199381: 158-165

JV Melo, DE Gordon, NC Cross and JM Goldman

The ABL-BCR fusion gene is expressed in chronic myeloid leukemia

http://bloodjournal.hematologylibrary.org/content/81/1/158.full.html
Updated information and services can be found at:
Articles on similar topics can be found in the following Blood collections
http://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#repub_requests
Information about reproducing this article in parts or in its entirety may be found online at:
http://bloodjournal.hematologylibrary.org/site/misc/rights.xhtml#reprints
Information about ordering reprints may be found online at:
http://bloodjournal.hematologylibrary.org/site/subscriptions/index.xhtml
Information 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 March 31, 2014. by guest bloodjournal.hematologylibrary.org From For personal use only. on March 31, 2014. by guest bloodjournal.hematologylibrary.org From
The ABL-BCR Fusion Gene Is Expressed in Chronic Myeloid Leukemia
By J unia V. Melo, D.E. Gordon, N.C.P. Cross, and J .M. Goldman
Although the BCR-ABL hybrid gene on chromosome 22q-
plays a pivotal role in t he pathogenesis of chronic myeloid
leukemia (CML), little is known of t he reciprocal chimeric
gene, ABL-BCR, formed on chromosome 9q+. By reverse
transcription/polymerase chain reaction amplification (RT/
PCR) we have detected ABL-BCR mRNA in cells from 31
of 44 BCR-ABL positive CML patients and 3 of 5 CML cell
lines. Of the 34 positive samples, 31 had classical t (9; 22)
(q34; ql l ) translocations; in 3 samples there was no Phil-
adelphia (Ph) and/or 9q+ chromosomes. ABL-BCR expres-
sion consisted of ABL(lb)-BCR mRNA in 26 patients and of
both ABL(lb)-BCR and ABL(la)-BCR mRNA species in 6 pa-
tients. The ABL-BCR transcripts encoded one or, more
HRONIC MYELOID leukemia (CML) is characterized
C cytogenetically by the presence of the Philadelphia (Ph)
chromosome, which originates from the reciprocal translo-
cation t(9;22) (q34;qll). In the formation of the Ph chro-
mosome, the bulk of the ABL protooncogene is translocated
from chromosome 9 onto the BCR gene in chromosome 22.
This gives rise to a novel chimeric BCR-ABL gene, which
encodes a 210-Kd (P210) fusion protein with prominent ty-
rosine kinase activity and transforming ability.
The product of the normal BCR gene is a 160-Kd (P160BCR)
cytosolic phosphoprotein, whose physiological role is not
clearly defined. It was shown to form cytoplasmic complexes
with P2 in Ph-positive CML cells, as well as with a
53-Kd protein of unknown function in both Ph-positive and
Ph-negative cell lines. Sequences encoded by the first exon
of BCR are responsible for the P160BCR serine/threonine ki-
nase a~ti vi ty.~ These sequences overlap the src-homology 2
(SH2)-binding regions of the BCR gene that are essential for
the activation of the ABL tyrosine kinase and the transform-
ing potential of the chimeric BCR-ABL In its
central segment, BCR has some homology to the dbl oncogene
and the yeast CDC24 gene.6 The product of the latter is in-
volved in the control of cell division after DNA replication.
The C-terminus of BCR has recently been shown to have a
GTPase-activating protein (GAP) activity for p2 1 , a mem-
ber of the RAS family of small GTP-binding proteins.*
In the t(9;22) translocation, the p2 1 GAP domain of
BCR is absent from the BCR-ABL chimeric protein. The 3
end of the BCR gene containing the coding sequence for this
domain is, in turn, fused to the 5 end of ABL on chromosome
9. The fate of the 9q+chromosome, the other partner in
From the MRC/LRF Leukaemia Unit, Department of Haematol-
ogy, Royal Postgraduate Medical School, London, UK.
Submitted July I , 1992; accepted September 3, 1992.
Address reprint requests to Junia V. Melo, MD, PhD, Department
of Haematology, Royal Postgraduate Medical School, Ducane Rd,
London WI 2 ONN, UK.
The publication costs of this article were defiayed in part by page
charge payment. This article must therefore be hereby marked
advertisement in accordance with 18 U.S.C. section I734 solely to
indicate this fact.
0 I993 by The American Society of Hematology.
0006-49 71/93/81 01 -001 5$3.00/0
rarely, both of the t wo potential junctions, designated ABL-
b3 and ABL-b4. which differed in size by 75 bp. In 2 pa-
tients, the BCR exon b3 was not present in either t he BCR-
ABL or the corresponding ABL-BCR transcript, whereas in
5 patients exon b3 was present in both transcripts. Direct
sequencing of PCR fragments representing t he full-length
coding sequence of ABL-BCR cDNAs type l b-b3,l a-b3, Ib-
b4, and la-b4 showed an open reading frame predicted t o
encode fusion proteins of 370 to 414 amino-acids. If an
ABL-BCR gene product is produced in CML cells, it may be
relevant as a mechanism for deregulating t he GTPase ac-
tivating protein (GAP) function of BCR.
0 1993 by The American Society of Hematology.
the reciprocal t(9;22), is not known. Whereas most of the
presently available data suggest that the P2 10 fusion protein
encoded by the BCR-ABL hybrid gene is involved in the
pathogenesis of CML;-l2 it is likely that other genetic changes
are also necessary for defining the full leukemic phenotype.
The expression ofthe 5ABL-3BCR hybrid gene in this trans-
location has not so far been studied, but it may have func-
tional consequences if it leads to abnormal activation of BCR-
GAP.
Weshow here that the reciprocal ABL-BCR gene is tran-
scriptionally active in over two thirds of Ph-positive CML
patients, and that translation of its cognate chimeric mRNA
into an ABL-BCR fusion protein is compatible with its se-
quence.
MATERIALS AND METHODS
Patients and cell lines. Cells from a total of 44 CML patients
were studied: 20 in chronic phase (CP) and 24 in blast crisis (BC).
Among the latter, 15 were myeloid BC, 4 lymphoid, 1 mixed myeloid/
lymphoid, and 4 of unknown phenotype. Thirty-eight patients showed
characteristic Ph and 9q+chromosomes, 1 had an atypical 22q-
without a 9q+, and 5 were Ph-negative. All patients had clonal BCR
gene rearrangement by Southern blot analysis.
Five BCR-ABL-positive cell lines were also investigated. These
were well characterized lines from patients with CML in blast crisis:
K562, KCL-22, KYO-1, BV173,I3 and LAMA-84.I4 The HL60 pro-
myelocytic cell line was used as a negative control for BCR-ABL
expression in all tests.
Amplifications
of specific sequences on the ABL, BCR, BCR-ABL, and ABL-BCR
genes were performed by reverse transcription (RT) of cDNA, fol-
lowed by PCR (RT/PCR) by standard methods. Briefly, PB white
blood cells (WBC) were obtained by dextran sedimentation or by red
cell lysis of centrifuged buffy coat ~reparati0ns.l~ The patients WBC
and freshly explanted cells from lines in culture were washed twice
in phosphate buffered saline and processed for RNA extraction by
the guanidinium thiocyanate/CsCl gradient method.16 The RNA was
reverse transcribed into cDNA with Mo-MuLV reverse transcriptase
(GIBCO-BRL, Gaithersburg, MD) using random hexamers and, in
some samples, oligo-dT primers. PCR amplification of cDNA was
performed as described elsewhere. Precautions towards eliminating
the possibility of false PCR results were based on the recommenda-
tions by Kwok and Higuchi. In brief: (1) cells, RNA, and cDNA
preparations were always handled in a room separate from that spe-
cifically dedicated to the analysis of PCR products; (2) different sets
of pipettes were dedicated to sample preparation and PCR product
Polymerase chain reaction (PCR) amplijication.
158 Blood, Vol 81, No 1 (J anuary 1). 1993: pp 158-165
For personal use only. on March 31, 2014. by guest bloodjournal.hematologylibrary.org From
ABL-BCR FUSION GENE IN CML 159
handling, and plugged tips (aerosol resistant) wereused in all steps;
(3) all reagents wereprepared under sterile conditions in a laminar-
flow cabinet, and stored as single-use aliquots; and (4) each PCR
experiment included 6 to 8 test (CML) cDNA samples plus 2 known
negative controls: BCR-ABL- and ABL-BCR-negativecDNA from
HL60 cells, and a H20-blank (ie, no cDNA). In no instance were
ABL-BCR products detected in either negativecontrol. Furthermore,
test samples showing no ABL-BCR amplification alongside samples
that did yield an ABL-BCR PCR product werealways found in any
given experiment, reinforcing the validity of the positive results. Five
microliters fromeach PCR was electrophoresed through ethidium
bromide stained 1% to 2% agarose mini-gels, visualized, and pho-
tographed under UV light. Samples that showed no ABLBCR prod-
uct weresubmitted to a second round of amplification with nested
primers and 1 pL of the original PCR products as template.
The
sequences of the synthetic oligonucleotide primers used in this in-
vestigation are shown on Table 1. Some of the primers weredesigned
to contain natural or forced restriction enzyme sites at their 5 ends
to facilitate future cloning of the PCR fragments into phageor plasmid
vectors. The size of PCR fragments amplified with each primer pair
is shown in Table 2.
Southern hybridization. Electrophoresed PCR products from
somesamples weretransferred fromagarosegels to nylon membranes
by Southern blotting, and tested for hybridization to a synthetic oli-
gonucleotide probe (primer BS+) 5-labeled with Y-~~P-ATP. Hybrid-
ization and high-stringency washings werecamed out at the appro-
priate discriminating temperatures for the oligonucleotide as described
elsewhere.
Direct sequencing of PCR products was performed
by the linear amplification sequencing method and/or by the Taq
cyclesequencing method [USB, Cleveland, OH]. PCR products were
used directly as templates on an estimated basis of 100 fmol ss-DNA
per baseper primer. Denaturing sequencing gels wereprepared, elec-
trophoresed, and autoradiographed by conventional techniques.
RESULTS
Primers for PCR, Southern hybridization, and sequencing.
Sequencing.
BCR-ABL gene. BCR-ABL amplification of cDNA with
primers B2+ and CA3- yielded fragments 385 and 465 bp
Tabl e 1. Syntheti c Oligonucleotide Primers Used in This Study
Primer Sequence ( 5 +. 3)
Gene
Locat i oPZ6
Aa+
Ab+
PAa+
PAb+
Jc-
CA3-
81 +
B2+
B3+
B4+
B7+
C4
84-
c5-
G-
PB-
ec-
TTGGAGATCTCCCTGAAG
CTTCTGGAAAGGGGTACCTATTA
tacggaattcATGTTGGAGATCTCCCTGAA
CGCTGAgaaTTCTGGAAGATCTTGAA
GGAGTGTTTCTCCAGACTGTTG
TGTTGACTGGCGTGATGTAGTTGCTTGG
GAGCGTGCAGAGTGGGGAGGGAGAACATCCGG
TTCAGAAGCTTCTCCCTGACAT
TCTGAATGTCATCGTCCACTCAGCCA
ttcaaATCTGTACTGCACCCTGGAGCT
GGTCgAaTTCAACAGCAGGGAGTTCA
attagatCTGAAGCTGTACTTCCGTGA
acgtgaaTTCCAGTTTGGCTCAGCTGT
tcaaggaTCCACGCACTGGCGCACGAT
ataggaTCCTTTGCAACCGGGTCTGAA
gatgggaTCCAGCTGCAGGAGTA
TCAGGAAGGaTCCCGCTCTACGGAT
ABL (la)
ABL (Ib)
ABL (la)
ABL (Ib)
ABL ( I l l )
ABL ( I l l )
BCR ( bl )
BCR (b2)
BCR (b3)
BCR (b4)
BCR (b7)
BCR (c4)
BCR (b4)
BCR (c3)
BCR (c5)
BCR (c7)
BCR ( 3 end)
Lower case indicates nucleotides that are not homologous t o the
cDNA sequence, but incorporate restriction enzyme recognition sites.
Tabl e 2. PCR Amplifications of ABL, BCR, BCR-ABL.
and ABL-BCR cDNA
PCR Product
Primers Gene (length in bp)
Aa+ - Jc- ABL (from la) 374
Ab+ - Jc- ABL (from Ib) 462
Bl + - G- BCR 1,208
385 (b2a2)
456 (b3a2)
PAa+ - G- ABL (la)-BCR 1,072 (la-b4)
1,147 (la-b3)
B2+ ++CA3- BCR-ABL
Ab+ ++ G- ABL (Ib)-BCR 1,148 (lb-b4)
1,223 (lb-b3)
PAa+ ++PB- ABL (la)-BCR 1,208 (la-b4)
1,283 (la-b3)
PAb+ - PB- ABL (Ib)-BCR 1.31 6 (lb-b4)
1,391 (lb-b3)
PAa+ - 84- * ABL (la)-BCR 189 (la-b4)
264 (la-b3)
PAb+ ++84- ABL (b-BCR 297 (lb-b4)
372 (lb-b3)
84 ++Bc- BCR (in ABL-BCR) 456
B7+ - C5- BCR (in ABL-BCR) 398
c4+ ++PB- * BCR (in ABL-BCR) 527
These were primers used for direct sequencing of overlapping seg-
ments of the template ABL-BCR amplified cDNAs as PCRs PAa+ - PB-
and PAb+ - PB-.
long, representing the b2a2 and b3a2 type transcripts re-
spectively (Fig 1A).
The types of BCR-ABL transcript expressed i n the 44 pa-
tients and the five CML cell lines were as follows. I n the CP
group, 45% of patients expressed only b2a2 and 50% only
b3a2 BCR-ABL transcripts. A relatively similar distribution
for b2a2 and b3a2 patients was observed in the BC group,
which includes the five cell lines (59% and 38%, respectively).
One CP and one BC patient expressed both b2a2 and b3a2
transcripts.
Expression of the
2 major alternative transcripts from the normal ABL allele
was observed in all the 44 patients. This was shown by RT/
PCR amplification of the sequence spanning exon I a to exon
111(PCR PAa+ * Jc-) and, likewise, the sequence from exon
I b to exon I11 (PCR Ab c-$ Jc-). Of the five cell lines studied,
only LAMA-84 and BV113 did not express either ABL tran-
script, which agrees with the absence of a normal chromosome
9 i n these
The normal BCR allele was also found to be expressed in
all patients and cell lines, as assessed by PCR amplification
of a 1,208-bp long fragment spanning the major breakpoint
cluster region (M-bcr), and extending downstream of exon
~ 7 . ~ ~ 3 ~ ~ This fragment includes the BCR-GAP coding region
in the normal
The formation of an ABL-BCR fusion
gene, the reciprocal product of the BCR-ABL translocation,
should theoretically yield different transcripts, depending on
the positions of the breakpoints in both ABL and BCR (Fig
2). If the breakpoint in ABL occurs upstream of exon Ib, no
Normal ABL and normal BCR genes.
ABL-BCR gene.
For personal use only. on March 31, 2014. by guest bloodjournal.hematologylibrary.org From
160
M 1 2 3 4 5 6 7 8 9 1 0 1 1 M
MELO ET AL
ABL-BCR transcript is formed. I f the breakpoint is between
exons Ib and la, only transcripts originating fromthe exon
Ib promoter are possible (Ib-BCR); and if between exons la
and 11, two RNA species can betranscribed fromthe inde-
pendent promoters in exons Ib and la (Ib-BCR and la-BCR).
The 49 samples weretested for expression of ABL-BCR
transcripts by RT/PCR amplification using sense primers on
ABL exons Ib (Ah') and la (PAa'). and anti-sense primers
on BCR 3' end (G- or PB- ) . A total of 34 out of 49 patients
(69%). including 2 Ph-negative patients and the patient with-
out a 9q+chromosome, showed ABL-BCR amplification of
the Ib-BCR type and 6 of these 34also expressed la-BCR.
No case expressed la-BCR alone (Table 3). Likewise, abnor-
mally large ABL-BCR fragments containing both exons Ib
and la were never found. In 7 of the 34 samples the ABL-
BCR products wereonly detected by nested PCR. In all the
others, the level of ABL-BCR transcripts seemed comparable
to that of ABL, BCR, and BCR-ABL amplified products, as
estimated within the limitations of a standard, nonquanti-
tative PCR assay. Among the I5 ABL-BCR negative samples,
4 (3 patients and the cell line K562) were Ph-negative, al-
though BCR-rearranged and BCR-ABL positive.
A
+ b3a2
+ b2a2
B
eb-b3 lb-b4
Fig 1. (A) BCR-ABL and (B)
ABL-BCR PCR products in 11
representative samples, as de-
tected on ethidium bromide-
stained agarose gels. BCR-ABL
bands represent fragments of
385 bp (b2a2) andlor 465 bp
(b3a2). corresponding to ampli-
fications with primers 82' and
CA3 . ABL-BCR products span
sequences between primersAb'
(in ABLexon Ib) and G (in BCR),
representing fragments 1,223
bp (lb-b3) and/or 1,148 bp (Ib-
b4) long. Lanes 1 to 10 are CML
patient samples; lane 11 is the
HL60 cell line (negative control).
M, DNA molecular weight
marker (pEMBL digested wi th
Tw I ) .
Like RCR-ABL. the ABL-BCR transcripts are predicted
to vary in length, depending on whether BCR exons b3 or
b4 are joined to the S'ABL. This was confirmed by Ib-BCR
products 1,148 bp and/or 1,223 bp long when primers
Ah' - G- were used for the PCR amplification (Fig IB).
The same 75 bp difference in the length of the PCR prod-
ucts was observed in la-BCR amplifications with primers
PAa' - G-.
Therefore, the ABL-BCR expressing cases fell into 3 cat-
egories (Table 4): 12 showing b b 4 junction, 19 with bb3
junction, and 3 cases with double transcripts (I bb4 and Ib-
b3). In the 6 patients who also showed la-BCR transcription,
the junction was of Ia-b4 type in 3 and Ia-b3 in 3. The pres-
ence or absence of BCR exon b3 in each transcript was con-
firmed bySouthem hybridization of the ABL-BCR fragments
with a 26-mer oligonucleotide (primer B3') spanning a se-
quence specific for exon b3 (Fig 3) and, in some cases, by
direct sequencing of the junction region in each PCR frag-
ment.
Overall, the proportion of ABLBCR-expressing patients
who showed only b2a2 (56%), only b3a2 (38%), and both
b2a2 and b3a2 (6%) BCR-ABL transcripts matched the pro-
For personal use only. on March 31, 2014. by guest bloodjournal.hematologylibrary.org From
ABL-BCR FUSION GENE IN CML 161
Fig 2. Schematic represen-
tation of the ABL, BCR, BCR-
ABL, and ABL-BCR genes. Ar-
rows indicate the most frequent
regions for breakpoints in the
ABL and the BCR genes. The
possible ABL-BCR transcripts
arising from the different break-
points in BCR are shown under-
neath the corresponding BCR-
ABL transcripts type b2a2 and
b3a2, respectively.
BCR-ABL
b2a2
E2 1
0
lb-b3 L I m"I--" . ---=TI- ~ 1
I b W
ABL-BCR
la-b3
la Z E2 1
XI
U W
E l W N
BCR-ABL I 1 IJ LU - 1
b3a2
E2 1
Ib g
lb-b4
ABL-BCR
la X E2 1
la-b4 - ~~~ ~
portion of patients in each BCR-ABL category in the whole
series. The frequency of ABL-BCR expression within each
BCR-ABL major group was 73% and 62%) in the b2a2 and
the b3a2 groups. respectively.
When the expression of the BCR-ABL and the ABL-BCR
fusion genes was compared in each individual sample (Table
5). it was found that, in 27 patients (79%). the junction type
in ABL-BCR was exactly reciprocal to the junction in BCR-
ABL. Thus, from the 19 b2a2-positive samples, 15 expressed
an ABL-b3 type of junction, and 2 expressed both ABL-b3
and ABL-b4 transcripts. Similarly, 8 out of I3 patients with
single b3a2 and 2 patients with both b2a2 and b3a2 transcripts
from BCR-ABL showed ABL-BCR expression of the ABL-
b4 type, as expected. However. in 7 samples the junction
type in the ABL-BCR transcripts did not correlate with the
reciprocal BCR-ABL products: these were 2 b2a2 patients
who expressed only ABL-W transcripts, and 5 b3a2 patients
in whom ABL-b3 type of transcripts were found either alone
(4 samples) or together with ABL-W transcripts ( 1 sample).
The junction region of BCR-ABL and ABL-BCR products
Table 3. ABL-BCR Transcripts Found in CML Patients
Stage of Disease
(no. of cases) None Ib-BCR Ib-BCR and la-BCR
~
Chronic phase (20) 5 (25%) 13 (65%) 2 (10%)
Blast crisis (29)' 10 (34%) 15 (52%)t 4 (14%)
TOTAL (49) 15 (31%) 28 (57%) 6 (12%)
Including the 5 CML cell lines.
t Including the cell lines KCL-22. KYO-1, and BV173.
from this group of patients was sequenced (primers PAa'
R4- ) and results confirmed that, in 2 cases, BCR exon b3
was not expressed in either BCR-ABL or ABL-BCR, whereas,
in 5 cases. this exon was present in both fusion-gene products.
The same result was obtained by Southern hybridization of
BCR-ABL and ABL-BCR products with the oligonucleotide
probe R3' (Fig 3). In 5 ofthe 7 cases it was possible to repeat
the BCR-ABL and ABL-BCR amplifications in duplicate
blood samples obtained at different times, and/or in duplicate
RNA/cDNA preparations from the original blood samples.
The initial results were always reproducible. In 2 patients
duplicate samples were not available for analysis.
PCR amplified
ABL-BCR fragments from cDNA of 2 samples were se-
quenced in overlapping segments. using 4 pairs of sense and
antisense primers (Table 2). The full-length PCR products
represented ABL-BCR type I bb3 and Ia-b3 from I patient.
and I bb4 and la-b4 from another. Each fragment started 60
bp (Ib) or IO bp (la) upstream of ABL's initiation ATG. and
ended 86 bp downstream the stop codon for PI 60RCR.
Coding seqircvce qj'clie ABL-BCR gene.
Table 4. J unction Type of the ABL(Ib)-BCR Transcripts
in the Positive Cases
Stage of Disease
(no. of cases) lb-b3 lb-b4 lbb3 and lbb4
Chronic phase (1 5) 7 (47%) 6 (40%) 2 (13%)
Blast crisis (1 9)' 12 (63%)' 6 (32%) 1(5%)
TOTAL (34) 19 (56%) 12 (35%) 3 (9%)
Including the cell lines KCL-22, KYO-1. and BV173.
For personal use only. on March 31, 2014. by guest bloodjournal.hematologylibrary.org From
162 MELO ET AL
In the 4 products. the sequences matched exactly ABL
exon or la. which joined in phase BCR exons b3 or
b4, maintaining the open reading frame (ORF) up to the
normal BCR stopcodon. The ORFs ofthe ARL-BCR hybrid
cDNAs code for predicted fusion proteins of 4 14 amino-acids
( AA) (Ihb3). 395 AA (la-b3). 389 AA (lkb4). and 370 AA
(Ia-b4).
DISCUSSION
I n Ph-positive CML. the coding sequences of two genes,
ABL and BCR. are disrupted as a result of the reciprocal
exchange between chromosomes 9 and 22. and two hybrid
genes are formed. BCR-ABL (on 22q-. Ph) and ABL-BCR
(on 9q+). Expression of the BCR-ABL gene was shown i n
all cases of Ph-positive CML when the sensitive method
Table 5. BCR-ABL Transcriot Versus ABLfIbl-BCR TranscriDt
ABL(lb)-BCR
BCR-ABL
(no. of cases) lb-b3 lb-b4 lbb3 and lbb4
b2a2 (1 9) 15 2 2
b3a2 (1 3) 4 8 1
b2a2 and b3a2 (2) 0 2 0
Cases in which the ABL-BCR transcripts are not reciprocal to the
BCR-ABL transcripts.
of RT/PCR amplification was used. However, it was
not known whether the reciprocal ABL-BCR hybrid gene
is likewise functionally active i n this disease. This possi-
bility was raised previously i n a single report when a 3.5
M 1 2 3 4 5 6 7 8 9 1 0 1 1 M
.)
2 3 4 5 6 7 8 9 10 11 M
b3a2
+lb-b3
Fig 3. Southern-blot hybrid-
ization with oligonucleotide
(primer) B3 of (A) BCR-ABL and
the same representative sam-
ples shown in Fig 1. The oligo-
nucleotide probe sequence is
specific for BCR exon b3. Note
that in samples 2 and 8 both
BCR-ABL and ABL-BCR tran-
scripts hybridize to this probe,
indicating the presence of BCR
exon b3 in the two hybrid genes.
Conversely, neither BCR-ABL
nor ABL-BCR transcripts from
sample 9 hybridize to the probe,
which confirms the absence of
exon b3 in both genes.
(B) ABL-BCR PCR products Of
For personal use only. on March 31, 2014. by guest bloodjournal.hematologylibrary.org From
ABL-BCR FUSION GENE IN CML 163
kb transcript, presumed to be a SABL-3BCR fusion mes-
sage, was detected in Northern blots of the BV173 cell
line.
In the present study, weshow that ABL-BCR expression
can be detected by RT/PCR in WBC from approximately
70% of CML patients. It is noteworthy that, among the ABL-
BCR positive patients, 2 were Ph-negative and 1 had no 9q+
chromosome, showing that, like BCR-ABL, the ABL-BCR
gene rearrangement can also take place in the absence of
cytogenetic evidence of a t(9;22).
In one third of our cases, no ABL-BCR expression was
detected, in spite of repeated PCR tests with several combi-
nations of primers. This group contains 4 of the Ph-nega-
tive, BCR-ABL positive samples that may represent com-
plex and not reciprocal chromosome trans location^.^^-^'
The remaining 14 samples (26% of the Ph-positive CML)
presumably result from translocation breakpoints in chro-
mosome 9 upstream of ABL exon Ib, or from deletions
in BCR sequences 3 to the chromosome 22 breakp~i nt.~~
In neither case would an ABL-BCR hybrid message be
present.
The higher proportion (82%) of patients expressing only
ABL-BCR transcripts of the Ib-BCR type, compared with
those (1 8%) with transcripts initiated from both exons Ib
and Ia (Ib-BCR and Ia-BCR) is probably a reflection of
the frequency of breakpoints between exons Ib and Ia, and
Ia and 11. Because the distance between exons Ib and Ia is
greater than 200 kb,24 approximately I O times that between
exons Ia and I1 ( 1 9 kb),34 the probability of a break oc-
curring within the first large ABL intron might be 10-fold
greater than within the second i ntr ~n. ~~- ~ The fact that a
single transcript carrying both exons Ib and Ia was never
found shows that the normal mechanism of alternative
splicing of these two exons3 is maintained in the ABL-
BCR hybrid gene.
The overall similar distribution of ABL-BCR-positive pa-
tients among the b2a2 and b3a2 BCR-ABL categories suggests
that there is no simple correlation between the breakpoint
sites in ABL and in BCR. On the other hand, the structure
of the BCR moiety of the BCR-ABL and ABL-BCR tran-
scripts from 7 patients showed unexpected patterns. In 2 of
these, neither BCR-ABL nor ABL-BCR transcripts include
BCR exon b3. This means that the breakpoint may be in
exon b3 itself, or that this exon was included in deletions at
the breakpoint site.39 Another possibility is that b3 was spliced
out in the mature mRNA from whichever of the two chimeric
primary transcripts that retained it. Precedents for alternative
splicing of exon b3 in BCR-ABL are well established in those
patients who express both b2a2 and b3a2 transcript^^^,^^,^^
and, in the present study, by the 3 patients with both Ib-b3
and Ib-b4 ABL-BCR. More surprising are the 5 patients in
this study in whom both BCR-ABL and ABL-BCR transcripts
contain BCR exon b3. In 3 of these (3 BC), the karyotype
showed single Ph and 9q+chromosomes, in 1 (CP), the 22q-
was longer than a typical Ph and there was no 9q+, and in
the fifth (CP),. additional i(Ph) chromosomes probably arose
from duplication of the original Ph. Therefore in these cases,
there is no evidence that the BCR-ABL and the ABL-BCR
hybrid genes are not part of the same clone, which implies
that BCR exon b3 alone was duplicated in the t(9;22).
Whether this b3 duplication took place before or during the
translocation is not known. The breakpoint regions of both
BCR-ABL and ABL-BCR in these 5 cases are being cloned
for further studies.
The nucleotide sequence of the 4 different ABL-BCR
hybrid cDNAs found in our series shows that in each case
the SABL-3BCR junction is in phase as expected. There-
fore, ABL-BCR fusion proteins of about 390 AA could be
translated, the exact size of each varying according to the
exon contribution of ABL and BCR to the transcript. The
fact that ABL-BCR amplifications were obtained from
oligo-dT primed cDNAs indicates the presence of poly-
(A) on these transcripts and suggests that translatable RNA
is produced. Investigations on the presence of such ABL-
BCR proteins in cells from patients with CML are in pro-
gress.
The significance of ABL-BCR expression in CML is
still unclear. It seems unlikely that the hybrid ABL-BCR
gene has any primary oncogenic role, because it is not
present in at least one third of the CML patients. Fur-
thermore, the fact that ABL-BCR is expressed in CP as
well as in BC of CML argues against a role in causing
disease progression. On the other hand, ABL-BCR
expression could well be associated with specific clinical
and/or hematological features in subsets of CML patients,
which may reflect prognosis and response to treatment.
Data from our present series are still insufficient to address
this question.
If a functional ABL-BCR fusion protein is indeed pro-
duced, as predicted from the cDNA coding sequences, it
would contain a GAP-BCR domain in its C-terminus linked
to an N-terminal ABL sequence. Such an arrangement could
alter the rucGAP function of BCR leading to either con-
stitutive activation or inactivation of the rucGAP activity.
Moreover, the Ib-BCR-encoded fusion protein, like the ABL
type I1P145,3,41,42 would have a myristoylation site at its
N-terminus and, therefore, unlike p 1 60BCR, it could become
associated with the cell membrane. Although mRNA
expression from the normal BCR allele can be detected in
CML cells,43 it is not clear whether the level of P160BCR
production is the same as in normal leukocytes. Coexpres-
sion of an abnormal ABL-BCR gene product could result
in competition for the same target protein, ruc, and imbal-
ance in the rate of ruc activation. The biologic effects of a
deregulated BCR are unknown. However, because the ruc
proteins display relative myeloid speci fi ~i ty~~ and are in-
volved in activation of the NADPH oxidase system of neu-
trophi l ~,~~ it is tempting to speculate that an ABL-BCR pro-
tein with altered GAP activity might have a role in
granulocyte functional defects.
ACKNOWLEDGMENT
We thank J ulie Bungey for reviewing the karyotypes of patients
and cell lines, and Dr Alan Hall (Chester Beatty Institute for Cancer
Research, London) for valuable discussion.
For personal use only. on March 31, 2014. by guest bloodjournal.hematologylibrary.org From
164 MELO ET AL
REFERENCES
I . Campbell ML, Li W, Arlinghaus RB: P210 BCR-ABL is com-
plexed to PI60 BCR and ph-P53 proteins in K562 cells. Oncogene
5:773, 1990
2. Li W, Dreazen 0, Kloetzer W, Gale RP, Arlinghaus RB: Char-
acterization of bcr gene products in hematopoietic cells. Oncogene
4:127, 1989
3. Maru Y, Witte ON: The BCR gene encodes a novel serine/
threonine kinase activity within a single exon. Cell 67:459, 1991
4. Muller AJ , Young J C, Pendergast AM, Pondel M, Landau NR,
Littman DR, Witte ON BCR first exon sequences specifically activate
the BCR/ABL tyrosine kinase oncogene of Philadelphia chromosome-
positive human leukemias. Mol Cell Biol 1 1: 1785, I99 1
5. McWhirter J R, Wang J YJ : Activation of tyrosine kinase and
microfilament-binding functions of c-ab1 by bcr sequences in bcr/abl
fusion proteins. Mol Cell Biol 11:1553, 1991
6. Ron D, Zannini M, Levis M, Wickner RB, Hunt LT, Graziani
G, Tronick SR, Aaronson SA, Eva A: A region of protodbl essential
for its transforming activity shows sequence similarity to a yeast cell
cycle gene, CDC24, and the human breakpoint cluster region, bcr.
New Biol 3:372, 1991
7. Ohya Y, Miyamoto S, Oshumi Y, Anraku Y Calcium sensitive
c/s4 mutant of Saccharomyces cerevisiae with a defect in bud for-
mation. J Bacteriol 165:28, 1986
8. Diekmann D, Brill S, Garrett MD, Totty N, Hsuan J , Monfries
C, Hall C, LimL, Hall A: BCR encodes a GTPase-activating protein
for p21mc. Nature 351:400, 1991
9. Daley GQ, van Etten RA, Baltimore D: Induction of chronic
myelogenous leukemia in mice by the P210kr'ab' gene of the Phila-
delphia chromosome. Science 247:824, I990
IO. Elefanty AG, Hariharan IK, Cory S bcr-ab/, the hallmark of
chronic myeloid leukaemia in man, induces multiple haemopoietic
neoplasms in mice. EMBO J 9:1069, 1990
1 I . Daley GQ, van Etten RA, Baltimore D: Blast crisis in a murine
model of chronic myelogenous leukemia. Proc Natl Acad Sci USA
88:11335, 1991
12. Gishizky M, Witte ON: Initiation of deregulated growth of
multipotent progenitor cells by bcr/abl in vitro. Science 256336,
1992
13. Keating A: Ph positive CML cell lines. Baillieres Clin Haematol
1:1021, 1987
14. Seigneurin D, Champelovier P, Mouchiroud G, Berthier R,
Leroux D, Prenant M, McGregor J , Starck J , Morle F, Micouin C,
Pietrantuono A, Kolodie L: Human chronic myeloid leukemic cell
line with positive Philadelphia chromosome exhibits megakaryocytic
and erythroid characteristics. Exp Hematol 15322, 1987
15. Roos D, Loos J A: Changes in the carbohydrate metabolism
of mitogenically stimulated human peripheral lymphocytes. I. Stim-
ulation by phytohaemagglutinin. BiochemBiophys Acta 222565,
1970
16. Chirgwin J M, Przybyla ARE, MacDonald RJ , Rutter WJ: Iso-
lation of biologically active ribonucleic acid from sources enriched
in ribonuclease. Biochemistry 18:5294, 1979
17. Melo J V, Goldman J M: Specific point-mutations that activate
v-ab/ are not found in Philadelphia-negative chronic myeloid leu-
kemia, Philadelphia-negative acute lymphoblastic leukemia or blast
transformation of chronic myeloid leukemia. Leukemia 6:786, 1992
18. Kwok S, Higuchi R Avoiding false positives with PCR. Nature
339:237, 1989
19. Cross NCP, de Franchis R, Sebastio G, Dazzo C, Tolan DR,
Gregori C, Odievre M, Vidailhet M, Romano V, Mascali G, Romano
C, Musumeci S, Steinmann B, Gitzelmann R, Cox TM: Molecular
analysis of aldolase B genes in hereditary fructose intolerance. Lancet
335:306, 1990
20. Craxton M: Linear amplification sequencing: A powerful
method for sequencing DNA. Methods: A companion to Methods
in Enzymology 3:20, I99 1
2 1. Westbrook CA, Rubin CM, Carrino J J , Le Beau MM, Bernards
A, Rowley J D Long-range mapping of the Philadelphia chromosome
by pulsed-field gel electrophoresis. Blood 71:697, 1988
22. Heisterkamp N, Stam K, Groffen J , de Klein A, Grosveld G:
Structural organization of the bcr gene and its role in the Ph' trans-
location. Nature 315:758, 1985
23. Lifshitz B, Fainstein E, Marcelle C, Shtivelman E, Amson R,
Gale RP, Canaani E: bcr genes and transcripts. Oncogene 2:113,
1988
24. Bernards A, Rubin CM, Westbrook CA, Paskind M, Baltimore
D The first intron in the human c-ab1 gene is at least 200 kilobases
long and is a target for translocations in chronic myelogenous leu-
kemia. Mol Cell Biol 7:3231, 1987
25. Fainstein E, Einat M, Gokkel E, Marcelle C, Croce CM, Gale
RP, Canaani E: Nucleotide sequence analysis of human ab/ and bcr/
ab/ cDNAs. Oncogene 4: 1477, 1989
26. Morgan GJ , Hernandez A, Chan LC, Hughes T, Martiat P,
Wiedemann LM: The role of alternative splicing patterns of BCR/
ABL transcripts in the generation of the blast crisis of chronic myeloid
leukaemia. Br J Haematol 76:33, 1990
27. Dobrovic A, Morley AA, Seshadri R, J anuszewicz EH:
Molecular diagnosis of Philadelphia-negative CML using the
polymerase chain reaction and DNA analysis: Clinical features
and course of M-bcr negative and M-bcr positive CML. Leukemia
5:187, 1991
28. Dhingra K, Talpaz M, Kantarjian H, Ku S, Rothberg J , Gut-
terman J U, Kurzrock R: Appearance of acute leukemia-associated
PI 90BCR-ABL in chronic myelogenous leukemia may correlate with
disease progression. Leukemia 5: 19 I , 199 1
29. de Braekeleer M: Variant Philadelphia translocations in chronic
myeloid leukemia. Cytogenet Cell Genet 44:215, 1987
30. Moms CM, Heisterkamp N, Kennedy MA, Fitzgerald PH,
Groffen J : Ph-negative chronic myeloid leukemia: Molecular analysis
of ABL insertion into M-BCR on chromosome 22. Blood 76: 18 12,
1990
3 1. Monis CM, Kennedy M, Heisterkamp N, Columbano-Green
L, Romeril K, Groffen J , Fitzgerald P A complex chromosome rear-
rangement forms the BCR-ABL fusion gene in leukemic cells with
a normal karyotype. Genes Chrom Cancer 3:263, 1991
32. Moms CM, Heisterkamp N, Groffen J , Fitzgerald PH: Entire
ABL gene is joined with 5'BCR in some patients with Philadelphia
positive leukemia. Blood 78: 1078, 199 1
33. Popenoe DW, Schaefer-Rego K, Mean J G, Bank A, Leibowitz
D: Frequent and extensive deletion during the 9,22 translocation in
CML. Blood 68:1123, 1986
34. Shtivelman E, Lifshitz B, Gale RP, Canaani E: Fused transcript
of ab1 and bcr genes in chronic myelogenous leukaemia. Nature 3 15:
550, 1985
35. Groffen J , Stephenson J R, Heisterkamp N, de Klein A, Bar-
tram CR, Grosveld G: Philadelphia chromosomal breakpoints are
clustered within a limited region, bcr, on chromosome 22. Cell 36:
93, 1984
36. Heisterkamp N, Stephenson J R, Groffen J , Hansen PF, de
Klein A, BartramCR, Grosveld G: Localization of the c-ab/ oncogene
adjacent to a translocation breakpoint in chronic myelocytic leukae-
mia. Nature 306:239, 1983
For personal use only. on March 31, 2014. by guest bloodjournal.hematologylibrary.org From
ABL-BCR FUSION GENE IN CML 165
37. Leibowitz D, Schaefer-Rego K, Popenoe DW, Mears J G, Bank
A: Variable breakpoints on the Philadelphia chromosome in chronic
myelogenous leukemia. Blood 66:243, 1985
38. Shtivelman E, Lifshitz B, Gale RP, Roe BA, Canaani E: Al-
ternative splicing of RNAs transcribed from the human ab/ gene and
from bcr-ab/ fused gene. Cell 47:277, 1986
39. Mills KI, Sprout AM, Leibowitz D, Bumett AK: Mapping of
breakpoints, and relationship to BCR-ABL RNA expression, in Phil-
adelphia-chromosome-positive chronic myeloid leukaemia patients
with a breakpoint around exon 14 (b3) of the BCR gene. Leukemia
5:937, 1991
40. Lee MS, Le Maistre A, Kantarjian HM, Talpaz M, Freireich
ET, Trujillo J M, Stass S A Detection of two alternative bcr/ab/ mRNA
junctions and minimal residual disease in Philadelphia chromosome
positive chronic myelogenous leukemia by polymerase chain reaction.
Blood 73:2165, 1989
41. J ackson P, Baltimore D N-terminal mutations activate the
leukemogenic potential of the myristoylated form of c-ab/. EMBO J
8:449, 1989
42. van Etten RA, J ackson P, Baltimore D The mouse type IV
c-ab/ gene product is a nuclear protein, and activation of trans-
forming ability is associated with cytoplasmic localization. Cell 58:
669, 1989
43. Collins S, Coleman H, Groudine M: Expression of BCR and
BCR-ABL fusion transcripts in normal and leukemic cells. Mol Cell
Biol 7:2870, 1987
44. Didsbury J , Weber RF, Bokoch GM, Evans T, Snyderman R
rac, a novel vas-related family of proteins that are botulin toxin sub-
strates. J Biol Chem 264:16378, 1989
45. Abo A, Pick E, Hall A, Totty N, Teahan CG, Segal A W Ac-
tivation of the NADPH oxidase involves the small GTP-binding pro-
tein p21"'. Nature 353:668, 1991
For personal use only. on March 31, 2014. by guest bloodjournal.hematologylibrary.org From