British Journal ofHaematology, 1977, 35, 71.
Myosin in Cultured Human Endothelial Cells ANNEMOORE,ERICA. JAFFE,CARLG. BECKER AND RALPH L. NACHMAN
Division of Haematology, Department of Medicine and Department of Pathology, Corne11 University Medical College, N e w York (Received I ]une 1976; accepted for publication ~ o ] u n e1976) SUMMARY. Myosin was isolated from cultured human endothelial cells by extraction with 0.6 M KC1 and chromatography on Sepharose 4B. The extracted endothelial cell protein was identified as myosin by the characteristic ATPase profile, that is, the ATPase was activated by Ca2 and EDTA and inhibited by MgZ +. On sodium dodecyl sulphate polyacrylamide gel el'ectrophoresis, the endothelial ccll myosin heavy chain migrated with a molecular weight of 200 ooo as did rabbit uterine and human platelet myosin heavy chains. A crude preparation of the cndothelial cell myosin reacted immunologically with an antiserum to platelet myosin, a smooth muscle type of myosin. In indirect immunofluorescence studies, antiserum to the purified endothelial cell myosin stained cultured endothelial cells in a fibrillar pattern. The fibrillar pattern was more intense when the endothelial cells were stained with antiserum to platelet myosin. The presence of myosin in the endothelial cell provides a basis for the contractility of these cells. This contractile property may plan an important role in the physiologic function of these cells. +
Contractile proteins have been found in many non-muscle cells and may be important for cell motility, cell division, secretion, and phagocytosis (Pollard & Weihing, 1974; Adelstein, 1975).Human endothelial cells both in vivo and in vitro have been shown by indirect methods to contain contractile proteins which are antigenically similar to platelet and uterine actomyosin (Becker & Nachman, 1973 ; Jaffe et a!, 1973). Endothelial cell contractility might play a role in such physiologic processes as the regulation of vascular permeability, the inflammatory response and the initiation of thrombosis. In order to study the contractile mechanisms of these cells, we have isolated and characterized endothelial cell myosin. The purification of this protein was made possible by the recent development of two techniques. First, pure human endothelial cells free of contaminating smooth muscle cells or fibroblasts have become available in culture (Jaffe et af, 1973). Second, a method for extracting myosin from a small amount of cell protein has been developed (Ostlund et al, 1974). In this report we describe the partial purification and characterization of myosin from cultured human umbilical vein endothelial cells. Correspondence: Dr Anne Moore, Division of Hematology, Department of Medicine, Cornell University Medical College, $25 East 68th Street, New York, N.Y. 10021, U.S.A.
72
A n n e Moore et al MATERIALS AND METHODS
Endothelid cell culture. Endothelial cells were obtained from human umbilical cord veins and cultured as previously described (Jaffe et al, 1973). Extraction of endothelial cell myosin. The method described by Ostlund et a! (1974) was used. All procedures were performed at 0-4"C. Endothelial cells grown to confluence were washed three times with phosphate buffered saline (0.145 M NaCI, 0.01M phosphate buffer, pH 7.4) containing 2 mM dithiothreitol (DTT) (Sigma Chemical Co., St Louis, Missouri) and once with 0.15M NaCl-2 I I ~ MDTT containing 0.5% bovine serum albumin (Miles Laboratories, Inc., Kankakee, Illinois) added to facilitate the uptake of the cells. The cells were removed with a rubber policeman and sedimented by centrifugation at 700 g for 15 min. The cell pellet was washed twice in 0.15 M NaC1-2 mM DTT by centrifugation at 700 g for 5 min and then frozen at - 70°C. The cell pellet was used usually within 2 or 3 weeks but equivalent results were found with pellets frozen up to 5 months. In a typical experiment, cells from approximately 40 Petri dishes ( IOO x 20 mm) or the equivalent number of cells from T-75 flasks or 2-litre roller bottles were used. This was equivalent to 30-40 mg of total cell protein. To extract myosin, the cell pellets were thawed, resuspended in buffer A (0.6 M KCl, 1 5 mM Tris-HC1, pH 7.5 containing I-butanol (3%, v/v) and 2 mM DTT) and stirred intermittently for I h. The suspension was then centrifuged at 12 ooog for 10 min. The cell lysate supernatant, concentrated to I ml in dialysis tubing packed in dry Sephadex G-zoo gel (Pharmacia Fine Chemicals, Inc., Piscataway, N.J.), was applied to a column (0.9 x 60 cni) of Sepharose 4B (Pharmacia) which had been washed with buffer B (buffer A without butanol). The specimen was eluted with buffer B at a rate of 3 ml/h and 4 ml fractions were collected and analysed within 12 h for ATPase activity. In some experiments, an aliquot of the cell lysate supernatant was dialysed against buffer B for 24-38 h and then tested for ATPase activity in an attempt to estimate the recovery of ATPase activity in the column fractions. Extraction ofplatelet myosin. Outdated platelets (3-10 d old), were kindly supplied by The New York Blood Center. The platelets were pooled, centrifuged at 300 g at room temperature, washed twice with 0.9% NaCl, 0.3% Na citrate, pH 7.2, and stored at - 70°C for up to 3 weeks. Myosin was extracted from 50 g of thawed platelets by the method of Pollard et al (1974). Rabbit crterine actomyosin. Rabbit uterine actomyosin was prepared as described previously (Becker & Nachman, 1973). ATPase assay. Myosin ATPase is active in the presence of potassium and either calcium or EDTA but is inactive in the presence of magnesium (Seidel, 1969). In order to define a myosin-like ATPase activity in the endothelial cell preparation, the assay describcd by Adelstein et al (1971) was used. A 0.25-0.5 ml aliquot of the chromatography fraction was incubated at 37°C for 30 niin in 2 ml containing 2 mM ATP (Sigma Chemical Co.), 10 mM imidazole-HC1, pH 7 (J. T. Baker Chemical Co., Phillipsburg, NJ.), 0.6 M KCl, and either 2 mM EDTA, 10 n m CaCl, or 5 niM MgCl,. The ATP solutions and the column samples were incubated alone as controls. An inorganic phosphorus standard was included in each assay, The amount of inorganic phosphorus at the beginning and at the end of the
Myosin in Endothelial Cells 73 incubation period was measured by a modification of the technique of Martin & Doty (1949).The results are expressed as ATPase units where I unit is that amount of enzyme that released I pmole of inorganic phosphorus per minute. Protcin determination. All samples for protein determination were diluted with 0.15 M NaCl, precipitated with a final concentration of 10% trichloroacetic acid, and the precipitate solubilized with I N NaOH for I h. The protein was measured by the method of Lowry using bovine serum albumin as a standard (Lowry et al, 1951). Immunization. Antibody to endothelial cell myosin was prepared by injecting a rabbit with a total of 350 pg of protein which had been dialysed against water and lyophilized. The protein was reconstituted with 0.9% NaCl, mixed with an equal volume of complete Freund’s adjuvant, and injected into footpads on two occasions 2 weeks apart. Two weeks later, the animal was boosted with an intramuscular injection of the antigen and this was repeated I month later. The antiserum was harvested 2 weeks after the last injection. Antibody to platelet myosin was prepared in a similar manner using a total of 360 pg of protein. Immunodifusion tests. These were performed in gels composed of 1.5% Agarose (Bio-Rad Laboratories, Richmond, California) in 0.6 M KC1 buffered to pH 7.4 with 0.02 M TrisHCI. Wells were filled several times and allowed to diffuse for 7-10 d at 4°C. The plates were then washed with 0.6 M KCl for 5 d, distilled water for 3 d, and 50% alcohol for I d, dried at 37°C and stained with Amido Black. The amount of antigen used was I mg protein/ ml in 0.6 M KCl. ImmtrnofIuoresccnce microscopy. Tissues were prepared for indirect immunofluorescence microscopy and the microscopy and photography performed as described previously (Jaffe et a!, 1973). Preparation of antisera for immunopuoresccnce microscopy. The antisera were routinely absorbed with group A, Rh positive human red blood cell stroma, and with platelet poor plasma. For these and subsequent absorptions, the concentration of absorbing protein was I mg/ml. Sodium dodecyl sulphate (SDS) polyacrylamide gel electrophoresis. SDS polyacrylamide gel electrophoresis (5% gels) was performed as described by Weber & Osborn (1975). Samples for analysis were added to an equal volume of a solution containing 8.5 M urea, 2% SDS, and 14 mM DTT and boiled for 5 min. The gels were stained with Coomassie Brilliant Blue. Densitometric scans of gels were carried out in a Gilford Model 240 spectrophotometer equipped with a gel scanning attachment (Gilford Instrument Laboratories, Inc., Oberlin, Ohio) and a Densicord recorder equipped with an integrator (Photovolt Corp., New York). The molecular weight of myosin was determined by SDS polyacrylamide gel electrophoresis as described (Weber & Osborn, 1975). Molecular weight markers included : phosphorylase A, mol wt 94 ooo (Sigma Chemical Co.); ovalbumin, mol wt 43 ooo (Pharmacia) ; chymotrypsin, mol wt 25 000 (Calbiochem, San Diego, California) ; and az-macroglobulin subunit, mol wt 185 ooo (kindly provided by Dr Peter Harpel). RESULTS Endothelial Cell Myosin A concentrated aliquot of a 0.6 M KCl extract of cultured endothelial cells was chromato-
Anne Moore et a/
74
Volume (ml)
FIG I. Sepharose 4B chromatograph of a 0.6 M KCl extract of cultured human endothclial cells. The myosin-like ATPase activity is localized to the void volunic. 0 , Ca2 ATPase; A, EDTA ATPase; m, Mg2 ATPase; - - - -, protein (% transmission at 280 nm).
TABLE I. Purification of myosin from cultured endothelial cells Specific ATPase activity? Sample* Extract supernatant Purified myosin
Ca2+
EDTA
0.016
0.0028
0.072
0.054
Mg2+ i0.01
0.014
* Endothelial cells from 33 Petri dishes were extracted with Buffer A yielding X I nig protein. The extract supernatant was divided and 5.5 mg was dialysed against Buffer B and assayed for ATPase activity. The remaining 5.5 mg was chromatographed on Sepharose 4B, eluted with Buffer B, and the column eluate assayed for ATPase activity. The purified myosin (600 pg protein) refers to the void volume peak where the ATPase activity was localized. t pmoles of inorganic phosphorus released/mg protein /min. graphed on a Sepharose 4B column and thecolumn eluate was monitored for ATPase activity. The calcium and EDTA-stimulated, magnesium-inhibited ATPasc activity was localized to the void volume. The chromatograph of a typical preparation is shown in Fig I. The predominant protein in the void volume peak as analysed by SDS polyacrylamide gel electrophoresis,had a molecular weight of 200 ooo and migrated to the same position on the gel as rabbit uterine myosin heavy chain (Fig 2). The void volume protein also contained a small amount of material of molecular weight 42 000,presumably endothelial cell actin.
Myosin in Endothelial Cells
75
The specific ATPase activities of the endothelial cell myosin before and after column chromatography were compared. Table I shows the ATPase specific activities obtained during a typical purification of endothelial cell myosin. There was a 4+-f0ld increase in calcium ATPase activity and a 20-fold increase in EDTA ATPase activity during the purification procedure.
Platelet Myosin For comparative purposes, human platelet myosin was partially purified. Platelet extract was chromatographed on a 4% agarose column (Bio-Gel A-r sm). The myosin-like ATPase activity was localized to the 200-275 ml region of the column (Fig 3). The ATPase specific activity of one platelet myosin preparation is shown in Table 11. The results are similar to those reported by Pollard et a1 (1974). The 200-275 ml fraction, analysed by SDS polyacrylamide gel electrophoresis,contained a 200 ooo molecular weight protein which migrated
Volume (mi)
FIG 3. Bio-Gel A-ISm chromatograph of platelet extract prepared as in text. The myosin-like ATPase activity is localized to the 200-275 ml fractions. 0 , C a z + ATPase; A, EDTA ATPase, W, MgZ ATPase; - - -,protein (% transmission at 280 nm).
-
TABLE 11. Platelet myosin ATPase activity Specific ATPase activityt Sample*
Platelet myosin
Caz+
EDTA
Mgzt
0.83
0.84
Myosin in Cultured Human Endothelial Cells ANNEMOORE,ERICA. JAFFE,CARLG. BECKER AND RALPH L. NACHMAN
Division of Haematology, Department of Medicine and Department of Pathology, Corne11 University Medical College, N e w York (Received I ]une 1976; accepted for publication ~ o ] u n e1976) SUMMARY. Myosin was isolated from cultured human endothelial cells by extraction with 0.6 M KC1 and chromatography on Sepharose 4B. The extracted endothelial cell protein was identified as myosin by the characteristic ATPase profile, that is, the ATPase was activated by Ca2 and EDTA and inhibited by MgZ +. On sodium dodecyl sulphate polyacrylamide gel el'ectrophoresis, the endothelial ccll myosin heavy chain migrated with a molecular weight of 200 ooo as did rabbit uterine and human platelet myosin heavy chains. A crude preparation of the cndothelial cell myosin reacted immunologically with an antiserum to platelet myosin, a smooth muscle type of myosin. In indirect immunofluorescence studies, antiserum to the purified endothelial cell myosin stained cultured endothelial cells in a fibrillar pattern. The fibrillar pattern was more intense when the endothelial cells were stained with antiserum to platelet myosin. The presence of myosin in the endothelial cell provides a basis for the contractility of these cells. This contractile property may plan an important role in the physiologic function of these cells. +
Contractile proteins have been found in many non-muscle cells and may be important for cell motility, cell division, secretion, and phagocytosis (Pollard & Weihing, 1974; Adelstein, 1975).Human endothelial cells both in vivo and in vitro have been shown by indirect methods to contain contractile proteins which are antigenically similar to platelet and uterine actomyosin (Becker & Nachman, 1973 ; Jaffe et a!, 1973). Endothelial cell contractility might play a role in such physiologic processes as the regulation of vascular permeability, the inflammatory response and the initiation of thrombosis. In order to study the contractile mechanisms of these cells, we have isolated and characterized endothelial cell myosin. The purification of this protein was made possible by the recent development of two techniques. First, pure human endothelial cells free of contaminating smooth muscle cells or fibroblasts have become available in culture (Jaffe et af, 1973). Second, a method for extracting myosin from a small amount of cell protein has been developed (Ostlund et al, 1974). In this report we describe the partial purification and characterization of myosin from cultured human umbilical vein endothelial cells. Correspondence: Dr Anne Moore, Division of Hematology, Department of Medicine, Cornell University Medical College, $25 East 68th Street, New York, N.Y. 10021, U.S.A.
72
A n n e Moore et al MATERIALS AND METHODS
Endothelid cell culture. Endothelial cells were obtained from human umbilical cord veins and cultured as previously described (Jaffe et al, 1973). Extraction of endothelial cell myosin. The method described by Ostlund et a! (1974) was used. All procedures were performed at 0-4"C. Endothelial cells grown to confluence were washed three times with phosphate buffered saline (0.145 M NaCI, 0.01M phosphate buffer, pH 7.4) containing 2 mM dithiothreitol (DTT) (Sigma Chemical Co., St Louis, Missouri) and once with 0.15M NaCl-2 I I ~ MDTT containing 0.5% bovine serum albumin (Miles Laboratories, Inc., Kankakee, Illinois) added to facilitate the uptake of the cells. The cells were removed with a rubber policeman and sedimented by centrifugation at 700 g for 15 min. The cell pellet was washed twice in 0.15 M NaC1-2 mM DTT by centrifugation at 700 g for 5 min and then frozen at - 70°C. The cell pellet was used usually within 2 or 3 weeks but equivalent results were found with pellets frozen up to 5 months. In a typical experiment, cells from approximately 40 Petri dishes ( IOO x 20 mm) or the equivalent number of cells from T-75 flasks or 2-litre roller bottles were used. This was equivalent to 30-40 mg of total cell protein. To extract myosin, the cell pellets were thawed, resuspended in buffer A (0.6 M KCl, 1 5 mM Tris-HC1, pH 7.5 containing I-butanol (3%, v/v) and 2 mM DTT) and stirred intermittently for I h. The suspension was then centrifuged at 12 ooog for 10 min. The cell lysate supernatant, concentrated to I ml in dialysis tubing packed in dry Sephadex G-zoo gel (Pharmacia Fine Chemicals, Inc., Piscataway, N.J.), was applied to a column (0.9 x 60 cni) of Sepharose 4B (Pharmacia) which had been washed with buffer B (buffer A without butanol). The specimen was eluted with buffer B at a rate of 3 ml/h and 4 ml fractions were collected and analysed within 12 h for ATPase activity. In some experiments, an aliquot of the cell lysate supernatant was dialysed against buffer B for 24-38 h and then tested for ATPase activity in an attempt to estimate the recovery of ATPase activity in the column fractions. Extraction ofplatelet myosin. Outdated platelets (3-10 d old), were kindly supplied by The New York Blood Center. The platelets were pooled, centrifuged at 300 g at room temperature, washed twice with 0.9% NaCl, 0.3% Na citrate, pH 7.2, and stored at - 70°C for up to 3 weeks. Myosin was extracted from 50 g of thawed platelets by the method of Pollard et al (1974). Rabbit crterine actomyosin. Rabbit uterine actomyosin was prepared as described previously (Becker & Nachman, 1973). ATPase assay. Myosin ATPase is active in the presence of potassium and either calcium or EDTA but is inactive in the presence of magnesium (Seidel, 1969). In order to define a myosin-like ATPase activity in the endothelial cell preparation, the assay describcd by Adelstein et al (1971) was used. A 0.25-0.5 ml aliquot of the chromatography fraction was incubated at 37°C for 30 niin in 2 ml containing 2 mM ATP (Sigma Chemical Co.), 10 mM imidazole-HC1, pH 7 (J. T. Baker Chemical Co., Phillipsburg, NJ.), 0.6 M KCl, and either 2 mM EDTA, 10 n m CaCl, or 5 niM MgCl,. The ATP solutions and the column samples were incubated alone as controls. An inorganic phosphorus standard was included in each assay, The amount of inorganic phosphorus at the beginning and at the end of the
Myosin in Endothelial Cells 73 incubation period was measured by a modification of the technique of Martin & Doty (1949).The results are expressed as ATPase units where I unit is that amount of enzyme that released I pmole of inorganic phosphorus per minute. Protcin determination. All samples for protein determination were diluted with 0.15 M NaCl, precipitated with a final concentration of 10% trichloroacetic acid, and the precipitate solubilized with I N NaOH for I h. The protein was measured by the method of Lowry using bovine serum albumin as a standard (Lowry et al, 1951). Immunization. Antibody to endothelial cell myosin was prepared by injecting a rabbit with a total of 350 pg of protein which had been dialysed against water and lyophilized. The protein was reconstituted with 0.9% NaCl, mixed with an equal volume of complete Freund’s adjuvant, and injected into footpads on two occasions 2 weeks apart. Two weeks later, the animal was boosted with an intramuscular injection of the antigen and this was repeated I month later. The antiserum was harvested 2 weeks after the last injection. Antibody to platelet myosin was prepared in a similar manner using a total of 360 pg of protein. Immunodifusion tests. These were performed in gels composed of 1.5% Agarose (Bio-Rad Laboratories, Richmond, California) in 0.6 M KC1 buffered to pH 7.4 with 0.02 M TrisHCI. Wells were filled several times and allowed to diffuse for 7-10 d at 4°C. The plates were then washed with 0.6 M KCl for 5 d, distilled water for 3 d, and 50% alcohol for I d, dried at 37°C and stained with Amido Black. The amount of antigen used was I mg protein/ ml in 0.6 M KCl. ImmtrnofIuoresccnce microscopy. Tissues were prepared for indirect immunofluorescence microscopy and the microscopy and photography performed as described previously (Jaffe et a!, 1973). Preparation of antisera for immunopuoresccnce microscopy. The antisera were routinely absorbed with group A, Rh positive human red blood cell stroma, and with platelet poor plasma. For these and subsequent absorptions, the concentration of absorbing protein was I mg/ml. Sodium dodecyl sulphate (SDS) polyacrylamide gel electrophoresis. SDS polyacrylamide gel electrophoresis (5% gels) was performed as described by Weber & Osborn (1975). Samples for analysis were added to an equal volume of a solution containing 8.5 M urea, 2% SDS, and 14 mM DTT and boiled for 5 min. The gels were stained with Coomassie Brilliant Blue. Densitometric scans of gels were carried out in a Gilford Model 240 spectrophotometer equipped with a gel scanning attachment (Gilford Instrument Laboratories, Inc., Oberlin, Ohio) and a Densicord recorder equipped with an integrator (Photovolt Corp., New York). The molecular weight of myosin was determined by SDS polyacrylamide gel electrophoresis as described (Weber & Osborn, 1975). Molecular weight markers included : phosphorylase A, mol wt 94 ooo (Sigma Chemical Co.); ovalbumin, mol wt 43 ooo (Pharmacia) ; chymotrypsin, mol wt 25 000 (Calbiochem, San Diego, California) ; and az-macroglobulin subunit, mol wt 185 ooo (kindly provided by Dr Peter Harpel). RESULTS Endothelial Cell Myosin A concentrated aliquot of a 0.6 M KCl extract of cultured endothelial cells was chromato-
Anne Moore et a/
74
Volume (ml)
FIG I. Sepharose 4B chromatograph of a 0.6 M KCl extract of cultured human endothclial cells. The myosin-like ATPase activity is localized to the void volunic. 0 , Ca2 ATPase; A, EDTA ATPase; m, Mg2 ATPase; - - - -, protein (% transmission at 280 nm).
TABLE I. Purification of myosin from cultured endothelial cells Specific ATPase activity? Sample* Extract supernatant Purified myosin
Ca2+
EDTA
0.016
0.0028
0.072
0.054
Mg2+ i0.01
0.014
* Endothelial cells from 33 Petri dishes were extracted with Buffer A yielding X I nig protein. The extract supernatant was divided and 5.5 mg was dialysed against Buffer B and assayed for ATPase activity. The remaining 5.5 mg was chromatographed on Sepharose 4B, eluted with Buffer B, and the column eluate assayed for ATPase activity. The purified myosin (600 pg protein) refers to the void volume peak where the ATPase activity was localized. t pmoles of inorganic phosphorus released/mg protein /min. graphed on a Sepharose 4B column and thecolumn eluate was monitored for ATPase activity. The calcium and EDTA-stimulated, magnesium-inhibited ATPasc activity was localized to the void volume. The chromatograph of a typical preparation is shown in Fig I. The predominant protein in the void volume peak as analysed by SDS polyacrylamide gel electrophoresis,had a molecular weight of 200 ooo and migrated to the same position on the gel as rabbit uterine myosin heavy chain (Fig 2). The void volume protein also contained a small amount of material of molecular weight 42 000,presumably endothelial cell actin.
Myosin in Endothelial Cells
75
The specific ATPase activities of the endothelial cell myosin before and after column chromatography were compared. Table I shows the ATPase specific activities obtained during a typical purification of endothelial cell myosin. There was a 4+-f0ld increase in calcium ATPase activity and a 20-fold increase in EDTA ATPase activity during the purification procedure.
Platelet Myosin For comparative purposes, human platelet myosin was partially purified. Platelet extract was chromatographed on a 4% agarose column (Bio-Gel A-r sm). The myosin-like ATPase activity was localized to the 200-275 ml region of the column (Fig 3). The ATPase specific activity of one platelet myosin preparation is shown in Table 11. The results are similar to those reported by Pollard et a1 (1974). The 200-275 ml fraction, analysed by SDS polyacrylamide gel electrophoresis,contained a 200 ooo molecular weight protein which migrated
Volume (mi)
FIG 3. Bio-Gel A-ISm chromatograph of platelet extract prepared as in text. The myosin-like ATPase activity is localized to the 200-275 ml fractions. 0 , C a z + ATPase; A, EDTA ATPase, W, MgZ ATPase; - - -,protein (% transmission at 280 nm).
-
TABLE 11. Platelet myosin ATPase activity Specific ATPase activityt Sample*
Platelet myosin
Caz+
EDTA
Mgzt
0.83
0.84
