Sphingomyelin Turnover Induced by Vitamin D3 in HL-60 Cells ROLE IN CELL DIFFERENTIATION*

Sphingolipid metabolism was examined in human promyelocytic leukemia HL-60 cells. Differentiation of HL-60 cells with la,25-dihydroxyvitamin D3 (vita-min D3; 100 nM) was accompanied by sphingomyelin turnover. Maximum turnover of [3H]choline-labeled sphingomyelin occurred 2 h following vitamin D3 treatment, with sphingomyelin levels decreasing to 77 -C 6% of control and returning to base-line levels by 4 h. Ceramide and phosphorylcholine were concomi- tantly generated. Ceramide mass levels increased by 55% at 2 h D3 and returned to base-line levels by 4 h. The amount of phos- phorylcholine produced equaled the amount of sphingomyelin hydrolyzed, suggesting the involvement of a sphingomyelinase. Vitamin D3 treatment resulted in a 90% increase in the activity of a neutral sphingomye- linase from HL-60 cells. The inferred role of sphingomyelin hydrolysis in the induction of cell differentia- tion was investigated using an exogenous sphingomyelinase. When a bacterial sphingomyelinase was added at concentrations that caused a similar degree of sphingomyelin nM vitamin D3, the ability of subthreshold of vitamin D3 to HL-60 cell

Sphingolipids and sphingolipid breakdown products are emerging as a new class of bioactive molecules that affect cell regulation, secretion, cell differentiation, and oncogenesis (1-3). Recently, we hypothesized the existence of a "sphingolipid cycle," analogous to the phosphatidylinositol cycle (I). This hypothesis led us to examine whether sphingolipid turnover occurred during human promyelocytic leukemia HL-60 cell differentiation induced by la,25-dihydroxyvitamin DB (vitamin D3).
In this report, vitamin DB is reported to induce the turnover of sphingomyelin (SM).' Phosphorylcholine and ceramide are the products of SM turnover, which occurs secondary to the activation of an endogenous neutral sphingomyelinase Grants EL00155, CA46738, and DK20205. The costs of publication * This work was supported in part by National Institutes of Health of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
7 To whom correspondence should be addressed Div. of Hematology/Oncology, Box 3355, Duke University Medical Center, Durham, NC 27710.
(SMase). Addition of bacterial neutral SMase enhanced the action of subthreshold vitamin D3 in inducing HL-60 cell differentiation. These results suggest that SM hydrolysis and its breakdown products may play a previously unrecognized role in cell differentiation.

Methods
Labeling of HL-60 Cells-HL-60 cells were grown in RPMI 1640 medium containing 10% fetal calf serum in 5% CO2 at 37 'C. The cells were washed three times with phosphate-buffered saline and incubated with [3H]palmitic acid (10 pM, 1 pCi/ml; specific activity: 200 mCi/ml) for 12 h or with [3H]choline chloride (0.5 pCi/ml; specific activity: 80 Ci/ml) for at least 48 h in serum-free RPMI 1640 medium containing insulin (5 mg/liter) and transferrin (5 mg/liter). The cells were then washed three times with phosphate-buffered saline and incubated in serum-free medium in the presence or absence of 100 nM vitamin D3.
Lipid Extraction and identification-After harvesting the cells at the indicated times, the lipids were extracted by the method of Bligh and Dyer (4). The samples were dried down under N, gas and dissolved in 100 p1 of chloroform; then, 20 pl were applied on Silica Gel 60 TLC plates (Merck), and 40 pl were used to measure phospholipid phosphate (duplicate measurements) (5). To identify SM and PC, TLC plates were developed in chloroform/methanol/acetic acid/ H 2 0 (50:30:8:5; solvent system A); chloroform, methanol, 2 N NHlOH (60:35:5; solvent system B); or chloroform/methanol/H20 (65:25:4; solvent system C). The combination of solvent systems A and B or A and C was also used for two-dimensional TLC separation. SM was further identified on TLC by alkaline hydrolysis. Chloroform extracts of cells were saponified in methanolic NaOH (0.1 N) at 37 "C for 1 h to eliminate ester-containing glycerolipids and then alkaline-hydrolyzed in methanolic NaOH (1 N) at 120 "C for 20 h to remove the Nacyl chains yielding sphingosylphosphorylcholine, which co-migrated with pure standard. The SM and PC spots were scraped and counted in 4 rnl of Safety-Solve (Research Products International Corp.) using a scintillation counter (Pharmacia LKB, Biotechnology Inc., 1209 RACKBETA). Radioactivity was corrected for the amount of phospholipids.
Sphingomyelin and Phosphatidylcholine Quantitation-Phospholipids were isolated and separated on TLC as described above. The sphingomyelin and phosphatidylcholine spots were scraped, and the lipids were eluted from silica gel in chloroform/methanol (2:l). SM and PC were then quantitated by measuring their phosphate content (5).
Ceramide Quantitation-Ceramide levels were measured enzymatically by using DAG kinase as described (6,22). Base-line ceramide SM was diluted with unlabeled SM to a specific activity of 10,000 cpm/nmol.
Assay of Neutral Sphingomyelinase-HL-60 cells were harvested; washed twice with phosphate-buffered saline; and suspended in 0.5 ml of 10 mM Tris/HCl (pH 7.5), 1 mM EDTA, and 0.1% Triton X-100 after treatment with 100 nM vitamin DS or with vehicle. The cells were homogenized by 30 strokes in a Dounce glass homogenizer and centrifuged at 100,000 X g for 1 h at 4 "C. The supernatant was used as an enzyme source. The assay mixture for the measurement of neutral sphingomyelinase contained 0.1 M Tris/HC1 (pH 7.5), 60 nmol of [methyl-"CISM, 6 mM MgCI,, 0.1% Triton X-100, and 50-300 pg of enzyme in a final volume of 0.1 ml. Incubation was carried out at 37°C for 30 min. The reaction was stopped by adding 1.5 ml of chloroform/methanol(2:1). Then, 0.2 ml of double distilled water was added to the tubes and vortexed. The tubes were centrifuged at 1,000 X g for 5 min to separate the two phases. The clear upper phase (0.4 ml) was removed and placed in a glass scintillation vial. Ten ml of scintillation fluid (Safety-Solve) were added. After shaking, the vials were counted. Control tubes contained boiled enzyme. Protein was measured by the method of Lowry et al. (9) with bovine albumin as a standard.
Analysis of Cell Differentiation-Nitro blue tetrazolium-reducing ability and nonspecific esterase activity were quantified as previously described (10).

RESULTS AND DISCUSSION
Detection of Sphingomyelin Turnover-Initially, we examined whether changes in sphingolipids occurred during the early phase of vitamin D3-induced HL-60 cell differentiation. Cells were labeled with [3H]palmitic acid and then treated with 100 nM vitamin D3, an optimal concentration for induction of differentiation. Among the labeled sphingolipids, SM showed significant changes in labeling during the first 4 h after addition of vitamin D3.' SM levels decreased to 77 k 6% of control 2 h after treatment and then returned to control levels by 4 h (Fig. 1A). SM metabolism was also followed by Metabolic Pathways Involved in Sphingomyelin Hydrolysis-SM turnover could result from a number of biochemical reactions ( Fig. 2 A ) . These include a sphingomyelinase ( Fig.  2.4 (Fig. 2B). Ceramide Mild alkaline hydrolysis of [3H]palmitic acid-labeled lipids selectively yields sphingolipids and alkyllysophospholipids. A number of these lipids changed on treatment of HL-60 cells with vitamin DB. Of these, sphingomyelin showed the earliest and most pronounced changes. production is consistent with activation of either SMase (reaction I) or an exchange enzyme (reaction IV).

Vitamin DS Induces Sphingomyelin Turnover
[3H]Cholinelabeled PC did not show any significant change during the same time interval (Fig. Therefore, it seems unlikely that the exchange enzyme was involved in SM turnover. This is further supported by the concomitant changes in the level of phosphorylcholine, a product of sphingomyelinase reaction Total cellular PC levels (400-500 pmol/nmol of phospholipids) were 10-fold greater than total SM levels as measured by head group phosphates. Therefore, changes in PC levels which could account for significant changes in SM levels could not be accurately assessed by choline labeling. Evidence against reaction IV is obtained by the changes in phosphorylcholine and from the detection of SMase of the phospholipase C-type in crude extracts of HL-60 cells treated with vitamin D3.
but not of the exchange reaction (Fig. 3). Moreover, the amount of SM breakdown (e.g. 251 f 13 cpm/nmol of phospholipids at 2 h) corresponded to that of phosphorylcholine generation in the cells (228 k 56 cpm/nmol of phospholipids at 2 h) at the same time points (Fig. 3). Therefore, these results suggest activation of a SMase by vitamin Ds. The mass of hydrolyzed SM was then compared to the mass of generated ceramide. Vitamin D3 treatment of HL-60 cells resulted in a maximum decrease of SM (32.9 & 4.0%), which corresponds to hydrolysis of 17 * 0.43 pmol of SM/nmol of phospholipids at 2 h. This was accompanied by the generation of 14 f 2.8 pmol of ceramide/nmol of phospholipids. These results quantitatively demonstrate that ceramide is the predominant product of SM hydrolysis. Other potential minor products, such as ceramide phosphate or lysosphingomyelin, were not detected.
SM mass levels returned to base-line levels (51 f 5.9 pmol/ nmol of phospholipids) by 4 h, indicating a resynthesis phase of SM accompanied by decreases in the levels of ceramide.4 These results are therefore consistent with the existence of a sphingomyelin cycle in response to vitamin D3 action.
Activation of Neutral Sphingomyelinase by Vitamin D3-Because of the inherent limitations of metabolic labeling studies, we next investigated the presence of endogenous SM hydrolyzing activity. Detergent extracts of HL-60 cells contained acid and neutral sphingomyelinase. Treatment of HL-60 cells with vitamin D3 resulted in a time-dependent increase in the neutral SMase activity (Fig. 4) which peaked at 1.5-2 h (no SM N-deacylase or D-type SMase was detected). These data show that the predominant effect of vitamin DS is the induction of SMase activity. ' The recovery of SM and ceramide levels by 4 h suggests that SM resynthesis occurs either by reaction IV or through de nouo synthesis (reaction V). did not induce DAG accumulation.' Exogenous (bacterial) SMase acted in synergy with vitamin D3 in inducing HL-60 differentiation. Simultaneous addition of 100 milliunits/ml SMase and 1 nM vitamin D3, a subthreshold concentration which barely induces differentiation, caused partial differentiation (Fig. 5). Concomitant with cell differentiation, there was a significant decrease in cell proliferation (Fig. 5, inset). Two parameters of differentiation showed significant changes. Nitro blue tetrazolium-reducing ability increased from 4 k 2 to 19 -+ 6%, and nonspecific esterase activity increased from 0 f 1 to 26 +-3% after 4 days. The synergy between vitamin D3 (1 nM) and SMase (1-100 milliunits/ml) was dose-dependent (Fig. 5). SMase itself did not show any induction of HL-60 cell differentiation in the absence of vitamin D3 (data not shown).6 Therefore, hydrolysis of SM by exogenous SMase synergistically enhanced the action of a subthreshold concentration of vitamin D3 in inducing HL-60 cell differentiation.

CONCLUDING REMARKS
Increases in SM levels have been observed in dexamethasone-treated neutrophils (11) and 3T3-Ll cells (12) and in HL-60 cells treated with phorbol12-myristate 13-acetate (13). The functional significance of these changes, however, has not been determined. In addition, it has been shown that high concentrations of DAG stimulate SM hydrolysis in GH3 pituitary cells (14). This effect was not reproduced by phorbol esters and occurred in cells where protein kinase C was downregulated, suggesting that SM hydrolysis occurred independently of protein kinase C activation. In similar studies (15), it was shown that exogenous bacterial SMase reduced membrane-associated protein kinase C activity, suggesting a role for SM hydrolysis in inhibiting protein kinase C activation. SMase treatment of HL-60 cells also caused an inhibition of phorbol 12-myristate 13-acetate-induced differentiation of HL-60 cells (16).
In this study, we show the induction of SM hydrolysis by vitamin D3 in HL-60 cells. This is accompanied by the generation of ceramide and phosphorylcholine. A neutral sphingomyelinase, detected in extracts of HL-60 cells, was induced by vitamin D3 treatment. SM, ceramide, and phosphorylcholine levels returned to base-line levels by 4 h, suggesting a resynthesis phase of SM, thus completing a sphingomyelin cycle. We also demonstrate a role for SM hydrolysis in enhancing HL-60 differentiation.
This observed turnover of SM may indicate the operation SMase (1-100 milliunits/ml) did not alter DAG mass measured by the method of Preiss et al. (6). Higher concentrations of SMase caused partial hydrolysis of PC and generation of DAG. These results suggest that phospholipase C activity of SMase may limit its selectively at higher concentrations, and care should be exercised in its use.
The inability of exogenous SMase to completely mimic the differentiating effects of 100 nM vitamin D3 (although both vitamin D3 and exogenous SMase result in similar hydrolysis of SM) suggests that the activation of SMase by vitamin D3 is not sufficient for the induction of differentiation or that exogenous SMase cannot fully mimic the intracellular hydrolysis of SM. In any case, the synergistic action of exogenous SMase and vitamin D, supports an important role for SM turnover. of a sphingomyelin cycle with a function in cell regulation. Unlike the phosphatidylinositol cycle, SM turnover occurred over a longer period and may be involved in longer term cell changes such as shown in this study with cell differentiation.
This study raises important questions as to how vitamin D3 regulates SMase activity. Vitamin D3 belongs to the steroid hormone family whose cellular actions are mediated through interaction with intracellular receptors (17-19). These receptors appear to mediate the action of steroid hormones by enhancing/suppressing gene activity (19). Preliminary studies suggest that the effects of vitamin D 3 on the induction of SMase are inhibited by cycloheximide.
Another major question raised by this study relates to the mechanism by which SM hydrolysis and the generation of ceramide and phosphorylcholine enhance cell differentiation. Studies (20) with exogenous SMase suggest an important role for ceramide, either as a second messenger or as a precursor for sphingosine or other metabolites. This latter possibility suggests a link between SM hydrolysis and regulation of protein kinase C since sphingosine inhibits protein kinase C in HL-60 cells (21). Further experiments are required to define the biochemical pathways leading from SM hydrolysis to cell differentiation.
A sphingomyelin cycle and its role in cellular regulation are being defined by this and other studies.