Download - Effect of the protein kinase C inhibitor staurosporine on chemosensitivity to daunorubicin of normal and leukemic fresh myeloid cells

Effect of the Protein Kinase C Inhibitor Staurosporine on Chemosensitivity to Daunorubicin of Normal and Leukemic Fresh Myeloid Cells

By Judith Laredo, Anne Huynh, Catherine Muller, Jean-Pierre Jaffrezou, Jean-Denis Bailly, Georges Cassar, Guy Laurent, and Cecile Demur

The effect of the protein kinase C (PKC) inhibitor staurosporine (ST) on the chemosensitivity of normal (colony- forming unit granulocyte-macrophage [CFU-GM]) and leukemic (acute myeloid leukemia-CFU [AML-CFUI) myeloid progenitors to daunorubicin (DNR) was evaluated. Primary colony inhibition assays allowed us to characterize two distinct groups of AML, a DNR-resistant group (patients no. 1 through 61, which displayed significantly lower DNR sensitivity than normal CFU-GM (Dso = 11.3 t 1.4 ng/mL v 1.8 rC_ 0.5 ng/mL, after 7 days of exposure, respectively; P < 0.01) and a DNR-sensitive group (patients no. 7 through 12) with 0 5 0

= 2.7 2 0.4 ng/mL. This classification remained unaltered when assessed by secondary colony inhibition assay (evaluating the self-renewal fraction of AML-CFU) or by viability assay (evaluating the ultimately differentiated blast cell population), suggesting that the DNR sensitivity profile is maintained throughout AML-CFU differentiation. DNR resistance

RUG RESISTANCE is the major cause of treatment failure in acute myeloid leukemia (AML). Among the

drugs widely used in the treatment of AML is the anthracycline daunorubicin (DNR), a potent antitumor antibiotic. One mechanism by which leukemic cells are, or become, resistant to chemotherapy implicates the multidrug-resistance (MDR) gene, mdrl, which encodes a transmembrane glycoprotein (P-gp) that is able to expel1 a number of structurally unrelated cytotoxic drugs from the cytosol.1 P-gp substrates include anthracyclines, Vinca alkaloids, epipodophyllotoxins, and actinomycin D. The increase of mdrl RNA or P-gp expression that has been reported in large series of AML patients at diagnosis or relapse, supports the hypothesis that the MDR phenotype is involved in the resistance of AML to DNR.’

P-gp has been described to be in a phosphorylated state in vivo: and it has also been reported that protein kinase C (PKC) can modulate P-gp’s activity and, therefore, drug re~istance.~ Indeed, inhibitors of PKC such as staurosporine (ST),5 isoquinoline sulfonamides,6 myristylated synthetic oc- tapeptides,’ and others lipophilic compounds’ can sensitize MDR-resistant cells by increasing the intracellular accumulation of cytotoxic drugs. On the other hand, PKC activators, such as phorbol esters9 and diacylglycerol derivatives,“ can increase MDR resistance by decreasing the intracellular drug accumulation. However, there are conflicting results. For example, P-gp phosphorylation is also promoted by agents that reverse MDR, such as verapamil and trifluoperazine, and that have been reported as PKC inhibitor^.^ Posada et al” reported that prolonged exposure of MDR cells to phor- bo1 esters, which downregulated PKC, actually increased drug resistance. Furthermore, inhibitors of PKC generally show little specificity and several PKC inhibitors such as ST and H-85 may also act by inhibiting P-gp function through their direct binding to Finally, as far as we know, the influence of agonist or antagonist of PKC on chemosensitivity to MDR-related drugs has not been yet evaluated on fresh tumor cells. In summary, present knowledge on the relationship between PKC activity, drug sensitivity, and P- gp function are wanting and need to be further elucidated

D

Blood, VOI 84, NO 1 (July l) , 1994: pp 229-237

of the differentiated blast cell population was not correlated with the level of P-glycoprotein (P-gp) expression but rather with the ability to extrude rhodamine 123 (Rh123). ST used at subtoxic concentrations induced a twofold to threefold enhancement of DNR cytotoxicity, increased Rh123 accumulation, and decreased Rh123 efflux kinetics in resistant AML cells. These effects were observed for ST concentrations much lower than those required to displace the P-gp-binding probe azidoprazosin, suggesting that ST might act through its PKC inhibitory effect and not through P-gp binding. Finally, this study provides evidence that DNR resistance in AML cells is, at least in part, related to the multidrug- resistance (MDR) phenotype. Because P-gp function can be downregulated by ST, it seems likely that the MDR phenotype can be functionally regulated by cellular signalization in AML cells. 0 1994 by The American Society of Hematology.

in P-gp-expressing cells, particularly in clinical relevant cellular models such as fresh leukemic cells. Therefore, we designed a study that was aimed at evaluating the impact of ST, a potent although nonspecific PKC inhibitor, on the chemosensitivity profile to DNR and P-gp function of fresh AML cells.

MATERIALS AND METHODS

Drugs. DNR (Cerubidine) was purchased from Laboratoire Roger Bellon (Neuilly-sur-Seine, France). All other drugs and re- agents were purchased from Sigma Chemical CO (St Louis, MO).

Monoclonal antibodies. C219 (IgG2a) was obtained from Cen- tocor (Malvern, PA). This antibody detects an intracellular epitope present on mdr-l and mdr3 isoforms of P-gp.” Antimouse IgG peroxydase conjugate was purchased from Sigma.

Leukemic myeloblasts. Before treatment, fresh leukemic cells were obtained from unselected AML patients at diagnosis or first relapse. Table 1 summarizes the main clinical features. Bone marrow (BM) aspirates were collected in heparinized syringes and mononuclear cells were separated by centrifugation through a Ficoll-Hy- paque density gradient. Cells were washed twice in Iscove’s modified Dulbecco’s medium (IMDM; GIBCO, Grand Island, NY) resuspended at a final concentration of 2 X lo7 cellslmL and cryopre-

From the Laboratoire de Phamcologie et de Toxicologie Fonda- mentales, CNRS; the Laboratoire de Cytologie et CytoghCtique, Centre de Transfusion Sanguine; and the Service d’Ht!mtologie, CHU Purpan, Toulouse, France.

Submined October 8, 1993; accepted March 7, 1994. Supported in part by grants from la Fkdkration Nationnale des

Centres de Lune Contre le Cancer and from INSERM (Contrat de Recherche Externe No. 920411).

Address reprint requests to Guy Laurent, MD, Service d’Hkmatolo- gie, CHU Purpan, Place du Dr Baylac, 31059 Toulouse, France.

The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked “advefiisement” in accordance with 18 U.S.C. section 1734 solely to indicate this fact. 0 1994 by The American Society of Hematology. OOO6-4971/94/8401-0002$3.00/0

229

230 LAREDO ET AL

Table 1. Patient's Characteristics

Percentage of Previous Patient No. Age 1/11 Type FAB Type Blast CellX Therapy PE 1 (%l PE 2 (%l Karyotypic Analysis

1 73 II M 2 53 - 0.1 1 0.02 47,XY,de1(5)(q22q23),i(8ql 2 19 I M 2 35 DNR-CA 2.3 0.6 46,XY,t(8;21)(q22;q22) 3 74 I M1 89 - 2.6 1.5 46,XX 4 71 I M 4 85 - 1.6 3.6 46,XY 5 19 I M 2 47 DNR-CA 2.5 1.5 46,XY,t(6;9)(p23;q34) 6 48 I MO 78 - 1.4 0.04 48,XY,-2,-7.+8,der(ll)

7 66 I1 (MDS) M 2 56 CA 0.33 0.22 N D 8 30 I M 4 49 + 14 Pm - 1.8 0.2 46 XX 9 60 I M5 93 DNR-CA 2.5 3.4 46,XX,t(5;13)(q35;q14)

10 59 I M 4 84 DNR-CA 1.4 0.6 N D 11 80 II (CML) M 2 37 HD; 6MP 1.04 2.1 46,XY,t(9;22)(q34;qlI) 12 6 I M5 83 - 1.14 0.83 46,XY,t(9; 1 l)(p21;q23)

t(2; 1 l)(q27;q23),+marx3

Abbreviations: I, de novo AML; H, secondary AML; FAB, French-American-British classification; MDS, myelodysplastic syndrome; CML, chronic myeloid leukemia; PE1, primary plating efficiency; PE2, secondary plating efficiency from pooled primary colonies; Pm, promonocytes; CA, cytosine-arabinoside; HD, hydroxyurea; 6MP. 6-mercaptopurine; ND, not determined.

* Blast cell percentage before Ficoll.

served in IMDM containing 10% dimethyl sulfoxide and 50% fetal calf serum (FCS; Flow Laboratory, Paris, France). Percentages of blast cells on BM smears at diagnosis varied from 50% to 99%. After processing (Ficoll separation, freezing, and thawing), the percentage of leukemic cells was higher than 90% in all cases.

Normal BM mononuclear cells. Normal BM samples were obtained from healthy allograft donors after informed consent. Mono- nuclear cells were collected as described above and processed without cryopreservation.

Leukemic cell lines. Human chronic myelogenous leukemia K562 cells and the doxorubicin-resistant subline K562/R7 were gen- erous gifts from Dr B.I. Sikic (Stanford University, CA). HEL eryth- roleukemia cells were obtained from the ATCC. Cell lines were grown in IMDM supplemented with 10% FCS and antibiotics: strep- tomycin (100 pg/mL) and penicillin (100 U/mL).

Clonogenic and cell viability assays. After thawing, blast cells were washed twice in IMDM and resuspended in IMDM supplemented with 15% FCS, 10% human bladder carcinoma cell line 5637 conditioned medium (5637 CM) in the presence or absence of ST used at ICl,,. In preliminary experiments, ST concentrations that induced a 10% inhibition of colony formation after 7-day exposure (ICl,,) were calculated from dose response curves for each BM sample.13 IClo varied from 0.05 to 4 nmoVL depending on the patients. After a 24 hour-incubation with ST in liquid culture at 37"C, blast cells were washed, and the blast primary colony assay (PEI) was performed as previously described.'3 A total of 1 X 10' cells/mL was resuspended in IMDM supplemented with 30% FCS, 0.5 mmoV L p-mercaptoethanol (GIBCO), 0.9% methylcellulose (Fisher Scien- tific, Boston, MA), 10% 5637 CM, and increasing concentrations of DNR in the presence or absence of ST at IClo. A total of l@ cells was then plated in 35-mm Petri dishes in 6 replicates. The plates were then incubated at 37°C in a humidified atmosphere of 5% CO2 for 7 days. Leukemic cell colonies (more than 20 cells) and clusters (more than 5 cells) were then scored under an inverted microscope. In each case, the leukemic nature of AML colony-fonning unit (AML-CFU) colonies was confirmed by the morphologic study after Giemsa staining, cytochemistry (Sudan black B and nonspecific es- terases), and immunophenotypic analysis from pooled colonies. In 6 cases, cytogenetic analysis of pooled blast cell colonies could be performed. In these cases, the same cytogenetic abnormality was identified in fresh AML cells and in 100% of the observed mitosis of pooled colonies.

Self-renewal (PES) was measured by the method described by Miyauchi et al.I4 Cells were preincubated during 24 hours with or without ST at IClo as described above. Cells were then washed, and IO6 blast cells were cultured in 1 mL IMDM containing 15% FCS, 10% 5637 CM, and increasing concentrations of DNR in the presence or absence of ST at IClo in 24-well dishes (Nunc, Roskilde, Denmark) for 7 days. Cells were then washed, enumerated using an automatic cell counter, and resuspended at a concentration of 1 X IO5 cells in the same semisolid medium as for PE1 without DNR or ST. Colonies and clusters were scored on day 7.

For the cell viability assay, trypan blue exclusion was performed after a 7-day suspension culture under the same conditions as described above for the liquid-phase culture of PES.

Normal myeloid progenitors were processed according to the method of CFU-granulocyte-macrophage (CFU-GM) as described e1~ewhere.l~ After a 24-hour preincubation with or without ST at the unique concentration of 1.75 nmol/L (IC10)'3 in the same condition as described for PE1, cells were washed, and IO5 cells were plated in the same semisolid medium as described for PE1 containing increasing concentrations of DNR in the presence or absence of 1.75 nmol/L ST. Colonies consisting of more than 50 cells were scored on day 7 and day 14.

Individual results for each AML-CFU or CFU-GM assay were expressed as DSo (drug concentration needed to induce 50% inhibition of colony formation). For each sample, D',, value was derived from the slope of the drug-effect curve by semilogarithmic regression. Each sample was tested in at least two independent experiments. The data are shown as the mean ? SEM. Wilcoxon's test was used for comparing the drug-induced cytotoxic effect on PE1 and PES as well as for comparing DNR cytotoxicity in resistant and sensitive groups. The Mann-Whitney test was used for comparing the cytotoxicity on leukemic and normal hematopoietic progenitor cells. Correlations were calculated with the Spearman test. All calcu- lations were made by using a computerized statistics program (PCS M stat; Deltasoft, Grenoble, France). Differences with a P value greater than .05 were not considered significant.

For K562/R7 cells, clonogenic assays were performed on methylcellulose semisolid medium without exogenous growth factor in the presence of 10% FCS. ST was used at IClo as for fresh blast cells.

Western blor. Cells were centrifuged at room temperature for 3 minutes at 900g, washed twice in phosphate-buffered saline (PBS; GIBCO), and lysed in single-detergent lysis buffer (50 mmoVL Tris-

PKC INHIBITOR STAUROSPORINE 231

HCl pH 8.0, 150 mmoVL NaCl, and 1% NP-40) in the presence of protease inhibitors (1 mmoUL phenylmethylsulfonyl fluoride (PMSF), 1 mg/mL aprotinin, and 30 pmoVL leupeptin) for 45 minutes at 4°C. The solution was then clarified by centrifugation at 10,000g for 20 minutes. Total cell proteins were quantitated using a modification of the Lowry method15 on the supernatant solution. Samples (200 pg of protein) were resolved using modified Fairbanks gel electrophoresis procedurei6 and transferred to nitrocellulose in Towbin bufferi7 at 200 mA overnight. Western blot procedure was performed using the enhanced chemoluminescence (ECL) detection system following the conditions recommended by the manufacturer (Amersham International, UK). The blots were incubated in blocking buffer (10% nonfat dried milk in Tris-buffered saline) at room temperature for 2 hours. Antibody incubation was performed at room temperature in diluted (1 : 1,000) C219 solution (Tris buffered saline- Tween 0.1%). Blots were washed 3 times for 5 minutes in the same buffer and incubated with goat antimouse peroxidase-labeled secondary antibody for 1 hour at room temperature in TBS-T buffer. The wash procedure was repeated, and blots were incubated in detection reagent for 1 minute at room temperature. Blots were then exposed to film for 1 to 5 minutes and analyzed by densitometry.

Rhodamine 123 (Rh123) eflux. To study the efflux of Rh123, cells (5 X 10s/mL) were stained with 150 ng/mL Rh123 for 10 minutes at 37°C in IMDM supplemented with 5% FCS." Cells were then centrifuged at 4°C and resuspended in Rhl23-free medium. A total of 5 X 10' cells was taken for flow cytometry analysis. The remaining cells were incubated at 37°C in the presence or absence of verapamil (15 pmol/L) and ST at various concentrations. At the indicated times, 5 X lo5 cells were taken for analysis. Cells were kept on ice until analysis. Samples were analyzed on a Coulter (Hialeah, FL) Epics Elite flow cytometer (excitation wave length, 505 nmoyL, emission wave length, 534 nmoVL).

Metabolic labeling of cells and immunoprecipitation of P-gp. Cells were washed twice with phosphate-free a-minimum essential medium (a-MEM) supplemented with 5% dialyzed FCS, then resuspended with the same buffer containing 0.20 mCi/mL (32P) (ortho- phosphate HCI-Free; Amersham, Little Chalfont, UK), and incubated at 37°C for 3 hours. The cells were incubated 1 more hour in the presence or absence of ST. After 2 washings with PBS, labeled cells were lysed 1% sodium dodecyl sulfate SDS, 50 mmoVL Tris-HC1, pH 7.4 (buffer A), in the presence of protease inhibitors (1 mmoY L PMSF, 30 leupeptin, and 4 p g / d pepstatin A) for 2 minutes at room temperature. Samples were diluted to 1.25% triton X-100, 190 mmoK NaCl in 50 mmol/L Tris buffer, pH 7.4 (buffer B), and disrupted by sonication. The solution was then clarified by centrifugation at 10,OOOg for 20 minutes. Total cell proteins were quantified using a modification of the Lowry methodI5 on the supernatant solution. Acid precipitable radioactivity was also counted after trichloro- acetic acid precipitation. Immunoprecipitations were performed by incubating cell extracts (50 pg of total protein from each sample) with 10 pg of monoclonal antibody C219 overnight at 4°C in buffer C (1: 10 buffer A:buffer B). A total of 100 pL of Protein A-Sepharose CL4B (6.7% voVvol in buffer C) was then added. After incubation for 3 hours at 4°C with constant gentle mixing, the precipitates were washed 4 times with 1 mL of buffer D (0.1% Triton-X 100, 0.03% SDS, 150 mmom NaCl in 50 mmol/L Tris, pH 7.4) and 1 time in buffer E (150 mmoVL NaCl in 50 mmoVL Tris, pH 7.4). Samples were solubilized in 2% SDS with 50 mmoK dithiothreitol in 10 mmoVL Tris (pH 8.0), incubated at room temperature, brought to a concentration of 4.5 mmol/L urea, and resolved on modified Fair- banks gels.16 The gels were fluorographed and exposed to x-ray film at -70°C.

Photoaffinity labeling. Cells were washed in PBS and resuspended in a buffer that consisted of 10 mmoVL Tris-HCI (pH 7.4) with 10 mmol/L NaCI, 1.5 mmom MgC12. and 0.02 mmoVL PMSF.

Cells were homogenized with 20 strokes using a dounce homoge- nizer; the cell homogenate was overlaid, onto a 35% sucrose solution and centrifuged for 60 minutes at 18,OOOg. The membrane fraction found at the interface was then centrifuged for 60 minutes at 100,OOOg. The membrane pellet was suspended in a Tris-HC1 (pH 7.4) buffer supplemented with 250 mmoVL sucrose. Protein concentration was determined by the method of Lowry et al . '5 Ideally, membranes were used the same day of isolation or frozen at -70°C for up to 3 days.

Photoaffinity labeling of cell membranes was performed according to previously described method." A total of 100 pg of membrane protein was incubated with 2.5 nmoVL of (iUI)-iodoaryl azidoprazosin (81.4 TBq/mmol; NEN-Dupont, Boston, MA) with and without drugs in a 10 mmoVL Tris-HC1 buffer (pH 7.4) at a final volume of 50 pL. These preparations were incubated at 25°C for 1 hour in the dark, followed by 20 minutes of exposure to a 366-nm W source (UVP, Inc, San Gabriel, CA). A total of 50 mL of loading buffer was added, and proteins were separated on a 6% polyacrylamide-SDS gel.

RESULTS

Characterization of the chemosensitivity projle to DNR of leukemic and n o m 1 myeloid clonogenic cells. Blast cells of the 12 patients samples were exposed to DNR at various concentrations (3, 6, 12, 25, 50, 100, and 200 ngl mL). Linear regression was calculated for each dose-effect curve (individual P values were always lower than .OS). Individual Dso values for PE1 and PES are listed in Table 2. Mean DsO values were 7.0 -t 1.5 ng/mL and 7.2 -t 1.5 ng/mL for PE1 and PES, respectively. In five different experiments performed on normal BM, we found that DNR induced a dose-effect toxicity on normal CFU-GM with mean DsO values of 1.8 -t 0.5 ng/mL and 1.9 ? 0.4 nglmL for day 7 and day 14, respectively (Table 3). Figure 1 shows the DNR dose-effect on PE1 (12 samples), PES (1 1 samples), and day- 7 CFU-GM (5 samples) recovery. DNR had the same effects on PE1 and PES, as determined by Wilcoxon test. CFU-GM were significantly more sensitive to DNR than were AML- CFU (P < .05).

Table 2. Individual DNR Sensitivity of AML-CFU of 12 Patients

DNR D50 (ngIrnLl

Patients PE1 PES

Group A 1 15.6 15.1 2 15.6 1 3 11 11.1 4 9.1 4 5 9 7 6 1 10.5 Mean 2 SEM 11.3 2 1.4 10.2 2 2.1

Group B 1 3.9 5.2 8 3.6 nd 9 2.8 3.9 10 2.6 11

5.1 1 .a 0.8

12 1.4 3 Mean 2 SEM 2.7 2 0.4 3.6 2 0.8

232 LAREDO ET AL

Table 3. Individual DNR Sensitivity of CFU-GM in the Presence or Absence of ST on Five Healthy Donors’ Bone Marrows

DNR D50 IngimL)

J7 J14

Normal BM Samples C ST C ST

A 1.6 1.1 1.8 2.5 B 3.6 2 1.2 1.5 C 1.8 3.5 3.4 3.3 D 0.4 0.7 1.7 1.8 E 1.8 1.4 1.6 1.1 Mean 2 SEM 1.8 2 0.5 1.7 2 0.5 1.9 2 0.4 2 2 0.4

Abbreviations: J7, CFU-GM after 7 days of culture; J14, CFU-GM after 14 days of culture; C, control; ST, ST used at D,,,.

According to the CFU-GM chemosensitivity to DNR, two groups of AML patients were considered, group A and group B (Table 2), the former being defined as samples with Dso values higher than 4 ng/mL (mean CFU-GM DsO + 2 stan- dard deviations). Subsequently, group A and B were termed resistant and sensitive, respectively. D50 values ranged from 7.0 to 15.6 ng/mL (mean, 11.3 t 1.4 ng/mL) for group A, and from 1.4 to 3.9 n g / d (mean, 2.7 ? 0.4 ng/mL) for group B. The difference between group A and B DsO values was highly significant ( P < .Ol), whereas no significant difference was found between group B and CFU-GM. The DNR dose effect on each group is shown in Fig 2.

DNR sensitivity of the whole AML cell population. To characterize the DNR chemosensitivity profile of total AML cell population (which is mainly composed of blast cells in terminal division), we performed viability assays (trypan

0 1 0 2 0 3 0 4 0 5 0

DNR (ng/ml)

Fig 1. DNR dose-effect cytotoxicity on normal and leukemic progenitor colony formation is shown. (01, CFU-GM J7; (m), AML-CFU PE1; and (01, AML-CFU PES. Bars are SEM.

100 n - 2 +.. C

10 rt 0

0.1 I- O 1 0 2 0 3 0 4 0 5 0

DNR (ng/ml)

Fig 2. DNR dose-effect cytotoxicity on normal and leukemic (group A and B) progenitor colony formation is shown. (01, CFU-GM 57; (A), group A PE1; and (0). group B PE1. Bars are SEM.

blue exclusion) on AML cells after 7 days in liquid culture. For this study, we selected five samples from each group. In both groups, DNR induced a semilogarithmic dose-effect toxicity on bulk AML cells (data not shown). As shown in Table 4, AML cells of group A displayed a resistant profile with mean DSO value of 13.3 t- 3.2 ng/mL versus 2.7 -t 0.4 ng/mL for group B ( P < .01).

Expression of P-gp in AML cells. To further explore the possible implication of P-gp in the resistance of AML cells to DNR, the expression of P-gp was investigated by Western blot. This study was performed with both group A and group

Table 4. Individual DNR Sensitivity of AML Fresh Cells in the Presence or Absence of ST Measured by V i a b i l i Assay

DNR50 (nalrnL)

Patients C ST F E

Group A 1 2 3 4 5

Mean 2 SEM Group B

7 8 9

10 12 Mean 2 SEM

22.3 9

20 8.3 6.8

13.3 t 3.2

4.2 2.1 1.9 3.1 2.3

2.1 2 0.4

8 5.9 7.2 4 0.5 l

5.1 -C 1.3

15.9 1.2 1.6 9.5 2.2

6.1 2 2.9

(2.78) (1.52) (2.77) (2.07)

I1 3.6)

(0.26) (1.75) (1.18) (0.32) 11.04)

Abbreviations: ST, ST used at D,,,; FE, fold enhancement of DNR cytotoxicity.


150 1 n ? a

600

500

400

300

ZOO

100

0

HEL 1 3 4 5 7 10 12

Fig 3. Expression of P-gp in AML cells detected by Western blot analysis using C219 monoclonal antibody is shown. Results are expressed as percentage of P-gp expression in HEL cells. Insert shows the relative level of P-gp expression in HEL, K562/R7, and K562 cell lines.

B AML cells. P-gp was found in 6 of the 7 samples tested (4 of 4 in group A and 2 of 3 in group B). P-gp expression was found to be highly variable with no correlation between P-gp expression and DNR sensitivity, as shown in Fig 3. When detected, the amount of P-gp was much lower than in drug-selected MDR-established AML cell lines. The most intense signal was obtained from a DNR-resistant (but not the most resistant) sample (patient no. 4). Patient no. 4 displayed a signal comparable with unselected parental HEL cell line, the latter representing 16% of the signal intensity found in R7 (see Fig 3). Negative controls were provided by K562 cells, because neither P-gp nor mdrl RNA expression were detectable using Western blot and reverse tran- scription-polymerase chain reaction analysis, respectively (data not shown).

Evaluation of P-gpfunction on AML cells. The lack of correlation between P-gp expression and sensitivity to DNR prompted us to investigate the function of P-gp in AML cells. This study was performed using an assay based on the ability of P-gp to extrude a fluorescent dye (Rh123) out of the cell." Rh123 efflux kinetics were monitored by flow cytometry. Both resistant (patients no. 1, 2, 3, 4, and 5) and sensitive (patients no. 7, 10, and 12) AML samples were investigated. We found that sensitive cells did not extrude Rh123, whereas, in each case, resistant cells displayed Rh123 efflux even if variations were observed (see Fig 4A). The kinetics of Rh123 efflux were not correlated with the amount of P-gp expression measured by Western blot analysis. For example, cells from patient no. 1 extruded Rh123, whereas cells from patient no. 12 did not; the latter expressed 5.8-fold more P-gp than did the former. The lack of correlation between P-gp expression and Rh123 efflux that we observed in fresh AML cells was also observed in P-gp-expressing AML cell lines. For example, as shown in Fig 4B, K562R7 and unselected HEL showed similar efflux capacity

while showing more than a 10-fold difference in the amount of P-gp (see above).

Furthermore, verapamil (15 pnoVL) blocked more than 80% of Rh123 efflux in patients no. 1, 3, and 4 (data not shown) and also blocked Rh123 efflux in K562/R7 and HEL cell lines (Fig 4B), proving that P-gp is directly involved in this process.

Effect of ST on DNR cytotoxicity. We investigated the effect of ST on the DNR chemosensitivity profile of leukemic progenitors (PE1; see Table 5) , total AML cell population (see Table 4), and normal CFU-GM (see Table 3). ST used at subtoxic concentration (IClo), was found to enhance DNR cytotoxicity on group A AML cells (both progenitors

CU- I \ F ..

= 20 U F 0 3 0 6 0 9 0 120 A Minutes

E 100 .I

F c U

- U

B 0 ! ' i ' 1 ' i ' i ' i ~ i

0 10 20 30 4 0 5 0 60 Minutes

Fig 4. Rh123 efflux kinetics in group A and B AML fresh cells and in AML cell lines expressed as percentage of "loading" value. (A) Fresh AML cells: (X), patient no. 1; (m), patient no. 2; (01, patient no. 3; (+l, patient no. 4; (A), patient no. 5; (A), patient no. 7; (0). patient no. 10; and (01, patient no. 12. (B) Leukemic cell lines in the absence (open symbols) or presence (closed symbols) of verapamil: (A), K562; (01, K562/R7; and (0). HEL.

234 LAREDO ET AL

Table 5. Individual DNR Sensitivity of AML-CFU (PE11 in the Presence or Absence of ST

DNR D50 (nghnL)

Patients C ST FE 200 -I Group A

l 2 3 4 5

Mean 2 SEM Group B 7 8 9 10 12 Mean 2 SEM

15.6 15.6 11 9.7 9

12.2 2 1.4

3.9 3.6 2.8 2.6 1.4

2.8 2 0.4

7 2.8 0.9 6.5 3

4.0 2 1.1

3.8 7 2.8 7.8 1.4

4.5 -c 1.2

a, 0 C Q) 0 v) Q)

Abbreviation: ST, ST used at Dlo.

and total population). Indeed, mean D50 values showed a threefold decrease ( P = .009) for both PE1 and viability assays, The magnitude of enhancement was found to be highly variable, ranging from 1.5 to 12.2 for PEl, and did not correlate with resistance levels. Interestingly, ST enhanced DNR cytotoxicity neither on group B AML cells nor on CFU-GM. Furthermore, we found that ST used at IC10 (35 nmol/L) in clonogenic assays partially reversed DNR resistance in K562/R7 cells (2.6-fold enhancement of DNR cytotoxicity) but had no effect on the sensitivity to DNR of K562 cells (data not shown).

Effect of ST on Rh123 accumulation. We evaluated the modification of Rh123 accumulation in the presence of ST in AML cells. Fresh leukemic cells were incubated with various doses (10, 35, and 100 nmol/L) of ST for 1 hour; then Rh 123 was added for 10 minutes, and cells were washed before analysis. For this study, we selected patients no. I and 3 of group A, and patient no. 10 of group B. K562/R7 and K562 cell lines were also used for comparison. No effect was observed for concentrations lower than 10 nmol/L. At 35 nmol/L ST, a 1 .S-foId increase of Rh1 23 accumulation was noted for patients no. 1 and 2 but not for patient no. IO (as shown in Fig 5). No greater effect was observed at 100 nmol/L. Furthermore, at the same concentrations, a twofold increase in Rh123 accumulation was observed in K562/R7 cell line but not in K562.

The kinetics of Rh123 efflux was also evaluated in the presence of ST for patients no. 1, 3, and 10. We found that ST induced a partial inhibition of Rh123 efflux for resistant patients no. 1 and 2 (32% and 38% inhibition, respectively) but had no effect for sensitive patient no. I O (data not shown).

Effect of ST on azidoprazosin binding to P-gp. Because we showed that ST may affect the function of P-gp, we tested the ability of ST to displace the P-gp probe '"I-azidoprazosin in K562/R7 cells. We could not detect a significant displacement of the radiolabeled probe by ST at concentrations below 1 pmol/L (results at 100 nmol/L are represented

r K

n v) .- c. 5 150

50

0 R 7 K562 1 3 1 0

Fig 5. Effect of ST on Rh123 accumulation on AML fresh cells and on AML cell lines. Cells were incubated for 1 hour with ( (m) 35 nmol/LST; then Rh123 was added for 10 minutes. Cells were washed, and Rh123 accumulation was evaluated by flow cytometry. For each cell sample, control value was arbitrarily fixed at 100%.

in Fig 6). As positive controls, 100 pmol/L cyclosporine A, cyclosporine SDZ PSC 833, and verapamil were used.

Effect of ST on P-gp phosphorylation. Because ST was able to decrease P-gp function without binding to P-gp, we investigated a possible impact of ST on P-gp phosphorylation. As shown in Fig 7, we found that ST significantly inhibited P-gp phosphorylation in K562/R7 cell line. This inhibition was dose-dependent because P-gp phosphorylation was 80% and 59% of control when K562/R7 cells were treated with 35 nmol/L and 100 nmol/L of ST, respectively (values determined by gel scanning).

The same experiments were performed using AML cells of patients no. I , 3, and 4. Despite a 32P incorporation into cell proteins comparable with K562/R7 cells (about 4500 c p d p g of protein), we could not identify phosphorylated P- gp in fresh AML cells, whether or not blast cells were incubated with ST, even after 2 weeks of exposure.

1 2

"l 3 4 5 6 7

Fig 6. Effect of ST on azidoprazosin binding to P-gp in K562/R7 cell line is shown. Left panel: Western blot analysis with C219 on K562 (lane 1) and K562/R7 (lane 21. Right panel: '*%azidoprazosin labeling on K562/R7 (lane 3) in the presence of 100 pmol/L SDZ PSC 833 (lane 41, 100 pmol/L cyclosporine A (lane 5). l 0 0 pmol/L verapamil (lane 6). and 100 nmol/L ST (lane 71.


1 2 3

Fig 7. Effect of ST on P-gp phosphorylation in K562/R7 cell line is shown. K562/R7 cells were labeled with 32P for 3 hours; then, ST was addedfor 1 hour. Immunoprecipitation with C219 and SDS-polyacrylamide gel electrophoresis were performed as described in Materials and Methods. Autoradiogram of the dried gel shows P-gp phosphorylation after treatment with 0 (lane l), 35 nmol/L (lane 21, and l00 nmollL (lane 3) of ST.

DISCUSSION

AML is a neoplastic disorder characterized by both the proliferation and the accumulation of leukemic blast cells of myeloid origin that can not enter normal granulo-monocytic differentiation?" The AML cell population consists of hierar- chical subpopulations including a minority of proliferating leukemic progenitors (AML-CFU). These AML-CFU may either self-renew or proceed to limited differentiation. The latter pathway provides the blast cells in terminal division, which represent the vast majority of the malignant cell population in AML patients (bulk leukemic cells)." Therefore, the aim in AML chemotherapy is to eradicate the bulk leukemic population but more importantly to kill the proliferating and self-renewing leukemic progenitors.

PE1 and PES are two assays that were designed to evaluate AML-CFU proliferation and self-renewal, respectively." Therefore, it is important to note that, in the present study, DNR induced similar cytotoxic effects in both PE1 and PES. These results are contrary to those described for another anthracycline, doxorubicin (DOX), which displayed greater cytotoxicity in PE1 .22 On the other hand, the anthracenedione derivative, mitoxantrone (MIT), has also been found to be highly effective in reducing both leukemic blast progenitor cell growth and leukemic cell self-renewal capacityF7 Simi- larly, in a recent study, we found etoposide (VP-16) to be equally effective against the self-renewal capacity of AML- CFU." These observations may explain the potent antileukemic effect of DNR, MIT, and VP-16 in AML therapy.

In this study, AML-CFU were much less sensitive to DNR than were normal CFU-GM. This was not unexpected because it has previously been shown that DNR, and also MIT, are much more effective on CFU-GM than on AML-CFU.25 Interestingly, VP-16, a potent topoisomerase I1 inhibitor, is

equally toxic on both AML-CFU and CFU-GM:4 which suggests that AML-CFU resistance to DNR may not involve a topoisomerase 11-mediated mechanism.

The individual analysis of DNR cytotoxicity in clonogenic assays allowed us to characterize two distinct groups of AML, group B (patients no. 7 through 12), which showed a sensitivity to DNR similar to that of normal CFU-GM, and group A (patients no. 1 through 6). which was more resistant, although some individual variations were found within each group. Viability assays, when available, validated these findings, suggesting that the DNR sensitivity profile of AML cells is maintained throughout AML-CFU differentiation.

To correlate the individual sensitivity of AML cells to P- g p expression and/or function, we found that, although most AML samples express P-gp, this expression was variable and generally modest compared with that for drug-selected MDR cell lines. No correlation between P-gp levels and chemosensitivity DNR was found. Such findings raise the possibility that DNR resistance in AML cells may be related to non-P-gp drug resistance-associated proteins such as multidrug resistance associated protein (MRP)," p95," p190,'8 or p1 the latter being expressed in high-risk AML cells.'" However, H69/AR MRP-expressing cells did not extrude Rh123 (data not shown); furthermore, as far it has been reported, there is no evidence that p95-, p1 lo-, and pl90-expressing cells display Rh123 efflux capacity. Therefore, the correlation we found between DNR sensitivity and the kinetics of Rh123 efflux strongly suggests a major role for P-gp in DNR resistance in group B patients. Al- though these observations certainly appear to implicate P- gp in AML cell susceptibility to DNR, they also raise the question of why does P-gp activity fluctuate between different AML cell samples. Among the several possibilities, one cannot rule out that certain AML cells may express mutated forms of P-gp. However, because such mutations appear to be rare events, this hypothesis seems unlikely." Posttransla- tional modifications of P-gp may also occur in AML cells. Indeed, it has already been established that the phosphorylation level of P-gp is a critical event that may profoundly affect its drug-extruding function? To test this latter hypothesis, we evaluated the impact of ST, a potent although nonspecific PKC inhibitor, on the cytotoxicity of DNR on AML cells.

When used at subtoxic concentrations, ST enhanced the cytotoxicity of DNR in DNR-resistant AML cells but did not modify the chemosensitivity of either CFU-GM- or DNR- sensitive AML cells. ST used at the nanomolar range significantly increased Rh123 accumulation. This increase in Rh123 accumulation was found to be similar in fresh AML cells and in the drug-selected MDR cell lines. Although it has previously been reported that ST can sensitize MDR cells and, therefore, increase the intracellular accumulation of cytotoxic drugs,5 it is still uncertain as to whether ST affects MDR indirectly through inhibition of PKC or whether it is a direct result of ST binding to the P-gp?'

To address this issue, we performed photoaffinity-labeling studies on AML cells. Because this technique is highly cell consuming, we did not use fresh leukemic cells but the K562/ R7 cell line. These P-gp-expressing cells were sensitized to

236 LAREDO ET AL

DNR by ST at the nanomolar range. These studies showed that verapamil (100 pmoVL), cyclosporine A (100 pmoVL), and SDZ PSC 833 (100 pmoVL), used as controls, displaced the P-gp probe, azidoprazosin, whereas ST did not present such an effect below 1 pmoVL (this concentration being over 10-fold higher that that required for modulation of P- gp function in K562/R7). Similar results have been described by others in murine cell lines.”,33 With these results under consideration, it would appear that, in AML cells with active P-gp, ST may act through downregulation of the P-gp-mediated DNR efflux without direct interaction with P-gp. This observation strongly supports the hypothesis that ST acts through inhibition of PKC with subsequent dephosphoryla- tion of P-gp, thereby decreasing P-gp-mediated drug efflux. The possibility of downregulation of mdrl gene expression is unlikely, because, at least in K562/R7 cells, we were unable to detect any modification of P-gp expression after several days exposure to ST (data not shown).

In this study, we confirmed that ST could inhibit P-gp phosphorylation in the K562/R7 cell line. This result is con- cordant with other MDR cell lines, HL60NCR34 and KB- Vl.4 However, under the same experimental conditions, we were unable to detect a phosphorylated P-gp signal in AML cells, in the presence or in the absence of ST. This could possibly be because of the low expression of P-gp in these cells compared with that in K562R7 cells, or because of a lack of P-gp-phosphorylating kinase activity such as the a and ,B isoforms of PKC,35.36 even if these PKC isoforms are commonly expressed in AML cells.37

In the perspective of developing new antileukemic thera- pies, metabolic signal transducers such as modulators of protein kinases, have recently emerged as one of the most prom- ising approaches in drug development. We have previously reported that ST induced a dose-dependent growth inhibition of AML progenitors, which are more sensitive than normal CFU-GM.’? This study proposed that the combination of DNR and modulators of protein kinases may offer an inter- esting potential for increasing both efficacy and specificity of AML therapy. Further studies to test this hypothesis using more specific PKC modulators3* and other active cytotoxic drugs are currently underway in our laboratory.

In conclusion, our study provides evidence that the chemosensitivity profile of AML cells to DNR may be, at least in part, a function of the P-gp-mediated outward process. Furthermore, the ability of P-gp to efflux DNR can be downregulated by protein kinase inhibitors such as ST. This result suggests that the MDR phenotype can be functionally regulated by cellular signalization in AML cells and that PKC modulators might be useful in combination with cytotoxic drugs for improving AML therapy.

ACKNOWLEDGMENT

We would like to thank George E. Duran (Stanford University, Stanford, CA) and Dr C. Savy (Sanofi-Recherche, Toulouse) for technical assistance.

REFERENCES 1. Bradley G, Juranka P, Ling V: Mechanism of multidrug resis-

tance. Biochim Biophys Acta 948237, 1988

2. Baer MR, Bloomfield CD: Multidrug resistance in acute myeloid leukemia. J Natl Cancer Inst 83:663, 1991

3 . Hamada H, Hagiwara KI, Nakajima T, Tsuruo T: Phosphoryla- tion of the Mr 170,000 to 180,000 glycoprotein specific to multidrug- resistant cells: Effect of verapamil, trifluoroperazine, and phorbol esters. Cancer Res 47:2860, 1987

4. Chambers T, McAvoy E, Jacobs J, Eilon G: Protein kinase C phosphorylates P-glycoprotein in multidrug resistant human KB carcinoma cells. J Biol Chem 265:7679, 1990 S. Sato W, Yusa K, Naito M, Tsuruo T: Staurosporine, a potent

inhibitor of C-kinase, enhances drug accumulation in multidrug- resistant cells. Biochem Biophys Res Commun 173:1252, 1990

6. Hagiwara M, Wakusawa S, Miyamoto IU, Hidaka H: Obviation of drug resistance and affinity purification of P-glycoprotein by iso- quinolinesulfonamides. Cancer Lett 60: 103, I991

7. O’Brian C, Ward N, Liskamp R, de Bont D, Earnest L, van Boom J, Fran D: A novel N-myristylated synthetic octapeptide inhib- its protein kinase C activity and partially reverses murine fibrosar- coma cell resistance to adriamycin. Invest New Drugs 9:169, 1991

8. JaffrCzou JP, Herbert JM, Levade T, Gau MN, Chatelain P, Laurent G: Reversal of multidrug resistance by calcium channel blocker SR33557 without photoaffinity labeling of P-glycoprotein. J Biol Chem 266:19858, 1991

9. Fine RL, Patel J, Chabner B: Phorbol esters induce multidrug resistance in human breast cancer cells. Proc Natl Acad Sci USA 85582, 1988

IO. Dong 2, Ward NE, Dan D, Gupta KP, O’Brian CA: In vitro model for intrinsic drug resistance: Effects of protein kinase C activators on the chemosensitivity of cultured human colon cancer cells. Mol Pharmacol 39563, 1991

I I . Posada J, Vichi P, Tritton T: Protein kinase C in adriamycin action and resistance in mouse sarcoma 180 cells. Cancer Res 49:6634, 1989

12. Georges E, Bradley G, Gariepy J, Ling V: Detection of P- glycoprotein isoforms by gene specific monoclonal antibodies. Proc Natl Acad Sci USA 87:152, 1990

13. Laredo J, Demur C, Muller C, Saivin S, Cassar G, Bousquet C, Dastugue N, Colombies P, Laurent G: Effects of H-7 and staurosporine on proliferation and self-renewal of acute myeloid leukemia progenitors. Leukemia 7:813, 1993

14. Miyauchi J, Kelleher CA, Minkin S, McCulloch EA: Growth factors influence the sensitivity of leukemic stem cells to cytosine arabinoside in culture. Blood 73:1272, 1989

IS. Peterson CL: A simplification of the protein assay method of Lowry et a1 which is generally more applicable. Anna1 Biochem 83:346, 1977

16. Fairbanks G, Steck TL, Wallach DHF Electrophoretic analysis of the major polypeptides of the human erythrocyte membrane. Biochemistry 10:2606, 1971

17. Towbin H, Staehelin T, Gordon J: Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: proce- dures and some applications. Proc Natl Acad Sci USA 76:4350, 1979

18. Chaudhary PM, Roninson IB: Expression and activity of P- glycoprotein, a multidrug efflux pump, in human hematopoietic stem cells. Cell 66:8S, l991

19. Greenberger LM, Yang CPH, Gindin E, Horwitz SB: Photoaf- finity probes for the aI-adrenergic receptor and the calcium channel bind to a common domain in P-glycoprotein. J Biol Chem 265:4394. I990

20. Metcalf D: The role of stem cell self-renewal and autocrine growth factor production in the biology of myeloid leukemia. Cancer Res 49:2305, 1989

21. McCulloch EA: Regulatory mechanisms affecting the blast stem cell of acute myeloblastic leukemia. J Cell Physiol4:27, 1986


22. Nara N, Suzuki T, Nagata K, Yamashita Y, Murohashi J, Adachi Y: Relationship between the in vitro sensitivity to cytosine arabinoside of blast progenitors and the outcome of treatment in acute myeloblastic leukemia patients. Br J Haematol70187, 1988

23. Grant S, Arlin Z , Gewirtz D, Feldman E Effect of pharmaco- logically relevant concentrations of mitoxantrone on the in vitro growth of leukemic blast progenitors. Leukemia 5:336, 1991

24. Chiron M, Demur C, Pierson V, Jaffkzou JP, Muller C, Sai- vin S , Bordier C, Bousquet C, Laurent G: Sensitivity of fresh acute myeloid leukemia cells to etoposide relationship with cell growth characteristics and DNA cleavage. Blood 80:209, 1992

25. Scholzel C, Van Putten W, Uwenberg B: A comparison of in vitro sensitivity of acute myeloid leukemia precursors to mitoxantrone, 4’deoxydoxorubicin, 4-demethoxydaunorubicin and daunorubicin. Leuk Res 10:1455, 1986

26. Cole SPC, Bhardwaj G, Gerlach JH, Mackie JE, Grant CE, Almquist KC, Stewart A J , Kurz EU, Duncan AMV, Deeley RG: Overexpression of a transporter gene in a multidrug-resistant human lung cancer cell line. Science 258:1650, 1992

27. Chen YN, Mickley LM, Schwartz AM, Acton EM, Hwang J, Fojo A T Characterization of adriamycin-resistant human breast cancer cells which display overexpression of a novel resistant-related membrane protein. J Biol Chem 265:10073, 1990

28. McGrath T, Center M: Adriamycin resistance in HL-60 cells in the absence of detectable P-glycoprotein. Biochem Biophys Res Commun 145:1171, 1987

29. Scheper RJ, Broxterman HJ, Scheffer GL, Kaaijk P, Dalton WS, van Heijningen THM, van Kalken CK, Slovak ML, de Vries EGE, van der Valk P, Meijer CJLM, Pinedo H: Overexpression of a Mr 110,000 vesicular protein in non-P-glycoprotein-mediated multidrug resistance. Cancer Res 53:1475, 1993

30. List AF, Spier CC, Abbaszadegan M, Grogan TM, Greer JP, Wolf SN, Scheper RJ, Dalton WS: Non-P-glycoprotein (PGP) mediated multidrug resistance (MDR): Identification of a novel drug resistance phenotype with prognostic relevance in acute myeloid leukemia (AML). Blood 82:443a, 1993 (abstr)

31. Holmes JA, Whittaker JA, Padua R Effect of position 185 mutations of the MDRl gene on drug resistance in leukemia. Leuke- mia 6:484, 1992

32. Miyamoto K, Inoko K, Wakusawa S , Kajita S, Hasegawa T, Takagi K, Koyama M: Inhibition of multidrug resistance by a new staurosporine derivative, NA-382, in vitro and in vivo. Cancer Res 53:1555, 1993

33. Wakusawa S , Inoko K, Miyamoto K Staurosporine derivatives reverse multidrug resistance without correlation with their protein kinase inhibitory activity. J Antibiot 46:353, 1993

34. Ma L, Marquardt D, Takemoto L, Center M: Analysis of P-gp phosphorylation in HL60 cells isolated for resistance to vincristine. J Biol Chem 266:5593, 1991

35. Gollapudi S , Pate1 K, Gupta S : Protein kinase C isoforms in multidrug resistant P388/ADR cells: a possible role in daunorubicin transport. Cancer Lett 62:69, 1992

36. Blobe G, Sachs C, Khan W, Fabbro D, Stabel S , Wetsel W, Obeid L, Fine R, Hannun Y: Selective regulation of expression of protein kinase C (PKC) isoenzymes in multidrug-resistant MCF-7 cells: Functional significance of enhanced expression of PKC a. J Biol Chem 268:658, 1993

37. Komada F, Nishikawa M, Uemura Y, Morita K, Hidaka H, Shirakawa S: Expression of three major protein kinase C isoenzymes in various types of human leukemic cells. Cancer Res 51:4271, 1991

38. Herbert JM, Augereau JM, Gleye J, Maffrand J P Cheler- ythrine is a potent and specific inhibitor of protein kinase C. Biochem Biophys Res Commun 172:993, 1990

Download - Effect of the protein kinase C inhibitor staurosporine on chemosensitivity to daunorubicin of normal and leukemic fresh myeloid cells

Top Related