Inhibition of cellular transport systems by alkyl phospholipid analogs in HL-60 human leukemia cells

Biochimica et Biophysica Acta, 1127 (1992) 74-80 © 1992 Elsevier Science Publishers B.V. All rights reserved 0005-2760/92/$05.00

BBALIP 53957

Inhibition of cellular transport systems by alkyl phospholipid analogs in HL-60 human leukemia cells Dennis R. Hoffman ~, Vickey L. Thomas ~ and Fred Snyder b o Department of Pediatrics, Unit~ersityof Texas Southwestern Medical Center, Dallas, TX (USA) and t, Medical and llcalth Sciences Dicision, Oak Ridge Associated Universities, Oak Ridge, TN (USA) (Re~ived 13 November 1991) (Revi~d manuscript received 18 March 1992)

Key words: Alkyl pht~pholipid: Antineoplastic agent; Nutrient transport system

Specific non-metabolizable alkyl-phospholipids selectively kill neoplastic cells, yet normal and more differentiated cells are relatively resistant. Although these highly selective anticanccr lipids appear to target the cell membrane, their mechanism of cytotoxic action remains to be defined. We report here that treatment of 'sensitive' HL-60 leukemia cells with one of the most potent lipid agents, I.alkyl-2.methoxy-glycero-3-phosphocholine,inhibits the cellular transport of multiple essential nutrients including choline, amino acids, fatty acids, and the non-metabolizable carbohydrate, 2-deoxy-o-glucosc. Minimal inhibitory responses of the varied transport systems were noted in HL-60 cells treated with the less potent, 2-1ysoanalog, and in 'resistant' K562 leukemia cells, treated with the 2.methoxy lipid. Although both the 2-methoxy lipid and 12-tetradacanoylphorbol 13-acetate induce differentiation in HL-60 cells, significant differences in the interactions of these lipids on cellular choline transport were found. Ba~d on these results, we conclude that multiple zmtrient deprivation induced by the detergent-like action of the methoxy-containing al~! phospholipid results in the selective destruction of neoplastic cells that are sensitive to this membrane-targeted antitumor agent.

Introduction l-Al~l-2-methoxy-glycero-3-phosphocholine (I-alkyl-2-methoxy-GPC) is the prototype for a group of antineoplastic agents generally termed alkyl-lysophospholipids or ET- 18.OC'H3 [1]` These ether-linked lipids possess highly selective cytotoxie actions towards certain ~ s of cancer cells [2,3]` in studies of a variety of ether lipid analogs screened for their antitumor action, we found the 2-methozy and 2-acetamide glycerophospholipid analogs were 43- to 94-times, respectively, more cytotoxic toward HL-60 human leukemic cells than normal cell types, such as human polymorphonuclear neutrophiis and Detroit 551 human skin fibroblasts [45], Structural requirements for the selective cytotoxic activity of glycerophospholipids include: (a) a long-chain a l ~ l moiety, (b) a quaternary phosphobase,

Abbreviations: GiZ~ m-81ycem-3-phospht~ho!ir,e; TPA, 12-O-tetrad e c a ~ 13-acetate;BSA, bovine serurr: albumin; PBS,phosp~tc-buffered ~line. Conespoadence to: D.R. Hoffman, Retina Foundation of the Southwest, 9900 North C~:ntral, #400, Dallas, "IX 75231, USA.

and; (c) a non-metabolizable group at the 2-position of glycerol [3]. Although the antineoplastic mechanism of action is, as yet undefined, these phospholipids appear to be membrane-targeted [6]. An early hypothesis for the selective antineoplastic action of these ether lipids involved the enzyme that removes the 1 -alkyl group from ether-linked phospholipids, alkyl cleavage enzyme [7-9]. A high activity of the enzyme in 'normal' cell types would degrade the lipid and protect normal cells whereas the lipid would accumulate to cytotoxic levels in susceptible cells. However, we reported earlier that the metabolism of l-al~l-2-methoxy-GPC was very limited in both susceptible cells (HL.60) and relatively resistant cells (K562, MDCK, PMN) and that cleavage of the alkTI moiety cannot explain the selective cytotoxicity of the ether-linked lipid [6]. Metabolic degrad~tion of l.alkyl-2-methoxy-GPC was a maximum of 7%, 8%, 3%, and 13% in HL-60, K562, MDCK cells, and PMN, respectively, over a 24 h incubation period, yet HL-60 cells are 20- to 70-fold more susceptible to destruction by the ether lipid during this incubation interval. These results were subsequently confrmed in HL-60, K562, and Raji human leukemia cells and mouse L1210 leukemia cells [10]` Thus, alternative mechanisms for

75 the selectivity and cytotoxic actions of these bioactive ether lipids are needed. We have reported an unusually high affinity of the plasma membrane of HL-60 cells for l-alkyl-2methoxT-GPC compared to that of neutrophils, supporting the membrune-targeted natvre of the analogs [6], Moreover, other studies have shown that the 2methoxy analog greatly inhibits the incorporation of choline [11-13] and oleic acid [11,14] into phosphatidylcholine of HL-60 cells and that incorporation of thymidine into DNA is markedly depressed in cells treated with the 2-methoxy analog [4,15,16]. These results, coupled with the membrane-targeted nature of the 2-methoxT ether lipid and preliminary studies [17], suggested t~ us that the deleterious effect of the analog could be due to its ability to inhibit cellular transport processes. Therefore, we examined the effect of the 2-metho~ ether lipid on the uptake of various nutrient molecules involving numerous transport systems in cytotoxically 'sensitive' HL-60 cells and a relatively 'resistant' line of human leukemia cells (K562). Our results suggest that these novel antitumor lipids exert a direct cytotoxic action on 'sensitive' neoplastic cells by inhibiting the transport of essential nutrients across their cell membranes. Materials and Methods

HL-60 human promyelocytic leukemia cells and K562 human erythroblastic leukemia cells were obtained from the American Type Culture Collection (ATCC, Rockville, MD). The cells were grown and maintained in RPMi 1640 media (GIBCO, Grand Island, NY)containing 10% heat-inactivated fetal bovine serum. Both cell types tested negative for mycoplasm contamination. The reported study utilized HL-60 cells between the 24th and 34th passages and unknown passages, as obtained from ATCC, for K562 cells. l-Oetadec-9',10'-enyl-2-methoxy-GPC and 1-hexadecyl-2-methoxT-GPC were purchased from R. Berchtold (Biochemisches Labor, Bern, Switzerland). l-Octadec-9',10'-enyl-2-methoxy-GPC was catalytically reduced with tritium by New England Nuclear (Boston, MA) to form l-[9',10'-3H]octadecyl-2-methoxy-GPC (8.2 mCi//~wol). The latter was purified to 94..5% (based on radioactivity) by thin-layer chromatography. l.[methyl-3H]Choline chloride (80 Ci/mmol), 2-amino[1-14C]isobutyric acid (59 mCi/mmol), L-[2,53H]histidine (.~7 Ci / mmol ), [ 1- ~4C]arachidonic acid (58 mCi/mmol) and 2-deoxy-~-[1-3H]glucose (17 Ci/mmol) were obtained from Amersham (Arlington Heights, !L), and 1.[methyl.3H]methionine (80 Ci/mmol) was from New England Nuclear. Unlabeled compounds (chofine ch|oride, 2-aminoisobutyrie acid, L-methionine, L-histidine, araehidonic acid, 2-deoxy-vglucose, 1-alkyl-2-1yso-GPC, 12-O-tetradecanoylphor-

bol 13-acetate (TPA) and fatty acid-free fraction V bovine serum albumin (BSA)were from Sigma Chemicals (St. Louis, MO). To determine the effect of 1-alkyl-2-methoxy-GPC on the cellular uptake of nutrients, cells were first pre-treated with the ether lipid, washed, then incubated with the radiolabeled nutrient. Typically, HL-60 or K562 cells (6.10r'/ml) were washed with Dulbecco's phosphate-buffered saline (PBS), then pre-incubated for times of 1 or 3 h, as defined for each experiment, in serum-free RPMI 1640 media (containing 100/zg/ml porcine insulin, 5/~g/ml transferrin, and 200/zg/ml BSA) at 37°C (95% air, 5% CO 2) with 12 /zM 2methoxy analog (suspended by sonication into serumfree media). In similar pre-ineubations, cells were exposed to 12/~M l-alkyi-2-1yso-GPC or vehicle (serumfree media). The treated cells were harvested, washed three times (centrifuged at 300 × g for 5 min) with PBS, and suspended in PBS/20 mM N-(2-hydroxyethyi)piperazine-N'-(2-ethanesulfonie acid) (Hepes) (pH 7.4). The determination of the cellular uptake of the labeled compounds was similar to the procedure described by Lewis et al. [18]. Pre-treated cells (1.106) were incubated (total assay volume, 250/~1) with one of the following radioiabeled compounds: 0.2 /~Ci/4.5 nmol [3H]choline, 5 ~Ci/25 nmol [3H]methionine, 1 /~Ci/25 nmol 2-[14C]aminoisobutyric acid, 0.2/~Ci/10 nmol [14C]arachidonic acid, 2/~Ci/25 nmol [3H]histidine or 0.2 /~Ci/2.5 izmol [3H]deoxyglucose. Final concentrations of nutrients in the reaction mixture were 100/~M for methionine, histidine and 2-aminoisobutyric acid, 40/~M for arachidonic acid, 10 mM for deoxyglucose, and 18 /~M for choline. The 30 m;n reaction was terminated by addition of 1 ml ice-cola PBS/0.3% BSA and the tube placed in ice-water. Cell suspensions were centrifuged at 13000 x g for 1 rain, supernatants aspirated, and cells washed three additional times with 1 ml ice-cold PBS/BSA. The bottoms of the microfuge reaction tubes, containing cell pellets, were cut off and placed in 5 ml of scintillant (BudgetSolve, Research Products International, Mount Prospect, IL). After sonication in an ultrasonic tank (5 rain, 25°C), the cell-associated radioactivity was measured in a Tracor Mark Ill liquid scintillation spectrometer. The number of cells kilied by a 3 h incubation with the 2-methoxy lipid was insignificant, such that the uptake of nutrient molecules expressed on the basis of protein content [19] was equivalent to that based on viable cell number, as determined by T~pan blue dye excluaion [20]. In several experiments, phosphatidv!cho!ine bin~,,nthesis was examined by determining the incorporation of ['~H]choline into cellv[ar phosphatidylcholine. A procedure was utilized to delineate inhibition of the biosynthetic pathway from inhibition of cellular choline

6 uptake. Pre-incubation of HL-60 cells with radiolabeled choline results in labeling of an intracellular pool of choline that serves as precursor for subsequent incorporation into phosphatidyicholine. Thus, any inhibitory action of the 2-methoxy lipid on cellular uptake of choline would not affect label incorporation

ml scram.flee media (containing 21.5 nmol/ml choline) were pre-labeled with [3H]choline (36/zCi) for 30 min at 37"C. Cells were washed four times with PBS (5 ml) to remove any free radiolabel, resuspended at 1.10 ~ ceils/mi, and incubated for 1 h at 37°C, with either PBS alone (control) or PBS plus l-alkyl-2-methox3,-GPC (12 ~tM). The reaction was terminated and cells repeatedly washed, as above. Cell pellets were resuspcnded in PBS and the amount of radiolabel associated with intact cells and lipids was determined. Total lipids were extracted from the cell suspensions by the procedure of Bligh and Dyer [21], except that the methanol contained 2% glacial acetic acid. Phospholipid phosphorus was determined according to the method of Rouser et al. [22]. Radiolabel incorporation into phosphatidylcholine was determined after fractionation of lipid classes by thin-layer chromatography on silica gel H layers, developed in chloroform/ methanol/acetic acid/water (50: 30:8: 6). The silica gel containing the phospholipid was suspended in scintiilant, sonicated, and the radioactivity determined. Student's t-test was utilized for statistical comparisons of treated vs. control groups of cells. Results

Gwwth curves for HL-60 cells over a 4 day incubation are shown in Fig. I. Treatment of the cells with I-alk~l-2-methoxy.GPC (12 ~tM) for short intervals (0.5 add I h, Fi~. IB and C), removal of the lipid by repeated washing and a subsequent 23 h incubation, killed 50% of the cells: however, after a second day of incubation, the cells recovered to a normal rate of growth, h, cGa~rast, if HL-60 cells are treated with the analog for at least 2 h (Fig. IE), the cell number continues to decline thro+~ghout the subsequent 4 day incubation. If the HL-60 cells were washed with 4% BSA after the 2 h treatment, they were protected from the cytotoxic effect of the analog (Fig. ID). Thus, the 'sensitive' HL-6~ cells could be treated for various time intervals with the 2-metho~ lipid to provide subthreshold, (I h) or beyond a threshold (2 h or greater), exposure level~ of the cytotoxic agent. These results confirm our p~evions report [4] ~hat the m~ljority of HL4g) ceI!s exposed for more than 90 min to the ether lipid will not survive a subsequent 24 h incubation and con'e+-q~onds to an inhibition of ['~H]th+.vmidine incorporation into nucleic acids. In separate experiments, the

3[

yA

~'A2"S

B

i

°

1.

o.s 0

0

........

.O

I 2 3 4 Incubation Daye After AMGPC Treatment

Fig. I. Growth curve of !tL-60 cells treated with I-alkyl-2-methoxy. GPC (AMGPC).Cellswere preincubated for (A) 0, (B) 30. (C) 60. or (E) 120 rain with 12 ~tM al~l phospholipid in serum-free media, washed three times with PBS, and then incubated in fresh media for 4 days. Additionalcells(D) were treated with the analog for 120 min, washed with 5 ml of 4% BSA/PBS then incubated further. The number of viable cells were determined by Trypan blue dye exclusion. Valuesare the averageof two experimentsdone in duplicate.

cellular uptake of l-[aH]alkyl-2-methoxy lipid (12 p.M, 0.37/tCi) by HL-60 cells was 34 and 87 pmol/I0 * cells in I h and 3 h incubations, respectively. From these values, we calculated that cell death would occur if the cells were exposed to the 2-metho~ analog, whereby the level of ether lipid exceeded 0.1% of the total phospholipid co,itcnt of th~ cells. The cellular u p t a k e o f [3H]choiine was selectively iahibited in the HL-60 cells by l-alkyl-2-methoxy.GPC (12 pM) (Fig. 2). Treatment of HL-60 cells with l-alkyl-2-1yso-OPC (12 IzM), which is 8 to I0 times less cytotoxically potent than the 2-methoxy analog [4], was 8.3-fold less inhibitory than the 2-metho~ lipid on choline uptake. Moreover, the inhibitory action of lalkyl-2-methoxy-OPC on choline uptake was greatly reduced in the 'resistant' K562 leukemia cells and was unchanged from the control following exposure of K562 cells to the 2-1yso analog. In separate experiments, the incorporation of ['~H]choline into phosphatidylcholine of HL-60 cells was inhibited to a greater extent by the 2-methoxy analog than the inhibition of choline uptake per se in HL-60 cells (data not shown). Therefore, it was important to determine if the inhibition of phosphatidylcholine biosynthesis reflects decreased incorporation into this lipid class via biosynthetic pathways or a blockade of choline transport across the plasma morn brane. Labeling of untreated HL-60 cells with [3H]choline resulted in a 50.1% cellular uptake of radiolabel (167 pmol × I06 cells-'). Radiolabel incorporation into phosphatidylcholine was 32.4% of the

77

cellular uptake (54.1 pmol x 10(' celis-~). Subsequent exposure of pro-labeled cells to l-aikyl-2-methoxy-GPC for 1 h inhibited [all]choline incorporation into phosphatidylcholine by 55%, compared to PBS-treated controls. In contrast, choline incorporation into phosphatidylcholine was only inhibited by 5% in HL-60 cells treated with the 2-1yso analog and only by 20% in K562 cells treated with the 2-mcthoxy ether lipid. These results suggest that 1-aikyl-2-methoxy-GPC specifically inhibits both nutrient uptake and phosphatidylcholine biosynthesis in the cytotoxically sensitive cells. To further characterize the inhibitory specificity of the cytotoxic ether lipid, 'sensitive' HL-60 cells and 'resistant' K562 cells were treated with 1-alkyl-2methoxy-GPC (12 /.tM) or the tumor promoter, TPA (0.1 ~M) and cellular [3H]choline uptake determined

laot

HL.80

lOOt" 80' eO' *

16°t

HL-60

K562

140 120

100

¢t

i

"

i¢

s ~

6O 4O ..f

2O 0

Vehicle AMGPC

TPA

Vehicle AMGPC

TPA

Fig. 3. Uptake of ['~H]choline by TPA- and I-alkyl-2-methoxy.GPC (AMGPC)-treated HL-60 and K562 leukemia cells. Cells (0.5. i0 t') were incubated with 12 /~M I-alkyi-2-methoxy-GPC, vehicle (PBS/Heoes), 0.1 g M TPA or its vehicle (DMSO, final reaction concentration, 0.8%) and [3H]choline for 30 rain. Cells were harvested and the cell-associated radioactivity determined. The listing of 100% uptake for PBS/Hepes vehicle was 370+26 S.D. and 836:1:17 pmol × 10~ cells- i hr- i for HL-60 and K562 cells, respectively. The 100% values for controls (DMSO only) were 433 + 14 and 1044.t-57 pmoi x 10c' cells- ~ hr- i for HL-60 and K562 cells, respectively. l-AllojI-2-methoxy-GPC and TPA values are compared vs. respective controls and each was the mean of four determinations. Vertical bars represent $.D. and the asterisks indicate values that differ significantl~ from controls (P < 0.001).

40 ~ elm

o

20, I

0

~,

I

I

0

15

K562 120

T 1001 t

'I"

---X

80 IlO 40 2O

Oo

I

I

I

1

2

3

Pre-lncubatlon with Ether Lipid (hr) Fig. 2. Uptake of ['~H]choline by cytotoxically 'sensitive' HL-60 or 'resi.~tant' K562 leukemia cells, treated with alkyl phospholipids. HL-60 cells or K562 cells were treated with the 2-methoxy ( x ) , (12 #.M) or 2-1yso analog (~,) (12 #M) ful 0, 1, ui 3 h. The uptake of [3H]choline was determined in a 30-rain incubation of cells in Dulbecco's PBS-Hepes (100% uptake = 264+9 S.D. and 949-1-39 I:,mol x 106 cells- i x hr- i for HL-60 and K562 cells, respectively). Each point is the mean + S.D.of four analyses. The asterisks indicate that the values are significantly different from controls (P < 0.001).

simultaneously (Fig. 3). As found previously, the 2methoxy lipid significantly inhibited choline uptake in HL-60 cells but :failed to affect uptake in K562 cells. Choline uptake remained unchanged in TPA-treated HL-60 cells, whereas it was significantly elevated, by 35%, in K562 cells. This response appears to conflict with that of Kiss et ai. [23] since they reported a 50% stimulation of choline incorporation into phosphatidylcholine in TPA-treated HL-60 ,.~l,~ incubated for 2 h. The discrepancy may be partially attributable to the use of differ¢~t incubation media, total choline concentration, incubation time, or the specific passage of cultured cells. The present results fimrther ~upport the existence of distinct differences between the responsiveness of the nutrient transport systems of HL-60 and K562 cells towards not only 1-alkyl-2-methoxy-GPC but also phorbol ester. The inhibitory action of 1-alkyl-2-methoxy-GPC on other r~utrient transport systems in HL-60 cells was also determined (Fig. 4). Treatment of HL-60 cells (3

h) with the 2-methoxy analog significantly inhibited cellular uptake of arachidonic acid. In addition, the 2-methoxy ether lipid also blocked the uptake of amino acids representative of several transport systems. The 'A' system for short-chaii~ amino acids, such as m~thionine [24], was sensitive to the 2.methoxy analog. Similarly, the sub-class 'ASC' sodium-dependent system

(e.g., 2.aminoisobutyric acid, see Ref. 25) and the 'L' system for amino acids greater than ~ve carbon units (e.g., L-histidine at pH 7.0,0.1 M, see Ref. 26) were inhibited by 1-aikyl-2-methoxy-GPC in HL-60 cells. Exposure of HL-60 cells to the 2-1yso analog did not effect methionine or histidine uptake. However, l-alkyl-2-1yso-GPC decreased 2-aminoisobutyric acid and arachidonic acid by about one.half compared to controis but not to the extent of treatment with the 2metho,xy analog. In contrast, nutrient transport in K562 cells was 'resistant' to the inhibitory actions of both the metheoty- and iyso-ether-lipid analogs. The carbohydrate transporter activity in HL-60 cells was significantly inhibited by 60-70%, following l-alkyl.2.metho0cy.GPC treatment for 1-3 h (Fig. 5). Treatment of these cells for either I or 3 h with the 2-1yso analog decreased transport of the non-metabolizable 2.deo0ty.~glucose by 30%. Cellular transport of deo,wglucose by K562 cells was relatively resistant to inhibition by both the 2-methoxy and 2-1ysoether lipids.

HL-60

,,°t

100 i 80

60 40

i '°t O,

,

~---~-

3

Discussion

@0

In previous studies we reported that cytotoxically 'sensitive' HL-60 leukemia cells express a greater affinity for l-alkyl-2-methoxy-GPC than the relatively 're-

40

,

i

K$~

20 0

0

t

t

|

1

2

S

Pre-lncubdon with Ether Lipid (hr)

1.0

Fig, 5, Inhibitionof 2-deoxy-o-glucose uptake by l-all~l-2-methoxyGPC in HL.60 cells but not in K562 leukemia cells. Cells were incubated with the 2-metho~ (×) (12 p.M) or 2-1yso ether lipid( A ), (12 ~M) for 0, I, or 3 h. The uptake of [3H]deoxyglucose was det©rmined in a 30 rain incubation of cellsin Dulbecco's PBS/Hepes (I00% uptake = 21.4+ 1.5 S.D. and 16.7± 1.0 nmol× 106 cells"'I hr-t for HL-00 and !<,562 cells, respectively). Each point is the mean ± S.D, of triplicate analyses and the asterisk indicates that the value is statistically different from control values (P < 0.001).

0,|

N

l lkt ~ ~ i

....

Y'|'

O, Q M

L,

HI.,410

O i

I.

KSe2

O M

I.

HL-60

C

M

L

KS62

Fig, 4, The effect of I-alkyl-2-methoxy.GFC on the uptake of various bioloOcal nutrients in HL-60 and K562 leukemia cells. Cells were preincubated with the 2-methoxy (M) or 2-1yso(L) ether lipid analogs (12 taM) or vehicle alone (C) for 3 h, washed, and then incubated for 30 rain with the various labeled compounds, (A) Control values 0"vehicle o~,) foe ~ake. of [~|i~methio,im: wcte 0.48 and 2.24 amolx 10e' cells-i hr-I in HL-60 and K562 cells, respectively, (R) ~ t m l values for uptake of 2-[14C]ami.oisobut~i© acid (AIB)were ZB7 ~ 4,94 nmol× 10~ cells-t hr-! in !JL-60 and K562 cells. ~ b ' , (C) Control values for uptake of [3H]histidine (oH 7, 0.1 M) were 040 and 1.07 nmolx I06 cells-I hr-i it; HL-60 and K562 cells, ~ i v e l y . (D)Control values for uptake of It4C]arachidonic acid were 1.99 and 6.49 nnml × 106 cells- t hr- I in HL~',O.and K562 cells, ruq)ex~i~ly. Standard deviations are represented by vertical bats and the asterisk indicates a statistically significant difference from control values (P < 0.001).

sistant' K562 leukemia cells or normal cell type, such as human polymorphonuclear nm!trophils [6], As demonstrated in the present report, the resultant cellular accumulation of l-alkyl-2-methoxy-OPC in 'sensitive' cells impacts a number of nutrient transport systems, which are quantitatively inhibited by the 2methoxy lipid, compared to the 2-1yso derivative. In contrast, nutrient transport is only marginally affected in 'resistant' ceils that accumulate the 2-memoxy analog to a lesser extent. The present results extend our earlier studies on l-ailo/l-2-metho~-GPC inhibition of nutrient transport systems [13,17]. We have reported that after short-term exposure of undifferentiated HL.60 cells, the 2~methoxy lipid markedly inhibited cellular choline uptake, whereas dimethylsulfoxide-differentiated HL-60 cells

79 remained unaffected by treatment with the alkyl phospholipid [ 13]. The present results confirm the inhibition o~ cho::ine [12,13,17] and l-aikyl-2-1yso-GPC [12] incorporation into phosphatidylcholine induced by l-aikyi-2-methoxTGPC in HL-60 cells. Vogler et al. [12] did not sep.arately evaluate cellular uptake mechanisms from biosynthetic pathways and, additionally, utilized longer treatment periods (24 h) where significant cell death occurs. 1-Alkyl-2-methoxy-GPC has also been reported to inhibit uptake and incorporation of tritiated thymidine into DNA of various neoplastic cells [4,15,16]. Similarly, Modolell et al. [11] have demonstrated that the 2-methoxy analog inhibits the incorporation of radiolabeled oleic acid, linoleic acid, and choline into phosphatidylcholine. Furthermore, Herrmann [14] reported that the 2-methoxy lipid inhibited oleic acid incorporation into phosphatidylcholine with a concomitant accumulation in triacylglycerols o[ sensitive tumor target cells (i.e., Meth A, HL-60, YAE and ABLS-8.1). Although a direct inhibition of phosphatidyicholine biosynthesis may be an integral component of the cytotoxic mechanism of action of these ether lipid analogs, the regulation of molecular transport systems at the cell membrane would appear to be an equally important target, as it deprives the 'sensitive' cells of essential nutrients for growth. Several of the aikyl phospholipid analogs have been reported to cause differentiation of HL-60 leukemia cells into macrophage-like cells [27,28] similar to the tumor promoter, TPA [29,30]. Although both l-alkyl2-methoxy-GPC and TPA are membrane-targeted and neither induces differentiation of I.:,562 cells [16,31], other similarities between these two agents are not apparent. A key component in the action of TPA involves activation of protein kinase c [32], whereas the 2-methoxy lipid inhibits this enzymatic activity [33]. Furthermore, TPA stimulates both the uptake and incorporation of choline into phosphatidylcholine in lymphocytes and select neoplastic cells [23,34,35 and Fig. 3], whereas 1-alkyl-2-methoxy-GPC has an opposite action [11,12,14 and Figs. 2 and 3]. These distinct characteristics may be explained by the receptor-mediated actions of TPA compared to detergent-like membrane disruption properties of non-metabolizable laikTI-2-methoxT-GPC. Several classical nuclear-targeted chemotherapeutic agents have been reported to also inhibit membraneassociated functions [36]. For example, cisplatin [37] and methotrexate [38] inhibit methionine and 2aminoisobutyric acid transport in both murine leukemia cells and human !ymphocytes, Inhibition of the Na+,K+-ATPase system was suggested as a common mechanism of action for these agents. ATPase may also be a critical mediator in 1-ali~w-2-methoxy-GPC cytotoxicity as its activity is reduced in 2-methoxy-

lipid-treated HL-60 cells and rat brain membrane preparations [39]. However, the growing list of functional targets of 1-alkyl-2-methoxy-GPC interaction appear to involve integral membrane components at the cell surface. Thus, the inhibitory action on nutrient transport systems, as described in this report, may reflect an ether lipid-induced membrane perturbation that is highly selective toward cytotoxically 'sensitive' neoplastic cells. Acknowledgements This work was funded, in part, by American Cancer Society grants IN-142 (DRH) and BC-70V (FS), a University of Texas Southwestern Institutional research grant (DRH), and the Office of Energy Research, U.S. Department of Energy (Contract No. DEAC05-760R00033) (FS). The authors thank Dr. W. Lai (UT Southwestern) for mycoplasm DNA analysis of HL-60 and K562 cells. References 1 Munder, P.G., Weltzien, H.U. and Modelell, M. (1976) in lmmunopathoFogy (Miescher, P.A, ed.), Vol. 7, pp. 411-424, Schwahe and Co, Basel. 2 Baumann, WJ., Berdel, W.E., van den Bosch, H., Eibl, H., Herrmann, D.BJ., Munder, P.G., Snyder, F.L. and Unger, C. (eds.) (1987) First international Symposium on Ether Lipids in Oncology. Lipids 22, 775-980. 3 Berdel, W.E. (1990) Onkologie 13, 245-250. 4 Hoffman, D.R., Hajdu, J. and Snyder, F. (1984) Blood 63, 545-552. 5 Hoffman, D.R., Stanley, LD., Berchtold, R. and Snyder, F. (1984) Res. Commun. Chem. Pathol. Pharmacol. 44, 293-306. 6 Hoffman, D.R., Hoffman, L.H. and Snyder, F. (1986) Cancer Res. 46, 5803-5809, 7 Andreesen, R., Modolell, M., Weltzien, H.U., Eibl, H., Common, H.H., Lohr, G.W. and Munder, P.G, (1978) Cancer Res. 38, 3894-3899. 8 Andreesen, R., Osterholz, J., Luckeni~ack, G.A., Costabel, U., Schulz, A., Speth, V., Munder, P.G. and Lohr, G.W. (1984) J. Natl. Cancer Inst.72, 53-59. 9 Berdel, W.E., Greiner, E., Fink, U., Stavrou, D., Reichert, A., Rastetter, J., Hoffman, D.R. and Snyder, F. (1983) Cancc~ R~. 43, 541-545. I0 linger, C., Eibi, H., Kim, DJ., Fleer, E.A., Kot.ting,J., Bartsch, H-H., Nagel, G.A. and Pfizenmaier. K. (1987) J. Natl. Cancer Inst. 78, 219-222. 11 Modolell, M., Andrees¢.n, R., Pahlke, W., Brugger, U. and MufideL P.G. t l n ' , n ~ r,~/._~. 12 Vogler, W.R., Whigham, E., Bennett. W.D. and Olson, A.C. (1985) Exp. Hematol, 13, 629-633. 13 Vallari, D.S., Smith, Z.L. and Snyder, F. (1988) Biochem. Biophys. Res. Commun. 156, I-8. 14 Herrmann, D.B.J. (1985) J. Natl. Cancer Inst. 75, 423-430. 15 Berdel, W.E., Fink, U., Egger, B.. Reichert, A., Munder, P G. and Rastetter, J. (1981) J. Natl. Cancer Inst. 66, 8i3-817. ' 16 Tidwell, T., Guzman, G. and Vogler, W.R. (1981) Blood 57, 794-797. 17 Snyder, F., Record, M., Sn:ith, Z., Blank, M.L. and Hoffman, D.R. (1987) Aktuel. Onkol. 34, 19-26. ,

j~,,

..tv~ 1 --.vvuv.

18 ~ NJ., Inloes, R.L, Hes, J., Matthews, R.H. and Milo, G. (1978) J. Med. Chem. 21, 1070-1073. 19 l . z ~ , O,H,, Rosebrough, NJ., Fan', A.L. and Randall, R.J. (1951) J. Biol. Chem. 193, 265-275. 20 Patterson, M.K. (1979) Methods Enzymoi. 58, 141-152. 21 Blish, E.G. and Dyer, W.J. (1954} Can. J. Biochem. Physiol. 37, 911-917. 22 Rotor, G., Siakotos, A.N. and Fleischer, S. (1966) Lipids 1, 85-86. 23 ~ , Dell, E., Girard, P.R., Pettit, G.R. and Kuo, J.F. (1987) Biechem. Biophys. Res. Commun. 146, 208-215. 24 ~ n d e r , D J- and Christensen, H.N. 0963) J. Biol. Chem. 238, 3686.-3699. 25 Van den ikrg, KJ. and Betel, !. 0973) FEBS Lett. 29, 149-152. 26 Matthews, R.H. (1972) Biochim. Biophys. Acta 282, 374-382. 27 Honme, Y., Kasukabe, T., Hozumi, M., Tsushima, S, and Nomere, H. (1~1) Cancer Res. 41, 3211-3216. 28 Honma, Y., lOisukabe, T., Okabe-Kada, J., Hozumi, M., Tsushime, S. and Nomura, H. (1983) Cancer Chemother. Pharmacol. II, 73-76. 29 Huberman, E. and Callham, M. (1979) Proc. Natl. Acad. Sci. USA 76, 1293-1297.

30 Rovera, G., Santoli, D. and Damski, C. (1979) Proc. Natl. Acad. Sci. USA 76, 2779-2783. 31 Lozzio, C.B. and Lozzio, B.B. (|975) Blood 45, 321-334. 32 Castagna, M., Takai, Y., Kaibuchi. K.. S~ilo, K..~Kiddowa, U. and Nishizuka, Y. (1982) J. Biol. Chem. 257, 7847-7851. 33 Helfman, D.M., Barnes, K.C., Kinkade, J.M., Vogler, W.R., Shoji, M. and Kuo, J.F. (1983) Cancer Res. 43, 2955-2961. 34 Kinzel, V., Kreibich, G., Hecker, E. and Suss, R. (1979) Cancer Res. 39, 2743-2750. 35 Rohrschneider, L.R. and Boutwell, R.K. (1973) Cancer Res. 33, 1945-1952. 36 Hickman, J.A., Scanlon, KJ. and Tritton, T.R. (1984) Trends Pharmacol. Sci. 5, 15-19. 37 $¢anlon, KJ., $afirstein, R.L., Thies, H., Gross, R.B., Waxman, S. and Guttenplan, J.G. (1983) Cancer Res. 43, 4211-4215. 38 Scanlon, KJ,, B©rkowitz, K., Pallai, M.E. and Waxman, S. (1983) Cancer Treatment Rep. 67. 631-639. 39 Oishi, K., Zheng, B., White, J.F., Vogler, W.R. and Kuo, J.F. (1988) Biochcm. Bi~ph~s. Res. Commun. 157, 1000-1006.