![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
*
Laboratoire de Biochimie Cellulaire,
Laboratoire de Biochimie de la Nutrition and
**
Laboratoire de Biologie Cellulaire, Université Catholique de Louvain, B-1348 Louvain-la-Neuve, Belgium
2To whom correspondence should be addressed.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
KEY WORDS: • Caco-2 cells • L-phenylalanine • intestinal transport • protein incorporation
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and 1984
), guinea pig (Satoh et al. 1989
) and mouse (Berteloot et al. 1982
)
have shown that the essential large neutral amino acid
L-phenylalanine (L-Phe) crosses the brush
border membrane via simple passive diffusion as well as via
Na+-independent and
Na+-dependent carriers. Munck and Munck (1994)
have shown that in the brush border membrane of the
rabbit small intestine, L-Phe is transported via an
Na+-dependent carrier system for neutral amino
acids but also shares both Na+-dependent and
Na+-independent transporters with
L-Lys. Other studies in the intestine have shown that large
neutral amino acids and cationic amino acids share
Na+-dependent and
Na+-independent transport systems (Harvey et al. 1993
, Munck et al. 1992
, Wolfram et al. 1984
). In humans, uptake studies performed on brush
border membrane vesicles from fetal and adult small intestine also
reported the presence of Na+-dependent and
Na+-independent transport systems for neutral
amino acids (Lücke et al. 1977
, Malo 1991
). Due to the wide interspecies variation that occurs in amino acid transport, it is difficult to extend the results obtained in animal models to humans. In addition, studies on human tissue are rare because of difficulties in obtaining experimental material. For these reasons, the use of a human cell culture system as an in vitro model of the intestinal barrier would be a useful complementary approach to characterize the mechanisms and the factors that regulate the intestinal transport of amino acids. One such system based on the cultivation of the Caco-2 cell line on microporous membranes is increasingly used to investigate the physiology of the enterocytes as well as the mechanisms involved in the transport of nutrients and pharmacological agents.
From the early 1990s, the intestinal transport of various amino acids
has been studied using the Caco-2 model system. Hidalgo et al. (1990)
described the transport of the large neutral amino acid
L-Phe across Caco-2 cell monolayers grown on microporous
filters. The results show that the AP-to-BL transport of
L-Phe is temperature dependent, saturable and inhibited by
metabolic inhibitors. In addition, the carrier system involved requires
Na+ and interacts with large neutral and cationic
amino acids but not with small neutral or anionic amino acids. In a
later report, the same workers (Hu and Borchardt 1992
)
characterized the processes involved in L-Phe uptake and
efflux from both the AP and BL sides of the Caco-2 cells. This study
shows that the AP uptake exhibits characteristics very different from
those of the BL uptake. The major carrier involved in the AP uptake
resembles the Na+-dependent
B0,+ system, whereas the carriers involved in the
BL uptake include the Na+-dependent
B0,+ and ASC systems as well as the
Na+-independent L system. Nevertheless, an
earlier report by Hidalgo et al. (1988)
indicated that
L-Phe transport across Caco-2 cell monolayers was not
reduced with Na+-free medium or in the presence
of ouabain.
The present study was undertaken to extend the knowledge about the transport of L-Phe across monolayers of Caco-2 cells grown in a serum-free, hormone-defined medium on microporous membrane supports. Moreover, we particularly sought to determine the relative implications of the paracellular pathway and the transcellular pathway with their various carrier systems localized in the AP and BL membranes, as well as the intracellular fate of L-Phe that was taken up and, in particular, its use for the protein synthesis.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Iscove’s modified Dulbecco’s medium, Ham’s F12 medium, NCTC 135 medium, Hanks’ balanced salt solution (HBSS, powder form) and HEPES were purchased from GIBCO BRL (Life Technologies Ltd., Paisley, U.K.). L-Phe, L-Leu, L-Trp, L-Tyr, L-Lys, L-His, glycine, ouabain, 2-deoxyglucose (2-DOG), choline chloride, deoxycholic acid sodium salt, EGTA and cycloheximide were purchased from Sigma Chemical Co. (St. Louis, MO). Trichloroacetic acid (TCA), sodium azide, NaOH, NaHCO3 and Tris were purchased from Merck (Darmstadt, Germany). Fetal bovine serum was purchased from BioWhittaker Europe (Verviers, Belgium). L-[2,3,4,5,6-3H]Phe (L-[3H]Phe, specific activity 4.85 TBq/mmol) and D-[1-14C]mannitol (D-[14C]mannitol, specific activity 2.07 GBq/mmol) were purchased from Amersham Life Sciences (Little Chalfont, U.K.).
Cell culture.
The intestinal Caco-2 cell line was cultivated as described
(Berger et al. 2000
). Briefly, the cells are routinely grown in plastic
tissue culture flasks (175 cm2; Greiner Labortechnik,
Frickenhaussen, Germany) using basal defined medium (Schneider 1989
), which consists mainly of a 5:5:1 (v/v/v) mixture of
Iscove’s modified Dulbecco’s, Ham’s F12 and NCTC 135 media and was
adapted for the cultivation of Caco-2 cells (Halleux and Schneider 1991
).
For transport experiments, efflux experiments and metabolism studies,
the cells were seeded onto a poly(ethylene terephthalate) microporous
membrane (1-µm pore diameter, 11-µm thickness, 1.6 x 106 pores/cm2; Whatman SA, Louvain-la-Neuve,
Belgium) as described (Berger et al. 2000
). All cells used in this
study were between passages 184 and 211. Experiments were always
performed 
24 h after the last feeding because the time after feeding
is known to affect the transport of amino acids. The integrity of the
monolayers was checked through measurement of the transepithelial flux
of D-[14C]mannitol and the transepithelial
electrical resistance (TEER) (Millicell-ERS Resistance System;
Millipore, Bedford, MA and Endohm 24; World Precision Instruments,
Sarasota, FL) across the monolayers. All of the monolayers used in this
study exhibited TEER values of 400–600 
· cm2 and
D-[14C]mannitol clearance values of <30 µL
· insert-1 · h-1.
Transport studies.
The culture medium was replaced by HBSS containing 5 mmol/L glucose and supplemented with 10 mmol/L HEPES and 5 mmol/L NaHCO3 (transport medium). The pH value of the transport medium was adjusted to pH 7.4 with NaOH or Tris for Na+-free medium. Before transport experiments, the cells were depleted of amino acids through incubation for 30 min at 37°C in transport medium, after which the donor and acceptor solutions were replaced with the appropriate solutions prewarmed at 37°C.
Transepithelial passage was quantified by the addition of
D-[14C]mannitol and
L-[3H]Phe in either the upper or the lower
compartment of the bicameral culture insert and by analysis of the
solution of the other compartment. The upper and lower compartments
contained 1.8 or 2.8 mL transport medium, respectively. The donor
solution contained the radioactive compounds (16.7 MBq/L
L-[3H]Phe and 16.7 MBq/L
D-[14C]mannitol) dissolved in transport
medium in the presence of a varying amount of nonlabeled
L-Phe to give the required final concentration. At each
sampling time (60, 120 and 180 min), 100 µL transport medium was
withdrawn from the acceptor compartment and replaced with fresh
transport medium. The rates of transport were calculated from the slope
of the L-Phe or D-mannitol appearance curve.
To determine sodium dependency of transport, sodium chloride in the transport medium was replaced with an equimolar amount of choline chloride and sodium salts were replaced by their potassium equivalents. Alternatively, ouabain (5 mmol/L) was added to the sodium-containing transport medium in both compartments to block the Na+,K+-ATPase responsible for maintaining the sodium gradient across the cell membrane. For energy depletion, transport medium depleted in glucose was used in both compartments. Transport medium depleted in glucose but supplemented with 1 mmol/L NaN3 and 50 mmol/L 2-DOG or with 5 mmol/L NaN3 alone was also used in both compartments. For protein synthesis inhibition, 50 mg/L cycloheximide was added to the transport medium in both compartments. For sodium replacement, energy depletion and protein synthesis inhibition, the cells were preincubated in the appropriate transport medium for 30 min. For the experiments at 4°C and 10°C, cells were preincubated for 30 min at 4°C or 10°C before the addition of the labeled probes. The absence of toxicity of these experimental conditions was checked with optical phase contrast microscopy examination and assayed by measurement of the TEER at the end of the transport experiment. To assess the effect of tight junction opening on transport, transport buffer was supplemented with 2.5 mmol/L EGTA in both compartments. For the competition experiments, an excess of unlabeled amino acid (1 mmol/L) was added to the donor compartment together with the radiolabeled L-Phe (10 µmol/L).
Intracellular accumulation.
At the end of the transport experiment (180 min), the cell monolayers
were washed six times with phosphate-buffered saline and then
treated as previously described (Berger et al. 2000
).
Efflux experiments.
The cell monolayers were loaded with 10 µmol/L L-[3H]Phe for 15 min at 37°C and then washed twice with ice-cold HBSS. The intracellular accumulation of L-[3H]Phe was determined on three samples as described earlier. For the other inserts, the preloaded L-[3H]Phe was then allowed to efflux into the media bathing the AP and BL sides of the monolayers. After various durations, ranging from 10 to 90 min, the amounts of L-[3H]Phe that were released were determined by measuring the radioactivity of 100-µL samples withdrawn from the AP and BL media.
Radioactive measurements.
For all of the experiments, the amount of radioactive material was analyzed by liquid scintillation spectrometry with a Packard Tri-Carb 1600 TR instrument (Packard Instrument Company, Meriden, CT) after dispersion in 2 mL Aqualuma cocktail (Lumac LSC, Groningen, the Netherlands).
Biotransformation.
The level of L-Phe biotransformation by the Caco-2 cells
was checked at the end of the transport experiment (180 min). For this
purpose, the identity of the 3H-label (either in the donor
compartment or that collected from the acceptor compartment as well as
that of the non–TCA-precipitable cell fraction) was determined through
thin layer chromatography (TLC) as described (Berger et al. 2000
).
The level of L-Phe biotransformation into other amino acids
by the Caco-2 cells was checked through HPLC analysis of cellular
extracts and culture media. Increases in various amino acids were
examined through comparison with a reference sample. After a
preincubation of 30 min at 37°C in HBSS, the cells were incubated for
an additional 3 h at 37°C with 100 µmol/L L-Phe in
HBSS or with HBSS alone (reference sample) in the donor compartment and
with HBSS in the acceptor compartment. At the end of the incubation
period, the solutions bathing the AP and BL poles were collected, and
the cells were removed by scraping in bidistilled water and disrupted
by ultrasonication. Samples were treated and analyzed as described
(Berger et al. 2000
).
Data analysis.
Results are expressed as means ± SD. Statistical analysis was performed using one-way ANOVA with a Dunnett post hoc test for comparisons between test assays and controls or a Tukey post hoc test for multiple comparisons. The computer program was Systat 5.2.1 (Systat, Evanston, IL).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The passages of D-mannitol and L-Phe were
studied on Caco-2 cells cultivated for 17–19 d after seeding over the
time range of 0–180 min with a 60-min sampling interval (Fig. 1
). To facilitate direct comparison, the results are presented as
clearance, which is, by definition, the volume of plasma containing the
amount of substance that is removed by the kidney per unit of time. By
analogy, in this study, we defined clearance as the equivalent volume
of solution (µL) containing the amount of labeled material that
passed across the cell monolayer. The results indicate that the passage
of D-mannitol in both the AP-to-BL and BL-to-AP directions
and the passage of L-Phe in the AP-to-BL direction were
continuous over the duration of the experiment. The clearance rate of
D-mannitol was similar in the AP-to-BL and BL-to-AP
directions, reaching values of 23 and 26 µL ·
insert-1 · h-1,
respectively. The passage of L-Phe was systematically
higher than that of D-mannitol. In addition, the AP-to-BL
transport of L-Phe was faster than the BL-to-AP transport.
|
, suggesting
that the transport of L-Phe was not affected by the
duration of the differentiation phase in culture as soon as the Caco-2
cells had reached their differentiation as estimated by the appearance
of TEER values of 
470 · cm2.
EGTA, a calcium chelator that opens intercellular tight junctions,
decreased the transepithelial electrical resistance of Caco-2
monolayers in our experimental medium from 506 ± 48 to 20 ± 13
· cm2 after 3 h of contact with the cells.
Simultaneously, the clearance rate of D-mannitol increased
from 24.2 ± 10.5 to 210.5 ± 45.5 µL ·
insert-1 · h-1 in the
AP-to-BL direction and from 19.6 ± 4.8 to 232.6 ± 23.1 µL
· insert-1 · h-1 in
the BL-to-AP direction. Similarly, the clearance rate of
L-Phe increased from 170.6 ± 27.3 to 223.5 ± 41.0 µL · insert-1 ·
h-1 in the AP-to-BL direction and from 57.8
± 1.8 to 211.6 ± 17.6 µL ·
insert-1 · h-1 in the
opposite direction. In the presence of EGTA, the passages of
L-Phe and D-mannitol became similar, and
L-Phe crossed the monolayers at almost the same rate in
AP-to-BL and BL-to-AP directions.
Intracellular accumulation of L-Phe.
After measurement of the transport across the cell monolayers for
3 h, the intracellular accumulation of L-Phe was
recorded. Caco-2 cell monolayers cultivated for 17–19 d and incubated
for 3 h with 10 µmol/L
L-[3H]Phe accumulated 1538 ± 529 pmol/mg cell protein (n = 37) from the AP pole and
2110 ± 529 pmol/mg cell protein (n = 36) from the
BL pole. The proportion of cell-accumulated
3H-labeled material precipitable by TCA was 88
± 4 and 90 ± 4% from the AP and BL poles of the cells,
respectively. In addition, the intracellular concentration of
L-Phe in the TCA-soluble fraction, calculated on the
basis of a previously determined cellular volume for Caco-2 cells of
3.66 µL/mg protein (Brunham et al. 1989
), was 48.3
± 20.2 µmol/L from the AP side and 57.9 ± 24.3 µmol/L
from the BL side. These results suggest that Caco-2 cells concentrated
L-Phe from both the AP and BL sides of the monolayers and
accumulated it preferentially from the BL pole (P < 0.05). In both cases, 90% of the accumulated L-Phe was
incorporated into newly synthesized proteins.
Biotransformation of L-Phe.
After 3 h of incubation of Caco-2 cell monolayers with 10 µmol/L
L-[3H]Phe, the identity of the
3H label in the AP and BL bathing solutions as
well as in the TCA-soluble cell fraction was determined with TLC.
In the medium bathing the BL pole and in the TCA-soluble cell
fraction as well as in the medium bathing the AP pole, when
L-[3H]Phe was added in the upper
compartment, 80% of the 3H radioactivity
migrated in the position corresponding to the
L-[3H]Phe standard and no other
major radioactive products were detected. In contrast, when
L-[3H]Phe was added in the lower
compartment, two major radioactive products were detected in the medium
bathing the AP pole, with each of them representing 40% of the total
radioactivity: one migrated in the position corresponding to the
L-[3H]Phe standard, and the other
migrated faster, therefore corresponding to an unknown compound but
more hydrophobic than L-Phe.
Because one possible pathway for the biotransformation of L-Phe is its hydroxylation into L-Tyr, a reaction catalyzed by the phenylalanine hydroxylase, the presence of L-Tyr in the samples was examined. HPLC analysis did not show a higher concentration of L-Tyr when the cells were incubated in the presence of 100 µmol/L L-Phe than in the reference sample (cells incubated with HBSS alone), indicating that Caco-2 cells did not significantly hydroxylate L-Phe into L-Tyr.
Concentration dependence of L-Phe transport and accumulation.
The transport of L-Phe was investigated at concentrations
ranging from 10 µmol/L to 10 mmol/L in the AP-to-BL and BL-to-AP
directions. The results indicate an absence of saturation (Fig. 2
). Moreover, transport in the AP-to-BL and BL-to-AP directions became
similar at concentrations of >1 mmol/L. This likely results from
saturation of the carrier systems involved in L-Phe
transport at concentrations of >1 mmol/L; transport then
resulted mainly from passive movement (via the paracellular
route, simple diffusion, or both).
|
) or BL (Fig. 3B
) poles was investigated at
different L-Phe concentrations after 3 h of incubation
at 37°C. The amount of radioactive material in the TCA-insoluble
portion was also measured. The results show that in both cases, the
total intracellular accumulation from both the AP and BL poles
increased at first and reached a quasiplateau at 1 mmol/L. A
saturation of the amount of
L-[3H]Phe incorporated
into TCA-precipitable material was observed. At concentrations of
<0.5 mmol/L, the bulk of the L-Phe molecules accumulated
from either the AP or the BL pole were incorporated into proteins,
whereas at higher concentrations, the proportion of the intracellular
L-Phe incorporated into proteins became progressively smaller. Regardless of the concentration, the accumulation from the BL
pole was about twice that from the AP pole (Fig. 3)
.
|
To establish whether the transport of L-Phe was energy
dependent, experiments were performed with monolayers incubated with
metabolic inhibitors or in a transport medium depleted in glucose.
Supplementation of the transport medium with 5 mmol/L
NaN3, a metabolic inhibitor, as well as the
absence of glucose in the transport medium did not affect the AP-to-BL
and BL-to-AP transports of L-Phe (not shown). The
simultaneous presence of 1 mmol/L NaN3 and 50
mmol/L 2-DOG to block glucose uptake and utilization had no effects on
the AP-to-BL transport, whereas it increased the BL-to-AP transport of
L-Phe to 183 ± 31% of the control value
(P < 0.05). This increase could result from tight
junction alteration by this strong energy depletion, because gate
function of tight junctions is itself ATP dependent (Mandel et al. 1993
). The BL-to-AP passage of
D-[14C]mannitol
in the presence of these metabolic inhibitors was higher (200 ± 62%) than in the control condition. We reported (Berger et al. 2000
)
that the increase in the passage of D-mannitol as a result
of treatment with metabolic inhibitors did not result from an osmotic
effect of 2-DOG.
At 10°C and 4°C, temperatures at which energy-dependent processes are considerably slowed, the AP-to-BL transport was reduced to 42 ± 15 and 23 ± 3% (P < 0.01) of the control values, respectively, and the BL-to-AP transport was reduced to 59 ± 7 and 44 ± 4% of the control values, respectively (P < 0.05). In addition, the transport at 4°C in both directions became close to the levels of the paracellular passage. Indeed, the AP-to-BL passage of L-Phe and D-mannitol at 4°C was 27.3 ± 0.7 and 11.2 ± 6.6 µL · insert-1 · h-1, respectively, whereas the BL-to-AP passage was 29.8 ± 3.3 and 12.9 ± 6.0 µL · insert-1 · h-1, respectively
The transport of amino acids has been described as coupled with
Na+ (Christensen 1990
,
Maillard et al. 1995
). Therefore, the transepithelial
transport of L-Phe in the absence of
Na+ was studied. The substitution of sodium
chloride in the transport medium with choline chloride failed to affect
the AP-to-BL transport. In contrast, it increased the BL-to-AP
transport to 154 ± 32% of the control value (P
< 0.05). The mechanism responsible for such increase is not yet
clear. In the presence of 5 mmol/L ouabain, an inhibitor of the
Na+,K+-ATPase, the
transepithelial transport of L-Phe was not significantly
different from control values. These results suggest that the AP-to-BL
transport of L-Phe is an
Na+-independent process.
Treatment of the monolayers with cycloheximide, an inhibitor of protein synthesis, increased the AP-to-BL transport of L-Phe to 137 ± 20% (P < 0.05) but did not significantly affect the BL-to-AP transport.
The intracellular accumulation of L-Phe from the AP and BL
poles was considerably inhibited by the presence of metabolic
inhibitors or of cycloheximide, the decrease in the incubation
temperature and, to a lesser extent, by glucose depletion (Fig. 4
). Under these conditions, the incorporation of L-Phe into
TCA-precipitable labeled material became very low. By contrast, the
absolute amount of intracellular 3H-labeled
material in the TCA-soluble fraction was not significantly affected
except for accumulation from the BL side at 4°C and 10°C (not
illustrated). In these cases, the amount of 3H
label soluble in TCA is higher than in the control condition. Sodium
removal or the addition of 5 mmol/L ouabain decreased the intracellular
accumulation by 40% from both the AP and BL sides of the cells
without affecting the incorporation of L-Phe into
TCA-precipitable material (Fig. 4)
. These results indicate that the
accumulation of L-Phe was sensitive to
Na+-free media and ouabain, which appears in
contrast to the transport process.
|
, the AP-to-BL transport was significantly decreased by the presence of
an excess of D-Phe, large neutral amino acids
(L-Leu, L-Tyr, L-Trp) and cationic
amino acids (L-His, L-Lys), whereas BL-to-AP
transport was unaffected by any of the tested amino acids. The small
neutral amino acid glycine had no effect on L-Phe transport
in either direction.
|
). The inhibitory effects of these amino acids were much more dramatic
from the AP side than from the BL side of the monolayers. Under the
control condition (i.e., 10 µmol/L
L-[3H]Phe), 89 ± 3 and 90
± 3% of the accumulated L-Phe from the AP and BL
sides, respectively, was incorporated into TCA-precipitable
material. These values were not significantly affected by the presence
of any of the other amino acids tested. These results are consistent
with the specificity of the aminoacyl-tRNA synthetases toward the
amino acid and its corresponding tRNA.
|
The effect of cycloheximide on L-Phe transport and
accumulation (Fig. 4)
indicates that a large proportion of the
L-Phe molecules taken up by the cells was incorporated into
proteins and that this incorporation reduced the amount of
L-Phe available for transcellular transport. Under these
conditions, it was difficult to determine whether the effects induced
by the various experimental conditions on L-Phe transport
and accumulation resulted from an action on the transport systems,
on the protein synthesis pathway or a combination. Therefore, the
effects of substrate concentration, temperature, inhibitors of
metabolism, absence of Na+ and other amino acids
on the processes of transport and intracellular accumulation were also
studied with cycloheximide-treated monolayers.
The rate of transport across cycloheximide-treated monolayers at
various L-Phe concentrations ranging from 5 µmol/L to 5
mmol/L was measured after 1 h of incubation and plotted against
the concentration. In the AP-to-BL direction, the amount of
L-Phe transported across the monolayers increased at first
and then tapered off at concentrations of >0.5 mmol/L (Fig. 7
). This curve likely results from the contribution of a saturable
process as well as of a nonsaturable one to the total L-Phe
transport. In contrast, BL-to-AP transport was strictly proportional
over the range of concentrations. The transport in the AP-to-BL and
BL-to-AP directions at different L-Phe concentrations also
was measured at 4°C. The results (not shown) demonstrate that the
transport at 4°C became similar in both directions and was
proportional to L-Phe concentration and lower than the
transport at 37°C, especially in the AP-to-BL direction.
|
0.25 mmol/L (Fig. 8
). At concentrations of >1 mmol/L, it increased again with increasing
concentration. These results suggest the existence of a saturable
component as well as of a nonsaturable one. By contrast, accumulation
from the BL side was not saturable.
|
30%
from both the AP and BL sides. The intracellular accumulation of
L-Phe from the AP pole was significantly inhibited (72
± 5%) by decreasing the temperature of incubation from 37°C to
4°C (P < 0.001). In contrast, the same condition
increased the accumulation from the BL pole by 43 ± 9%
(P < 0.01), suggesting that the efflux of
L-Phe across the AP membrane occurred not only via a simple
passive diffusion process. Coincubation with the stereoisomer
D-Phe, the large neutral amino acids L-Leu and
L-Tyr and the cationic amino acid L-His
decreased the accumulation from the AP pole by 29 ± 22, 63
± 8, 54 ± 13 and 22 ± 6%, respectively (P
< 0.05). L-Tyr inhibited BL accumulation by 34
± 11% (P < 0.01), whereas D-Phe and
L-His had no effect. L-Leu tended (P
= 0.08) to have a small inhibitory (32 ± 3%) effect on BL
accumulation. The cationic amino acid L-Lys and the small
neutral amino acid glycine failed to modify the intracellular
accumulation of L-Phe from both poles of the cells. Uptake and efflux of L-Phe.
Monolayers were preloaded from the AP side (for BL efflux studies) or
from the BL side (for AP efflux studies) of the cells with 10 µmol/L
L-[3H]Phe for 15 min. The
accumulation of L-Phe from either the AP or BL side of the
monolayers achieved under these conditions is given in Figure 9
(inset). After the cells were washed with ice-cold HBSS,
the efflux of the preloaded
L-[3H]Phe from the BL or
AP side of the monolayers was monitored as a function of time at 37°C
(Fig. 9)
. The amount that effluxed at the BL side increased until the
end of the experiment (90 min), and 30% of the preloaded
L-Phe from the AP side of the cells was released at their
BL pole. In addition, 77 ± 11% of the cell-associated
radioactivity at the end of the efflux experiment was precipitable by
TCA, indicating that the 3H label that was not
effluxed principally corresponded to
L-[3H]Phe incorporated
into proteins. On BL loading, only a very small proportion (<5%) of
the preloaded L-Phe was released at the AP pole, and 89
± 6% of the cell-associated radioactivity at the end of the
efflux experiment was precipitable by TCA. These results indicate that
L-Phe accumulated by the cells was selectively effluxed at
their BL side.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). The transport of L-Phe in the AP-to-BL and BL-to-AP directions and the accumulation from both the AP and BL sides of the cells were studied. AP-to-BL transport and AP accumulation are useful to characterize the processes involved in nutrient absorption. BL-to-AP transport and BL accumulation were investigated to better identify the polarity of distribution of the carriers involved in the transport of L-Phe but also to understand how the nutritional supply to the intestinal cells is maintained when there is no exogenous source of nutrient in the absorption sites. In addition to the transport mechanisms, the incorporation of L-Phe into newly synthesized proteins was taken into account.
TLC and HPLC analyses indicated that L-Phe was not appreciably metabolized by the Caco-2 cells when added in the upper compartment of the bicameral insert, mimicking the AP pole of the enterocytes. In contrast, the appearance of an unknown radiolabeled compound was observed in the medium bathing the AP pole when L-[3H]Phe was added in the lower compartment. Therefore, the bulk of 3H-labeled material accumulated by the Caco-2 cells monolayers as well as transported in the AP-to-BL direction reflected intact L-[3H]Phe, whereas only the half of tritium transported in the BL-to-AP direction still reflected intact L-[3H]Phe.
The transepithelial transport of L-Phe was not affected by
the duration of the differentiation phase in culture. Similar results
were previously reported for the transport of large neutral amino acids
(Hidalgo et al. 1990
) and of the cationic amino acid
L-Lys (Ferruzza et al. 1995
) across Caco-2
cell monolayers.
L-Phe was transported faster from the AP to the BL side of the Caco-2 cells than in the opposite direction, indicating the presence in the Caco-2 cells of one or more polarized systems for the intestinal absorption of large neutral amino acids. The passage of L-Phe in the AP-to-BL and BL-to-AP directions was much faster than that of D-mannitol, which crosses the cell monolayer exclusively through the paracellular route. These results suggest that the transcellular transport pathway of L-Phe outweighed the paracellular pathway.
No energy dependency of the transepithelial bidirectional transport of L-Phe was demonstrated by treatment of the monolayers with metabolic inhibitors. In contrast, AP-to-BL and BL-to-AP transports of 10 µmol/L L-[3H]Phe were reduced by >70 and 50%, respectively, when the temperature of incubation was lowered from 37°C to 4°C. However, this temperature dependency may in part result from a decrease in the cellular membrane fluidity at 4°C. Kinetics of the AP-to-BL and BL-to-AP transports did not allow us to identify a clearly saturable component.
L-Phe transport in the AP-to-BL direction was significantly
increased in the presence of cycloheximide, suggesting that a
proportion of L-Phe taken up by the Caco-2 cells was
diverted to protein synthesis pathways during the transport process.
This was confirmed by the fact that most of L-Phe
accumulated by the Caco-2 cells was incorporated into newly synthesized
proteins and that L-Phe accumulation was sensitive to
cycloheximide. A similar effect of cycloheximide on transport was
observed in a study characterizing the transendothelial transport of
L-Leu across bovine brain microvessel endothelial cell
monolayers (Audus et al. 1986
).
The addition of metabolic inhibitors, a decrease in the temperature of incubation and, to a lesser extent, glucose depletion significantly reduced both AP and BL accumulations. These reductions seem to result from an inhibitory action on protein synthesis, because the incorporation of L-[3H]Phe into newly synthesized proteins was significantly decreased by these experimental conditions, whereas the amount of L-[3H]Phe in the TCA-soluble fraction was not affected.
The absence of Na+ or the presence of ouabain in
the transport medium did not affect the transepithelial transport of
L-Phe, indicating that L-Phe transport across
Caco-2 cell monolayers was Na+ independent. The
absence of Na+ dependency of the AP-to-BL
transport of L-Phe disagrees with the previously published
observation that L-Phe transport across Caco-2 cell
monolayers was reduced by 33% in the presence of 100 µmol/L ouabain
(Hidalgo et al. 1990
). Nevertheless, an earlier report
by the same workers indicated that L-Phe transport across
Caco-2 cell monolayers was not reduced from
Na+-free transport medium or by 2–5 mmol/L
ouabain (Hidalgo et al. 1988
).
In contrast to transport, the accumulation of L-Phe from
both poles was Na+ dependent. Two hypotheses
could explain the difference observed between transport and
accumulation. The first hypothesis is based on the presence of
Na+-dependent carrier systems at both the AP and
BL poles of the Caco-2 cells. However, the Na+
dependency of the transport would have been masked because
L-Phe transported by the Na+-dependent carrier
systems would be preferentially diverted to an intracellular pool
committed to protein incorporation and not available for transport. The
second hypothesis is also based on the presence of
Na+-dependent carrier systems at both sides of
the cells. It implies that in Na+-free medium or
in the presence of ouabain, the Na+-dependent carrier
system localized on the pole opposite to that on which
L-[3H]Phe was added become
implicated in the efflux of the accumulated L-Phe molecules
as a result of the greater intracellular-to-extracellular
Na+ gradient. It therefore compensates for the
decrease in uptake resulting from the absence of
Na+. The observation of Hu and Borchardt (1992)
that the BL efflux of
L-[14C]Phe was increased in the
absence of Na+ or in the presence of ouabain
could support this hypothesis.
In the absence of protein synthesis (cycloheximide-treated monolayers), the kinetics of the AP-to-BL transport at 37°C identified a saturable component as well as a nonsaturable one. In contrast, the BL-to-AP transport was not saturable. The transport at 4°C was similar in the two directions and proportional to concentration. This nonsaturable transport component probably represents the contributions of the paracellular passage and/or the transcellular passive diffusion to the total transport of L-Phe across Caco-2 cell monolayers. Similarly, the kinetics of L-Phe accumulation into cycloheximide-treated CaCo-2 cells identified a saturable component and a nonsaturable one for AP accumulation, whereas BL accumulation was not saturable.
The results of the efflux studies indicate that the accumulated L-Phe molecules were selectively effluxed at the BL side of the Caco-2 cells. In addition, the 3H-labeled material not effluxed at the end of the experiment was mainly incorporated into proteins.
To characterize the specificity of the carriers responsible for
the transepithelial transport of L-Phe, we investigated the
effect of selected amino acids on L-Phe transport and
accumulation. The AP-to-BL transport and AP accumulation were
significantly inhibited by elevated extracellular levels of large
neutral, aromatic and cationic amino acids, indicating that
L-Phe was transported from the AP-to-BL side via a set of
carrier systems interacting with bulky neutral and cationic amino
acids. The lack of effect of glycine most likely suggests that these
carriers have no affinity for small neutral amino acids. The low but
significant inhibitory effect of the D-isomer on AP-to-BL
transport suggests that the carriers involved in the AP-to-BL transport
of L-Phe have only moderate stereospecificity. These
results are consistent with previous studies performed on Caco-2 cells
in which it was suggested that large neutral amino acids and cationic
amino acids may share the same carrier systems (Ferruzza et al. 1995
, Hidalgo et al. 1990
, Hu and Borchardt 1992
). In contrast to transport and accumulation from
the AP side, transport and accumulation from the BL side were only
moderately inhibited by coincubation with large neutral amino acids,
suggesting that different carrier systems mediate the AP-to-BL and
BL-to-AP transports of L-Phe.
To determine the various amino acid carrier types involved in the
transport of L-Phe, our results were considered in the
context of the classification of the neutral amino acid transport
systems described by Christensen (1990)
. The transport
of L-Phe in the AP-to-BL direction seems to be mediated
mainly via a combination of systems b0,+
and L. The evidence supporting this hypothesis includes the absence of
an effect of sodium removal, ouabain and metabolic inhibitors on
transport (Na+-independent carrier systems
characteristics) and the inhibition of transport by large neutral amino
acids and cationic amino acids (system
b0,+ characteristics) and by bicyclic
amino acids (system L characteristics). However, it is likely that an
Na+-dependent system may participate in the
transport of L-Phe from the AP pole because AP accumulation
was decreased by sodium removal and ouabain. The results also suggest
that L-Phe transport from the BL pole at the concentration
of 10 µmol/L does not occur via passive diffusion only. Indeed,
BL-to-AP transport was significantly inhibited at 4°C; BL
accumulation was significantly decreased by sodium removal and ouabain
as well as by coincubation with large neutral amino acids, and
L-Phe was concentrated intracellularly on accumulation from
the BL pole. Our results in part agree with previous reports on large
neutral amino acid uptake in filter-grown Caco-2 cell monolayers
(Chen et al. 1994
, Hu and Borchardt 1992
)
that showed the uptake of the large neutral amino acids
L-Met and L-Phe was mediated via the
combination of systems B0,+, ASC and L.
In conclusion, the study goal was characterization of the transepithelial transport and intracellular accumulation of the essential amino acid L-Phe in the Caco-2 cells. The transport of L-Phe in the AP-to-BL direction was shown to be mediated by a combination of the Na+-independent b0,+ and L systems, whereas it is likely that an Na+-dependent carrier may also participate. These carrier systems perform an important nutritional function because they ensure the absorption of L-Phe from the intestinal lumen to the blood circulation. This is particularly important in the case of L-Phe, which must be obtained from the diet. By contrast, transport in the BL-to-AP direction results mainly from passage through nonspecific pathways (paracellular route and/or transcellular passive diffusion). However, the transport of L-Phe across the BL membrane was shown to be, at least in part, mediated by carrier systems, indicating that L-Phe that enters the intestinal cells from the BL side via carrier systems is not delivered at their AP side but is for the large part used for their own maintenance and metabolism.
Finally, the results obtained confirm the usefulness of the Caco-2 cell culture system to study intestinal absorption and transport regulation of the amino acids. Nevertheless, the Caco-2 cell culture system has some limitations that must be taken into account in interpretation of the results of transport experiments, such as its tumoral origin, its colonic instead of small intestine origin, the absence of the mucin-producing goblet cells and the unstirred conditions under which the transport experiments have been performed.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
3 Abbreviations used: AP, apical; BL, basolateral; 2-DOG, 2-deoxyglucose; HBSS, Hanks’ balanced salt solution; TCA, trichloroacetic acid; TEER, transepithelial electrical resistance; TLC, thin layer chromatography.
Manuscript received February 8, 2000. Initial review completed March 15, 2000. Revision accepted July 25, 2000.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1. Audus K. L., Borchardt R. T. Characteristics of the large neutral amino acid transport system of bovine brain microvessel endothelial cell monolayers. J. Neurochem. 1986;47:484-488[Medline]
2.
Berger V., De Bremaeker N., Larondelle Y., Trouet A., Schneider Y.-J. Transport mechanisms of the imino acid L-proline in the human intestinal epithelial Caco-2 cell line. J. Nutr. 2000;130:2772-2779
3. Berteloot A., Khan A. H., Ramaswamy K. K+- and Na+-gradient-dependent transport of L-phenylalanine by mouse intestinal brush border membrane vesicles. Biochim. Biophys. Acta 1982;691:321-331[Medline]
4.
Brunham D. B., Fondacaro J. D. Secretagogue-induced protein phosphorylation and chloride transport in Caco-2 cells. Am. J. Physiol. 1989;256:G808-G816
5. Chen J., Zhu Y., Hu M. Mechanisms and kinetics of uptake and efflux of L-methionine in an intestinal epithelial model (Caco-2). J. Nutr. 1994;124:1907-1916
6.
Christensen H. N. Role of amino acid transport and countertransport in nutrition and metabolism. Physiol. Rev. 1990;70:43-77
7. Ferruzza S., Ranaldi G., Di Girolamo M., Sambuy Y. The transport of lysine across monolayers of human cultured intestinal cells (Caco-2) depends on Na+-dependent and Na+-independent mechanisms on different plasma membrane domains. J. Nutr. 1995;125:2577-2585
8. Ganapathy V., Brandsch M., Leibach F. H. Intestinal transport of amino acids and peptides. Johnson L. R. eds. Physiology of the Gastrointestinal Tract 3rd ed. 1994:1773-1794 Raven Press New York, NY.
9. Halleux C., Schneider Y.-J. Iron absorption by intestinal epithelial cells: 1. Caco 2 cells cultivated in serum-free medium, on polyethyleneterephthalate microporous membranes, as an in vitro model. In Vitro Cell Dev. Biol. 1991;27A:293-302
10.
Harvey C. M., Muzyka W. R., Yao S.Y.M., Cheeseman C. I., Young J. D. Expression of rat intestinal L-lysine transport systems in isolated oocytes of Xenopus laevis. Am. J. Physiol. 1993;265:G99-G106
11. Hidalgo I. J., Borchardt R. T. Amino acid transport in a novel model system of the intestinal epithelium (Caco-2 cell). FASEB J 1988;2:2538
12. Hidalgo I. J., Borchardt R. T. Transport of large neutral amino acid (phenylalanine) in a human intestinal epithelial cell line: Caco-2. Biochim. Biophys. Acta 1990;1028:25-30[Medline]
13. Hu M., Borchardt R. T. Transport of a large neutral amino acid in a human intestinal epithelial cell line (Caco-2): Uptake and efflux of L-phenylalanine. Biochim. Biophys. Acta 1992;1135:233-244[Medline]
14. Lücke H., Haase W., Murer H. Amino acid transport in brush-border membrane vesicles isolated from human small intestine. Biochem. J. 1977;168:529-532[Medline]
15. Maillard M. E., Stevens B. R., Mann G. E. Amino acid transport by small intestinal, hepatic and pancreatic epithelia. Gastroenterology 1995;108:888-910[Medline]
16. Malo C. Multiple pathways of amino acid transport in brush border membrane vesicles isolated from the human fetal small intestine. Gastroenterology 1991;100:1644-1652[Medline]
17. Mandel L. J., Bacallao R., Zampighi G. Uncoupling of the molecular ‘fence’ and paracellular ‘gate’ functions in epithelial tight junctions. Nature (Lond.) 1993;361:552-555[Medline]
18.
Munck B. G., Munck L. K. Phenylalanine transport in rabbit small intestine. J. Physiol. 1994;480:99-107
19. Munck L. K., Munck B. G. Variation in amino acid transport along the rabbit small intestine: Mutual jejunal carriers of leucine and lysine. Biochim. Biophys. Acta 1992;1116:83-90[Medline]
20. Satoh O., Kudo H., Yamada K., Kawasaki T. Characterization of amino-acid transport systems in guinea-pig intestinal brush-border membrane. Biochim. Biophys. Acta 1989;985:120-126[Medline]
21. Schneider Y.-J. Optimisation of hybridoma cell growth and monoclonal antibody secretion in a chemically defined serum- and protein-free culture medium. J. Immunol. Methods 1989;116:65-77[Medline]
22. Stevens B. R., Kaunitz J. D., Wright E. M. Intestinal transport of amino acids and sugars: advances using membrane vesicles. Annu. Rev. Physiol. 1984;46:417-433[Medline]
23. Stevens B. R., Ross H. J., Wright E. M. Multiple transport pathways for neutral amino acids in rabbit jejunal brush border vesicles. J. Membr. Biol. 1982;66:213-225[Medline]
24. Thwaites D. T., Markovich D., Murer H., Simmons N. L. Na+-independent lysine transport in human intestinal Caco-2 cells. J. Membr. Biol. 1996;151:215-224[Medline]
25. Wolfram S., Giering H., Scharrer E. Na+-gradient dependence of basic amino acid transport into rat intestinal brush border membrane vesicles. Comp. Biochem. Physiol. 1984;78A:475-480
26. Zheng L., Chen J., Zhu Y., Yang H., Elmquist W., Hu M. Comparison of the transport characteristics of D- and L-methionine in a human intestinal epithelial model (Caco-2) and in a perfused rat intestinal model. Pharm. Res. 1994;11:1771-1776[Medline]
This article has been cited by other articles:
|
|
 |
|
  R. Martin-Venegas, M. J. Rodriguez-Lagunas, Y. Mercier, P.-A. Geraert, and R. Ferrer Effect of pH on L- and D-methionine uptake across the apical membrane of Caco-2 cells Am J Physiol Cell Physiol, March 1, 2009; 296(3): C632 - C638. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Broer Amino Acid Transport Across Mammalian Intestinal and Renal Epithelia Physiol Rev, January 1, 2008; 88(1): 249 - 286. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Uc, J. B. Stokes, and B. E. Britigan Heme transport exhibits polarity in Caco-2 cells: evidence for an active and membrane protein-mediated process Am J Physiol Gastrointest Liver Physiol, December 1, 2004; 287(6): G1150 - G1157. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
