Alteration in the cellular response to retinoic acid of a human acute promyelocytic leukemia cell...

Leukemia Research 28 (2004) 959–967

Alteration in the cellular response to retinoic acid of a humanacute promyelocytic leukemia cell line, UF-1, carryinga patient-derived mutant PML-RAR� chimeric gene

Atsushi Satoa, Masue Imaizumia,b,∗, Yoshiyuki Hoshia, Takeshi Rikiishia, Kunihiro Fujiia,Masahiro Kizakic, Hiroyuki Kagechikad, Akira Kakizukae, Yutaka Hayashib, Kazuie Iinumaa

a Department of Pediatrics, Tohoku University School of Medicine, 1-1 Seiryo-machi, Aoba-ku, Sendai 980-8574, Japanb Department of Pediatric Hematology and Oncology, Tohoku University School of Medicine, 1-1 Seiryo-machi, Aoba-ku, Sendai 980-8574, Japan

c Division of Hematology, Keio University School of Medicine, Tokyo, Japand Faculty of Pharmaceutical Sciences, Tokyo University, Tokyo, Japan

e Department of Functional Biology, Kyoto University Graduate School of Biostudies, Kyoto, Japan

Received 20 August 2003; accepted 31 December 2003

Available online 17 April 2004

Abstract

Cellular response to all-trans retinoic acid (ATRA) of acute promyelocytic leukemia (APL) with patient-derived mutant PML-retinoicacid receptor-� (PML-RAR�) was investigated using an APL cell line, UF-1, carrying Arg611Trp mutation in PML-RAR�. Although themutant protein showed a decreased ligand-dependent transcriptional activity and retained a dominant-negative effect on normal RAR�,UF-1 cells underwent growth inhibition, maturation and apoptosis in response to ATRA at 1�M, but not≤100 nM, after 4 days of treatmentwith ATRA. Moreover, in the presence of 1�M ATRA, approximately 50% of UF-1 cells expressing annexin V, an early-apoptotic marker,was negative for CD11b and showed immature morphology. These findings suggest that UF-1 cells, despite expressing mutant PML-RAR�

protein, can be induced by ATRA to undergo differentiation and apoptosis through RA-inducible mechanism(s), in which a proportionof apoptosis may occur independent of terminal differentiation. This unique cell line may be useful for investigating the pathogenesis ofATRA resistance and the mechanism of ATRA-induced apoptosis in APL.© 2004 Elsevier Ltd. All rights reserved.

Keywords: Acute promyelocytic leukemia; Retinoic acid; Apoptosis; Differentiation; Mutant PML-RAR�

1. Introduction

Acute promyelocytic leukemia (APL) is characterized bya blockade of myeloid maturation at the promyelocyte stageand by a 15;17 chromosomal translocation generating thePML-retinoic acid receptor-� (PML-RAR�) chimeric pro-tein, which inhibits normal RAR� transcriptional activityin a dominant-negative fashion[1]. Differentiation of APLcells is induced by treatment all-trans retinoic acid (ATRA)at pharmacological doses, and, under such conditions, PML-RAR� can respond to ATRA and show a ligand-dependenttranscriptional activity as well as a decreased dominant-

Abbreviations: APL, acute promyelocytic leukemia; PML-RAR�,PML-retinoic acid receptor-�; ATRA, all-trans retinoic acid; 9CRA,9-cis-retinoic acid; RXR, retinoid X receptor

∗ Corresponding author. Tel.:+81-22-717-7287; fax:+81-22-717-7290.E-mail address: [email protected] (M. Imaizumi).

negative effect on normal RAR� [2]. Although ATRA ther-apy is highly effective in inducing complete remission ofpatients with APL[3,4], some patients who become ATRA-resistant during relapse have a poor prognosis[3,5,6]. Re-cently, however, it has been shown that arsenic trioxide caninduce complete remission in the majority of these relapsedpatients[7,8].

ATRA resistance may be caused by increased oxidativecatabolism of ATRA[9–11], or by mutations in the RAR�portion of the chimeric PML-RAR� gene[12–15]. Analysisof APL patients with ATRA resistance revealed three clus-ters of missense mutations in the PML-RAR� chimeric gene.These mutations are located in the RAR� portion of thechimeric protein that functions in ligand binding, receptordimerization and ligand-dependent transcriptional activity[16]. In vitro transfection experiments revealed that the dif-ferent mutations identified in patients with ATRA resistancehad different effects on the RA-dependent transcription

0145-2126/$ – see front matter © 2004 Elsevier Ltd. All rights reserved.doi:10.1016/j.leukres.2003.12.017

960 A. Sato et al. / Leukemia Research 28 (2004) 959–967

process, suggesting that diverse molecular pathogenesescontribute to ATRA resistance due to mutant PML-RAR�[15]. Experiments with primary cultures of APL cellsfrom patients with ATRA resistance showed that ATRAsensitivity varied with different mutations in PML-RAR�[14,17]. Although experimentally derived ATRA-resistantHL-60 or NB4 subclones have either complete loss of ora maturation defect in the response to RA[12,18–20], itis not known how APL cells derived from patients withATRA resistance would respond to ATRA at the cellularlevel.

During ATRA therapy, APL cells are induced to enter theterminal stage of differentiation and are then cleared fromthe blood[21]. This process may be mediated by apoptosisbecause aged mature neutrophils are eliminated via apop-tosis [22] and APL cells undergo apoptosis in response toATRA therapy in vivo[21,23]. It has been suggested thatrestoration of PML function and normal RAR signaling maybe involved in the mechanism(s) by which APL cells arestimulated to undergo apoptosis[20]. Thus, it is thought, butnot yet proven, that ATRA induces apoptosis of APL cellsindependent of or prior to terminal differentiation.

Recently, Kizaki et al.[24] established an APL cell line,UF-1, from a patient who became ATRA resistant at relapse.This cell line expresses a short form of the PML-RAR�protein with a missense mutation (Arg611Trp). This muta-tion impairs ATRA binding and reduces ATRA-sensitivityin vitro [25]. In the present study, we examined RA-induceddifferentiation and apoptosis of UF-1 cells and comparedthe cellular responses of UF-1 to RA with the molecular re-sponse of the mutant PML-RAR� protein. Our findings sug-gest that regulation of RA signaling other than that throughmutant PML-RAR� may be involved in the cellular responseof UF-1 to RA, and that ATRA may induce APL cells to un-dergo apoptosis not only as consequence of differentiationbut also prior to terminal differentiation.

2. Materials and methods

2.1. Cell line

The UF-1 cell line was established and characterized byKizaki et al.[24]. Cells were maintained in suspension cul-ture in RPMI 1640 medium (Gibco-BRL, Grand Island, NY)with 10% fetal bovine serum (FBS) (Hyclone LaboratoriesInc., Logan, UT) at 37◦C in a humidified atmosphere of 5%CO2 in air. Cell numbers and viability were assessed witha hemocytometer and by trypan blue dye-exclusion test, re-spectively.

2.2. Retinoids

ATRA and 9-cis-retinoic acid (9CRA) were purchasedfrom Sigma Chemical Co. (St. Louis, MO). HX600, asynthetic retinoid that acts as an retinoid X receptor

(RXR)-specific agonist, was provided by Dr. H. Kagechika[26]. Stock solutions (1 mM) of retinoids in ethanol werediluted serially in ethanol and then added to cell cultures atthe final concentrations indicated. The final concentrationof ethanol in each culture was less than 0.1%.

2.3. Morphologic evaluation of apoptosis

Cell morphology was examined by light microscopy oncytospin slides stained with Wright-Giemsa. More than 200cells for each sample were scored and evaluated for morpho-logic features of apoptosis, including nuclear condensationand fragmentation.

2.4. Nitroblue tetrazolium (NBT) reduction assay

The ability of cells to reduce NBT was assessed as pre-viously described[27]. Briefly, 106 viable cells were sus-pended in 1 ml phosphate-buffered saline (PBS) containing0.5 mg/ml NBT (Sigma) and 162 nM 12-O-tetradecanoyl-phorbol-13-acetate (TPA) (Sigma). After incubation at 37◦Cfor 25 min, the percentage of formosan-positive cells wasdetermined on cytospin slides by Wright-Giemsa staining.With exclusion of dead cells, at least 200 intact cells perspecimen were scored under light microscopy.

2.5. Flow cytometric analysis of CD11b expression

To evaluate cell differentiation, we analyzed expressionof CD11b. Cells were washed twice with PBS and incu-bated with fluorescein isothiocyanate (FITC)-conjugatedanti-CD11b monoclonal antibody (Coulter, Hialeah, FL) orwith FITC-conjugated non-immune antibody as a negativecontrol at 4◦C for 15 min. Cells were washed with PBSand analyzed on a FACScan (Becton Dickinson, Moun-tain View, CA) with Cell Quest software (Becton Dickin-son).

2.6. Flow cytometric analysis of early apoptotic cells

To detect cells in the early phase of apoptosis, we analyzedexpression of annexin V by flow cytometry. This method re-quires no permeabilization of the cell membrane and, there-fore, is suitable for investigating the kinetics of differentia-tion and apoptosis at the single-cell level. After incubationwith phycoerythrin (PE)-conjugated anti-CD11b antibody(Coulter) at 4◦C for 15 min, cells were washed with PBSand stained with FITC-conjugated annexin V and propid-ium iodide (PI) according to the manufacturer’s instructions(Genzyme, Cambridge, MA). Flow cytometric analysis wasperformed with a FACScan, or three-color sorting was per-formed with an Epics Elite (Coulter). After gating PI (−)viable cells, annexin V (+) and CD11b (−) cells were an-alyzed and sorted for morphological evaluation. Annexin V(+) and PI (−) cells were considered to be in the early stageof apoptosis[28].

A. Sato et al. / Leukemia Research 28 (2004) 959–967 961

2.7. DNA extraction and agarose gel electrophoresis

DNA was extracted as previously reported but with slightmodification [29]. Briefly, cell pellets were suspended in1 ml TEN solution (10 mM Tris–Cl (pH 7.4), 25 mM EDTA,100 mM NaCl). After addition of 100�l 10% sodium dode-cyl sulfate (SDS), 30�g RNase A (Sigma), and 1 mg pro-teinase K (Sigma), the reactions were incubated overnight at55◦C. DNA was extracted with phenol and precipitated withsodium acetate and isopropanol, and then DNAs (2�g persample) were separated by electrophoresis on 2% agarosegels, and stained with ethidium bromide.

2.8. Transfection assays of ligand-dependenttranscriptional activity of PML-RARα chimeric protein

A full length of cDNA clone of PML-RAR� withArg611Trp mutation was derived from pCMX expressionvector harboring a short form of the wild type PML-RAR�cDNA using a site-directed mutagenesis kit (Stratagene,Cambridge, United Kingdom)[13]. The wild-type andmutant cDNA clones were used in transient transfectionexperiments for evaluating in vitro ligand-dependent tran-scriptional activity. COS-1 cells were seeded in six-wellplates at 2× 105 cells per well in 2 ml DMEM with 10%FBS 1 day prior to transfection. Transfection was performedwith the Lipofectamine Plus kit (Gibco-BRL) as describedpreviously [13]. Briefly, the transfection mixture (200�lOpti-MEM I, 0.4�g of pCMX expression vector encod-ing the short form of wild-type or mutant (Arg611Trp)PML-RAR�, 0.4�g of R140-Luciferase reporter plas-mid, 0.2�g of pCMV-�-galactosidase reporter plasmid,6�l PLUS reagent, and 6�l Lipofectamine reagent) wasoverlaid onto cells and incubated for 6 h at 37◦C. Thetransfection mixture was then replaced with 2 ml DMEMcontaining 10% FBS and ATRA, 9CRA, or HX600 wasadded at various concentrations indicated. After 48 h ofincubation, the cells were lyzed in 200�l Report LysisBuffer (Promega) and centrifuged at 15,000 rpm for 2 min,and then 20�l of supernatant of the cell lysate was mixedwith 100�l luciferase assay reagent (Promega), and lu-ciferase activity was measured with a luminometer (AttoCo., Tokyo, Japan). Values were normalized in referenceto �-gal activity. For the analysis of the dominant-negativeaction of mutant PML-RAR�, pCMX expression vec-tors encoding normal RAR� mixed with that encodingwild-type or mutant-form PML-RAR� chimeric gene, orpCMX vector alone (total 0.4�g with a weight ratio of 1:3,respectively) were transfected into COS-1 cells with sameprotocol.

2.9. Statistical analysis

Student’st-test was used to determine statistical signifi-cance. Probability (P) values lower than 0.05 were consid-ered significant.

3. Results

3.1. Time course of cell growth, viability, and NBTreduction of UF-1 cells induced by ATRA or 9CRA

Growth of UF-1 cells was slow (doubling time: 96 h), andproliferation was inhibited in the presence of 1�M, but not≤100 nM, ATRA after 4 days of culture (Fig. 1A), whereasgrowth inhibition was observed in the presence of 9CRA at1�M and 100 nM after 4 days of culture (Fig. 1B). Viabilityof UF-1 cells was markedly decreased after 6 days of cul-ture in the presence of 1�M ATRA or 9CRA (Fig. 1C andD). Viability of UF-1 remained above 90% in the presenceof ATRA at doses below 100 nM, whereas cell viability de-creased slightly in the presence of 100 nM 9CRA (Fig. 1Cand D).

3.2. Morphology and NBT reduction of UF-1 in thepresence of ATRA or 9CRA

Morphological analysis of UF-1 cells treated with 1�MATRA for 8 days revealed myeloid maturation and forma-tion of apoptotic bodies, suggestive of differentiation withapoptosis (Fig. 2A). NBT Reduction by UF-1 cells was notobserved at≤100 nM, but it was increased at 1�M ATRA,while NBT reduction was observed at 1�M and 100 nM of9CRA in a dose dependent manner (Fig. 1E and F). At 8days of culture, the degree of NBT reduction induced by10 nM or 100 nM 9CRA was low but significantly higherthan that induced by 10 nM or 100 nM ATRA (P < 0.01)(Fig. 2B). Similar findings were observed for expression ofCD11b (data not shown).

3.3. Apoptosis of UF-1 cells induced by ATRA and 9CRA

Morphological evaluation revealed that ATRA at 1�M,but not≤100 nM, induced apoptosis of UF-1 cells, whereas9CRA at≥100 nM induced apoptosis in a dose-dependentmanner (Fig. 3). Examination of DNA ladder formation con-firmed that apoptosis occurred in response to 1�M ATRAand≥100 nM 9CRA (Fig. 4).

3.4. Effect of HX600 on ATRA-sensitivity of UF-1 cells

To evaluate the involvement of RXR activation in alteredATRA sensitivity, UF-1 cells were cultured in the presenceof 100 nM ATRA, 1�M HX600, or both. NBT reduction andapoptosis were not observed in response to ATRA or HX600alone. However, in the presence of both compounds, thepercentages of cells showing NBT reduction and apoptosisincreased to 14.1±2.6% and 8.3±2.8%, respectively, whichare levels similar to those induced by 100 nM 9CRA (Fig. 5).

3.5. Ligand-dependent transcriptional activity

As described previously[25], UF-1 cells contain amutation (Arg611Trp) in the RAR� E-domain of the


Via

ble

cell

coun

ts (

x10

/ml)

Culture Period, days

1086420

.1

100

20

40

60

NB

T r

educ

tion

(%

)

Culture Period, daysC

ell v

iabi

lity

(%)

0

20

40

60

80

100

1086420

Cel

l via

bilit

y

0

20

40

10

(%

)

108

NB

T r

educ

tion

(%

)

10864200

20

40

60

80

100


60

80

100

6420

Via

ble

cell

coun

ts (

x10

/ml)

86420.1

1

5

10


5

00

10

80



108642

1

(A) (C) (E)

(B) (D) (F)

Fig. 1. Time course of viable cell number (A, B), cell viability (C, D) and NBT reduction (E, F) for UF-1 cells cultured in the absence (�) or presenceof ATRA (A, C, E) or 9CRA (B, D, F). RA doses were: 10 nM (�), 100 nM (�), and 1�M (�). NBT reduction analysis was performed with exclusionof dead cells on slides. Results are mean± S.D. of three experiments.

Fig. 2. (A) Morphological changes in UF-1 cells cultured for 8 days in the absence or presence of 1�M ATRA. Wright-Giemsa staining of cells (1000×magnification). (B) NBT reduction in UF-1 cells treated with ATRA or 9CRA. After 8 days of culture in the presence of different concentrations ofATRA or 9CRA, UF-1 cells were tested for NBT reduction. Bars aremean ± S.D. of three experiments (∗P < 0.05).


Fig. 3. Apoptotic morphology of UF-1 cells treated with ATRA or 9CRA.After 8 days of culture in the absence or presence of different concen-trations of ATRA or 9CRA, UF-1 cells were examined for morpholog-ical features of apoptosis. Bars are mean± S.D. of three experiments(∗P < 0.05).

PML-RAR� chimeric gene (data not shown). To assessligand-dependent transcriptional activity of the mutant pro-tein in comparison with that of the wild-type, we performedtransient transfection experiments. Increased transcriptionof the reporter constructs was observed when cells ex-pressing wild-type PML-RAR� were treated with variousconcentrations of ATRA or 9CRA alone. Although cellsexpressing mutant PML-RAR� showed increased transcrip-tion in response only to 1�M ATRA, the level reachedonly that of the wild-type PML-RAR� stimulated by 10 nMATRA (Fig. 6). There was no significant difference be-tween ATRA and 9CRA for RA-dependent transcriptionalactivity of the mutant protein, and the combination ofHX600 and ATRA had no effect on the response of mu-tant PML-RAR� to RA (Fig. 6A–C). Furthermore, when

Fig. 4. DNA fragmentation in UF-1 cells treated with ATRA (A) or 9CRA (B). Total DNA was extracted from UF-1 cells cultured for 7 days in the absenceor presence of various concentrations of ATRA or 9CRA and then separated by electrophoresis in 2% agarose gels. Markers are shown in bp in size.

Fig. 5. Apoptosis of and NBT reduction by UF-1 cultured for 7 days inthe absence or presence of 100 nM ATRA or 1�M HX600, or both. Barsindicate mean± S.D. of three experiments.

cells were co-transfected with vectors expressing normalRAR� and PML-RAR�, the mutant protein showed sig-nificantly higher dominant-negative inhibition of RAR� incomparison with that of wild-type PML-RAR� (Fig. 6D).

3.6. Three-color FACS analysis of early apoptoticcells

To detect cells in early apoptosis, the percentages ofdead cells or viable, mature cells were analyzed duringATRA-induced differentiation and apoptosis (Fig. 7). In thepresence of 1�M ATRA, the percentage of CD11b (+), PI(−) viable, mature cells gradually increased over a period of6 days, and then the number of PI (+) dead cells increasedgradually (Fig. 8A). A small fraction (5.0 ± 0.4%) of earlyapoptotic cells with annexin V (+), PI (−) phenotypes were


Fig. 6. Ligand-dependent transcriptional activities of wild-type or mutant PML-RAR� proteins. Luciferase and�-gal reporter constructs were co-transfectedwith expression vector encoding wild-type or mutant PML-RAR� into COS-1 cells, and the cells were cultured in the absence or presence of variousconcentrations of ATRA (A) or 9CRA (B) or the combination of 100 nM ATRA with 1�M HX600 (C). To evaluate the dominant-negative effect ofPML-RAR� on normal RAR� activity, vector expressing normal RAR� was co-transfected with vector expressing wild-type or mutant PML-RAR� (D).Luciferase activity was normalized to�-gal activity and expressed as fold-increase of activities in the absence of retinoids.

detected after 6 days when the cells started to show mor-phologic features consistent with apoptosis. Interestingly,approximately 50% of the annexin V (+), PI (−) cells didnot express CD11b (Fig. 8B). Furthermore, annexin V (+),PI (−), CD11b (−) UF-1 cells sorted after 8 days culture inthe presence of 1�M ATRA showed immature morphology(Fig. 8C).

4. Discussion

In the present study, we showed that UF-1 cells had aunique response to RA at the cellular level. RA sensitivitywas markedly altered in connection with the dose and timeof RA treatment. UF-1 cells were unresponsive to ATRA atdoses≤100 nM, but showed growth inhibition, differentia-tion and apoptosis in response to 1�M ATRA. UF-1 cellsshowed a similar but more sensitive response to 9CRA. In-terestingly, the reduced sensitivity of UF-1 cells to RA wasobvious within 4 days of ATRA treatment, but we foundthat culture with 1�M ATRA for longer periods inducedUF-1 cells to undergo growth inhibition, differentiation andapoptosis.

It is likely that the impaired cellular response to≤100 nMATRA is due to the mutant PML-RAR�, because the UF-1cell line was established from an ATRA-resistant APL pa-tient with the mutation at relapse. Furthermore, the mu-tant PML-RAR� protein showed no transcriptional activityin response to≤100 nM ATRA and lacked ATRA binding[25,30]. However, it is not clear how APL cells with the

mutant PML-RAR� can respond to 1�M ATRA when ex-posed for longer times. To assess this question, we com-pared RA responses of UF-1 cells with those of the mutantPML-RAR� as assessed by RA-dependent transcriptionalactivity in vitro.

In response to 1�M ATRA, ≥100 nM 9CRA, or 1�MHX600 in combination with ATRA, UF-1 cells showedmorphological and functional maturation. By contrast,the extents of in vitro transcriptional activity of mutantPML-RAR� in response to the same conditions were as lowas that of wild-type PML-RAR� in response to 10 nM RAand had no influence by HX600. Therefore, it is suggestedthat there are differences between the cellular responseof UF-1 to RA and the molecular response of mutantPML-RAR�. Furthermore, mutant PML-RAR� had an in-crease in dominant-negative effect on the RA-dependenttranscriptional activity of normal RAR� as compared withthat by wild-type PML-RAR�. Therefore, the response ofUF-1 cells can not be due to the response of the mutantPML-RAR�, suggesting that signaling pathways other thanthat of the mutant protein may alter the UF-1 response to1�M ATRA.

Although the mechanism that underlies APL cell mat-uration remains unclear, it is thought that PML-RAR�causes a differentiation block that can be released by RAR�agonist [31]. In this context, the response of APL cellsexpressing mutant PML-RAR� may be determined primar-ily by the altered response of mutant PML-RAR� to RA.An in vitro transfection study revealed that patient-derivedPML-RAR� carrying distinct mutations have diverse effect


Fig. 7. Flow cytometric detection of undifferentiated early-apoptotic cells. UF-1 cells were cultured for 6 days in the absence (A, B) or presence (C,D) of 1�M ATRA and then stained with PI, annexin V, and anti-CD11b antibody. The proportion of early-apoptotic cells with the annexin V (+), PI(−) phenotype were analyzed among the whole cell population (A, C). After gating PI (−) viable cells, undifferentiated early-apoptotic cells with theannexin V (+), CD11b (−) phenotype were analyzed in the gated populations (B, D).

on RA-dependent transcriptional regulation[15]. It wasreported that the Arg611Trp mutant PML-RAR� does nei-ther release the co-repressor SMRT nor recruit co-activatorACTR even when stimulated with≥1�M ATRA, whereas

Fig. 8. Three-color cytometric analysis of UF-1 cells stained for annexin V, CD11b, and PI. (A) Chronological changes in proportion of PI (+) deadcells (�) or viable CD11b (+), PI (−) differentiating cells (�). Results are mean± S.D. of three expreiments. (B) Among the gated PI (−) viable cells,chromological changes in proportion of annexin V (+) early-apoptosic cells (�) or undifferentiated annexin V (+), CD11b (−) early-apoptotic cells (�)were evaluated. Results are mean± S.D. of three expreiments. (C) Morphology of annexin V (+), CD11b (−) UF-1 cells stained with Wright-Giemsa(1000× magnification).

the mutant PML-RAR� retained the ability to interact withDRIP, a ligand-dependent transcription activators of vitaminD3 receptor, with RA sensitivity similar to that of wild-typePML-RAR� [16,32].


There have been two reports of the cellular response ofprimary cultures of APL cells with distinct mutations inPML-RAR� [14,17]. Such APL cells may be categorizedinto two groups: (1) APL cells with moderately decreasedresponses to 1�M ATRA by day 4 and partially reducedATRA binding and (2) cells with little response to 1�MATRA by day 4 and almost complete loss of ATRA binding.Although UF-1 cells seem to belong to the latter category, itneeds further investigation to know whether patient-derivedAPL cells with features of the latter category would have analtered RA response similar to that of UF-1.

It should be noted that UF-1 cells gradually underwentmaturation with subsequent apoptosis after 4 days of treat-ment with 1�M ATRA. However, it is not known howUF-1 cells would respond to ATRA if UF-1 cells have hadwild-type PML-RAR� protein instead of the mutant. Thistime course of RA response of UF-1 cells is relatively slowif compared with that of HL-60 or NB4 cells, which showa maximal response by day 4[33,34], suggesting that RAsignaling other than that through PML-RAR� may under-lay the delayed response of UF-1 to 1�M ATRA. Recently,it was shown that RAR� is upregulated in response to RAstimulation and is involved in RA-induced apoptosis of tu-mor cells[35]. Thus, activation of RAR� may play a role inthe delayed RA response of UF-1, although further analysesare needed.

The response of APL to RA may be regulated by signalsother than PML-RAR�. Fanelli et al.[36] showed that thesensitivity of APL cells to RA is altered by the proteosomepathway through constitutive degradation of PML-RAR�protein. Moreover, Benoit et al.[37] reported the exis-tence of an RAR�-independent RXR signaling pathways,in which a combination of RXR agonists and protein ki-nase A stimulators induce differentiation of ATRA-sensitiveand ATRA-resistant NB4 cells without degradation ofPML-RAR� protein or reconstitution of PML nuclear bod-ies. Alternatively, post-transcriptional modification throughcovalent binding of RA to cellular proteins, a process called“retinoylation”, may be involved in the RA response ofUF-1 cells[38]. Further studies are needed.

ATRA at pharmacological doses can restore RA signal-ing in APL cells by releasing the dominant-negative effectof PML-RAR�, leading to maturation and subsequent deathof cells via apoptosis. However, hematological findings inpatients who had undergone ATRA therapy suggested thatAPL cells can be induced to undergo apoptosis as a con-sequence of maturation, and independent of differentiation[23]. Interestingly, in the presence of 1�M ATRA, we de-tected a small fraction of viable UF-1 cells that were annexinV-positive, CD11b-negative, and morphologically imma-ture, indicating that a portion of the ATRA-treated APL wasin the early stage of apoptosis. UF-1 cells may be predis-posed to undergo apoptosis independent of differentiation asobserved in response to stimuli involving the STAT signalpathway[39] or prior to terminal differentiation in responseto arsenic trioxide[8]. This separation of apoptosis from

differentiation in RA-treated APL suggest that there maybe a cellular imbalance in the restored functions of proteinssuch as normal RAR� and PML restored by ATRA[20,40].

In conclusion, the UF-1 cell line, which was establishedfrom a patient with ATRA resistance, may be of great valuefor investigating the relation between the cellular and molec-ular responses to ATRA of APL with mutant PML-RAR�chimeric gene. Moreover, this cell line may be useful for elu-cidating the mechanism(s) by which APL cells are inducedinto apoptosis independent of differentiation.

Acknowledgements

This study was supported by grants from the Ministry ofEducation, Science and Culture, and the Ministry of Healthand Public Welfare, Japan. We thank Dr. T.R. Breitman forhelpful advice.

References

[1] Perez A, Kastner P, Sethi S, Lutz Y, Reibel C, Chambon P. PMLRARhomodimers: distinct DNA binding properties and heteromericinteractions with RXR. EMBO J 1993;12:3171–82.

[2] Grignani F, De Matteis S, Nervi C, Tomassoni L, Gelmetti V, CioceM, et al. Fusion proteins of the retinoic acid receptor-� recruit histonedeacetylase in promyelocytic leukaemia. Nature 1998;391:815–8.

[3] Chen ZX, Xue YQ, Zhang R, Tao RF, Xia XM, Li C, et al. Aclinical and experimental study on all-trans retinoic acid-treated acutepromyelocytic leukemia patients. Blood 1991;78:1413–9.

[4] Warrell RPJ, Frankel SR, Miller WHJ, Scheinberg DA, Itri LM,Hittelman WN. Differentiation therapy of acute promyelocyticleukemia with tretinoin (all-trans-retinoic acid). N Engl J Med 1991;324:1385–93.

[5] Cornic M, Chomienne C. Induction of retinoid resistance by all-transretinoic acid in acute promyelocytic leukemia after remission. LeukLymphoma 1995;18:249–57.

[6] Degos L, Dombret H, Chomienne C, Daniel MT, Miclea JM,Chastang C, et al. All-trans-retinoic acid as a differentiating agent inthe treatment of acute promyelocytic leukemia. Blood 1995;85:2643–53.

[7] Shen ZX, Chen GQ, Ni JH, Li XS, Xiong SM, Qiu QY, et al. Useof arsenic trioxide (As2O3) in the treatment of acute promyelocyticleukemia (APL): II. Clinical efficacy and pharmacokinetics inrelapsed patients. Blood 1997;89:3354–60.

[8] Soignet SL, Maslak P, Wang ZG, Jhanwar S, Calleja E, Dardashti LJ.Complete remission after treatment of acute promyelocytic leukemiawith arsenic trioxide. N Engl J Med 1998;339:1341–8.

[9] Delva L, Cornic M, Balitrand N, Guidez F, Miclea JM, Delmer A, etal. Resistance to all-trans retinoic acid (ATRA) therapy in relapsingacute promyelocytic leukemia: study of in vitro ATRA sensitivityand cellular retinoic acid binding protein levels in leukemic cells.Blood 1993;82:2175–81.

[10] Kizaki M, Ueno H, Yamazoe Y, Shimada M, Takayama N, Muto A,et al. Mechanisms of retinoid resistance in leukemic cells: possiblerole of cytochrome P450 andP-glycoprotein. Blood 1996;87:725–33.

[11] Kizaki M, Ueno H, Matsushita H, Takayama N, Muto A, AwayaN, et al. Retinoid resistance in leukemic cells. Leuk Lymphoma1997;25:427–34.

[12] Shao W, Benedetti L, Lamph WW, Nervi C, Miller WHJ. Aretinoid-resistant acute promyelocytic leukemia subclone expressesa dominant negative PML-RAR� mutation. Blood 1997;89:4282–9.


[13] Imaizumi M, Suzuki H, Yoshinari M, Sato A, Saito T, Sugawara A,et al. Mutations in the E-domain of RAR portion of the PML/RARchimeric gene may confer clinical resistance to all-trans retinoic acidin acute promyelocytic leukemia. Blood 1998;92:374–82.

[14] Ding W, Li YP, Nobile LM, Grills G, Carrera I, Paietta E, et al.Leukemic cellular retinoic acid resistance and missense mutationsin the PML-RAR� fusion gene after relapse of acute promyelocyticleukemia from treatment with all-trans retinoic acid and intensivechemotherapy. Blood 1998;92:1172–83.

[15] Cote S, Zhou D, Bianchini A, Nervi C, Gallagher RE, MillerJr WH. Altered ligand binding and transcriptional regulation bymutations in the PML/RAR� ligand-binding domain arising inretinoic acid-resistant patients with acute promyelocytic leukemia.Blood 2000;96:3200–8.

[16] Cote S, Rosenauer A, Bianchini A, Seiter K, Vandewiele J, NerviC, et al. Response to histone deacetylase inhibition of novelPML/RAR� mutants detected in retinoic acid-resistant APL cells.Blood 2002;100:2586–96.

[17] Zhou DC, Kim SH, Ding W, Schultz C, Warrell Jr RP, GallagherRE. Frequent mutations in the ligand-binding domain of PML-RAR� after multiple relapses of acute promyelocytic leukemia:analysis for functional relationship to response to all-trans retinoicacid and histone deacetylase inhibitors in vitro and in vivo. Blood2002;99:1356–63.

[18] Robertson KA, Emami B, Collins SJ. Retinoic acid-resistant HL-60R cells harbor a point mutation in the retinoic acid receptorligand-binding domain that confers dominant negative activity. Blood1992;80:1885–9.

[19] Li YP, Said F, Gallagher RE. Retinoic acid-resistant HL-60 cellsexclusively contain mutant retinoic acid receptor-alpha. Blood1994;83:3298–302.

[20] Duprez E, Benoit G, Flexor M, Lillehaug JR, Lanotte M. A mutatedPML/RARA found in the retinoid maturation resistant NB4 subclone,NB4-R2, blocks RARA and wild-type PML/RARA transcriptionalactivities. Leukemia 2000;14:255–61.

[21] Vyas RC, Frankel SR, Agbor P, Miller WHJ, Warrell RPJ,Hittelman WN. Probing the pathobiology of response to all-trans retinoic acid in acute promyelocytic leukemia: prematurechromosome condensation/fluorescence in situ hybridization analysis.Blood 1996;87:218–26.

[22] Savill J, Fadok V, Henson P, Haslett C. Phagocyte recognition ofcells undergoing apoptosis. Immunol Today 1993;14:131–6.

[23] Sato A, Imaizumi M, Saito T, Yoshinari M, Suzuki H, Hayashi Y,et al. Detection of apoptosis in acute promyelocytic leukemia cellsin vivo during differentiation-induction with all-trans retinoic acidin combination with chemotherapy. Leuk Res 1999;23:827–32.

[24] Kizaki M, Matsushita H, Takayama N, Muto A, Ueno H, Awaya N, etal. Establishment and characterization of a novel acute promyelocyticleukemia cell line (UF-1) with retinoic acid-resistant features. Blood1996;88:1824–33.

[25] Takayama N, Kizaki M, Hida T, Kinjo K, Ikeda Y. Novel mutationin the PML/RAR� chimeric gene exhibits dramatically decreased

ligand-binding activity and confers acquired resistance to retinoicacid in acute promyelocytic leukemia. Exp Hematol 2001;29:864–72.

[26] Umemiya H, Kagechika H, Fukasawa H, Kawachi E, EbisawaM, Hashimoto Y, et al. Action mechanism of retinoid-synergisticdibenzodiazepines. Biochem Biophys Res Commun 1997;233:121–5.

[27] Imaizumi M, Sato A, Koizumi Y, Inoue S, Suzuki H, Suwabe N,et al. Potentiated maturation with a high proliferating activity ofacute promyelocytic leukemia induced in vitro by granulocyte orgranulocyte/macrophage colony-stimulating factors in combinationwith all-trans retinoic acid. Leukemia 1994;8:1301–8.

[28] Vermes I, Haanen C, Steffens Nakken H, Reutelingsperger CP. Anovel assay for apoptosis. Flow cytometric detection of phosphatidyl-serine expression on early apoptotic cells using fluorescein labelledAnnexin V. J Immunol Methods 1995;184:39–51.

[29] Tominaga T, Kure S, Yoshimoto T. DNA fragmentation in focalcortical freeze injury of rats. Neurosci Lett 1992;139:265–8.

[30] Marasca R, Zucchini P, Galimberti S, Leonardi G, Vaccari P,Donelli A, et al. Missense mutations in the PML/RAR� ligandbinding domain in ATRA-resistant As(2)O(3) sensitive relapsed acutepromyelocytic leukemia. Haematologica 1999;84:963–8.

[31] Slack JL, Gallagher RE. The molecular biology of acutepromyelocytic leukemia. Cancer Treat Res 1999;99:75–124.

[32] Shao W, Rosenauer A, Mann K, Chang CP, Rachez C, Freedman LP,et al. Ligand-inducible interaction of the DRIP/TRAP co-activatorcomplex with retinoid receptors in retinoic acid-sensitive and -resistant acute promyelocytic leukemia cells. Blood 2000;96:2233–9.

[33] Imaizumi M, Breitman TR. Retinoic acid-induced differentiationof the human promyelocytic leukemia cell line, HL-60, and freshhuman leukemia cells in primary culture: a model for differentiationinducing therapy of leukemia. Eur J Haematol 1987;38:289–302.

[34] Taimi M, Breitman TR. Growth, differentiation, and death of retinoicacid-treated human acute promyelocytic leukemia NB4 cells. ExpCell Res 1997;230:69–75.

[35] Chambon P. A decade of molecular biology of retinoic acid receptors.FASEB J 1996;10:940–54.

[36] Fanelli M, Minucci S, Gelmetti V, Nervi C, Gambacorti-PasseriniC, Pelicci PG. Constitutive degradation of PML/RAR� throughthe proteasome pathway mediates retinoic acid resistance. Blood1999;93:1477–81.

[37] Benoit G, Altucci L, Flexor M, Ruchaud S, Lillehaug J, RaffelsbergerW, et al. RAR-independent RXR signaling induces t(15;17) leukemiacell maturation. EMBO J 1999;18:7011–8.

[38] Takahashi N, Breitman TR. Covalent modification of proteins byligands of steroid hormone receptors. Proc Natl Acad Sci USA1992;89:10807–11.

[39] Yoshinari M, Imaizumi M, Sato A, Minegishi M, Fujii K, Suzuki H.G-CSF induces apoptosis of a human acute promyelocytic leukemiacell line, UF-1: possible involvement of Stat3 activation and alteredBax expression. Tohoku J Exp Med 1999;189:71–82.

[40] Rego EM, Wang ZG, Peruzzi D, He LZ, Cordon-Cardo C, PandolfiPP. Role of promyelocytic leukemia (PML) protein in tumorsuppression. J Exp Med 2001;193:521–9.

Alteration in the cellular response to retinoic acid of a human acute promyelocytic leukemia cell...

Documents

Transcript of Alteration in the cellular response to retinoic acid of a human acute promyelocytic leukemia cell...