Mechanisms and kinetics of citrulline uptake in a model of human intestinal epithelial cells

Mechanisms and kinetics of citrulline uptake in a model of human intestinal epithelial cells
of 9
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  ORIGINAL ARTICLE Mechanisms and kinetics of citrulline uptake ina model of human intestinal epithelial cells Senda Bahri a , Emmanuel Curis b , Fatima-Zahra El Wafi a , Christian Aussel c ,Jean-Claude Chaumeil a , Luc Cynober c,d , Naı¨ma Zerrouk a, * a Laboratoire de Pharmacie Gale´nique, France b Laboratoire de Biomathe´matiques, France c Laboratoire de Biologie de la Nutrition, EA-2498, Faculte´ des Sciences Pharmaceutiques et Biologiques, Universite´ ParisDescartes, 4, Avenue de l’Observatoire, 75270 Paris Cedex 06, France d Service de Biochimie, Hoˆtel Dieu, AP-HP, Paris, France Received 5 March 2008; accepted 1 August 2008 KEYWORDS Citrulline;Caco-2 cells;Amino acid transporter;Transport kinetic Summary Background & aims:  Citrulline is a major precursor of arginine by  de novo  synthesis in thekidneys. Oral citrulline supplementation may be beneficial in some clinical conditions.However, citrulline bioavailability depends on its intestinal absorption. Since the mechanismof citrulline transport across the intestine has not been established yet, this study was de-signed to characterize  L -[ 14 C]-citrulline uptake by Caco-2 cells. Methods:  Caco-2 cells were cultured in a bicameral insert system. Inhibition studies were con-ducted in the presence of neutral, cationic, acidic and non-metabolized amino acids. We per-formed control inhibition studies for arginine uptake. Results:  Citrulline uptake was pH-independent whereas the uptake rate was reduced in theabsence of Na þ . Kinetic analysis indicated the involvement of Na þ -dependent and Na þ -inde-pendent saturable transport components. For competition studies, both the transport compo-nents were markedly inhibited by large, small neutral and cationic amino acids. It was alsonoticed that specific inhibitor of system  L BCH inhibited uptake. The inhibition profile of argi-nine transport was different from that of citrulline transport as arginine uptake was insensitiveto BCH. * Corresponding author. Tel.:  þ 33 153739586; fax:  þ 33 153739771. E-mail address:  (N. Zerrouk).0261-5614/$ - see front matter   ª  2008 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.doi:10.1016/j.clnu.2008.08.003 available at www.sciencedirect.com Clinical Nutrition (2008)  27 , 872 e 880  Conclusions:  These characteristics suggest that system B 0, þ might be responsible for the Na þ -dependent uptake of citrulline, whereas Na þ -independent uptake may include systems L andb 0, þ . Our results show that systems involved in citrulline transport are partly different fromthose involved in arginine transport. ª  2008 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rightsreserved. Introduction L -citrulline, a non-protein amino acid, was first identifiedfrom the juice of watermelon,  Citrullus vulgaris  Schrad. 1 Watermelon is particularly rich in free citrulline. This aminoacid was long considered as a mere metabolic interme-diate, especially in the urea cycle, and therefore attractedrelatively little interest. 2 However,  L -citrulline has emergedas an important amino acid both as a pr oduct of the NOcycle and as a precursor for arginine. 3 In vivo  studiessuggest that citrulline is synthesized and released from theintestine as a product of glutamine and arginine metabo-lism. 4 Unlike arginine, which is taken up by the liver andextensively metabolized into urea, citrulline passes freelythrough the liver. 5 Even though recent data suggest thatcitrulline can be taken up by the liver in some situations 6 the message remains unclear. 7 Thus circulating citrullineappears to be a masked form of arginine to avoid liver uptake and inadequate ureagenesis in certain situations. 2 Uptake and metabolism of citrulline may contribute to themaintenance of plasma arginine levels, which appears to beimportant for sustained  in vivo  NO production. Recentdata 8 also suggest that citrulline could have regulatoryaction on protein synthesis. Thus, it could be an importantcomponent in the regulation of protein metabolism.Citrulline may be useful for patients with impaired argininemetabolism and oral supplementation of citrulline can beused in a variety of clinical conditions. 3 Since citrullinebioavailability from food is dependent on its intestinalabsorption, it seems essential to clarify the mechanism ofcitrulline transport across the intestinal membrane.Currently, transport of  L -citrulline in the intestine is poorlyunderstood, with only one report 9 describing the involve-ment of a Na þ -dependent system in citrulline transportacross everted sacs of the rat small intestine. More gener-ally, only little information is available on the mechanismmediating cellular uptake of citrulline. Distinct pathwaysfor citr ulline transport have been descr ibed in macro- phages, 10 in rat aortic smooth muscle cells, 11 in neural cellcultures 12 and in bovine aortic endothelial cells. 13 All theseresults are difficult to extrapolate to intestinal cells.Therefore, characterizing citrulline transport acrossintestinal cells is necessary. Several models are availablefor that purpose; among them, the Caco-2 cells have beenextensively used as a relevant physiological model for uptake studies of the small intestinal epithelium. 14 Thismodel is known to express several transport systems thatare present in the intestinal epithelium such as glucose,amino acids and dipeptides transport systems. 15 The present work was designed to characterize citrullineuptake by Caco-2 cells. Because of the strong structural andmetabolic relationship between citrulline and arginine, wehavealsoinvestigatedtheuptakeofargininebyCaco-2cells,to detect whether separate pathways mediate the entry ofcitrulline and arginine into these cells. Since arginine trans-port across Caco-2 cells was already reported in the litera-ture,arginineuptake experiments werealso awaytoensurethat our experimental conditions were correct. Materials and methods Materials L -[ 14 C]-citrulline (specific activity 56.3 mCi/mmol) and  L -[ 14 C]-arginine (specific activity 360 mCi/mmol) werepurchased from Perkin Elmer (Courtabœuf, France). Allculture media and antibiotics were obtained from Invi-trogen Life Technologies (Cergy Pontoise, France). Allamino acids ( L -isomers) were obtained from Sigma e Aldrich(Saint-Quentin, France).Caco-2 cells at passage 8 were kindly provided by Dr.Zweibaum and Dr. Rousset (INSERM U710, Villejuif, France).The Caco-2 cell line derives from a human colon adeno-carcinoma. In culture, Caco-2 cells spontaneously undergoenterocytic differentiation and exhibit several morpholog-ical and functional characteristics of mature enterocytes. 15 Cell culture Cells(0.9  10 6 )weregrownon75 cm 2 plasticflasks(CorningCostar,France)andthemediumwaschanged everydaywithDulbecco’s modified Eagle’s medium (DMEM) supplementedwith 10% fetal calf serum and 1% non-essential amino acids(‘‘complete medium’’). Caco-2 cells were incubated inacontrolledatmosphereat37   C,with95%relativehumidityand 10% CO 2  and reached confluence after 7 days. Cells ofpassage 21 e 34 were used, since Caco-2 monolayer grown onpermeable filter supports exhibits adequate functionalcriteria at this stage of differentiation. 16 For uptake studies, cells were seeded on polycarbonatefilters (Costar Transwell  6-well plate inserts, pore diam-eter 0.4  m m, Costar, MA, USA) at a density of 1.5  10 6 cellsper well. The cells were allowed to grow and differentiateduring 21 days in complete medium supplemented withpenicillin (110 IU/mL) and streptomycin (110  m g/mL). Thecells were used for experiments 21 e 28 days post seeding. Measurement of the transepithelial electricalresistance (TEER) In order to check the monolayer confluence and integrity,the TEER values were measured using a Millicell-ERSepithelial voltohmmeter with planar electrodes (MilliporeCitrulline uptake in Caco-2 cells 873  Co., Bedford, MA, USA) and calculated according to theequation TEER Z R total    A , where  R total  ( U ) is the measuredresistance, taking into account the values for cells andmembranes, and  A  is the surface area (cm 2 ).Before each experiment, TEER values were measured ineach well; they were between 1000 and 1200  U  cm 2 . HighTEER and low variability indicate both acceptable congru-ence and integrity in our experimental conditions (litera-ture range: 150 e 1600  U  cm 2 ). 15 Uptake studies To measure the uptake of  L -[ 14 C]-citrulline from the apicalside of Caco-2 cells, the cells were first deprived of aminoacids by incubation for 15 min at 37   C in Krebs buffer (NaCl140 mmol/L, KCl 5.4 mmol/L, MgSO 4  1.2 mmol/L, KH 2 PO 4 0.3 mmol/L, CaCl 2  2.8 mmol/L, NaH 2 PO 4  0.3 mmol/L,glucose 10 mmol/L, Hepes 10 mmol/L) adjusted to pH 7.4with Tris base. The buffer was then aspirated and the cellswere rinsed three times with this buffer. Citrulline uptakewas initiated by adding 1 mL Krebs buffer containing 200  m Mcitrulline (10  m M  L -[ 14 C]-citrulline,  i.e . 0.56  m Ci/mL, and190  m M unlabeled  L -citrulline) in the apical compartment.After incubation at controlled temperature for theappropriate times (see Results), apical uptake was stoppedby removing the Krebs buffer and immediately washing thecells three times with ice-cold buffer. Cells were harvestedin 1 mL NaOH 0.1 N and removed by scraping after 1 h atroom temperature.A set of experiments was performed at 4   C to quantifypassive diffusion across the cellular membrane. All other experiments were performed at 37   C.Radioactivity was measured by liquid scintillationcounting with a beta-Wallac 1409 instrument (Perkin Elmer,USA) on sample aliquots.The linearity of citrulline uptake with time was checkedwith the usual linear regression analysis, with threemeasures for each time. Two series of experiments wereperformed and analyzed independently, and the obtainedslope and intercept compared with the F test for jointcomparison of linear regression parameters, to check thereproducibility of the results.For studies at 4   C, we tested in addition whether theslope was equal to zero. Protein assay The protein content of cell monolayers was determined byBradford’s method 17 using bovine serum albumin as a stan-dard. Results for uptake are expressed as nanomoles ofcitrulline or arginine per milligram of protein per minute(nmol/mg prot/min). pH-dependence The effect of pH on citrulline uptake was examined bychanging the pH of the solution used to measure thetransport rates (pH 6 e 8.5). Krebs buffer was maintained for the solutions at pH 7 and 7.4. Hepes in the Krebs buffer wassubstituted by MES (2- N  -morpholinoethanesulfonic acid) for solutions at pH 6 and 6.5 and by the Tris base for solutionsat pH 8 and 8.5. The pH of the medium was adjusted witha solution of HCl or NaOH. Na D -dependence and kinetic studies of citrullineuptake In order to characterize the apical uptake of citrulline, theconcentration-dependence of the uptake by Caco-2 cellswas examined in the presence and in the absence ofextracellular Na þ , using total citrulline concentrationsranging from 25  m M to 2.5 mM, with a constant concentra-tion of 10  m M labeled citrulline completed with unlabeledcitrulline, leading to solutions of constant radioactivity. Allexperiments were performed at 37   C, with 5 min incuba-tion time for which uptake was linear with time (seeResults, below).Na þ -independent citrulline uptake was assessed byomission of NaH 2 PO 4  and equimolar  substitution of NaCl bycholine chloride in the Krebs buffer. 18 The Na þ -dependentuptake of citrulline was then determined as the differencebetween the uptakes measured in the sodium and in thecholine uptake medium, respectively.Kinetic parameters of citrulline uptake ( K  m  and  V  max )were determined by non-linear regression using theMichaelis e Menten equation with a diffusion term: V  Z V  max  ½ S  K  m  þ½ S  þ D ½ S  ; where  V   is the initial uptake rate, [ S ] the initial citrullineconcentration,  V  max  the maximum uptake rate,  K  m  theMichaelis e Menten constant and  D  the diffusion coefficient,according to Fick’s law. This diffusion term was only takeninto account when necessary, that is when modeling theNa þ -independent uptake, since in the Na þ -dependentmodel it is cancelled out in the subtraction process (totaluptake minus Na þ -independent uptake).In both cases, the predominant experimental uncer-tainty arises from the radioactive counting process, which(for the observed count rates) is well described by a normaldistribution. Consequently, we used the following expres-sion to fit the model to the experimental data: x 2 Z X ni Z 1  V  i  V  th ð½ S  i ; K  m ; V  max ; D Þ s i =  ffiffiffiffi m p   20 ; where  n  is the number of different citrulline concentra-tions,  m  the number of repetitions,  V  i  the initial velocityobserved for the  i th citrulline concentration, [ S ] i ,  V  th  thepredicted velocity and  s i  the experimental standard devi-ation on  V  i . This expression, which is a weighted leastsquare, corresponds to the maximum likelihood method inthat case and is known to give the most accur ate values for the three fitted parameters,  D ,  K  m  and  V  max . 19 Fitting was accomplished with the LASE software,developed by one of us, 2 based on a gradient-like non-linear fitting algorithm with analytical expression for thederivatives, allowing one to have the best accuracyperformances.The adequacy of the fitted model to the experimentaldata was checked with the chi-square criterion: if themodel is acceptable, then  x 2 should be approximately lower than  ð n  q Þ þ  ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2 ð n  q Þ p   , with  q  the number of fitted874 S. Bahri et al.  parameters and  n  the number of experimental points.However, since the model is not linear, this criterion is onlyto be used as a rule of thumb; in particular,  p -values are notrelevant here (and will therefore not be given).Uncertainties on the fitted parameters were determinedusing the Monte-Carlo method, 20 on a basis of 1000randomly generated ‘‘experiments’’. Results of thisanalysis were also used to construct an  ad-hoc  test for thenon-parametric, simultaneous comparison of the fittedparameters between the Na þ -independent and the Na þ -dependent models. Competition studies of citrulline uptake Competition experiments were performed to characterizethe substrate specificity and to identify the transportsystems. For these experiments, unlabeled amino acidswere added to the transport medium in 100-fold excess thatis 200  m M of citrulline in the presence of 20 mM of thestudied unlabeled amino acid competitor. Competition studies of arginine uptake Cross-inhibition studies of arginine uptake were performedin the same conditions as those created for competitionstudies of citrulline uptake. Arginine uptake (200  m M) wasstudied in the presence of a 100-fold excess of inhibitor amino acids. However, since the main purpose wascomparison with results in the literature, we only testeda subset of the competitors: BCH, arginine, leucine,glutamine, ornithine and citrulline. Statistical analysis Each experimental point was performed in triplicate;each experiment was repeated to ensure reproducibilityof the results. This leads to a total of six repetitions for each experimental point. There were no outliers and allexperimental points were considered in the analysis.Results are expressed as mean  SEM. Unless otherwisespecified, a result was considered to be significant if  p < 0.05.A global study of the srcin of the experimental uncer-tainties was performed by application of the error propa-gation theorem. This study showed that, in our conditions,besides the intrinsic variability between cell cultures, themain sources of uncertainties are, firstly, detection of theradioactive citrulline and, secondly, the protein assay. Bothmethods rely on the hypothesis that the signal-to-doserelation is linear; this was validated by linear regressionanalysis: unless the non-linear term was evidenced at the95% confidence level with the classical F test, linearity wasassumed.For the pH-dependence study, experiments wereanalyzed using a two-way hierarchical analysis of variancewith the ‘‘experiment’’ factor nested in the ‘‘pH’’ factor,the first one being fixed and the second one random.For the competition studies of citrulline transport,because of important discrepancies in the experimentalvariances between the different competitors, most statis-tical tools are not applicable, especially all hierarchical andclassification methods. That is why a simple Dunnett’s test(to be taken with caution because of these unequal vari-ances) and a Kruskal e Wallis analysis of variance wereperformed. Results Characteristics of citrulline apical uptake The time course of citrulline uptake by Caco-2 cells isshown in Fig. 1. At 37   C, citrulline uptake is not linear onthe total time interval of the study but is very accuratelydescribed by a parabolic relation. However, up to 12 min,the line is not statistically different from this parabola;a linear model can therefore be used to calculate theuptake kinetic values. For this reason, all subsequentexperiments were done with 5 min incubation time if theincubation time is less than 12 min.Citrulline uptake in Caco-2 cells was not affected bychanging pH in the selected interval in the incubatedmedium (data not shown). Removing Na þ from the mediumstrongly reduced citrulline uptake (Fig. 2A).The sodium-independent uptake and the sodium-dependent uptake (obtained by the difference of citrullineuptake between the Na þ -containing and the Na þ -freemedia) are both saturable, suggesting the involvement ofcarrier-mediated uptakes. Kinetic parameters for bothuptakes are given in Table 1. The affinity and capacity ofthe sodium-dependent and sodium-independent uptakesare different, as evidenced by different  K  m  and  V  max Figure 1  Time course of  L -citrulline uptake in Caco-2 cells.Cells were deprived of amino acids by incubation for 15 min at37   C in Krebs buffer. The buffer was then aspirated and thecells were rinsed three times with this buffer. Citrulline uptakewas initiated by adding 1 mL Krebs buffer containing 200  m Mcitrulline (10  m M  L -[ 14 C]-citrulline) in the apical compartment.Cells were incubated at 37   C ( ) and 4   C ( 6 ) in the presenceof  L -citrulline for 1, 2, 4, 6, 8, 12 and 15 min. Uptake at 4   C canbe considered as almost inexistent. Uptake at 37   C can beconsidered as linear up to 12 min (  p Z 0.46 for the non-linearity test), but not up to 15 min (  p Z 0.002). Citrulline uptake in Caco-2 cells 875  (  p < 0.001), the Na þ -independent uptake being lessefficient.Experiments performed at 4   C indicated that passivediffusion is negligible. This is confirmed by values obtainedfor   D  in complete fits which are not significantly differentfrom zero. Thus, it is legitimate to set  D Z 0 in fits. Setting D Z 0 also overcomes the problem of the strong correlationbetween  D  and  K  m  and allows a more precise estimation of K  m . However, a very strong correlation still exists between K  m  and  V  max . This explains the quite high individualuncertainties on these parameters despite the joint confi-dence interval, which is a very narrow, but long, ellipse.The Na þ -independent transport is very well fitted by thissimple Michaelis e Menten model, despite the lack ofa diffusion term (Fig. 2B, Table 1). The Na þ -dependentuptake is less well fitted, which may be due to the over-simplification of the Michaelis e Menten approach or to thepresence of one unidentified aberrance in the data. Modelsinvolving more than one transport system were considered,but statistical analysis shows that much more data would beneeded to correctly estimate  K  m  and  V  max  for each trans-port system. Another difficulty is that, if the other trans-port systems have properties too close to those ofcitrulline, it is impossible to distinguish among them. Competition studies Competition studies of citrulline In the presence of Na þ , neutral amino acids, cationic aminoacids and synthetic analogues showed a differential abilityto inhibit citrulline uptake into Caco-2 cells.Aspartate, an acidic amino acid, caused a marginalinhibition of citrulline uptake. The synthetic  N  -(methyl-amino)-isobutyric acid (MeAIB) only partially inhibitedcitrulline uptake. Moderate inhibition was observed byglycine, 2-aminobicyclo-(2,2,1)-heptane-2-carboxylic acid(BCH) and valine. In contrast, glutamine, alanine, leucineand arginine were potent inhibitors of citrulline uptake butthe most effective inhibitors were citrulline (self-inhibi-tion), phenylalanine, ornithine and lysine (Fig. 3A).In Na þ -free conditions, inhibition studies showeda significant decrease in citrulline uptake in the presence ofneutral amino acids (phenylalanine, leucine and ornithine)and cationic amino acids (arginine and lysine). BCH,a model substrate for system L, induced a significant inhi-bition of citrulline uptake (Fig. 3B).The inhibition profile of citrulline uptake in the presenceof Na þ (Fig. 3A) represents the inhibition of total citrullineuptake involving both Na þ -dependent and Na þ -indepen-dent amino acid transport systems. To highlight the role ofNa þ -dependent transport systems, we determined thedifference between the inhibition of total uptake and thatof the Na þ -independent uptake. Fig. 3C shows the inhibi-tion of only Na þ -dependent amino acid transportersinvolved in citrulline uptake. This figure shows that this or these systems have a strong affinity for neutral, cationicamino acids and BCH. Competition studies of arginine Neutral amino acids (leucine and glutamine) and cationicamino acids (arginine and ornithine) significantly inhibitedarginine uptake (Fig. 4). However, the inhibition profile ofarginine uptake was different from that of citrulline as BCHfailed to inhibit arginine uptake. Both citrulline and argininecaused a marked self-inhibition of uptake and both aminoacids have a strong inhibitory effect on each other (Fig. 5). BA Figure 2  (A)Concentration-dependenceof L -citrullineuptakeby Caco-2 cells. After amino acid depletion, Caco-2 cells wereincubated at 37   C for 5 min with varying concentrations of  L -citrulline in the presence ( ) and in the absence ( - ) of extra-cellular Na þ . Total citrulline concentrations ranged from 25  m Mto 2.5 mM, with a constant concentration of labeled citrulline(10  m M). Na þ -dependent uptake ( : ) was determined as beingthe difference between  L -citrulline uptakes in the presence andin the absence of extracellular Na þ . Data are expressed as themean  SEM,  n Z 6. (B) Eadie e Hofstee plots of the total uptake( ), Na þ -dependent uptake ( 6 ), and Na þ -independent uptake( , ) of  L -citrulline by Caco-2 cells. Table 1  Kinetic parameters: results of the fit of theMichaelis e Menten model to the experimental data for citrulline uptakeNa þ -dependentuptakeNa þ -independentuptake K  m  (mM) 0.46  0.07 0.67  0.10 V  max  (nmol/mgprot/min)3.58  0.39 2.16  0.23 r  ( K  m ,  V  max ) 0.94 0.96 c 2 expected(approx.)14.5 or lower 14.5 or lower  c 2 obtained 25.6 3.56 r  ( K  m ,  V  max ): correlation coefficient between  K  m  and  V  max ; c 2 : quality of fit criteria (see text for details). 876 S. Bahri et al.
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks