Vol.
166,
January
No. 30,
2, 1990
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
Pages
1990
715-722
CHARACTERIZATION OF PHOSPHATIDYLINOSITOL PHOSPHOLIPASE C ACTIVITY IN HUMAN MELANOMA Frank W. Perrella Medical Products Department E. I. Du Pont de Nemours & Co. 500 South Ridgeway Avenue Glenolden, Pennsylvania 19036 Received
November
3, 1989
SUMMARY: Phosphoinositide phospholipase C activity was investigated in human melanoma grown as solid tumor xenografts in nude mice. The enzyme was dependent on calcium for activity and was stimulated by the detergent deoxycholate. The pH optimum was 5.5 in the absence of detergent, and in the presence of deoxycholate two pH maxima were present, 5.5 and 7.2. Phospholipase C activity was inhibited by the sulfhydryl reagent dithionitrobenzoate with an IC56 in the micromolar range. Phospholipase C activity was distributed widely in mouse tissues. The enzyme showed a progressive increase in activity from heart, liver, lung, colon, spleen, to brain Mouse and human melanomas grown as solid tumors had higher tissue. phospholipase C activity than mouse brain. The relatively high activity of this enzyme in melanoma may suggest a biological role for phospholipase C in solid tumor growth. 01990 Academic Press, Inc.
There is considerable evidence that the metabolism of inositol phospholipids is stimulated in cells in response to hormones and growth factors (for reviews, see Refs. 1-5). Although the exact mechanisms by which these signals are transduced in the cell are not yet known, phospholipase C (PLC) is believed to play a major role (5). This enzyme hydrolyzes selectively phosphoinositides to the products diacylglycerol and inositol phosphate(s), which are second messengers that transmit external signal(s) inside the cell. DAG stimulates the activity of protein kinase C, and inositol phosphates regulate the cytosolic concentration of free calcium. It is believed that these phenomena are important in cell growth (1,2). The purpose of this study is to characterize PLC activity from tumor tissue with emphasis on the human melanoma RPM1 7272. Abbreviations.
PLC, phosphoinositide phospholipase C; PI, phosphatidylinositol; InPs, phosphate(s); DAG, diacylglycerol; SDS, sodium dodecylsulfate, phenylmethylsulfonyl fluoride, DTNR, 5,5’-dithiobis (2- nitrobenzoate); 7,12-dimethylbenz[alanthracene.
715
inositol PMSF, DMBA,
0006-291x/90 $1.50 Copyright 0 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.
Vol.
166, No. 2, 1990
BIOCHEMICAL
AND BlOPHYSlCAL
RESEARCH COMMUNICATIONS
The characterization of PLC activity in tumor tissue would contribute understanding of the role of phospholipid turnover in solid tumor Portions of this work have been presented in a preliminary form (6). MATERIALS
AND
to our growth.
METHODS
. was purchased from Avanti Polar Lipids Inc., Materials, - Phosphatidylinositol Birmingham, Alabama. Alkaline phos hatase (bovine intestinal mucosa, type VII-T) and other chemicals were from 8 igma Chemical Co., St. Louis, Missouri. Buffer A: 10 mM imidazole, 10 mM EDTA, 0.25 M sucrose, pH 7.2; Buffer B: 10 mM imidazole, pH 7.2. Animals and Tlampr Cella; Mice for xeno afts were six week old NuiNu Swiss weighing from 24 to 30 grams and bre r at E. I. Du Pont. Animals were maintained in a regulated lighting cycle (light 12 h, dark 12 h) for 6 weeks before use. The human melanoma line RPM1 7272, obtained from D. B. R&in (New York University Medical Center, N. Y. C.), was originally derived from a patient with malignant melanoma (7). The melanoma was introduced by S.C. implantation of cultured tumor cells in nude mice. The mice were sacrificed by cervical dislocation and tumors (weighing 1 to 2 g) were rapidly removed and frozen at -70 oC in hexane on dry ice. The B16 mouse melanoma tumor line was maintained by serial passage in C57BLJ6 mice. The melanoma was introduced by S.C. implantation of tumor brei prepared by homogenizing tumor tissue in saline. The mice were sacrificed by cervical dislocation and tumors (weighing 1 to 2 g) or normal tissue were rapidly removed and frozen at -70 oC in hexane on dry ice. Primary colon tumors were induced by ten weekly subcutaneous injections of methylazoxymethanol acetate to CFl mice (8). The mice were sacrificed and the colon tumors were removed at 25 weeks. Tissue nrenaration; Frozen tumors or mouse tissues were thawed, excess fat was removed, and they were placed in 6.5 volumes of homogenization Buffer A containing the protease inhibitors 1 mM PMSF and 1 uM leupeptin. The tissue was homogenized on ice by mechanical disruption and sonication using a Tekmar ultrasonic homogenizer for 30 seconds at 20,000 rpm. The homogenates were centrifu ed at 2000 rpm for 20 minutes at 4 oC. The sediment was washed once with b d er A and centrifuged at 2000 rpm for 20 minutes. The washed pellet is referred to as the crude nuclear fraction. The supernatant and wash were combined and centrifuged at 105,000 g for 1 h at 4 oC. The fatty layer was removed using a cotton swab, and the su ernatant was dialyzed twice against one liter of Buffer B containing 1 mM PM 8 F and 1 uM leupeptin overni ht using dialysis tubing with a molecular weight cutoff of 25 k Daltons. The dia Kyzed supernatant is referred to as the cytosolic fraction and the high speed pellet as the microsomal fraction. Automated PLC asPLC activity was measured by determining the formation of inositol l- hosphate from phosphatidylinositol using a modification of the method of Pa Pmer (9). All concentrations are final assay values unless stated otherwise. The assay was prepared in 96-well microtiter plates using a mixture of 25 mM imidazole, pH 7.2, 100 mM KCI, 1 mM CaC12, 0.8 mg/ml deoxycholate, 1 mM phos hatidylinositol, and 20 to 40 ug of PLC protein fraction in a final volume of 100 u.P Phosphatidylinositol was prepared as a 10 mM stock solution in deionized water by brief sonication for 30 sec. The enzyme reaction was started by adding 1 mM substrate and incubating for 30 min at 37 oC. The calcium de endent PLC reaction was stopped b the addition of 9 mM EDT+, .pH 7.0. A& a1’me hos hatase (10 units) was a cfded and incubated for an additional 30 minutes P37 o 6 1 to hydrolyze the product of the reaction inositol l-phosphate to 716
Vol.
166,
No.
2, 1990
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
inorganic phosphate. The phosphatase reaction was stopped by adding 3.7 % SDS and 29 mM EDTA (pH 4.0) to the assay mixture. Inorganic phosphate was quantitated by adding 36.5 mM ZnCl2 and 5.5 mM ammonium molybdate. The mixture was incubated for 20 min at 37 oC before the absorbance was read spectrophotometrically at 360 run in a Titertek Multiskan microtiter plate reader connected to an IBM PC. Phosphatidylinositol hydrolysis was measured using the cytosolic fraction of human melanoma PLC. The rate of PI hydrolysis was linear at protein concentrations up to 30 ug per assay volume. Approximately, 25 percent of the phospholipid substrate was hydrolyzed during this time period. PI hydrolysis was constant for the first 30 min of incubation and leveled off by 60 minutes. PLC activity was determined routinely for 30 min at 30 ug of protein producing a specific activity of about 35 nmol/min/mg protein. Other methods; Protein was determined by the method modified Lowry procedure in the presence of SDS (11).
of Bradford
(10) or a
RJZSULTS Effect of nH. calcium. and substrate, The pH optima of the tumor enzyme activity was studied using a pH range of 5.0 to 8.0. PLC activity was maximum at pH 5.5 in the absence of detergent and declined rapidly at higher pH values (Fig. 1). In contrast, biphasic pH optima of 5.5 and 7.2 were observed in the presence of 1.8 The melanoma PLC was dependent on calcium for activity mM deoxycholate. (Fig. 2A), as is expected for this enzyme (5). The concentration of calcium producing one half maximal activity was 100 uM, while maximal activity was obtained at a calcium concentration of 500 uM. In support of the calcium dependence of this enzyme, little activity was observed in the presence of 1 mM EGTA. When calcium was replaced with magnesium or manganese very little enzyme activity was observed (data not shown). Substrate saturation studies on the tumor enzyme using PI as the substrate are shown in Fig. 2B. The initial rate of phosphatidylinositol hydrolysis was proportional to substrate concentration up
w. Effect of pH on cytosolic PLC activity of human melanoma solid tumors. The deoxycholate concentration in the assay was 0.8 mg/ml. For other assay conditions, see the Materials and Methods section. Each point represents the mean of quadruplicate determinations and is representative of the results obtained in a similar experiment. The SEM of each data point was less than 10 percent. 717
Vol.
166, No. 2, 1990
BIOCHEMICAL
AND BIOPHYSICAL
1000 -
RESEARCH COMMUNICATIONS
B
600 .’
400
..
0
“:“‘:“‘:-“:“‘:“‘i 200
400
CALCIUM
600
CONCENTRATION
600
1000
0
1200
(UK)
200
400
600
PH0SPHATIDYIJN0SlT0L
600
1000
(JlM)
FiP 2, Effect of (A) calcium and (B) phosphatidyl-inositol concentration on cytosolic PLC activity of human melanoma solid tumors. The zero calcium concentration in (A) was determined in the absenceof any divalent cation chelator and no added calcium. For assay conditions, see the Materials and Methods section. Each point represents the mean of quadruplicate determinations and is representative of results obtained in other experiments. The SEM of each data point was less than 10 percent.
to 200 r&I. PLC exhibited apparent saturation mM PI with an apparent Km of 2’70 uM.
kinetics
at concentrations
above 1
The effect of sulfhydryl reagents on tumor PLC was Effect of sulfhvdrvl reapents. explored to determine if the enzyme required cysteine for activity (12). Preincubation of the enzyme for 10 min at 37 OC with DTNB inhibited the enzyme in a concentration dependent manner. The ICgO for this reagent was 20 uM. Addition of 10 mM beta-mercaptoethanol during the preincubation period eliminated the inhibitory effect (data not shown). To verify that the inhibitory effect of DTNB on PLC activity was groupspecific, preincubation studies were done at various concentrations of DTNB (Fig. 3A). A plot of the logarithm of PLC activity versus preincubation time showed The data lines intersected near linear relationships at all DTNB concentrations. a single point on the ordinate of the graph, characteristic of group-specific reagents (13).
against the estimate the The data fit modification .
ue dism
A log-log plot of the reciprocal
.
.
The activity
tissues and in mouse and human activity
of the half-time
of inactivation
DTNB concentration is presented in Fig. 3B. This plot is used to average order of the reaction with respect to DTNB concentration. a straight line with a slope of 1.08, indicating that on the average the of one cysteine may be responsible for inhibiting the tumor enzyme. of cytosolic PLC was compared
(70 - 80 %), in all tissues, was in the cytosolic fraction;
as an operational
definition
in various mouse of the PLC
solid tumors (Fig. 4). The majority
of soluble
enzyme 718
cytosolic was used
(data not shown).
In the mouse,
Vol.
166,
No.
2, 1990
PREINCURATION
BIOCHEMICAL
Tilde
AND
BIOPHYSICAL
RESEARCH
D’MB
(min)
COMMUNICATIONS
be/d
Fig. 3, Time-dependent inhibition of the oup-specific reagent DTNB of cytosolic PLC of human melanoma solid tumors. F ytosohc. PLC was preincubated for 0 to 90 minutes at 37 oC in the presence of 10, 20, 49, or 80 ug/ml of DTNB and then assayed for enzyme activity. (A) Plotted according to the equation log(%control activity) = -k’(lOO%)t. (B) Plotted according to the equation log(l/tY2) = log(k) + n(log[I]). For assay conditions, see the Materials and Methods section. Each point represents the mean of quadruplicate determinations and is representative of results obtained in a second experiment. The SEM of each data point was less than 10 percent.
there was a progressive increase in the activity of PLC that followed the order: heart, liver, lung, colon, spleen, and brain. The brain had the highest enzyme activity of all the normal tissues examined. Primary mouse colon tumors, induced by the chemical methylazoxymethanol (8), had PLC activity comparable to that from mouse brain and greater than that found in normal colon. Moreover, the transplanted mouse B16 melanoma had much higher levels of PLC activity than any of the normal mouse tissues including brain. The human melanoma RPM1 7272 had the greatest PLC activity, higher than any of the other tissues
FiP 4, Tissue distribution of cytosolic PLC activity. For assay conditions, see the Materials and Methods section. Each point re resents the mean of quadruplicate determinations and is representative of res up ts obtained in other experiments. The SEM of each data point was less than 10 percent. 719
Vol.
166, No. 2, 1990
BIOCHEMICAL
examined. The specific activity that from mouse brain.
AND BIOPHYSICAL
of human
melanoma
RESEARCH COMMUNICATIONS
PLC was a-fold greater
than
DISCUSSION The studies presented here demonstrate that human melanoma grown as xenografts in nude mice contain high PI-specific PLC activity. Although most of the tumor PLC activity can be accounted for in the cytosolic fraction, this should be interpreted as a soluble fraction operationally since it is likely that this enzyme is loosely associated with intracellular membranes in vivo and then becomes disassociated upon homogenization (5,14-l@. PI- specific PLC activity from tumor tissue had a pH optimum of pH 5 The addition of deoxycholate to the cytosolic in the absence of deoxycholate. fraction revealed a second pH maximum of pH 7.2 for PLC. Both pH maxima may be properties of the same enzyme, with the activity at pH 5.0 representing the less biologically significant activity. Alternatively, multiple pH optima may reflect the existence of different forms of PLC (5,19-25). The apparent Km of PI for the tumor enzyme was 270 uM when the substrate was dispersed as mixed micelles with deoxycholate. In brain, heart, and intestinal mucosa, the activity of PLC was inhibited by sulfhydryl reagents (16,26,27). In bovine brain, two immunologically distinct PLCs were differentially inhibited by HgCl2 (28). In the present study, tumor PLC activity was inhibited by the sulfhydryl reagent DTNB in the micromolar range. The tumor enzyme in the present study was more sensitive to the inhibitory effects of DTNB than was the form of PLC from heart (28). These observations may suggest that different forms of PLC are more prevalent in some tissues than in others (30). Using DTNB as a model group-specific reagent, tumor PLC inhibition studies revealed that the modification of a single cysteine may be responsible for It is not known at present whether this labile the inhibition of the enzyme. cysteine is in a calcium binding site, the catalytic site, or elsewhere in the enzyme. The activities of PLC in mouse tissues were studied to determine the relative activities in various organs. Brain had the highest enzyme activity of all the mouse tissues examined. This high activity could be associated with the degree of signal transduction in the brain. In contrast, the high PLC activity in tumor tissues may correspond to the growth rate of tumor tissue since the PI 720
Vol.
166, No. 2, 1990
BIOCHEMKAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
signaling pathway is elevated in growing cells (2). In this regard, chemically induced mouse colon tumors had elevated PLC activity compared with that of normal colon. This is in agreement with the observation that PLC activity was elevated in DMBA-induced rat mammary tumors compared to normal mammary tissue (29). The present studies demonstrate that melanoma has high PI-PLC activity. Further studies of the tumor enzyme are in progress so that we may better understand the role of PLC in solid tumor growth.
OwLEIII;ME1yTs; I would like to thank Rosemarie Jankewicz (Medical Products, E. I. Du Pont) for excellent technical assistance in these studies, and Davette Behrens (Medical Products, E. I. Du Pont) for generously supplying primary colon tumors for these studies.
REFERENCES 1. 2. 3. 4.
Rozengurt, E. (1986) Science 234,161-166. Berridge, M. J. (1987) IS1 Atlas of Sci.: pharmacol. 91-97 . Bishop, W. R. and Bell, R. (1987) Oncogene Res. 2,1-14. Majerus, P. W., Connolly, T. M., Bansal, V. S., Inhorn, R. C., Ross, T. S., and Lips, D. L. (1988) J. Biol. Chem. 263,3051-3054. 5. Majerus, P. W., Connolly, T. M., Deckmyn, H., Ross, T. S., Bross, T. E., Ishii,H., Bansal, V. S., and Wilson, D. B. (1986) Science 234,1519-1526. 6. Perrella, F. W. (1988) Proc. Am. Assoc. Cancer Res. 29,lO. 7. Quinn, L. A., Woods, L. K., Merrick, S. B., Arabasz, N. M., and Moore, G. E. (1977) J. Natl. Cancer Inst. 59,301-305. 8. Pratesi, G. and Deschner, E. E. (1984) Cancer 54,18-24. 9. Palmer, F. B. S. C. (1985) Anal Biochem. 150,345-352. 10. Bradford, M. (1976) Anal. Biochem. 72,248-254. 11. Markwell, M. A. K., Haas, S. M., Bieber, L. L., and Tolbert, N. E. (1978) Anal. Biochem. 87,206-210. 12. Russo, A. and Bump, E. A. (1988) in Methods of Biochemical Analysis (GIick, D., Ed.), Vol. 33, pp. 165-241, John Wiley & Sons, New York. 13. Schloss, J. V. (1988) in Target Sites of Herbicide Action (Boger, P. and Sandmann, G., Eds.), Boca Raton, FL, CRC Press, Inc., Boca Raton, FL. 14. Irvine, R. F. and Dawson, R. M. C. (1978) J. Neurochem. 31,1427-1434. 15. Wang, P., Toyoshima, S., and Osawa, T. (1986) J. Biochem. 100,1015-1022. 16. Schwertz, D. W., Halverson, J. B., Palmer, J. W., and Feinberg, H. (1987) Arch. Biochem. Biophys. 253,388-398. 17. Ho, A. K. and KIein, D. C. (1987) J. Neurochem. 48,1033-1038. 18. Ribbes, H., Plantavid, M., Bennet, P. J., Chap, H., and Douste- Blazy, L. (1987) Biochim. Biophys. Acta 919,245-254. 19. Hofinann, S. L. and Majerus, P. W. (1982) J. Biol. Chem. 2576461-6469. 20. Katan, M. and Parker, P. J. (1987) Eur. J. Biochem. 168,413-418. 21. Carter, H. R. and Smith, A. D. (1987) Biochem. J. 244,639~645. 22. Low, M.G., Carroll, R. C., and Weglicki, W. B. (1984) Biochem. J. 221,813-820. 23. Hirasawa, K., Irvine, R. F., and Dawson, R. M. C. (1982) Biochem. Biophys. Res. Commun. 107,533-537. 24. Nakanishi, H., Nomura, H., Kikkawa, U., Kishimoto, A., and Nishizuka, Y. (1985) Biochem. Biophys. Res. Commun. 132,582-590. 25. Rebecchi, M. J. and Rosen, 0. M. (1987) J. Biol. Chem. 262,12526-12532. 26. Thompson, W. 0967) Can. J. Biochem. 45,853-861. 721
Vol.
166, No. 2, 1990
BIOCHEMICAL
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
27. Atherton, R. S. and Hawthorne, J. N. (1968) Eur. J. Biochem. 4,68-75. 28. Ryu, S. H., Cho, K. S., Lee, K. Y., Suh, P. G., and Rhee, S. G. (1987) J. Biol. Chem. 262,12511-12518. 29. Rillema, J. A. (1986) Proc. Sot. Exptl. Biol. Med. 181,450-453. 30. Homma, Y., Takenawa, T., Emori, Y., Sorimachi, H., and Suzuki, K. (1989) Biochem. Biophys. Res. Commun. 164,406-412.
722
166,
January
No. 30,
2, 1990
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
Pages
1990
715-722
CHARACTERIZATION OF PHOSPHATIDYLINOSITOL PHOSPHOLIPASE C ACTIVITY IN HUMAN MELANOMA Frank W. Perrella Medical Products Department E. I. Du Pont de Nemours & Co. 500 South Ridgeway Avenue Glenolden, Pennsylvania 19036 Received
November
3, 1989
SUMMARY: Phosphoinositide phospholipase C activity was investigated in human melanoma grown as solid tumor xenografts in nude mice. The enzyme was dependent on calcium for activity and was stimulated by the detergent deoxycholate. The pH optimum was 5.5 in the absence of detergent, and in the presence of deoxycholate two pH maxima were present, 5.5 and 7.2. Phospholipase C activity was inhibited by the sulfhydryl reagent dithionitrobenzoate with an IC56 in the micromolar range. Phospholipase C activity was distributed widely in mouse tissues. The enzyme showed a progressive increase in activity from heart, liver, lung, colon, spleen, to brain Mouse and human melanomas grown as solid tumors had higher tissue. phospholipase C activity than mouse brain. The relatively high activity of this enzyme in melanoma may suggest a biological role for phospholipase C in solid tumor growth. 01990 Academic Press, Inc.
There is considerable evidence that the metabolism of inositol phospholipids is stimulated in cells in response to hormones and growth factors (for reviews, see Refs. 1-5). Although the exact mechanisms by which these signals are transduced in the cell are not yet known, phospholipase C (PLC) is believed to play a major role (5). This enzyme hydrolyzes selectively phosphoinositides to the products diacylglycerol and inositol phosphate(s), which are second messengers that transmit external signal(s) inside the cell. DAG stimulates the activity of protein kinase C, and inositol phosphates regulate the cytosolic concentration of free calcium. It is believed that these phenomena are important in cell growth (1,2). The purpose of this study is to characterize PLC activity from tumor tissue with emphasis on the human melanoma RPM1 7272. Abbreviations.
PLC, phosphoinositide phospholipase C; PI, phosphatidylinositol; InPs, phosphate(s); DAG, diacylglycerol; SDS, sodium dodecylsulfate, phenylmethylsulfonyl fluoride, DTNR, 5,5’-dithiobis (2- nitrobenzoate); 7,12-dimethylbenz[alanthracene.
715
inositol PMSF, DMBA,
0006-291x/90 $1.50 Copyright 0 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.
Vol.
166, No. 2, 1990
BIOCHEMICAL
AND BlOPHYSlCAL
RESEARCH COMMUNICATIONS
The characterization of PLC activity in tumor tissue would contribute understanding of the role of phospholipid turnover in solid tumor Portions of this work have been presented in a preliminary form (6). MATERIALS
AND
to our growth.
METHODS
. was purchased from Avanti Polar Lipids Inc., Materials, - Phosphatidylinositol Birmingham, Alabama. Alkaline phos hatase (bovine intestinal mucosa, type VII-T) and other chemicals were from 8 igma Chemical Co., St. Louis, Missouri. Buffer A: 10 mM imidazole, 10 mM EDTA, 0.25 M sucrose, pH 7.2; Buffer B: 10 mM imidazole, pH 7.2. Animals and Tlampr Cella; Mice for xeno afts were six week old NuiNu Swiss weighing from 24 to 30 grams and bre r at E. I. Du Pont. Animals were maintained in a regulated lighting cycle (light 12 h, dark 12 h) for 6 weeks before use. The human melanoma line RPM1 7272, obtained from D. B. R&in (New York University Medical Center, N. Y. C.), was originally derived from a patient with malignant melanoma (7). The melanoma was introduced by S.C. implantation of cultured tumor cells in nude mice. The mice were sacrificed by cervical dislocation and tumors (weighing 1 to 2 g) were rapidly removed and frozen at -70 oC in hexane on dry ice. The B16 mouse melanoma tumor line was maintained by serial passage in C57BLJ6 mice. The melanoma was introduced by S.C. implantation of tumor brei prepared by homogenizing tumor tissue in saline. The mice were sacrificed by cervical dislocation and tumors (weighing 1 to 2 g) or normal tissue were rapidly removed and frozen at -70 oC in hexane on dry ice. Primary colon tumors were induced by ten weekly subcutaneous injections of methylazoxymethanol acetate to CFl mice (8). The mice were sacrificed and the colon tumors were removed at 25 weeks. Tissue nrenaration; Frozen tumors or mouse tissues were thawed, excess fat was removed, and they were placed in 6.5 volumes of homogenization Buffer A containing the protease inhibitors 1 mM PMSF and 1 uM leupeptin. The tissue was homogenized on ice by mechanical disruption and sonication using a Tekmar ultrasonic homogenizer for 30 seconds at 20,000 rpm. The homogenates were centrifu ed at 2000 rpm for 20 minutes at 4 oC. The sediment was washed once with b d er A and centrifuged at 2000 rpm for 20 minutes. The washed pellet is referred to as the crude nuclear fraction. The supernatant and wash were combined and centrifuged at 105,000 g for 1 h at 4 oC. The fatty layer was removed using a cotton swab, and the su ernatant was dialyzed twice against one liter of Buffer B containing 1 mM PM 8 F and 1 uM leupeptin overni ht using dialysis tubing with a molecular weight cutoff of 25 k Daltons. The dia Kyzed supernatant is referred to as the cytosolic fraction and the high speed pellet as the microsomal fraction. Automated PLC asPLC activity was measured by determining the formation of inositol l- hosphate from phosphatidylinositol using a modification of the method of Pa Pmer (9). All concentrations are final assay values unless stated otherwise. The assay was prepared in 96-well microtiter plates using a mixture of 25 mM imidazole, pH 7.2, 100 mM KCI, 1 mM CaC12, 0.8 mg/ml deoxycholate, 1 mM phos hatidylinositol, and 20 to 40 ug of PLC protein fraction in a final volume of 100 u.P Phosphatidylinositol was prepared as a 10 mM stock solution in deionized water by brief sonication for 30 sec. The enzyme reaction was started by adding 1 mM substrate and incubating for 30 min at 37 oC. The calcium de endent PLC reaction was stopped b the addition of 9 mM EDT+, .pH 7.0. A& a1’me hos hatase (10 units) was a cfded and incubated for an additional 30 minutes P37 o 6 1 to hydrolyze the product of the reaction inositol l-phosphate to 716
Vol.
166,
No.
2, 1990
BIOCHEMICAL
AND
BIOPHYSICAL
RESEARCH
COMMUNICATIONS
inorganic phosphate. The phosphatase reaction was stopped by adding 3.7 % SDS and 29 mM EDTA (pH 4.0) to the assay mixture. Inorganic phosphate was quantitated by adding 36.5 mM ZnCl2 and 5.5 mM ammonium molybdate. The mixture was incubated for 20 min at 37 oC before the absorbance was read spectrophotometrically at 360 run in a Titertek Multiskan microtiter plate reader connected to an IBM PC. Phosphatidylinositol hydrolysis was measured using the cytosolic fraction of human melanoma PLC. The rate of PI hydrolysis was linear at protein concentrations up to 30 ug per assay volume. Approximately, 25 percent of the phospholipid substrate was hydrolyzed during this time period. PI hydrolysis was constant for the first 30 min of incubation and leveled off by 60 minutes. PLC activity was determined routinely for 30 min at 30 ug of protein producing a specific activity of about 35 nmol/min/mg protein. Other methods; Protein was determined by the method modified Lowry procedure in the presence of SDS (11).
of Bradford
(10) or a
RJZSULTS Effect of nH. calcium. and substrate, The pH optima of the tumor enzyme activity was studied using a pH range of 5.0 to 8.0. PLC activity was maximum at pH 5.5 in the absence of detergent and declined rapidly at higher pH values (Fig. 1). In contrast, biphasic pH optima of 5.5 and 7.2 were observed in the presence of 1.8 The melanoma PLC was dependent on calcium for activity mM deoxycholate. (Fig. 2A), as is expected for this enzyme (5). The concentration of calcium producing one half maximal activity was 100 uM, while maximal activity was obtained at a calcium concentration of 500 uM. In support of the calcium dependence of this enzyme, little activity was observed in the presence of 1 mM EGTA. When calcium was replaced with magnesium or manganese very little enzyme activity was observed (data not shown). Substrate saturation studies on the tumor enzyme using PI as the substrate are shown in Fig. 2B. The initial rate of phosphatidylinositol hydrolysis was proportional to substrate concentration up
w. Effect of pH on cytosolic PLC activity of human melanoma solid tumors. The deoxycholate concentration in the assay was 0.8 mg/ml. For other assay conditions, see the Materials and Methods section. Each point represents the mean of quadruplicate determinations and is representative of the results obtained in a similar experiment. The SEM of each data point was less than 10 percent. 717
Vol.
166, No. 2, 1990
BIOCHEMICAL
AND BIOPHYSICAL
1000 -
RESEARCH COMMUNICATIONS
B
600 .’
400
..
0
“:“‘:“‘:-“:“‘:“‘i 200
400
CALCIUM
600
CONCENTRATION
600
1000
0
1200
(UK)
200
400
600
PH0SPHATIDYIJN0SlT0L
600
1000
(JlM)
FiP 2, Effect of (A) calcium and (B) phosphatidyl-inositol concentration on cytosolic PLC activity of human melanoma solid tumors. The zero calcium concentration in (A) was determined in the absenceof any divalent cation chelator and no added calcium. For assay conditions, see the Materials and Methods section. Each point represents the mean of quadruplicate determinations and is representative of results obtained in other experiments. The SEM of each data point was less than 10 percent.
to 200 r&I. PLC exhibited apparent saturation mM PI with an apparent Km of 2’70 uM.
kinetics
at concentrations
above 1
The effect of sulfhydryl reagents on tumor PLC was Effect of sulfhvdrvl reapents. explored to determine if the enzyme required cysteine for activity (12). Preincubation of the enzyme for 10 min at 37 OC with DTNB inhibited the enzyme in a concentration dependent manner. The ICgO for this reagent was 20 uM. Addition of 10 mM beta-mercaptoethanol during the preincubation period eliminated the inhibitory effect (data not shown). To verify that the inhibitory effect of DTNB on PLC activity was groupspecific, preincubation studies were done at various concentrations of DTNB (Fig. 3A). A plot of the logarithm of PLC activity versus preincubation time showed The data lines intersected near linear relationships at all DTNB concentrations. a single point on the ordinate of the graph, characteristic of group-specific reagents (13).
against the estimate the The data fit modification .
ue dism
A log-log plot of the reciprocal
.
.
The activity
tissues and in mouse and human activity
of the half-time
of inactivation
DTNB concentration is presented in Fig. 3B. This plot is used to average order of the reaction with respect to DTNB concentration. a straight line with a slope of 1.08, indicating that on the average the of one cysteine may be responsible for inhibiting the tumor enzyme. of cytosolic PLC was compared
(70 - 80 %), in all tissues, was in the cytosolic fraction;
as an operational
definition
in various mouse of the PLC
solid tumors (Fig. 4). The majority
of soluble
enzyme 718
cytosolic was used
(data not shown).
In the mouse,
Vol.
166,
No.
2, 1990
PREINCURATION
BIOCHEMICAL
Tilde
AND
BIOPHYSICAL
RESEARCH
D’MB
(min)
COMMUNICATIONS
be/d
Fig. 3, Time-dependent inhibition of the oup-specific reagent DTNB of cytosolic PLC of human melanoma solid tumors. F ytosohc. PLC was preincubated for 0 to 90 minutes at 37 oC in the presence of 10, 20, 49, or 80 ug/ml of DTNB and then assayed for enzyme activity. (A) Plotted according to the equation log(%control activity) = -k’(lOO%)t. (B) Plotted according to the equation log(l/tY2) = log(k) + n(log[I]). For assay conditions, see the Materials and Methods section. Each point represents the mean of quadruplicate determinations and is representative of results obtained in a second experiment. The SEM of each data point was less than 10 percent.
there was a progressive increase in the activity of PLC that followed the order: heart, liver, lung, colon, spleen, and brain. The brain had the highest enzyme activity of all the normal tissues examined. Primary mouse colon tumors, induced by the chemical methylazoxymethanol (8), had PLC activity comparable to that from mouse brain and greater than that found in normal colon. Moreover, the transplanted mouse B16 melanoma had much higher levels of PLC activity than any of the normal mouse tissues including brain. The human melanoma RPM1 7272 had the greatest PLC activity, higher than any of the other tissues
FiP 4, Tissue distribution of cytosolic PLC activity. For assay conditions, see the Materials and Methods section. Each point re resents the mean of quadruplicate determinations and is representative of res up ts obtained in other experiments. The SEM of each data point was less than 10 percent. 719
Vol.
166, No. 2, 1990
BIOCHEMICAL
examined. The specific activity that from mouse brain.
AND BIOPHYSICAL
of human
melanoma
RESEARCH COMMUNICATIONS
PLC was a-fold greater
than
DISCUSSION The studies presented here demonstrate that human melanoma grown as xenografts in nude mice contain high PI-specific PLC activity. Although most of the tumor PLC activity can be accounted for in the cytosolic fraction, this should be interpreted as a soluble fraction operationally since it is likely that this enzyme is loosely associated with intracellular membranes in vivo and then becomes disassociated upon homogenization (5,14-l@. PI- specific PLC activity from tumor tissue had a pH optimum of pH 5 The addition of deoxycholate to the cytosolic in the absence of deoxycholate. fraction revealed a second pH maximum of pH 7.2 for PLC. Both pH maxima may be properties of the same enzyme, with the activity at pH 5.0 representing the less biologically significant activity. Alternatively, multiple pH optima may reflect the existence of different forms of PLC (5,19-25). The apparent Km of PI for the tumor enzyme was 270 uM when the substrate was dispersed as mixed micelles with deoxycholate. In brain, heart, and intestinal mucosa, the activity of PLC was inhibited by sulfhydryl reagents (16,26,27). In bovine brain, two immunologically distinct PLCs were differentially inhibited by HgCl2 (28). In the present study, tumor PLC activity was inhibited by the sulfhydryl reagent DTNB in the micromolar range. The tumor enzyme in the present study was more sensitive to the inhibitory effects of DTNB than was the form of PLC from heart (28). These observations may suggest that different forms of PLC are more prevalent in some tissues than in others (30). Using DTNB as a model group-specific reagent, tumor PLC inhibition studies revealed that the modification of a single cysteine may be responsible for It is not known at present whether this labile the inhibition of the enzyme. cysteine is in a calcium binding site, the catalytic site, or elsewhere in the enzyme. The activities of PLC in mouse tissues were studied to determine the relative activities in various organs. Brain had the highest enzyme activity of all the mouse tissues examined. This high activity could be associated with the degree of signal transduction in the brain. In contrast, the high PLC activity in tumor tissues may correspond to the growth rate of tumor tissue since the PI 720
Vol.
166, No. 2, 1990
BIOCHEMKAL
AND BIOPHYSICAL
RESEARCH COMMUNICATIONS
signaling pathway is elevated in growing cells (2). In this regard, chemically induced mouse colon tumors had elevated PLC activity compared with that of normal colon. This is in agreement with the observation that PLC activity was elevated in DMBA-induced rat mammary tumors compared to normal mammary tissue (29). The present studies demonstrate that melanoma has high PI-PLC activity. Further studies of the tumor enzyme are in progress so that we may better understand the role of PLC in solid tumor growth.
OwLEIII;ME1yTs; I would like to thank Rosemarie Jankewicz (Medical Products, E. I. Du Pont) for excellent technical assistance in these studies, and Davette Behrens (Medical Products, E. I. Du Pont) for generously supplying primary colon tumors for these studies.
REFERENCES 1. 2. 3. 4.
Rozengurt, E. (1986) Science 234,161-166. Berridge, M. J. (1987) IS1 Atlas of Sci.: pharmacol. 91-97 . Bishop, W. R. and Bell, R. (1987) Oncogene Res. 2,1-14. Majerus, P. W., Connolly, T. M., Bansal, V. S., Inhorn, R. C., Ross, T. S., and Lips, D. L. (1988) J. Biol. Chem. 263,3051-3054. 5. Majerus, P. W., Connolly, T. M., Deckmyn, H., Ross, T. S., Bross, T. E., Ishii,H., Bansal, V. S., and Wilson, D. B. (1986) Science 234,1519-1526. 6. Perrella, F. W. (1988) Proc. Am. Assoc. Cancer Res. 29,lO. 7. Quinn, L. A., Woods, L. K., Merrick, S. B., Arabasz, N. M., and Moore, G. E. (1977) J. Natl. Cancer Inst. 59,301-305. 8. Pratesi, G. and Deschner, E. E. (1984) Cancer 54,18-24. 9. Palmer, F. B. S. C. (1985) Anal Biochem. 150,345-352. 10. Bradford, M. (1976) Anal. Biochem. 72,248-254. 11. Markwell, M. A. K., Haas, S. M., Bieber, L. L., and Tolbert, N. E. (1978) Anal. Biochem. 87,206-210. 12. Russo, A. and Bump, E. A. (1988) in Methods of Biochemical Analysis (GIick, D., Ed.), Vol. 33, pp. 165-241, John Wiley & Sons, New York. 13. Schloss, J. V. (1988) in Target Sites of Herbicide Action (Boger, P. and Sandmann, G., Eds.), Boca Raton, FL, CRC Press, Inc., Boca Raton, FL. 14. Irvine, R. F. and Dawson, R. M. C. (1978) J. Neurochem. 31,1427-1434. 15. Wang, P., Toyoshima, S., and Osawa, T. (1986) J. Biochem. 100,1015-1022. 16. Schwertz, D. W., Halverson, J. B., Palmer, J. W., and Feinberg, H. (1987) Arch. Biochem. Biophys. 253,388-398. 17. Ho, A. K. and KIein, D. C. (1987) J. Neurochem. 48,1033-1038. 18. Ribbes, H., Plantavid, M., Bennet, P. J., Chap, H., and Douste- Blazy, L. (1987) Biochim. Biophys. Acta 919,245-254. 19. Hofinann, S. L. and Majerus, P. W. (1982) J. Biol. Chem. 2576461-6469. 20. Katan, M. and Parker, P. J. (1987) Eur. J. Biochem. 168,413-418. 21. Carter, H. R. and Smith, A. D. (1987) Biochem. J. 244,639~645. 22. Low, M.G., Carroll, R. C., and Weglicki, W. B. (1984) Biochem. J. 221,813-820. 23. Hirasawa, K., Irvine, R. F., and Dawson, R. M. C. (1982) Biochem. Biophys. Res. Commun. 107,533-537. 24. Nakanishi, H., Nomura, H., Kikkawa, U., Kishimoto, A., and Nishizuka, Y. (1985) Biochem. Biophys. Res. Commun. 132,582-590. 25. Rebecchi, M. J. and Rosen, 0. M. (1987) J. Biol. Chem. 262,12526-12532. 26. Thompson, W. 0967) Can. J. Biochem. 45,853-861. 721
Vol.
166, No. 2, 1990
BIOCHEMICAL
AND BIOPHYSICAL RESEARCH COMMUNICATIONS
27. Atherton, R. S. and Hawthorne, J. N. (1968) Eur. J. Biochem. 4,68-75. 28. Ryu, S. H., Cho, K. S., Lee, K. Y., Suh, P. G., and Rhee, S. G. (1987) J. Biol. Chem. 262,12511-12518. 29. Rillema, J. A. (1986) Proc. Sot. Exptl. Biol. Med. 181,450-453. 30. Homma, Y., Takenawa, T., Emori, Y., Sorimachi, H., and Suzuki, K. (1989) Biochem. Biophys. Res. Commun. 164,406-412.
722
