrast to previous observations of sig-
ni®cant amounts found in liver (Alonso Aperte & Varela-
Moreiras, 1996). Interestingly, DNA was signi®cantly
(P < 0.05) undermethylated in treated rats. In consequence,
this drug appears to impair the capacity of tissues, either a
peripheral one such as liver or a central one such as brain,
to incorporate folate.
In humans, the antiepileptic drug valproic acid (VPA) is
associated with two major adverse effects: teratogenicity
and folate de®ciency (Smith & Carl, 1982): The mechan-
isms by which VPA exerts the teratogenic or antifolate
effect remain unclear. An alteration in the methionine cycle
is the strongest hypothesis proposed. Wegner and Nau
(1992) have demonstrated that VPA alters embryonic
folate distribution in mice, producing elevated levels of
tetrahydrofolate and reduced levels of both formylated
forms of folate. Folate distribution is highly dependent on
the functioning of the methionine cycle. These previous
observations suggest a link between VPA and the methio-
nine cycle that could be related to the biochemical mechan-
isms involved in VPA teratogenesis, although the speci®c
effect of VPA on the reactions involved in the aforemen-
tioned cycle has not been studied. In consequence, we have
recently evaluated the in¯uence of VPA administered
during neural tube formation in the rat on the methionine
cycle. VPA induced a reduction in methionine serum
concentration (P < 0.05) caused by a 24% reduction of
methionine synthase activity. This provoked hepatic DNA
hypomethylation, and both, impaired methionine synthesis
and DNA hypomethylation may be involved in VPA-
induced teratogenesis (Alonso-Aperte et al, 1999). In
addition, we also observed that VPA alters normal gestation
reducing the fertility index, and induces skeletal modi®ca-
tions on the skull, appendicular bones, vertebrae and ribs
(Ubeda et al, 1996).
Alcohol
Ethanol causes many pharmacological effects. Alcoholic
drinks are consumed for their mood-altering actions and
pleasing tastes, and in moderate amounts for stimulating
appetite.
Vitamin de®ciencies occurring in heavy alcohol drinkers
are complex in their etiology. Causes of de®ciency include
dietary insuf®ciency, malabsorption, hyperexcretion, and a
decreased rate of activation. Of all the vitamins, B-vitamin
de®ciencies are the most common among alcohol drinkers.
We should not forget that adolescents are now dramatically
increasing the alcohol intake while at the same time are
borderline with regard to several vitamins intakes, mainly
due to prolonged slimming diets, poor diets, or wrong
choice of food (Hoyumpa & Schenker, 1982).
Thiamin de®ciency is very frequent among alcohol
drinkers, producing two disorders of the central nervous
system. Folic acid de®ciency also occurs in alcoholics.
Alcohol with empty calories often replaces food in the
diet. It can cause in¯ammation of the stomach, pancreas
and intestine, and interferes with the normal process of
European Journal of Clinical Nutrition
 Drug ± nutrient interaction
E Alonso-Aperte and G Varela-Moreiras
S74
digestion and absorption. The cytoplasm alcohol dehydro-
genate (AHD) requires vitamins such as thiamin, ribo¯avin,
nicotinic acid and pentatonic acid. Prolonged excessive
alcohol consumption can therefore increase the body
requirements for these vitamins.
The microsomal ethanol oxidizing system (MEOS) is an
alternative pathway for the oxidation of alcohol to acet-
aldehyde, taking place especially when the ADH pathway
is exhausted due to the shortage of the B-vitamins. The
metabolic product, acetaldehyde, can interfere with the
activation of vitamins in the liver, the vitamins particu-
larly affected being, thiamin, vitamin B6, folic acid and
vitamin D.
Symptoms of vitamin A de®ciency (night blindness)
accompanied by low circulatory levels of the vitamin
have been found among alcoholics with liver damage.
Retinol is oxidized to retinalaldehyde (the active form of
vitamin A) by alcohol dehydrogenase. This latter enzyme
has a 50-fold greater af®nity for ethanol than retinol.
Therefore, it appears that a possible explanation mechan-
ism could be a competitive inhibition of retinal formation
by ethanol.
Finally, chronic alcoholics have also been reported to be
more susceptible to bone fracture than non-alcoholics.
Association of alcoholism with bone fracture is attributed
to abnormal vitamin D metabolism. The liver is important
for both storage and metabolism of vitamin D, and inter-
ference with its function by alcohol can cause secondary
vitamin de®ciency.
References
Alonso Aperte E, Varela-Moreiras G (1996): Brain folates and DNA
methylation in rats fed a choline de®cient diet or treated with low doses
of methotrexate. Int. J. Vit. Nutr. Res. 66, 232 ± 236.
Alonso-Aperte E, Ubeda N, AchoÂn M, PeÂrez-Miguelsanz J & Varela-
Moreiras G (1999): Impaired methionine synthesis and hypomethylation
in rats exposed to valproate during gestation. Neurology 52, 750 ± 756.
Basu TK (1983a): Effects of protein malnutrition and ascorbic acid levels
on drug metabolism. Can. J. Physiol. Pharmac. 61, 295 ± 301.
Basu TK (1983b): Avitaminosis and congenital malformation. Int. J. Vit.
Nutr. Res. 24(Suppl), 9 ± 14.
Clark AJ, Mossholder S, Gates R (1987): Folacin status in adolescent
females. Am. J. Clin. Nutr. 46, 302 ± 306.
Guengerich FP (1995): In¯uence of nutrients and other dietary materials on
cytochrome P450 enzymes. Am. J. Clin. Nutr. 61(Suppl) 651S ± 658S.
Hamilton Smith C (1995): Drug ± Food=Food ± drug interactions. In: Ger-
iatric Nutrition 2nd edn., JE Morley, Z Glick, LZ Rubensaein. New
York: Raven Press.
European Journal of Clinical Nutrition
Hoyumpa AM & Schenker S (1982): Major drug interactions: effect of
liver disease, alcohol and malnutrition. A. Rev. Med. 33, 113 ± 149.
Knapp HR (1996): Nutrient ± drug interactions. In: Present Knowledge in
Nutritioned EE Ziegler, LJ Filer Jr, pp 530 ± 539. Washington, DC: ILSI
Press.
Labadarios D (1993): The effect of chronic drug administration on folate
status and drug toxicity. In: Food, Nutrition and Chemical Toxicity DV
Parks, C loannides, R Walker pp 71 ± 80. London: Smith-Gordon.
Lambie DG & Johnson RH (1985): Drugs and folate metabolism. Drugs
30, 145 ± 155.
Ragel Prudencio MC, Ruiz-Roso B & Varela-Moreiras G (1994): Effects
of short and long-term administration of acetylsalicylic acid, acetami-
nophen or phenacetin on diet utilization in rats. Nutr. Res. 14, 399 ± 410.
Roe DA (1985): Prediction of the cause, effects, and prevention of drug ±
nutrient interactions using attributes and attribute values. Drug ± Nutri-
ent Interactions 3, 187 ± 189.
Roe DA (1990): Drug ± folate interrelationships: historical aspects and
current concerns. In: Folic Acid Metabolism in Health and Disease, ed.
MF Picciano, ELP Stokstad, JF Gregory III, New York: Wiley ± Liss.
pp 277 ± 287.
Sauberlich HE (1990): Evaluation of folate nutrition in population groups.
In: Folic Acid Metabolism in Health and Disease, MF Picciano, ELR
Stokstad, JF Gregory, Sauberlich HE pp 212 ± 235. New York: Wiley ±
Liss.
Selhub J, Seyoum E, Pomfret EA & Zeisel SH (1991): Effects of choline
de®ciency and methotrexate treatment upon liver folate content and
distribution. Cancer Res. 51, 16 ± 21.
Skaar DJ (1991): Drug ± nutrient interactions: Implications for pharmaceu-
tical care. Partners Pharm. Care Oct: 11 ± 19.
Smith DB & Carl GF (1982): Interactions between folates and carbam-
azepine or valproate in the rat. Neurology 32, 965 ± 969.
Thomas JA, Tschanz C (1994): Nutrient ± drug interactions. In: Nutritional
Toxicology, ed. FN Kotsonis, M Mackey, J Hjelle New York: Raven
Press.
Ubeda N, Alonso E, Martin-Rodriguez JC, Varela-Moreiras G, Puerta J &
PeÂrez-Miguelsanz J (1996): Valproate-induced developmental modi®-
cations maybe partially prevented by coadministration of folinic acid
and S-adenosylmethionine. Int. J. Dev. Biol. (Suppl 1), 291S ± 292S.
Varela-Moreiras G, Ruiz-Roso B, Varela G (1991): Effects of long-term
administration of acetaminophen on the nutritional utilization of dietary
protein. Ann. Nutr. Metab. 35, 303 ± 308.
Varela-Moreiras G, Ragel C & Ruiz-Roso B (1993): Effects of prolonged
aspirin or acetaminophen administration to rats on liver folate content
and distribution. Relation to DNA methylation and S-adenosylmethio-
nine. Int. J. Vit. Nutr. Res. 63, 41 ± 46.
Varela-Moreiras G, Ragel C & PeÂrez de Miguelsanz J (1995): Choline
de®ciency and methotrexate treatment induces marked but reversible
changes in hepatic folate concentrations, serum homocysteine, and
DNA methylation rates in rats. J. Am. College Nutr. 14, 480 ± 485.
Wegner C & Nau H (1992): Alteration of embryonic folate metabolism by
valproic acid during organogenesis: implications for mechanism of
teratogenesis. Neurology 42(Suppl 5): 17 ± 24.
Welling PG (1996): Effects of food on drug absorption. A. Rev. Nutr. 16,
383 ± 415.