Glucosphingolipid dependence of hormone-stimulated inositol trisphosphate formation.

The modulatory role of endogenous cellular glycosphingolipids in bradykinin-stimulated myo-inositol 1,4,5-trisphosphate (InsP3) formation by MDCK cells was evaluated utilizing the glucosylceramide synthase inhibitor, threo-1-phenyl-2-decanoylamino-3-morpholino-1-propanol (PDMP). Bradykinin-stimulated InsP3 formation in intact cells and in isolated plasma membranes was significantly enhanced when cells were first depleted of their glucosphingolipids. The effect of glucosphingolipid depletion on phospholipase C activity was dependent on the duration of exposure to the inhibitor and the cellular level of glucosylceramide. Inclusion of glucosylceramide in the culture medium prevented the stimulatory effect of PDMP on InsP3 formation. It is concluded that membrane glucosphingolipids may regulate phospholipase C activity.

The modulatory role of endogenous cellular glycosphingolipids in hradykinin-stimulated myo-inositol 1,4,5-trisphosphate (InsPa) formation by MDCK cells was evaluated utilizing the glucosylceramide synthase inhibitor, threo-1-phenyl-2-decanoylamino-3-morpholino- . Bradykinin-stimulated InsP, formation in intact cells and in isolated plasma membranes was significantly enhanced when cells were first depleted of their glucosphingolipids. The effect of glucosphingolipid depletion on phospholipase C activity was dependent on the duration of exposure to the inhibitor and the cellular level of glucosylceramide.
Inclusion of glucosylceramide in the culture medium prevented the stimulatory effect of PDMP on InsPs formation.
It is concluded that membrane glucosphingolipids may regulate phospholipase C activity.
Glycosphingolipids are ubiquitous and structurally diverse compounds.
Glucosylceramide is a fundamental glycosphingolipid made from ceramide and UDP-glucose by a glucosyltransferase.
It has been stated that there are at least 300 glycosphingolipids, including gangliosides, most of them formed from glucosylceramide by the attachment of additional sugars and sulfate. Known properties of these lipids include binding to viruses (I), toxins (2,3), bacteria (4), and matrix proteins (5), the regulation of ionic transport (6), and immunoprotection (7). Additionally, these compounds have been implicated in a variety of cellular growth and differentiation phenomena (8) including cancer (9). Tyrosine kinase activities stimulated by platelet-derived growth factor and epider-ma1 growth factor have been shown to be modulated by structurally distinct gangliosides (10). Protein kinase C, like tyrosine kinases, has recently been shown to be modulated by the addition to cell cultures of sphingolipids, particularly the free base, sphingosine isolated following exposure to hypotonic buffer by the protocol of Hepler and Harden (18). Prior to the exposure to agonist, the membranes were pelleted at 12,000 x g. Protein was determined utilizing the fluorescamine reagent (19). Inositol 2,4,5Trisphosphate Determination-Hormone-stimulated InsP:, formation was measured in intact MDCK cells as previously detailed (20). For broken cell studies, membranes were suspended in a buffer consisting of 10 mM Hepes, 2 mM EGTA, 424 FM CaC&, 0.91 mM MgSO,, 115 mM KCl, and 5 mM KH2P04, pH 7.0. Thereafter additions consisting of the same buffer, containing 1.0 pM GTP-yS, 0.1 PM bradykinin, or GTPrS and bradykinin together, were made. After 10 min 6% trichloracetic acid was added, and the resultant precipitate was sedimented by centrifugation at 5000 X g. The supernatant was treated with trioctylamine-Freon to extract the acid. InsPs formation was measured by a competitive binding assay utilizing high specific activity myo-[2-3H]inositol 1,4,5-trisphosphate binding to calf adrenocortical microsomes following the method of Palmer et al. (21) as recently applied for measurements in intact MDCK cells (20).

AND DISCUSSION
Madin-Darby canine kidney cells generate InsPs and an increase in intracellular calcium in response to bradykinin (22). The formation of InsPs in these cells has been shown to be regulated by a guanine nucleotide regulatory protein (22). These cells can be grown in defined (serum-free) media, and therefore exposure to exogenous sphingolipids can be strictly controlled. The cells were exposed to DL-or D-PDMP for 24 h and then metabolically labeled in the presence of PDMP another 24 h to yield an estimate of the rates of synthesis of the major sphingolipids. Exposure of cells to DL-PDMP resulted in a marked decrease in the conversion of [3H]galactose to ["Hlglucosylceramide ( Fig. 1). This change in radiolabeling paralleled changes in cerebroside mass as evidenced by charring of TLC plates (Fig. 1). Decreased labeling of acidic glycolipids, including ganglioside GM3 , which utilize glucosylceramide as their precursor was also observed (Fig. 2). Ceramide and sphingomyelin labeling with [3H]palmitate was observed to increase (Table I). Apparently cells do not possess a readily available feedback mechanism for slowing the synthesis of ceramide when its rate of utilization is reduced. The rate of utilization for sphingomyelin synthesis seems to be insufficient to prevent ceramide accumulation.
When cells were stimulated with lo-' M bradykinin, maximal InsPs formation was observed at 15 s (data not shown). Cells which had been exposed to 20 pM PDMP for 24 h and    3. InsPa formation at 15 s in glucosphingolipid-depleted cells exposed to lo-' M bradykinin.
Basal levels of InsPs were not different between control and PDMP-treated cells. However, statistically significant differences in stimulated InsPa levels were observed by the paired t test. *, p < 0.05 uersus unstimulated control (CON) and PDMP-treated cells; 3, p < 0.05 uersus control cells; n, p < 0.05 versus control bradykinin (BK)-stimulated cells.
then stimulated with bradykinin also showed maximal production at 15 s. Multiple determinations of InsPs accumulation 15 s after stimulation demonstrated a statistically significant increase in the product by the glucosphingolipid-depleted cells (Fig. 3). The stimulatory effects of glucosphingolipid depletion were further assessed in membranes isolated from MDCK cells following 24 h of exposure of the cells to 20 PM PDMP. The membranes were incubated for 10 min with buffer, 10e6 M  (32) 19.5 (290) 10.4 (24) 13.6 (527) 5.60 (57) 45. 3 (806) 19.6 (133) 28.9 (1230) 6.27 (76) 50.2 (904) 18.5 (120) 16.3 (651) 8.43 (136) 77. 3 (1450) 29.6 (252) 42. 4 (1850) " p < 0.05 uersus parallel control conditions. GTP+, 10m7 M bradykinin, or GTPyS and bradykinin together. The formation of InsPs was stimulated in the control membranes by both agonists, particularly the combination of the two (Table II). The stimulatory effects of GTP-yS and bradykinin were enhanced in glucosphingolipid-depleted membranes. This observation is consistent with the interpretation that enhanced InsPB levels are due to an increase in phospholipase C activity and not simply the result of impaired metabolism of InsPs. This effect was not the result of enhanced phosphatidylinositol 4,5-bisphosphate formation since levels of this lipid were unchanged in control and glucosphingolipid-depleted cells. For these experiments MDCK cells were grown in the presence of 20 PM PDMP for 24 h. Phospholipids were extracted by the addition of chloroform, methanol, 1 N HCl (1:2:1) to scraped cells. The aqueous phase was extracted with chloroform:methanol (2:1), and the lower phases were pooled and evaporated under nitrogen. Phosphatidylinositol4,5-bisphosphate was separated by HPTLC (23) and quantitated by phosphate analysis (24). Under these conditions control levels were 1.42 f 0.16 nmol/mg protein; PDMP cells contained 1.55 f 0.19 nmol/mg protein (n = 10, p = 0.57 by the unpaired t test).
The enhanced formation of InsPB was dependent on the duration of exposure of the cells to the glucosylceramide synthase inhibitor. Increased formation of hormone-stimulated InsPa paralleled changes in glucosylceramide mass (Fig.  4). The close temporal association between decreased glucosylceramide and increased InsPs suggests that glucosylceramide itself or a closely related sphingolipid mediates the observed changes. Finally, addition of glucosylceramide to the culture medium prevented the stimulatory effect of glucolipid depletion (Table III). Galactosylceramide, which is also absorbed by the cells and hydrolyzed to ceramide, fatty acid, and sphingosine,' did not block the stimulatory action of ' Incorporation of exogenously added cerebrosides into MDCK cells was documented by incubating cells with liposomes containing galactosylceramide under identical conditions to those described in Table III MDCK cells were exposed to control medium or to medium containing 20 pM PDMP for 6, 12, 24, or 48 h. A, charred TLC plates showing cerebroside levels; B, InsP3 formation by isolated plasma membranes following exposure to GTP-# and bradykinin (0) or buffer alone (Cl) as described in Table II. C denotes control conditions; P denotes PDMP-exposed cells. The data represent the mean of duplicate determinations. InsPB content (pmol/mg protein) was measured in MDCK cells treated with D-PDMP as described in Table II. The cells were simultaneously incubated for 24 h with PDMP and liposomes made from either phosphatidylcholine alone (lines 1 and 2) or from phosphatidylcholine plus glucosyl-or galactosylceramide. Cerebroside incorporation was documented with ["Hlglucosyl-and ["Hlgalactosylceramide containing liposomes in parallel incubations under identical conditions. The liposomes consisted of egg phosphatidylcholine (3 mg) and cerebroside (1 mg) in 1 ml of 20 mM Tris-Cl, pH 7.4. Sufficient liposomes were added to produce a final concentration of 1 pM cerebroside. Plasma membranes were then isolated and InsPa formation determined as in Table II PDMP. This experiment confirms the assumption that the PDMP effect on phospholipase C is due to a cellular deficiency in glucosylceramide. In many tissues the concentration of phospholipase C is sufficiently high, if assayed under ideal conditions in uitro, to cause the rapid hydrolysis of the entire pool of cellular inositol lipids within seconds. On this basis it has been postulated that there exists an endogenous negative regulator of phospholipase C (25). Our experiments support the hypothesis that glycosphingolipids may serve as such regulators. Specifically, the findings suggest that cellular glucosphingolipids, when present at normal levels in MDCK cells, regulate the hormone-stimulated formation of InsP3.

Sphingolipid Modulation of Inositol T&phosphate Formation
Several potential mechanisms may explain the enhanced generation of InsP, under glucosphingolipid-depleted conditions. First, alterations in the plasma membrane sphingolipid content may directly affect the activity of phospholipase C. The activity of phospholipase C in hydrolyzing phosphatidylcholine in artificial membranes, for example, was observed to be dependent on the content of ceramide (26). Second, enhanced phospholipase C activity may be the indirect result of inhibition of protein kinase C. The activation of protein kinase C has recently been shown to produce a decrease in phospholipase C activity in several cell types (27). Free sphingosine and N-deacylated glycolipids inhibit protein kinase C and may accumulate under conditions of glucosylceramide synthesis inhibition. Third, the y-isoenzyme of phospholipase C is a substrate for epidermal growth factor-and platelet-derived growth factor-mediated tyrosine kinase activity (28). These kinase activities have been demonstrated to be modulated by the exogenous addition of a variety of sphingolipids (10). Thus the observed changes in phospholipase C activity may be secondary to the changes in tyrosine kinase activity.
In summary, these data demonstrate the modulation of hormone-stimulated InsPa formation by endogenous glycosphingolipids. Although the mechanisms whereby glucosylceramide levels regulate bradykinin-stimulated phospholipase C activity remain to be elucidated, glycosphingolipid metabolism appears to play an important role in the regulation of InsPa generation.