Download Intestinal Absorption of Dipeptides Containing

Clinical Science and Molecular Medicine (1973)45,849-858.
INTESTINAL ABSORPTION OF DIPEPTIDES
CONTAINING GLYCINE, PHENYLALANINE, PROLINE,
B-ALANINE OR HISTIDINE I N THE RAT
D. J. BOULLIN, R . F. CRAMPTON, C H R I S T I N E E. H E A D I N G
A N D D. P E L L I N G
The British Industrial Biological Research Association,
Carshalton, Surrey
(Received 21 June 1973)
SUMMARY
1 . The intestinal absorption of carnosine, glycylglycine, glycyl-D-phenylalanine,
glycyl-L-phenylalanine, glycyl-L-proline and L-prolylglycine were investigated after
intraluminal injection of dipeptide into anaesthetized rats.
2. With all six dipeptides, the intact substance was detected by ion-exchange
chromatography in blood samples taken from the superior mesenteric vein.
3. The rate of hydrolysis of the dipeptides in tissue homogenates was measured in
vitro.
4. The relative rates of hydrolysis varied by a factor of 300; there was an apparent
inverse relationship between rate of hydrolysis and detection of intact peptide.
5. Peptide absorption was accompanied by increases in venous concentrations of
the component amino acids, which appeared in proportions appropriate to the view
that peptide absorption preceded hydrolysis.
6. It is suggested that slowly hydrolysed dipeptides may pass intact through the
intestine wall under physiological conditions.
Key words : dipeptides, intestinal absorption.
It is now clearly established that certain peptides can be absorbed from the gastrointestinal
tract without prior hydrolysis to their component amino acids. Most of this work has been
carried out in vitro using rings or sacs of everted intestine (e.g. Addison, Burston & Matthews,
1972; Burston, Addison & Matthews, 1972; Cheeseman 8z Smyth, 1973), procedures which
are useful for examining the absorption and transport of peptides into and across intestinal
cells. The conditions in vivo are more complicated since absorption into the portal blood
involves passage through an additional tissue layer, the walls of the blood vessels. Also,
peptides which are incompletely hydrolysed in the intestine may undergo hydrolysis in the
Correspondence: Dr R. F. Crarnpton, The British Industrial Biological Research Association, Woodmansterne Road, Carshalton, Surrey.
I
849
850
D. J. Boullin et al.
liver and elsewhere so that changes in peptide and amino acid concentrations in peripheral
blood do not accurately reflect their absorption. Despite these complications, the intact
absorption of peptides into the circulation has been reported for glycylglycine (Newey &
Smyth, 1959; Peters & MacMahon, 1970; Adibi, 1971), carnosine (Perry, Hansen, Tischler,
Bunting & Berry, 1967) and prolylhydroxyproline (Hueckel & Rogers, 1972).
An alternative to the sampling of peripheral blood is the sampling of mesenteric blood,
but even this technique can be unsatisfactory. First, hydrolysis of peptides proceeds rapidly,
and secondly, it is difficult to sample mesenteric or portal blood. In the work described here,
we have surmounted the first of these problems by injection of high concentrations of dipeptide into the lumen of the small intestine. The second was overcome by collecting samples
of blood from the superior mesenteric vein of rats, before the peptides had reached the systemic
circulation and had been subjected to enzymic hydrolysis by the liver and other tissues. This
technique, devised by one of us (D.P.) has been described in detail elsewhere (Butterworth &
Pelling, 1973). Using this method we have been able to detect intact dipeptide in the mesenteric
venous blood in the case of every compound examined.
METHODS
Peptide hydrolase activity
The ability of mucosa from the rat small intestine to hydrolyse a series of dipeptides was
assessed by a modification of the qualitative screening test described by Heizer & Laster
(1969).
Male Charles River rats (450-600 g) were starved overnight and killed by a blow on the
head. The entire small intestine was removed, flushed with sucrose solution (250 mmol/l)
at 4°C and the mucosa then scraped off with a scalpel blade. The mucosa was weighed, suspended in sucrose solution (250 mmol/l; 5 ml/g wet wt.) and disrupted by ultrasound using
an MSE 150 W ultrasonic disintegrator (Measuring and Scientific Equipment Ltd, Crawley,
Sussex), fitted with a titanium exponential microprobe tip. Ultrasonic impulses were applied
to the tissue, kept at 4“C, at the rate of 20 MHz in intermittent bursts for a total time of
60 s. The disintegrator settings used were ‘low’ and ‘I’ and under these conditions the current
strength did not exceed 8 pA.
This suspension was used for assay except when pilot experiments had indicated the hydrolase activity was greater than 20 pmol g-’ 10 min-’ (weight as wet tissue). In such cases, the
tissue solution was diluted 1/10 with sucrose (250 mmol/l).
Peptide solutions (20 mmol/l) were prepared in Tris-maleate buffer (200 mmol/l; pH 7.5).
After being warmed, 100 pl of this buffer-substrate solution was mixed with 100 pl of tissue
solution and incubated at 37°C for 10 min. A portion (50 p1) was then removed from the incubation mixture and added to sulphosalicylic acid solution (200 g/l). Distilled water (2 ml)
was added to the sample before assay by ion-exchange chromatography. Pure peptide solutions
and sonified tissue extract without added peptide were used as controls and blanks respectively.
Boiled tissue and tissue precipitated with equal volumes of sulphosalicylic acid solution
(200 g/l) were also examined for peptide hydrolase activity. Hydrolase activity was expressed
as mol of peptide hydrolysed/g wet wt. of mucosa during the 10 min incubation (pmol g-’
10 min-’) and an arbitrary value relative to the highest activity (100) assigned to each peptide.
Intestinal absorption of dipeptides
851
Absorption in vivo
Male Charles River rats of 350-600 g body weight were starved overnight and anaesthetized
with ether followed by pentobarbitone sodium (40 mg/kg intravenously). The right external
jugular vein was cannulated with polyethylene tubing (PP 100, Portland Plastics Ltd). The
abdomen was opened by mid-line incision and the superior mesenteric vein exposed by carefully parting the fatty tissue overlying it. The drawn and bevelled tip of a length of PP 100
tubing was inserted into a hole cut in the vein. The blood flow was interrupted for 1-2 min
during insertion of this cannula which conveyed the portal blood to the jugular vein. The
cannula was inserted into the superior mesenteric vein below the splenic and pyloric tributaries.
Thus blood was collected from a loop 40-60 cm in length composed of jejunum and ileum,
and extending to within 5 cm of the ileo-caecal junction. Branches draining the gut distal
to the ileo-caecal junction were tied off.
The blood pressure and flow in the mesenteric-central venous shunt were recorded continuously by means of a pressure transducer, drop counter and recorder (Devices Ltd). The shunt
was primed with heparinized sodium chloride solution (150 mmol/l) and had a total volume
of 0-5 ml. Short plugged lengths of PP 100 tubing were tied into each end of the ileal loop
drained by the shunt. The ileum remained in situ throughout the procedure; manipulation
of the intestine was necessary only at the two ends of the loop to insert the luminal cannulae
and to inject the test substance. After establishing the shunt the abdomen was closed with
Michel clips, leaving space for exit of the cannulae and for a thermometer placed in the peritoneal cavity. At least 15 min were allowed for restoration of normal temperature before
taking control blood samples from the shunt. Further details are given in the original description of this procedure (Butterworth & Pelling, 1973).
Samples (4O-400 pl) of mesenteric and jugular venous blood were taken from respective
ends of the shunt, before and at intervals after introducing 1.5 ml of peptide solution into the
lumen. The peptides (100 mmol/l, unless otherwise stated) were dissolved in sodium chloride
solution (150 mmol/l). After injection, the mesenteric blood was sampled for 30 min in at
least two experiments and for 60 min in at least one experimentwith each peptide. The samples
were kept in an ice bath for not more than 20 min and then centrifuged at 3000 g. The plasma
was removed and mixed with 3 volume of sulphosalicylic acid solution (200 g/l) and diluted
to 2-3 ml for assay.
Materials and analytical techniques
The peptides, B-alanyl-L-histidine (carnosine), glycylglycine (Gly-Gly), glycyl-D-phenylalanine (Gly-D-Phe), glycyl-L-phenylalanine (Gly-L-Phe), glycyl-L-proline (Gly-L-Pro) and
L-prolylglycine (L-Pro-Gly) were obtained from Calbiochem Ltd and the Sigma Chemical
Co. and were found to be chromatographically pure. Amino acid analysis was performed
using a JLC-5AH automatic analyser [Japan Electron Optics (U.K.) Ltd], solutions of relevant
amino acids and peptides being assayed as required as external standards. For the assay of
peptide hydrolase activity, analytical programmes were designed to measure one of the amino
acid products of the peptide hydrolysis. For the analysis of peptides and amino acids in plasma
collected during the experiments in vivo, programmes were designed to measure plasma
concentration of the dipeptide in question, its constituent amino acids and L-alanine. In
three pilot experiments, peptides were added to plasma before precipitation of protein and
subsequent analysis. Assay values of not less than 97% of the theoretical value were obtained.
852
D. J. Boullin et al.
RESULTS
Peptide hydrolase activity
As we did not investigate the kinetics of dipeptide hydrolysis in detail, the quantitative
aspects of our results must be interpreted with caution. Nevertheless, under the experimental
conditions used, it is possible to obtain a qualitative picture of the range of enzyme activity
of the mucosa towards the peptides studied. The range of activity was over 300-fold; Gly-L-Phe
being hydrolysed the most, and carnosine the least. The maximum variation between rats for
any one peptide was fourfold (Gly-L-Phe) but the mean variation was considerably less as
indicated by the SD values shown in Table 1 .
No hydrolytic activity was seen in any of the control procedures described in the Methods
section.
Absorption of dipeptides and amino acids in vivo
Time-course of absorption. After intraluminal injection, the dipeptide or its constituent
amino acids appeared in the superior mesenteric circulation. Intact dipeptides were detected
in most experiments and amino acid concentrations were invariably increased. In all except
one of the fifteen time-course experiments, maximum peptide concentration occurred within
30 min. Peak concentrations of amino acids were noted between 5 min and 60 min, which
was the longest time investigated. There was no obvious relationship between peptide structure
and the time of appearance of peak concentrations.
Figs. 1 and 2 present results from individual experiments showing changes in plasma
peptide concentrations with time. Each graph on Fig. 1 and Fig. 2 represents data from a
single animal. The cumulated data are presented in Table 1, which shows that there were
wide individual variations in plasma concentrations.
Figs. l(a) and l(b) show examples of changes in plasma peptide concentrations with time
for five of the six compounds investigated. Fig. 2 shows examples of changes in plasma peptide
and amino acid concentration with time, using one experiment with Gly-L-Phe as an example.
Generally the increase in plasma concentration was substantially greater for amino acids
than for peptides.
Relationship between peptide hydrolysis and detection of intact peptide. Table 1 shows the
results of all experiments in which mesenteric plasma was sampled 15 min after peptide injection. These data are compared with information on the rate of hydrolysis of dipeptides
under standard conditions, the peptides being arranged in decreasing order of hydrolysis.
We found that the two peptides which were most rapidly hydrolysed, Gly-L-Phe and GlyL-Pro (Table l), were detected in two out of four and three out of four experiments; all others,
whose hydrolysis rate was at least an order of magnitude less, were detected in every instance
(four out of four or five out of five experiments). Further experiments would be required to
establish the precise relationship between incidence of detection and hydrolysis rate.
Peptide and amino acid concentrations in mesenteric venous plasma
Quantitative comparisons of the relationship between peptide and amino acid concentrations
after injection of different peptides were possible. However, due to experimental variations in
time of appearance of peak concentrations in the plasma, it was necessary to treat the data
with caution. The qualitative picture that emerged showed that increases in plasma peptide
Intestinal absorption of dipeptides
0
5
Time (mid
853
I
I
I
I
I
10
15
20
25
30
Time (mid
FIG.1. Examples of changes in peptide concentration in mesenteric plasma with time after intraluminal injection; each curve represents an individual experiment. Peptide (150 pmol) in 1.5 ml of
NaCl(l50 mmol/l) was introduced into the gut loops at zero times. Fig. l(a) shows Gly-L-Pro (0)
and Gly-L-Phe (0). Fig. l(b) shows Gly-Gly (O), Gly-D-Phe (0) and carnosine (A).
Tme after peptide injeCtion (mn)
FIG.2. Example of changes in peptide and amino acid concentration in mesenteric plasma with
time. Gly-L-Phe (150 pmol) in 1.5 ml of NaCl (150 mmol/l) was introduced into the gut loop at
zero time. 0 represents Gly-L-Phe concentration, 0 phenylalanine and A glycine.
concentrations were inversely related to the rate of peptide hydrolysis (Table l), whereas
no such relationship exists for the component amino acids. It can be seen that with Gly-L-Phe,
Gly-L-Pro, L-Pro-Gly the increase in plasma concentration of the N-terminal component
amino acid was similar to the increase of the C-terminal amino acid. This was not the
case with Gly-D-Phe; with carnosine the N-terminal amino acid was not studied. Table 1
D.J. Boullin et al.
854
illustrates the increases in plasma peptide and amino acid concentration 15 min after intraluminal injection of peptide, but we found that no matter what time-interval was selected
for comparison, the above qualitative picture was seen.
TABLE
1. Increases in plasma concentrations of peptides and component amino acids, 15 min after intraluminal
injection of 150 pmol of peptide in 1.5 ml of NaCl (150 mmol/l). Each trio of peptide and amino acid concentrations (uuland uu2)represents data obtained from a single experiment. ual and uuz refer to N- and Cterminal amino acids respectively. For Gly-Gly, plasma amino acid is expressed as glycine. Absolute hydrolysis
rates are expressed as meansf SD with the number of estimations in parentheses.
Peptide
Hydrolysis rate
Absolute
Relative
Peptide
detection limit
(pmolirnl)
100
0.01
Plasma concentrations (pnollml)
Peptide
UUl
< 0.01
<0.01
0.05
0.06
< 0.01
0.05
0.22
0.44
0.10
0.18
0.25
0.99
0.31
0.89
0.97
1.36
1.92
0.1 1
0.29
0.23
0.50
0.12
0.30
0.62
0.97
0.12
1.83
2.10
2.66
0.89
0.97
6.02
2.97
1.76
0.31
0.01
140
(pn-101 g-llO min-')
Gly-L-Phe
6505241.9 (6)
Gly-L-Pro
137k28.1 ( 5 )
L-Pro-GIy
15.65 3.7 (6)
GIy-Gly
1 2 3 k 3.3 (5)
Gly-D-Phe
Carnosine
5.82k 3.6 (5)
< 2.0 (5)
21.2
0.01
0.10
2.2
0.9
< 0.3
0.02
0.01
0.04
0.12
1.81
2.52
2.60
1.20
0.58
5.68
1.85
1.33
0.55
-0.03
1.19
13.02
8.96
6.59
0.54
1.37
0 24
0.83
0.26
0.25
1.90
0.96
0.14
048
0.64
1.12
-
Increases in mesenteric plasma alanine concentration were frequently observed, but similar
increases also occurred in jugular venous plasma. Table 2, compiled from results of a single
experiment with Gly-L-Pro, illustrates this phenomenon and contrasts jugular plasma concentrations with those of mesenteric plasma.
Relationship between intraluminal and plasma concentrations
Mesenteric plasma concentrations of peptide and amino acids, after injection of different
In test inal absorption of dipeptides
855
TABLE
2. External jugular (EJ)and superior mesenteric (SM) plasma concentrationsof glycine, proline, alanine
and G ~ Y - L -before
~ O and after intraluminal injection of 150 -01
of G ~ Y - L -in~ 1.5
o ml of NaCl(l50 mmol/l).
Values are from a single experiment and peptide was injected at zero time.
_ _ _ _ _ _ _
~
~~
Plasma concentration (pmol/ml)
Time after
injection (min)
0
15
30
Glycine
Proline
Alanine
Gly-L-Pro
ELI
SM
ELI
SM
EJ
SM
El
0.32
0.41
3-38
2.82
0.09
0.23
2.08
2.72
0.70
-
1.26
2.17
001
-
0.44
1.12
1.41
0.01
0.19
-
0.65
-
0.46
SM
0.01
concentrations of peptide into the gut lumen, were studied using Gly-Gly. We found considerable individual variation in values obtained in separate experiments 15 min after injection of
Gly-Gly in concentrations ranging from 20 to 100 mmol/l (Fig 3). With the lowest concentra-
Concn.of peptide Wcted (pLmol/rnl)
FIG.3. Effect of different concentrationsof Gly-Gly injected intraluminally on mesentefic plasma
concentration of glycine and Gly-Gly. The ordinates represent increases in plasma concentration
from originalvalue (pmol/ml) and the abscissae indicate the intraluminalconcentration(pmol/ml).
Each value of the histograms represents a separate experiment, being obtained from a sample collected 15 min after intraluminal injection of 1.5 ml of peptide in NaCl(l50 mmol/l).
856
D. J. Boullin et al.
tion, plasma values were approximately 0.2 pmol/ml of plasma. With 50 mmol/l they ranged
between 0.2 and 1.6 and with 100 mmol/l between 0.3 and 2.0 pmol/ml. It appears therefore
that the plasma concentrations of Gly-Gly are related to the concentration of peptide introduced into the lumen. Similar remarks apply to the amino acid glycine (Fig. 3).
DISCUSSION
We have been able to show that six dipeptides containing the amino acids p-alanine, glycine,
histidine, phenylalanine and proline are absorbed intact by the small intestine of the rat when
injected intraluminally in high concentrations (20-100 pmol/ml). We demonstrated absorption
directly, detecting the various peptides in the mesenteric venous blood by ion-exchange
chromatography.
Previous workers have obtained evidence for the absorption intact of some of the peptides
we have studied. Peters & MacMahon (1970) and Adibi (1971) have described the absorption
of intact Gly-Gly into the blood in vivo, although Agar, Hird & Sidhu (1953) showed passage
of intact Gly-Gly across the intestine in vitro. Rubino, Field & Shwachman (1971) have obtained data which were highly suggestive of intact absorption of Gly-L-Pro. In human experiments Perry et al. (1967) reported that carnosine could pass intact from the intestinal
lumen into the blood and Navab & Asatoor (1970) have obtained data suggesting mucosal
uptake of intact carnosine in a case of Hartnup disease. Previously, Craft, Geddes, Hyde,
Wise & Matthews (1968) showed that intact absorption of di- and tri-glycine probably occurred
before intracellular hydrolysis. However, this appears to be the first unequivocal demonstration
of the absorption of intact L-Pro-Gly, Gly-L-Phe and Gly-D-Phe.
It is of interest to compare our results relating to increased plasma amino acid concentration
after intraluminal peptide injection, with data obtained by Burston et al. (1972) on peptide
and amino acid uptake into rings of everted rat ileum in vitro. In both cases there was no
obvious relationship between the rate of peptide hydrolysis by intact rings or tissue homogenates and the extent of uptake ofthe constituent amino acids. These results were not surprising,
because there are many factors which will affect amino acid concentrations. Steady-state
conditions do not exist in the situations either in vivo or in vitro. There are various processes
which will influence the concentrations of amino acids in any particular location. Thus in
our experiments in vivo we do not know the amino acid concentrations within the cell, nor
the extent of efflux of amino acids into the intestinal lumen.
The absence of any differences in plasma concentration of C- and N-terminal amino acid
components of the peptides we have studied provides additional support for the idea that
intact dipeptides are removed from the intestinal lumen before significant hydrolysis occurs.
This hypothesis, first put forward for Gly-Gly by Newey & Smyth (1959), would lead one
to expect equal concentrations of C- and N-terminal amino acids of the peptide on the serosal
side of the intestine.
Our observations of increased plasma alanine concentration are in agreement with those
of other workers, since Craft et al. (1968) and Peters & MacMahon (1970) have reported
similar results after the administration of glycine and peptides containing glycine. The mechanism of these increases has not been established. As too little is known of the interactions
between amino acids in blood and tissues it is not possible for us to comment further on this
phenomenon.
Intestinal absorption of dipeptides
857
One question arising as a result of our observations concerns the relevance of the results
in relation to peptide concentrations likely to occur in the gastrointestinal tract after normal
food intake. First, we found that the frequency of detection of unhydrolysed peptide was
inversely related to the rate of peptide hydrolysis. Although our hydrolysis system was in all
probability quite different from conditions in the intact animal, relative comparisons of the
various peptide hydrolysis rates appear valid. Thus it is reasonable to suggest that absorption
of intact peptide into the blood may occur under physiological conditions in the case of those
dipeptides where the rate of hydrolysis is very slow.
Secondly, the experiments with Gly-Gly indicated that absorption of intact peptide and
glycine was approximately proportional to intraluminal concentration of dipeptide, although
the lowest concentration studied (20 ,umol/ml) is higher than any likely to occur under physiological conditions.
Despite the high intraluminal concentrations used in all our studies, our experimental
procedure probably approximates fairly closely to physiological conditions. Whether or not
a significant proportion of protein digestion products will enter the blood as peptides under
normal conditions, and the relevance of such observations to human physiology, can only be
determined by further investigation in vivo.
ACKNOWLEDGMENTS
It is a pleasure to thank the British Nutrition Foundation Ltd for financial support and
Professor D. M. Matthews for his advice and criticism.
REFERENCES
ADDISON,J.M., BURSTON,
D. & MAITHEWS,
D.M. (1972) Evidence for active transport of the dipeptide glycylsarcosine by hamster jejunum in vitro. Clinical Science, 43, 907-911.
ADIBI,S.A. (1971) Intestinal absorption of dipeptides in man: relative importance of hydrolysis and intact
absorption. Journal of Clinical Investigation, 50,2266-2275.
AGAR,W.T., HIRD, F.J.R. & SIDHU,G.S. (1953) The active absorption of amino acids by the intestine. Journal
of Physiology, 121,255-263.
BURSTON,
D., ADDISON,J.M. & MATTHEWS,
D.M. (1972) Uptake of dipeptides containing basic and acidic
amino acids by rat small intestine in vitro. Clinical Science, 43, 823-837.
BUITERWORTH,
K.R. & PELLING,
D. (1973) Mesenteric venous blood sampling in vivo in the rat. Journal of
Physiology, 232, 6 0 ~ 6 1 ~ .
CHEESEMAN,
C.I. & SMYTH,
D.H. (1973) Specific transfer process for intestinal absorption of peptides. Journalof
Physiology, 2 2 9 , 4 5 ~ - 4 6 ~ .
CRAFT,I.L., GEDDES,D., HYDE,C.W., WISE,I.J. & MATTHEWS,
D.M. (1968) Absorption and malabsorption
of glycine and glycine peptides in man. Gut, 9,425-437.
HEIZER,
W.D. & LASTER,
L. (1969) Hydrolases in the mucosa of rat small intestine for phenylalanine-containing
dipeptides. Biochimica et Biophysica Acta, 185,409-423.
HUECKEL,
H.J. & ROGERS,Q.R. (1972) Prolylhydroxyproline absorption in hamsters. Canadian Journal of
Biochemistry, 50,782-790.
NAVAB,F. & ASATOOR,
A.M. (1970) Studies on intestinal absorption of amino acids and a peptide in a case
of Hartnup disease. Gut, 11,373-379.
NEWEY,H. & SMYTH,
D.H. (1959) The intestinal absorption of some dipeptides. Journal of Physiology, 145,
48-56.
858
D. J. Boullin et al.
PERRY,T.L., HANSEN,
S., TISCHLER,
B., BUNTING,
R. & BERRY,K. (1967) A new metabolic disorder associated
with neurological disease and mental defect. New England Journal of Medicine, 227, 1219-1227.
PETERS,
T.J. & MACMAHON,
M.T. (1970) The absorption of glycine and glycine oligopeptides by the rat. Clinical
Science, 39, 811-821.
RUBINO,A., FIELD,M. & SHWACHMAN,
H. (1971) Intestinal transport of amino acid residues of dipeptides.
Journal of Biological Chemistry, 246,3542-3548.