all-trans-retinal stimulates superoxide release and phospholipase C

Transcription

1 Proc. Natl. Acad. Sci. USA Vol. 83, pp , October 1986 Biochemistry all-trans-retinal stimulates superoxide release and phospholipase C activity in neutrophils without significantly blocking protein kinase C (cell-stimulation/phospholipids/inositol phosphates) JANis E. LOCHNER*, JOHN A. BADWEY, WILHELM HORN, AND MANFRED L. KARNOVSKYt Department of Biological Chemistry, Harvard Medical School, Boston, MA 2115 Communicated by Eugene P. Kennedy, July 3, 1986 ABSTRACT all-trans-retinal was previously shown to stimulate high levels of superoxide release by guinea pig neutrophils. When the cells, previously labeled with [3H]inositol, are treated with all-trans-retinal, they exhibit a decrease in the levels of [3H]inositol phospholipids and an increase in the accumulation of [3H]inositol phosphates. The maximal accumulation of inositol phosphates and the optimal rate of superoxide release occurred together at -7 min after stimulation. The levels of [3H]inositol phosphates accumulated were comparable to those observed when the cells were stimulated with a chemotactic peptide. In direct measurements, using concentrations that stimulate intact cells maximally, all-trans-retinal was found not to inhibit protein kinase C from the cytosol of neutrophils significantly. This contrasts with the situation with this kinase obtained from other sources. These observations represent additional effects of vitamin A on cells. Retinoids have recently attracted considerable attention due to their ability to affect cellular differentiation and inhibit carcinogenesis (for review, see ref. 1). These substances may exert their effects by influencing the expression of certain genes (e.g., see ref. 2), by functioning as glycolipid intermediates in glycosylation reactions (e.g., see ref. 3), by increasing the liquid crystalline fraction of membranes (4, 5), and by acting as inhibitors of protein kinase C (e.g., see refs. 6-8). The last-mentioned enzyme is itself the cellular receptor for the tumor-promoting phorbol esters (9, 1). Diglyceride, formed by a phospholipase C-catalyzed hydrolysis of inositol phospholipids, is the physiological activator of protein kinase C (e.g., see ref. 1), and phorbol esters can substitute for diglyceride in this role (1). Certain phorbol esters [e.g., phorbol 12-myristate 13- acetate (PMA)] have been widely used as inducers of high levels of superoxide (O 2) release by neutrophils (e.g., see refs. 11 and 12). Superoxide is a key component of the oxygen-dependent antimicrobial mechanisms of phagocytic leukocytes (for review, see ref. 13). Activation of phospholipase C and of protein kinase C are considered to be critical initial events in the stimulation of this release (for review, see ref. 14). Retinoids are known to block the effects of phorbol esters on a wide variety of cells (6, 15, 16). However, we have recently reported that they stimulate high levels of O2 release by neutrophils (17). The mechanism of the stimulation by retinoids is of interest since this is a situation in which retinoids mimic, rather than inhibit, a cellular effect of phorbol esters. In this paper, we report effects of all-transretinal on phospholipase C and protein kinase C of neutrophils. The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C solely to indicate this fact. MATERIALS AND METHODS myo-[2-3h]inositol (1-2 Ci/mmol; 1 Ci = 37 GBq) and [y-32p]atp (2-1 Ci/mmol) were purchased from Dupont- NEN Products. all-trans-retinal (Type XVI), phosphatidylserine, 1,2-dioleoyl-rac-glycerol, histone (type III-S), neomycin sulfate, and inositol phospholipid standards were obtained from Sigma. Anion exchange resin AG1-X8 (formate form) (2-4 mesh) was a product of Bio-Rad. Glycophase G glass beads (CPG/2) were obtained from Pierce. Sources of all other materials used have been described elsewhere (12, 17). Preparation of Neutrophils. Guinea pig peritoneal neutrophils were prepared as described (18). Superoxide Release. Superoxide release by neutrophils was measured as outlined earlier (12). Solutions of all-transretinal (1 mm) were prepared in dimethyl sulfoxide immediately prior to use (17). Stock solutions of fmet-leu-phe and PMA were prepared in dimethyl sulfoxide and stored at -2 C until needed. These compounds were diluted with dimethyl sulfoxide so that the final concentration of solvent in the assays was.25% (vol/vol) in all cases. This concentration of solvent did not itself cause the effects noted. Labeling of Cells with [3H]Inositol. Neutrophil inositol phospholipids were labeled by incubating the cells (18 cells per ml) for 2 hr at 37 C in an isotonic Hepes buffer (124 mm NaCl/5 mm KCl/1 mm Hepes, ph 7.35) containing [3H]inositol (125,Ci/ml). The cells were washed in a modified Hanks' Hepes buffer (124 mm NaCl/5 mm KCl/.5 mm CaCl2/.2 mm MgCl2/1 mm Hepes, ph 7.35/5 mm glucose), resuspended in 5 ml of this buffer at a concentration of 2 x 16 cells per ml, and equilibrated to 37 C. They were stimulated with all-trans-retinal (25,uM), with fmet- Leu-Phe (1,uM), or with PMA (.5 AM) for various times, and the reactions were terminated by rapidly pouring the cell suspensions into 5-ml flasks containing 1 ml of the modified Hanks' Hepes buffer chilled to 4 C. The cells were transferred to 5-ml centrifuge tubes and collected by centrifugation for 1 min at 4 C. Cell pellets were lysed by the addition of.67 ml of 4.5% perchloric acid, and the water extracts were combined and neutralized by the procedure of Bradford and Rubin (19). The water-insoluble lipid residue was saved for analysis of inositol phospholipid labeling. Separation and Measurement of [3H]Inositol Phosphates. The combined neutralized extracts from 18 cells were applied to 1.-ml columns of AG1-X8 resin (formate form), and myo-inositol and glycerophosphoinositol were eluted as outlined in ref. 2. The inositol phosphates-inositol mono- Abbreviations: PMA, phorbol 12-myristate 13-acetate; InsP3, inositol trisphosphate; InsP2, inositol bisphosphate; InsP, inositol monophosphate. *Present address: Department of Chemistry, Lewis and Clark College, Portland, OR tto whom reprint requests should be addressed. 7673

2 7674 Biochemistry: Lochner et al. phosphate, inositol bisphosphate, and inositol trisphosphate (InsP, InsP2, and InsP3)-were then sequentially eluted with a stepwise gradient of ammonium formate (ref. 2; see legend to Fig. 2). Fractions (1. ml each) were collected and.5 ml was used to determine the 3H content by scintillation counting Ṁeasurement of [3H]Inositol Phospholipids. Lipids were extracted from the insoluble residue that remained from the perchlorate extraction procedure and were washed by the method of Schacht (21). The organic phase was washed twice more with.5 M KCl. The extracts were applied to immobilized neomycin glycophase columns, and the individual inositol phospholipids were eluted with increasing concentrations of ammonium formate (22). It was necessary to raise the final formate concentration to 2. M in order to elute all of the phosphatidylinositol 4,5-bisphosphate, as noted previously (23). Phospholipids were measured as total lipid phosphorous after washing (24). Protein Kinase C. This enzyme was assayed by measuring the incorporation Of 32p from [y32p]atp into histone type III-S. The assay conditions were those described by Hirota et al. (6). The neutrophil cytosol fraction was prepared by disrupting 19 cells in 4. ml of buffer A (6) by sonication or freeze-thawing and centrifuging the resulting homogenate at 1, x g for 1 hr at 4C. Cytosol fractions prepared from cells disrupted by either method exhibited identical results. RESULTS Stimulation of Release by Neutrophils. Upon exposure to all-trans-retinal neutrophils release O2, detected by the reduction of ferricytochrome c. A lag period of several minutes was observed between the addition of this stimulus and the attainment of the linear steady-state rate. In contrast, a latency period was not observed when the chemotactic peptide fmet-leu-phe served as the stimulus, and only a brief lag ('3 sec) occurred when the phorbol ester PMA was used (Fig. 1). The maximal rates of O2 release, derived from the slopes of the steepest linear regions of the progress curves, were virtually identical for all of these stimuli at optimal concentrations (i.e., -5 nmol of O2 per min per 17 cells) (12, 17). When the combined effects of retinal and PMA were checked, the progress curve for O2 release was very similar E Time, min FIG. 1. Time course of superoxide release by neutrophils stimulated with various agents. The curves show a portion of the progress curves for superoxide release from neutrophils (1.5 x 16 per ml) stimulated with 1,uM fmet-leu-phe (curve A),.5,uM PMA (curve B), and 25,uM all-trans-retinal (curve C), which were added to the assays last to initiate the reactions (arrow). All other assay conditions are described in Methods. The time (t) required to reach the maximal activity was approximated by back-extrapolation to zero absorbance change (dashed lines) as indicated for curves B and C. to that observed for retinal alone-i.e., a single sustained rate was seen; the maximal rate was that for either stimulus alone-i.e., the rates were not additive (cf. Fig. 1). Neutrophils labeled with [3H]inositol exhibited lag times identical to those observed for freshly harvested unlabeled cells. The maximal rates of O- release exhibited by these labeled cells stimulated with all-trans-retinal (25,M), PMA (.5,uM), or fmet-leu-phe (1,uM) were 91.% ± 5.% (SD; n = 3), 71.% ± 16.% (SD; n = 3), and 67% (n = 2; range, 5-84), respectively, of the values for unlabeled cells; i.e., cells not subjected to a previous incubation step. Thus, the labeling procedure only modestly affected O2 release when the cells were stimulated later. Accumulation of Inositol Phosphates in Neutrophils. Unstimulated neutrophils contain measurable quantities of [3H]labeled inositol phosphates, particularly InsP (Fig. 2). The large amount of this entity is probably a reflection of the higher concentration and greater specific activity of [3H]phosphatidylinositol in these cells compared to the other inositol phospholipids (Table 1). Stimulation of neutrophils with all-trans-retinal resulted in increases of the radioactivity in InsP2 and InsP3 of and =2.7-fold, respectively, whereas that in InsP was increased by a factor of only 1.5 (Fig. 3). Maximal accumulation of radioactive inositol phosphates occurred at '-7 min after stimulation with all-trans-retinal, the same time at which optimal O2 release was observed with this stimulus (Fig. 3). These increases in inositol phosphates were comparable to those observed upon stimulation of the cells with fmet-leu- Phe (Table 2). In contrast, stimulation with the tumor promoter PMA, which bypasses the phospholipase C step and activates protein kinase C directly (e.g., see ref. 1), O 3.5 _ 1- x 3. u 2.5.' 2. * F- 1. _.5 Proc. Natl. Acad. Sci. USA 83 (1986) A B C Fraction FIG. 2. Separation of inositol phosphates of neutrophils by anion exchange chromatography. The elution profiles of [3H]inositol phosphates from unstimulated neutrophils (x) and neutrophils stimulated with all-trans-retinal (25,M) for 1 min (o) and 7 min (e) are shown. Labeling of cells with [3H]inositol, extraction of the inositol phosphates, and separation of these compounds on Dowex AG1-X8 (formate form) columns are described in Methods. Inositol phosphates were eluted sequentially with the following eluants: region A,.1 M formic acid/.2 M ammonium formate (8 ml); region B,.1 M formic acid/.4 M ammonium formate (12 ml); region C,.1 M formic acid/1. M ammonium formate (1 ml). Regions A, B, and C contained InsP, InsP2, and InsP3, respectively. Fractions of 1. ml each were collected. (Inset) Expanded depiction of region C.

3 Biochemistry: Lochner et al. Table 1. Labeling of the inositol phospholipids of guinea pig neutrophils with [3H]inositol Radioactivity, Specific activity, Phospholipid Mass, nmol cpm cpm/nmol PtdIns 3. ± ,945 ± ± 444 PtdlnsP 2.2 ±.24 6,167 ± ± 443 PtdInsP2 1.4 ±.1 2,236 ± ± 229 Guinea pig neutrophils were labeled with [3H]inositol, and the inositol phospholipids were subsequently extracted and quantified. Data are expressed as mean ± SEM for three different preparations of cells. Values are for 18 cells. Ptdlns, phosphatidylinositol; PtdlnsP, phosphatidylinositol 4-phosphate; PtdInsP2, phosphatidylinositol 4,5-bisphosphate. produced only marginal effects on the levels of these substances. The time periods chosen in Table 2 to compare the levels of inositol phosphates in cells stimulated with various agents were those at which the cells responded maximally in terms of O- release (Fig. 1). The distribution of [3H]inositol in the membrane lipids (inositol phospholipids) was examined in cells stimulated for 7 min with all-trans-retinal (25 um), and compared to the values observed in unstimulated cells. The percentages of radioactivity in phosphatidylinositol, phosphatidylinositol 4-phosphate, and phosphatidylinositol 4,5-bisphosphate were 93 ± 12, 51 ± 8, and 75 ± 6 (SEM; n = 3), respectively, Proc. Natl. Acad. Sci. USA 83 (1986) 7675 Table 2. Effects of various stimulators on inositol phosphate accumulation in neutrophils Effect of stimulation, % of control Stimulant InsP InsP2 InsP3 all-trans-retinal (25 AM) 149 ± ± ± 49 fmet-leu-phe (1 MM) 141 ± ± ± 5 PMA (.5 MM) 115 ± ± 2 92 ± 1 Neutrophils labeled with [3H]inositol were incubated with the stimuli listed for a period of time during which each agent evoked maximal superoxide release. These time periods for all-trans-retinal, PMA, and fmet-leu-phe were 7 min, 1 min, and 3 sec, respectively (Fig. 1). Reactions were terminated and the levels of [3H]inositol phosphates were determined as described in Methods. Data are expressed as the percentages of the values for these substances in unstimulated cells. The data for all-trans-retinal are the means + SEM of three separate experiments. The data for fmet-leu-phe and PMA represent mean values ± ranges for two separate experiments. Each separate experiment used a different preparation of cells. of the values observed for unstimulated cells. (The last two values are significantly lower than 1%, P <.1.) Effect of all-trans-retinal on Protein Kinase C Activity. Treatment of intact cells with retinal (25,uM) for 1 min had no effect on the total amount of protein kinase C activity that could be recovered-for example, by lysis of the cells in the presence of 5 mm EDTA. The values for recovered protein kinase C were as follows: control cells, 6389 ± 873 (n = 6); retinal-treated cells, 5814 ± 839 (n = 6) (pmol of Pi per 4 min per 18 cells ± SEM). The physiological substrates of protein kinase C in neutrophils are not presently known. We therefore evaluated the effects of all-trans-retinal on this kinase in vitro by using an exogenous substrate. Neutrophil protein kinase C activity is in the cytosol (25); it is stimulated severalfold by calcium in the presence of phosphatidylserine and diolein (protein kinase C) (Fig. 4). Calcium alone produced some stimulation (=5%), whereas diolein alone was without effect on the enzyme activity (unpublished data). all-trans-retinal at 1 and 1 AM inhibited this protein kinase C activity by at most 2% and 47%, respectively, and did not significantly affect 6. x E 5. C.) >' 4. C 3. 4 Time, min FIG. 3. Time course of accumulation of inositol phosphates in neutrophils stimulated with all-trans-retinal. Neutrophils labeled with [3H]inositol were stimulated with all-trans-retinal (25 MuM), and the 3H content of InsP (x), InsP2 (e), and InsP3 (o) was determined at the times indicated. Inositol phosphates were separated as described in Fig. 2 and the total radioactivity in regions A, B, and C was determined. Data are expressed as the percentages of the respective values for these substances in unstimulated cells. Values are means ± SEM of three independent determinations performed on three different cell preparations. The 3H contents of InsP, InsP2, and InsP3 in unstimulated cells at zero time were 6166 ± 54, 198 ± 26, and 287 ± 42, respectively. These values did not change significantly upon incubation of the cells at 37 C in the absence of all-trans-retinal. (Inset) Derivative of the progress curve for 2 release from neutrophils stimulated with all-trans-retinal (25 MM) (see Fig. 1). Rate is expressed as nmol of O2 per min per 17 cells..e 2. Z 1. A u o6 1-1 all-trans-retinal, M FIG. 4. Effect of all-trans-retinal on activity of protein kinase C from neutrophil cytosol. The hatched bar depicts the level of protein kinase C activity from the cytosol of neutrophils measured in the presence of Ca2l (.5 mm), phosphatidylserine (2 ug/ml), and diolein (.8,g/ml) (complete system). Stippled bar represents the level of activity exhibited in the absence of these components. The same value was also observed in the absence of the cytosolic fraction. All other assay conditions are described in Methods., Effect of different concentrations of all-trans-retinal on protein kinase C activity in the complete system; o, effect of all-trans-retinal at various levels on protein kinase activity in the absence of Ca2", phosphatidylserine, and diolein.

4 7676 Biochemistry: Lochner et al. the basal activity (i.e., activity measured in the absence of Ca2", diolein, and phosphatidylserine) (Fig. 4). Under identical conditions, all-trans-retinal at 1 ttm completely inhibits protein kinase C from cultured pituitary cells and increases the basal kinase activity severalfold (6). The assay conditions used in Fig. 4 were chosen for comparative purposes but are not optimal. No inhibition of protein kinase C activity from neutrophils by all-trans-retinal (1 or 1 AM) was observed when optimal conditions (i.e., those used in ref. 26) were employed (W.H., unpublished data). DISCUSSION We have recently reported that compounds known to increase the liquid crystalline fraction of membranes [e.g., cis-unsaturated fatty acids (27, 28), retinoids (4, 5)] induce high levels of release by neutrophils (17, 29). Cisunsaturated fatty acids stimulate purified phosphatidylinositol-specific phospholipase C activity in vitro (3, 31), and this effect has been attributed to alteration of the physical properties of the lipid substrate (3). The loss of cellular [3H]inositol phospholipids (see text) and the accumulation of [3H]inositol phosphates (Fig. 3; Table 2) upon treatment of neutrophils with all-trans-retinal are consistent with activation of phospholipase C in intact neutrophils by this substance. Similar studies on inositol phospholipid metabolism of neutrophils stimulated with cis-unsaturated fatty acids were not undertaken since these stimuli have damaging effects on cells shortly (c3 min) after stimulation. Phosphatidylinositol-specific phospholipase C catalyzes the hydrolysis of phosphatidylinositol 4,5-bisphosphate to produce diglyceride and InsP3. Both products have important "second messenger" roles in stimulating certain cellular responses (for review, see ref. 32). As noted earlier, diglyceride is the physiological activator of protein kinase C. InsP3 stimulates the release of Ca2+ from intracellular storage sites. These messengers may interact synergistically to stimulate neutrophils to produce (for review, see ref. 14). Neutrophils stimulated with all-trans-retinal may similarly utilize these "second messengers," since all-trans-retinal does not significantly inhibit protein kinase C from these cells at concentrations at which it elicits production (Fig. 4). This contrasts with the situation regarding the kinase obtained from other sources (6). Previous studies have shown that neutrophils treated with fmet-leu-phe degrade their inositol phospholipids and accumulate inositol phosphates, with the maximal effects occurring at 2-3 sec after stimulation (19, 33-35). This rapid response is compatible with the absence of a lag period in the progress curve for release with this stimulus (Fig. 1). In contrast, neutrophils stimulated with all-trans-retinal exhibit a pronounced lag period in release (Fig. 1) and do not exhibit maximal accumulation of labeled InsP2 and InsP3 until several minutes after stimulation (Fig. 3). The lag period of 1 release may therefore correlate with attainment of a critical level of a particular component. In our experiments, fmet-leu-phe and all-trans-retinal induced similar levels of inositol phosphates, with InsP2 predominating in the case of both stimuli (Table 2). Reasons for the apparently high levels of labeled InsP2 (measured as 3H) may include the larger quantity and greater specific activity of the substrate phosphatidylinositol 4-phosphate compared to phosphatidylinositol 4,5-bisphosphate (Table 1), the possibility that phosphatidylinositol 4-phosphate is the preferred substrate (36), and/or the existence of phosphatases of high activity in guinea pig neutrophils that degrade InsP3 (37). Retinoids at concentrations similar to those used here affect differentiation and growth of normal and malignant cells, even in cells that lack specific cytosolic-binding proteins for these substances (ref. 38 and refs. cited therein). In Proc. Natl. Acad. Sci. USA 83 (1986) this paper, we report an effect of all-trans-retinal on a particular cell type (i.e., activation of phospholipase C). In addition, we show that the neutrophil protein kinase C is largely insensitive to all-trans-retinal when examined in cell extracts (Fig. 4). Studies to determine whether retinoids activate phospholipase C in cells other than neutrophils would be of interest. This work was supported by U.S. Public Health Service Grant 1 ROI GM J.A.B. is the recipient of a Research Career Development Award (1 K4 AI 672-1) from the National Institute of Allergy and Infectious Diseases. W.H. was supported by the Deutsche Forschungsgemeinschaft. 1. Goodman, D. S. (1984) N. Engl. J. Med. 31, Sporn, M. B. & Roberts, A. B. (1984) J. Natl. Cancer Inst. 73, DeLuca, L. B. (1977) Vitam. Horm. (N. Y.) 35, Stillwell, W., Ricketts, M., Hudson, H. & Nahmias, S. (1982) Biochim. Biophys. Acta 699, Valtersson, C., van Diyn, G., Verkleij, A. J., Chojnacki, T., de Kruijff, B. & Dallner, G. (1985) J. Biol. Chem. 26, Hirota, K., Hirota, T., Aguilera, G. & Catt, K. J. (1985) J. Biol. Chem. 26, Taffet, S. M., Greenfield, A. R. L. & Haddox, M. K. (1983) Biochem. Biophys. Res. Commun. 114, Kapoor, C. L. & Chader, G. J. (1984) Biochem. Biophys. Res. Commun. 122, Vandenbark, G. R., Kuhn, L. J. & Niedel, J. E. (1984) J. Clin. Invest. 73, Castagna, M. Y., Takai, K., Kaibuchi, K., Sano, U., Kikkawa, Y. & Nishizuka, Y. (1982) J. Biol. Chem. 257, Repine, J. E., White, J. G., Clawson, C. C. & Holmes, B. M. (1974) J. Clin. Invest. 54, Robinson, J. M., Badwey, J. A., Karnovsky, M. J. & Karnovsky, M. L. (1985) J. Cell Biol. 11, Badwey, J. A. & Karnovsky, M. L. (198) Annu. Rev. Biochem. 49, Badwey, J. A. & Karnovsky, M. L. (1986) Curr. Topics Cell. Regul. 28, Kensler, T. W. & Mueller, G. C. (1978) CancerRes. 38, Verma, A. K., Shapes, B. G., Rice, H. M. & Boutwell, R. K. (1979) Cancer Res. 39, Badwey, J. A., Robinson, J. M., Curnutte, J. T., Karnovsky, M. J. & Karnovsky, M. L. (1986) J. Cell. Physiol. 127, DePierre, J. W. & Karnovsky, M. L. (1974) J. Biol. Chem. 249, Bradford, P. G. & Rubin, R. P. (1985) Mol. Pharmacol. 27, Berridge, M. J., Dawson, R. M. C., Downes, C. P., Heslop, J. P. & Irvine, R. F. (1983) Biochem. J. 212, Schacht, J. (1981) Methods Enzymol. 72, Palmer, F. B. St. C. (1981) J. Lipid Res. 22, Rittenhouse, S. E. & Allen, C. L. (1982) J. Clin. Invest. 7, Ames, B. N. (1966) Methods Enzymol. 8, Wolfson, M., McPhail, L. C., Nasrallah, V. N. & Snyderman, R. (1985) J. Immunol. 135, Hannun, Y. A., Loomis, C. R. & Bell, R. M. (1985) J. Biol. Chem. 26, Usher, J. R., Epand, R. M. & Papahadjopoulos, D. (1979) Chem. Phys. Lipids 22, Klausner, R. D., Kleinfeld, A. M., Hoover, R. L. & Karnovsky, M. J. (198) J. Biol. Chem. 255, Badwey, J. A., Curnutte, J. T., Robinson, J. M., Berde, C. B., Karnovsky, M. J. & Karnovsky, M. L. (1984) J. Biol. Chem. 259, Irvine, R. F., Hemington, N. & Dawson, R. M. C. (1979) Eur. J. Biochem. 99, Takenawa, T. & Nagai, Y. (1981) J. Biol. Chem. 256, Berridge, M. J. & Irvine, R. F. (1984) Nature (London) 312, Volpi, M., Yassin, R., Naccache, P. H. & Sha'afi, R. I. (1983)

5 Biochemistry: Lochner et al. Biochem. Biophys. Res. Commun. 112, Dougherty, R. W., Godfrey, P. P., Hoyle, P. C., Putney, J. W. & Freer, R. J. (1984) Biochem. J. 222, Ohta, H., Okajima, F. & Ui, M. (1985) J. Biol. Chem. 26, Wilson, D. B., Bross, T. E., Hofmann, S. L. & Majerus, Proc. Natl. Acad. Sci. USA 83 (1986) 7677 P. W. (1984) J. Biol. Chem. 259, Badwey, J. A., Sadler, K. L., Robinson, J. M., Karnovsky, M. J. & Karnovsky, M. L. (1986) Fed. Proc. Fed. Am. Soc. Exp. Biol. 45, 1191a (abstr.). 38. Oreffo, R.. C., Francis, M. J. A. & Triffitt, J. T. (1985) Biochem. J. 232,