Antiluminal Membrane Research Articles

A novel method to determine the localization of high and low-affinity GABA transporters to the luminal and antiluminal membranes of brain capillary endothelial cells

Some evidence has shown that the luminal plasma membrane and the antiluminal plasma membrane of the blood–brain barrier (BBB) are functionally distinct. This polarity permits brain capillary endothelial cells (BCECs) to actively regulate the internal milieu of the brain. Especially, polarity also exists concerning the transport of some endogenous substances such as amino acids and glucose. It is important to determine the luminal and antiluminal membrane localization of some transporters of endogenous substances, which will contribute to a better understanding of the transport mechanisms of the endogenous substances at the BBB. The present paper describes a novel procedure for combining isolated brain capillaries with primary cultured BCECs to determine the subcellular localization of some transporters, which is relatively simpler than the methods described previously. This method was used for the first time to determine successfully the luminal and antiluminal distribution of high and low-affinity GABA transporters. The low-affinity GABA transporter is probably localized to the luminal membrane of BCECs, and the high-affinity GABA transporter may be localized to the antiluminal membrane. Themes: Transporters Topics: GABA

Sodium and chloride-dependent high and low-affinity uptakes of GABA by brain capillary endothelial cells

The mechanisms of carrier-mediated transport of γ-aminobutyric acid (GABA) at the blood–brain barrier (BBB) were examined by investigating [ 3 H ]GABA uptake by isolated bovine brain capillaries, monolayers of primary cultured brain capillary endothelial cells (BCECs) attached to plates or suspended BCECs. The uptake of [ 3 H ]GABA was concentration-dependent and saturable. Nonlinear regression analysis of the original data indicated the existence of two distinct high and low-affinity GABA transporters on isolated brain capillaries or suspended BCECs, with K m1, K m2, V m1 and V m2 equal to 25.3 μM, 485.2 μM, 3.6 and 8.4 nmol/5 min/mg protein, respectively, for the capillaries, and 21.3 μM, 322.0 μM, 6.1 and 15.7 nmol/5 min/mg protein, respectively, for the suspended BCECs. In contrast, a single low-affinity transporter was found for monolayers of BCECs attached to plates with K m and V m equal to 338.7 μM and 18.8 nmol/5 min/mg protein, respectively. Subcellular location of the two distinct transporters on BCECs is discussed, suggesting that the low-affinity GABA transporter is probably localized to the luminal membrane of BCECs, and the high-affinity GABA transporter is probably localized to the antiluminal membrane. Low temperature (4°C) and metabolic inhibitors markedly diminished both high and low-affinity uptakes of [ 3 H ]GABA by isolated brain capillaries. The substitution of Na + with choline +, K + or Li + with the counter anion Cl − almost completely abolished both uptakes. Substitution of Cl − with Br −, I −, F − or NO 3 − in the presence of Na + significantly reduced both uptakes to different extents. Alanine, leucine, phenylalanine, arginine, glutamate and pyruvate had no obvious effect on either uptake. Probenecid, amino-oxyacetic acid, β-alanine, taurine, betaine, and nipecotic acid significantly reduced both uptakes. These data suggested that both the GABA transporters at the BBB were temperature, metabolic energy, Na + and Cl −-dependent, and may be specific and different from the known monocarboxylic acid, GABA and other amino acid transporters, which may play a role in the disposition of GABA in the brain.

P-glycoprotein is strongly expressed in the luminal membranes of the endothelium of blood vessels in the brain.

Luminal membranes of the vascular endothelium were isolated from brain, heart and lungs by modification of their density. The presence of P-glycoprotein (P-gp) was detected by Western blotting in luminal membranes from the endothelium of the three tissues. Strong enrichment in brain capillary luminal membranes, compared with brain capillaries (17-fold) and whole membranes (400-500-fold), indicates that P-gp is mainly located on the luminal side of the brain endothelium. Western blotting was also performed with antibodies directed against GLUT1, glial fibrillary acidic protein, adaptin, IP3R-3, integrins alphav and collagen IV as controls to determine whether the preparations were contaminated by other membranes. Strong enrichment of GLUT1 in brain capillary luminal membranes (9.9-fold) showed that the preparation consisted mainly of endothelial cell plasma membranes. Poor enrichment of glial fibrillary acidic protein (1.4-fold) and adaptin (2.4-fold) and a decreased level of IP3R-3, integrins alphav and collagen IV excludes the possibility of major contamination by astrocytes or internal and anti-luminal membranes. High levels of P-gp in the luminal membranes of brain capillary endothelial cells suggests that it may play an important role in limiting the access of anti-cancer drugs to the brain.

Brain-to-blood active transport of β-alanine across the blood–brain barrier

A high-affinity antiluminal uptake system for β-alanine was demonstrated in primary cultured bovine brain capillary endothelial cells (BCEC) for which K t is 66.9 μM. β-alanine uptake was energy-, sodium- and chloride ion-dependent. β-amino acids strongly inhibited the uptake, while α- and γ-amino acids had a little or no inhibitory effect. In ATP-depleted cells, the uptake was stimulated by preloading β-alanine or taurine but not by l-leucine. These results suggest that β-alanine is actively transported across the antiluminal membrane of BCECs that is common to β-amino acids. The system may function for the efflux from the brain to blood.

Kidney targeting of glycosylated peptides

Arg-vasopressin (AVP) derivatives modified at the N-terminal amine with various sugars via an octamethylene were synthesized and tissue uptake was evaluated after i.v. injection in rats. Glucosyl, mannosyl and 2-deoxyglucosyl derivatives showed significantly higher tissue uptake clearance by the kidney than galactosyl and α-mannosyl derivatives. In vivo tissue uptake analysis of the glucosyl derivatives showed that only the kidneys had a saturable uptake mechanism via the antiluminal membrane. The derivatives which showed high renal uptake clearance showed specific binding to the kidney membrane in vitro. These results suggested that there was a novel sugar transport mechanism in the kidneys which could be utilized as a delivery method for glycosylated peptides.

Expression of epidermal growth factor and its receptor in normal and diseased human kidney: An immunohistochemical and in situ hybridization study

The kidney is one of the major sites of EGF production and there it seems to play several biological functions, such as modulation of cell growth, renal repair following injury, regulation of cellular metabolism and glomerular haemodinamics. The present study was first aimed at localizing EGF and its receptor (R) in normal human kidney by immunohistochemical and in situ hybridization techniques. Then, the distribution of the growth factor and its R was explored in biopsy specimens from eight patients with acute tubulointerstitial damage. In the normal human kidney, both EGF immunoreactivity and EGF mRNA were localized in tubular profiles corresponding to Henle's loop and, although to a lesser intensity, to distal convoluted tubule. EGF immunostaining was remarkable mainly at the apical surface of tubular cells. EGF-R protein expression was detected in glomerular endothelial cells, in peritubular capillaries and arteriolar walls, as well as along the thick ascending limb of Henle's lop and distal convoluted tubule, where it colocalized with Tamm-Horsfall protein. Immunohistochemical analysis of tubular profiles revealed that EGF-R was located especially along the basolateral membrane of tubular cells and within the basal part of cytoplasm. Endogenous alkaline phosphatase and CHIP28 positive tubules did not show any signal for EGF and its receptor. Kidneys with acute tubulointerstitial injury exhibited a dramatic decrease of EGF expression, whereas EGF-R showed only minor modifications. Interestingly, EGF-R was localized to both apical and antiluminal membranes of positive tubular cells. It is concluded that EGF-EGF receptor loop may be relevant in the pathogenesis of acute tubulointerstitial damage and recovery from tubular injury, while its role in the physiological renewal of the urothelium remains speculative.

Na +- and Cl −-Dependent transport of taurine at the blood-brain barrier

The characteristics of carrier-mediated transport of taurine at the blood-brain barrier (BBB) were studied by using primary cultured bovine brain capillary endothelial cells (BCECs), in situ brain perfusion and brain capillary depletion methods in rats. The uptake of [ 3H]taurine by cultured cells showed that the active transporter functions on both the luminal and antiluminal membranes of BCECs. The kinetic parameters for the saturable transport of taurine were estimated to be: for the luminal uptake, the Michaelis constant, K v , was 12.1 ± 0.5 μM, and the maximum uptake rate, J max, was 4.32 ± 0.05 nmol/30 min/mg protein; for the antiluminal uptake, K t was 13.6 ± 2.4 μM and J max was 2.81 ± 0.22 nmol/30 min/mg protein. The luminal and antiluminal uptakes of [ 3H]taurine were each dependent on both Na + and Cl −. Stoichiometric analyses suggest that two Na + and one Cl − are associated with the luminal uptake of one taurine molecule. β-Amino acids such as β-alanine and hypotaurine strongly inhibited the uptake of [ 3H]taurine, whereas α- and γ-amino acids had little or no effect. Furthermore, by in situ brain perfusion and in vivo brain capillary depletion methods, the carrier-mediated transport found by in vitro experiments was confirmed to function for the translocation of the taurine molecule from the vascular space into the brain. From these results, it was concluded that a Na + and Cl − gradient-dependent transport (uptake) system for taurine exists in both the luminal and the antiluminal membranes of BCECs.

Mechanism of valproic acid uptake by isolated rat brain microvessels.

In an effort to characterize putative transport systems of valproic acid (VPA) at the blood-brain barrier, the effects of various substrates and inhibitors of known anion transporters on the equilibrium vessel-to-medium concentration (vessel/medium) ratio of VPA were investigated using isolated rat brain microvessels. The equilibrium vessel/medium ratio of VPA was decreased by the presence of high millimolar concentration of unlabeled VPA, indicating that a saturable transport system was involved in VPA transport from medium to microvessels. Short-chain monocarboxylates such as propionic acid, pyruvic acid, and L-lactic acid did not alter the vessel/medium ratio, whereas medium-chain fatty acids and unsaturated metabolites of VPA significantly inhibited the net transport of VPA. Dicarboxylates, tricarboxylate, and p-aminohippuric acid did not affect VPA accumulation in the brain microvessels. Several anionic drugs including salicylic acid, penicillin G, cefazolin, and probenecid significantly reduced the vessel/medium ratio of VPA. In addition, disulfonate inhibitors of inorganic anion exchangers, SH-group modifying reagent, and metabolic inhibitor showed remarkable inhibitory effects on the net transport of VPA between brain microvessels and medium. These results suggest that VPA may be actively transported through the antiluminal membrane via a carrier-mediated system shared by other anionic drugs.

Effects of l-carnitine on the renal tubular transport of cephaloridine

It has been demonstrated recently that cephaloridine (Cld) inhibits the tubular reabsorption of filtered carnitine (Carn) in the rabbit kidney. This interaction has suggested that the limited net cell-to-luminal fluid movement of Cld following its secretory transport across the antiluminal membrane might result from a balance of active Cld reabsorption by a Carn carrier at the brush border approximately equal to its secretion into the tubular fluid, rather than the previously proposed luminal membrane block. Studies were done to determine the effects of l-Carn, 750 mg/kg, i.v. on the tubular secretion and cortical concentrations of Cld, infused i.v. at a dose of 28 mg/kg (Carn:Cld molar ratio = 70:1). The fractional excretions of Cld during three consecutive periods of 10 min each, one before and two following the bolus infusion of Carn, were (means ± SEM): 1.18 ± 0.14, 1.20 ± 0.14, and 1.16 ± 0.11, respectively ( N = 6 each; differences NS). Cortex-to-serum concentration ratios of Cld in control and Carn-treated rabbits were 10.43 ± 0.32 and 10.16 ± 0.86, respectively (NS). The data provide evidence against the reabsorptive transport of Cld by a Carn carrier, and do not support a model of balanced secretion and reabsorption as the cause of limited clearance despite significant secretory transport of Cld into the tubular cell.

Inhibitors of nonselective cation channels in cells of the blood-brain barrier.

In the antiluminal membrane of isolated capillaries of rat and porcine brain (blood-brain barrier) nonselective cation channels with g = 31 pS were observed in cell-excised membrane patches. The channel inactivated by decreasing cytosolic Ca2+ below 1 microM and was inhibited by 1 mM ATP on the intracellular side. Anions and divalent cations did not pass the channel, but Na+ and K+ were equally permeant. Like the nonselective cation channel of rat exocrine pancreatic cells, the channel in cerebral capillary endothelial cells was inhibited reversibly by derivatives of diphenylamine-2-carboxylate (DPC), like 3',5-dichlorodiphenylamine-2-carboxylic acid (DCDPC, ki = 1 microM), and flufenamic acid (ki = 4.9 microM). 4'-methyldiphenylamine-2-carboxylic acid (4-MDPC), 5-chloro-2(3-trifluormethylphenylamino)-3-nitrobenzoic acid, and 5-nitro-2-(3-phenylpropylamino)-2-carboxylic acid (NPPB), as well as the antiinflammatory drug ((Z)-5-chloro2,3-dihydro-3-(hydroxy-2-thienylmethylene)-2-ox o-1H-indole-1- carboxamide (Tenidap)) had a relatively low blocking potency (ki > 10 microM). Gadolinium (10 microM), a blocker of stretch-activated channels, inhibited the nonselective cation channel potently.

Mechanosensitive nonselective cation channels in the antiluminal membrane of cerebral capillaries (blood-brain barrier).

Single stretch-activated (SA) cation channels have been investigated in the antiluminal membrane of freshly isolated brain capillaries. SA-channels did not distinguish between K+ and Na+ ions and were also permeable to Ca2+ and Ba2+ ions. With monovalent cations in the patch pipette the single-channel conductance was 37 pS and with the divalent cations Ba2+ and Ca2+ slope conductance was 16 and 19 pS, respectively. The open probability of the SA-channel increased with increasing negative pressure as well as with depolarization. Cell swelling induced by hypotonic shock activated the SA-channels in cell-attached experiments. The contribution of SA-channels to the regulation of cerebrospinal fluid in brain edema is discussed.

Stretch-activated non-selective cation channels in the antiluminal membrane of porcine cerebral capillaries.

1. Single stretch-activated (SA) channels have been studied in isolated brain capillary endothelial cells as well as in the antiluminal membrane of intact porcine cerebral capillaries using the patch-clamp recording technique. 2. The SA channels were found to be cation selective and permeable to Na+, K+, Ba2+ and Ca2+. 3. With monovalent cations in the patch pipette, the channels showed inward rectification in cell-attached patches with a single-channel conductance of 37 pS at negative and 24 pS at positive clamp potentials. 4. With either 70 mM-Ca2+ or Ba2+ in the patch pipette, the current-voltage relation was linear with slope conductances of 16 and 19 pS, respectively. 5. Mean channel open probability increased with increasing pressure and with depolarizing clamp potentials. 6. Cell swelling induced by hypotonic shock activated the SA channels in cell-attached experiments. 7. The SA channel may be involved in cell volume or blood flow regulation. The contribution of these channels to the regulation of cerebrospinal salt and water content, especially in brain oedema, is discussed.

A calcium and ATP sensitive nonselective cation channel in the antiluminal membrane of rat cerebral capillary endothelial cells

The patch-clamp technique was applied to the antiluminal membrane of freshly isolated capillaries of rat brain (blood-brain barrier). With 1.3 mM Ca 2+ in the bath, excision of membrane patches evoked ion channels, which could not be observed in cell-attached mode. The channel was about equally permeable to Na + and K + ions, but not measurable permeable to Cl − and and the single-channels conductance was 31 + 2 pS ( n = 22). The channel open probability was not dependent on the applied potential. Lowering of Ca 2+ to 1 μM or below on the cytosolic side inactivated the channels, whereas addition of cytosolic ATP (1 mM) inhibited channel activity completely and reversibly. The channel was blocked by the inhibitor of nonselective cation channels in rat exocrine pancreas 3′,5-dichlorodiphenylamine-2-car☐ylic acid (DCDPC, 10 μM) and by the antiinflammatory drugs flufenamic acid (> 10 μM) and tenidap (100 μM), as well as by gadolinium (10 μM). Thus, these nonselective cation channels have many properties in common with similar channels observed in fluid secreting epithelia. The channel could be involved in the transport of K + ions from brain to blood side.

Increased distal nephron EGF content and altered distribution of peptide in compensatory renal hypertrophy.

Epidermal growth factor (EGF) precursor is synthesized within kidney in the thick ascending limbs of Henle's loop and in distal tubule. Under baseline conditions EGF precursor is localized to the luminal membrane. In contrast, functional EGF receptors are present in basolateral membranes of sensitive renal cells. Immunostainable EGF is increased in contralateral kidneys following uninephrectomy of rats. To confirm this increase and determine whether the distribution of EGF changes in this setting, we measured immunostainable EGF in kidneys originating from rats 1, 2, 5, or 14 days following unilateral nephrectomy or sham surgery. There was a suggestion of an increase in immunostainable EGF in distal tubules 5 days postnephrectomy and a definite increase 14 days postnephrectomy. At 1 or 2 days postnephrectomy, and following sham surgery, immunostainable EGF was present predominantly at luminal membranes. In contrast, immunostainable EGF was present more diffusely throughout distal tubular cells at 5 and 14 days postnephrectomy and clearly localized adjacent to both luminal and antiluminal membranes in kidneys obtained 14 days postnephrectomy. EGF extractable from kidneys was increased significantly 5 and 14 days postnephrectomy. This material is the size of mature EGF. The altered localization of immunostainable peptide indicates that a redistribution of intracellular EGF accompanies increased synthesis postnephrectomy. Antiluminal EGF precursor or mature EGF present within kidney could act as a paracrine growth factor.

The distribution of enzymes involved in purine metabolism in rat kidney

Adenosine produced from 5′-AMP has been proposed as a mediator of intrinsic renal regulation. The rates of 5′-AMP and adenosine metabolism are dependent on the activities of enzyme involved in purine metabolism. The activities of adenosine kinase (AK), adenosine deaminase (ADA), 5′-nucleotidase (5′-NT). AMP deaminase, xanthine, oxidase and purine nucleoside phosphorylase were measured in cytosolic and membrane fractions from glomeruli, cortical tubules, medulary thick ascending limb of Henle (MTAL) and collecting duct prepared from rat kidney by combinations of sieving and sucrose density gradient centrifugation techniques. In the cytoplasm of glomeruli cells, the activity ratios of ADA/AK and AMP deaminase/5′-NT were and 70 and 2.4,respectively. The highest activity of 5′-NT was found in membrane fractions of cortical tubules where it was equally distributed between luminal and antiluminal membranes. Membrane fractions of MTAL did not contain detectable amounts of adenosine deaminase activity. The highest activity of xanthine oxidase and purine nucleoside phosphorylase was in the cytoplasm fraction of glomeruli. These results suggest that deamination of AMP and adenosine may be favored in the cytoplasm of glomeruli cells. In contrast, in the extracellular space of glomeruli and especially in the cortical tubule, AMP can be converted preferentially to adenosine by 5′-NT.

Substrate-induced modulation of ATP turnover in dog and rabbit proximal tubules.

In dog proximal tubules in suspension, the addition of glucose increased significantly the ouabain-sensitive fraction of respiration, a response suppressed by phlorizin. The addition of alpha-methyl-D-glucoside (alpha-MG) had a modest effect and 3-O-methyl-D-glucoside (3-O-MG) had no effect. The different stimulation of the Na+,K(+)-ATPase activity elicited for each hexose could be explained by a different increment of net transepithelial flux of sodium induced by the sodium: hexose cotransport. This flux is a direct function of the transport characteristics of both luminal and antiluminal membranes of proximal cells for these sugars: glucose is rapidly transported by both membranes (allowing a large transepithelial flux of glucose: sodium) while alpha-MG is poorly transported by the basolateral, and 3-O-MG by the luminal, membrane of the dog proximal tubule (allowing a small transepithelial flux of hexoses and sodium). However the overall tubular respiration of dog proximal tubules was not increased by glucose addition because the increment in the ouabain-sensitive fraction was accompanied by a reciprocal decrement in an ouabain-insensitive but oligomycin- or N',N' dicyclohexylcarbodiimide (DCCD)-sensitive (or in the bafilomycin-sensitive) component of respiration. This component reflects the activity of a large BBM-bound H(+)-ATPase found in this species. The intracellular pH of dog proximal tubules in suspension was measured using the proton-sensitive fluorescent probe 2',7'-bis-2-(carboxyethyl)-5, (and 6)-carboxyfluorescein. Glucose application significantly alkalinized the cells. In contrast, other substrates such as lactate or acetate simultaneously acidified the cells and increased the ouabain-insensitive phosphorylative respiration of dog tubules. These observations suggest that a modulation of the activities of both the sodium and most probably the proton pump is elicited by substrate availability in suspensions of proximal tubules.

Localization of epidermal growth factor (EGF) binding sites on antiluminal plasma membrane of rat kidney: autoradiographic study using nonfiltering perfused rat kidney.

We previously demonstrated that the specific binding of EGF to the antiluminal plasma membrane was a prerequisite step for the renal uptake of EGF. In the present study, the localization of 125I-EGF binding sites on the antiluminal plasma membrane was investigated by tissue sampling and X-ray autoradiography in the nonfiltering kidney. The binding of 125I-EGF was recognized over the whole kidney and was highest in the inner medulla followed by the cortex and outer medulla. The binding of 125I-EGF in the nonfiltering kidney was completely inhibited in the presence of 20 nM unlabeled EGF, suggesting specific binding of 125I-EGF to its receptor. Further, we used a histologic tissue staining method to confirm the location of the 125I-EGF binding sites. Binding of 125I-EGF was demonstrated on the proximal straight tubules (PST), cortical collecting ducts (CCD), inner medullary collecting ducts (IMCD), and thin limb of Henle in the inner medulla (IMTLH). We found that the binding of 125I-EGF was high in the IMTLH. In addition, we determined the grain density both on the cell surface membrane and in the intracellular space of the proximal straight tubules, where the grain density on the antiluminal plasma membrane was approximately 50% that in the intracellular space at 20 min after the start of 125I-EGF perfusion, suggesting the internalization of 125I-EGF from the antiluminal plasma membrane to the intracellular compartment. In conclusion, the binding sites of 125I-EGF, which were accessible from the antiluminal side, were broadly distributed over the whole kidney and were most dense around the IMTLH.

Renal tubular handling of p-aminohippurate and epidermal growth factor (EGF) in filtering and nonfiltering perfused rat kidneys.

We examined the integrity of renal tubular function in filtering and nonfiltering isolated perfused rat kidneys by using p-amino-3H-hippurate (3H-PAH) and the multiple indicator dilution method with 14C-creatinine as a reference. The influx clearance (PSu,1) of unbound 3H-PAH was 0.37 and 0.38 ml/sec in the filtering and nonfiltering kidneys, respectively. The efflux rate constants were comparable between filtering and nonfiltering kidneys, while the sequestration rate constant in the filtering kidney was approximately three times larger than that in the nonfiltering kidney. These data suggest that the nonfiltering kidney maintains 3H-PAH transporting ability through the antiluminal plasma membrane. The renal handling of epidermal growth factor (EGF) by filtering and nonfiltering kidneys was compared. The ratio of the total uptake of tracer 125I-EGF over 20 min in the nonfiltering kidney to that in the filtering kidney was 0.8. This ratio was reduced to 0.2 when the kidneys were perfused with tracer 125I-EGF plus 20 nM EGF. Furthermore, the total uptake of tracer 125I-EGF in the nonfiltering kidney was reduced 20-fold in the presence of 20 nM unlabeled EGF. These findings suggest that the tubular uptake of tracer 125I-EGF by filtering kidney takes place mainly via the antiluminal plasma membrane and that this uptake is a saturable process.

Localization of binding sites for epidermal growth factor (EGF) in rat kidney: evidence for the existence of low affinity EGF binding sites on the brush border membrane.

We investigated the renal distribution of 125I-EGF in the filtering perfused rat kidney using an acid washing technique. Trichloroacetic acid-precipitable 125I-EGF radioactivity was eluted from both the renal vein and the urinary cannulae, the former regarded as representing the antiluminal, and the latter the luminal, cell surface bound 125I-radioactivity. The addition of excess unlabeled EGF (20 nM) to the perfusate completely inhibited the binding of 125I-EGF to the antiluminal membrane but did not inhibit that of 125I-EGF to the luminal membrane. On the other hand, the order of relative density of 125I-EGF binding sites in the in vivo kidney determined by autoradiography was cortex > inner medulla > outer medulla. After the i.v. administration of excess unlabeled EGF together with 125I-EGF, the renal uptake of 125I-EGF was inhibited completely in the inner medulla, but only by 50% in the cortex and outer medulla, suggesting the presence of nonsaturable luminal uptake of EGF in the cortex and outer medulla. After i.v. administration of 125I-EGF, a change in position of silver grains from the luminal cell surface membrane to the intracellular space was observed in the proximal convoluted tubules. In conclusion, in addition to the previously identified uptake mechanisms of circulating EGF through high-affinity binding sites on the antiluminal cell surface membrane, the reabsorption mechanism of filtered EGF through low-affinity binding sites on the luminal cell surface membrane was demonstrated. In vivo autoradiography showed the gradual internalization of EGF from the luminal cell surface membrane to the intracellular space of the proximal convoluted tubule.

Moment analysis of drug disposition in kidney. VI: Assessment ofin vivo transmembrane transport ofp-aminohippurate in tubular epithelium

This paper describes a novel method to assess the antiluminal membrane (ALM) and luminal membrane (LM) transport in vivo across renal tubular epithelial cells. The method is based upon a noncompartmental moment analysis of the plasma concentration and urinary excretion rate curves following renal artery injection. Quantitative relationships are represented between the noncompartmental parameters (clearance, volume of distribution, and the mean transit time) and the first-order rate constants associated with transmembrane transport processes. The in vivo transepithelial transport of [14C]p-aminohippurate (PAH) was examined using the rat kidney in the absence or presence of various plasma concentrations of unlabeled PAH, cefazolin, and methotrexate. The tubular secretion intrinsic clearance was reduced with an increase in the plasma concentration of concurrent unlabeled organic anions. The distribution volume of PAH in the kidney decreased in association with a decrease in the amount of PAH secreted, whereas the mean transepithelial (artery-to-lumen) transit time (Tcell) remained constant. These findings indicate that ALM transport is a capacity-limited process determining the amount of tubular secretion, and that LM transport is linear over the concentration range examined and independent of the amount of secretion. The contribution of ALM and LM transport to transcellular transport was first clarified in vivo. The present method will be useful for analyzing the transmembrane transport processes in vivo for highly diffusible substances in the kidney.