Download The Tel-PDGFRß fusion gene produces a chronic

Leukemia (1999) 13, 1790–1803
 1999 Stockton Press All rights reserved 0887-6924/99 $15.00
http://www.stockton-press.co.uk/leu
The Tel-PDGFR␤ fusion gene produces a chronic myeloproliferative syndrome in
transgenic mice
KA Ritchie1,2 AAG Aprikyan1,3, DF Bowen-Pope2, CJ Norby-Slycord1,2, S Conyers1, E Sitnicka4,5 SH Bartelmez4,5
and DD Hickstein1,3
1
Medical Research Service, VA Puget Sound Health Care System, Seattle, WA; 5Seattle Biomedical Research Institute Seattle, WA;
Department of 4Pathobiology School of Public Health, and Departments of 2Pathology, and 3Medicine, School of Medicine, University of
Washingon, Seattle, WA, USA
Chronic myelomonocytic leukemia (CMML) is a pre-leukemic
syndrome that displays both myelodysplastic and myeloproliferative features. The t(5;12) chromosomal translocation,
present in a subset of CMML patients with myeloproliferation
fuses the amino terminal portion of the ets family member, Tel,
with the transmembrane and tyrosine kinase domains of platelet-derived growth factor receptor ␤ (PDGFR␤) gene. To investigate the role of this fusion protein in the pathogenesis of
CMML, we expressed the Tel-PDGFR␤ fusion cDNA in hematopoietic cells of transgenic mice under the control of the human
CD11a promoter. Transgenic founders and their offspring
express the transgene specifically in hematopoietic tissues and
develop a myeloproliferative syndrome characterized by: overproduction of mature neutrophils and megakaryocytes in the
bone marrow; splenomegaly with effacement of splenic architecture by extramedullary hematopoiesis; an abnormal population of leukocytes co-expressing lymphoid and myeloid markers; and increased numbers of colonies in in vitro bone marrow
CFU assays. All mice expressing the transgene exhibited at
least one of these features of dysregulated myelopoiesis, and
20% progressed to a myeloid or lymphoid malignancy. This
murine model of CMML parallels a myeloproliferative syndrome
in humans and implicates the Tel-PDGFR␤ fusion protein in
its pathogenesis.
Keywords: Tel-PDGFR␤; myeloproliferative syndrome; transgenic
mice
Introduction
Myelodysplastic syndromes (MDS) represent heterogeneous
clinicopathologic conditions generally characterized by ineffective hematopoiesis.1,2 Peripheral blood counts are low
despite a usually hypercellular marrow. Also, peripheral blood
erythrocytes, granulocytes, and platelets, and their precursors
in the marrow demonstrate abnormal morphologies
(dysplasia) (see Bain for review and description of the French–
American–British (FAB) classification scheme57). Those MDS
with increased numbers of blasts in the marrow may progress
to acute myeloid leukemia, and therefore represent pre-leukemic states.
The chronic myeloproliferative syndromes (MPS) are hematologic diseases characterized by hematopoietic hyperproliferation.3 The bone marrow is hypercellular, but hematopoiesis
is effective and maturation is progressive, producing increased
numbers of mature elements in the peripheral blood. Classification of a particular MPS is based on the lineage that is
predominant in the periphery, ie erythrocytes in polycythemia
vera (PV), platelets in essential thrombocythemia (ET), megakaryocytes (in the spleen) with myelofibrosis in myelosclerosis
Correspondence: KA Ritchie, Pathology and Lab Medicine Service
113, VA Puget Sound Health Care System, 1660 South Columbian
Way, Seattle, WA 98108, USA; Fax: 206-764-2001
A Aprikyan and K Ritchie contributed equally to this work
Received 16 March 1999; accepted 20 May 1999
with myeloid metaplasia (MMM), and granulocytes in chronic
myeloid leukemia (CML). All the MPS are characterized by
splenomegaly, and may display subacute clinical problems
associated with the hyperplastic lineage (eg hyperviscosity in
PV, thrombosis in ET). Over time, these diseases may evolve
into marrow failure. All of the MPS are associated with an
increased risk of developing leukemia, best exemplified by
CML, which progresses from a chronic phase to an acute
leukemia in almost all patients.
The progression of myelodysplastic syndromes to increasingly severe dysplasias, cytopenias, and ultimately to acute
myeloid leukemia (AML) signifies these conditions as pre-leukemic states, and supports the multi-hit hypothesis of malignant transformation. Similarly, the increased risk of leukemia
associated with the chronic myeloproliferative syndromes suggests that they, too, are pre-leukemic states. In both syndromes, the dysregulation of hematopoiesis predisposes to
leukemia. Investigation of possible mechanisms of this dysregulation might elucidate early events in leukemogenesis.
Chronic myelomonocytic leukemia (CMML) is an unusual
myelodysplastic syndrome with features of both myelodysplasia and myeloproliferation. It is characterized by increased
numbers of monocytes and sometimes granulocytes in the
peripheral blood, and monocyte precursors in the marrow.
CMML may be difficult to distinguish both clinically and histopathologically from CML, however, the presence of the Philadelphia chromosome and the bcr/abl fusion cDNA are diagnostic of CML. CMML may be further subtyped into a
myelodysplastic form (eg abnormal morphologies in several
lineages), or a myeloproliferative form, similar to CML yet
negative for the Philadelphia chromosome. A subset of CMML
shows a distinctive t(5,12) chromosomal translocation, and
shows features of both MDS and chronic myeloproliferative
syndrome.4
Insight into CMML was obtained by the cloning and
sequencing of the t(5;12) chromosomal breakpoint, associated
with a subset of CMML.5,6 This translocation creates a fusion
gene between Tel, a newly identified member of the ETS family of transcription factors, and the platelet-derived growth factor receptor beta gene (PDGFR␤).7 The fusion gene contains
the first 154 amino acids of Tel, including a putative helixloop-helix domain thought to mediate protein–protein interactions. The breakpoint occurs such that the fusion mRNA encodes a protein which lacks the extracellular ligand binding
domain of PDGFR␤, but includes the transmembrane and
intracellular domains of PDGFR␤, including the split tyrosine
kinase domains that mediate intracellular signalling. Both the
oligomerization domain of Tel and the tyrosine kinase activity
of PDGFR␤ are operative in the transformation of Ba/F3
cells.8,9 Both Tel10–19 and PDGFR␤20,21 have been found in a
number of different translocation fusion genes associated with
hematopoietic malignancies. The participation of PDGFR␤ in
malignant transformation and normal signalling pathways is
Tel-PDGFR␤ transgenic mice develop a myeloproliferative syndrome
KA Ritchie et al
well characterized.7,22–24 However, the respective roles of Tel
and PDGFR␤ in the pathogenesis of CMML has yet to be
elucidated.
We used the human CD11a promoter, which directs tissuespecific expression of a human CD4 reporter gene in the leukocytes of transgenic mice,25 to express the human TelPDGFR␤ fusion gene in transgenic mice. Parallel to the
endogenous mouse CD11a gene, this promoter directs
expression of the reporter gene early in hematopoiesis, and
also in both myeloid and lymphoid leukocyte lineages. By 4
months of age, mice expressing the Tel-PDGFR␤ fusion protein exhibit dysregulated hematopoiesis with features similar
to a human myeloproliferative syndrome. Older mice demonstrate a low frequency of hematopoietic malignancies.
Materials and methods
Gene construct and transgenic mice
The construct used to generate the Tel-PDGFR␤ transgenic
mice consists of a 1.7-kb fragment from the human CD11a
proximal promoter, a 463 bp cDNA fragment encoding the
first 154 amino acids of Tel, a 3.3-kb cDNA fragment containing the entire transmembrane and intracellular domains of
PDGFR␤ (amino acids 511–1106 and 1787 bp of 3⬘ untranslated region),26 and a 1.9 kb fragment containing the human
growth hormone (hGH) mini-gene (Figure 1). The human CD4
cDNA was released from the CD11a promoter-hCD4-hGH
construct25 by BamHI digestion, and the portion containing
the human CD11a promoter, the human growth hormone
mini-gene and p0GH (Nichols Institute, San Juan Capistrano,
CA, USA) was gel isolated and purified on a Qiagen column.
A custom multicloning site (MCS) containing 5⬘-BamHI–SalI–
HincII–NotI–BglII-3⬘ restriction sites was synthesized (Applied
Biosystems Model 392, Foster City, CA, USA) and ligated into
the purified plasmid between the CD11a promoter and the
hGH mini-gene. Orientation was confirmed by sequencing.
Primer (b) with a 5⬘ HincII site and primer (d) with a 3⬘ NotI
site were used to amplify by RT-PCR the 463 bp Tel fragment
from mRNA prepared from the HL-60 cell line. This fragment
Figure 1
Construct used to generate the Tel-PDGFR␤ transgenic
mice. The PvuI–NsiI fragment containing 1.7 kb of the human CD11a
promoter, the portions of human Tel and human PDGFR␤ cDNAs
corresponding to the Tel-PDGFR␤ fusion gene, and a human growth
hormone mini-gene was isolated and sent to DNX for microinjection.
Arrows indicate oligos used during construction: (a) 5⬘GCCAGTGTCACCAGCCTGT-3⬘; (b) 5⬘-CCCCGAATATTGGCGTCGACATCTCTCTCGCTGTGAGACATG-3⬘; (c) 5⬘-CTCAAGTGGGCTGAAAATGAGTT-3⬘; (d) 5⬘-TGTCACGTGCTGTGCTGCGGCCGCCTTCTTCATGGTTCTGATGCAG-3⬘; (e) 5⬘-ACGTAGATGTACTCATGGCCGTCA-3⬘; (f) 5⬘-TTGGGAAGGCACTGCCCTGAA-3⬘.
was then ligated into the HincII site of the MCS. Orientation
and possible mutations were ruled out by DNA sequencing
using the primer pairs (a–d) and (c–e), and an automated DNA
sequencer (Applied Biosystems, Model 373). To preserve the
reading frame and amino acid composition of the junction,
the NotI site in the MCS was digested and filled in with
Klenow and dNTPs. Finally, the FspI–SspI fragment containing
the transmembrane and tyrosine kinase domains of PDGFR␤
was isolated from full length PDGFR␤ cDNA26 and ligated
into this NotI site. Orientation and maintenance of the reading
frame across the junction were confirmed by DNA sequencing
using primer pair (c–e). The hGH mini-gene was included in
the construct because it is reported to increase mRNA
stability.27
The entire construct (7.4 kb) was excised by Pvul–NsiI
digestion, isolated from an agarose gel, purified using a
Qiagen column, quantitated by absorbance at 260 nm, and
sent to DNX (Princeton, NJ, USA) for microinjection. Mice
received from DNX were screened for the presence of the
transgene as described,25 using a probe for the hGH portion
of the construct.
Detection of Tel-PDGFR␤ expression by Western
blotting
Approximately 100 ␮l of mouse peripheral blood were
obtained by retro-orbital bleeding. Erythrocytes were lysed by
an NH4 Cl-Tris hypotonic lysis method28 and the peripheral
blood leukocytes were washed twice in cold PBS, resuspended in lysis buffer (50 mm HEPES, pH 7.5; 150 mm NaCl;
1% Triton X-100) containing protease inhibitors (1.0 mm
PMSF, 1 ␮g/ml leupeptin, 1 ␮g/ml aprotinin), and centrifuged
at 10 000 g for 10 min at 4°C. Resultant supernatants were
used for analysis. Protein lysates from tissues were similarly
prepared. Protein concentration was determined by
absorbance at 280 nm. Fifty to 100 ␮g of total protein were
loaded on to a reducing, 10% SDS-polyacrylamide gel. Prestained molecular weight markers were run in one lane of
each gel. After electrophoresis, the proteins were electroblotted on to either a polyvinylidene difluoride (PVDF) membrane
(BioRad, Hercules, CA, USA) or nitrocellulose-nylon membrane (Genescreen; BioRad) in electroblot buffer (25 mm Tris,
192 mm glycine, 20% methanol) in an electroblotting apparatus (BioRad). The membrane was then incubated 30 min at
room temperature with blocking buffer (50 mm Tris pH 7.4,
150 mm NaCl, 0.1% Tween-20, 0.2% BSA), and probed with
a 1/2000 dilution of a rabbit polyclonal antibody (7649) that
reacts with the C-terminal portion of PDGFR␤.29 The blots
were washed four times with blocking buffer, incubated with a
1/5000 dilution of alkaline phosphatase conjugated goat antirabbit IgG secondary antibody (GIBCO-BRL, Gaithersburg,
MD, USA) for 1 h, washed again, and developed with either a
chromogenic substrate (p-nitrobluetetrazolium chloride (NBT)
and 5-bromo-4chloro3-indoylphosphate-toluidine salt (BCIP)
(BioRad) or chemiluminescent substrate (Chemisubstrate;
Pierce, Rockford, IL, USA).
Histology
Mouse tissues were fixed in 10% buffered formalin, pH 7.0,
sectioned and processed in a Miles Scientific (Naperville, IL,
USA) Tissue Tek according to manufacturer’s instructions.
Routinely sampled tissues were spleen, thymus, femur, lung,
1791
Tel-PDGFR␤ transgenic mice develop a myeloproliferative syndrome
KA Ritchie et al
1792
heart, kidney, liver, pancreas, brain, and skeletal muscle.
Bones were fixed in formalin, decalcified in perfix (1.8% zinc
chloride, 2% trichloroacetic acid, 4% paraformaldehyde, 17%
isopropyl alcohol in water), then processed routinely. Hematoxylin and eosin-stained histologic sections were prepared
according to routine procedures.30 Histologic sections were
photographed using a Zeiss photomicroscope (Carl Zeiss,
Oberkochen, Germany) and Nikon UFX exposure metering
system (Tokyo, Japan).
Flow cytometry
The following directly conjugated antibodies were used to
analyze spleen cells and peripheral blood leukocytes: fluoroscein isothiocyanate (FITC) rat anti-mouse CD45R/B220 (B
lymphocyte marker, RA3-6B2, IgG2a, PharMingen, San Diego,
CA, USA); FITC conjugated rat anti-mouse CD2 (T lymphocyte
marker, RM2–5, IgG2b, PharMingen); FITC conjugated hamster anti-mouse CD3 (T cell marker, 145-2C11, hamster IgG,
PharMingen); phycoerythrin (PE) conjugated rat anti-mouse
granulocyte (GR-1, RB6-8C5, IgG2b PharMingen); phycoerythrin (PE) conjugated rat anti-mouse CD11b (myeloid marker,
M1/70, IgG2b, PharMingen); FITC conjugated rat anti-mouse F
4/80 (monocyte/macrophage marker, C1:A3-1, IgG2b, Serotec,
Washington, DC, USA); FITC conjugated rat anti-mouse CD14
(monocyte marker, rmC5-3, IgG1, PharMingen); FITC conjugated rat IgG1 (R3-34, isotype control, PharMingen); FITC conjugated rat IgG2a (R35–95, isotype control, PharMingen); FITC
conjugated rat IgG2b (R53–38, isotype control, PharMingen);
PE conjugated rat IgG2b (R53-38, isotype control,
PharMingen). Rat IgG (Sigma Chemicals, St Louis, MO, USA)
or Fc BLOCK (Pharmingen) were used to block non-specific
binding.
Spleen and peripheral blood leukocytes from transgenic
founders and normal littermates were analyzed by flow cytometry. Peripheral blood was obtained from anesthetized mice
by orbital bleeding. Mice were sacrificed by cervical dislocation, tissues were removed for histology and cell suspensions were made by gently rubbing one half spleen over a fine
stainless steel screen. Erythrocytes were lysed as described,28
leukocytes were washed twice in PBS, and resuspended in
PBS, 2% fetal bovine serum (FBS), and 0.1% sodium azide.
Fifty ␮l of cell suspension (5 × 105 cells) were dispensed into
a 5 ml polystyrene test tube, non-specific binding was blocked
with 5 ␮l of rat IgG (10 mg/ml stock) or 2 ␮l of Fc BLOCK
(0.5 mg/ml stock), 1 ␮l of appropriate antibody was added to
each tube, and the cells were incubated on ice for 60 min.
The cells were then washed twice with PBS, 2% fetal bovine
serum, 0.1% sodium azide, fixed with 1% paraformaldehyde,
and analyzed on a FACScan (Becton Dickenson, San Jose, CA,
USA) flow cytometer using CONSORT 30 software. A gate
was drawn around all live leukocytes on the forward scatter
(FSC) vs side scatter (SSC) plot, and analyzed on an FL-2 (PE)
vs FL-1 (FITC) two dimensional plot. In some experiments antibody capping was inhibited by 4°C instead of sodium azide,
propidium iodide was used to label dead cells, and the cells
were analyzed using FACSCAN Research software. Both procedures yielded the same results. Analysis quadrants were
placed such that 98% of the cells stained with isotype controls
fell into the lower left quadrant. Single antibody controls
showed the location of each different, positive population on
the contour graph. The relative fluoresences of the FITC-CD2
positive cells and the FITC-B220 positive cells were sufficiently different to clearly separate T cells from B cells on
the x axis. Simultaneous addition of PE-GR-1, FITC-CD2 and
FITC-B220 to a single tube allowed for distinct separation of
granulocytes, T cells and B cells on the two-dimensional
contour graph.
Colony-forming cell (CFC) assays
Bone marrow cells were harvested from one femur by flushing
the marrow cavity with 3 ml of cold PBS, 1% FBS using a 1 cc
syringe and a 25 gauge needle. Cells were counted and were
grown in a double layer nutrient agar culture in the presence
of stem cell factor (SCF) 50 ng/ml, interleukin-3 (IL-3),
50 ng/ml, interleukin-6 (IL-6), 20 ng/ml, and granulocyte–
macrophage colony-stimulating factor (GM-CSF), 5 or
20 ng/ml, as described.31 In most experiments cells were
plated at 50 000 cells per plate and these data were then normalized to CFC per 5000 cells. Plating densities of either 5000
or 50 000 gave similar results. All CFC assays were set up in
triplicate. Plates were incubated in 5% CO2, 95% ambient air
gas mixture at 37°C for 10 days, and colonies counted at low
magnification by individuals blinded to the plates’ composition. Raw data were statistically analyzed and graphed using
Excel (Microsoft, Redmond, WA, USA) or Kaleidagraph
(Synergy Software, Reading, PA, USA) or Instat for DOS
(Graphpad, San Diego, CA, USA) softwares.
To determine the types of cells in colonies for each growth
condition, 10 individual colonies from each growth condition
were picked with Pasteur pipettes, separately resuspended in
0.5 ml of sterile PBS, 2% FBS, cytospun (Cytospin 2, Shandon,
Pittsburgh, PA, USA), Wright–Giemsa-stained (Hemastainer;
Geometric Data, Wayne, PA, USA), and a 200 cell differential
count was performed on each colony.
Copy number determination and Southern blots
Genomic DNA was prepared from tissues and treated with
RNase according to standard procedures.32 DNA concentration was determined by OD.260 To accomodate a range of
copy numbers, 10, 5, 2.5, and 1 ␮g of transgenic and normal
littermate control DNAs were applied to nylon membranes
(Genescreen Plus; New England Nuclear, Boston, MA, USA)
using a dot-blot apparatus (BioRad) according to the manufacturer’s directions. Dot blots were hybridized with a radiolabelled (Ready-to-Go DNA Labelling Beads; Pharmacia, Piscataway, NJ, USA) fragment of the human growth hormone gene25
and autoradiographed. Densitometry was performed using a
Beckman DU640 spectrophotometer (Palo Alto, CA, USA). To
derive relative copy numbers, all values were normalized to
the transgenic mouse with the lowest signal (Transgenic 935).
Immunoglobulin heavy chain (IgH) and T cell receptor beta
(TCR␤) gene rearrangements were detected by Southern blotting. Purified genomic DNA was separately digested with
EcoRI, BamHI, of BglII (New England Biolabs) and probed
with either pJ11 (mouse IgH J region, kindly provided by Dr
Ursula Storb)33 or TCR␤ (1.4 kb EcoRI–SacI fragment, kindly
provided by Dr Dennis Willerford).34,35 Because the transgenic mice are derived from both the C57B1/6 and SJL inbred
mouse strains, unrearranged bands were determined by
including kidney DNA from both parent strains. These controls ruled out restriction fragment length polymorphisms that
might otherwise be misinterpreted as rearranged genes.
Tel-PDGFR␤ transgenic mice develop a myeloproliferative syndrome
KA Ritchie et al
Results
Generation of transgenic mice
The human CD11a promoter was used in the transgene construct to direct expression of the Tel-PDGFR␤ fusion cDNA
(Figure 1). The 1.7 kb CD11a proximal promoter directs the
expression of a human CD4 reporter gene in all leukocytes of
transgenic mice.25 Fifteen out of 44 mice received from DNX
were positive for the transgene, as determined by tail DNA
slot blotting or by PCR amplification of tail DNA using primers
that flank the Tel-PDGFR␤ junction (c and e, Figure 1). One
female and all of the male founders were bred to establish
transgenic lines. Normal littermates were reserved for controls. Two founders died prior to the time of analysis and two
founders died at over 2 years of age without evidence of
malignancy. Eleven founders, 11 littermates and a number of
offspring from different transgenic lines were analyzed in
detail by Western blotting, histopathology, FACS analysis, and
in vitro bone marrow colony-forming cell (CFC) assay.
(Figure 2a, left two lanes). The additional bands on the blot
represent proteins which cross react with the polyclonal antiPDGFR␤ antibody. These cross reacting proteins do not represent endogenous mouse PDGFR␤, which has a molecular
weight of 180 kDa,7 and is not expressed on PBLs.23
The human CD11a promoter directs expression of transgenes in hematopoietic tissues.25 Western blots of different
tissues from transgenic mice reveal that the Tel-PDGFR␤
fusion protein is expressed only in spleen, thymus and bone
marrow (Figure 2b). Seven different founders and offspring
which were tested displayed expression only in the spleen,
thymus and bone marrow. Although the amount of TelPDGFR␤ protein expressed by each founder varied,
expression in all transgenics was restricted to leukocytes, and
was not related to transgene copy number (see below). These
studies indicate that the human CD11a promoter directs
tissue-specific expression of the Tel-PDGFR␤ transgene in
leukocytes of transgenic mice.
Histology
Analysis of Tel-PDGFR␤ expression by Western
blotting
To establish that the Tel-PDGFR␤ transgenic mice express the
fusion protein, lysates from peripheral blood leukocytes (PBL)
were analyzed by Western blotting. Western blots of PBL protein lysates from founders and normal littermates were probed
with a rabbit polyclonal antibody that reacts with the C-terminal portion of PDGFR␤. All of the transgenic founders
(these seven plus eight others not shown) express variable
amounts of the 85 kDa Tel-PDGFR␤ fusion protein in their
PBLs (Figure 2a). The 85 kDa band representing the fusion
protein is not present in leukocytes from normal littermates
Figure 2
(a) Expression of the Tel-PDGFR␤ fusion protein in peripheral blood leukocytes. PBL lysates from transgenic mice and normal littermates were blotted and probed with a rabbit polyclonal primary antibody that recognizes the C-terminal portion of human
PDGFR␤ and a secondary, alkaline phosphatase conjugated goat antirabbit antibody. (b) Expression of the Tel-PDGFR␤ fusion protein in
different tissues. Protein lysates from different tissues were blotted and
probed with a rabbit polyclonal primary antibody that recognizes the
C-terminal portion of human PDGFR␤ and a secondary, alkaline
phosphatase conjugated goat anti-rabbit antibody.
Representative histology of a normal littermate and two transgenic founders is shown (Figure 3). Low power views of the
spleens (Figure 3, top row) indicate that the normal littermate
has intact follicular architecture with evenly distributed
lymphoid follicles (darker areas) with paler, central germinal
centers, and pale interfollicular areas. The spleens of rodents
normally contain small amounts of extramedullary hematopoiesis (EMH), usually seen as a few megakaryocytes and
erythroid precursors scattered in the subcapsular regions.
EMH does not normally efface the splenic architecture, however the splenic architecture of the transgenics is effaced by
EMH, with only irregular residual lymphoid regions (darker
areas). The large, pale cells scattered throughout the effaced
spleen and visible even at low power are megakaryocytes,
best shown by transgenic 941. While the megakaryocyte
nuclei in normal littermate 926 and transgenic 907 display
normal morphology, those of transgenic 941 are dark and
irregular, suggestive of myelodysplasia in 941. High power
views of the spleens (second row) show intact follicular architecture in the normal littermate with a germinal center at the
upper left corner, and surrounding small lymphocytes. High
power views of both transgenic spleens show immature
erythroid cells (smaller darker cells), granulocytes with irregular and doughnut-shaped nuclei, and large, pale, polyploid
megakaryocytes. Eight out of the 11 transgenics analyzed
show partial to total effacement of the splenic architecture by
EMH. Although the EMH contained erythroid, myeloid and
megakaryocytic precursors in all cases, six mice had a predominance of granulocytes at all stages of maturation
(demonstrated by transgenic 907), and two had large numbers
of megakaryocytes (exemplified by transgenic 941).
Normal mouse bone marrow has a cellularity of greater
than 90%, therefore, the increase in cellularity up to 100%
that is seen in the transgenics is not readily demonstrable.
However, high power views of the femoral marrows (bottom
row) show that the marrow of normal littermates contains a
mixture of trilineage hematopoiesis, while the marrows of the
Tel-PDGFR␤ transgenics show a predominance of either granulocytes or megakaryocytes. For example, transgenic 907
marrow is packed with mature neutrophils and scattered
megakaryocytes but no erythroid precursors. Trilineage hematopoiesis with greatly increased numbers of dysplastic megakaryocytes is present in transgenic 941. The Tel-PDGFR␤
1793
Tel-PDGFR␤ transgenic mice develop a myeloproliferative syndrome
KA Ritchie et al
1794
Figure 3
Histology of the spleen and bone marrow from a normal littermate and two transgenic founders at 5 months of age. Normal
littermate tissues are shown in the left hand column and transgenic tissues appear in the middle and right-hand columns. The first row contains
low power views of spleens; the second row contains high power views of spleens, and the bottom row shows high power views of the
bone marrows.
transgene does not appear to cause any relative increase in
erythroid precursors. Of the eight transgenics showing abnormal bone marrow histology, four show increased numbers of
granulocytes, three increased numbers of megakaryocytes,
and one shows a malignant histiocytosis (see below).
The Tel-PDGFR␤ founders and their normal littermates have
been followed for evidence of hematologic abnormalities by
peripheral blood smears, white blood cell counts (WBC),
hematocrit and hemoglobin determinations and FACS analysis
of peripheral blood leukocytes (data not shown). These studies
did not reveal a significant difference in the peripheral blood
between most transgenics and normal littermates, which is
likely due to the wide variability of peripheral total WBC (2
to 20 × 109/liter) and percent granulocytes (5 to 25%) between
individuals and within individuals over time.36
FACS analysis of transgenic mice
Spleen cells and peripheral blood leukocytes from normal littermates and transgenics were analyzed by FACS. Profiles of
representative founders and normal littermates are shown
(Figure 4). Simultaneous staining of cells with the three antibodies PE-GR-1, FITC-CD2, and FITC-B220 allows separation
and enumeration of granulocytes (upper left), T lymphocytes
(lower right) and B lymphocytes (lower far right) on a single
contour plot.
A few transgenics displayed increased neutrophils in the
peripheral blood and spleen. For example, normal littermate
926 had 5% neutrophils in its spleen, while transgenic 907
had 48% neutrophils (Figure 4a). Also, the transgenic mouse
(907) had markedly decreased numbers of CD2 positive T
cells, as shown by the absence of the population of cells just
to the right of the vertical line. FACS analysis of peripheral
blood showed the same results (data not shown). In addition
to histologic abnormalities, transgenic 907 also exhibited
abnormal leukopoiesis as evidenced by greatly increased
numbers of neutrophils and decreased numbers of T cells in
the spleen and peripheral blood.
Other transgenics displayed an abnormal population of leukocytes that simultaneously coexpress both lymphoid and
myeloid markers. Transgenic founder (940) and normal littermate (913) were chosen at random and analyzed by histology and FACS analysis (Figure 4b). FACS analysis of their
spleens showed a population of abnormal leukocytes in the
transgenic abnormally co-expressing variable amounts of the
granulocyte marker GR-1 and constant amounts of the T lymphocyte marker CD2 (Figure 4b). This abnormal population is
not present in the normal littermate. The same result was
obtained when CD3, another T cell marker, was used in place
of CD2 (data not shown). FACS analysis of another normal
littermate (910) and transgenic (929) again showed an abnormal population of cells in the transgenic which co-expressed
GR-1 and CD2 (Figure 4c, upper panel). However in this
Tel-PDGFR␤ transgenic mice develop a myeloproliferative syndrome
KA Ritchie et al
1795
Figure 4
FACS analysis of spleen cells from representative normal littermates (left panels) and representative transgenics (right panels). (a
and b) Two-color fluorescence PE-GR-1 vs FITC CD2 + FITC B220. (c) Lower panels also show PE-CD11b vs FITC CD2+FITC B220. Single
antibody controls demonstrate that the B220 positive cells are brighter than (to the right of) CD2 positive cells, and that CD2 positive cells are
found to right of the vertical line. The scales of both axes are log10 fluoresence. The mice were analyzed at (a) 5; (b) 8; and (c) 13 months of age.
transgenic, the expression of GR-1 was uniformly high, rather
than variable, as seen in transgenic 940 (Figure 4b). Coexpression of both myeloid and lymphoid markers was confirmed by demonstrating that this abnormal population of leukocytes in trangenic 929 also co-expressed another myeloid
marker, CD11b, along with CD2 (Figure 4c, lower panel).
The nine transgenic founder mice that were analyzed by
FACS demonstrated four immunophenotypes: one mouse displayed no abnormalities on FACS; three mice had a slightly
increased percentage of granulocytes (data not shown); one
mouse had a markedly elevated (48%) percentage of granulocytes, and four mice showed an abnormal population of leukocytes that simultaneously co-expressed both lymphoid and
myeloid cell surface antigens. These findings, particularly the
biphenotypic picture, suggest that the Tel-PDGFR␤ transgene
disrupts leukopoiesis and alters lineage fidelity.
Progenitor cell levels
To investigate whether the Tel-PDGFR␤ transgenic phenotype
reflects changes present in hematopoietic progenitors, colonyforming assays were performed on the marrow from normal
littermates and transgenic founders (Figure 5). All of the transgenic founders (and their offsrping, see below) analyzed,
regardless of age or sex, displayed more CFCs per 5000 cells
plated than normal littermate controls. This increase in bone
marrow CFCs was a consistent abnormality in Tel-PDGFR␤
founders (Figure 5) and in their progeny (Figure 7).
Tel-PDGFR␤ offspring
Six Tel-PDGFR␤ founders were backcrossed into C57B1/6
normal mice to generate independent transgenic lines. Histopathologic, Western blot, and FACS analysis of these offspring
Figure 5
Colony-forming cell (CFC) numbers of transgenic founders and age matched normal littermates at the folowing ages: 5
months (926, 907); 10 months (912, 943); 13 months (909, 937, 910,
929); and 15 months (921, 919). All founders tested showed a statistically significant increase in the number of CFC on soft agar colony
forming assay.
showed the same features of a myeloproliferative syndrome
as described for the founders: leukocyte-specific expression,
distortion of splenic architecture by EMH, biphenotypic leukocytes, and increased numbers of CFC.
Consistent with a single integration site, approximately 50%
of the progeny of transgenic founders inherit the transgene
(data not shown). At 4 months of age, non-transgenic offspring
Tel-PDGFR␤ transgenic mice develop a myeloproliferative syndrome
KA Ritchie et al
1796
a
c
b
d
Figure 6
Histopathology and FACS analysis of spleens from transgenic offspring at 4 months of age. (a, b, e and g) Offspring littermates that
did not inherit the transgene. (c, d, f and h) Offspring that inherited the transgene.
had a normal sized spleen with intact follicular architecture
(Figure 6a), and interfollicular regions containing small lymphocytes (Figure 6b). Four-month-old transgenic offspring had
enlarged spleens that maintained follicular architecture but
showed enlarged interfollicular regions (Figure 6c). High
power view demonstrates that the interfollicular areas are
expanded by extramedullary hematopoiesis (large arrowheads
point to megakaryocytes, small arrowheads to clusters of granulocytes, the smallest, dark cells are erythroid precursors,
Figure 6d). Regardless of founder lineage, all offspring that
inherited the transgene displayed the increased splenic extramedullary hematopoiesis seen in the parent.
FACS analysis of offspring revealed that leukocyte biphenotypism follows the inheritance of the transgene. Non-transgenic offspring displayed normal granulocytes, T cells and B
cells on FACS analysis of peripheral blood leukocytes (Figure
6e, g). (The 16% granulocytes demonstrated by this normal
littermate is within the normal range of 5–25%.) Transgenic
offspring showed biphenotypic leukocytes (Figure 6f, h) like
those seen in their transgenic parent, and additionally, the pattern of biphenotypism was inheritable. Founders with CD2+,
uniformly high GR-1+, CD11b+ leukocytes (Figure 4c) produced transgenic offspring that also displayed CD2+, uniformly high GR-1+, CD11b+, CD11b+ leukocytes (Figure 6f,
h). Founders that displayed leukocytes that coexpressed CD2
with a range of GR1 positivity (Figure 4b) produced transgenic
offspring that also displayed CD2+, variable GR-1+ biphenotypic leukocytes (data not shown).
CFC analysis of offspring (Figure 7) showed that the transgenic lines had increased numbers of CFCs compared to normal littermate controls. One individual from each transgenic
line was analyzed; two offspring that did not inherit the transgene served as controls; cells were plated out in triplicate. The
offspring of founders 929, 941 and 919 showed the greatest
increase in CFCs. These founders also demonstrated the earliest and most severe manifestations of the myeloproliferative
Tel-PDGFR␤ transgenic mice develop a myeloproliferative syndrome
KA Ritchie et al
Figure 7
CFC numbers for 4-month-old offspring of different transgenic founders. The founder number is indicated at the bottom.
phenotype (see spleen of 941, Figure 3), and relatively larger
increases in CFCs in the founders (see founders 919 and 929
CFCs, Figure 5). The founders whose offspring had milder
increases in CFCs (907) also tended to have intermediate
increases in CFCs themselves, milder splenomegaly, less
extramedullary hematopoiesis and less striking abnormalities
on FACS analysis. The onset and severity of the myeloproliferative features most closely correlate with the degree of
increased bone marrow CFCs of each founder and its offspring. The particular features of the myeloproliferative syndrome breed true within each lineage and the myeloproliferative phenotype tracks with the inheritance of the TelPDGFR␤ transgene.
Malignancies in founders
Three out of 13 (23%) of the Tel-PDGFR␤ founders developed
a malignancy. No malignancies were observed in agematched, normal littermate controls. At 14 months of age,
transgenic 911 was found to have a large tumor in the mesentery of the small bowel (Figure 8a, large arrowhead), hepatomegaly with white nodules of tumor infiltration (medium
arrowhead), and a spleen 10 times normal size (small
arrowheads). The histology of the tumor showed a mixture of
early myeloid cells, granulocytes and lymphocytes (Figure 8b).
The malignant nature of the tumor was demonstrated by tumor
infiltration of spleen and lung parenchyma (data not shown)
and by tumor infiltration among hepatocytes (Figure 8c, L).
Founder 902 was found to have an enlarged spleen and a
retrosplenic tumor. The histology was identical to that of the
tumor in transgenic 911 and consisted of a mixture of granulocytes and lymphocytes with numerous mitotic figures (Figure
8d). FACS analysis of the 911 and 902 tumors were identical
and confirmed the histopathology (Figures 8e and f showing
902 tumor). The cells were uniformly positive for CD11a,
which confirms the hematopoietic origin of the tumor and
rules out a poorly differentiated carcinoma. Also, the tumors
consisted of a mixture of CD3-positive T cells, B220-positive
B cells, and a few GR-1-positive granulocytes (not shown).
DNA dot blots confirmed the presence and Western blots confirmed the protein expression of the transgene in the tumors
(not shown). Morphologically and immunophenotypically
these tumors represent dysregulated hematopoietic neoplasms
that defy classification into specific types or lineages.
Because of the indefinite histology and FACS results of 902
and 911 tumors, antigen receptor gene rearrangement studies
were performed to help determine the lineages of the malignancies. Genomic Southern blots of tumor DNAs were probed
with either a mouse IgH J-region probe (Figure 9a) or a mouse
TCR␤ probe (Figure 9b). Kidney DNA of the parental mouse
strains (C57B1/6 and SJL) indicate the unrearranged, germline
bands (G). Mice 902 and 230 have rearranged IgH alleles,
and germline TCR␤ alleles. This established the existence of
a monoclonal B cell lymphoma in the mixture of cells in the
902 tumor, and confirms the B cell immunophenotype (CD19)
of the 230 tumor (see below). Mouse 911 has a definitive
TCR␤ gene rearrangement, and a probable IgH rearranged
band slightly above the germline band. This indicates that the
911 tumor most likely contains a malignant lymphoma that
has rearranged both IgH and TCR␤ loci. This observation is
occasionally seen in human lymphomas, and is a manifestation of ‘molecular biphenotypism’. Alternatively but less
likely, the 911 tumor contains a mixture of two separate
malignant lymphomas, one of B cell and the other of T cell
lineage.
Founder 937 suddenly developed hind limb paralysis and
was sacrificed. Post-mortem analysis showed a population of
biphenotypic leukocytes, increased bone marrow CFCs but no
gross evidence of tumor. Therefore, only histologic assessment
was possible. Histology of the spine showed that a histiocytic
infiltrate replaced the marrow of some vertebral bodies
(arrowhead, Figure 10a), while the marrow of adjacent vertebral bodies had normal hematopoiesis (asterisk, Figure 10a).
In clinical terms this mouse had suffered a pathologic fracture
of its spine. The malignant nature of this process is demonstrated by tumor infiltrating paraspinal skeletal muscle (Figure
10b, arrowhead). While femoral marrow had regions of normal hematopoiesis (Figure 10c, asterisk), the myeloproliferation replaces adjacent marrow of (Figure 10c, arrowhead),
indicating the systemic nature of this process. High power
view of the infiltrate (Figure 10d) shows a histiocytic appearance. A negative reticulin stain (data not shown), ruled out
marrow fibrosis. Therefore, similar to human myeloproliferative syndromes, the Tel-PDGFR␤ mice display a predisposition towards malignant transformation.
Malignancies in offspring
Although most offspring were sacrificed at a young age for
analysis, several lineages were maintained for breeding and
over time several of these offspring developed malignancies.
At 4 months of age, transgenic mouse 424 (offspring of founder 937) died suddenly. Necropsy discovered a large, mediastinal tumor compressing the trachea. Histology showed a
1797
Tel-PDGFR␤ transgenic mice develop a myeloproliferative syndrome
KA Ritchie et al
1798
a
c
b
d
e
Figure 8
f
Histopathology and FACS analysis of 911 (a, b, c) and 902 (d, e, f) tumors. The mice were analyzed at 14 months of age.
a
b
Figure 9
Gene rearrangements in tumors. Unrearranged bands (G) are indicated by kidney DNA from the parental mouse strains C57B1/6
and SJL. Tumors are denoted by their mouse number (902, 911, 230). (a) Genomic DNA was digested with BamHI and hybridized with a
mouse immunoglobulin heavy chain J region probe; (b) Genomic DNA was digested with BglII and hybridized with a mouse T cell receptor
beta chain probe.
Tel-PDGFR␤ transgenic mice develop a myeloproliferative syndrome
KA Ritchie et al
1799
a
c
b
d
Figure 10
Founder 937 suffered a pathologic spine fracture and paralysis at 13 months of age, due to a systemic and aggressive histiocytic
infiltration of bone marrow and paraspinal muscle.
malignant lymphoid tumor that infiltrated around pulmonary
vessels (Figure 11a). This same finding was recently described
for a bcr-abl transgenic mouse (Figure 3).37
At age 14 months, transgenic mouse 230 (offspring of founder 929) developed an enlarged abdomen and was sacrificed.
The mouse had large mesenteric and mediastinal tumors. Peripheral smear showed increased numbers of small, dark lymphocytes (Figure 11b). The tumors consisted of small, dark,
lymphoblastoid cells with a high mitotic rate. These malignant
cells also packed the spleen and bone marrow. DNA dot blot
confirmed the presence of Tel-PDGFR␤ DNA, and Western
blot showed that the tumor expressed the Tel-PDGFR␤ protein
(not shown). FACS analysis showed that the tumor cells
expressed CD2, CD5, CD19 and co-expressed variable
amounts of the myeloid marker CD11b. The histology, peripheral smear, co-expression of CD5 and CD19, and the IgH
gene rearrangement (Figure 9) are analogous to human
chronic lymphocytic leukemia. The co-expression of the
myeloid marker CD11b, T cell markers CD2 and CD5, and
the B cell marker CD19 exemplify the biphenotypism exhibited by dysregulated leukemic cells.
Because the presentation of several different founders and
offspring can be difficult to follow, Table 1 is included to indicate the founder and offspring designations. Offspring of each
founder were identified with numbers within a particular hundred series. Table 1 also summarizes relative transgene copy
numbers. Similar to that found for CD11a-promoter-hCD4
transgenic mice, there is no correlation between copy number
and levels of expression, CFC, phenotypic features, and the
development of malignancy. Lack of correlation with copy
number may be due to effects of different integration sites,
timing of expression during hematopoiesis, and/or genetic
variability in these non-inbred, F2 mice.
Table 2 summarizes the number of founders showing the
various features of the Tel-PDGFR␤ phenotype.
Discussion
Transgenic mice expressing the Tel-PDGFR␤ fusion protein
display a phenotype characterized by splenomegaly, efface-
ment of the splenic architecture by extramedullary hematopoiesis, increased granulocytes and/or megakaryocytes in the
spleen and bone marrow, abnormal leukocytes that coexpress both myeloid and lymphoid markers, and increased
numbers of colonies on bone marrow soft agar CFC assays.
While normal littermates show normal lymphoid splenic
architecture and normal, trilineage hematopoiesis in the marrow, Tel-PDGFR␤ transgenics show markedly abnormal hematopoiesis consistent with a chronic myeloproliferative syndrome. Transgenics like 907 have histology very similar to that
seen in the chronic phase of human chronic myeloid leukemia,3 and transgenics like 941 have histology very similar to
that seen in human essential thrombocythemia or early stages
of myelosclerosis with myeloid metaplasia.3 With long-term
follow-up, 23% of the transgenic founders developed malignancy. Thus, the features of leukocyte-specific, yet both
myeloid and lymphoid expression of Tel-PDGFR␤ in transgenic mice are those of a chronic myeloproliferative syndrome, and are similar to those seen in transgenic mice harboring the CML-associated, bcr-abl fusion gene, and in human
chronic myeloproliferative syndromes.
Cytogenetic and other studies have shown that the underlying abnormality in the myeloproliferative syndromes exists in
a pluripotent progenitor cell capable of differentiation into
several lineages.38,39 However, the pre-leukemic syndromes
and most of the associated leukemias are expressed as overgrowth in the myeloid lineages (granulocytes, monocytes).
The lymphoid lineage is rarely affected in MDS, but often
affected in MPS. Immunophenotyping of leukemias and MPS
has shown that the leukemic clone can simultaneously or serially manifest itself as an overproliferation of both lymphoid
and myeloid cells (termed a bilineage leukemia). Alternatively, the leukemic clone can simultaneously express both
myeloid and lymphoid surface markers on the same cell
(termed a biphenotypic leukemia). The ability of a leukemic
process to exhibit lineage heterogeneity is well demonstrated
by CML, in which the myeloid chronic phase of the leukemia
transforms to an acute lymphoblastic leukemia with an immature B cell phenotype. Similarly, myeloid plus B cell and
myeloid plus T cell biphenotypism has been described in
human CML.40,41 Therefore, lineage heterogeneity and lineage
Tel-PDGFR␤ transgenic mice develop a myeloproliferative syndrome
KA Ritchie et al
1800
a
b
Figure 11
(a) Histopathology of transgenic offspring 424 tumor (5 months of age). (b) FACS analysis and histopathology of transgenic offspring
230 tumor (14 months of age).
promiscuity are features of the dysregulated leukopoiesis seen
in human acute leukemia and MPS, as well as in bcr-abl42–44
and Tel-PDGFR␤ transgenic mice.
The CD11a promoter-Tel-PDGFR␤ phenotype resembles
the myeloproliferative phenotype seen in transgenic mice that
express the bcr-abl fusion gene from the t(9,22) translocation
of the Philadelphia chromosome positive CML.42–44 Both the
Tel-PDGFR␤ and bcr-abl transgenic mice develop myeloproliferative syndromes characterized by splenomegaly, effacement of the splenic architecture, extramedullary hematopoiesis, granulocytic and megakaryocytic hyperplasia, and
biphenotypic leukocytes (GR-1 and B220 coexpression).44
This myeloproliferative pre-leukemia is not seen in transgenic
mice in which Tel-PDGFR␤ transgene expression is directed
Tel-PDGFR␤ transgenic mice develop a myeloproliferative syndrome
KA Ritchie et al
Table 1
Relative copy numbers and offspring designations for TELPDGFR␤ mice
Founder
902
905
907
911
919
928
929
935
937
941
Table 2
Offspring designation
Relative copy number
not bred
500
700
not bred
0–100
600
200
300
400
100
0.4
1.3
10
5.7
34
3.3
3.0
1.0
21.7
5.3
Features of the tel-PDGFR␤ myeloproliferative syndrome
Feature
Splenomegaly (n = 11)
Spleen histology (n = 11)
Extramedullary hematopoiesis
Intact architecture
Bone marrow histology (n = 11)
Increased granulocytes,
megakaryocytes, or histiocytes
No abnormality
FACS analysis (n = 9)
Biphenotypic leukocytes
Increased granulocytes
No abnormality
Increased colonies on colony-forming
cell assay (n = 6)
Malignancy (n = 13)
Number of founders
showing feature
9
8
3
8
3
4
4
1
6
3
to the lymphoid compartment by the immunoglobulin heavy
chain enhancer/promoter.45 It is also noteworthy that both
bcr-abl and Tel-PDGFR␤ transgenic mice exhibit the same
histopathology of malignant lymphoid infiltrates surrounding
pulmonary vessels (Figure 11a and Figure 337). Both bcr-abl
and Tel-PDGFR␤ transgenic mice have a myeloproliferative
chronic phase that often evolves into lymphoid malignancy.
These similarities suggest that bcr-abl and Tel-PDGFR␤ share
similar mechanisms and/or similar targets for transformation,
and that like bcr-abl, myeloproliferation that evolves into
lymphoid malignancy is an intrinsic property of the TelPDGFR␤ fusion gene.
The Tel-PDGFR␤ and bcr-abl mice demonstrate some differences in their phenotype. The bcr-abl mice displayed peripheral blood granulocytosis and/or leukemia, while the TelPDGFR␤ mice did not show significant abnormalities in their
peripheral blood. The myeloproliferative syndrome of the TelPDGFR␤ mice manifests itself primarily as histopathologic
changes in spleen and bone marrow. The majority of the bcrabl transgenic mice developed a malignancy within a short
time, while a minority of the Tel-PDGFR␤ transgenic mice
developed a malignancy only after a long latent period. These
differences between the Tel-PDGFR␤ and bcr-abl mice probably reflect differences in the relative oncogenicity of these
two fusion genes, or in the promoters used to express them in
transgenic mice. Introduction of additional oncogenic
mutations (‘second hits’) by either physical or chemical
exposure or breeding with transgenic mice harboring other
oncogenes might accelerate the development of malignancy
in Tel-PDGFR␤ mice.
The similarities between the Tel-PDGFR␤ and bcr-ab1 mice
suggest that the activity of their tyrosine kinases are important
in producing the myeloproliferative phenotype. The TelPDGFR␤ and the bcr-abl fusion genes share some important
features: the genes at N-terminal portions (bcr and Tel) were
discovered solely by their participation in their respective
translocations, and the C-terminal portions (abl and PDGFR␤)
are tyrosine kinases that participate in multiple signalling
pathways and also signal to ras.2,23,24,46,47 In addition, both of
these fusion genes are associated with human myeloproliferative syndromes. v-fms, the viral homologue of the normal
cellular proto-oncogene c-fms (the macrophage colony-stimulating factor-1 (MCSF-1) receptor), belongs to the same structural family as PDGFR␤.23 v-fms also produces chronic myeloproliferative syndromes and hematopoietic malignancies of
different lineages in transgenic mice.48 These observations,
plus the similar phenotypes in human myeloproliferative syndromes and in transgenic mice, suggest that v-fms, TelPDGFR␤ and bcr-ab1 share similar mechanisms.
Abnormal tyrosine kinase activity may account for the histologic and immunophenotypic abnormalities seen in the
spleen, peripheral blood leukocytes and bone marrow of these
mice. Aberrant tyrosine kinase activity acting through ras signalling pathways may deregulate growth control by either
increasing intracellular signals to proliferate or hindering normal feedback signals that restrict proliferation. The CFC results
demonstrate that, like human myeloproliferative syndromes,
Tel-PDGFR␤ transgenic mice have an intrinsic difference in
hematopoietic progenitors, evidenced by increased numbers
on soft agar cloning assays. Increased numbers of CFCs suggest that increased numbers of hematopoietic progenitors and
their progeny ultimately pack the marrow and produce extramedullary hematopoiesis. Similarly, since myeloid and
lymphoid lineages have distinct kinase signalling pathways,
one hypothesis to explain the biphenotypism seen in bcr-ab1
and Tel-PDGFR␤ transgenic mice as well as human leukemias, is that ectopic tyrosine kinase activity allows excessive
‘cross-talk’ between signalling pathways.49,50
The TEL-PDGFR␤ fusion gene is associated with a pre-leukemic syndrome that progresses to myeloid leukemia in
humans, but lymphoid malignancies in transgenic mice. This
difference may reflect subtle differences between human and
mouse hematopoiesis, an interspecies difference in the transformation of lymphoid or myeloid lineages, the nature of
‘second hits’ between mouse and man, the activity of
endogenous retroviruses present in the murine genome, differences in signalling pathways, or the timing of transgene
expression. Although the human CD11a promoter expresses
transgenes relatively early in hematopoiesis (manuscript in
preparation), it is not known when during hematopoiesis the
tel-promoter-driven fusion gene is expressed. In general, the
same genetic abnormality often produces similar, but not
identical, phenotypes in humans and transgenic mice. Differences in the timing of expression were demonstrated when
the human fetal globin gene behaved as an embryonic globin
gene in transgenic mice.51,52 A bcl-2-Ig construct isolated from
the t(14,18) translocation in human, monoclonal, malignant
follicular lymphoma produces a polyclonal lymphoproliferation that does not progress to monoclonal malignancy in
transgenic mice.53 Whereas mutations in dystrophin produce
Duchenne muscular dystrophy in humans, mdx mice compensate for the lack of dystrophin, suffer only a transient
myopathy, and live a normal lifespan without disease.54,55
1801
Tel-PDGFR␤ transgenic mice develop a myeloproliferative syndrome
KA Ritchie et al
1802
Such discrepancies are a feature of the transgenic approach
to understanding human disease, and further investigation of
these differences often elucidates the underlying mechanisms
of disease. Further investigation of the TEL-PDGFR␤ transgenic mice with respect to the possible mechanisms mentioned above may account for the TEL-PDGFR␤ fusion gene
association with myeloid leukemia in humans, but lymphoid
malignancies in transgenic mice.
Patients with the t(5;12) translocation exhibit a syndrome
characterized by splenomegaly, eosinophilia, monocytosis
and absence of the Philadelphia chromosome, and have been
diagnosed as CMML. However, it is not clear whether this
syndrome represents CMML-myelodysplastic or CMML-myeloproliferative. It has been suggested that the syndrome represents a clinical entity with features of both CMML and
CML.4,56 Since the diagnosis of human myelodysplasia
depends on recognition of specific dysmorphologies in hematopoietic elements (for example, binucleated erythroid presursors, Pseudo-Pelger–Huet neutrophils, micromegakaryocytes),
and since the morphologies of mouse granulocytes differ from
human granulocytes, and because a mouse MDS has never
been described, the degree to which the changes seen in TelPDGFR␤ mice represent a myelodysplastic syndrome is difficult to assess. The Tel-PDGFR␤ mice exhibit splenomegaly,
chronic myeloid or megakaryocytic proliferation in the spleen
and bone marrow, effacement of splenic architecture,
increased bone marrow CFCs, decreased T cells, and
biphenotypic cells that co-express myeloid and lymphoid
markers, and development of myeloid, lymphoid and
biphenotypic malignancies. Therefore, the Tel-PDGFR␤ mice
demonstrate features of hematopoietic dysregulation more
consistent with a myeloproliferative syndrome, and suggest
that the human t(5;12) syndrome might be best classified as a
myeloproliferative rather than a myelodysplastic disorder.
Acknowledgements
This work was supported by National Institutes of Health
grants K08HL02959 (NHLBI; KAR) and DK48708-02 (NIDDK;
SHB). DDH was supported by the Veterans Affairs Career
Development and Merit Review programs, and the National
Institutes of Health. The authors wish to acknowledge the
expert technical assistance of Lisa Embree, the photographic
expertise of Dale Tilly and Eden Palmer, and the statistical
help of Dan Bankson. We would also like to thank Ursula
Storb, Department of Molecular Genetics and Cellular
Biology, University of Chicago for the IgH J region probe, and
Dennis Willerford, Department of Medicine, University of
Washington for the TCR␤ probe.
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
References
1 Brunning RD. Myelodysplastic syndromes. In: Knowles DM (ed).
Neoplastic Hematopathology. Williams and Wilkins: Baltimore,
1992, pp 1367–1404.
2 Greenberg P, O’Brien S, Goldman J, Sawyers C. Myelodysplatic
syndromes and myeloproliferative disorders: clinical, therapeutic
and molecular advances. In: McArthur JR (ed). Hematology 1995,
Education Program American Society of Hematology. University
of Washington: Seattle, 1995, pp 10–21.
3 Vardiman JW. Chronic myelogenous leukemia and the myeloproliferative disorders. In: Knowles DM (ed). Neoplastic Hematopathology. Williams and Wilkins: Baltimore, 1992, pp 1405–1438.
4 Wessels JW, Fibbe WE, van der Keyr D, Landegent JE, van der
Plas DC, den Ottolander GJ, Roozendaal KJ, Beverstock GC.
20
21
22
t(5;12)(q13;p12): a clinical entity with features of both myeloid
leukemia and chronic myelomonocytic leukemia. Cancer Genet
Cytogenet 1993; 65: 7–16.
Golub TR, Barker GF, Lovett M, Gilliland DG. Fusion of PDGF
receptor ␤ to a novel ets-like gene, tel, in chronic myelomonocytic
leukemia with t(5;12) chromosomal translocation. Cell 1994; 77:
307–316.
Wlodarska I, Mecucci C, Marynen P, Guo C, Franckx D, La Starza
R, Aventin A, Bosly A, Martelli MF, Cassiman JJ, Van den Berghe
H. TEL gene is involved in myelodysplastic syndromes with either
the typical t(5;12)(q33;p13) translocation or its variant
t(10;12)(q24;p13). Blood 1995; 85: 2848–2852.
Yarden Y, Escobedo JA, Kuang W-J, Yang-Feng TL, Daniel TO,
Tremble PM, Chen EY, Ando ME, Harkins RN, Franke U, Fried
VA, Ullrich A, Williams LT. Structure of the receptor for plateletderived growth factor helps define a family of closely related
growth factor receptors. Nature 1986; 323: 226–232.
Carroll M, Tomasson MH, Barker GF, Golub TR, Gilliland DG.
The TEL/platelet-derived growth factor ␤ receptor (PDGFR␤) fusion
in chronic myelomonocytic leukemia is a transforming protein that
self-associates and activates PDGFR␤ kinase-dependent signalling
pathways. Proc Natl Acad Sci USA 1996; 93: 14845–14850.
Jousset C, Carron C, Boureux A, Quang CT, Oury C, DusanterFourt I, Charon M, Levin J, Bernard O, Ghysdael J. A domain of
TEL conserved in a subset of ETS proteins defines a specific oligomerization interface essential to the mitogenic properties of the
TEL-PDGFR␤ oncoprotein. EMBO J 1997; 16: 69–82.
Wasylyk B, Hahn SL, Giovane A. The ets family of transcription
factors. Eur J Biochem 1993; 211: 7–18.
Macleod K, Leprince D, Stehelin D. The ets gene family. TIBS
1992; 17: 251–256.
Golub T, McLean T, Stegmaier K, Ritz J, Sallan S, Neuberg D,
Gilliland DG. TEL-AML1: the most common gene rearrangement
in childhood ALL. Blood 1995; 86: 597a.
Cave H, Baens M, Raynaud S, Vilmer E, Marynen P, Grandchamp
B. Study of a patient with acute lymphoblastic leukemia points
out to tel as the target of the deletions at chromosome 12p12–13.
Blood 1995; 86: 36a.
Wlodarska I, Marynen P, LaStarza R, Mecucci C, Uyttebroeck A,
Verhoef G, Berghe VD. TEL rearrangements in a childhood acute
leukemia associated with a t(7;12)(p15;p13). Blood 1995; 86:
787a.
Buijs A, Sherr S, van Baal S, van Bezouw S, van der Plas D, Geurts
van Kessel A, Riegman P, Lakanne-Deprez R, Zwarthoff E, Hagemeijer A. Translocation (12;22)(p13;q11) in myeloproliferative disorders results in fusion of the ETS-like TEL gene on 12p13 to the
MN1 gene of 22q11. Oncogene 1995; 10: 1511–1519.
Papadopoulos P, Ridge SA, Boucher CA, Stocking C, Wiedemann
LM. The novel activation of ABL by fusion to an ets-related gene
TEL. Cancer Res 1995; 55: 34–38.
Golub TR, Barker GF, Bohlander SK, Stegmaier K, Bray-Ward P,
Ward DC, Sato Y, Suto Y, Rowley JD, Gilliland DG. Evidence
implicating the tel gene at 12p13 in multiple mechanisms of leukemogenesis. Blood 1994; 84: 229a.
Romana SP, Poirel H, Macintyre EA, Berger R, Bernard OA. TELAML1 gene fusion is a common occurrence in human childhood
B lineage acute lymphoblastic leukemia. Blood 1995; 86: 777a.
Shurtleff SA, Buijs A, Behm FG, Rubnitz JE, Raimondi SC, Hancock
ML, Head DR, Pui C-H, Grosveld G, Downing JR. TEL/AML1
fusion occurs in 28% of pediatric B-progenitor cell acute lymphoblastic leukemia (ALL) and defines a subgroup of patients with
a favorable prognosis. Blood 1995; 86: 598a.
Ross TS, Bernard OA, Berger R, Gilliland DG. Fusion of Huntingtin
interacting protein 1 to platelet-derived growth factor ␤ receptor
(PDGFR␤) in chronic myelomonocytic leukemia with
t(5;7)(q33;q11.2). Blood 1998; 91: 4419–4426.
Abe A, Emi N, Tanimoto M, Terasaki H, Marunouchi T, Saito H.
Fusion of the platelet-derived growth factor receptor beta to a
novel gene CEV14 in acute myelogenous leukemia after clonal
evolution. Blood 1997; 90: 4271–4281.
Yan X-Q, Brady G, Iscove NN. Platelet-derived growth factor
(PDGF) activates primitive hematopoietic precursors (preCFCmulti) by up-regulating IL-1 in PDGF receptor-expressing
macrophages. J Immunol 1993; 150: 2440–2448.
Tel-PDGFR␤ transgenic mice develop a myeloproliferative syndrome
KA Ritchie et al
23 Fantl WJ, Johnson DE, Williams LT. Signalling by receptor tyrosine
kinases. Annu Rev Biochem 1993; 62: 453–481.
24 McCormick F. How receptors turn ras on. Nature 1993; 363:
15–16.
25 Ritchie KA, Aprikian A, Gollahon KA, Hickstein DD. The human
leukocyte integrin CD11a promoter directs expression in leukocytes of transenic mice. Blood 1995; 86: 147–155.
26 Gronwald RGK, Grant F, Haldeman BA, Hart CE, O’Hara PJ,
Hagen FS, Ross R, Bowen-Pope DF, Murray MJ. Cloning and
expression of a cDNA coding for the human platelet-derived
growth factor receptor: evidence for more that one receptor class.
Proc Natl Acad Sci USA 1988; 85: 3435–3439.
27 Palmiter RD, Sandgren EP, Avarbock MR, Allen DD, Brinster RL.
Heterologous introns can enhance expression of transgenes in
mice. Proc Nat Acad Sci USA 1991; 88: 478–482.
28 Mishell BB, Shiigi SM. Selected Methods in Cellular Immunology.
WH Freeman: San Francisco, 1980, p 486.
29 Iida H, Seifert R, Alpers CE, Gronwald RGK, Phillips PE, Pritzl PE,
Gordon K, Gown AM, Ross R, Bowen-Pope DF, Johson RJ. Platelet
derived growth factor (PDGF) and PDGF receptor are induced in
mesangial proliferative nephritis in the rat. Proc Natl Acad Sci USA
1991; 88: 6560–6564.
30 Sheehan DC, Hrapchak BB. Theory and Practice of Histotechnology, 2nd edn. CV Mosby: St Louis, 1980.
31 Sitnicka E, Ruscetti FW, Priestly GV, Wolf NS, Bartelmez SH.
Transforming growth factor ␤1 directly and reversibly inhibits the
initial cell divisions of long-term repopulating hematopoietic stem
cells. Blood 1996; 88: 82–88.
32 Sambrook J, Fritsch EF, Maniatis T. Molecular Cloning, A Laboratory Manual, 2nd edn. Cold Spring Harbor: Cold Spring Harbor
Press, 1989.
33 Marcu KB, Banerji J, Penncavage NA, Lang R, Arnheim N. 5⬘
Flanking region of immunoglobulin heavy chain constant region
genes displays length heterogeneity in germlines of inbred mouse
strains. Cell 1980; 22: 187–196.
34 Hood L, Koop BF, Rowen L, Wang K. Human and mouse T-cellreceptor loci: the importance of comparative large-scale DNA
sequence analysis. Cold Spring Harbor Symp Quant Biol 1993;
58: 339–348.
35 Koop BF, Hood L. Striking sequence similarity over almost 100
kilobaes of human and mouse T-cell receptor DNA. Nat Genet
1994; 7: 48–53.
36 Bannerman RM. Hematology. In: Foster HL, Small JD, Fox JG (ed).
The Mouse in Biomedical research Volume 3, Normative Biology,
Immunology, and Husbandry. Academic Press: New York, 1983,
pp 294–308.
37 Honda H, Oda H, Suzuki T, Takahashi T, Witte ON, Ozawa K,
Ishikawa T, Yazaki Y, Hirai H. Development of acute lymphoblastic leukemia and myeloproliferative disorder in transgenic mice
expressing p210.bcr/ab1 a novel transgenic model for human Ph1
positive leukemias. Blood 1998; 91: 2067–2075.
38 Bain BJ, Leukemia Diagnosis, A Guide to the FAB Classification.
JB Lippincott Philadelphia, 1990.
39 Fialkow PJ, Jacobson RJ, Papayannopoulou T, Penfold GK, Jacobson RT, Hansen JA. Chronic myelocytic leukemia: clonal origin
in a stem cell common to the granulocyte, erythrocyte, platelet,
and monocyte/macrophage. Am J Med 1977; 63: 125–132.
40 Akashi K, Taniguchi S, Nagafuji K, Harada M, Shibuya T, Hayashi
S, Gondo H, Niho Y. B-lymphoid/myeloid stem cell origin in Phpositive acute leukemia with myeloid markers. Leukemia Res
1993; 17: 549–557.
41 Akashi K, Mizuno S, Harada M, Kimura N, Kinjyo M, Shibuya T,
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
Shimoda K, Takeshita M, Okamura S, Matsumoto I, Kikuchi M,
Niho Y. T-lymphoid/myeloid bilineal crisis in chronic myelogenous leukemia. Exp Hematol 1993; 21: 743–751.
Daley GQ, Van Etten RA, Baltimore D. Induction of chronic myelogenous leukemia in mice by the P210 bcr/abl gene of the Philadelphia chromosome. Science 1990; 247: 824–830.
Kelliher MA, McLaughlin J, Witte ON, Rosenberg N. Induction of
a chronic myeogenous leukemia-like syndrome in mice with vabl and BCR/ABL. Proc Natl Acad Sci USA 1990; 87: 6649–6653.
Voncken JW, Kaartinen V, Pattengale PK, Germeraad WTV,
Groffen J, Heisterkamp N. BCR/ABL P210 and P190 cause distinct
leukemia in transgenic mice. Blood 1995; 86: 4603–4611.
Tomasson MH, Williams IR, Hasserjian R, Udomsakdi C, McGrath
SM, Schwaller J, Druker B, Gilliland DG. TEL/PDGF␤R induces
hematologic malignancies in mice that respond to a specific tyrosine kinase inhibitor. Blood 1999; 93: 1707–1714.
Satoh T, Fantl WJ, Escobedo JA, Williams LT, Kaziro Y. Plateletderived growth factor receptor mediates activation of ras through
different signalling pathways in different cell types. Mol Cell Biol
1993; 13: 3706–3713.
Srivastava A, Boswell HS, Heerema NA, Nahreini P, Lauer RC,
Antony AC, Hoffman R, Tricot G. KRAS2 oncogene overexpression in myelodysplastic syndrome with translocation (5;12).
Cancer Genet Cytogenet 1988; 35: 61–71.
Heard JM, Roussel MF, Rettenmier CW, Sherr CJ. Multilineage
hematopoietic disorders induced by transplantation of bone marrow cells expressing the v-fms oncogene. Cell 1987; 51: 663–673.
Ilaria RL, Van Etten RA. Constitutive activation of stat family members in BaF3 cells transformed by p210 BCR/ABL. Blood 1995;
86: 264a.
Carlesso N, Frank DA, Griffin JD. Tyrosyl phosphorylation and
DNA binding activity of STAT proteins in hematopoietic cells
transformed by p210 BCR/ABL. Blood 1995; 86: 264a.
Chada K, Magram J, Constantini F. An embryonic pattern of
expression of a human fetal globin gene in transgenic mice. Nature 1986; 319: 685–689.
Kollias G, Wrighton N, Hurst J, Grosveld F. Regulated expression
of human A␥-, ␤-, and hybrid ␥␤-globin genes in transgenic mice:
manipulation of the developmental expression patterns. Cell
1986; 46: 89–94.
McDonnell TJ, Deane N, Platt FM, Nunez G, Jaeger U, McKearn
JP, Korsmeyer SJ. bcl-2-immunoglobulin transgenic mice demonstrate extended B cell survival and follicular lymphoproliferation.
Cell 1989; 57: 79–88.
Deconinck AE, Rafael JA, Skinner JA, Brown SC, Potter AC, Metzinger L, Watt DJ, Dickson JG, Tinsley JM, Davies KE. Utrophin
dystrophin-deficient mice as a model for Duchenne muscular dystrophy. Cell 1997; 90: 717–727.
Grady RM, Teng H, Nichol MC, Cunningham JC, Wilkinson RS,
Sanes JR. Skeletal and cardiac myopathies in mice lacking utrophin and dystrophin: a model for Duchenne muscular dystrophy.
Cell 1997; 90: 729–738.
Baranger L, Szapiro N, Gardais J, Hillon J, Derre J, Francois S,
Blanchet O, Boasson M, Berger R. Translocation t(5;12)(q31–
33;p12–13): a non-random translocation associated with a
myeloid disorder with eosinophilia. Br J Haematol 1994; 88:
343–350.
Reference added in proof
57 Bain BJ. Leukemia Diagnosis: A Guide to the FAB Classification.
JB Lippincott: New York, 1990.
1803