Protamine MICHAEL Pulmonary University
interaction
with the epithelial
Wc PETERSON AND DIANE GRUENHAUPT Division, Department of Medicine, College of Medicine, of hua, Iowa City, Ioura 52242
PETERSON, MICHAEL W., AND DIANE GRIJENHAUPT. Protamine interaction with the epitheliul cellsurface.J. Appl. Physiol. 72(l): 236-241, 1992.-We have previously reported that exposing cultured Madin Darby canine kidney (MDCK) cells to the polycation protamine (PRO) results in increasedshort-circuit current, and decreasedbarrier integrity as measuredby mannitol permeability and transepithelial electrical resistance. To further investigate the interaction of PRO with the surface of epithelial cells, we labeled PRO with [‘“Cl with useof reductive alkylation. [14C]PR0 bound to the cells in a biphasic pattern. Approximately 10% of the [14C]PR0 was bound to the cells in the first 5 min, followed by an additional 10%that was bound over the next 25 min. No additional [14C]PR0 bound to the cellsafter the initial 30 min. Binding of [14C]PR0wasinhibited by “cold” PRO, which suggestedspecificity. Binding was alsoinhibited by polyanions, serum,and albumin, agentspreviously found to protect MDCK cells from PRO-induced injury. The binding of PRO to MDCK cellswas not inhibited by incubation of the MDCK cells with neuraminidase,to remove surface sialic acid residues,or with heparinase,to remove surface heparan sulfate, even though metabolic labeling experiments demonstratedthat neuraminidasedecreasedcell sialic acid and heparinasedecreasedcell heparan sulfate. Neuraminidaseand heparinaseoffered no protection from PRO injury and had no effect themselveson mannitol permeability. Incubation of the cellswith trypsin, however, blunted both the binding of PRO to the cells and the increasein mannitol permeability after exposure of the cellsto PRO. These results suggestthat PRO binds to a trypsin-sensitive site on MDCK cells,that protamine is not binding to sialic acid or heparan sulfate moieties,that there is specificity to the PRO binding, and that agents that prevent this binding protect the cells from injury. Becausea number of inflammatory cells contain polycations that are released at sitesof inflammation, theseobservationshave implications for understanding inflammatory epithelial injury.
cationic proteins; epithelial permeability; inflammation
THE POLYCATION PROTAMINE affects the barrierintegrity of both cultured endothelial monolayers (17) and
cultured epithelial monolayers (16,20) and increases the albumin permeability in isolated perfused lung models (7, 22). In the isolated perfused lung model it has been suggested that protamine acts by binding to anionic sites on the endothelial cell surface and neutralizing the negative surface charge and that this neutralization of the luminal charge alters the way the anionic protein, albumin, moves across the capillary bed (22). In support of this hypothesis the investigators demonstrated that removing surface heparan sulfate by perfusing the lung with heparinase resulted in the same increase in perme236
cell surface
ability to albumin as that induced by perfusion with protamine. Simple surface charge neutralization seems an unlikely explanation for the results of experiments with cultured epithelial cells, however, for several reasons. First, studies with Madin Darby canine kidney (MDCK) monolayers have shown that protamine exposure increases permeability to mannitol and decreases transepithelial electrical resistance (16). These two findings together suggest that protamine causes alterations in the parace llular pathway or tight j unctions. Second, microscopic studies of the glome&li- and lung after protamine exposure have reported changes in the microscopic appearance of epithelial ceils associated with cell injury. Finally, not all epithelial cells respond to protamine in the same way even though they all have the same negative charge on their surface @,8,W . One potential explanation for these findings is that protamine binds to specific sites on protamine-sensitive epithelial cells. On the basis of the observation that polyanions protect the MDCK monolayers from protamine injury, these sites may be anionic sites. The most prev ,alent anionic compopronents on the cell surface are sialic acid-containing teins and the sulfated glycosaminoglycan, heparan sulfate. An alternative explanation is that protamine interacts electrostatically with proteins on the epithelial cell surface (9). The present study was undertaken to investigate the .teraction of the polycati .on protamine with the surface of MDCK epithelial Cells. Using radiolabeled protamine, we first addressed the question whether protamine binds to the epithelial cell surface and whether polyanions and serum, agents that prevent protamine-induced increase in permeability, prevent protamine binding to the epithelial cell surface. Next, we addressed the nature of the binding of protamine to the epithelial cell surface by selectively removing heparan sulfate, sialic acid, or proteins from the cell surface. Finally, we compared the degree of protamine binding after treatment with these agents with their ability to protect the monolayers from injury after protamine exposure. MATERIALS
AND
METHODS
Reagents. Eagle’s minimal essential medium (MEM) was purchased from Gibco, [l*C]mannitol (52 mCi/ mmol) was purchased from New England Nuclear, and neuraminidase, heparinase, essen tial fatty acid free bovine serum albumin, protamine, heparin, and sulfated
0161-7567&Z $2.00 Copyright 0 1992 the American Physiological Society
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 14, 2019.
PROTAMINE
BINDING
dextran (7,000 avg mol wt) were purchased from Sigma Chemical (St. Louis, MO). Cell culture. High-resistance MDCK cells were maintained in culture in MEM containing Earle’s salts with 10% fetal bovine serum (FBS), penicillin (100 U/ml), and streptomycin (100 pg/ml) at 37*C in 95% air-5% CO,. They were passaged weekly after exposure to 0.25% trypsin and 0.1% EDTA with a plating density of 1~10. They maintained the typical appearance of sheets of polygonal cells with domes in culture. For experiments measuring mannitol permeability (P,,,,), the cells were plated at the same density on cellulose filter inserts (Millicell-HA, 0.45pm pore size, 0.6 cm2 surface area), which were free standing in 24-well tissue culture dishes. The cells were cultured for 5-7 days before use. Measurement of pm,,. P,m, was measured with cells cultured on Millicell-HA inserts as previously described (12, 16,20). Cells were cultured for 5-7 days before use. Before addition of [ 14C]mannitol-containing solutions, the medium was aspirated from the inserts and the cells were rinsed in warm MEM. The inserts were then placed in a 24.well plate, 0.33 ml of MEM containing 0.5 &i/ml (9.1 nmol/ml) of [14C]mannitol was layered over the monolayer, and 0.54 ml plain MEM was added to the well. These volumes ensured no hydrostatic pressure was exerted across the monolayer. Because it has previously been found that the flux of mannitol is the same in this system whether it is measured in the apical to basolateral or basolateral to apical direction, the experiments were all done measuring flux in apical to basolateral. At the end of 1 h the monolayers were moved to a second plate containing fresh MEM, and we added the agents of interest to the apical surface of the monolayer. The P,,,, determined for each filter during the 1st h was defined as its baseline pma,, . We found that there was some variation in the absolute value of P,,,, among monolayers in the same experiment but that the P,,,, of each monolayer remained constant for as long as 24 h. For this reason our results are expressed as percentage baseline P,,,, for each monolayer. After each hour the monolayers were transferred to new 24.well plates, and the medium in the wells containing the radiolabeled mannitol that had crossed the monolayer was added to scintillation cocktail and counted in a liquid scintillation counter. At the conclusion of the experiment, the medium was removed from the apical surface and counted. The permeability of the monolayers to mannitol for each hour was calculated as P = F, X [mann];’
X s-l
where P is permeability, F, is flux from apical to basolatera1 as counts per minute, [mann], is concentration of mannitol in the apical bathing media at the beginning of the time period, and S is surface area (0.6 cm2 in our system). Because a small amount of mannitol crossed the monolayers each hour, the apical concentration of mannito1 decreased slightly each hour. We accounted for this decreased [mann], in calculating P,=, by subtracting the amount of mannitol that had crossed the monolayer during earlier incubations from the [man& at time 0. The resulting value for [mann], was used as the [mann],
TO EPITHELIA
237
at the beginning of that time period. In all experiments each group was composed of at least six filters. Protein labeling and adherence assays. Protamine was labeled with 14C using reductive alkylation, as previously described (6), and dialyzed extensively to remove unincorporated 14C. Incubation of the labeled protamine in cold 6% trichloroacetic acid resulted in precipitation of 93% of the counts. These results are consistent with most of the 14C remaining after dialysis being bound to the protein. Monolayers of MDCK cells were grown to confluence, washed with MEM to remove serum, and incubated with MEM containing 2 pg/ml [14C]protamine for varying time periods. The medium was then removed and the cells were washed twice with MEM to remove unbound [14C]protamine. The cells were then scraped from the plates and added to scintillation fluid. Aliquots from each of the washes and the scraped cells were separately added to scintillation vials, combined with scintillation fluid, and counted in a liquid scintillation counter. The results were expressed as percentage of added counts remaining adherent to the cells. Less than 50 cpm/ml were removed with the second wash with this method. Metabolic labeling experiments. Sialic acid was labeled with [3H]glucosamine to monitor the release of sialic acid after neuraminidase exposure (18). To label with [3H] glucosamine, confluent plates of MDCK cells were incubated for 12 h with MEM modified to contain only 100 mg/lOO ml of glucose (low glucose). The monolayers were then incubated for 24 h with low-glucose MEM containing 2 &i/ml [3H] glucosamine followed by incubation for 24 h in MEM + 10% FBS containing no [3H]glucosamine. The media was removed and the cells washed three times with MEM. The monolayers were then incubated with medium containing 0, 25, 50, or 75 PUlml of neuraminidase. After 30 min, we removed medium and washed the cells twice with MEM. An aliquot from each of these washes was added to liquid scintillation cocktail and counted in a liquid scintillation counter. After washing, the cells were scraped from the plates, added to liquid scintillation cocktail, and counted in a liquid scintillation counter. The results of these experiments were expressed as percentage total counts (all washes plus scraped cells) released into the media. The release of heparan sulfate after heparinase exposure was measured by labeling the cells with 35S04 followed by anion exchange chromatography (15,20). Confluent monolayers of MDCK cells were incubated with MEM + 10% FBS containing 35S04 (50 &Yml) as H,S04 for 48 h. The unincorporated 35S04 was removed from the cells by washing the cells four times with MEM. The monolayers were then incubated with MEM containing either 0,50,100, or 150 mU/ml heparinase. After 6 h the medium was removed and the cells were washed twice with MEM. The media and washes from each monolayer were spun at 1,000 g to remove any cell debris, combined, and passed over a DEAE sephadex column (0.5-ml bed volume) equilibrated with buffer A (0.01 M K,PO,, 0.5% Triton X-100, and 0.15 M NaCl, pH 7.4). Each column was then washed with 6 ml of bufferA followed by 6 ml of buffer B (0.01 M K,PO,, 0.5% Triton X-100, and 0.25 M NaCl, pH 7.4). The heparan sulfate was then eluted from the column by washing with 6 ml of buffer C (0.01 M
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 14, 2019.
238
PROTAMINE
BINDING
TO EPITHELIA
-
-
0
”
10 Time
”
20
”
30
’
”
40
’
50
”
60
of Exposure
1. Addition of [14C]protamine to cultured Madin Darby canine kidney monolayers results in an initial rapid binding over first 5 min followed by slower binding that reached a plateau by 30 min. No additional protamine bound to cells between 30 and 60 min. CPM, counts/ minute. FIG.
K,PO,, 0.05% Triton X-100, and 0.5 M NaCl, pH 7.4). Aliquots of buffer C washes were added to scintillation cocktail and counted in a liquid scintillation counter. The cell-associated heparan sulfate was solubilized by incubating the cells with buffer B. This cell lysate was also added to DEAE sephadex columns, and heparan sulfate was eluted from the columns and counted as described for the cell supernatants. The results again are expressed as percentage total counts removed after heparinase treatment. Statistical analysis. Comparison of results within groups was carried out by performing one-way analysis of variance. Post hoc testing within the groups was done using the Newman-Keuls test. P < 0.05 was considered significant. RESULTS
Adherence of protamine. Radiolabeled protamine adhered to the surface of the cultured epithelial monolayer in a time-dependent and saturable fashion (n = 3 for each data point; Fig. 1). Approximately 10% of the added p&amine adhered to the monolayer in the first 5 min. An additional lo-15% of the added protamine adhered over the next 25 min, but no further protamine bound after another 30 min. We used these data to estimate the number of protamine binding sites on the monolayers. Each well had a surface area of 4.91 cm2 and held 0.5 ml of medium containing 1 pg (200 pmol) of protamine. If 20% of the protamine bound to the cells, we estimated the number of binding sites as 49,000/ym2 (moles X Avogadro’s number/surface area). In some parallel experiments, we found similar degrees of binding to cells grown on porous supports. There is some specificity of the binding because incubating the epithelial monolayers with [ 14C]protamine in the presence of either an equal concentration or a IO-fold excess concentration of unlabeled protamine significantly reduced the binding of the radiolabeled protamine (Fig. 2). Polyanions and serum have previously been shown to protect the epithelial monolayers from protamine-induced increase in Pman, and decrease in transepithelial
............... ............... ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ............... ............... .............. ............... .............. ............... ............... ............... ............... ............... ............... ............... .............. ............... .............. ............... .............. ............... .............. ............... .............. ............... .............. ............... ...............
. ... . . ......... ............... ............... ............... ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ............... ............... ............... ............... ............... ............... ................ ............... ................ ............... ............... ............... ............... ............... ............... ............... ............... ............... ............... ............... ............... ............... .............. ............... ............... ............... ................ ............... ................ ............... ................ ............... ................ ............... ............... ............... ..............
T I
.............. ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ............... ............... ............... ............... ............... ............... ............... ............... ............... ............... ................ ............... ................ ............... ................ ............... ................ ............... ............... ............... ................ ............... ................ ............... ................ ............... ............... ............... ..............
I!
1:o
1O:l
1:l
T .............. ............... .............. ............... ............... ............... .............. ............... .............. ............... .............. ............... .............. ............... .............. ............... .............. ............... ............... .............. n.............. 15
1:lO
Ratio of Radiolabeled: Cold Protamine FIG. 2. There is some specificity to binding of protamine because there is less [14C]protamine bound to cells in presence of unlabeled protamine. These binding assays were done after 30 min of incubation at 37OC.
electrical resistance. Incubation of the epithelial monolayers with [14C]protamine in the presence of the polyanions heparin or sulfated dextran or in the presence of serum prevented adherence of the protamine to the MDCK cell surface (Table 1). Because the major anionic protein in serum is albumin, the binding assay was repeated in the presence of albumin. The protective effect of serum appeared to be due to the albumin because the binding inhibition was duplicated by 3 mg/ml of bovine serum albumin. These results are consistent with the hypothesis that polyanions, including the serum protein albumin, protect MDCK cells in part by preventing binding of the protamine to the cell surface. To investigate the cell surface components involved in protamine binding, the MDCK monolayers were treated with heparinase to remove surface heparan sulfate, with neuraminidase to remove surface sialic acid residues, or with trypsin to remove surface proteins, before incubation with [14C]protamine. Incubation of the cells with heparinase at 50,100, or 150 mu/ml for 6 h had no effect on adherence of the radiolabeled protamine (Table 2). Exposure of the cells to 100 and 150 mu/ml for 6 h did, however, increase the release of 35S0,-labeled heparan 1. Effects of polyanions, serum, and albumin on binding of [‘*C]protamine to MDCK monolayers TABLE
Treatment Group
% P&amine Bound
Control Heparin Control Sulfated dextran Control 10% Serum Control Albumin (3 mglml)
20.7kO.5 1.4t0.2 17.7kl.O 2.3tO. 1 9.9zt2.7 2.0tl.l 7.1+1.8 3.6kO.4
P
x0.01 co.01 -co.05
interaction
with the epithelial
Wc PETERSON AND DIANE GRUENHAUPT Division, Department of Medicine, College of Medicine, of hua, Iowa City, Ioura 52242
PETERSON, MICHAEL W., AND DIANE GRIJENHAUPT. Protamine interaction with the epitheliul cellsurface.J. Appl. Physiol. 72(l): 236-241, 1992.-We have previously reported that exposing cultured Madin Darby canine kidney (MDCK) cells to the polycation protamine (PRO) results in increasedshort-circuit current, and decreasedbarrier integrity as measuredby mannitol permeability and transepithelial electrical resistance. To further investigate the interaction of PRO with the surface of epithelial cells, we labeled PRO with [‘“Cl with useof reductive alkylation. [14C]PR0 bound to the cells in a biphasic pattern. Approximately 10% of the [14C]PR0 was bound to the cells in the first 5 min, followed by an additional 10%that was bound over the next 25 min. No additional [14C]PR0 bound to the cellsafter the initial 30 min. Binding of [14C]PR0wasinhibited by “cold” PRO, which suggestedspecificity. Binding was alsoinhibited by polyanions, serum,and albumin, agentspreviously found to protect MDCK cells from PRO-induced injury. The binding of PRO to MDCK cellswas not inhibited by incubation of the MDCK cells with neuraminidase,to remove surface sialic acid residues,or with heparinase,to remove surface heparan sulfate, even though metabolic labeling experiments demonstratedthat neuraminidasedecreasedcell sialic acid and heparinasedecreasedcell heparan sulfate. Neuraminidaseand heparinaseoffered no protection from PRO injury and had no effect themselveson mannitol permeability. Incubation of the cellswith trypsin, however, blunted both the binding of PRO to the cells and the increasein mannitol permeability after exposure of the cellsto PRO. These results suggestthat PRO binds to a trypsin-sensitive site on MDCK cells,that protamine is not binding to sialic acid or heparan sulfate moieties,that there is specificity to the PRO binding, and that agents that prevent this binding protect the cells from injury. Becausea number of inflammatory cells contain polycations that are released at sitesof inflammation, theseobservationshave implications for understanding inflammatory epithelial injury.
cationic proteins; epithelial permeability; inflammation
THE POLYCATION PROTAMINE affects the barrierintegrity of both cultured endothelial monolayers (17) and
cultured epithelial monolayers (16,20) and increases the albumin permeability in isolated perfused lung models (7, 22). In the isolated perfused lung model it has been suggested that protamine acts by binding to anionic sites on the endothelial cell surface and neutralizing the negative surface charge and that this neutralization of the luminal charge alters the way the anionic protein, albumin, moves across the capillary bed (22). In support of this hypothesis the investigators demonstrated that removing surface heparan sulfate by perfusing the lung with heparinase resulted in the same increase in perme236
cell surface
ability to albumin as that induced by perfusion with protamine. Simple surface charge neutralization seems an unlikely explanation for the results of experiments with cultured epithelial cells, however, for several reasons. First, studies with Madin Darby canine kidney (MDCK) monolayers have shown that protamine exposure increases permeability to mannitol and decreases transepithelial electrical resistance (16). These two findings together suggest that protamine causes alterations in the parace llular pathway or tight j unctions. Second, microscopic studies of the glome&li- and lung after protamine exposure have reported changes in the microscopic appearance of epithelial ceils associated with cell injury. Finally, not all epithelial cells respond to protamine in the same way even though they all have the same negative charge on their surface @,8,W . One potential explanation for these findings is that protamine binds to specific sites on protamine-sensitive epithelial cells. On the basis of the observation that polyanions protect the MDCK monolayers from protamine injury, these sites may be anionic sites. The most prev ,alent anionic compopronents on the cell surface are sialic acid-containing teins and the sulfated glycosaminoglycan, heparan sulfate. An alternative explanation is that protamine interacts electrostatically with proteins on the epithelial cell surface (9). The present study was undertaken to investigate the .teraction of the polycati .on protamine with the surface of MDCK epithelial Cells. Using radiolabeled protamine, we first addressed the question whether protamine binds to the epithelial cell surface and whether polyanions and serum, agents that prevent protamine-induced increase in permeability, prevent protamine binding to the epithelial cell surface. Next, we addressed the nature of the binding of protamine to the epithelial cell surface by selectively removing heparan sulfate, sialic acid, or proteins from the cell surface. Finally, we compared the degree of protamine binding after treatment with these agents with their ability to protect the monolayers from injury after protamine exposure. MATERIALS
AND
METHODS
Reagents. Eagle’s minimal essential medium (MEM) was purchased from Gibco, [l*C]mannitol (52 mCi/ mmol) was purchased from New England Nuclear, and neuraminidase, heparinase, essen tial fatty acid free bovine serum albumin, protamine, heparin, and sulfated
0161-7567&Z $2.00 Copyright 0 1992 the American Physiological Society
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 14, 2019.
PROTAMINE
BINDING
dextran (7,000 avg mol wt) were purchased from Sigma Chemical (St. Louis, MO). Cell culture. High-resistance MDCK cells were maintained in culture in MEM containing Earle’s salts with 10% fetal bovine serum (FBS), penicillin (100 U/ml), and streptomycin (100 pg/ml) at 37*C in 95% air-5% CO,. They were passaged weekly after exposure to 0.25% trypsin and 0.1% EDTA with a plating density of 1~10. They maintained the typical appearance of sheets of polygonal cells with domes in culture. For experiments measuring mannitol permeability (P,,,,), the cells were plated at the same density on cellulose filter inserts (Millicell-HA, 0.45pm pore size, 0.6 cm2 surface area), which were free standing in 24-well tissue culture dishes. The cells were cultured for 5-7 days before use. Measurement of pm,,. P,m, was measured with cells cultured on Millicell-HA inserts as previously described (12, 16,20). Cells were cultured for 5-7 days before use. Before addition of [ 14C]mannitol-containing solutions, the medium was aspirated from the inserts and the cells were rinsed in warm MEM. The inserts were then placed in a 24.well plate, 0.33 ml of MEM containing 0.5 &i/ml (9.1 nmol/ml) of [14C]mannitol was layered over the monolayer, and 0.54 ml plain MEM was added to the well. These volumes ensured no hydrostatic pressure was exerted across the monolayer. Because it has previously been found that the flux of mannitol is the same in this system whether it is measured in the apical to basolateral or basolateral to apical direction, the experiments were all done measuring flux in apical to basolateral. At the end of 1 h the monolayers were moved to a second plate containing fresh MEM, and we added the agents of interest to the apical surface of the monolayer. The P,,,, determined for each filter during the 1st h was defined as its baseline pma,, . We found that there was some variation in the absolute value of P,,,, among monolayers in the same experiment but that the P,,,, of each monolayer remained constant for as long as 24 h. For this reason our results are expressed as percentage baseline P,,,, for each monolayer. After each hour the monolayers were transferred to new 24.well plates, and the medium in the wells containing the radiolabeled mannitol that had crossed the monolayer was added to scintillation cocktail and counted in a liquid scintillation counter. At the conclusion of the experiment, the medium was removed from the apical surface and counted. The permeability of the monolayers to mannitol for each hour was calculated as P = F, X [mann];’
X s-l
where P is permeability, F, is flux from apical to basolatera1 as counts per minute, [mann], is concentration of mannitol in the apical bathing media at the beginning of the time period, and S is surface area (0.6 cm2 in our system). Because a small amount of mannitol crossed the monolayers each hour, the apical concentration of mannito1 decreased slightly each hour. We accounted for this decreased [mann], in calculating P,=, by subtracting the amount of mannitol that had crossed the monolayer during earlier incubations from the [man& at time 0. The resulting value for [mann], was used as the [mann],
TO EPITHELIA
237
at the beginning of that time period. In all experiments each group was composed of at least six filters. Protein labeling and adherence assays. Protamine was labeled with 14C using reductive alkylation, as previously described (6), and dialyzed extensively to remove unincorporated 14C. Incubation of the labeled protamine in cold 6% trichloroacetic acid resulted in precipitation of 93% of the counts. These results are consistent with most of the 14C remaining after dialysis being bound to the protein. Monolayers of MDCK cells were grown to confluence, washed with MEM to remove serum, and incubated with MEM containing 2 pg/ml [14C]protamine for varying time periods. The medium was then removed and the cells were washed twice with MEM to remove unbound [14C]protamine. The cells were then scraped from the plates and added to scintillation fluid. Aliquots from each of the washes and the scraped cells were separately added to scintillation vials, combined with scintillation fluid, and counted in a liquid scintillation counter. The results were expressed as percentage of added counts remaining adherent to the cells. Less than 50 cpm/ml were removed with the second wash with this method. Metabolic labeling experiments. Sialic acid was labeled with [3H]glucosamine to monitor the release of sialic acid after neuraminidase exposure (18). To label with [3H] glucosamine, confluent plates of MDCK cells were incubated for 12 h with MEM modified to contain only 100 mg/lOO ml of glucose (low glucose). The monolayers were then incubated for 24 h with low-glucose MEM containing 2 &i/ml [3H] glucosamine followed by incubation for 24 h in MEM + 10% FBS containing no [3H]glucosamine. The media was removed and the cells washed three times with MEM. The monolayers were then incubated with medium containing 0, 25, 50, or 75 PUlml of neuraminidase. After 30 min, we removed medium and washed the cells twice with MEM. An aliquot from each of these washes was added to liquid scintillation cocktail and counted in a liquid scintillation counter. After washing, the cells were scraped from the plates, added to liquid scintillation cocktail, and counted in a liquid scintillation counter. The results of these experiments were expressed as percentage total counts (all washes plus scraped cells) released into the media. The release of heparan sulfate after heparinase exposure was measured by labeling the cells with 35S04 followed by anion exchange chromatography (15,20). Confluent monolayers of MDCK cells were incubated with MEM + 10% FBS containing 35S04 (50 &Yml) as H,S04 for 48 h. The unincorporated 35S04 was removed from the cells by washing the cells four times with MEM. The monolayers were then incubated with MEM containing either 0,50,100, or 150 mU/ml heparinase. After 6 h the medium was removed and the cells were washed twice with MEM. The media and washes from each monolayer were spun at 1,000 g to remove any cell debris, combined, and passed over a DEAE sephadex column (0.5-ml bed volume) equilibrated with buffer A (0.01 M K,PO,, 0.5% Triton X-100, and 0.15 M NaCl, pH 7.4). Each column was then washed with 6 ml of bufferA followed by 6 ml of buffer B (0.01 M K,PO,, 0.5% Triton X-100, and 0.25 M NaCl, pH 7.4). The heparan sulfate was then eluted from the column by washing with 6 ml of buffer C (0.01 M
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 14, 2019.
238
PROTAMINE
BINDING
TO EPITHELIA
-
-
0
”
10 Time
”
20
”
30
’
”
40
’
50
”
60
of Exposure
1. Addition of [14C]protamine to cultured Madin Darby canine kidney monolayers results in an initial rapid binding over first 5 min followed by slower binding that reached a plateau by 30 min. No additional protamine bound to cells between 30 and 60 min. CPM, counts/ minute. FIG.
K,PO,, 0.05% Triton X-100, and 0.5 M NaCl, pH 7.4). Aliquots of buffer C washes were added to scintillation cocktail and counted in a liquid scintillation counter. The cell-associated heparan sulfate was solubilized by incubating the cells with buffer B. This cell lysate was also added to DEAE sephadex columns, and heparan sulfate was eluted from the columns and counted as described for the cell supernatants. The results again are expressed as percentage total counts removed after heparinase treatment. Statistical analysis. Comparison of results within groups was carried out by performing one-way analysis of variance. Post hoc testing within the groups was done using the Newman-Keuls test. P < 0.05 was considered significant. RESULTS
Adherence of protamine. Radiolabeled protamine adhered to the surface of the cultured epithelial monolayer in a time-dependent and saturable fashion (n = 3 for each data point; Fig. 1). Approximately 10% of the added p&amine adhered to the monolayer in the first 5 min. An additional lo-15% of the added protamine adhered over the next 25 min, but no further protamine bound after another 30 min. We used these data to estimate the number of protamine binding sites on the monolayers. Each well had a surface area of 4.91 cm2 and held 0.5 ml of medium containing 1 pg (200 pmol) of protamine. If 20% of the protamine bound to the cells, we estimated the number of binding sites as 49,000/ym2 (moles X Avogadro’s number/surface area). In some parallel experiments, we found similar degrees of binding to cells grown on porous supports. There is some specificity of the binding because incubating the epithelial monolayers with [ 14C]protamine in the presence of either an equal concentration or a IO-fold excess concentration of unlabeled protamine significantly reduced the binding of the radiolabeled protamine (Fig. 2). Polyanions and serum have previously been shown to protect the epithelial monolayers from protamine-induced increase in Pman, and decrease in transepithelial
............... ............... ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ............... ............... .............. ............... .............. ............... ............... ............... ............... ............... ............... ............... .............. ............... .............. ............... .............. ............... .............. ............... .............. ............... .............. ............... ...............
. ... . . ......... ............... ............... ............... ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ............... ............... ............... ............... ............... ............... ................ ............... ................ ............... ............... ............... ............... ............... ............... ............... ............... ............... ............... ............... ............... ............... .............. ............... ............... ............... ................ ............... ................ ............... ................ ............... ................ ............... ............... ............... ..............
T I
.............. ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ................ ............... ............... ............... ............... ............... ............... ............... ............... ............... ............... ............... ................ ............... ................ ............... ................ ............... ................ ............... ............... ............... ................ ............... ................ ............... ................ ............... ............... ............... ..............
I!
1:o
1O:l
1:l
T .............. ............... .............. ............... ............... ............... .............. ............... .............. ............... .............. ............... .............. ............... .............. ............... .............. ............... ............... .............. n.............. 15
1:lO
Ratio of Radiolabeled: Cold Protamine FIG. 2. There is some specificity to binding of protamine because there is less [14C]protamine bound to cells in presence of unlabeled protamine. These binding assays were done after 30 min of incubation at 37OC.
electrical resistance. Incubation of the epithelial monolayers with [14C]protamine in the presence of the polyanions heparin or sulfated dextran or in the presence of serum prevented adherence of the protamine to the MDCK cell surface (Table 1). Because the major anionic protein in serum is albumin, the binding assay was repeated in the presence of albumin. The protective effect of serum appeared to be due to the albumin because the binding inhibition was duplicated by 3 mg/ml of bovine serum albumin. These results are consistent with the hypothesis that polyanions, including the serum protein albumin, protect MDCK cells in part by preventing binding of the protamine to the cell surface. To investigate the cell surface components involved in protamine binding, the MDCK monolayers were treated with heparinase to remove surface heparan sulfate, with neuraminidase to remove surface sialic acid residues, or with trypsin to remove surface proteins, before incubation with [14C]protamine. Incubation of the cells with heparinase at 50,100, or 150 mu/ml for 6 h had no effect on adherence of the radiolabeled protamine (Table 2). Exposure of the cells to 100 and 150 mu/ml for 6 h did, however, increase the release of 35S0,-labeled heparan 1. Effects of polyanions, serum, and albumin on binding of [‘*C]protamine to MDCK monolayers TABLE
Treatment Group
% P&amine Bound
Control Heparin Control Sulfated dextran Control 10% Serum Control Albumin (3 mglml)
20.7kO.5 1.4t0.2 17.7kl.O 2.3tO. 1 9.9zt2.7 2.0tl.l 7.1+1.8 3.6kO.4
P
x0.01 co.01 -co.05
