![]() |
|
|
Federal Dairy Research Centre, Department of Physiology and Biochemistry of Nutrition, D-24103 Kiel, Germany and * Research Institute for the Biology of Farm Animals, Division of Nutritional Physiology, D-18059 Rostock, Germany
3To whom correspondence should be addressed.
| ABSTRACT |
|---|
|
|
|---|
KEY WORDS: miniature pigs true protein digestibility D-amino acids lysinoalanine 15N-labeling
| INTRODUCTION |
|---|
|
|
|---|
The existing literature provides conflicting information concerning
whether levels of free as well as peptide- and protein-bound
D-amino acids or cross-linked products such as LAL are
responsible for loss of digestibility. Short D-amino
acidcontaining peptide fragments are less digestible than the
L-enantiomers (Lister et al. 1995
,
Pappenheimer et al. 1997
). Schwass et al. (1983)
observed a delayed absorption of enzymatically prepared
hydrolysates of alkali-treated proteins compared with the
respective nontreated proteins. This group and Bunjapamai et al. (1982)
suggested that this phenomenon is explained mainly by
racemization before hydrolysis because in both studies, the suppressed
formation of LAL and other cross-links did not attenuate loss of
digestibility. However, in a study of similar design, Possompes et al. (1983)
found that inhibition of cross-linking during
alkali-treatment of proteins prevented the loss of digestibility.
We therefore decided to measure in vivo whether and how partial
racemization of amino acids within a protein affects true digestibility
and thus the nutritional value of the proteins, using Göttingen
miniature pigs as a model. The in vivo determination of prececal
digestibility is the method of choice because all other approaches are
fraught with shortcomings. Growth studies might be affected by other
causes of reduced growth in addition to reduced proteolysis. In vitro
studies cannot take into account aspects relevant to absorption and
thus overall digestion, such as the rate of D-amino acid
transport in the mucosa. Results from measurements of in vivo
digestibility over the entire digestive tract may be distorted by
microbial activity in the hind gut or endogenous protein secretion
(De Groot and Slump 1969
, Possompes et al. 1983
).
All of these limitations are avoided in this study. Pigs compare well
with humans with respect to both the anatomy of the digestive tract and
the overall digestive physiology (Erbersdobler 1990
,
Moughan et al. 1994
). Although protein digestibility
through the entire tract is somewhat higher in pigs, there is no
difference in the degree of digestion up to the terminal ileum
(Moughan and Rowan 1989
). With respect to technical
aspects, pigs are particularly suited because they may be fitted with a
fistula in the distal ileum, allowing prececal chyme sampling. Wheat
and milk proteins were chosen as the test proteins because they are of
high nutritional value and are widely used in human nutrition.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Components of the purified experimental diets and providers are listed
in Table 1
. 15N-ammonium sulfate (10 atom%) was obtained
from Medgenix (Ratingen, Germany) and 99%
o-methyl-iso-urea hydrogen sulfate from
Fluka (Deisenhofen, Germany). All other chemicals were of
analytical grade and were obtained from Merck (Darmstadt, Germany).
|
15N-labeled cows milk was produced by continuously
infusing 15N-ammonium sulfate (50 g/d, 10 d) into the
rumen of lactating cows via a permanent cannula. 15N-casein
was isolated by acid precipitation of the defatted milk at 37°C, pH
4.6, acid-washed three times and lyophilized. Purity was determined
by SDS gel electrophoresis. 15N-ß-lactoglobulin was
isolated by microfiltration (Maubois et al. 1987
) in the
State Institute for Dairy Research, Jokionen, Finland;
15N-labeled wheat was provided by Dr. E. Fern,
Nestlé, Vevey, Switzerland.
Aliquots of the labeled proteins were subjected to heat and alkali
treatment. Suspensions (10%) of 15N-labeled casein,
ß-lactoglobulin and wheat in 0.01 mol/L borate buffer were adjusted
to pH 10.511.5 (depending on the protein) with 5 mol/L NaOH, and were
incubated at 65°C for 6 or 24 h. Then, alkali-treated casein
and ß-lactoglobulin were precipitated at their respective isoelectric
points (4.6 and 5.2), washed and lyophilized. The high starch content
of the wheat protein preparation interfered with precipitation.
However, when the incubate was first lyophilized and resuspended in
water (22°C), the wheat protein precipitated readily. Conditions of
treatment are outlined in Table 2
.
|
-NH2-group. Then the material was
acid-precipitated at the isoelectric point (pH 4.6) and washed
three times with water at pH 4.6, according to Schmitz et al. (1991)Animals.
Experimental animals were adult male Göttingen miniature pigs
(n = 9; age 1618 mo; initial body weight 3235
kg; final weight 4555 kg), bred in our animal facility from a strain
supplied by the Institut für Tierzucht und Haustiergenetik,
Universität Göttingen, Germany. Pigs were housed
individually in metabolic cages at 1921°C and 5570% relative
humidity. The pigs were fitted with a T-cannula in the distal
ileum, 10 cm proximal to the ileocecal valve (Schmitz et al. 1991
).
All experimental procedures were approved by the Animal Care and Animal Ethics Committee of the Ministry of Environment of Schleswig-Holstein, Germany, and were carried out according to established guidelines for the care and use of laboratory animals.
Experimental protocol.
Two weeks before the onset of the experiments, the pigs were accustomed
to a purified basal diet [390410 g/d air-dried matter; protein
(casein), 150 g/kg; metabolizable energy, 15.3 MJ/kg dry matter (DM);
Table 1
], given in two equal meals at 0600 and 1600 h, together
with 1 L of water. Food was apportioned according to energy
requirements [0.44 MJ/(kg body weight0.75 · d), i.e.,
body weight at the beginning of the respective protein periods
described below]. Pigs had free access to water.
15N-labeled proteins were added to the basal diet at the
expense of nonlabeled casein, such that the 15N
concentration was identical in all diets and sufficiently high for
measurement (15N-casein, 75 g/kg;
15N-ß-lactoglobulin, 50 g/kg; and 15N-wheat
protein, 40 g/kg DM). Furthermore, 20 g/kg of the undigestible marker
chromic oxide was added, to correct for losses of chyme not recorded by
the T-cannula technique applied (Roos et al. 1994
,
Schmitz et al. 1991
). Overall there were nine test
proteins (Table 2)
, i.e., four casein preparations (native form or
heat- and alkali-treated for 6 or 24 h, and the 24-h treated
and guanidinated casein), three ß-lactoglobulin preparations (native
form or heat- and alkali-treated for 6 or 24 h) and two wheat
protein preparations (native form or heat- and alkali-treated for
24 h), labeled with 15N each. All diets were tested in
all nine miniature pigs.
There were three separate experimental periods for casein, ß-lactoglobulin and wheat protein. Within each protein period, diets with the various modifications of the protein were distributed to the nine pigs as a twofold complete cross classification. Therefore, the minor weight gain of the adult pigs between the separate trials within one period was not taken into account.
After the pigs were deprived of food for 14 h, the test diets were given as single morning meals (204 g), at least 2 wk apart. Ileal chyme was collected postprandially for 33 h. Chyme flow was blocked caudally to the fistula by inflating a balloon catheter (Rösch, Rommerlshausen, Germany). To avoid intestinal atonia, a complete occlusion of the intestine had to be avoided. Because of the ensuing minor loss of chyme to the ileocecal valve, the recovery of chromic oxide was <100%.
Chyme appearing at the fistula was immediately frozen in liquid nitrogen and stored at -20°C until lyophilization. The freeze-dried samples of the first 3 h and the successive 5-h periods were pooled, ground and passed through a sieve (0.5-mm pore size) before chemical analysis.
Analytical methods.
The test proteins and chyme lyophilizates were hydrolyzed for 24 h
in 6 mol/L HCl at 105°C, and amino acids were determined on an amino
acid analyzer (LC 5001, Biotronic, Maintal, Germany). The ratio of
D- to L-amino acids was estimated by chiral
phase capillary gas chromatography (Dani 8521, DANI, Monza, Italy) on
chirasil-L- and D-Val stationary phases
(Chrompack, Frankfurt, Germany), using N(O)-trifluoroacetyl- and
pentafluoropropionyl-amino acid propyl esters (Frank 1990
). The degree of racemization was determined only for some
particularly racemization-prone amino acids, which were present in
the treated protein at a relatively high concentration. Slowly
racemizing amino acids were not determined because it would be
difficult to distinguish between the low racemization due to alkali
treatment and the unavoidable formation of D-amino acids
during preparation for amino acid analysis.
Total nitrogen in the diet and in freeze-dried chyme samples was
determined using the Kjeldahl method. Chromium was determined according
to Stevenson and De Langen (1960)
. Measurement of the
15N-enrichment of the samples was performed on an isotope
ratio mass spectrometer (delta e, Finnigan MAT, Bremen, Germany) as
described previously by Roos et al. (1994)
.
Prececal digestibility of proteins and individual L- and
D-amino acids is calculated as follows:
![]() |
![]() |
where Pd, Pc (µmol), Md and Mc (mg) denote the concentration/g DM of either the parameter of interest (P; 15N for protein digestibility, L- or D-amino acids) or the marker chromic oxide (M) in diet (d) or chyme (c), respectively.
Relatively high amounts of protein-bound D-amino acids, but not of lysinoalanine, were already measured in untreated proteins, mainly as a result of acidic hydrolysis during preparation for amino acid determination. D-Amino acid concentrations in treated proteins were corrected for these basal D-amino acid levels.
LAL was chosen as an indicator of alkali-induced cross-linking
because LAL levels were higher in the alkali-treated proteins than
levels of other cross-links such as lanthionine, ornithinoalanine
and ß-aminoalanine (Friedman 1979
) and because LAL is
particularly relevant because of its nephrotoxicity and its inhibitory
effect on digestion.
Statistics.
If not indicated otherwise, data shown are means ± pooled
SEM, n = 9 pigs. Statistical evaluation
was performed using the Statgraphics statistical package version 6.1,
1993 (Statistical Graphics, Rockwell, MD). The three test proteins were
evaluated separately. Individual treatments of the same protein were
compared by one-way ANOVA, followed by the Scheffé test
(Scheffé 1953
), except for comparison of
D-amino acid digestibility of 6- vs. 24-h treated proteins,
which was done using one-sided, paired t tests.
Differences were considered significant at P < 0.05.
| RESULTS |
|---|
|
|
|---|
Levels of D-amino acids and LAL of all test proteins
increased with both duration of treatment and temperature (Table 3
).
|
Effect of heat and alkali treatment on true prececal protein digestibility.
Heat and alkali treatment decreased prececal digestibility by up to
18% compared with control native proteins. This was true for all
proteins tested and for all degrees of treatment (Table 4
).
|
When the casein preparations containing varying amounts of
protein-associated D-amino acids and LAL were tested,
prececal 15N-recovery and consequently true
prececal digestibility changed in parallel with the increase in
D-amino acid (D-Asx) content, but were
independent of LAL levels (Fig. 1
).
|
Apparent prececal amino acid digestibility was determined for some of
the particularly racemization-prone amino acids present in the
labeled casein and wheat protein. Heat and alkali treatment
significantly (P < 0.05) increased recovery in the
ileal chyme of the L-enantiomers of some of the amino acids
studied (L-Asx, L-serine and
L-Glx), indicating a decrease in apparent digestibility by
up to 17% (equivalent to 25% of the native protein digestibility),
independent of the length of treatment (Table 5
). However, recovery of other amino acids, such as
L-phenylalanine and L-tyrosine, was not
affected.
|
| DISCUSSION |
|---|
|
|
|---|
The question whether either D-amino acids and
D-amino acidcontaining peptides (Bunjapamai et al. 1982
) or cross-links (Possompes et al. 1983
)
alone are responsible for this lower protein digestibility, or whether
their effects are additive (Jenkins et al. 1984
), had
not been answered until now. Both racemization and cross-links were
shown to inhibit proteolysis and decrease protein and peptide
digestibility in in situ experiments using isolated loops of rat
intestine (Lister et al. 1995
, Schwass et al. 1983
).
The results of the experiments with the guanidinated caseins, in which
the loss of digestibility was independent of LAL levels, offer strong
evidence that D-amino acids are mainly responsible for the
impaired digestibility, whereas LAL plays at most a minor role. This
conclusion is in contrast to previously observed inhibitory effects of
LAL on proteolysis (Friedman et al. 1985
, Savoie 1984
). In contrast to earlier work in this field
(Bunjapamai et al. 1982
, Jenkins et al. 1984
, Schwass et al. 1983
), the different LAL
concentrations obtained in this experiment were achieved by treatment
of the protein preparation after racemization, which guaranteed that
proteins were practically identical with respect to amino acid pattern.
The slower absorption of free (Jervis and Smyth 1959
)
and peptide-bound (Lister et al. 1995
,
Pappenheimer et al. 1997
) D- compared with
L-amino acids might contribute to this decrease in protein
digestibility. Clues concerning further mechanisms were provided by the
measurement of the digestibility of individual amino acids.
D-Amino acids and LAL showed the same low prececal
digestibility of < 40%. This value was lower than might have
been expected on the basis of earlier in vitro studies and was largely
independent of the amino acid itself.
The apparent digestibility of the L-amino acids in native
proteins was twice as high as that of the D-enantiomers.
Heat and alkali treatment decreased prececal digestibility of the
L-amino acids, L-serine, L-Asx and
L-Glx, but not of others, including
L-phenylalanine and L-tyrosine (Table 5)
. This
loss of digestibility obviously does not correlate with the degree of
racemization because the effect was already maximal after 6 h,
although racemization increased further with a longer time of treatment
(Table 3)
.
This phenomenon may be explained by the following line of arguments.
During the digestive process, the endopeptidases pepsin and
chymotrypsin attack mainly the peptide chain at bonds involving
phenylalanine or tyrosine. Therefore, proteolysis of dietary proteins
due to pepsin (in the stomach) and chymotrypsin (in the upper section
of the small intestine) creates mainly peptides with the terminal amino
acids phenylalanine or tyrosine. During further degradation of the
peptides by carboxy- and aminopeptidases, these amino acids are the
first to be cleaved and absorbed. Peptidic degradation proceeds up to
the point at which there is a D-amino acid at the end of
the peptide chain, inhibiting further activity of the exopeptidases and
the release of absorbable peptides and free amino acids (Paquet et al. 1985
). This means that digestibility of all amino acids
remaining in these peptides is impaired. Amino acids that are
substrates for exopeptidases would be largely unaffected.
Therefore, although many amino acids show only a low susceptibility toward racemization, their digestibility may nevertheless be markedly impaired. This implies that treatments such as those used here may have nutritional disadvantages. Even if essential amino acids were not racemized to D-enantiomers and would, after breakdown of the protein, be available for absorption, adjacent D-amino acids might nevertheless interfere with their absorption.
These findings are of considerable importance in animal nutrition, in which an impairment of protein digestibility far less than the 18% observed here translates into major economic costs. Feeding roller-dried milk powder to calves may impair their growth. Some NaOH and/or heat-treated feedstuffs may contain considerable amounts of D-amino acids, for example, as a consequence of alkaline detoxification of aflatoxines. Another example are protein concentrates that are left over when alcohol is produced from barley and wheat residues of starch production.
These results also have relevance for human nutrition. They show a diminished nutritive value for racemized, heat- and alkali-treated dietary proteins not only due to a reduction in L-amino acid content but also because of a diminished digestibility.
These findings would also be relevant to scientists who evaluate feed
because the evaluation score of protein quality may become misleading.
The "protein digestibility-corrected amino acid score," favored
by the FAO/WHO (1990)
, calculates protein quality from
the amino acid pattern, corrected for standard true protein
digestibility and/or bioavailability of limited amino acids in rats.
Because proteolysis of undigested dietary protein by the intestinal
microflora is generally not accounted for, the protein digestibility is
overestimated compared with the true prececal digestibility.
Nevertheless, at least in industrial countries, the impaired digestibility observed in this study should not be rated too highly with respect to its relevance for humans. First, more and more mild processing techniques have been used in recent years and second, there is generally a sufficient to superfluous protein supply.
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
2 Supported by Deutsche Forschungsgemeinschaft (Grant HA 456 21). ![]()
4 Abbreviations used: Asx, aspartic acid + asparagine; DM, dry matter; Glx, glutamic acid + glutamine; LAL, lysinoalanine. ![]()
Manuscript received September 28, 1999. Initial review completed October 29, 1999. Revision accepted April 3, 2000.
| REFERENCES |
|---|
|
|
|---|
1. Bunjapamai S., Mahoney R. R., Fagerson I. S. Determination of D-amino acids in some processed foods and effect of racemization on in vitro digestibility of casein. J. Food Sci. 1982;47:1229-1234
2. Chung S.-Y., Swaisgood H. E., Catignani G. L. Effects of alkali treatment and heat treatment in the presence of fructose on digestibility of food proteins as determined by an immobilized digestive enzyme assay (IDEA). J. Agric. Food Chem. 1986;34:579-584
3. De Groot A. P., Slump P. Effects of severe alkali treatment of proteins on amino acid composition and nutritive value. J. Nutr. 1969;98:45-56
4. de Vrese M., Middendorf K., Hagemeister H. Prevention of amino acid racemization during guanidinationa prerequisite for measurement of protein digestibility by homoarginine labeling. Z. Ernährwiss. 1994;33:310-312
5. Drescher K., Hagemeister H., de Vrese M., Roos N., Middendorf K., Rantamäki P. Guanidination (homoarginine-labeling) of proteins does affect pre-caecal 15N-recovery in pigs. Souffrant W.-B. Hagemeister H. eds. VIth International Symposium on Digestive Physiology in Pigs 1994:64-66 EAAP Publication no. 80 Dummerstorf, Germany.
6. Erbersdobler H. F. Tiermodelle im Protein- und Aminosäurenstoffwechsel. Kirchgeßner M. eds. Tiermodelle für die menschliche Ernährungsphysiologie 1990:27-38 Verlag Paul Parey Hamburg, Germany.
7. FAO/WHO Protein quality evaluation. Report of a Joint FAO/WHO Expert Consultation. December 1989, Bethesda, MD 1990 Food and Agriculture Organization of the United Nations Rome.
8. Frank H. Gas chromatography of enantiomers on chiral stationary phase. Holmstedt B. Frank H. Testa B. eds. Chirality and Biological Activity 1990:33-55
9. Friedman M. Alkali-induced lysinoalanine formation in structurally different proteins. Pour-El A. eds. Functionality and Protein Structure 1979:225-235 ACS Symposium Series 92, American Chemical Society Washington, DC.
10. Friedman M. Formation, nutritional value and safety of D-amino acids. Friedman M. eds. Nutritional and Toxicological Consequences of Food Processing 1991:447-481 Plenum Press New York, NY.
11. Friedman M., Grosjean O.-K., Zahnley J. C. Carboxypeptidase inhibition by alkali-treated food proteins. J. Agric. Food Chem. 1985;33:208-213
12. Hagemeister H., Roos N. Comparison of protein digestibility measured by 15N and homoarginine. Eggum B. O. Boisen S. Børsting C. Danfær A. Hvelplund T. eds. Protein Metabolism and Nutrition 1991;Vol. 2. Short Communications:36-38 EAAP-publication no. 59 National Institute of Animal Science, Research Center Foulum.
13. Jenkins W. L., Tovar L. R., Schwass D. E., Liardon R., Carpenter K. J. Nutritional characteristics of alkali-treated zein. J. Agric. Food Chem. 1984;32:1035-1041
14. Jervis E. L., Smyth D. H. The effect of concentrations of amino acids on their rate of absorption from the intestine. J. Physiol. 1959;149:433-441[Medline]
15. Liardon R., Ledermann S. Racemization kinetics of free and protein-bound amino acids under moderate alkaline treatment. J. Agric. Food Chem. 1986;34:557-565
16. Lister N., Sykes A. P., Baily P. D., Boyd C.A.R., Bronk J. R. Dipeptide transport and hydrolysis in isolated loops of rat small intestine: effects of stereospecificity. J. Phys. 1995;484:173-182
17. Man E. H., Bada J. F. Dietary D-amino acids. Annu. Rev. Nutr. 1987;7:209-225[Medline]
18. Maubois L., Pierre A., Fauquant J., Piot M. Industrial fractionation of main whey proteins. Bull. Int. Dairy Fed. 1987;212:154-159
19. Moughan P. J., Cranwell P. D., Derragh A. J., Rowan A. M. The domestic pig as a model animal for studying digestion in humans. Souffrant W.B. Hagemeister H. eds. FBN Schriftenreihe JSSNO 19461981 1994;vol. 4:389-396 Dummerstorf Germany.
20. Moughan P. J., Rowan A. M. The pig as a model animal for human nutrition research. Proc. Nutr. Soc. N.Z. 1989;14:116-123
21.
Pappenheimer J. R., Karnovsky M. L., Maggio J. E. Absorption and excretion of undegradable peptides: role of lipid solubility and net charge. J. Pharmacol. Exp. Ther. 1997;280:292-300
22. Paquet A., Thresher W. C., Swaisgood H. E., Catignani G. L. Synthesis and digestibility determination of some epimeric tripeptides occurring in dietary proteins. Nutr. Res. 1985;5:891-901
23. Possompes B., Cuq J. L., Guenoun D., Besançon P. Use of cysteine as a reducing agent during alkali treatment of proteins. Biochemical and nutritional effects. Food Chem. 1983;11:15-26
24. Roos N., Pfeuffer M., Hagemeister H. Labeling with 15N compared with homoarginine suggests a lower prececal digestibility of casein. J. Nutr. 1994;124:2404-2409
25. Savoie L. Effect of protein treatment on the enzymatic hydrolysis of lysinoalanine and other amino acids. Friedman M. eds. Advances in Experimental Medicine and Biology, Vol. 177: Nutritional and Toxicological Aspects of Food Safety 1984:413-422 Plenum Press New York, NY.
26.
Scheffé H. A method for judging all contrasts in the analysis of variance. Biometrika 1953;40:87-104
27. Schmitz M., Hagemeister H., Erbersdobler H. F. Homoarginine labeling is suitable for determination of protein absorption in miniature pigs. J. Nutr. 1991;121:1575-1580
28. Schwass D. E., Finley J. W. Heat and alkaline damage to proteins: racemization and lysinoalanine formation. J. Agric. Food Chem. 1984;32:1377-1382
29. Schwass D. E., Tovar L. R., Finley J. W. Absorption of altered amino acids from the intestine. Finley J. W. Schwass D. E. eds. Xenobiotics in Foods and Feeds 1983:187-201 ACS Symposium Series 234, American Chemical Society Washington, DC.
30. Stevenson A. E., De Langen H. Measurement of feed intake by grazing cattle and sheep. VII. Modified wet digestion method for determination of chromic oxide in faeces. N.Z. J. Agric. Res. 1960;3:314-319
31. Zagon J., Dehne L.-L., Bögl K.-W. D-Amino acids in organisms and food. Nutr. Res. 1994;14:445-463
This article has been cited by other articles:
![]() |
A. Deglaire, P. J. Moughan, C. Bos, K. Petzke, S. M. Rutherfurd, and D. Tome A Casein Hydrolysate Does Not Enhance Gut Endogenous Protein Flows Compared with Intact Casein When Fed to Growing Rats J. Nutr., March 1, 2008; 138(3): 556 - 561. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Bos, B. Juillet, H. Fouillet, L. Turlan, S. Dare, C. Luengo, R. N'tounda, R. Benamouzig, N. Gausseres, D. Tome, et al. Postprandial metabolic utilization of wheat protein in humans Am. J. Clinical Nutrition, January 1, 2005; 81(1): 87 - 94. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||