Inhibitors of Choline Transport in Alveolar Type II Epithelial Cells Chandra Dodia, Aron B. Fisher, Avinash Chander, and Arnost Kleinzeller Institute for Environmental Medicine and Department of Physiology, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania
Isolated alveolar type II epithelial cells (granular pneumocytes) from rat lung accumulate free choline against a concentration gradient by an energy-dependent saturable transport process with apparent K, approximately 18 JLM. In order to evaluate the structural requirements for choline transport by these cells, the inhibition of the initial rate of cellular uptake of PH]choline (5 JLM) by its analogue was measured. There was no significant inhibition of substrate uptake by analogues lacking an amino group while the presence of a quaternary nitrogen was most effective. N, N'-dimethylethanolamine (apparent K, 7 JLM) and n-decylcholine (apparent K, 0.5 JLM) were potent competitive inhibitors of choline transport. Substitution of the hydroxyl group in choline greatly diminished the inhibitory effect; fluorocholine, thiocholine, betaine, and betaine aldehyde showed little or no inhibition. This requirement for a hydroxyl group raises the possibility of hydrogen bonding of choline with the transport protein. The choline transport system in granular pneumocytes appears to differ from that in synaptosomes by the lower affinity of the carrier for substrate and for hemicholinium-3 and from that in erythrocytes by the role of the hydroxyl in the substrate molecule. The availability of inhibitory analogues for choline transport will facilitate isolation and study of the granular pneumocyte choline transport protein.
Phosphatidylcholine, a major component of the secretory product (i.e., lung surfactant) of granular pneumocytes, is synthesized de novo with choline as an important substrate precursor (1). Plasma choline, derived from dietary sources, is actively taken up by granular pneumocytes by a process with a K; of approximately 18 JLM (2). The properties of this transport process appear to be similar to those in some other cells, e. g., intestinal mucosa (3-5), guinea pig placental tissue (6), and renal collecting duct cells (7), but differ in several respects from the well-studied active transport process in synaptosomes (8). First, the affinity of the transport protein for choline in synaptosomes is some 20-fold higher (8) than in granular pneumocytes (2). Second, Naand choline are co-transported in synaptosomes (8), whereas there appears to be little or no involvement of Na' in choline transport in granular pneumocytes (2). Finally, hemicholinium-3 is a powerful inhibitor of the synaptosomal transport process (8), whereas the affinity of hemicholinium for the transport site in granular pneumocytes was relatively low (50% inhibition at 20-fold greater concentration) (2). A third choline transport system, essentially behaving as an equilibrating (facilitated diffusion) transport, has been described in red blood cells (9). The present study aims at char-
(Received in original form July 15, 1991 and in revised form September 30, 1991) Address correspondence to: A. B. Fisher, M.D., Institute for Environmental Medicine, University of Pennsylvania, Philadelphia, PA 19104-6068. Abbreviation: Eagle's modified essential medium, MEM. Am. J. Respir. Cell Mol. BioI. Vol. 6. pp. 426-429, 1992
acterization of the choline transport system in granular pneumocytes by analyzing its transport specificity with the goal to develop analogue inhibitors for further study and possible isolation of the choline transporter.
Materials and Methods The methods employed in this study for cell isolation and measurement of choline uptake were described in detail previously (2) and are summarized here. Granular pneumocytes were isolated by elastase digestion of blood-free rat lungs and were purified by "panning" on IgG-coated bacteriologic petri dishes to remove contaminating macrophages followed by differential adherence on tissue culture plastic dishes or collagen-coated Nuclepore polycarbonate membranes (3-JLm pore size, 5-cm diameter) during 21 h of incubation at 37° C in Eagle's minimal essential medium (MEM). In a usual experiment, 10 to 12 filters, each containing about 230 JLg protein (equivalent to approximately 2 X 1()6 cells), were prepared from the lungs of two rats. Cells on the filters were > 85 % granular pneumocytes as determined by specific fluorescence with phosphine 3R (2). The cells were washed 3 times in choline-free MEM (pH 7.4) and then preincubated in this medium at 37° C for 1 h. Measurement of uptake was carried out by addition of PH]choline (5 JLM choline; sp act, 100 JLCil JLmol) to choline-free MEM and incubation at 37° C for 90 s, during which periodthe uptake was linear (2). After incubation, adhering choline was removed by washing the filters 3 times (total wash time approximately 5 s) in ice-cold saline and then placing the filters in 1 ml of 0.1 M NaOH. The extract
Dodia, Fisher, Chander et al.: Lung Epithelial Choline Transport
was used for the determination of cell protein by the Coomassie blue dye binding assay and for measurement of dpm corresponding to internalized PH]choline (2). The rate of choline uptake is expressed in pmol choline/mg cell protein x 90 s. Our previous studies (2) expressed choline uptake as a function of cell water volume. Cell protein was used as the standard in the present study in order to avoid possible variation due to effects of inhibitors on cell volume. The H20 space in normal granular pneumocytes at 24 h in culture is approximately 4.1 j.tl!mg protein (10). Experiments with analogues were always carried out with a control (no inhibitor), and results are expressed as a percentage of control uptake in order to account for the variability of choline uptake with different preparations of cells. Efflux of PH]choline from cells was measured as previously described (2). Cells on plastic dishes (60-mm diameter) were preloaded for 10 min with PH]choline (5 j.tM, 100 j.tCi/nmol). Subsequently, the cells were washed with successive 3-ml portions of medium at 0.5, 1, 2, 3, 4, 5, and 10 min, and each wash was assayed for radiolabel. Parallel incubations were carried out under control conditions and with the presence of inhibitor in the wash solutions. At the end of the washout, the residual PH]choline in the cells was determined and expressed as log percent of the label remaining in the cells as a function of time. All incubations were carried out in duplicate, and results were averaged to give a mean value; duplicate incubations in one experiment generally did not differ by more than 15%. Data are generally expressed as mean ± SEM. Mean percent uptake in the presence of inhibitors was evaluated for deviation from control (100% uptake) by t test (11). Because inhibitors were studied in 20-fold molar excess over choline, we confined our statistical analysis to those compounds in which mean inhibition exceeded 50 %. The levelof statistical significance was taken as P < 0.05.
Materials Fluorocholine ([2-fluoroethyl]trimethylammonium) halides (chloride, bromide, iodide) were purchased from Lee's Bioorganic Laboratories (Marcus Hook, PA); the compounds were characterized by H-NMR and 19F-NMR and showed a single peak by ion chromatography. Dimethyl, n-decyl(hydroxyethyl)ammonium bromide (n-decylcholine) was prepared and analyzed for purity as described by Deves and Krupka (9). All other reagents were commercial preparations of the highest available purity.
Results The mean rate of choline uptake in the absence of inhibitors was 330 ± 25 pmol/mg cell protein/90 s (n = 20). This is equivalent to an uptake of approximately 80 pmol!j.tl cell H20/90 s, similar to our previous measurements (2). The effect of various structural analogues of choline on the rate of (3H]choline uptake is given in Table 1, expressed as a percentage of the rate of choline uptake in control cells, i.e., no added inhibitor. The first inhibitors studied evaluated the role of the amino group for the transport of choline. 3,3'-Dimethylbutanol, an analogue lacking the amino N, did not affect choline uptake (Table 1), indicating that the presence of a nitrogen atom in
427
TABLE 1
PH]Choline
uptake by granular pneumocytes: inhibition by structural analogues*
Choline Analogue
3,3' -Dimethyl-a-butanol Ethanolamine N-Methy lethanolamine N,N' -Dimethylethanolamine n-Decylcholine (dimethyl,n-decyl[2-hydroxyethyl]ammonium bromide) Trimethylamine Tetraethylammonium chloride Betaine ([carboxymethyl]trimethylammonium chloride) Betaine aldehyde ([formylmethyl)trimethylammonium chloride]) Hexamethonium (hexamethylenebis[trimethylammonium bromide]) Fluorocholine ([2-fluoroethyl]trimethylammonium bromide) Thiocholine ([2-thioethyl]trimethylammonium chloride)
Percentage of Control
105 88 90 32
± ± ± ±
5.3 (2) 4.0 (3) 1.4 (3) 7.4t (5)
11±3.1 t(3) 90 ± 0.5 (2) 90 ± 0.5 (2) 87 ± 4.2 (3) 86 ± 3.9 (3) 80
± 10 (3)
93 ± 3.0 (3) 86 ± 4.1 (3)
* Cell layers on filters were incubated with Eagle's minimal essential medium containing 5 J.l.M [3H]choline for 90 s without (control) and with 100 J.l.M of the respective analogue and uptake measured in pmol/mg cell protein. Values of choline uptake for number of experiments indicated in parentheses are given as a percentage of control (mean ± SEM, or mean ± range for n = 2). t p < 0.05.
the molecule is an absolute requirement for the interaction of the substrate with the transport protein. In order to evaluate further the amino requirement, the primary amine, ethanolamine, and several substituted amines were studied. Ethanolamine showed at most weak inhibition of choline transport: at a 20-fold molar excess, the mean inhibition was only 12%. This finding is consistent with a report for renal collecting duct cells where the K, for ethanolamine inhibition of choline uptake was one order of magnitude higher than the K; for choline uptake (7), although ethanolamine has been somewhat more potent as an inhibitor in other systems (12). Monomethylethanolamine, a secondary amine, also did not significantly inhibit choline uptake (Table 1). By contrast, the tertiary amine, dimethylethanolamine, exerted a significant inhibition of choline uptake (Table 1), pointing to a requirement for at least two alkyl substituents on the nitrogen atom in order to compete with choline in this system. The n-decyl derivative of choline (dimethyl-n-decyl[2-hydroxyethyl]ammonium ion), a quaternary amine, was an even more potent inhibitor of choline transport, resulting in nearly 90% inhibition when present in 20-fold greater concentration (Table 1). When n-decylcholine and choline were present in equimolar concentration (5 j.tM), PH]choline uptake was 14 ± 4.3% of control (mean ± SEM, n = 4). Figures 1 and 2 show the dose response for the inhibitory effectsof dimethylethanolamine and n-decylcholine. A linear relationship was obtained when plotting the reciprocal of the choline transport rate against the concentration of both analogues. Assuming competition of these analogues and choline for the transporter, the K was estimated from the equation K = [1]50/(1 + [S]/Km) , where [1]50 is the analogue
428
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 6 1992
TABLE 2
0.04
Effect of preincubation of granular pneumocytes with ts-decylcholine on PHlcholine uptake* 0.03
Presence of n-Decylcholine Preincubation
>
During Uptake
Rate of PH]Choline Uptake (pmol/mg protein X 90 s)
+
297 210 31
Isolated alveolar type II epithelial cells (granular pneumocytes) from rat lung accumulate free choline against a concentration gradient by an energy-dependent saturable transport process with apparent K, approximately 18 JLM. In order to evaluate the structural requirements for choline transport by these cells, the inhibition of the initial rate of cellular uptake of PH]choline (5 JLM) by its analogue was measured. There was no significant inhibition of substrate uptake by analogues lacking an amino group while the presence of a quaternary nitrogen was most effective. N, N'-dimethylethanolamine (apparent K, 7 JLM) and n-decylcholine (apparent K, 0.5 JLM) were potent competitive inhibitors of choline transport. Substitution of the hydroxyl group in choline greatly diminished the inhibitory effect; fluorocholine, thiocholine, betaine, and betaine aldehyde showed little or no inhibition. This requirement for a hydroxyl group raises the possibility of hydrogen bonding of choline with the transport protein. The choline transport system in granular pneumocytes appears to differ from that in synaptosomes by the lower affinity of the carrier for substrate and for hemicholinium-3 and from that in erythrocytes by the role of the hydroxyl in the substrate molecule. The availability of inhibitory analogues for choline transport will facilitate isolation and study of the granular pneumocyte choline transport protein.
Phosphatidylcholine, a major component of the secretory product (i.e., lung surfactant) of granular pneumocytes, is synthesized de novo with choline as an important substrate precursor (1). Plasma choline, derived from dietary sources, is actively taken up by granular pneumocytes by a process with a K; of approximately 18 JLM (2). The properties of this transport process appear to be similar to those in some other cells, e. g., intestinal mucosa (3-5), guinea pig placental tissue (6), and renal collecting duct cells (7), but differ in several respects from the well-studied active transport process in synaptosomes (8). First, the affinity of the transport protein for choline in synaptosomes is some 20-fold higher (8) than in granular pneumocytes (2). Second, Naand choline are co-transported in synaptosomes (8), whereas there appears to be little or no involvement of Na' in choline transport in granular pneumocytes (2). Finally, hemicholinium-3 is a powerful inhibitor of the synaptosomal transport process (8), whereas the affinity of hemicholinium for the transport site in granular pneumocytes was relatively low (50% inhibition at 20-fold greater concentration) (2). A third choline transport system, essentially behaving as an equilibrating (facilitated diffusion) transport, has been described in red blood cells (9). The present study aims at char-
(Received in original form July 15, 1991 and in revised form September 30, 1991) Address correspondence to: A. B. Fisher, M.D., Institute for Environmental Medicine, University of Pennsylvania, Philadelphia, PA 19104-6068. Abbreviation: Eagle's modified essential medium, MEM. Am. J. Respir. Cell Mol. BioI. Vol. 6. pp. 426-429, 1992
acterization of the choline transport system in granular pneumocytes by analyzing its transport specificity with the goal to develop analogue inhibitors for further study and possible isolation of the choline transporter.
Materials and Methods The methods employed in this study for cell isolation and measurement of choline uptake were described in detail previously (2) and are summarized here. Granular pneumocytes were isolated by elastase digestion of blood-free rat lungs and were purified by "panning" on IgG-coated bacteriologic petri dishes to remove contaminating macrophages followed by differential adherence on tissue culture plastic dishes or collagen-coated Nuclepore polycarbonate membranes (3-JLm pore size, 5-cm diameter) during 21 h of incubation at 37° C in Eagle's minimal essential medium (MEM). In a usual experiment, 10 to 12 filters, each containing about 230 JLg protein (equivalent to approximately 2 X 1()6 cells), were prepared from the lungs of two rats. Cells on the filters were > 85 % granular pneumocytes as determined by specific fluorescence with phosphine 3R (2). The cells were washed 3 times in choline-free MEM (pH 7.4) and then preincubated in this medium at 37° C for 1 h. Measurement of uptake was carried out by addition of PH]choline (5 JLM choline; sp act, 100 JLCil JLmol) to choline-free MEM and incubation at 37° C for 90 s, during which periodthe uptake was linear (2). After incubation, adhering choline was removed by washing the filters 3 times (total wash time approximately 5 s) in ice-cold saline and then placing the filters in 1 ml of 0.1 M NaOH. The extract
Dodia, Fisher, Chander et al.: Lung Epithelial Choline Transport
was used for the determination of cell protein by the Coomassie blue dye binding assay and for measurement of dpm corresponding to internalized PH]choline (2). The rate of choline uptake is expressed in pmol choline/mg cell protein x 90 s. Our previous studies (2) expressed choline uptake as a function of cell water volume. Cell protein was used as the standard in the present study in order to avoid possible variation due to effects of inhibitors on cell volume. The H20 space in normal granular pneumocytes at 24 h in culture is approximately 4.1 j.tl!mg protein (10). Experiments with analogues were always carried out with a control (no inhibitor), and results are expressed as a percentage of control uptake in order to account for the variability of choline uptake with different preparations of cells. Efflux of PH]choline from cells was measured as previously described (2). Cells on plastic dishes (60-mm diameter) were preloaded for 10 min with PH]choline (5 j.tM, 100 j.tCi/nmol). Subsequently, the cells were washed with successive 3-ml portions of medium at 0.5, 1, 2, 3, 4, 5, and 10 min, and each wash was assayed for radiolabel. Parallel incubations were carried out under control conditions and with the presence of inhibitor in the wash solutions. At the end of the washout, the residual PH]choline in the cells was determined and expressed as log percent of the label remaining in the cells as a function of time. All incubations were carried out in duplicate, and results were averaged to give a mean value; duplicate incubations in one experiment generally did not differ by more than 15%. Data are generally expressed as mean ± SEM. Mean percent uptake in the presence of inhibitors was evaluated for deviation from control (100% uptake) by t test (11). Because inhibitors were studied in 20-fold molar excess over choline, we confined our statistical analysis to those compounds in which mean inhibition exceeded 50 %. The levelof statistical significance was taken as P < 0.05.
Materials Fluorocholine ([2-fluoroethyl]trimethylammonium) halides (chloride, bromide, iodide) were purchased from Lee's Bioorganic Laboratories (Marcus Hook, PA); the compounds were characterized by H-NMR and 19F-NMR and showed a single peak by ion chromatography. Dimethyl, n-decyl(hydroxyethyl)ammonium bromide (n-decylcholine) was prepared and analyzed for purity as described by Deves and Krupka (9). All other reagents were commercial preparations of the highest available purity.
Results The mean rate of choline uptake in the absence of inhibitors was 330 ± 25 pmol/mg cell protein/90 s (n = 20). This is equivalent to an uptake of approximately 80 pmol!j.tl cell H20/90 s, similar to our previous measurements (2). The effect of various structural analogues of choline on the rate of (3H]choline uptake is given in Table 1, expressed as a percentage of the rate of choline uptake in control cells, i.e., no added inhibitor. The first inhibitors studied evaluated the role of the amino group for the transport of choline. 3,3'-Dimethylbutanol, an analogue lacking the amino N, did not affect choline uptake (Table 1), indicating that the presence of a nitrogen atom in
427
TABLE 1
PH]Choline
uptake by granular pneumocytes: inhibition by structural analogues*
Choline Analogue
3,3' -Dimethyl-a-butanol Ethanolamine N-Methy lethanolamine N,N' -Dimethylethanolamine n-Decylcholine (dimethyl,n-decyl[2-hydroxyethyl]ammonium bromide) Trimethylamine Tetraethylammonium chloride Betaine ([carboxymethyl]trimethylammonium chloride) Betaine aldehyde ([formylmethyl)trimethylammonium chloride]) Hexamethonium (hexamethylenebis[trimethylammonium bromide]) Fluorocholine ([2-fluoroethyl]trimethylammonium bromide) Thiocholine ([2-thioethyl]trimethylammonium chloride)
Percentage of Control
105 88 90 32
± ± ± ±
5.3 (2) 4.0 (3) 1.4 (3) 7.4t (5)
11±3.1 t(3) 90 ± 0.5 (2) 90 ± 0.5 (2) 87 ± 4.2 (3) 86 ± 3.9 (3) 80
± 10 (3)
93 ± 3.0 (3) 86 ± 4.1 (3)
* Cell layers on filters were incubated with Eagle's minimal essential medium containing 5 J.l.M [3H]choline for 90 s without (control) and with 100 J.l.M of the respective analogue and uptake measured in pmol/mg cell protein. Values of choline uptake for number of experiments indicated in parentheses are given as a percentage of control (mean ± SEM, or mean ± range for n = 2). t p < 0.05.
the molecule is an absolute requirement for the interaction of the substrate with the transport protein. In order to evaluate further the amino requirement, the primary amine, ethanolamine, and several substituted amines were studied. Ethanolamine showed at most weak inhibition of choline transport: at a 20-fold molar excess, the mean inhibition was only 12%. This finding is consistent with a report for renal collecting duct cells where the K, for ethanolamine inhibition of choline uptake was one order of magnitude higher than the K; for choline uptake (7), although ethanolamine has been somewhat more potent as an inhibitor in other systems (12). Monomethylethanolamine, a secondary amine, also did not significantly inhibit choline uptake (Table 1). By contrast, the tertiary amine, dimethylethanolamine, exerted a significant inhibition of choline uptake (Table 1), pointing to a requirement for at least two alkyl substituents on the nitrogen atom in order to compete with choline in this system. The n-decyl derivative of choline (dimethyl-n-decyl[2-hydroxyethyl]ammonium ion), a quaternary amine, was an even more potent inhibitor of choline transport, resulting in nearly 90% inhibition when present in 20-fold greater concentration (Table 1). When n-decylcholine and choline were present in equimolar concentration (5 j.tM), PH]choline uptake was 14 ± 4.3% of control (mean ± SEM, n = 4). Figures 1 and 2 show the dose response for the inhibitory effectsof dimethylethanolamine and n-decylcholine. A linear relationship was obtained when plotting the reciprocal of the choline transport rate against the concentration of both analogues. Assuming competition of these analogues and choline for the transporter, the K was estimated from the equation K = [1]50/(1 + [S]/Km) , where [1]50 is the analogue
428
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 6 1992
TABLE 2
0.04
Effect of preincubation of granular pneumocytes with ts-decylcholine on PHlcholine uptake* 0.03
Presence of n-Decylcholine Preincubation
>
During Uptake
Rate of PH]Choline Uptake (pmol/mg protein X 90 s)
+
297 210 31
