Alfred Berteloot

Common characteristics for Na+-dependent sugar transport in Caco-2 cells and human fetal colon

SummaryThe recent demonstration that the human colon adenocarcinoma cell line Caco-2 was susceptible to spontaneous enterocytic differentiation led us to consider the question as to whether Caco-2 cells would exhibit sodium-coupled transport of sugars. This problem was investigated using isotopic tracer flux measurements of the nonmetabolizable sugar analog α-methylglucoside (AMG). AMG accumulation in confluent monolayers was inhibited to the same extent by sodium replacement, 200 μm phlorizin, 1mm phloretin, and 25mm d-glucose, but was not inhibited further in the presence of both phlorizin and phloretin. Kinetic studies were compatible with the presence of both a simple diffusive process and a single, Na+-dependent, phlorizin-and phloretin-sensitive AMG transport system. These results also ruled out any interaction between AMG and a Na+-independent, phloretin-sensitive, facilitated diffusion pathway. The brush-border membrane localization of the Na+-dependent system was inferred from the observations that its functional differentiation was synchronous with the development of brush-border membrane enzyme activities and that phlorizin and phloretin addition 1 hr after initiating sugar transport produced immediate inhibition of AMG uptake as compared to ouabain. Finally, it was shown that brush-border membrane vesicles isolated from the human fetal colonic mucosa do possess a Na+-dependent transport pathway(s) ford-glucose which was inhibited by AMG and both phlorizin and phloretin. Caco-2 cells thus appear as a valuable cell culture model to study the mechanisms involved in the differentiation and regulation of intestinal transport functions.

Analysis of kinetic data in transport studies: New insights from kinetic studies of Na+-d-glucose cotransport in human intestinal brush-border membrane vesicles using a fast sampling, rapid filtration apparatus

SummaryUsing the fast sampling, rapid filtration apparatus (FSRFA) recently developed in our laboratory (Berteloot et al., 1991.J. Membrane Biol.122:111–125), we have studied the kinetic characteristics of Na+-d-glucose cotransport in brush-border membrane vesicles isolated from normal adult human jejunum. True initial rates of transport have been determined at both 20 and 35°C using a dynamic approach which involves linearregression analysis over nine time points equally spaced over 4.5 or 2.7 sec, respectively. When the tracer rate of transport was studied as a function of unlabeled substrate concentrations added to the incubation medium, a displacement curve was generated which can be analyzed by nonlinear regression using equations which take into account the competitive inhibition of tracer flux by unlabeled substrate. This approach was made imperative since at 20°C, in the presence of high substrate concentrations or 1mm phlorizin, no measurable diffusion was found and the resultant zero slope values cannot be expressed into a classicalv versus S plot. All together, our results support the existence of a single Na+-d-glucose cotransport system in these membranes for which Na+ is mandatory for uptake. This conclusion is at variance with that of a recent report using the same preparation (Harig et al., 1989.Am J. Physiol.256:8618–8623). Since the discrepancy seems difficult to resolve on the consideration of experimental conditions alone, we have determined the kinetic parameters ofd-glucose transport using one time point measurements and linear transformations of the Michaelis-Menten equation, in order to investigate the potential problems of such a widely used procedure. Comparing these approaches, we conclude that: (i) the dynamic uptake measurements give a better understanding of the different uptake components involved; (ii) it does not matter whether a dynamic or a one time point approach is chosen to generate the uptake data provided that a nonlinear-regression analysis with proper weighting of the data points is performed; (iii) analytical procedures which rely on linearization of Michaelian process(es) are endowed with a number of difficulties which make them unsuitable to resolve multicomponent systems in transport studies. A more general procedure which uses a nonlinear-regression analysis and a displacement curve is proposed since we demonstrate that it is far superior in terms of rapidity, data interpretation, and visual information.

Caco-2 Cell Line Used as an In Vitro Model to Study Cadmium Accumulation in Intestinal Epithelial Cells

Abstract.109Cd uptake was studied using the highly differentiated TC7 clone of Caco-2 cells as a model of human enterocyte function. Intracellular accumulation of 0.3 μm109Cd involved a rapid and a slow uptake phase, which resulted in complete equilibration (t½= 17.3 ± 1.3 min) with an apparent in-to-out distribution ratio (αe) of 11.6 ± 0.8. The amplitude of the rapid phase (U0) and the rate of the slow phase (V) were similarly reduced in the less differentiated PF11 clone, but comparable αe values were observed at equilibrium. In both clones, the t½ and αe values increased and decreased, respectively, upon addition of unlabeled Cd to the uptake media. In TC7 cells, 109Cd uptake at 1 min (U1) was unaffected by Ca concentrations four order of magnitude in excess, but both U0 and V demonstrated similar sensitivities to unlabeled Cd, Zn and sulfhydryl-reactive agents. Only U0 disappeared when EDTA was present in the wash solutions. U1 showed saturation kinetics and the data were found compatible with a model assuming rapid initial Cd binding and transport through a unique transport protein (Km= 3.8 ± 0.7 μm). Cd efflux kinetics demonstrated partial reversibility in EDTA-containing solutions, suggesting that the taken up Cd might be both tightly and loosely bound to intracellular binding sites. However, the displacement of 109Cd measured at 65 min failed to reveal this heterogeneity: the data were found compatible with a model equation assuming the presence of one class of high-capacity high-affinity binding sites. We conclude that a slow-transport fast-intracellular binding mechanism of Cd uptake best accounts for these results and that Cd transport most likely involves a carrier-type of protein unrelated to Ca absorption.

Electrogenic amino acid exchange via the rBAT transporter

A cDNA clone was isolated from rabbit renal cortex using DNA‐mediated expression cloning, which caused alanine‐dependent outward currents when expressed in Xenopus oocytes. The cDNA encodes rBAT, a Na‐independent amino acid transporter previously cloned elsewhere. Exposure of cDNA‐injected oocytes to neutral amino acids led to voltage‐dependent outward currents, but inward currents were seen upon exposure to basic amino acids. Assuming one charge/alanine, the outward current represented 38% of the rate of uptake of radiolabelled alanine, and was significantly reduced by prolonged preincubation of oocytes in 5 mM alanine. The currents were shown to be due to countertransport of basic amino acids for external amino acids using the cut‐open oocyte system. This transport represents a major mode of action of this protein, and may help in defining a physiological role for rBAT in the apical membrane of renal and intestinal cells.

Characteristics of glutamic acid transport by rabbit intestinal brush-border membrane vesicles. Effects of Na+-, K+- and H+-gradients.

In the presence of a Na+-gradient (out greater than in), L-glutamic acid and L-and D-aspartic acids were equally well concentrated inside the vesicles, while no transport above simple diffusion levels was seen by replacement of Na+ by K+. Equilibrium uptake values were found inversely proportional to the medium osmolarity, thus demonstrating uptake into an osmotically sensitive intravesicular space. The extrapolation of these lines to infinite medium osmolarity (zero space) showed only a small binding component in acidic amino-acid transport. When the same experiment was performed at saturating substrate concentrations, linear relationships extrapolating through the origin but showing smaller slope values were recorded, thus indicating that the binding component could be more important than suspected above. However, binding to the membrane was neglected in our studies as it was absent from initial rate measurements. Na+-dependent uphill transport of L-glutamic acid was stimulated by K+ present on the intravesicular side only but maximal stimulation was recorded under conditions of an outward K+-gradient (in greater than out). Quantitative and qualitative differences in the K+ effect were noted between pH 6.0 and 8.0. Initial uptake rates showed pH dependency in Na+-(out greater than in) + K+-(in greater than out) gradient conditions only with a physiological pH optimum between 7.0 and 7.5. It was also found that a pH-gradient (acidic outside) could stimulate both the Na+-gradient and the Na+ + K+-gradient-dependent transport of L-glutamic acid. However, pH- or K+-gradient alone were ineffective in stimulating uptake above simple diffusion level. Finally, it was found that increased rates of efflux were always observed with an acidic pH outside, whatever the conditions inside the vesicles. From these results, we propose a channel-type mechanism of L-glutamic acid transport in which Na+ and K+ effects are modulated by the surrounding pH. The model proposes a carrier with high or low affinity for Na+ in the protonated or unprotonated forms, respectively. We also propose that K+ binding occurs only to the unprotonated carrier and allows its fast recycling as compared to the free form of the carrier. Such a model would be maximally active and effective in the intestine in the in vivo physiological situations.

Proximo-distal gradient of Na+-dependent D-glucose transport activity in the brush border membrane vesicles from the human fetal small intestine.

Brush‐border membrane vesicles were isolated from the jejunum and ileum of 17–20‐week‐old normal human fetuses and found to be highly enriched in sucrase activity with less than 5% contamination by basolateral membranes. Time course studies of D‐glucose uptake clearly showed a transient, phlorizin‐sensitive, and Na+‐dependent accumulation of sugar into these vesicles. Higher maximum overshoot values and initial rates of D‐glucose uptake were recorded in jejunal as compared to ileal vesicles while low substrate binding to the membranes, identical intravesicular volumes and equivalent dissipation of the Na+‐gradient were found in the two preparations. It was concluded that a fully functional Na+‐D‐glucose cotransport system is present with a proximo‐distal gradient of activity during the early gestation period.

Glucose Transport and Glucose 6-Phosphate Hydrolysis in Intact Rat Liver Microsomes

Glucose transport was investigated in rat liver microsomes in relation to glucose 6-phosphatase (Glu-6-Pase) activity using a fast sampling, rapid filtration apparatus. 1) The rapid phase in tracer uptake and the burst phase in glucose 6-phosphate (Glu-6-P) hydrolysis appear synchronous, while the slow phase of glucose accumulation occurs during the steady-state phase of glucose production. 2) [14C]Glucose efflux from preloaded microsomes can be observed upon addition of either cold Glu-6-P or Glu-6-Pase inhibitors, but not cold glucose. 3) Similar steady-state levels of intramicrosomal glucose are observed under symmetrical conditions of Glu-6-P or vanadate concentrations during influx and efflux experiments, and those levels are directly proportional to Glu-6-Pase activity. 4) The rates of both glucose influx and efflux are characterized by t values that are independent of Glu-6-P concentrations. 5) Glucose efflux in the presence of saturating concentrations of vanadate was not blocked by 1 mM phloretin, and the initial rates of efflux appear directly proportional to intravesicular glucose concentrations. 6) It is concluded that glucose influx into microsomes is tightly linked to Glu-6-Pase activity, while glucose efflux may occur independent of hydrolysis, so that microsomal glucose transport appears unidirectional even though it can be accounted for by diffusion only over the accessible range of sugar concentrations.

Reduction of an Eight-State Mechanism of Cotransport to a Six-State Model Using a New Computer Program

A computer program was developed to allow easy derivation of steady-state velocity and binding equations for multireactant mechanisms including or without rapid equilibrium segments. Its usefulness is illustrated by deriving the rate equation of the most general sequential iso ordered ter ter mechanism of cotransport in which two Na+ ions bind first to the carrier and mirror symmetry is assumed. It is demonstrated that this mechanism cannot be easily reduced to a previously proposed six-state model of Na+-D-glucose cotransport, which also includes a number of implicit assumptions. In fact, the latter model may only be valid over a restricted range of Na+ concentrations or when assuming very strong positive cooperativity for Na+ binding to the glucose symporter within a rapid equilibrium segment. We thus propose an equivalent eight-state model in which the concept of positive cooperativity is best explained within the framework of a polymeric structure of the transport protein involving a minimum number of two transport-competent and identical subunits. This model also includes an obligatory slow isomerization step between the Na+ and glucose-binding sequences, the nature of which might reflect the presence of functionally asymmetrical subunits.

Two-step mechanism of phlorizin binding to the SGLT1 protein in the kidney.

Abstract. The relationships between phlorizin binding and Na+-glucose cotransport were addressed in rabbit renal brush-border membrane vesicles. At pH 6.0 and 8.6, high affinity phlorizin binding followed single exponential kinetics. With regard to phlorizin concentrations, the binding data conformed to simple Scatchard kinetics with lower apparent affinities of onset binding (Kdi= 12–30 μm) compared to steady-state binding (Kde= 2–5 μm), and the first-order rate constants demonstrated a Michaelis-Menten type of dependence with Km values identical to Kdi. Phlorizin dissociation from its receptor sites also followed single exponential kinetics with time constants insensitive to saturating concentrations of unlabeled phlorizin or d-glucose, but directly proportional to Na+ concentrations. These results prove compatible with homogeneous binding to SGLT1 whereby fast Na+ and phlorizin addition on the protein is followed by a slow conformation change preceding further Na+ attachment, thus occluding part of the phlorizin-bound receptor complexes. This two-step mechanism of inhibitor binding invalidates the recruitment concept as a possible explanation of the fast-acting slow-binding paradigm of phlorizin, which can otherwise be resolved as follows: the rapid formation of an initial collision complex explains the fast-acting behavior of phlorizin with regard to its inhibition of glucose transport; however, because this complex also rapidly dissociates in a rapid filtration assay, the slow kinetics of phlorizin binding are only apparent and reflect its slow isomerization into more stable forms.

Evidence for a Membrane Exchangeable Glucose Pool in the Functioning of Rat Liver Glucose-6-phosphatase

We have investigated the kinetics of tracer uptake into rat liver microsomes in relation to [14C]glucose 6-phosphate (Glu-6-P) hydrolysis by glucose 6-phosphatase (Glu-6-Pase). 1) The steady-state levels of intravesicular tracer accumulated during the rapid (AMP1) and slow (AMP2) phases of uptake both demonstrate Michaelis-Menten kinetics relative to outside Glu-6-P concentrations with K values similar to those observed for the initial burst (V) and steady-state (V) rates of Glu-6-P hydrolysis. 2) The AMP1/AMP2 ratio is constant (mean value = 0.105 ± 0.018) over the whole range of outside Glu-6-P concentrations and is equal to the AMP/AMP ratio (0.109 ± 0.032). 3) Linear relationships are observed between the initial rates of glucose transport during the slow uptake phase (V2) and [AMP1], and between [V] and [AMP2]. 4) The value of V exceeds by more than 10-fold that of V. 5) It is concluded that the substrate transport model is incompatible with those results and that AMP1 represents a membrane exchangeable glucose pool. 6) We propose a new version of the conformational model in which the catalytic site lies deep within a hydrophilic pocket of an intrinsic membrane protein and communicates with the extra- and intravesicular spaces through channels with different glucose permeabilities.