Characterization of Acute Promyelocytic Leukemia Cases With PML-RARa Break/Fusion Sites in PML Exon 6: Identification of a Subgroup With

Decreased In Vitro Responsiveness to All-Trans Retinoic Acid

By Robert E. Gallagher, Yun-Ping Li, Sreenivas Rao, Elisabeth Paietta, Janet Andersen, Polly Etkind, John M. Bennett, Mart in S. Tallman, and Peter H. Wiernik

Of 113 acute promyelocytic leukemia cases documented to have diagnostic PML-RARa hybrid mRNA, 10 cases (8.8%) had fusion sites in PML gene exon 6 (V-forms) rather than in the two common hybrid mRNA configurations resulting from breaksites in either PML gene intron 6 (L-forms) or intron 3 (S-forms). In 4 V-form cases, a common break/fusion site was discovered at PML gene nucleotide (nt) 1685, abut- ting a 3' cryptic splice donor sequence. The fusion site was proximal to the common site in 1 case and more distal in 5 cases. The open reading frame encoding a PML-RARa gene was consistently preserved, either by an in-frame fusion site or by the insertion of 3 to 127 unidentified nts. In 2 V-form cases, hybridization analysis of the reverse transcriptase- polymerase chain reaction products with a PML-RARa juc-

T HE RECIPROCAL t( 15; 17) chromosome translocation specific for acute promyelocytic leukemia (APL) gen-

erates two hybrid gene products, PML-RARa and RARa- PML."' The PML-RARa fusion product seems of greatest significance, because it contains most functional components of each gene and is expressed in all t( 15; l7)-positive cases, whereas RARa-PML is not coexpressed in 20% to 30% of cases.' Both the PML and RARa genes have multidomain structures with sequences characteristic of transcriptional regulators (see Grignani et a16 for a recent review). The more extensively studied RARa gene is a member of the steroid- thyroid hormone receptor gene superfamily with typical DNA and ligand (RA) binding domains and can positively or negatively regulate many different gene promoters after binding to adjacent RA response elements (RAREs) in the liganded or nonliganded state. The ligand binding domain

From the Montejiore Medical Center, Albert Einstein Cancer Cen- ter, Bronx, NY: the Dana Farber Cancer Institute, Boston, MA: the University of Rochester Medical Center, Rochester, NY: und the Northwe.stern University Medical School, Chicago, IL.

Submitted November 22, 1994: accepted April 12, 1995. Supported by Public Health Service Grants No. CA14985.

CAI7145, CA56771, P30CA13330. CA66636, CA11083, CA21115. and CA23318 from the National Cancer Institute, National Institutes of Health, and the Department of Health and Human Senices and by the Elsa U. Pardee Foundation and Chemotherapy Foundation.

This study was conducted in the majority by the Eastern Cooperu- tive Oncology Group (Douglass C. Tormey, MD, PhD. Chuirmun).

The contents of this article are solely the responsibility of the authors and do not necessarily represent the oficial views of the National Cancer Institute.

Address reprint requests to Robert E. Gallagher, MD, Department of Oncology, Montejiore Medical Center, 111 E 210th St, Bronx, NY 10467.

The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereb-y marked "advertisement'' in accordance with 18 U.S.C. srctiotz 17.34 solely to indicate this fi~firct. 0 1995 by The American Society of Hematology. 0006-4971/95/8604-0034$3.00/0

1540

tion probe was required for discrimination from L-form cases. Two V-form subgroups were defined by in vitro sensi- tivity to all-trans retinoic acid (tRA)-induced differentiation: 4 of 4 cases tested with fusion sites at or 5' to nt 1685 (subgroup E6S) had reduced sensitivity (E& z I O " mol/L), whereas 4 of 4 cases with fusion sites at or 3' to nt 1709 (subgroup E6Ll had high sensitivity (E& <IO-' mol/L) indis- tinguishable from that of L-form and S-form cases. These results provide the first link between PML-RARa configura- tion and tRA sensitivity in vitro and support the importance of subclassifying APL cases according to PML-RARa tran- script type. 0 1995 by The American Society of Hematology.

of RARa also contains subdomains essential for receptor dimerization and ligand-dependent transcriptional activation. Although the PML gene has yet to be definitively established as a transcriptional regulator, the amino end contains three cysteine-rich regions followed by a long a helical region with a leucine zipper that, respectively, provide interfaces for DNA-protein and protein-protein interactions character- istic of transcriptional regulatory proteins.'

Two predominant types of PML-RARa hybrid mRNA have been observed in APL patients caused by alternative breakage of the 9 exon-long PML gene in either intron 3 or intron 6.'~3."'' After the appropriate removal of intronic sequences, the 3' end of PML exon 3 or 6 is spliced to a common 5' end of exon 3 of the RARa gene, which is invariably broken in intron 2 (Fig 1A). These two different mRNA products, which differ in length by 474 nucleotides (nts) are referred to as the short (S)- and long (L)-forms of PML-RARa. Each contains the DNA-protein and protein dimerization regions of both component peptides. The S- form lacks a 158 amino acid-long, prolinekerine-rich seg- ment of PML that contains several potential phosphorylation sites that could be important in modulating transcriptional

Functional testing in in vitro cotransfection ex- periments of S- and L-form PML-RARa expression plas- mids with reporter gene vectors linked to various RAREs in a variety of target cells showed relatively minor differences between the However, these results may not be repre- sentative, because genes that regulate differentiation in re- sponse to all-trans retinoic acid (tRA) in APL cells are not known and because potential differential effects of the two forms of PML-RARa may be effected through alternative regulatory mechanisms involving the PML component of PML-RARcx.'~"' In two recent clinical reports of APL pa- tients initially treated with tRA followed by chemotherapy, S-form cases had a significantly worse prognosis than did L-form c a ~ e s . ~ " ~ " An earlier study by one of these groups, apparently involving the same cohort of APL patients,'" had shown no difference between the in vitro tRA sensitivity of APL cells with either the L- or S-form configuration of PML- RARa."

Blood, Vol 86, No 4 (August 15). 1995: pp 1540-1547

For personal use only.on February 12, 2018. by guest www.bloodjournal.orgFrom

APL SUBGROUP WITH REDUCED TRA SENSITIVITY 1541

Fig 1. RT-PCR analysis of V- form PML-RARa mRNAs. (A) Schematic representation of in- tron-exon structure of regions of PML and RARn genes re- arranged in association with t(15;17) in APL. Intronsarerepre- sented by lines, exons by boxes ([W] conserved in PML-RARa hy- brids; [RI variably removed by breakpoint site or alternative splicing in different PML-RARa hybridslisoforms; [U1 excluded from PML-RARn hybrid, variably included in reciprocal RARa-PML hybrids). Vertical arrows indi- cate PML gene breaksites in in- tron 3, exon 6, or intron 6, re- spectively, in S-, V-, and L-forms of PML-RARcr. The horizontal, double-headed arrow indicates the universal site of breakage in RARa intron 2. Asterisks indicate the sites of four PCR primers; 5‘ PML primers were chosen from exon 3 (P31 and exon 6 (P61 and matched to 3‘ RARa primers from exon 4 (R4b and R4a). (B) Ethidium bromide-stained RT- PCR DNA products generated by amplification with the indicated primer pairs and after gel elec- trophoresis. M, Hae Ill-digested @X 174 DNA marker with indi- cated fragment sizes; 0, no RNA; N. NB4 cell RNA; S, S-form APL patient RNA; V, V-form APL pa- tient (no. 5); L, L-form APL pa- tient RNA. IC) Ethidium bro- mide-stained RT-PCR products from indicated V-form APL pa- tients with little or no size de- crease relative t o L-form con- figuration (NB4 cell control). (D) Schematicsummaryof break/fu- sion sites in nucleotide sequence of the 3’4erminus of PML exon 6 in V-form APL patients. Double- headed arrows indicate break/ fusion sites; heavy arrows indi- cate nonunique sites. Underlin- ing indicates the consensus splice acceptor nucleotide se- quence. Asterisks indicate the phosphorylation site of single- letter-coded amino acid se- quence.

A PML RARa

Exon 3 4 5 6 7 4 R R B B *

3 4 All Sites ..

V L I , Breaksite

S

B Primers

M ~

1353 bp - 872 bp - 603 bp -

310 bp - 234 bp -

C

P3 x R4a P6 x R4b

O N S L V N L V ~ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

Case No.

M O N 9 1 0 1 1353 b p - , I

I 872 bp - 603 bp -

310 bp -

D PR 1

PHL Exon 6

For personal use only.on February 12, 2018. by guest www.bloodjournal.orgFrom

1542 GALLAGHER ET AL

In addition to S- and L-form types of PML-RARa, there have been two reports of APL cases in which the PML gene break/fusion site occurred within exon 6.7,9 In one study, this form was detected in 20% of cases (apparently 30 to 40 total case^).^ In the other study, it was detected in 1 of 31 cases9 In three other reports, no apparent exon 6 breawfusion site cases were detected in a total of about 100 cases,3.8.’1 al- though l of these cases was subsequently reclassified as an apparent exon 6 case by higher resolution reverse tran- scriptase-polymerase chain reaction (RT-PCR).IX Sequence analysis in 4 cases disclosed that the breaksite within exon 6 occurred at different nts in each case,7.“ for which reason w e refer to these cases as variable (V)-forms.

W e report here a comprehensive analysis of the molecular, biologic, and pretreatment clinical characteristics of 10 APL patients with V-form PML-RARn mRNA, identified from a total of 113 PML-RARa-positive patients with this rela- tively rare form of acute myelogenous leukemia. Our results provide a plausible explanation for the wide disparity in detection of this PML-RARn type in previous reports. More importantly, they indicate that a molecularly definable subset of V-form cases has reduced tRA sensitivity in vitro and not unlikely a poor overall prognosis. Additionally, the identifi- cation and molecular analysis of such cases may provide important information about the mechanism of APL cellular resistance to tRA.

MATERIALS AND METHODS

Patients andpretreatment clinical laboratory studies. Of the 1 13 patients involved in this study, 94 were participants in the Eastern Cooperative Oncology Group (ECOG) intergroup clinical trial E2491 and/or the laboratory ancillary study E1485. The other 19 patients were studied at Albert Einstein Cancer Center hospitals. Cytologic evaluation with French-American-British (FAB) classifi- cation,’’ cytogenetic analysis of 20 or more metaphase spreads, and immunophenotyping with a panel of monoclonal antibodies” were routinely performed with central review in ECOG cases and in all V-form APL cases. Patient materials were obtained under an ECOG protocol-approved consent form and/or individual institutional re- view board approval. Baseline patient data, eg, white blood cell (WBC) counts, were obtained from the central data registry in ECOG cases.

Cell preparation and culture. Heparinized bone marrow (BM) and peripheral blood (PB) specimens were received by overnight express mail from ECOG institutions at room temperature, except for specimens for immunophenotype studies, which were received on wet ice. A low-density WBC fraction (density 51.077 g/mL) was isolated by standard sodium metrizamide step gradient centrifugation (Histopaque 1077; Sigma, St Louis, MO). Cell recovery was evalu- ated in the presence of trypan blue dye and, as measured by dye exclusion, cell viability was virtually always 290%. Cell slide mounts were prepared using a Cytospin 2 centrifuge (Shandon, Pitts- burgh, PA) and stained with a modified Wright’s stain (Dif-Quik; Scientific Products, McGraw Hill, IL). Only low-density cell frac- tions with 275% myeloblasts/promyelocytes were used for cell cul- ture studies.

For tRA sensitivity testing, washed low-density cells were sus- pended at 5 X 10’ cells/mL in Iscove’s modified Dulbecco’s medium (IMDM) containing 10% fetal bovine serum (GIBCO, Grand Island, NY or HyClone, Logan, UT), 50 U/mg/mL of penicillin/streptomy- cin, and 0.02% ethanol 5 IO-* to mol& tRA (Sigma). tRA was solubilized in 100% ethanol as mol& stocks, stored at

-20°C in light-protected vials for up to I month, and handled under light-subdued conditions throughout the experiments. Cultures were incubated in a totally humidified, 5% COz atmospere at 37°C. After S days, viable cell counts, Dif-Quik-stained Cytospin cell mounts. and nitroblue tetrazolium (NBT; Sigma) tests were performed and analyzed as previously described.”

RT-PCR analysis. Total cellular RNA was prepared by a modi- tication of the guanidine isothiocyanate extraction-cesium chloride gradient ultracentrifugation procedure. as previously described,” or using the TRIzol reagent (Life Technologies, lnc, Gaithersburg. MD), according to the supplier’s directions.

Reverse transcription of RNA from random hexamer primers (Pharmacia, Piscataway, NJ) and PCR were performed essentially as previously described,’2 except that amplification conditions were modified to be optimal for each primer pair used. Primer pairs were selected with the assistance of Oligo software (National Biosciences, Inc, Plymouth, MN). The primers used to amplify first-strand cDNA for PML-RARa mRNA were P3. S‘-ACCGATGGCTTCGACGAG- TTC-3’ (sense orientation), and R4a, 5’-AGCCCTTGCAGCCCT- CACAG-3’ (antisense orientation). In V-form cases, a I : 100 dilution of the first-round PCR amplification product was reamplified from the following nested primers: P6,5’-AATACAACGACAGCCCAG- AAG-3‘ (sense), and R4b, 5’-CTCACAGGCGCTGACCCCAT-3‘ (antisense). PCR products (8-pL aliquots) were fractionated on 3 8 agarose gels run in Tris-acetate buffer, pH 7.2, and were poststained with ethidium bromide (0.5 pg/mL for 20 minutes).

PCR products from all putative L-form PML-RARa cases were rerun in a common gel, Southern blot transferred to nylon mem- branes (Nytran, 0.45 pm pore size; Scleicher & Schuell, Keene, NH), and hybridized and washed under stringent conditions with the following [”Plend-labeled 15-mer probe that spans the PML exon 6-RARa exon 3 junction (Fig ID): PRl, 5”TCAATGGCTGCC- TCC-3’ (antisense). After the hybridization procedure, the nylon filters were radioautographed overnight using Kodak XAR-5 film (Eastman Kodak, Rochester, NY) and Dupont lightening plus inten- sifying screens (DuPont, Wilmington, DE) at -80°C and were then inspected for the presence or absence radioautographic bands corre- sponding to the upper and middle ethidium bromide staining bands (Fig lB, lane L). PCR products of radiographically negative cases (2/S9) were further analyzed by sequence analysis.

DNA sequence analysis. The nucleotide sequence of the trun- cated 3”terminus of PML exon 6 and proximately joined segment from the B-region of RARa in all PML-RARa V-form cases was determined by direct sequencing of P6 X R4b PCR products, using the following antisense primer anchored in RARa exon 3: RSI, S’- GACCCCATAGTGGTAGCCTGAGGA-3’. In most cases, the PCR reaction product was used without further processing, but in a few instances in which additional, light, apparently extraneous gel bands appeared, the principal band of interest was first cleaned up by cutting from the gel and purifying it using a PCR Magic Prep kit (Promega, Madison, WI). Sequencing reactions were performed US-

ing the Sequenase PCR Product Sequencing Kit (US Biochemical, Cleveland, OH), according to the supplier’s detailed instructions. The reaction mixture, radiolabeled with a[’S]dATP ( 3 pL; Dupont NEN, Boston, MA), was electrophoresed on a 5% Long Ranger sequencing gel (AT Biochem, Malvern, PA) for 2 hours at 60 W. Gels were dried for I hour at 80°C and then exposed to X-OMAT- AR film (Kodak) at room temperature for various periods.

Datu analysis. Differences in in vitro sensitivity among the dif- ferent PML-RARa forms (L., S-, V/E6S-, and VE6L-forms) in different APL cases were evaluated with repeated measures analysis of variance in SAS (V 6.07; SAS Institute, Cary, NC). Differences between E6S and E6L cases in presenting clinical data were evalu- ated with a Wilcoxon rank sum test. P values <.05 were considered

For personal use only.on February 12, 2018. by guest www.bloodjournal.orgFrom

APL SUBGROUP WITH REDUCED TRA SENSITIVITY 1543

Table 1. APL V-Form Cases: PML-RARrr Junction Nucleotide and Amino Acid Changes

~~~~~~~~~

Net Net

Case Junction nt Deletedlnt Change Substitution Change Length Amino Acid Length

No. (nt) Added (ntl (Junction) (aa)

Exon 6

1 1678 61/127* +66 42 aat +22 2-5 1685 54/0 -54 T -18

6 1702 37/25$ -12 QGLDTEVG -4 7 1709 30lGAG -27 R -9 8 1718 211GAG -18 G -6 9 1727 1210 -12 - -4

10 1731 81AGmGGG o w 0 ~~ ~ ~ ~~~~~ ~ ~ ~~

* Insert nt sequence: 5'-TGCCATCCTAACCTTCCATCTTGGCAAGG- GGCACTGGGTCCTTATGGGGTTGlTGTCCTGGCCCCAGACAClTGG- CTGTCATCTTTGAGGCTITCATCCCCAGGAGTGGAGGGGAGAAGC- TGCTCTG-3'.

t Insert amito acid (aa) sequence: CHPNLPSWQGALGPYGVWLAP-

i Insert nt sequence: 5'-GAAGGACTGGACACACAGGTTGE-3'. DTWLSSLRLEGVEGRSCS.

significant. No adjustment for multiple comparisons was made in this small hypothesis-generating study.

RESULTS

IdentiJcation of V-form cases by PML-RARa RT-PCR analysis. For orientation purposes, Fig 1A shows the in- tron-exon genomic structure of the PML and RARa genes in the region of the break sites associated with formation of the 3 types of PML-RARa hybrid genes in APL. The primer sites used for PCR amplification after conversion by an RT reaction of PML-RARa mRNA into first-strand cDNA from random primers are indicated by asterisks.

Lanes 3 through 6 of Fig 1B show typical results obtained for each of the PML-RARa types after primary PCR ampli- fication from primers P3 X R4a. S-form cases, in which exons 4 to 6 are spliced out, resulting in the direct fusion of the 3' end of PML exon 3 to the 5' end of RARa exon 3, were identified by a single band of 220 bp. L-form cases yielded 3 bands: (I) a full-length band of 694 bp from the direct fusion of PML exon 6 to RARa exon 3, (2) an interme- diate band of 550 bp that differs from the full-length form by the splicing out of PML exon 5, and (3) a lower band of 291 bp caused by the splicing out of PML exons 5 and 6, resulting in the direct fusion of the 3' end of PML exon 4 to the 5' end of RARa exon 3',' (and our confirmatory se- quencing results). Several V-form cases were recognized by the presence of relatively short upper and intermediate bands relative to their L-form counterparts. Confirmation that these V-form cases contain a shortened exon 6 was secured by reamplifying an aliquot of the P3 X R4a PCR product with a primer anchored in the 5' part of exon 6 (P6) with another primer from RARa exon 4 (R4b), nested to the 5' side of primer R4a, as shown compared with L-forms in lanes 7 through 9 in Fig 1B.

In one case (Table 1, case no. l), PCR showed that the upper and middle bands were longer than those from L-form cases (Fig 1C). This finding suggested that in some cases an insert sequence at the breaksite in PML exon 6 might

compensate the deletion and render it undetectable by stan- dard gel electrophoretic analysis. Consequently, PCR prod- ucts from all cases that had been assigned the L-form type were screened for their ability to hybridize to an oligonucleo- tide probe that spanned the L-form PML-RARa junction (probe PR1, Fig 1D) after Southern blot transfer of the gel electrophoretic DNA bands. This analysis confirmed that case no. 1, indeed, lacked the 3'-terminus of exon 6 and led to the identification of two additional V-form cases (no. 9 and 10; hybridization data not shown). In case no. 10, the insert sequence was exactly equal to the deleted sequence such that no difference in PCR product length could be discerned by gel electrophoretic analysis (Fig 1C).

Sequence analysis of V-form PML-RARa splice junctions. The results of performing direct DNA sequence analysis on R6 X P4b PCR products from a sequencing primer anchored in RARa exon 3 (RS1, Fig 1D) are summarized in Table 1 and Fig 1D. The same breaWfusion site after PML nt 1685 was identified in 4 cases (no. 2 through 5), which was imme- diately followed by the consensus splice donor sequence GTGAG (underlined in Fig 1D). All other exon 6 breaks occurred at unique sites between PML nts 1678 to 1731, the latter only 8 nts from the normal 3'-terminus of exon 6. In 5 cases (no. 2 through 5 and 9), the last PML nt was directly fused to RARa exon 3. In the other 5 cases (no. 1 ,6 through 8, and lo), there was an insert of 3 to 127 nts. In all cases, the translational open reading frame (OW) for PML-RARa was maintained, encoding a putative protein from - 18 to 1-22 amino acids different in length from that encoded by the full-length L-form mRNA. Other than the tri-nt GAG, which likely is a cryptic splice donor signal(s) in intron 2 from the RARa gene,' there was no sequence homology between the various inserts or to any region of normal PML or RARa exonic sequences. A search of the EMBL and GenBank sequence data bases (CD-ROM Release 12.0 PC/ GENE, 3/94; IntelliGenetics, Inc, Mountain View, CA) failed to disclose any perfect matches for the 25 nt or 127 nt insert sequences. The 42 amino acid insert in case no. 1 contained a potential phosphorylation site (Table 1 legend, asterisked) that resembled such sites in the serine/proline- rich region of the normal PML gene.'*

Incidence of V-form PML-RARa in APL. The 10 V-form cases were derived from 113 successive APL cases docu- mented in our laboratory to be positive for PML-RARa hybrid mRNA by RT-PCR analysis, ie, an incidence of 8.8%.

In vitro tRA sensitivity. Eight of the 10 V-form cases were tested for sensitivity to tRA-induced differentiation, as measured by the NBT dye reduction test, in liquid suspension culture. At the lowest tRA concentration tested, lo-' moll L, 4 cases (no. 7 through IO), all with exon 6 breafusion sites at or 3' to PML nt 1709, showed NBT positivity in 270% cells, whereas 4 cases (no. 2 through 5) with exon 6 break sites at or 5' to PML nt 1685 showed positivity in 535% of cells (Fig 2A). Even at 10" and m o m tRA, 3 of 4 cases with relatively short exon 6 segments failed to or barely reached 50% NBT positivity. The tRA dose-response curve generated from the mean NBT positivity values of the V-form cases with relatively short exon 6 segments (E6S subgroup) significantly differed from that of the V-form

For personal use only.on February 12, 2018. by guest www.bloodjournal.orgFrom

1544

bu 1 , l , // /

404 I

0 0 7 6

tRA CONCENTRATION: -LW [M]

100 1 B

80 1 h

60 1

T

0-1 1

0 8 7 6

tRA CONCENTRATION: -Log [M]

Fig 2. NBT responsivity of APL cells from 8 patients with V-form PML-RARa hybrid mRNAs to various concentrations of tRA. (A) Re- sults of individual patient testing. Open symbols indicate E6L cases; solid symbols indicate E S cases. (B) Grouped results of V-form E 6 S (n = 4; squares), V-form E6L In = 4; circles), L-form In = 44; triangles), end S-form (n = 31; diamonds) cases ? SE bars.

cases with relatively long exon 6 segments (E6L subgroup), as well as from those generated from 44 L-form and 31 S- form cases (P s .0001 by repeated measure ANOVA; Fig 2B). There were no significant differences among the tRA dose-response curves of the E6L subgroup, L-form, and S- form cases.

Pretreatment clinical features. As summarized in Table 2, all of the V-form APL cases were studied at the time of presentation with a single exception (case no. 3), in which the molecular and tRA sensitivity studies were performed at first relapse. The median age of the 10 V-form cases was 38 years. Six of the cases were male and 4 were female. Six of 10 cases had presentation WBC counts greater than 10 X 109/L (median, 31 X 109/L). However, most notably, S of

GALLAGHER ET AL

these high WBC cases belonged to the E6S subgroup (me- dian, 67 X 10'L). For the E6L subgroup, the median WBC count was 3 X IO'/L, which is similar to that for many reports of APL in general. The distribution of the WBC counts of the E6S subgroup significantly differed from that of the E6L subgroup ( P < .OS , Wilcoxon rank sum test).

Using cytologic analysis, 6 of l0 V-form cases had classi- cal M3 characteristics. Four of 10 cases, including 3 of S high WBC count E6S cases, had microgranular variant (M3v) features." Overall, the leukemic promyelocytes had typical immunophenotypic features of APL cells, ie, low expression of HLA-DR, CD34, CD1 lb, CD14, CDlS, and P-glycopro- tein antigens2" Notably, in 4 cases HLA-DR or/and CD1 1 b expression was atypically higher than the median observed in a large series of APL cases (Table 2), although the median percentage of leukemic promyelocytes expressing HLA-DR in the V-form cases (7%) was similar to that reported for overall APL cases (S%).2" On the other hand, the median value for CD1 1 b (13%) as well as CD14 ( 1 1 %) in the V- form cases was greater than the intraquartile (IQ) values reported for all APL cases (4S% to 10% and 2% to 1096, respectively). Expression of HLA-DR, CD1 lb, and CD14 antigens between the E6S and E6L subgroups appeared to be similar, although it seems noteworthy that all but one of the atypical levels of antigen expression occurred in the E6S subgroup. One M3v case (no. 5 ) expressed relatively increased CD2 antigen.

Two of the V-form cases lacked the characteristic t( 1 S ; 17) chromosome, despite the presence of PML-RARa hybrid mRNA. In case no. 4, this may have been caused by an inadequate specimen, because only 8 metaphase spreads were evaluable. However, in case no. S , no t( 15; 17) chromo- some marker was observed, despite an analysis of 100 meta- phases [at relapse, the t( 15; 17) was identified in 2 of 21 metaphases]. Three E6S cases (no. 1, 4, and S ) had addi- tional/other chromosome abnormalities.

DISCUSSION

The incidence of APL patients with PML-RARa genes with exon 6 breaklfusion sites has varied widely in previous reports from 0% in most study series to 3% to 20% of

RARa-positive APL patients (8.8%) is consistent with a true incidence in the range of 5% to 10% of cases, as pro- posed in a recent review.' Because some of our cases were detected only by special RT-PCR studies, including hybrid- ization analysis with a PML-RARa junction-specific probe to detect cases with little or no length differences from L- form cases, it seems likely that some V-form cases were overlooked in previous studies.

In addition to the 10 cases reported here, sequence analysis of the PML-RARa junction has been reported in 4 other V- form cases, which identified the following PML mRNA breaklfusion sites: nt 1587 with a 28 nt insert; nt 1675 with a 4 nt insert; nt 1678 with a 25 nt insert; and nt 1685 with no Two of these sites at nt 1678 and 1685 coincide with those found in the current report, although completely different insert sequences were present in the two nt 1678 break/fusion site cases. Pandolfi et al' reported that the insert

cases,3.7-9.1 I Our finding of 10 cases in 113 successive PML-

For personal use only.on February 12, 2018. by guest www.bloodjournal.orgFrom

APL SUBGROUP WITH REDUCED TRA SENSITIVITY 1545

Table 2. Pretreatment Clinical Features of PML-RARa V-Form Cases

PB Cytogenetics

Case No.

1

Disease WBC BIPros FAB Atypical Cell Ag Expression* t(15;17) Other AgelSex Stage ClO’IU (96) Class (% Positive) Marker Marker

19/M U 11 40 M3 - + (de19q12q22) 2 7 1/M U 67 62 M3v HLA-DR(32) + -6,+6ring 3 331F U 51 78 +

R 25 75 M3v HLA-DR(4OL CDllb(46) 4 66/F U 120 95 M3 CD1 1 b(72) -t 5 32/F U 79 80 M3v CDZ(41) -* + l 3 6 31/M U 3 1 M3 - + 7 29/F U 70 86 M3 - + 8 44/M U 6 76 M3 9 56/M U 2 0 M3v CD1 1 b(32) +

- + -

-

- -

- + -

10 43/M U 1 10 M3 - + +8

-

Abbreviations: U, previously untreated; R, first relapse; BVPros, blasts/promyelocytes. Cell antigen (Ag) expression regarded as atypical if it differed by 2x median -t IQ value for 35 previously reported APL cases.2o

t Cytogenetic evaluation based on only 8 metaphase figures; all others are based on 2 2 0 metaphase figures. * 46,XX(97/100);47,XX,+13(3/100).

sequence from their nt 1678 case was derived from intron 2 of RARa. Indeed, it seems probable that most, possibly all, of the insert sequences, many of which are terminated by consensus splice donor sequences (eg, GAG in 3 of our cases), were derived from this genomic source. In 5 of the 14 PML exon 6 brewfusion sites so far reported the junction has been localized to PML nt 1685, which has a consensus intronic splice donor sequence (GTGAG) immediately downstream. This finding suggests that this fusion site could result from different downstream breaksites at the DNA level, likely in close proximity to nt 1685, because this cryp- tic splice donor sequence was not used in 5 cases with exon 6 bredfusion sites at or distal to nt 1702. Alternatively, nt 1685 could be an unusually susceptible site for chromosome 15 breakage and translocation in APL. In all cases, the trans- lational O W for PML-RARa is preserved, thus supporting the importance of this gene product for the development and maintenance of APL cells.

This is the first report suggesting that molecular resolution of PML-RARa V-form cases is of biologic importance and of potential clinical relevance. Although the number of cases tested was limited because of the rarity of this PML-RARa type of APL, the reduced sensitivity to tRA-induced NBT test positivity of APL cells from the V-form subgroup with relatively short residual exon 6 segments (PML fusion site s n t 1685; subgroup E6S) was statistically highly significant compared with the tRA sensitivity of cells from V-form cases with relatively long residual exon 6 segments (PML fusion site 2 1709; subgroup E6L) and from the predominant L-form and S-form cases. As described in many previous rep0rts,2~-** tRA-induced NBT test positivity correlates well with terminal differentiation of APL cells by cytologic crite- ria but provides a much less subjective measure of this pro- cess than differential cell counts of differentiating APL cells, which frequently manifest atypical cytologic features after short-term tissue culture. The present study emphasizes the importance of performing tRA titration experiments, because reduced tRA sensitivity was most apparent at lo-’ m o m

tRA (Fig 2) and because testing at only m o m tRA, as performed in some previous correlative clinical reports,25’28 could fail to detect reduced APL cellular tRA sensitivity.” Most of the few documented instances of reduced in vitro tRA sensitivity were derived from tRA-treated cases at the time of disease relapse, ie, with acquired clinical tRA resis- tan~e,”.’~ whereas virtually universal tRA sensitivity in vivo has been reported for APL cases not previously treated with tRA.”*’* Thus, it remains uncertain if the minor subgroup of APL patients with the E6S form of PML-RARa have re- duced clinical sensitivity to tu-induced remission induc- tion. However, a poor prognosis of E6S cases is suggested by two additional considerations. First, all 5 of these cases presented with a high WBC count, which has been the most consistent adverse risk factor in APL.3’,32 Also, 3 of 5 of these cases expressed atypical immunophenotypic markers” andor had abnormal chromosome markers in addition to or instead of the t(15; 17). Second, the E6S cases represent the only PML-RARcu-positive APL subgroup so far identified with reduced in vitro tRA sensitivity before tRA therapy. In contrast, the S-form subgroup, reported to have an increased incidence of clinical relapse” or of adverse outcome mea- sured by increased early death in combination with relapse,“ did not manifest reduced pretherapy in vitro tRA sensitivity (Chen et a19 and this report, Fig 2B). A more definite answer regarding a possible correlation of PML-RARa genotype and clinical outcome should become available when analysis of data from the international phase 111 study of tRA versus chemotherapy for the remission induction therapy of APL (protocol E2491) has been completed.

The association of reduced tRA sensitivity with relatively short exon 6 brewfusion sites in E6S V-form cases implies that this type of PML-RARa gene rearrangement could be mechanistically involved in APL cellular resistance to tRA. Based on our preliminary observations, we profer two possi- ble molecular mechanisms. First, the disturbed protein se- quence introduced by the brewfusion sites at nts 1678 or 1685 could interfere with the phosphorylatioddephosphory-

For personal use only.on February 12, 2018. by guest www.bloodjournal.orgFrom

1546 GALLAGHER ET AL

lation of nearby putative serine phosphorylation sites speci- fied by nts 1661 to 1663 and 1670 to 1672 (Fig lD).” Al- though we are unaware of any evidence for this related to PML or PML-RARa, the phosphorylation state of serine/ threonine residues has been shown to play an important role in the regulation of several transcriptional regulatory pro- teins 33.14 . . Second, in all 4 V-form cases with fusion sites at nt 1685, we have observed an additional PML-RARa isoform, which is not evident by RT-PCR analysis in all other types of PML-RARa (our unpublished results). This isoform lacks all of exon 6 due to alternative splicing and encodes a pro- spective, novel, aberrant PML pr~tein.’~ Because this trun- cated protein retains the putative DNA binding and protein dimerization domains of PML, it could function as a domi- nant negative inhibitor of PMLPML-RARa proteins.’ Fur- ther work is required to determine if either of these possible mechanisms is operative in the primary APL cellular tRA resistance of E6S cases and whether similar alterations could lead to the development of acquired (secondary) cellular tRA resistance in the common PML-RARa types after chronic therapy with tRA.27329

ACKNOWLEDGMENT

The authors thank the physicians, nursing staffs, and data manag- ers of the Albert Einstein Cancer Center and other ECOG-affiliated institutions for their cooperation in obtaining and sending the clinical specimens used for performing these studies. We also thank Shan Peng and Lucy Qiu for technical assistance and Dr Anna Velcich for assistance with the computer search for sequences homologous to PML-RARa gene inserts and for review of the manuscript.

REFERENCES 1. de The H, Lavau C, Marchio A, Chomienne C, Degos L,

Dejean A: The PML-RARa fusion mRNA generated by the t(15; 17) translocation in acute promyelocytic leukemia encodes a functionally altered RAR. Cell 66~675, 1991

2. Kakizuka A, Miller WH Jr, Umesono K, Warrell RP Jr, Frankel SM, Murty VV, Dmitrovsky E, Evans RM: Chromosomal transloca- tion t(l5; 17) in human acute promyelocytic leukemia fuses RARa with a novel putative transcription factor, PML. Cell 66:663, 1991

3. Kastner P, Perez A, Lutz Y, Rochette-Egly C, Gaub M-P, Durand B, Lanotte M, Berger R, Chambon P: Structure, localization and transcriptional properties of two classes of retinoic acid receptor- a fusion proteins in acute promyelocytic leukemia (APL): Structural similarities with a new family of oncoproteins. EMBO J 11:629, 1992

4. Pandolfi P, Grignani F, Alcalay M, Mencarelli A, Biondi A, LoCoco F, Grignani F, Pelicci P Structure and origin of the acute promyelocytic leukemia myURARa cDNA and characterization of its retinoid-binding and transactivation properties. Oncogene 6 1285, 1991

5. Alcalay M, Zangrilli D, Fagioli M, Pandolfi P, Mencarelli A, Lo Coco F, Biondi A, Grignani F, Pelicci P: Expression pattern of the RARa-PML fusion gene in acute promyelocytic leukemia. Proc Natl Acad Sci USA 89:4840, 1992

6. Grignani F, Fagioli M, Alcalay M, Longo L, Pandolfi P, Donti E, Biondi A, Lo Coco F, Grignani F, Pelicci P: Acute promyelocytic leukemia: From genetics to treatment. Blood 83: 10, 1994

7. Pandolfi P, Alcalay M, Fagioli M, Zangrilli D, Mencarelli A, Diverio D, Biondi A, Lo Coco F, Rambaldi A, Grignani F, Rochette- Egly C, Gaube M-P, Chambon P, Pelicci P: Genomic variability and alternative splicing generate multiple PML/RARn transcripts that

encode aberrant PML proteins and PML/RARa isoforms in acute promyelocytic leukaemia. EMBO J 1 1: 1397, 1992

8. Miller WH Jr, Kakizuka A, Frankel SR, Warrell RP Jr, De- Blasio A, Levine K, Evans RM, Drnitrovsky E: Reverse transcription polymerase chain reaction for the rearranged retinoic acid receptor- a clarifies diagnosis and detects minimal residual disease in acute promyelocytic leukemia. Proc Natl Acad Sci USA 89:2694, 1992

9. Chen S-J, Chen 2, Chen A, Tong J-H, Dong S, Wang Z-y, Waxman S, Zelent A: Occurrence of distinct PML-RARa fusion gene isoforms in patients with acute prornyelocytic leukemia de- tected by reverse transcriptase/polymerase chain reaction. Oncogene 7: 1223, 1992

10. Huang W, Sun G-L, Li X-S. Cao Q, Lu Y, Jang G-S, Zhang F-Q, Chai J-R, Wang Z-y, Waxman S, Chen Z, Chen S-J: Acute promyelocytic leukemia: Clinical relevance of two major PML- RARa isofoms and detection of minimal residual disease by retro- transcriptase/polymerase chain reaction to predict relapse. Blood 82: 1264, 1993

I 1. Naoe T, Ohno R: Molecular heterogeneity of the PML gene rearrangement in acute promyelocytic leukemia in Japan: Prevalence and clinical significance. Leukemia 8:1086, 1994

12. Fagioli M, Alcalay M, Pandolfi P, Venturini L, Mencarelli A, Simeone A, Acampora D, Grignani F, Pelicci P: Alternative splicing of PML transcripts predicts coexpression of several carboxy- terminally different protein isoforms. Oncogene 7: 1083, 1992

13. Perez A, Kastner P, Sethi S, Lutz Y , Reibel C, Chambon P: PMLRAR homodimers: Distinct DNA binding properties and heterodimeric interactions with RXR. EMBO J 12:3171, 1993

14. Weis K, Rambaud S, Laqvau C, Jansen J, Carvalgo T, Carmo- Fonseca M, Lamond A, Dejean A: Retinoic acid regulates aberrant nuclear localization of PML-RARa in acute promyelocytic leuke- mia. Cell 76:345, 1994

15. Koken M, Puvion-Dutilleul F, Guillemin M, Viron A, Li- nares-Cruz G, Stuurman N, de Jong L. Szostecki C, Calvo F, Cho- mienne C, Degos L, Puvion E, de The H: The t( 15; 17) translocation alters a nuclear body in a retinoic acid-reversible fashion. EMBO J 13:1073, 1994

16. Dyck J, Maul G, Miller W Jr, Chen J, Kakizuka A, Evans R: A novel macromolecular structure is a target of the promyelocyte- retinoic acid receptor oncoprotein. Cell 76:333, 1994

17. Vahdat L, Maslak P, Miller WH Jr, Eardley A, Heller G, Scheinberg DA, Warrell RR Jr: Early mortality and the retinoic acid syndrome in acute promyelocytic leukemia: Impact of leukocytosis, low-dose chemotherapy, PMLIRAR-a isoform, and CD13 expres- sion in patients treated with all-trans retinoic acid. Blood 84:3843, 1994

18. Levine K, DeBlasio A, Miller Jr WH: Molecular diagnosis and monitoring of acute promyelocytic leukemia treated with retinoic acid. Leukemia 8:S116, 1994 (suppl 1)

19. Bennett J, Catovsky D, Daniel M-T, Flandrin G, Galton D, Gralnick H, Sultan C: Proposed revised criteria for the classification of acute myeloid leukemia. Ann Intern Med 103:620, 1985

20. Paietta E, Andersen J, Gallagher R, Bennett J, Yunis J, Cassi- leth P, Rowe J, Wiernik P The immunophenotype of acute promy- elocytic leukemia (APL): An ECOG study. Leukemia 8:1108, 1994

21, Gallagher RE, Said F, Pua 1, Papenhausen PR, Paietta E, Wiernik PH: Expression of retinoic acid receptor alpha mRNA in human leukemic cells with variable responsiveness to retinoic acid. Leukemia 3:789, 1989

22. Li Y-P, Said F, Gallagher R: Retinoic acid-resistant HL-60 cells exclusively contain mutant retinoic acid receptor-alpha. Blood 83:3298, 1994

23. Bennett J , Catovsky D, Daniel M, Flandrin G, Galton D. Gralnick H, Sultan C: A variant form of hypergranular promyelo- cytic leukaemia (M3). Br J Haernatol 44:169, 1980

For personal use only.on February 12, 2018. by guest www.bloodjournal.orgFrom

APL SUBGROUP WITH REDUCED TRA SENSITIVITY 1547

24. Breitman T, Collins S, Keene B: Terminal differentiation of human promyelocytic leukemic cells in primary culture in response to retinoic acid. Blood 57:1000, 1981

25. Huang M-e, Ye Y-c, Chen S-r, Chai J-r, Lu J-X, Lin Z, Gu L-j, Wang Z-y: Use of all-trans retinoic acid in the treatment of acute promyelocytic leukemia. Blood 72:567, 1988

26. Chomienne C, Ballerini P, Balitrand N, Daniel MT, Fenaux P, Castaigne S, Degos L: All-trans retinoic acid in acute promyelocytic leukemias. 11. In vitro studies: Structure-function relationship. Blood 76:1710, 1990

27. Chen Z-X, Xue Y-Q, Zhang R, Tao R-F, Xia X-M, Li C, Wang W, Zu W-Y, Yao X-Z, Ling B-J: A clinical and experimental study on all-trans retinoic acid-treated acute promyelocytic leukemia patients. Blood 78:1413, 1991

28. Warrell RP Jr, Frankel SR, Miller WH Jr, Steinberg DA, Itri LM, Hittleman W, Vyas R, Andreeff M, Tafuri A, Jakubowski A, Gabrilove J, Gordon M, Dmitrovsky E: Differentiation therapy of acute promyelocytic leukemia with tretinoin (all-trans-retinoic acid). N Engl J Med 324:1385, 1991

29. Delva L, Comic M, Balitrand N, Guidez F, Miclea J-M, Del- mer A, Teillet F, Fenaux P, Castaigne S, Degos L, Chomienne C:

Resistance to all-trans-retinoic acid (ATRA) therapy in relapsing acute promyelocytic leukemia: Study of in vitro sensitivity and cellu- lar retinoic acid binding protein levels in leukemic cells. Blood 82:2175, 1993

30. Degos L: Is acute promyelocytic leukemia a curable disease? Treatment strategy for a long-term survival. Leukemia 8:911, 1994

31. Warrell RP Jr, Maslak P, Eardley A, Geller G, Miller WH Jr, Frankel SR: Treatment of acute promyelocytic leukemia with all- trans retinoic acid: An update of the New York experience. Leuke- mia 8:929, 1994

32. Cunningham I, Gee T, Reich L, Kempin S, Naval A, Clarkson B: Acute promyelocytic leukemia: Treatment results during a decade at Memorial Hospital. Blood 73: 11 16, 1989

33. Hunter R, Karin M: The regulation of transcription by phos- phorylation. Cell 70:375, 1992

34. Hill CS, Treisman R: Transcriptional regulation by extracellu- lar signals: Mechanisms and specificity. Cell 80:199, 1995

35. Gallagher R, Etkind P, Peng S, Qiu L, Rao S, Sagayadan G, Waxman S, Zelent A, Wiemik P: Aberrant PML-RARaPML fusion products in a subset of acute promyelocytic leukemia (APL) patients. Blood 80:300, 1992

For personal use only.on February 12, 2018. by guest www.bloodjournal.orgFrom

1995 86: 1540-1547  

WiernikRE Gallagher, YP Li, S Rao, E Paietta, J Andersen, P Etkind, JM Bennett, MS Tallman and PH acidsubgroup with decreased in vitro responsiveness to all-trans retinoicPML-RAR alpha break/fusion sites in PML exon 6: identification of a Characterization of acute promyelocytic leukemia cases with 

http://www.bloodjournal.org/content/86/4/1540.full.htmlUpdated information and services can be found at:

Articles on similar topics can be found in the following Blood collections

http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requestsInformation about reproducing this article in parts or in its entirety may be found online at:

http://www.bloodjournal.org/site/misc/rights.xhtml#reprintsInformation about ordering reprints may be found online at:

http://www.bloodjournal.org/site/subscriptions/index.xhtmlInformation about subscriptions and ASH membership may be found online at:

  Copyright 2011 by The American Society of Hematology; all rights reserved.Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American

For personal use only.on February 12, 2018. by guest www.bloodjournal.orgFrom