Carrinogenesis vol.12 no. 10 pp 1791-1794, 1991
The effect of diacylglycerols on fibronectin release and its reversal by retinoic acid in cell culture
Helmut Scheidl, Trissia Chandra, Bernhard Gmeiner1, Gerold Zerlauth2, Giorgio Scita and George Wolf3 Department of Nutritional Sciences, University of California, Berkeley, CA 94720, USA, 'institute for Medical Chemistry, University of Vienna, A-1090 Austria and 2Immuno AG, Vienna, A-1220 Austria 'To whom correspondence should be addressed
Introduction Cells in culture respond in a great variety of ways to tumor promoters (1). Many of the early responses seem to be plasma membrane-associated effects. One of the most striking and rapid responses to the tumor promoter 12-0-tetradecanoylphorbol13-acetate (TPA*) is the stimulation of the release of fibronectin (FN) from the cell surface of cultured cells into the medium (2,3). By the use of a line of carcinogen-initiated but not promoted mouse epidermal cells (BALB/c), work from our laboratory demonstrated that stimulation of the release of FN by the tumor promoters TPA and mezerein appeared to be related to tumor promotion, and suggested that the increased release of FN may be a necessary (but not sufficient) event in tumor promotion (4). In view of these results, it was of importance to determine if diacylglycerols (DAGs), the endogenous ligands of the TPA receptor, would also stimulate the release of FN from the cell surface. Calcium-activated, phospholipid-dependent protein kinase C (PKC) acts as the receptor for phorbol esters which substitute for DAG, the endogenous PKC activator, produced in cell membranes as a result of phospholipase C action. Thus, the actions of tumor promoters appear to be mediated by PKC (5). It is significant in that respect that DAG was found to be •Abbreviations: TPA, 12-O-tetradecanoylphortiol-13-acetate; FN, fibronectin; DAG, diacylglycerol; PKC, calcium-activated phospholipid-dependent protein kinase; D1C8, 1,2-dioctanoyl-sn-glycerol; OAG, l-oleoyl-2-acctyl-i/i-glycerol; RA, retinoic acid; HLF, human lung fibroblasts; MEM, minimum essential medium, DiC18, 1,2-dioleoyl-OT-glycerol; PKA, cAMP-dependent protein kinase. © Oxford University Press
Materials and methods Cell cultures Human lung fibroblasts (HLF) (CCL 137) were obtained from the American Type Culture Collection at passage 7 (Rockville, MD). Frozen cells were thawed and the cell suspension was diluted with an appropriate volume of Eagle's minimum essential medium (MEM; Gibco Laboratories, Grand Island, NY), supplemented with 0.1% lactalbumin hydrolysate (Sigma Chemical Co., St Louis, MO), 1 mM sodium pyruvate, 100 U/ml penicillin, 100 /ig/ml streptomycin and 10% fetal bovine serum (all from Gibco). The cells were inoculated into 84 cm2 tissue culture flasks (Coming Glassware, Coming, NY) and incubated at 37°C in 95% air/5% COj. For experiments, cells between passages 9 and 14 were used. Confluent cells from the flasks were then inoculated at a density of 1.0 x 106 cells/dish onto 60 mm diameter culture dishes (Coming), and grown in the supplemented MEM and under the conditions described above. After 20 h, the medium was removed and the cell layer washed three times with PBS. For the concentration dependence experiments, the cells were incubated with 3 ml of different concentrations of treatment media (see below) for 2 h. For the time dependence experiments, incubations were with 5 x 10~4 M DiC8. Following incubation, the treatment medium was collected and 20 /ii of 1 mM PMSF (Sigma) was added to inhibit proteases. The medium was centrifuged and FN assayed on an aliquot by ELISA (12). The cell layer remaining on the dish was incubated with 2 ml 0.25% trypsin solution for 5 min. Released cells were counted by a hemocytometer. In some experiments (where indicated) PKC activities and FN release were reported on a cell protein basis. For this, cells remaining on dishes were dissolved in 0.1 M NaOH plus 2% SDS and incubated for 30 min
1791
Downloaded from http://carcin.oxfordjournals.org/ at McMaster University Library on June 30, 2015
Previous work from our laboratory showed that tumor promoters such as phorfool ester (TPA) stimulated the release of fibronectin (FN) from the surface of several cell types in culture, and that this stimulation was counteracted by retinoic acid. Diacylglycerols (DAGs) are the endogenous ligands of the TPA receptor and can activate and translocate protein kinase C (PKC) in a manner similar to TPA. To show that the release of FN is related to activation of PKC, we tested the action of DAGs on FN release from human lung fibroblasts and its counteraction by retinoic acid. We found that DAGs stimulated the release of FN in a concentrationand time-dependent manner. The stimulation of the release of FN correlated with the translocation-activation of PKC by DAG. Retinoic acid reversed the action of DAG with respect to stimulation of FN release and inhibited this release even in the absence of DAG. These results suggest that the release of FN is in some way related to translocation — activation of PKC.
a complete tumor promoter in mouse skin (6). PKC can be activated by the endogenous generation of DAG as well as by the addition of synthetic DAG analogs or tumor promoters such as phorbol esters (5). When activated, PKC is translocated from the cytosol to the membrane compartment (7). The enzyme then phosphorylates cellular proteins, resulting in subsequent responses such as growth regulation, cell proliferation, modulation of gene expression, tumor promotion and probably oncogenesis (8). Various synthetic DAGs, such as 1,2-dioctanoyl-.rn-glycerol (DiC8) and l-oleoyl-2-acetyl-s/i-glycerol (OAG), have been reported to activate PKC in vitro (9). Boni and Rando (10) have demonstrated that only the physiologically relevant 1,2-sndiacylglycerols are capable of activating the enzyme in vitro. Neither 2,3-sn-DAG nor 1,3-sn-DAG is effective in activating the kinase. Phorbol esters and DAGs seem to have the same binding site on PKC, leading to the activation of the kinase through a similar mechanism (11). However, there are some differences between phorbol esters and DAG as kinase activators. Since phorbol ester is not readily metabolized, it remains active for a period of time once it is intercalated into the membranes. By comparison, the endogenous kinase activator exists transiently after the breakdown of phosphatidylinositol to DAG, and is rapidly degraded (5). Phorbol esters are much more potent than DAGs as PKC activators. To demonstrate that the release of FN from the cell surface of cells in culture is related to activation of PKC, we tested the effect of DAGs on FN release and correlated this to the translocation-activation of PKC. Since in earlier work we found that retinoic acid (RA) inhibited the TPA-induced stimulation of FN release (3), then the same effect of RA should follow DAG treatment.
H.ScheidJ et at. at room temperature. The protein was then precipitated with 10% TCA. The protein in the precipitate was determined by the Pierce reagent (Pierce Co., Rockford, IL), a modified Lowry procedure. Treatment medium Serum-free treatment medium was prepared from serum-free MEM containing 0.1 mg/ml of BSA (Sigma), 1 mM sodium pyruvate, 100 U/ml penicillin and 100 jig/ml streptomycin. Stock solutions of DiC8, 1,2-dk4eoyl-OT-glycerol (DiC18) and OAG (Sigma) were prepared in spectra] grade DMSO (Burdick and Jackson, Muskegon, Ml) and stored at -20°C. The stock solutions were diluted into the BSA media at the time of use, with never a greater concentration of DMSO than 0.1 %. For the RA experiment, RA (from Eastman Kodak Co., Rochester, NY) (added in DMSO) at final concentrations indicated, was preincubated for 1 h at 37°C (as described in rcf. 13) in the treatment medium (serum-free MEM) described by Zerlauth and Wolf (3), except that BSA was at a concentration of 0.01% with 2 mM glutamine. The RA used was checked for purity by HPLC routinely before each incubation.
Results Figures 1 and 2 show that both DiC8 and DiC18 stimulated release of FN into the medium of HLF in a concentrationdependent manner. The extent of stimulation over the control release of FN is quite similar to that observed with TPA (3), namely 1.5 — 2 times the control rate (i.e. in the absence of promoter), within 2 h. With DiC8 the highest level of stimulation was with a very high concentration, 5 x 10~4 M. DiC8 forms a sufficiently stable emulsion which allows it to be easily delivered to cells. It has sufficient hydrophobic character enabling it to be cell permeable (17). This synthetic compound has been used extensively as a model for endogenous DAG studies on tumor promotion and PKC activation (11). DiC8 is hydrolyzed rapidly, thus high concentrations ranging from 5 x 10~6 to 5 x 10~4 M had to be used in this study to obtain effective action (11). That the DiC18 was more active at a much lower concentration (Figure 2) may be due to the much greater lipophilic character of this compound, allowing it to penetrate into the plasma membrane more easily and more extensively or both. On the other hand, at a higher concentration (10~6 M), it was inactive, possibly because of the difficulty of forming stable emulsions. Time dependence is shown in Figure 3, with a 2 h incubation giving a large increase in FN release. The increase from 2 to 3 h was smaller, possibly because of destruction of the DAG after that time (11). No stimulation was observed with OAG. Figure 4 shows the known biologic activity of DAG, translocating PKC from cytosol to membrane, in conjunction with 1792
d
800
o z o
600
tii
-I
-
400
z
200 0 5 \O~* 5 IO" S 5I0"4 CONCENTRATION OF DIC8 ( M )
Fig. 1. Effect of 1,2-dioctanoyl-sn-glycerol on fibronectin release from human lung fibroblasts. Incubations were for 2 h with conditions as described in Materials and methods. Each bar represents the mean ± SE of four different experiments Each experiment was done with triplicate dishes of cells. A significant (P < 0.05) difference was found only between the 5 x 10~4 M and control groups.
m UJ
o
1000 800
O 600 5 400 200
0 IO" H IO-t IO" T IO" # CONCENTRATION DICI8(M) Fig. 2. Effect of 1,2-dioleoyl-sn-glycerol on fibronectin release from human lung fibroblasts. The bars of 0 and IO~7 M groups represent the means ± SE of three different experiments, while those of 1 0 " " , 10" 9 and 10~6 M groups represent the average and separate values of two different experiments. (Each experiment was done with triplicate dishes of cells.) Significant difference (P < 0.05) between the 10" 7 M and the control group.
1000 v>
l! 800
CONTROL
•
DiC8
LU
o
2 O
600
_i
d 400 2
i
z 200
I INCUBATION
2 TIME
3 (hours)
Fig. 3. Effect of incubation time on fibronectin release from human lung fibroblasts upon treatment with 5 x IO~5 M of 1,2-dioctanoyl-jn-glycerol as described in Materials and methods. Each bar represents the mean ± SE of three different experiments. (Each experiment was done with mplicate dishes of cells.) Statistically significant difference (P < 0.05) between 2 h treatment and control.
Downloaded from http://carcin.oxfordjournals.org/ at McMaster University Library on June 30, 2015
PKC assay Cells were inoculated at a density of 1.5 x 106 cells/flask into 150 cm2 tissue culture flasks and grown in supplemented MEM under 95% air/5% CO2. Upon reaching confluence, the medium was removed and the cell layer was washed three times with PBS. The cells were then incubated with 10 ml treatment medium. Membrane and cytosol fractions for PKC assays were isolated and purified from the cells by a procedure modified from that of Thomas et al. (14) and Smith and Colbum (15) as follows. Following incubation, the medium was removed and the cells were scraped into homogenization buffer A. Buffer A was 20 mM Tris-HCl, pH 7.5, containing 2 mM EDTA, 0.5 mM ethyleneglycol bis(/3-aminoethylether)-MM/V:,A''-tetraacetic acid, 2 mM PMSF, 25 n%lm\ leupeptin and 0.33 M sucrose (all from Sigma). Then the cell suspension was centnfuged at 800 g for 5 min at 4°C. The cell pellet was resuspended in 1 5 ml buffer A. The suspension was then homogenized by sonication on ice with three 10 s bursts interspersed by 10 s pauses. Aliquots of this preparation were taken for protein determination. The homogenate was then centnfuged at 15 000 g for 10 nun at 4"C to separate the supernatant (i.e. cytosolk) fraction from the membrane fraction. The latter was resuspended in 1 ml buffer A containing 1 % Triton X-100. After letting it stand at 4°C for 1 h, the suspension was centnfuged at 15 000 g for 10 min to obtain membrane-associated, detergent-soluble PKC in this supernatant fraction. PKC was assayed by the method of Hannun et al. (16), using a kit obtained from Amersham Corp. (Arlington Heights, IL).
1000
Effect of diacytglycerob on fibronectin release
1500
30
z
ui I-
o
(T Q.
20
1000
UJ O
500
10
0
I INCUBATION TIME (hours)
z tu
B
o cr a
z z \
20
to UJ
o o y
10
DAG
0AG/RA
Fig. 5. Effect of dioctanoyl-OT-glycerol on the release of fibronectin and its counteraction by retinoic acid. Human lung fibroblasts were grown on 60 mm diameter Petri dishes in MEM with 10% fetal calf serum and 2 mM glutamine. At subconfluency, this medium was replaced by 2 ml treatment medium: control medium, MEM with 0.01% BSA, 2 mM glutamine, 0 02% DMSO; treatment medium, additionally 10~ 4 M dioctanoyl-snglycerol. Every 30 min, 40 y\ of 10~3 M DAG-containing medium was added, over a total of 2.5 h. The RA-containing medium contained additionally 2.2 x 10~ 6 M RA, added in DMSO (final DMSO concentration, 0.02%). The medium was preincubated for 1 h at 37°C with RA, in the dark. FN in the media was determined by ELISA (3). Values are averages of results from two separate incubations varying by < 5 % from each other.
100
I INCUBATION TIME
(hours)
Fig. 4. Effect of dioctanoyl-OT-glycerol on the release of fibronectin and PKC activity as a function of time Human lung fibroblasts were grown to confluency in 150 cm2 cell culture flasks as described in Materials and methods. After washing, the cells were incubated with 10 ml of 10~4 M DAG in MEM (with 0.01 % BSA and 2 mM glutamine) as a loading dose or with MEM/BSA/glutamine. Every 30 min, 200 n\ of 10" 3 M DAG in the same medium (or medium alone) was added as maintenance dose. Fibronectin was assayed, and cytosolic and membrane fractions were prepared and assayed for PKC as described in Materials and methods. (A) Open circles FN in medium of cells treated with DAG; open squares, FN in medium without DAG. The values are averages of three incubations which differed from each other by < 5 % . (B) Open triangles. PKC activity in cytosol; full triangles, PKC activity in membranes. The values are averages of duplicate assays from one incubation which differed from each other by < 5 % .
the stimulation of the release of FN, similar to that which we previously observed with phorbol ester (3). Since previous work from our laboratory (3) demonstrated that RA inhibited the TPA-stimulated release of FN, we tested the action of RA on FN release by DAG; as shown in Figure 5, RA does indeed inhibit such FN release. It is of interest to note that, even in the absence of DAG (Figure 6), RA inhibited the release of FN, though effectively only at 10~6 M RA and not at lower concentrations. Discussion Earlier work from our laboratory showed that tumor promoters such as TPA or mezerein (3,4) stimulated the release of FN from the surface of HLF. This suggested that FN release may be related to activation of PKC; in that case, DAG, the endogenous ligand of the TPA receptor, should also stimulate the release of FN from these cells. This effect was indeed observed both with DiC8 and DiC18 in a time- and concentration-dependent manner. As expected, activation by translocation of PKC accompanied the
x
The effect of diacylglycerols on fibronectin release and its reversal by retinoic acid in cell culture
Helmut Scheidl, Trissia Chandra, Bernhard Gmeiner1, Gerold Zerlauth2, Giorgio Scita and George Wolf3 Department of Nutritional Sciences, University of California, Berkeley, CA 94720, USA, 'institute for Medical Chemistry, University of Vienna, A-1090 Austria and 2Immuno AG, Vienna, A-1220 Austria 'To whom correspondence should be addressed
Introduction Cells in culture respond in a great variety of ways to tumor promoters (1). Many of the early responses seem to be plasma membrane-associated effects. One of the most striking and rapid responses to the tumor promoter 12-0-tetradecanoylphorbol13-acetate (TPA*) is the stimulation of the release of fibronectin (FN) from the cell surface of cultured cells into the medium (2,3). By the use of a line of carcinogen-initiated but not promoted mouse epidermal cells (BALB/c), work from our laboratory demonstrated that stimulation of the release of FN by the tumor promoters TPA and mezerein appeared to be related to tumor promotion, and suggested that the increased release of FN may be a necessary (but not sufficient) event in tumor promotion (4). In view of these results, it was of importance to determine if diacylglycerols (DAGs), the endogenous ligands of the TPA receptor, would also stimulate the release of FN from the cell surface. Calcium-activated, phospholipid-dependent protein kinase C (PKC) acts as the receptor for phorbol esters which substitute for DAG, the endogenous PKC activator, produced in cell membranes as a result of phospholipase C action. Thus, the actions of tumor promoters appear to be mediated by PKC (5). It is significant in that respect that DAG was found to be •Abbreviations: TPA, 12-O-tetradecanoylphortiol-13-acetate; FN, fibronectin; DAG, diacylglycerol; PKC, calcium-activated phospholipid-dependent protein kinase; D1C8, 1,2-dioctanoyl-sn-glycerol; OAG, l-oleoyl-2-acctyl-i/i-glycerol; RA, retinoic acid; HLF, human lung fibroblasts; MEM, minimum essential medium, DiC18, 1,2-dioleoyl-OT-glycerol; PKA, cAMP-dependent protein kinase. © Oxford University Press
Materials and methods Cell cultures Human lung fibroblasts (HLF) (CCL 137) were obtained from the American Type Culture Collection at passage 7 (Rockville, MD). Frozen cells were thawed and the cell suspension was diluted with an appropriate volume of Eagle's minimum essential medium (MEM; Gibco Laboratories, Grand Island, NY), supplemented with 0.1% lactalbumin hydrolysate (Sigma Chemical Co., St Louis, MO), 1 mM sodium pyruvate, 100 U/ml penicillin, 100 /ig/ml streptomycin and 10% fetal bovine serum (all from Gibco). The cells were inoculated into 84 cm2 tissue culture flasks (Coming Glassware, Coming, NY) and incubated at 37°C in 95% air/5% COj. For experiments, cells between passages 9 and 14 were used. Confluent cells from the flasks were then inoculated at a density of 1.0 x 106 cells/dish onto 60 mm diameter culture dishes (Coming), and grown in the supplemented MEM and under the conditions described above. After 20 h, the medium was removed and the cell layer washed three times with PBS. For the concentration dependence experiments, the cells were incubated with 3 ml of different concentrations of treatment media (see below) for 2 h. For the time dependence experiments, incubations were with 5 x 10~4 M DiC8. Following incubation, the treatment medium was collected and 20 /ii of 1 mM PMSF (Sigma) was added to inhibit proteases. The medium was centrifuged and FN assayed on an aliquot by ELISA (12). The cell layer remaining on the dish was incubated with 2 ml 0.25% trypsin solution for 5 min. Released cells were counted by a hemocytometer. In some experiments (where indicated) PKC activities and FN release were reported on a cell protein basis. For this, cells remaining on dishes were dissolved in 0.1 M NaOH plus 2% SDS and incubated for 30 min
1791
Downloaded from http://carcin.oxfordjournals.org/ at McMaster University Library on June 30, 2015
Previous work from our laboratory showed that tumor promoters such as phorfool ester (TPA) stimulated the release of fibronectin (FN) from the surface of several cell types in culture, and that this stimulation was counteracted by retinoic acid. Diacylglycerols (DAGs) are the endogenous ligands of the TPA receptor and can activate and translocate protein kinase C (PKC) in a manner similar to TPA. To show that the release of FN is related to activation of PKC, we tested the action of DAGs on FN release from human lung fibroblasts and its counteraction by retinoic acid. We found that DAGs stimulated the release of FN in a concentrationand time-dependent manner. The stimulation of the release of FN correlated with the translocation-activation of PKC by DAG. Retinoic acid reversed the action of DAG with respect to stimulation of FN release and inhibited this release even in the absence of DAG. These results suggest that the release of FN is in some way related to translocation — activation of PKC.
a complete tumor promoter in mouse skin (6). PKC can be activated by the endogenous generation of DAG as well as by the addition of synthetic DAG analogs or tumor promoters such as phorbol esters (5). When activated, PKC is translocated from the cytosol to the membrane compartment (7). The enzyme then phosphorylates cellular proteins, resulting in subsequent responses such as growth regulation, cell proliferation, modulation of gene expression, tumor promotion and probably oncogenesis (8). Various synthetic DAGs, such as 1,2-dioctanoyl-.rn-glycerol (DiC8) and l-oleoyl-2-acetyl-s/i-glycerol (OAG), have been reported to activate PKC in vitro (9). Boni and Rando (10) have demonstrated that only the physiologically relevant 1,2-sndiacylglycerols are capable of activating the enzyme in vitro. Neither 2,3-sn-DAG nor 1,3-sn-DAG is effective in activating the kinase. Phorbol esters and DAGs seem to have the same binding site on PKC, leading to the activation of the kinase through a similar mechanism (11). However, there are some differences between phorbol esters and DAG as kinase activators. Since phorbol ester is not readily metabolized, it remains active for a period of time once it is intercalated into the membranes. By comparison, the endogenous kinase activator exists transiently after the breakdown of phosphatidylinositol to DAG, and is rapidly degraded (5). Phorbol esters are much more potent than DAGs as PKC activators. To demonstrate that the release of FN from the cell surface of cells in culture is related to activation of PKC, we tested the effect of DAGs on FN release and correlated this to the translocation-activation of PKC. Since in earlier work we found that retinoic acid (RA) inhibited the TPA-induced stimulation of FN release (3), then the same effect of RA should follow DAG treatment.
H.ScheidJ et at. at room temperature. The protein was then precipitated with 10% TCA. The protein in the precipitate was determined by the Pierce reagent (Pierce Co., Rockford, IL), a modified Lowry procedure. Treatment medium Serum-free treatment medium was prepared from serum-free MEM containing 0.1 mg/ml of BSA (Sigma), 1 mM sodium pyruvate, 100 U/ml penicillin and 100 jig/ml streptomycin. Stock solutions of DiC8, 1,2-dk4eoyl-OT-glycerol (DiC18) and OAG (Sigma) were prepared in spectra] grade DMSO (Burdick and Jackson, Muskegon, Ml) and stored at -20°C. The stock solutions were diluted into the BSA media at the time of use, with never a greater concentration of DMSO than 0.1 %. For the RA experiment, RA (from Eastman Kodak Co., Rochester, NY) (added in DMSO) at final concentrations indicated, was preincubated for 1 h at 37°C (as described in rcf. 13) in the treatment medium (serum-free MEM) described by Zerlauth and Wolf (3), except that BSA was at a concentration of 0.01% with 2 mM glutamine. The RA used was checked for purity by HPLC routinely before each incubation.
Results Figures 1 and 2 show that both DiC8 and DiC18 stimulated release of FN into the medium of HLF in a concentrationdependent manner. The extent of stimulation over the control release of FN is quite similar to that observed with TPA (3), namely 1.5 — 2 times the control rate (i.e. in the absence of promoter), within 2 h. With DiC8 the highest level of stimulation was with a very high concentration, 5 x 10~4 M. DiC8 forms a sufficiently stable emulsion which allows it to be easily delivered to cells. It has sufficient hydrophobic character enabling it to be cell permeable (17). This synthetic compound has been used extensively as a model for endogenous DAG studies on tumor promotion and PKC activation (11). DiC8 is hydrolyzed rapidly, thus high concentrations ranging from 5 x 10~6 to 5 x 10~4 M had to be used in this study to obtain effective action (11). That the DiC18 was more active at a much lower concentration (Figure 2) may be due to the much greater lipophilic character of this compound, allowing it to penetrate into the plasma membrane more easily and more extensively or both. On the other hand, at a higher concentration (10~6 M), it was inactive, possibly because of the difficulty of forming stable emulsions. Time dependence is shown in Figure 3, with a 2 h incubation giving a large increase in FN release. The increase from 2 to 3 h was smaller, possibly because of destruction of the DAG after that time (11). No stimulation was observed with OAG. Figure 4 shows the known biologic activity of DAG, translocating PKC from cytosol to membrane, in conjunction with 1792
d
800
o z o
600
tii
-I
-
400
z
200 0 5 \O~* 5 IO" S 5I0"4 CONCENTRATION OF DIC8 ( M )
Fig. 1. Effect of 1,2-dioctanoyl-sn-glycerol on fibronectin release from human lung fibroblasts. Incubations were for 2 h with conditions as described in Materials and methods. Each bar represents the mean ± SE of four different experiments Each experiment was done with triplicate dishes of cells. A significant (P < 0.05) difference was found only between the 5 x 10~4 M and control groups.
m UJ
o
1000 800
O 600 5 400 200
0 IO" H IO-t IO" T IO" # CONCENTRATION DICI8(M) Fig. 2. Effect of 1,2-dioleoyl-sn-glycerol on fibronectin release from human lung fibroblasts. The bars of 0 and IO~7 M groups represent the means ± SE of three different experiments, while those of 1 0 " " , 10" 9 and 10~6 M groups represent the average and separate values of two different experiments. (Each experiment was done with triplicate dishes of cells.) Significant difference (P < 0.05) between the 10" 7 M and the control group.
1000 v>
l! 800
CONTROL
•
DiC8
LU
o
2 O
600
_i
d 400 2
i
z 200
I INCUBATION
2 TIME
3 (hours)
Fig. 3. Effect of incubation time on fibronectin release from human lung fibroblasts upon treatment with 5 x IO~5 M of 1,2-dioctanoyl-jn-glycerol as described in Materials and methods. Each bar represents the mean ± SE of three different experiments. (Each experiment was done with mplicate dishes of cells.) Statistically significant difference (P < 0.05) between 2 h treatment and control.
Downloaded from http://carcin.oxfordjournals.org/ at McMaster University Library on June 30, 2015
PKC assay Cells were inoculated at a density of 1.5 x 106 cells/flask into 150 cm2 tissue culture flasks and grown in supplemented MEM under 95% air/5% CO2. Upon reaching confluence, the medium was removed and the cell layer was washed three times with PBS. The cells were then incubated with 10 ml treatment medium. Membrane and cytosol fractions for PKC assays were isolated and purified from the cells by a procedure modified from that of Thomas et al. (14) and Smith and Colbum (15) as follows. Following incubation, the medium was removed and the cells were scraped into homogenization buffer A. Buffer A was 20 mM Tris-HCl, pH 7.5, containing 2 mM EDTA, 0.5 mM ethyleneglycol bis(/3-aminoethylether)-MM/V:,A''-tetraacetic acid, 2 mM PMSF, 25 n%lm\ leupeptin and 0.33 M sucrose (all from Sigma). Then the cell suspension was centnfuged at 800 g for 5 min at 4°C. The cell pellet was resuspended in 1 5 ml buffer A. The suspension was then homogenized by sonication on ice with three 10 s bursts interspersed by 10 s pauses. Aliquots of this preparation were taken for protein determination. The homogenate was then centnfuged at 15 000 g for 10 nun at 4"C to separate the supernatant (i.e. cytosolk) fraction from the membrane fraction. The latter was resuspended in 1 ml buffer A containing 1 % Triton X-100. After letting it stand at 4°C for 1 h, the suspension was centnfuged at 15 000 g for 10 min to obtain membrane-associated, detergent-soluble PKC in this supernatant fraction. PKC was assayed by the method of Hannun et al. (16), using a kit obtained from Amersham Corp. (Arlington Heights, IL).
1000
Effect of diacytglycerob on fibronectin release
1500
30
z
ui I-
o
(T Q.
20
1000
UJ O
500
10
0
I INCUBATION TIME (hours)
z tu
B
o cr a
z z \
20
to UJ
o o y
10
DAG
0AG/RA
Fig. 5. Effect of dioctanoyl-OT-glycerol on the release of fibronectin and its counteraction by retinoic acid. Human lung fibroblasts were grown on 60 mm diameter Petri dishes in MEM with 10% fetal calf serum and 2 mM glutamine. At subconfluency, this medium was replaced by 2 ml treatment medium: control medium, MEM with 0.01% BSA, 2 mM glutamine, 0 02% DMSO; treatment medium, additionally 10~ 4 M dioctanoyl-snglycerol. Every 30 min, 40 y\ of 10~3 M DAG-containing medium was added, over a total of 2.5 h. The RA-containing medium contained additionally 2.2 x 10~ 6 M RA, added in DMSO (final DMSO concentration, 0.02%). The medium was preincubated for 1 h at 37°C with RA, in the dark. FN in the media was determined by ELISA (3). Values are averages of results from two separate incubations varying by < 5 % from each other.
100
I INCUBATION TIME
(hours)
Fig. 4. Effect of dioctanoyl-OT-glycerol on the release of fibronectin and PKC activity as a function of time Human lung fibroblasts were grown to confluency in 150 cm2 cell culture flasks as described in Materials and methods. After washing, the cells were incubated with 10 ml of 10~4 M DAG in MEM (with 0.01 % BSA and 2 mM glutamine) as a loading dose or with MEM/BSA/glutamine. Every 30 min, 200 n\ of 10" 3 M DAG in the same medium (or medium alone) was added as maintenance dose. Fibronectin was assayed, and cytosolic and membrane fractions were prepared and assayed for PKC as described in Materials and methods. (A) Open circles FN in medium of cells treated with DAG; open squares, FN in medium without DAG. The values are averages of three incubations which differed from each other by < 5 % . (B) Open triangles. PKC activity in cytosol; full triangles, PKC activity in membranes. The values are averages of duplicate assays from one incubation which differed from each other by < 5 % .
the stimulation of the release of FN, similar to that which we previously observed with phorbol ester (3). Since previous work from our laboratory (3) demonstrated that RA inhibited the TPA-stimulated release of FN, we tested the action of RA on FN release by DAG; as shown in Figure 5, RA does indeed inhibit such FN release. It is of interest to note that, even in the absence of DAG (Figure 6), RA inhibited the release of FN, though effectively only at 10~6 M RA and not at lower concentrations. Discussion Earlier work from our laboratory showed that tumor promoters such as TPA or mezerein (3,4) stimulated the release of FN from the surface of HLF. This suggested that FN release may be related to activation of PKC; in that case, DAG, the endogenous ligand of the TPA receptor, should also stimulate the release of FN from these cells. This effect was indeed observed both with DiC8 and DiC18 in a time- and concentration-dependent manner. As expected, activation by translocation of PKC accompanied the
x
