Changes in Nucleoside Transport of HL-60 Human Promyelocytic Cells During Tvjaf-Dimethylformamide Induced Differentiation1

[CANCER RESEARCH 46, 3449-3455, July 1986] Changes in Nucleoside Transport of HL-60 Human Promyelocytic Cells during TVjAf-Dimethylformamide Induced Differentiation1

Shih-Fong Chen,2Jeffrey S. Cleaveland, Ann B. Hollmann, Michael C. Wiemann, Robert E. Parks, Jr., and Johanna D. Stoeckler3

Division of Biology and Medicine, Brown University, Providence, Rhode Island 02912 fS-F. C., J. S. C., R. E. P., J. D. S.J and Department of Medicine, Roger Williams General Hospital, Providence, Rhode Island 02908 [A. B. H., M. C. WJ

ABSTRACT A study of uridine transport in Friend murine erythroleuke- mia cells during erythroid differentiation induced by dimethyl The rate of nucleoside transport decreased profoundly in human pro- sulfoxide detected only a 2- to 3-fold decrease in rate which myelocytic leukemia HL-60 cells after myeloid differentiation was in duced by 5-6 days of exposure to 0.8% A,,\-dimethylfnrniamidc (DMF). was correlated with the decrease in surface area (18). Differ The facilitated diffusion of 100 /IM radiolabeled adenosine and 2'-deox- ences in the uptake rates of purine and pyrimidine nucleosides yadenosine, measured by rapid transport assays, decreased 10- to 20- in normal versus activated or neoplastic cells have generally fold. The transport of 2 /IM coformycin or 2'-deoxycoformycin, which is been attributed to differences in the levels of the activating mediated by the same mechanism and was monitored by the adenosine enzymes, rather than to changes in transport. The transport deaminase titration assay, decreased 29-fold. The reduction in nucleoside kinetics of uridine in serum-activated or SV40-transformed 3T3 transport capacity after DMF treatment was confirmed by a 19-fold mouse fibroblasts did not differ significantly from those in decrease in the number of specific binding sites per cell (from 24-30 x quiescent cells (19); similarly, the transpon of uridine and IO4 to 1.2-1.7 x IO4) for [3H]-6-p-nitrobenzylthioinosine, a nucleoside thymidine was independent of culture age in Novikoff rat transport inhibitor. The binding affinity of 6-/Â»-nitrobenzylthioinosinewas hepatoma cells (20); however, the number of binding sites for not altered significantly and nucleoside transport remained sensitive to the transport inhibitor, NBMPR, fluctuated 2- to 8-fold during the transport inhibitors, 6-y-nitrobenzylthioinosine, dipyridamole, and dilazep after DMF-induced maturation. the replication cycle of synchronized human HeLa carcinoma Time-dependence studies showed that the rate of 100 MMdeoxyaden- cells (21). osine transport was unchanged for the first 24 h of exposure to DMF but Evidence is presented below that a profound decrease in fell to about 36% of control rates at 24-26 h and then gradually decreased nucleoside transport occurs in HL-60 cells after DMF-induced further to about 4-5% of control rates after 5-6 days. In contrast, myeloid differentiation. DMF was selected as the maturationai transport rates of the purine bases were reduced only 2- to 3-fold in HL- agent because its congener, ./V-methylformamide, currently un 60 cells after 5 days of DMF treatment. The rates of adenosine and der study in clinical trials, is a less effective inducer of this cell deoxyadenosine transport were unchanged or reduced by no more than line (6). Part of this work has been presented in a preliminary 2-fold after 5-6 days of exposure to 0.8% DMF in the following human report (22). tumor cell lines that are not inducible with DMF: ARH-77 (multiple myeloma), KG-1 (acute myelogenous), and K-562 (chronic myelogenous). Thus, changes in nucleoside transport may serve as an early, membrane- MATERIALS AND METHODS associated marker of differentiation of the HL-60 cell line. Cells

INTRODUCTION The HL-60 cell line, derived from a patient with acute promyelocytic Human promyelocytic leukemia HL-60 cells can be induced leukemia (1,2) was maintained in RPMI 1640 medium supplemented to mature toward either the myeloid (1, 2) or monocytic (3-5) with penicillin (100 units/ml) and streptomycin (100 ng/ml) and 15% heat-inactivated fetal bovine serum. K-562 cells, derived from a patient phenotype when they are exposed to different inducers. Polar solvents such as dimethyl sulfoxide or DMF4 (6) and many with chronic myelogenous leukemia in the blastic phase (23), were cultured in the same medium containing 10% heat-inactivated fetal other agents (7-12) induce terminal myeloid differentiation. A bovine serum. KG-1 cells, derived from a patient with acute myeloge number of morphological, functional, and biochemical changes nous leukemia (24), were maintained in a-MEM without nucleosides, have been observed during dimethyl sulfoxide-induced differ supplemented with gentamycin (20 //u/mlj and 20% heat-inactivated entiation. These changes include a decrease in size (13), alter fetal bovine serum. ARH-77 cells, derived from a patient with multiple ations in the composition of membrane glycoproteins (14, 15), myeloma (25), were maintained in Ham's FIO medium supplemented glycopeptides (16), and antigens (17), and disappearance of the with penicillin (100 units/ml), L-glutamine (1 m\i), MEM vitamins (2.5 ml of lOOx solution/500 ml) and 20% heat-inactivated fetal bovine membrane receptor for transferrin, an important growth factor serum. Cultures were incubated at 37Â°Cin humidified air containing (14). These membrane-associated changes raised the possibility 5% CO2. For routine maintenance, cells were seeded at 2 x IO5cells/ that differentiation-induction might cause significant altera ml and subcultured every 3-5 days by dilution with fresh medium. Cells tions in the facilitated diffusion of nucleosides across the plasma used for nucleoside transport studies were harvested during exponential membrane. growth, 2-3 days after subculturing. DMF-treated cells were continu ously exposed to 0.8% DMF for the designated time intervals. Nitroblue Received 10/25/85; revised 2/25/86; accepted 3/25/86. The costs of publication of this article were defrayed in part by the payment tetrazolium dye reduction was used as an indicator for differentiation. of page charges. This article must therefore be hereby marked advertisement in The percentage of cells containing blue-black formazan deposits was accordance with 18 U.S.C. Section 1734 solely to indicate this fact. determined on Wright-Giemsa-stained cytospin (Shandon Southern 1This work was supported by USPHS grants CA 37901, 20892, and 13943 awarded by the National Cancer Institute, Department of Health and Human Instruments, Inc., Sewickley, PA) preparations of cell suspensions as Services. This report is a publication of the Roger Williams Cancer Center. described by Collins et al. (26). After 5-6 days of treatment with 0.8% 2Present address: Biomedicai Products Department, Glenolden Laboratory, DMF, 60-80% cells showed a positive reaction with the dye. Cell Du Pont Company, 500 S. Ridgeway Avenue, Glenolden, PA 19036. concentrations were determined with a Coulter Counter Model B 3To whom requests for reprints should be addressed. 4 The abbreviations used are: DMF, JV.A'-dimethylformamide; NBMPR, 6-p- equipped with a Coulter Channelyzer (Coulter Electronics, Inc., I li nitrobenzylthioinosine, 6-[(4-nitrobenzyl)thio]-9-/3-D-ribofuranosylpurine, 6-(p- aleah, FL) and cell viability determinations were based on the exclusion nitrobenzyl)mercaptopurine ribonucleoside; a-MEM, a minimal Eagle's medium. of trypan blue dye by cells counted on a hemocytometer (Hausser 3449 Downloaded from cancerres.aacrjournals.org on September 29, 2021. © 1986 American Association for Cancer Research. NUCLEOSIDE TRANSPORT AND DIFFERENTIATION IN HL-60 CELLS

Scientific, Blue Bell, PA). To prepare the cells for transpon studies, carried out for 1 min at 650 x g. For time points ranging from 2-24 the cell suspensions were centrifugea at 800 x g for 5 min in a Son all min, the permeant solution was added to a large volume of cell suspen GLC-4 centrifuge (Sorvall Instruments, Wilmington, DE). The result sion in a round-bottomed 15-ml plastic culture tube. At the indicated ing cell pellets were resuspended in 10 ml "standard transport medium" time intervals, the tubes were vortexed and 0.5-ml aliquots were with (RPMI1640 medium without NaHCO3 and supplemented with 20 IHM drawn and injected into centrifuge tubes containing the cold stopping 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid-NaOH, pH 7.2), solution. Adenosine deaminase activity was determined as described and centrifuged at 800 x g for 2 min in a clinical centrifuge. The cell previously after the washed cells were resuspended in 2.25 ml of 50 pellets were finally resuspended in an appropriate volume of the stand HIMpotassium phosphate buffer, pH 7.5, and sonicated. ard transport medium to give cell concentrations of 1-5 x IO7 cells/ ml. Specific Binding of NBMPR The procedures for determination of A",,ofNBMPR and the number Furine Base and Nucleoside Transport of binding sites per cell in control and DMF-treated HL-60 cells were Oil-stop Procedure for Rapid Transport Assays. A detailed descrip similar to published methods (32). Briefly, cell suspensions in standard tion of the transpon assays for nucleosides and purine bases using a transport medium were preincubated at room temperature in the pres mixing chamber connected to 2 Eppendorf Repeater pipetters has been ence and absence of 50 ^M nonradioactive transport inhibitors, 6-(2- documented elsewhere (27). Briefly, equal volumes (50 p\) of cell hydroxyl-5-nitrobenzyl)thioguanosine or dipyridamole, for 15 min. suspensions (2-5 x 10' cells/ml) in standard transpon medium and After addition of graded concentrations of [â€¢'!I|NBMPR,the cell sus transport medium containing 3H- or MC-labeled permeants were ejected pensions were incubated for another 15 min. The radioactivities asso at timed intervals through the mixing chamber into 1.5 nil microcen ciated with the cell pellets were determined after centrifugation through trifuge tubes containing 0.25 ml 1-bromododecane. After the last injec 1-bromododecane. Specific binding is defined as the difference between the binding of [3H]NBMPR to the cells in the absence and presence of tion, the centrifuge was activated within 0.5 s and the transpon process was terminated by centrifuging the cells through the 1-bromododecane. the competitive transport inhibitors, which displace specifically bound NBMPR from the cellular binding sites. The A',,and numbers of specific An interval of 2 s for separation of cells from medium was consistent with the results obtained for control cells versus cells pretreated with binding sites were determined from linear double-reciprocal plots of transport inhibitors. molecules bound per cell versus the free NBMPR concentration and Inhibitor-stop Procedure. For time points as early as 0.5 s, transpon from Scatchard plots, which gave similar values. was stopped by dilution with medium containing the highly soluble transport inhibitor, dilazep, at 4'C. Each 1.5-ml microcentrifuge tube Chemicals containing 100 ,,1 of cell suspension at room temperature was held Coformycin (3-/J-D-ribofuranosyl-6,7,8-trihydroimidazo[4,5-d][ 1-3]- continuously on a vortex mixer during the addition of 100 /Â¿Iof diazepin-8-(/?)-ol) and deoxycoformycin [3-(2'-deoxy-|8-D-erythropen- transport medium containing radiolabeled permeant at zero time and tofuranosyl)-6,7,8-trihydroimidazo[4,5-d][l-3]diazepin-8(A)-ol] were until transport was stopped after a measured time interval by addition provided by the Drug Development Branch of the National Cancer of 200 Â»Iof ice-cold transport medium containing 750 JIM dilazep. Institute, Bethesda, MD. Their concentrations were determined from After mixing was complete, 0.25 ml 1-bromododecane was added and their spectra in water (33) and confirmed by measuring the pseudo first the cells were pelleted by centrifugation for 1 min in an Eppendorf order rate of inactivation of calf intestinal adenosine deaminase (30, Centrifuge 5414 (15,000 x g) 15 s after the dilazep addition. 31). NBMPR was a gift from A. R. P. Paterson of the University of For both rapid transport assay procedures the medium remaining Alberta, Edmonton, Alberta, Canada. Dilazep was generously provided above the 1-bromododecane was removed by aspiration. The inner walls by M. Gudenzi of Asta-Werke AG, Frankfurt/Main, Federal Republic of the microcentrifuge tubes above the 1-bromododecane were washed of Germany. Dipyridamole (Persantin), adenosine, 2'-deoxyadenosine, gently with 1 ml of water and recentrifuged for 2 min. After removing adenine, hypoxanthine, 1-bromododecane, and Triton X-100 were pur the wash and most of the 1-bromododecane, the cell pellets were incubated overnight at 37Â°Cwith0.75 ml of tissue solubilizer consisting chased from Sigma Chemical Company, St. Louis, MO. ,V,.V-I)imeth- ylformamide was obtained from Aldrich Chemical Co., Milwaukee, WI. of 2 g NH4HCO3, l g Triton X-100, and 1 g trypsin in 100 ml H2O Guanine was obtained from Pharmacia P-L Biochemicals, Inc., Pisca- (28). The centrifuge tubes were then placed in scintillation vials. After taway, NJ. RPMI 1640 (Ix, lOx), a-MEM without nucleosides, MEM mixing with 10 ml of Aqueous Counting Scintillant (Amersham Corp., vitamin (lOOx) penicillin-streptomycin solution, trypsin, try pun blue, Arlington Heights, IL), the radioactivity was determined in a Packard fetal bovine serum, and glutamine were purchased from GIBCO Lab Tricarb 460 liquid scintillation spectrometer. oratories, Grand Island, NY. 6-(2-Hydroxyl-5-nitrobenzyl)thio- The total water space and the extracellular space in cell pellets were guanosine hemiisopropanol was obtained from Calbiochem-Behring determined by using 3H2O and [14C]inulin, respectively (29). Under the Corp., La Jolla, CA. [i/-14C]Adenosine (specific activity, 300 mCi/ conditions of the oil-stop method, intracellular water spaces in the mmol), |2,K-'H|adenosinc (specific activity, 48 Ci/mmol), and [8-3H] pellets from untreated and DMF-treated cells were 0.952 Â±0.156 (SD) deoxyadenosine (specific activity, 21 Ci/mmol) were purchased from and 0.409 Â±0.097 jil/10* cells, respectively, and were significantly ICN Pharmaceuticals, Inc., Irvine, CA. [C/-3H]/7-Nitrobenzylthioinosine different (P < 0.006). The extracellular pellet spaces of 0.101 Â±0.036 (specific activity, 17 Ci/mmol), [8-'4C]adenine (specific activity, 59 and 0.073 Â±0.054 Â¿jl/106cells, respectively, were not significantly mCi/mmol), [8-14C]guanine (specific activity, 56 mCi/mmol), and [8- different. These values are from 6 and 4 separate determinations, 14C]hypoxanthine (specific activity, 56 mCi/mmol) were obtained from respectively. Adenosine Deantinase Titration Assay for Deoxycoformycin and C'o- Moravek Biochemicals Inc., Brea, CA. The radiochemical purities of nucleosides were determined by high pressure liquid chromatography formycin Transport. Intracellular concentrations of these tight-binding using a Cu-MBondapak column (Waters Associates, Milford, MA) inhibitors of adenosine deaminase (30, 31) were determined by measur eluted with H^O containing 12.5% methanol. When necessary, the ing residual enzyme activity in an ammonia liberation assay. The radioactive materials were repurified by high pressure liquid chroma procedure was modified slightly from that described previously (27). tography to >95-99% radiochemical purity before use. 3H2O and For time points ranging from 0.25-2 min, 400 Â»\of cell suspension [car6ojr>'-l4C]inulin were obtained from New England Nuclear, Boston, (0.4-1.5 x IO7cells) were injected into 15-ml conical centrifuge tubes. MA. After addition of 50 //I of transport medium or medium containing a transport inhibitor, the tubes were preincubated in a shaking water bath at 30"C. Transport was initiated by addition of 50 n\ of deoxycofor- RESULTS mycin or coformycin (final concentration, 2 ^M). The tube was vortexed and replaced in the bath until the transport was stopped by addition of Transport of Adenosine and Deoxyadenosine in Control and 2 ml of ice-cold Puck's saline G containing 5 ^M NBMPR. The cells DMF-treated Leukemia Cells. Fig. 1 compares the time courses were washed 5 times as described earlier, except that centrifugation was of the influx of 100 MMadenosine in HL-60 cells as determined 3450

120

10 20 30 SECONDS Fig. 2. Transport of 100 Â¿iMadenosinein HL-60 cells after 5 days of exposure 456 8 9 to 0.8% DMF. Assays were performed by the oil-stop procedure. A rate of 37 pmol/min/10' cells was determined from the slope. No correction was made for SECONDS extracellular trapping of adenosine. Fig. I. Transport of udcnosine Â¡nHL-60 cells. The transport of 100 pM radiolabeled adenosine was measured by the oil- (A) and inhibitor-stop (â€¢) procedures as described in "Materials and Methods." Rates of 279 and 236 pmol/ rnin/lO6 cells, respectively, were determined by linear regression analysis of the slopes. The data point at zero time was determined by addition of the cold inhibitor solution before the permeant. No correction for extracellular trapping of adenosine was made; therefore, the X-intercept for the oil-stop line is approx imately double that of the inhibitor-stop line since the inhibitor solution effects a 2-fold dilution of the permeant in the extracellular fluid.

Table 1 Transport of Â¡00Â¡IMadenosineand deoxyadenosine in HL-60 cells and noninducible leukemia cell lines before and after 6 days of incubation with 0.8% DMF cellsTransport pmol/min/106

of adenosineCell of deoxyadenosineControl367 60 90 120 06 12 18 24 lineHL-60 SECONDS MINUTES Â±110Â° Â±20 Â±132 Â±23 Fig. 3. Effect of DMF on the transport of deoxycoformycin and coformycin ARH-77 554 (387)* 491 (442) in HL-60 cells. The transport of the nucleoside analogues was determined by the 736 375 adenosine deaminase titration assay as described in "Materials and Methods." KG-1 542 520 870 (732) 573 (529) K-562Control4251142DMF-treated42608Transport1169(1200)DMF-treated38570 (810) Influx of 2 fiM deoxycoformycin (â€¢)and 2 MMcoformycin (â€¢)caused (a) 47 and " Mean Â±SD of six or more determinations in separate experiments. 21% inactivation of enzyme/min in control cells and (b) 3.1 and 1.6% inactiva- * Numbers in parentheses, rates of deoxyadenosine transport in a separate tion/min in cells exposed for 5 days to DMF, respectively. experiment. exposed to 0.8% DMF for 5-6 days, transport of adenosine or by the oil- and inhibitor-stop assay procedures. The inhibitor- deoxyadenosine was extremely slow and the uptake remained stop method provided time points as early as 0.5 s and lower linear for longer than 30 s in most cases; therefore, the initial intercepts for the extrapolated zero time point because of the velocities were estimated by linear regression analyses of data 2-fold dilution of the permeant by the stopping solution. These obtained between 5 and 30 s. The mean transport rates of both intercepts were in agreement with the values obtained when the nucleosides decreased approximately 10-fold in HL-60 cells inhibitor was added before or at the same time as the permeant (Table 1). In contrast, the transport rates of 100 Â¿Â¿Madenosine solution. Table 1 lists the transport rates of 100 Â¿IMadenosine and deoxyadenosine in 3 other human leukemia cell lines, and deoxyadenosine in control and DMF-treated HL-60 and ARH-77, KG-1, and K-562 were unchanged or reduced by only other human tumor cells. Adenosine transport rates in HL-60 50%. Five days of treatment with 0.8% DMF did not evoke cells ranging from 295-620 pmol/min/106 cells gave a mean morphological changes in these cell lines, all of which continued value of 425 Â±110pmol/min/106 cells in a total of 16 deter to proliferate. minations. Four of these determinations were inhibitor-stop Transport of Deoxycoformycinand Coformycinin Control and assays, which gave a mean of 374 Â±100; 12 were oil-stop DMF-treated HL-60 Cells. The transport of deoxycoformycin assays, which gave a mean of 443 Â±110pmol/min/106 cells; and coformycin was measured by monitoring the inactivation thus, the results obtained by the 2 methods were not signifi of intracellular adenosine deaminase. These tight-binding in cantly different. Deoxyadenosine transport rates ranged from hibitors (30, 31) do not dissociate from adenosine deaminase 231-567 pmol/min/106 cells with a mean of 367 Â±132pmol/ during the time course of the experiment (34) and it was shown min/106 cells in 7 determinations. On some occasions a slower that each molecule of deoxycoformycin that initially enters a rate of adenosine or deoxyadenosine uptake was observed after cell inactivates a molecule of enzyme (27). In both control and 3-7 s. In those cases the transport rate was determined from DMF-treated HL-60 cells, the titration of the enzyme was linear the initial portion of the curve. There was no obvious relation with respect to time up to >80% inactivation. In control cells, ship between the initial transport rate and the occurrence of 2 /IM concentrations of deoxycoformycin and coformycin nonlinearity during the first 10s. Since both methods yielded caused >90 and >40% inactivation, respectively,of intracellular similar linear rates of transport for up to 10 s in most experi adenosine deaminase within 120 s (Fig. 3Â«).After a 5-day ments, the less laborious oil-stop assay was used for many of exposure to DMF, a comparable percentage of inactivation of these studies. the enzyme required more than 18 min (Fig. 3b). As shown in Fig. 2 shows a representative time course for the transport of Table 2, the activity of adenosine deaminase in HL-60 cells 100 /Â¿Madenosinein DMF-treated HL-60 cells. After cells were decreased from 3 units to 1 unit/109 cells after 5-6 days of 3451

Table 2 Changes in adenosine deaminase activity and transpon of 2 \im deoxycoformycin and coformycin in HL-60 cells after 5 days of exposure to DMF In cells exposed to DMF for 5 days, 5 //M NBMPR and dipyridamole both caused 80% inhibition of deoxycoformycin inactivation/minDeoxy (pmol/min/cells)Deoxy 10' influx and 63% inhibition of coformycin influx. The residual must- influx rate for both nucleosides was approximately 0.7 pmol/ (units/10' min/109 cells in the presence of either transport inhibitor and cells)"3.00 coformycin37.4 mycin18.9 coformycin113 mycin56 Control Â±0.13e Â±9 Â±27 may represent passive diffusion into the smaller differentiated 0.8% DMFAdenosineili-imi1.09 Â±0.04% 3.3 Â±0.5Cofor 1.8Transport*3.9 Â±0.6Cofor 1.9 cells. " One unit represents 1 umul Ml, liberated/min at 30*C. Adenosine influx was similarly affected by these transport " Calculated on the basis of 1:1 stoichiometry for adenosine deaminase inacti- inhibitors. In untreated cells, 5 HÃ•Aconcentrations of NBMPR vation and deoxycoformycin or coformycin influx. 1Mean Â±SD of two to four separate experiments. and dilazep caused 88% inhibition while dipyridamole caused 95% inhibition of adenosine influx. Deoxyadenosine transport was inhibited to a greater extent by all three inhibitors, almost 300- 99% by NBMPR and dilazep and about 99.5% by dipyridamole. In the DMF-treated cells, NBMPR and dipyridamole were

â€¢O NO INHIBITOR almost equipotent as inhibitors of adenosine transport, as seen above with coformycin and deoxycoformycin. A A 5 pM NBMPR Specific NBMPR Binding to Control and DMF-treated HL- â€¢O 5 uM DIPYRIDAMOLE 60 Cells. The modulation of adenosine and deoxyadenosine transport by DMF in HL-60 cells was further confirmed by the decrease in the number of NBMPR binding sites in the HL-60 cells after DMF treatment. A typical NBMPR binding study with untreated HL-60 cells is shown in Fig. 5. The specific binding of NBMPR was saturable in both control and DMF- treated HL-60 cells. The binding affinities and numbers of 10 14 18 22 specific NBMPR binding sites per cell of the HL-60 cells before MINUTES and after 5 days of treatment with 0.8% DMF are summarized in Table 3. A 20-fold reduction in the number of specific b. NBMPR binding sites per cell was observed in the DMF-treated 80 cells. In contrast, the affinity constant (A'(l)was not greatly â€¢O NO INHIBITOR affected by DMF. A A 5 pM NBMPR 060 Time Course of Effect of DMF on Transport of 100 UM â€¢D 5 ;jM DIPYRIDAMOLE Deoxyadenosine and on Cell Volume of HL-60 Cells. The time dependence of the effect of DMF on transport of 100 /tM 40

O TOTAL NBMPR BINDING T NON-SPECIFIC NBMPR BINDING 10- â€¢SPECIFIC NBMPR BINDING

\z 24 36 48 MINUTES Fig. 4. Effects of transport inhibitors on the influx of deoxycoformycin and coformycin in control (a) and DMF-treated (6) HL-60 cells. The rates of influx for 2 MMdeoxycoformycin (â€¢,A. â€¢)and 2 MMcoformycin (O, A, D) were determined from the rates of inactivation of adenosine deaminase (see Fig. 3). When the cells were preincubated without transport inhibitor, with 5 MMNBMPR, or with 5 MMdipyridamole, the respective influx rates per min per 10' cells were (a) 140, 36, and 1.1 pmol of deoxycoformycin and 59, 12, and 1 pmol of coformycin in control cells and (A) 3.5, 0.7, and 0.7 pmol of deoxycoformycin and 1.9, 0.7, and 0.7 pmol of coformycin in cells exposed to 0.8% DMF for 5 days. 50 100 NBMPR (nM) DMF treatment; therefore, the transport of both nucleosides, Fig. 5: Binding of ^-nitrobenzylthioinosine to HL-60 cells. Cell suspensions when expressed as pmol/min/IO9 cells was 29 times slower in in standard transport medium were preincubated at room temperature in the presence and absence of 50 MMdipyridamole for 15 min. After addition of graded the DMF-treated cells. concentrations of [3H]NBMPR, the cell suspensions were incubated for an addi tional 15 min. The radioactivities associated with the cell pellets were determined Effects of Nucleoside Transport Inhibitors. The influx of after centrifugation through 1-bromododecane as described in "Materials and deoxycoformycin and coformycin was sensitive to transport Methods." The specific [3H]NBMPR binding was determined from the difference inhibitors both before and after DMF-induced differentiation, between the binding of Â¡3H]NBMPRto the cells in the absence and presence of as illustrated in Fig. 4, a and b. A 1S-min preincubation with 5 dipyridamole. Â¿IMNBMPR caused 74% inhibition of deoxycoformycin influx and 80% inhibition of coformycin influx in control HL-60 cells. Table 3 Specific NBMPR binding to HL-60 cells before and after DMF In contrast, 5 MMdipyridamole caused 99 and 98% inhibition, treatment respectively. The residual influx rate in the dipyridamole-inhib- DMF, ited cells of approximately 1 pmol/min/10' cells may result 6days0.8 Apparent A,,(nM) (3.0)" (1.0) chiefly from passive diffusion. The effects of the transport Specific NBMPR binding 24 (30) x IO40.8% 1.2 (1.7) x IO4 inhibitor, dilazep (35) were similar to those of NBMPR (data sites/cellControl2.3 not shown). " Numbers in parentheses, results of a separate experiment. 3452

Downloaded from cancerres.aacrjournals.org on September 29, 2021. © 1986 American Association for Cancer Research. NUCLEOSIDE TRANSPORT AND DIFFERENTIATION IN HL-60 CELLS deoxyadenosine and on the cell volume of HL-60 cells is shown Both the oil- and inhibitor-stop assays showed that the rapid in Fig. 6. The rate of transport of deoxyadenosine was un influx of adenosine or deoxyadenosine was linear with respect changed during the first 24 h after treatment of HL-60 cells to time for up to 10 s in the untreated cells. After 5-6 days of with 0.8% DMF; then a sharp decrease in the rate of deoxy exposure to 0.8% DMF, influx rates were 10- to 20-fold slower adenosine transport to about 35-40% of the control rate oc and remained linear for 30 s or longer. Since extrapolation to curred between 24 and 26 h of incubation. The rate was further zero time gave data points similar to those obtained by addition decreased to about 4-5% of the control rate after cells were of transport inhibitors prior to the perineum, it is reasonable exposed to DMF for 4-6 days; thus, there was an overall to assume that initial velocities of transport were measured. In decrease of more than 20-fold observed in this experiment. this respect the HL-60 cells resemble HeLa cells which exhibit Morphological changes associated with granulocytic matura linear initial transport for up to 10s (29) whereas erythrocytes tion were not seen until after 3 days of DMF treatment. The show linear transport only prior to 3 s (35). The duration of cell volume also decreased upon treatment with 0.8% DMF but the linear initial transport velocities should be a function of the to a much lesser degree than the rate of transport. The volume surface area to volume ratios and of the initial rates and routes decreased progressively during the first 3 days of DMF exposure of intracellular metabolism, since the latter can serve to main to about one-half the original volume and remained essentially tain low intracellular permeant concentrations. unchanged thereafter. The influx of the nucleoside analogues, coformycin and deox Transport of Furine Bases in Control and DMF-treated HL- ycoformycin, was even more dramatically reduced after DMF- 60 Cells. Since profound effects on the rates of transport of induced differentiation than that of the natural permeants. nucleosides were observed, the transport rates of the purine Earlier studies with human erythrocytes (27) showed that the bases adenine, guanine, and hypoxanthine were compared in control and DMF-treated HL-60 cells. As shown in Table 4, rate of adenosine deaminase inactivation by deoxycoformycin was stoichiometric with the rate of influx of the radiolabeled the changes in the transport rates of these bases were much less pronounced than those seen with nucleoside transport and may compound; furthermore, the initial rate of inactivation was reflect the decreases in cell surface area related to the changes independent of the intraerythrocytic adenosine deaminase ac in volume that occur during myeloid differentiation. tivity over a wide range of enzyme concentrations. Although DMF-treated HL-60 cells have 33% of the adenosine deaminase activity of control cells, the transport of coformycin and deox DISCUSSION ycoformycin occurred at only 3% of the control rates. It is The induction of myeloid differentiation in HL-60 cells by important to note that since >80% of the enzyme was inacti exposure to 0.8% DMF is accompanied by profound changes vated in a linear fashion, it is highly unlikely that the slow rate in nucleoside transport. The reduction in nucleoside transport seen with the differentiated cells is due to the absence of capacity in the DMF-treated cells has been demonstrated by facilitated diffusion in the induced cells and a normal rate of three independent experimental procedures: (a) rapid oil- and transport in a small percentage of cells that have not responded inhibitor-stop assays for the influx of the radiolabeled natural to DMF. permeants, adenosine and deoxyadenosine; (b) the adenosine The transport rate of 2 Â¿Â¿Mdeoxycoformycininto erythrocytes at 30Â°Cwas 18 pmol/min/1 ml packed cells (27). This ex deaminase titration assay for the influx of the nucleoside ana logues, coformycin and deoxycoformycin; and (c) the equilib tremely slow rate of influx at micromolar concentrations could rium binding assay for the number of specific binding sites for be attributed to the very low affinity (K-,= 10 HIM)of deoxyco the nucleoside transport inhibitor, NBMPR. formycin for the erythrocytic nucleoside transporter. Although deoxycoformycin transport at 30Â°Cwas approximately 6 times faster into HL-60 cells (113 pmol/min/109 cells), this rate was only about 0.2% of that seen with 2 ^M adenosine at room 100- temperature;5 thus, it is likely that the nucleoside transporter of HL-60 cells also has very low affinity for both deoxycofor mycin and coformycin. In 6 of 7 human leukemia cell lines ife 50- studied to date (36),5 deoxycoformycin influx was 1.3-2.4 times faster than that of coformycin. In HL-60 cells, deoxycoformycin was transported approximately twice as fast as coformycin both before and after DMF treatment. 23456 Further confirmation that the profound decreases in the rates EXPOSURE (DAYS) of nucleoside uptake represent decreased transport capacity was Fig. 6. Time dependence of the change in deoxyadenosine transport and in the cell size of 1II, 60 cells. Exposure of 5 x IO9cells/ml (initial concentration) provided by the observation of a 19-fold reduction in the num to 0.8% DMF was performed as described in "Materials and Methods." At the ber of specific NBMPR binding sites in HL-60 cells after 6 indicated time intervals, a portion of the cell suspension was removed and the days of treatment with 0.8% DMF. Jarvis and Young (32, 37) rate of transport of 100 ^M deoxyadenosine and cell volume were determined. have presented genetic evidence that high affinity binding of Table 4 Effect of DMF on parine base transpon in HL-60 cells NBMPR represents interaction with functional nucleoside transport sites. In nucleoside impermeable sheep erythrocytes, cells)"Control50.9(pmol/min/1 06 site-specific binding was not detected, whereas such binding (K,i

tration about 0.5 HM)was demonstrable with erythrocytes from related BaseAdenine G.M)3.533.72 control49.1 sheep in which an allelomorphic gene controlling nucleoside transporter function was expressed; furthermore, it was shown Guanine 40.9 15.3 37.4 that species differences in nucleoside transport capacity are HypoxanthineConcen 3.72Transport 31.4DMF-treated25.011.5%of 36.5 " Mean of two determinations. * Unpublished data. 3453

Downloaded from cancerres.aacrjournals.org on September 29, 2021. © 1986 American Association for Cancer Research. NUCLEOSIDE TRANSPORT AND DIFFERENTIATION IN HL-60 CELLS attributable to differences in the cell surface content of specific examining the regulation of nucleoside transport capacity in a NBMPR binding sites (38). human cell. Experiments in progress will characterize the ef Although NBMPR-insensitive tumor cell lines have been fects of a variety of maturational agents and attempt to elucidate characterized (39, 40), the bulk of the influx of nucleosides in the mechanism responsible for the rapid decrease in functional the majority of cell types studied to date can be blocked by nucleoside transport sites. NBMPR as well as other transport inhibitors (41). HL-60 cells were still highly sensitive to transport inhibitors after DMF- ACKNOWLEDGMENTS induced differentiation, which indicates that the slow influx rates seen with the induced cells were largely attributable to facilitated diffusion. If the dipyridamole-insensitive transport We thank Drs. A. R. P. Paterson and M. Gudenzi for providing of deoxycoformycin, 1 pmol/min/109 untreated cells and 0.7 NBMPR and dilazep and T. P. Zimmerman for helpful suggestions regarding the inhibitor-stop procedure. We are also grateful to Jeffrey pmol/min/109 induced cells, represents passive diffusion, then M. Stein, Matthew Abbate, and Shih-Ying Li for their fine technical it is not likely that DMF-treated HL-60 cells undergo a large assistance in some of these studies. change in surface area; consequently, the reduced transport capacity would represent a decrease in the number of functional REFERENCES transport sites per unit cell surface area. About 10-20% of the influx of adenosine, coformycin, and deoxycoformycin in control HL-60 cells is insensitive to 5 /tM 1. Collins, S. .1.. Gallo, R. C., and Gallagher, R. E. Continuous growth and differentiation of human myeloid leukemic cell in suspension culture. Nature NBMPR but can be blocked by 5 /iM dipyridamole. This (Lond.), 270: 347-349, 1977. NBMPR-insensitive transport component was not observed 2. Gallagher, R., Collins, S., Trujillo, J., McCredie, K., Aheam, M., Tsai, S., Metzgar, R., Aulakh, G., Ting, R., Ruscetti, F., and Gallo, R. Characteriza with deoxyadenosine in the control cells and could not be tion of the continuous differentiating myeloid cell line (HL-60) from a patient detected at all after DMF treatment. The 2-fold decrease in the with acute promyelocytic leukemia. Blood, 54:713-733, 1979. high-affinity dissociation constant of NBMPR following DMF 3. Lotern. J., and Sachs, L. Regulation of normal differentiation in mouse and human myeloid leukemic cells by phorbol esters and the mechanism of tumor treatment (Table 3) may not be statistically significant or be promotion. Proc. Nati. Acad. Sci. USA, 76: 5158-5162, 1979. 4. Miyaura, (.'., Abe, E., Kuribayashi, T., Tanaka, II., Konno, K., Nishii, V.. expected to influence significantly the effects seen at 5 /Ã•M and Suda, T. la,25-Dihydroxyvitamin I), induces differentiation of human NBMPR concentrations. The differences in sensitivity to myeloid leukemia cells. Biochem. Biophys. Res. Commun., 102: 937-943, NBMPR could be due to changes in low-affinity binding sites 1981. 5. Rovcra, G., Santoli, D., and Damsky, C. Human promyelocytic leukemia or allosteric mechanisms. Studies are in progress to determine cells in culture differentiate into macrophage-like cells when treated with a whether the affinities of permeants or other inhibitors change phorbol diester. Proc. Nati. Acad. Sci. USA, 76:2779-2783, 1979. after a 5- to 6-day exposure of HL-60 cells to DMF. 6. Collins, S. J., Ruscetti, F. W., Gallagher, R. E., and Gallo, R. C. Terminal differentiation of human promyelocytic leukemia cells induced by dimethyl In summary, the profound reduction in the rates of nucleoside sulfoxide and other polar compounds. Proc. Nati. Acad. Sci. USA, 75:2458- transport after DMF exposure is corroborated by a comparable 2462, 1978. decrease in the number of high affinity-binding sites for 7. Bodner, A. J., Ting, R. C., and Gallo, R. C. Induction of differentiation of human promyelocytic leukemia cells (HL-60) by nucleosides and methotrex- NBMPR. This decrease in the nucleoside transport capacity in ate. J. Nati. Cancer Inst., 67: 1025-1030, 1981. HL-60 cells appears to be a consequence of DMF-induced 8. Breitman, T. R., Selonick, S. E., and Collins, S. J. Induction of differentiation of the human promyelocytic leukemia cell line (HL-60) by retinole acid. differentiation rather than a result of nonspecific direct effects Proc. Nati. Acad. Sci. USA, 77: 2936-2940, 1980. of this polar solvent on the transporter protein or the plasma 9. Collins, S. J., Bodner, A. J., Ting, R. C., and Gallo, R. C. Induction of membrane. This conclusion is based on the finding that there morphological and functional differentiation of human promyelocytic leu kemia cells (HL-60) by compounds which induce differentiation of murine is no change in the transport rate of 100 tiM deoxyadenosine leukemia cells. Int. J. Cancer, 25: 213-218,1980. during the first 24 h of incubation with DMF (Fig. 6) and there 10. Honma, Y., Fujita, Y., Okabe-Kado, J., Kasukabe, T., and Hozumi, M. is little or no change in the nucleoside transport rates of KG-1, Induction of differentiation of human promyelocytic leukemia cells (HL-60) by arginase. Cancer Lett., 10:287-292, 1980. K-562, and ARH-77 cells that are not inducible with DMF 11. Mendelsohn, N., Michl, J., Gilbert, H.S., Acs, G., and Christman, J. K. L- (Table 1); furthermore, preliminary studies5 indicate that other Ethionine as an inducer of differentiation in human promyelocytic leukemia cells (HL-60). Cancer Res., 40: 3206-3210, 1980. myeloid differentiation inducers also cause a reduction in the 12. Schwartz, E. L., and Sartorelli, A. C. Structure-activity relationships for the nucleoside transport capacity of HL-60 cells and that the rate induction of differentiation of HL-60 human acute promyelocytic leukemia of nucleoside transport in normal granulocytes resembles that cells by anthracyclines. Cancer Res., 42: 2651-2655,1982. 13. Stendati!, O., Dahlgren, C., and Hed, J. Physicochemical and functional of the DMF-treated HL-60 cells. changes in human leukemic cell line HL-60. J. Cell. Physio)., 112: 217-221, The rapid 60-65% reduction in the transport capacity after 1982. 24-26 h of exposure to DMF precedes the reported onset of 14. Felsted, R. L., Gupta, S. K., Glover, C. J., Fischkoff, S. A., and Gallagher, R. E. Cell surface membrane protein changes during the differentiation of morphological maturation and is therefore a very early marker cultured human promyelocytic leukemia HL-60 cells. Cancer Res., 43: 2754- of the induction of myeloid differentiation. In HL-60 cells 2761, 1983. exposed to dimethyl sulfoxide there is an irreversible 70-80% 15. Gahmberg, C. G., Nilsson, K., and Andersson, L. C. Specific changes in the surface glycoprotein pattern of human promyelocytic leukemic cell line HL- decrease in cloning efficiency after 24 h even though cells 60 during morphologic and functional differentiation. Proc. Nati. Acad. Sci. transferred to drug-free medium do not differentiate to granu USA, 76:4087-4091,1979. 16. Cossu, G., Kuo, A. L., l'essano. S., Warren, L., and Cooper, R. A. Decreased locytes (42). Detailed studies of DMF effects on functional or synthesis of high-molecular-weight glycopeptides in human promyelocytic biochemical parameters of HL-60 cells have not been reported, leukemic cells (HL-60) during phorbol ester-induced macrophage differentia but DMF may resemble dimethyl sulfoxide in its actions. tion. Cancer Res., 42:484-489, 1982. 17. Mulder, A., Alexander, S., Engelfriet, C. P., von dem Borne, A. E. G., Kr., The reduction in functional nucleoside transporter molecules and Strominger, J. L. Characterization, by immunoprecipitation, of myeloid- appears to be a specific feature of the induction process in HL- and monocyte-specific antigens present on the human promyelocytic cell line (HL-60) in three stages of differentiation. Proc. Nati. Acad. Sci. USA, 78: 60 cells; thus, modulation of nucleoside transport provides a 5091-5095, 1981. new parameter for exploring the mechanisms of action of 18. Gordon, P. B., and Rubin, M. S. Membrane transport during erythroid differentiation-inducing agents. Such agents, it is hoped, will differentiation. J. Membr. Biol., 64: 11-21, 1982. 19. Koren, R., Shohami, E., Bibi, O., and Stein, W. D. Vridine transport lead to more specific and less toxic cancer chemotherapy (43, properties of mammalian cell membranes are not directly involved with 44); furthermore, this cell line provides a convenient model for growth control or oncogenesis. FEBS Lett., 86: 71-75, 1978. 3454 Downloaded from cancerres.aacrjournals.org on September 29, 2021. © 1986 American Association for Cancer Research. NUCLEOSIDE TRANSPORT AND DIFFERENTIATION IN HL-60 CELLS

20. MÃ¤rz,K.. Wohlhueter, R. M.. and Plagemann, P. G. W. Metabolic stability 33. Woo, P. W. K., Dion, H. W., Lange, S. M., Dahl, L. F., and Durham, L. J. of the nucleoside transport system of Novikoff rat hepatoma cells. J. Supra- J. A novel adenosine and ara-A deaminase inhibitor, (A)-3-(2-deoxy-0-r> mol. Struct., 6:433-440, 1978. erKf/iro-pentofuranosyl)-3,6,7,8-tetrahydroimidazo[4,5-d)[l,3]diazepin-8-ol. 21. Cass, C.E., Dahlig, E., Lau, E. Y., Lynch, T. P., and Paterson, A. R. P. J. Heterocycl. Chem., //: 641-645, 1974. Fluctuations of nucleoside uptake and binding of the inhibitor of nucleoside 34. Rogler-Brown, T., Agarwal, R. P., and Parks, R. E., Jr. Tight-binding transport, nitrobenzylthioinosine, during the replication cycle of Hoi .a cells. inhibitorsâ€”VI. Interactions of deoxycoformycin and adenosine deaminase Cancer Res., 39:1245-1252, 1979. in intact human erythrocytes and Sarcoma 180 cells. Biochem. Pharmacol., 22. Chen, S. F., Stoeckler, J. D., Wiemann, M. C., Hollmann, A. B., and Parks, 27: 2289-2296, 1978. R. E., Jr. Modulation of nucleoside transport of HL-60 cells by the differ 35. Paterson, A. R. P., Harley, E. R., and Cass, C. E. Inward fluxes of adenosine entiation inducer, WV-dimethylformamide. Proc. Am. Assoc. Cancer Res., in erythrocytes and cultured cells measured by a quenched flow method. 25:64, 1984. Biochem. J., 24:1001-1008, 1985. 23. Lozzio, C. B., and Lozzio, B. B. Human chronic myelogenous leukemia cell 36. Parks, R. E., Jr., Dawicki, D. D., Agarwal, K. C., Chen, S. F., and Stoeckler, line with positive Philadelphia chromosome. Blood, 45: 321-331,1975. J. D. Role of nucleoside transpon in drug action: the adenosine deaminase 24. Koeffler, H. P., and Golde, D. W. Acute myelogenous leukemia: a human inhibitor, deoxycoformycin, and the antiplatelet drugs, dipyridamole and cell line responsive to colony-stimulating activity. Science (Wash. DC), 200: dilazep. Ann. NY Acad. Sci., 4SI: 188-203, 1986. 1153-1154,1978. 37. Jarvis, S. M., and Young, J. D. Direct evidence for the specific interaction 25. Burk, K. H., Drewinko, B., Trujillo. J., and Ahearn, M. J. Establishment of of nitrobenzylthioinosine with functional nucleoside transpon sites in sheep a human plasma cell line in vitro. Cancer Res., 38: 2508-2513, 1978. erythrocyte membranes. J. Physiol. (I .mid.). 284: 96P-97P, 1978. 26. Collins, S. J., Ruscelli, F. W., Gallagher, R. E., and Gallo, R. C. Normal 38. Jarvis, S. M., Hammond, J. R., Paterson, A. R. P., and Clanachan, A. S. function;!] characteristics of cultured human promyelocytic leukemia (HL- Species differences in nucleoside transponâ€”a study of uridine transpon and 60) after induction of differentiation by dimethyl sulfoxide. J. Exp. Mod.. nitrobenzylthioinosine binding by mammalian erythrocyles. Biochem. J., Â¡49:969-974, 1979. 208: 83-88, 1982. 27. Chen, S. F., Sloeckler, J. D., and Parks, R. E., Jr. Transpon of deoxycofor- 39. Belt, J. A. Nitrobenzylthioinosine-insensitive undine transpon in human mycin in human erythrocytes: measurement by adenosine deaminase titrat ion lymphoblastoid and murine leukemia cells. Biochem. Biophys. Res. Com and radioisolope assay. Biochem. Pharmacol., 33:4069-4079, 1984. mun., 110:417-423, 1983. 28. Michaels, E. B., Hahn, E. C., and Kenyon, A. J. An improved procedure for 40. 11ir, L. H., and Dubach, U. C. Uphill transpon of pyrimidine nucleosides in solubilization and assay of blood and feces in liquid scintillation counting. renal brush border vesicles. Pfluegers Arch. Eur. J. Physiol., 404: 238-243, Anal. Biochem., 99:288-296, 1979. 1985. 29. Hurley. E. R., Paterson, A. R. P., and Cass, C. E. Initial rate kinetics of the 41. Paterson, A. R. P., Jakobs, E. S., Harley, E. R., Fu, N. W., Robins, M. J., transpon of adenosine and 4-amino-7-(/3-D-ribofuranosyl)pyrrolo[2,3-d]py- and Cass, C. E. Inhibition of nucleoside transpon. In: R. M. Berne, T. W. rimidine (tubercidin) in cultured cells. Cancer Res., 42: 1289-1295, 1982. Rail, and R. Rubio (eds.), Regulalory Function of Adenosine, pp. 203-220. 30. Agarwal, R. P., Spector, T., and Parks, R. E., Jr. Tighl-binding inhibitorsâ€” Boston: Martinus Nijhoff Publishers, 1983. IV. Inhibition of adenosine deaminase by various inhibitors. Biochem. Phar 42. Fibach, E., Peled, T., and Rachmilewitz, E. A. Self-renewal and commit meni macol., 26:359-367,1977. to differentiation of human leukemic promyelocytic cells (HL-60). J. Cell. 31. Cha, S., Agarwal, R. P., and Parks, R. E., Jr. Tight-binding inhibitorsâ€”II. Physiol., 113: 152-158, 1982. Non-steady state nature of inhibition of milk xanthine oxidase by allopurinol 43. Koeffler, H. P. Induction of differentiation of human acute myelogenous and alloxanthine and of human erythrocytic adenosine deaminase by cofor- leukemia cells: therapeutic implications. Blood, 62: 709-721, 1983. mycin. Biochem. Pharmacol., 24: 2187-2197, 1975. 44. Menimi. H., and Breitman, T. R. Induction of differentiation of the human 32. Jarvis, S. M., and Young, J. D. Nucleoside transpon in human and sheep promyelocytic cell line HL-60 and primary cultures of human leukemia cells: erythrocylesâ€”evidence that nitrobenzylthioinosine binds specifically to func a model for clinical treatment. In: R. I. Glazer (ed.), Developments in Cancer tional nucleoside transpon sites. Biochem. J., 790:377-383, 1980. Chemotherapy, pp. 247-280. Boca Raton: CRC Press Inc., 1984.

3455

Downloaded from cancerres.aacrjournals.org on September 29, 2021. © 1986 American Association for Cancer Research. Changes in Nucleoside Transport of HL-60 Human Promyelocytic Cells during N,N-Dimethylformamide Induced Differentiation

Shih-Fong Chen, Jeffrey S. Cleaveland, Ann B. Hollmann, et al.

Cancer Res 1986;46:3449-3455.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/46/7/3449

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/46/7/3449. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.