Content uploaded by Dietmar Fuchs
Author content
All content in this area was uploaded by Dietmar Fuchs on Jun 23, 2014
Content may be subject to copyright.
Fructose Malabsorption is Associated with Decreased Plasma Tryptophan
M. Led ochowski, B . Widner, C. Murr, B. Sperner-Unterweger & D. Fuchs
Dept. of Clinical Nutrition, Institute for Medica l Chemistry and Biochemistry, and Dept. of
Psychiatry, University of Innsbruck , Innsbruck, Austria
Ledochowski M, Widner B, Murr C, Sperner-Unterwege r B, Fuchs D. Fructose malabsorptio n is
associated with decreased plasma tryptophan. Scand J Gastroenterol 2001;36:367 –371.
Background:
Fructose malabsorption is characterize d by the inability to absorb fructose ef ciently. As a
consequenc e fructose reaches the colon where it is broken down by bacteria to short fatty acids, CO
2
, H
2
,
CH
4
and lactic acid. Bloating, cramps, osmotic diarrhea and other symptoms of irritable bowel syndrome
are the consequence and can be seen in about 50% of fructose malabsorbers. Recently it was found that
fructose malabsorptio n was associated with early signs of depressive disorders. Therefore, it was
investigated whether fructose malabsorption is associated with abnormal tryptophan metabolism.
Method s:
Fifty adults (16 men, 34 women) with gastrointestina l discomfort were analyzed by measuring
breath hydrogen concentration s after an oral dose of 50 g fructose after an overnight fast. T hey were
classi ed as normals or fructose malabsorbers according to their breat h H
2
concentrations . All patients
lled out a Beck depression inventory questionnaire . Blood samples were taken for plasma tryptophan
and kynurenine measurements.
R esults:
Fructose malabsorption (breath
D
H
2
production
>
20 ppm) was
detected in 35 of 50 individuals (70%). Subjects with fructose malabsorption showed signi cantly
lower plasma tryptophan concentration s and signi cantly higher scores in the Beck depression inventory
compared to those with normal fructose absorption .
Conclusions:
Fructose malabsorption is associated
with lower tryptophan levels that may play a role in the development of depressive disorders.
High intestinal fructose concentratio n seems to interfere with
L
-tryptophan metabolism, and it may
reduce availability of tryptophan for the biosynthesi s of serotonin (5-hydroxytryptamine) . Fructose
malabsorption should be considered in patients with symptom s of depression and disturbance s of
tryptophan metabolism.
K ey words:
Fructose loading test; fructose malabsorption ; H
2
exhalation; tryptophan
Maximilian Ledochowski, Universita
¨
tsklinik Innsbruck, Abteilung fu
¨
r E rna
¨
hrungsmedizin , Anichstraße
35, A-6020 Innsbruck, Austria (fax.
‡
43 512 504 2017, e-mail. maximilian.ledochowsk i
@uibk.ac.at)
F
ructose malabsorption syndrome was rst described
some years ag o (1–3). Patients with fructose malab-
sorption are unable to absorb the ingested mono-
saccharide in a suf cient way so that large quantities of
fructose reach the colon, where it is broken down by colon
bacteria into short fatty ac ids, CO
2
, CH
4
, lactic acid and H
2
which can be measured in the expired air. Bloating,
abdominal discomfort and sometimes osm otic diarrhea are
the consequences induced by the degradation products built
by the colonic bacteria. It is believed that up to 36% of the
Europea n population have fructose malabsorption in a more
or less severe form, and about half of them are symptomatic
(4). Recently we have observed that fructose malabsorption
was associated with early signs of depressive disorders (5),
and that these signs were ameliorated upon a fructose-and
sorbitol-reduced diet (6). The data available so far suggest
that abnormalities of tryptophan availability could be
involved in the development of fructose malabsorption
associated depression. It was therefore of interest to investi-
gate tryptophan concentrations in relation to fructose malab-
sorption.
Mate rial and Methods
Patients
All adult outpatients who visited the physician’s of ce
between November 1997 and March 1998 for a medical
health check and admitte d gastrointestinal discomfort were
studied. Subjects were included in the study if any of the
following symptoms were present: stool irregularities, bloat-
ing, abdominal cramps, diarrhea, constipation or na usea. The
50 patients (16 men, 34 women) were aged from 16 to 72
years (mean, 43.3 years) and otherwise he althy. Physical
examination and routine laboratory assessment did not reveal
abnormalities. None of the patients showed signs of in am-
matory bowel disease, any other chronic disease or infectious
diseases and were—except for oral contraceptives in some
cases—under no medication. Body weight and height were
ORIGINAL ARTICLE
Ó 2001
Taylor & Francis
measured, and blood samples for plas ma tryptophan and
kynurenine measurements were taken after an overnight fast
before breath hydrogen testing was performed.
Hydrogen (H
2
) breath tests
Breath (H
2
) was measured using a Bedfont gastrolizer
(Bedfont Ltd, Kent, U K) which has been validated by several
authors (7–9). After a 12 h overnight fast a baseline H
2
breath
test was performed. An oral dose of 50 g fructose was given in
250 ml of tap water and H
2
exhalation was monitored in
30 min intervals for at least 2 h. A l l tests were started between
0800 h and 0830 h. Maximum H
2
exhalation (H
2
-max) after
fructose load was monitored and the differences to baseline
levels (
D
H
2
) were calculated. Patients with fasting breath
hydrogen concentrations of
>
10 ppm were excluded from the
study. Between the monitoring of breath H
2
all patients lled
out a Beck depression inventory questionnaire (10).
Tryptophan measurement
Tryptophan was measured simultaneously with kynure-
nine, a metabolite of the tryptophan catabolism, by high
performance liquid chromatography (HPLC) according to a
recently established method (11). In brief, 100
m
l serum was
diluted with 100
m
l buffer solution (pH = 6.4) containing
10
m
M 3-nitro-
L
-tyrosine as internal standard. Protein was
then precipita ted with 25
m
l trichloroacetic acid (2 M). The
specimens were centrifuged, and 100
m
l of the supernatant
was injected in the HPLC column. A Lichrochart RP-18
reversed phase column (grain size, 5
m
m; Merck, Darmstadt,
Germany) was applied. The elution buffer was a 15 mM
phosphate buffer (pH = 6.4). The pump and the data system
were from Varian (Palo Alto, CA, USA). Tryptophan was
detected by it s natural uorescence (excitation, 285 nm;
emission, 365 nm) with a HP 1046A uorescence detector
(Hewlett Packard, Vienna, Austria). Kynurenine and nitrotyr-
osine were detected by UV absorption at 360 nm with a UV
detector (UV 975, Jasco, Tokyo, Japan). External calibration
was done by an albumin-based standard, containing 10
m
M
kynurenine, 50
m
M tryptophan and 10
m
M nitrotyrosine. All
chemicals used (Merck) were of high analytical grade.
Data analysis
The cut-off points for the diagnosis of fructose malabsorp-
tion were breath H
2
concentrations greater than 20 ppm over
baseline (12). Subjects with an increase of breath H
2
concentration equal or less than 20 ppm over baseline w ere
considered to be normal fr uctose absorbers. For comparison
of groups the Mann–Whitney U test was employed using a
standard PC statistical program (STATISTICA for Windows)
(13), for correlation analyses Spearman rank correlation
coef cients were calculated. Frequencies were compared by
the Fisher exact test.
R esults
The main results are summarized in Table I. In 35 patients
breath H
2
concentrations increased more than 20 ppm over
basal fasting values. They we re classi ed as fructose
malabsorbers. The remaining 15 subjects with lower H
2
exhalation were classi ed as normal fructose absorbers. The
two groups of individuals showe d no difference in age. There
was only a trend to higher Beck inventory depression scores
in fructose malabsorbers (9.47
§
7.35) than in normal
fructose a bsorbers (7.07
§
4.62; see Table I) but no s ig-
ni cant differenc e was observed between the two groups of
individuals. When subjects were split into two groups by sex,
the Beck inventory depression scores we re higher in female
fructose malabsorbers (12.30
§
7.16) than in females with
normal fructose absorption (6.66
§
5.50; P = 0.02). No such
difference was observed in males.
Mean plasma tryptophan conc entrations were signi cantly
lower in fructose malabsorbers than in normal fruc tose
absorbers (P = 0.02 ; Table I). Plasma kynurenine concentra-
tions and tryptophan per kynurenine quotients w ere within the
normal range of healthy controls in most individuals (4/50
had kynurenine concentrations
>
3
m
M, 4/50 individuals
presente d with a kynurenine per tryptophan quotient
>
40)
and they did not signi cantly differ between the two groups.
When patients were divided into two groups by sex, serum
tryptophan concentrations were lower in individuals with
fructose malabsorption compared to normals only in females
(fructose malabsorbers: 61.3
§
14.0
m
M, normals: 74.7
§
16.5
m
M, P = 0.03; Fig. 1) but not in males (fructose
malabsorbers: 70.3
§
10.4
m
M, normals: 76.4
§
12.5
m
M, P =
not signi cant). Kynurenine concentrations (females, fructose
malabsorbers: 1.81
§
0.58
m
M, normals: 1.98
§
0.64
m
M;
males, fructose malabsorbers: 2.12
§
0.46
m
M, normals:
1.93
§
0.30
m
M) and kynurenine per tryptophan ratios
(females, fructose malabsorbers: 30
§
9 mM/M, normals:
Table I. Comparison of fructose malabsorbers and normals
Fructose
malabsorbers Normals
n = 35 n = 15
Age (years) 45.4 (24–72 ) 38.4 (16–57 )
D
H
2
concentration
(ppm)
43.8 (21–111) 0.95 (
¡
12 to
¡
6)
Maximum H
2
concentratio n (ppm)
46.8 (22–112) 7.79 (0–51)
Beck depressio n score 9.47 (0–38) 7.01 (0–17)
Tryptophan (
m
M) 64.1 (39.0–97.2) 75.2 (53.9–110)*
Kynurenine (
m
M) 1.91 (0.95–3.25) 1.97 (0.97–3.39)
Kyn/Try (mM/M) 30 (18–49) 27 (18–50)
Characteristic s (mean and range in parentheses ) of individuals, H
2
concentration s (after fructose load, see Materials and Methods) and
depression score, serum tryptophan and kynurenine concentration s
and kynurenine per tryptophan ratios (Kyn/Try) split int o two groups
according to maximum H
2
concentration s after fructose load
µ
20
ppm (normals) and
>
20 ppm (fructose malabsorbers) ; * P = 0.02.
Scand J Gastroenterol 2001 (4)
368 M. Ledochowski et al.
27
§
10 mM /M; males, fructose malabsorbers: 31
§
8
mM/M, normals: 26
§
3 mM/M) did not differ between the
groups.
When com paring tryptophan concentrations with the Beck
depression inventory scores, there was no s igni cant re lation-
ship in the whole group of individuals (n = 50; r
s
=
¡
0.182,
not signi cant). There was also no signi cant relationship
when groups were separated by sex. However, individu-
als with tryptophan concentrations lower than the median
(= 67.0
m
M) more often presented with a Beck depression
inventory score above the median value of six (P = 0.036;
Fisher exact test). When analyses were restricted to fructose
malabsorbers, a s igni cant inverse relationship between
tryptophan concentration and Beck score were foun d for the
whole group of individuals (n = 35; r
s
=
¡
0.348, P = 0.043)
and for females (n = 24; r
s
=
¡
0.503, P = 0.014; Fig. 2). There
was no such association between the Beck score and serum
tryptophan levels in ma le fructose malabsorbers (n = 11;
r
s
= 0.205, P = not signi cant).
Dis cussion
A pathogenic link between fructose malabsorption a nd func-
tional bowel disease—a typical psychosomatic disorder—was
discussed by se veral authors (14–17). However, this associa-
tion could no t be found by other investigators (18, 19). In a
previous study (5) we described an association between
fructose malabsorption and early signs of depressive disorders
as re ected by the Beck depression inventory score especially
in females. The data in the present study con rm and extend
this observation. Moreover, the association between fructose
malabsorption and decreased serum tryptophan concentra-
tions found in our study supports the view that abnormal
tryptophan availability could be involved in the higher risk for
developing signs of mental depression in female patient s with
fructose malabsorption. Earlier studies imply that distur-
bances of
L
-tryptophan metabolism are involved in inducing
depression (20–22) and pre- menstrual syndrome (23), since
low tryptophan levels may limit the biosynthesis of serotonin
(5-hydroxytryptamine).
The nding of decreased serum tryptophan concentrations
in patients with fructose malabsorption supports the view that
fructose malabsorption interferes with tryptophan metabo-
lism. On the one hand, fructose malabsorption may reduce
transit time in the gut and thus reduce the absorption of the
essential amino acid similar as with foli c acid (24). However,
reduce d transit time obviously does not contribute largely to
decreased serum tryptophan concentrations as we could not
Fig. 1. Serum tryptopha n concentration s in females and males with fructose m alabsorption and healthy controls with normal fructose
absorption (plots show medians = horizontal lines, 25th–75th percentiles = dotted box, and 5th–9th percentile s = bars); the differenc e is
signi cant for females only (P = 0.03).
Fig. 2. Association between serum tryptophan and the Beck
depression scores in female patients with fructose malabsorption
(r
s
=
¡
0.503, P = 0.014).
Scand J Gastroenterol 2001 (4)
Tryptophan and Fructose Malabsorption 369
nd such lower serum tryptophan c oncentrations in subjects
with lactose maldigestion (data not included).
On the other hand, fructose, as other saccharides, reacts
with proteins and amino acids such as
L
-tryptophan (25),
thereby a fructose–
L
-tryptophan comple x can be formed
which results in a decrease in protein quality due to the loss of
amino acid residues and decreased protein digestibility. This
chemical interaction according to the so-called Maillard
reaction (26) could provide an explanation for the possible
association between fructose malabsorption and disturbed
tryptophan m etabolism. Maillard products can also inhibit the
uptake and metabolism of other free amino acids such as
L
-
tryptophan and of other nutrients such as zinc (26).
In our study, two-thirds of subjects were classi ed as
fructose malabsorbers, and there were no non-H
2
-producers.
We are aware that a large dose of 50 g of fructose may be
insuf ciently absorbed by many individuals and increase the
percentage of fructose malabsorption in the study population.
In this study, usin g 50 g of fructose th e highest
D
H
2
value in
normal fructose absorbers was 8 ppm and the lowest
D
H
2
value of fructose malabsorbers w as 20 ppm, so that a cut-off
value of 10 does not change the overall message. However, in
a few patients 50 and 25 g were administered on two distinct
days. As expected the overall H
2
exhalation was lower when
subjects received 25 g of fructose and some of these subjects
had
D
H
2
values between 10 and 20 ppm. However, also when
we strati ed the data of H
2
breath tests with administration of
25 g fructose there was practically no difference in the
relationship to tryptophan and Beck depression scores.
Interestingly the association of fructose malabsorption with
signs of mental depression is more expressed in females than
in males. However, the number of male patients is still small
in this study for a nal judgment. Since blood concentrations
of
L
-tryptophan are already signi cantly lower in healthy
females than in males (11, 27), disturbed absorption of
L
-
tryptophan may have more impact to precipitate clinically
relevant disturbances of serotonin metabolism in females than
in males. In 7/24 female patients tryptophan concentrations
below 50
m
M were observed, this is almost the same range
seen in patients with, e.g . endogenous depression (27). This
goes along with ndings of sex differences in mood responses
to acute tryptophan depletion by several authors (21, 23, 28).
The impact of fructose malabsorption may be more pro-
nounced in females with symptoms of ges tagen de ciency as
it is the case in women wit h pre-menstrual syndrome and in
perimenopausal women.
Reduced tryptophan concentrations in general could be due
to enhance d degradation of tryptopha n. Immune activation in
patients, e.g. due to clinically inapparent infections or
autoimmune disorders, could activate indoleamine (2, 3)-
dioxygenase which degrades tryptophan to form kynurenine
metabolites (11, 28). Of our patients with fructose malabsorp-
tion only 4/50 had an elevated tryptophan per kynurenine
quotient. When these subjects wer e excluded from the
statistical evaluation, the association between fructose mal-
absorption and lower tryptophan concentrations remained
signi cant (n = 46: P = 0.035 versus n = 50: P = 0.022). Thus,
enhanced degradation of tryptophan is unlikely the reason of
decreased tryptophan in patients with fructose malabsorption.
We conclude that fructose malabsorption is associated with
lower tryptophan levels. High intestinal fructose concentra-
tions seem to interfere with
L
-tryptophan metabolis m and
hence reduce the availability of serotonin (5-hydroxytrypta -
mine). Low tryptophan concentrations may play a role in the
development of symptoms of mental depression. Although
the corre lations found do not necessarily con rm a causal
relationship, this observation suggests that fructose malab-
sorption may be implicated in the pathogenesis of mood
disturbances and depressive disorders.
Acknowledgements
Presented in part at the 9th International Meeting of the
International Study Group for Tryptophan Research, 10–14
October 1998, Hamburg, Germany.
R eferences
1. Rumessen JJ , Gudmand-Hoye r E. Malabsorptio n of fructose–
sorbitol mixtures. Interactions causing abdominal distress.
Scand J Gastroentero l 1987;22:431 –6.
2. Gotza H, Mahdi A. Fructose malabsorption and dysfunctiona l
gastrointestina l manifestation s [see comments]. Monatssch r
Kinderheilkd 1992;140:814 –7.
3. Rumessen JJ. Fructose and related food carbohydrates . Sources,
intake, absorption , and clinical implications. Scand J Gastro-
enterol 1992;27:819 –28.
4. Born P, Zech J, Lehn H, Classen M, Lorenz R. Colonic bacterial
activity determines the symptoms in people with fructose-
malabsorption . H epatogastroenterolog y 1995;42:778 –85.
5. Ledochowski M, Sperner-Unterwege r B, Widner B, Fuchs D.
Fructose malabsorption is associated with early signs of mental
depression. Eur J Med Res 1998;3:295–8.
6. Ledochowski M, Widner B, Bair H, Propst T, Fuchs D.
Fructose-and sorbitol-reduce d diet improves mood and gastro-
intestinal disturbance s in fructose malabsorbers. Scand J Gas-
troenterol 2000. In press.
7. Braden B, Braden CP, Klutz M, Lembcke B. [Analysis of breath
hydrogen (H
2
) in diagnosis of gastrointestina l function: valida-
tion of a pocket breath H
2
test analyzer]. Z Gastroenterol 1993;
31:242–5.
8. Fleming SC. Evaluation of a hand-held hydrogen monitor in the
diagnosis of intestinal lactase de ciency. Ann Clin Biochem
1990;27:499–500.
9. Duan LP, Braden B, Clement T, Caspary W F, Lembcke B.
Clinical evaluation of a miniaturized desktop breath hydrogen
analyzer. Z Gastroentero l 1994;32:575 –8.
10. Hautzinger M, Bailer M, Worall H, Keller F. Beck Depression
Inventory—German version/author. 1994, Beck Depressions
Inventar (BDI) 1994; Huber.
11. Widner B, Werner ER, Schennach H, Wachte r H, Fuchs D.
Simultaneous measurement of serum tryptopha n and kynurenine
by HPLC. Clin Chem 1997;43:2424 –6.
12. Veligati LN, Treem WR, Sullivan B, Burke G, Hyams JS. Delta
10 ppm versus delta 20 ppm: a reappraisal of diagnostic criteria
for breath hydrogen testing in children. Am J Gastroenterol
1994;89:758–61.
13. STATISTICA for Windows [computer program manual]. Tulsa,
OK: StatSoft, Inc., 2325 East 13th Street. StatSoft I; 1995. (5.0).
14. Ferna´ndez-Banares F, Esteve-Pardo M, de Leon R, Humbert P,
Scand J Gastroenterol 2001 (4)
370 M. Ledochowski et al.
Cabre E, Llovet JM, et al. Sugar malabsorptio n in functional
bowel disease: clinical implications. Am J Gastroentero l 1993;
88:2044–50.
15. Born P, Vierling T, Barina W. Fructose malabsorptio n and the
irritable bowel syndrom e [letter; comment]. Gastroenterolog y
1991;101:1454.
16. Ferna´ndez-Banares F, Esteve-Pardo M, Humbert P, de Leon R,
Llovet JM, Gassull MA. Role of fructose–sorbitol malabsorpti o n
in the irritable bowel syndrom e [letter; comment]. Gastroenter -
ology 1991;101:1453 –4.
17. Rumessen JJ, Gudmand-Hoye r E. Functional bowel disease: the
role of fructose and sorbitol [letter; comment]. Gastroenterolog y
1991;101:1452 –3.
18. Nelis GF, Vermeeren MA, Jansen W. Role of fructose–sorbitol
malabsorption in the irritable bowel syndrome [see comments].
Gastr6oenterolog y 1990;99:1016 –20.
19. Evans PR, Piesse C, Bak YT, Kellow JE. Fructose–sorbitol
malabsorption and symptom provocation in irritable bowel
syndrome: relationshi p to enteric hypersensitivity and dysmo-
tility. Scand J Gastroenterol 1998;33:1158 –63.
20. Benkelfat C, Ellenbogen MA, Dean P, Palmour RM, Young SN.
Mood-lowering effect of tryptophan depletion. Enhanced
susceptibilit y in young men at genetic risk for major affective
disorders. Arch Gen Psychiatry 1994;51:687 –97.
21. Ellenbogen MA, Young SN, Dean P, Palmou r RM, Benkelfat C.
Mood response to acute tryptopha n depletion in healthy
volunteers: sex difference s and temporal stability . Neuropsy-
chopharmacolog y 1996;15:465 –74 .
22. Neumeister A, Praschak Rieder N, Besselmann B, Rao ML,
Gluck J, Kasper S. Effects of tryptophan depletion on drug-free
patients with seasonal affective disorder during a stable response
to bright light therapy. Arch Gen Psychiatry 1997;54:133 –8.
23. Menkes DB, Coates DC, Fawcett JP . Acute tryptophan depletion
aggravates premenstrual syndrome. J Affect Disord 1994;32:37–
44.
24. Ledochowsk i M, U
¨
berall F, Props t T, Fuchs D. Fructose mal-
absorption is associated with lower plasma folic acid concentra-
tion in middle aged subjects. Clin Chem 1999;45:2013 –4.
25. Davis EA. Functionality of sugars: physicochemica l interactions
in foods. Am J Clin Nutr 1995;62:170S –7S.
26. Dills WL. Protein fructosylation: fructose and the Maillard
reaction. Am J Clin Nut r 1993;58:779S –87S.
27. Gasse T, Widner B, Baier-Bittelich G, Sperner-Unterwege r B,
Leblhuber F, Wachter H, et al. Abnormal tryptophan metabo-
lism, neurologic/psychiatric disturbance s and its relationship to
immune activation . In: Quereshi GA, editor. Neurochemical
markers of degenerative nervous diseases & drug addiction.
Progres s in HPLC-HPCE. 7th ed. The N etherlands : VSP Press,
Zeist; 1997 .
28. Salomon RM, Delgado PL, Licinio J, Krystal JH, Heninger GR,
Charney DS. Effects of sleep deprivation on serotoni n function
in depression. Biol Psychiatry 1994;36:840 –6.
Received 8 June 2000
Accepted 15 September 2000
Scand J Gastroenterol 2001 (4)
Tryptophan and Fructose Malabsorption 371
