Oncogene (2004) 23, 665–678
& 2004 Nature Publishing Group All rights reserved 0950-9232/04 $25.00 www.nature.com/onc
Myeloid leukemia with promyelocytic features in transgenic mice expressing hCG-NuMA-RARa Mahadeo A Sukhai1,2, Xuemei Wu3, Yali Xuan2, Tong Zhang2, Patricia P Reis2, Karina Dube´2, Eduardo M Rego3, Mantu Bhaumik3, Denis J Bailey2,4, Richard A Wells3, Suzanne Kamel-Reid*,1,2,4,5 and Pier Paolo Pandolfi3 1
Department of Medical Biophysics, University of Toronto, Toronto, Ontario, Canada; 2Department of Cellular and Molecular Biology, the Ontario Cancer Institute/University Health Network, Toronto, Ontario, Canada; 3Molecular and Developmental Biology Lab, Molecular Biology Program and Department of Pathology, Memorial Sloan Kettering Cancer Center, Sloan Kettering Institute, New York, NY, USA; 4Department of Pathology, The Ontario Cancer Institute/University Health Network, Toronto, Ontario, Canada; 5Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Ontario, Canada
Acute promyelocytic leukemia (APL) is characterized by the accumulation of abnormal promyelocytes in the bone marrow (BM), and by the presence of a reciprocal chromosomal translocation involving retinoic acid receptor alpha (RARa). To date, five RARa partner genes have been identified in APL. NuMA-RARa was identified in a pediatric case of APL carrying a translocation t(11;17)(q13;q21). Using a construct containing the NuMA-RARa fusion gene driven by the human cathepsin G promoter (hCG-NuMA-RARa), two transgenic mouse lines were generated. Transgenic mice were observed to have a genetic myeloproliferation (increased granulopoiesis in BM) at an early age, and rapidly developed a myeloproliferative disease-like myeloid leukemia. This leukemia was morphologically and immunophenotypically indistinguishable from human APL, with a penetrance of 100%. The phenotype of transgenic mice was consistent with a blockade of neutrophil differentiation. NuMARARa is therefore sufficient for disease development in this APL model. Oncogene (2004) 23, 665–678. doi:10.1038/sj.onc.1207073 Keywords: acute promyelocytic leukemia; RARa; transgenic model; immunophenotype
NuMA-
Introduction Acute promyelocytic leukemia (APL) is characterized by accumulation of hematopoietic cells with promyelocytic features (Bennett et al., 1976), association with balanced chromosomal translocations involving the retinoic acid receptor alpha (RARa) locus on chromosome 17q21 (Rowley et al., 1977a, b), and, in most cases, sensitivity of the APL blasts to differentiation upon treatment with all trans-retinoic acid (ATRA) (Grignani et al., 1994). *Correspondence: S Kamel-Reid, Princess Margaret Hospital/The Ontario Cancer Institute, Room 9-622, 610 University Avenue, Toronto, ON, Canada M5G 2M9; E-mail:
[email protected] Received 13 June 2003; revised 25 July 2003; accepted 4 August 2003
Leukemic clones accumulate in the bone marrow (BM) and inhibit the growth of normal cells, resulting in pancytopenia (Harmon, 1991). In more than 95% of APL cases, the translocation t(15;17) results in an in-frame fusion of the promyelocytic leukemia (PML) gene to RARa (de The´ et al., 1990, 1991). Four other rare RARa fusion partner genes, collectively referred to as ‘X,’ have been identified: PLZF (Chen et al., 1993), nucleophosmin (NPM) (Redner et al., 1996; Hummel et al., 1999), NuMA, the nuclear mitotic apparatus protein (Wells et al., 1996, 1997), and STAT5b (Arnould et al., 1999). Thus, disruption of RARa, a nuclear hormone receptor and transcription factor involved in embryonic development and granulopoiesis (Labrecque et al., 1998; Mark et al., 1999), is thought to be the underlying mechanism of disease formation in APL (Melnick and Licht, 1999). The variant APL fusion gene NuMA-RARa was identified (Wells et al., 1996, 1997) in a pediatric case of APL, initially considered to have a myeloproliferative disorder (MPD) and found to carry the translocation t(11;17)(q13;q21). NuMA-RARa retains the N-terminal globular ‘head’ domain and long central coiled-coil motif of NuMA, including NuMA’s oligomerization domain (Compton et al., 1992). The fusion protein also contains, in its C-terminus, domains C–F of RARa, including the ATRA-binding site, RA-dependent DNAbinding site, and retinoid X receptor alpha and nuclear corepressor/coactivator interfaces (Melnick and Licht, 1999). NuMA-RARa lacks the ligand-independent transactivation domain of RARa and most of the C-terminal globular domain of NuMA (Wells et al., 1997). Wild-type (WT) NuMA protein is ubiquitously and highly expressed (200 000 copies per cell), and plays a key role in mitotic progression (Compton and Cleveland, 1993). It has been suggested that, while the formation of a dominant-negative form of RARa is the major causative event in APL, the N-terminal portion of the fusion protein may help determine the overall disease phenotype (reviewed in Melnick and Licht, 1999). This hypothesis may be examined in more detail by
NuMA-RARa transgenic mice develop APL MA Sukhai et al
666
comparative phenotype analysis of APL transgenic mice (TMs). Several transgenic models of APL have been previously characterized, using PML-, PLZF-, and NPM-RARa (Early et al., 1996; Brown et al., 1997; David et al., 1997; Grisolano et al., 1997; He et al., 1997, 1998, 1999; Westerveldt and Ley, 1998; Chen et al., 1999; Rego and Pandolfi, 2001). Two different promoters have been used to develop PML-RARa TMs, the human cathepsin G (hCG) and hMRP8 promoters, while only the hCG promoter has been used in the in vivo analysis of PLZF- and NPM-RARa. Mice expressing hMRP8-PML-RARa developed a more severe end phenotype (a myeloid leukemia without maturation vs myeloid leukemia with maturation in hCG-PML-RARa mice) with shorter onset than did mice expressing hCGPML-RARa (Brown et al., 1997; Grisolano et al., 1997; He et al., 1997; Kogan et al., 2002). However, a direct phenotype comparison is possible between all hCG-XRARa models: Mice carrying hCG-X-RARa developed leukemias with variable penetrance and onset. Since the features of the leukemias observed in the individual transgenic systems were not identical, it is reasonable to conclude that the fusion partners of RARa may affect the phenotypes of the mice. To examine the role of NuMA-RARa in APL pathogenesis, we developed a TM model of NuMARARa. Our goals were to characterize the phenotype of hCG-NuMA-RARa TMs, to ascertain the immunophenotype of their leukemic cells (if present), to determine whether NuMA-RARa TMs are a good model system for APL, and to compare their phenotype to previous APL models and to the index patient.
rapidly progressed to a nonfatal myeloproliferative disease-like myeloid leukemia with promyelocytic features in all animals (median latency 6 months). Furthermore, this phenotype progressed to a myeloid
a B
b
E N
S
NuMA-RARa
Rd F2
hCG mini gene + 3' flanking Cat (b) B2 CT
M W WT W W WT T T T T + - 0
NuMA-RARα 251 bp
c
M T T T T T T T T T T W T WT W T T T + - 0 M
F2 Mice NuMA-RARα 251 bp
M WWWW W W WWWW W T T T T T T T T T + - 0
NuMA-RARα 251 bp
d
3000
2500 Bone Marrow Expression of NuMA-RARα
2000
1500
1000
500
0 TM founder #2
WT
e
TM, founder #1
3500 2 month old 5 month old
3000 2500 2000 1500 1000 500
Oncogene
iu
ey
rd
dn Ki
en
ca
le
yo M
r
ng
Sp
Lu
ve Li
TMs were observed to carry a genetic myeloproliferation (median latency 8 weeks, 100% frequency) that
In te st in Sk e M ele us ta cl l e Th ym us
0 m
Overview of TM phenotype
M
F3 Mice
Relative Expression
Two founder TMs were generated with the full-length NuMA-RARa cDNA, under the control of hCG regulatory sequences (a schematic of the construct is shown in Figure 1a). The transgene’s expression is thus targeted only to immature cells of the neutrophil lineage. The presence and expression of hCG-NuMA-RARa in the two founder lines was confirmed by PCR and Southern blot (data not shown). Figure 1b illustrates PCR genotyping of F1 mice descended from both founders, demonstrating the integration of hCGNuMA-RARa into the genomes of mice. All subsequent generations of mice retained copies of the transgene (shown in Figure 1c for representative F2 and F3 mice). All TMs expressed NuMA-RARa in the BM (Figure 1d). No expression of the transgene was detected in organs of 2-month-old TMs. Transgene expression was, however, detected by RT–PCR in multiple organs (liver, lungs, spleen, heart and skeletal muscle, kidney, intestine, thymus) in mice 4.5 months of age and older (Figure 1e).
C
hCG promoter + 5' flanking
Results Construction and genotyping of mice
E
NuMA-RARa transgenic mice develop APL MA Sukhai et al
667
leukemia with maturation in all mice older than 10 months of age. No mice were observed to die from this disease, even when followed to 21 months of age. There were no differences in the phenotype of leukemias observed in mice descended from the two founder animals. BM phenotype Transgenic BM was hypercellular, with increased granulopoiesis, based on the histological analysis of hematoxylin and eosin (H&E)-stained hindlimb bone sections (data not shown). Granulocytic maturation was still observed in the marrow, despite the significant presence of blasts. This was consistent with the observations of the index patient (Wells et al., 1996). Cells of the granulocytic series constituted a median of 45% of BM cellularity in 8-week-old TMs, with a median of 8% being promyelocytes. This is in contrast to 25% granulocytes and 2% promyelocytes in agematched controls. TMs older than 8 weeks of age exhibited progression of the phenotype (Figure 2a), showing increasingly elevated numbers of promyelocytes and granulocytes in their BM. Mice 21 months old had 33% promyelocytes and 80% granulocytes in their BM, with morphology as shown in Figure 2b. Flowcytometry analysis of transgenic BM corroborated these findings, showing progressively greater percentages of cells expressing the neutrophil marker Gr-1, the myeloid marker CD11b, and the stem cell marker CD117 with increasing age (Figure 2d, WT vs TM BM, 1, 11, and 21 months of age). Coincident with this, we observed an abnormal cell population in the BM (Figure 2c, shown for mice 21 months of age), which possessed moderately decreased CD45 expression levels relative to mature neutrophils. :——————————————————————— Figure 1 (a) Schematic of the hCG-NuMA-RARa transgene. (b) TMs carry hCG-NuMA-RARa in their genomes. PCR genotyping of five representative wild-type (W) and six representative transgenic (T) F1 mice derived from both transgenic founder lines, showing integrations of the hCG-NuMA-RARa transgene, are present in TM genomes. M ¼ 100 bp marker, ‘ þ ’ ¼ positive DNA control, ‘’ ¼ negative DNA control, ‘0’ ¼ no template control. (c) Vertical transmission of transgene from the parent to offspring is evident in TMs. PCR genotyping gels of representative F2 and F3 mice, demonstrating that the transgene is present in the germ line, and thus can be stably transmitted through multiple generations. WT mice are indicated by W, transgenic by T. M ¼ 100 bp marker, ‘ þ ’ ¼ positive DNA control, ‘’ ¼ negative DNA control, ‘0’ ¼ no template control. (d) TMs express NuMA-RARa in their BM. Quantitative RT–PCR of BM, demonstrating the presence of NuMA-RARa mRNA transcripts in the BM of TMs, but not in any WT animals. Expression levels of NuMA-RARa in the BM of six representative TM from both founders is shown, relative to two WT mice. (e) TMs 44 months of age exhibit transgene expression at multiple sites outside the BM. Quantitative RT–PCR of RNA from representative TM (2 and 21 months) organs (liver, lung, spleen, heart and skeletal muscle, kidney, intestine, and thymus) and BM, demonstrating the expression of NuMA-RARa in multiple tissues of the older TM, but not the younger animal. Mouse-tomouse variation in the identity of organs affected and in the level of transgene expression was observed in all animals studied. Expression was calculated relative to the RNA sample with the lowest expression level
Peripheral blood (PB) phenotype In contrast to their BM phenotype, TMs did not display any differences in PB leukocyte composition (Figure 3a) or flow-cytometry profile, relative to WT mice, until 44 months of age. However, all TMs over 4 months of age developed a PB MPD, characterized by persistent neutrophilia, with mild leukocytosis, no anemia, and moderate thrombocytopenia (Table 1). This condition increased in severity with time, as seen by an increase in the abundance of myeloid forms in the PB (Figure 3b: increased myeloid/lymphoid (My/Ly) ratio), and by the appearance of abnormal myeloid forms and blasts (data not shown). Flow-cytometry analysis of PB from mice with MPD indicated increased proportions of cells expressing the myeloid markers Gr-1 and CD11b, but no increase in the proportion of cells expressing CD117 (Figure 3d). Mice that demonstrated 420% promyelocytes in the BM also showed further progression of hematologic changes in the PB, with 450% neutrophils, 5–10% promyelocytes (Figure 3d; Table 1), and the emergence of a population of Gr-1 þ /CD11b þ /CD117 þ cells in the PB (Figure 3c). These changes were associated with the most severe form of phenotype observed in the TMs. Immunophenotype of leukemic cells The abnormal cell populations observed in both PB and BM possessed moderately decreased CD45 expression relative to mature neutrophils (compare Figures 2c and 3c). Based on forward- and side-scatter characteristics, these cells were identified as relatively large cells with low to moderate cytoplasmic complexity, closely matching the characteristics of human APL blasts (Weir and Borowitz, 2001). These cells have a cellsurface immunophenotype that is highly analogous to the human APL phenotype (Paietta et al., 1994; Rizzatto et al., 2002), differing only in the expression of CD11b (Table 2). Previous APL transgenic models have also demonstrated CD11b þ leukemias (Brown et al., 1997; He et al., 1997), in comparison to human APLs, which are generally CD11b, possibly indicating a basic difference between human and murine APLs. In vitro assays of hematopoietic progenitor function In order to enumerate the functionality of WT vs transgenic hematopoietic progenitors, the BM from WT and TM was also analysed by colony-forming unit (CFU) assays and long-term marrow cultures (LTMCs). TMs displayed decreased total viable CFU (Figure 4a), an increased fraction of CFU-GM (Figure 4b: 70% in WT CFU assays, compared to 490% in TM CFU assays), decreased BFU-E and CFU-GEMM, and increased numbers of abortive colonies (Figure 4a). Although there was increased proliferation of CFU-GM (number and size of colonies) in response to G-CSF and GM-CSF in WT mice, hematopoietic progenitors from TMs failed to proliferate on treatment with G-CSF, and demonstrated decreased proliferation in response to Oncogene
NuMA-RARa transgenic mice develop APL MA Sukhai et al
668
GM-CSF (Figure 4c). These defects in cytokine response were overcome by cotreatment of transgenic CFUs with pharmacological concentrations (106 M) of ATRA
(Figure 4d, shown for GM-CSF). ATRA treatment alone failed to induce proliferation, but caused the differentiation of CFU-GM into mature granulocytes
b
a
35
% Promyelocytes
30 25 (i) WT 400X
(ii) TM 400X
(iii) TM 1000X
(iv) TM 1000X
20 15 10 5 0 2mo.
c
WT
104
Ly
103
Gr FL3-H
FL3-H
TM
104
Ly
103 CD45
6mo. 10mo. 14mo. 21mo.
102
102
101
101
100
100 0
Gr
B
0
1000
1000
Side Scatter/Cytoplasmic Complexity WT
0 100
102
20
FL2-H
0 100
Counts
12.5%
102 FL2-H
Oncogene
23.7%
102
40
0 100
104
104
20
40
10 0 100
102 FL2-H
102
104
0 100
104
70
6.0%
102 FL2-H
100
18.4%
102
104
0 100
40
FL2-H
102
40
104
0 100
104
19.6%
100
102 FL2-H
48.6%
102
104
FL2-H
20 0
104
20
FL2-H
40
102 FL1-H
32.2%
100
40.6%
20 0 100
104
20 0
104
18.2%
102
102
40
FL1-H
FL2-H
20 0 100
0
104
Counts
Counts
70
17.7%
FL2-H
7.6%
102
20
FL1-H
20
FL2-H
Counts
CD117 70
FL1-H
10 0 100
104
0 100
104
25.5%
40
Counts
Counts
16.2%
102
21mo.
Counts
FL1-H
Counts
Mac-1 70
20 0 100
104
20.4%
Counts
FL1-H
102
70
20.6%
11mo.
Counts
0 100
104
40
1mo.
Counts
10
Counts
102
28.2%
Counts
0 100
20
21mo.
Counts
Counts
26.5%
Counts
70
11mo.
Counts
1mo. Gr-1
TM
Counts
d
104
40
30.5%
20 0 100
102 FL2-H
104
NuMA-RARa transgenic mice develop APL MA Sukhai et al
669
(not shown), consistent with previous reports (Kelly et al., 2002). LTMCs are often used as in vitro models of myeloid hematopoiesis, thus allowing us the opportunity to examine in detail the mechanism by which NuMA-RARa may interfere with neutrophil differentiation. LTMCs established from TMs showed expression of NuMARARa (Figure 5a), as well as increased numbers of immature forms (Figure 5b, compare WT with TM; quantitated in Figure 5c). Flow-cytometry profiles of TM LTMCs indicated an immature cell-surface phenotype, as compared to WT LTMCs (Figure 5d), as shown by increased proportions of cells expressing CD117 and CD13. We also tested the responsiveness of these immature forms to treatment with ATRA. In transgenic cultures treated with ATRA, the proportion of morphologically immature forms decreased, while that of mature neutrophils was significantly elevated (Po0.05; Figures 5b and c). Flow-cytometry analysis also showed the emergence of a population of cells with more mature neutrophil characteristics in TM cultures after ATRA treatment (Figure 5d). Organ phenotype Gross visual abnormalities of the lung, kidney, and thymus, along with massively enlarged spleens, were noted in the very oldest leukemic mice (420 months of age), correlating with 430% promyelocytes in the BM and 410% in the PB. Spleens from TMs exhibited enlarged white pulp with large germinal centers, and leukemic infiltrates in the red pulp (not shown). However, livers from TMs were not enlarged, nor were histologic changes apparent. Finally, we examined the white-cell populations of the spleen by flow cytometry. We detected an infiltration of Gr-1 þ , CD11b þ , and CD117 þ cells in the spleens of all TMs, most marked in the oldest transgenic animals studied (data not shown). A hallmark of diagnosis of murine leukemia, as presented by the authors of the Bethesda proposals for community standards in the diagnosis of murine nonlymphoid hematopoietic neoplasms (Kogan et al., 2002), is the infiltration of organs other than BM, PB, and spleen by leukemic cells. Given the presence of major organ abnormalities in the oldest TMs, and the presence of NuMA-RARa transcript expression
in multiple organs of mice 44.5 months of age (Figure 1d), we may characterize hCG-NuMA-RARa TMs as having a genetic myeloproliferation that rapidly progressed to a nonfatal MPD-like myeloid leukemia with promyelocytic features.
Discussion We examined the role of the variant APL fusion oncogene NuMA-RARa in APL pathogenesis using the hCG-NuMA-RARa TM model. The underlying BM phenotype of TMs was a genetic myeloproliferation (Kogan et al., 2002), characterized by increased granulopoiesis, impaired neutrophil differentiation, abnormal cytokine response, and responsiveness to ATRA. Our CFU and LTMC results also clearly demonstrated accumulation of promyelocytes and failure to respond to G-CSF and GM-CSF, characteristic of impaired neutrophil differentiation. The reversion in vitro to a WT morphological and flow-cytometry profile observed in transgenic LTMCs on ATRA treatment is consistent with the index patient’s response to ATRA. This mouse has a phenotype very similar to that of previous APL transgenic models (Grisolano et al., 1997; He et al., 1997, 1998). However, the observed leukemia, while closely resembling that seen in the index patient (Wells et al., 1996), differed from those seen in the previous models in several respects. The disease we observed in our mice was progressive, as seen by increasing levels of promyelocytes and granulopoiesis in the BM, increasing My/Ly ratio in the PB, and cells expressing Gr-1, CD11b, and CD117 in the BM and PB. We therefore postulate a continuous accumulation of cells blocked at the promyelocyte stage of differentiation in the marrow, rather than the acquisition of a second mutation leading to disease, to be the root cause of leukemia in hCG-NuMA-RARa mice. This is in contrast to previous models which have suggested that a secondary mutation is necessary for progression from a genetic myeloproliferation to a fullblown leukemia (reviewed in the supplemental online dataset to Kogan et al., 2002: http://www.bloodjournal. org/cgi/content/full/100/1/238/DC1). Thus, we present a novel model of phenotype progression in our hCG-NuMA-RARa TMs: Mice
:————————————————————————————————————————————————— Figure 2 (a) Increased promyelocyte percentage in the BM of TMs. WT (dark bars) vs transgenic (light bars) promyelocyte percentages are reported for mice aged 2–21 months. Between three and six WT and TM were analysed for each time point. Promyelocyte percentages in TMs increased with age, whereas in WT mice these percentages remained within a narrow range. (b) BM morphology. BM cytospins from mice aged 21 months were prepared and stained with May-Gru¨nwald and Giemsa. (i) Magnification ( 400) of WT mouse BM. (ii) Magnification ( 400) of TM BM. (iii) Magnification ( 1000) of TM BM, showing a preponderance of immature cells that bear resemblance to promyelocytes (indicated in all panels by black arrows). (c) Physical characteristics of transgenic BM cells. CD45 vs side-scatter dot plots are presented for a representative WT and TM pair, aged 21 months, representative of n ¼ 5 WT and n ¼ 5 TM. Lymphoid (Ly), granulocyte (Gr), and leukemic blast cell (B) populations are marked on the plots; the TM displayed elevated numbers of granulocytes and a distinct population of leukemic cells (B, indicated by the circle). (d) BM flow-cytometry analysis of WT vs TMs. All BM samples collected for flow cytometry were washed and resuspended in PBS. Samples were blocked with purified CD16/ CD32 receptor antibodies (BD Pharmingen), and incubated with antibodies (BD Pharmingen) to cell-surface antigens CD45, CD117, Gr-1, and CD11b. Gr-1, CD11b, and CD117 histogram profiles are presented for BM (y-axis: cell counts; x-axis: mean fluorescence intensity). Data are shown for age-matched WT and TM animals 1, 11, and 21 months of age, representative of n ¼ 5 WT and n ¼ 5 TM per time point. The percentages of Gr-1 þ , CD11b þ or CD117 þ cells in the BM samples are indicated on the histograms. All TMs had greater numbers of cells expressing these three markers, a phenomenon that increased with age Oncogene
NuMA-RARa transgenic mice develop APL MA Sukhai et al
670
initially have a genetic myeloproliferation, as described above. As the percentage of promyelocytes in the BM of mice increases, they develop an MPD-like myeloid
a 100
leukemia that manifests as follows: MPD in the PB, as the PB neutrophil cell percentage exceeds 20% (My/Ly ratio 40.33), and expression of NuMA-RARa in multi-
4
b
WT TM
3 My/Ly Ratio
% Leukocyte
80 60 40
2
1
20 0
0 Neutrophil/ Monocyte
12
Lymphocyte
14
16 18 Age (mo.)
WT
c
20
22
TM
104
104
Ly
103
Ly
103
Gr
FL3-H
CD45 FL3-H
Gr 102
Ery
101
102
B
101
Ery
0
0
10
10
0
0
1000
1000
SSC-H
WT
11mo.
21mo.
200 7.5%
12.6%
5.2%
20 0
Counts
Counts
40
0 100
102
100
104
FL1-H
102
1mo.
11mo.
500
0
104
12.0%
40 Counts
Gr-1
20
FL1-H
102
104
18.1%
40
0 100
21mo.
20
20
0 100
FL1-H
102
43.2%
40 Counts
1mo.
TM
Counts
d
Counts
SSC-H
0 100
104
102
104
100
104
FL1-H
FL1-H
FL1-H
102
Mac-1 16.7%
6.1%
5.2%
0 100
102
0 100
104
FL2-H
102
104
Counts
Counts
Counts
20
20
0
FL2-H
102
104
20 0
0 100
100
FL2-H
102
104
44.8%
40
40
40
Counts
500
14.6%
Counts
Counts
200
7.5%
40
20 0
100
102
104
100
FL2-H
FL2-H
102 FL2-H
104
CD117
100
102 FL2-H
Oncogene
4.7%
104
0
100
102 FL2-H
104
0
8.5%
20 0
100
102 FL2-H
104
12.1%
40 Counts
40 Counts
Counts
Counts
Counts
20 0
500 8.7%
20
102 FL2-H
104
20 0
0 100
24.1%
40 Counts
200 6.7%
40
100
102 FL2-H
104
100
102 FL2-H
104
NuMA-RARa transgenic mice develop APL MA Sukhai et al
671 Table 1 Peripheral blood characteristics of hCG-NuMA-RARa mice
Wild-type mice (N ¼ 21) MPD mice (N ¼ 23) Leukemic mice (N ¼ 8)
WBC ( 109/l)
Hb (g/l)
Plt ( 109/l)
% Neutrophils
4.45–12.46 [10.0] 4.5–24.0 [11.5] 10.1–47.0 [15.3]
135.0–235.8 [194.85] 132.0–242.9 [211.55] 160.0–246.0 [212.8]
756–1862 [983.5] 228–1862 [710.0] 552–1416 [792.0]
5–22 [16] 33–48 [36] 50–78 [52.5]
% Blasts o1 o1 5–15 [7.5]
Data are reported for animals at the time of diagnosis of leukemia, along with age-matched MPD and wild-type animals. Values are presented as data ranges (lowest–highest), [median]
Table 2 Immunophenotype of myeloid/leukemic cell population from hCG-NuMA-RARa transgenic mice, in comparison to the human APL immunophenotype (see the text section for description of methods used) Antibody
# Positive mice
Human APL immunophenotype
Gr-1 Mac-1 CD13 CD14 CD19 CD33 CD34 CD117 HLA-DR
9/9 9/9 7/9 1/9 3/9 5/9 5/9 9/9 1/9
Not tested (4%) +(97%) (10%) (11%) +(97%) +(23%) +(77%) (4%)
Numbers in parentheses represent the percentage of human APLs that test positive for the given cell surface marker
ple organs. Further accumulation of promyelocytes in the BM leads to the detection of blasts in the PB, and continued elevation of the My/Ly ratio. Gross organ abnormalities and massive splenomegaly are detected in this most advanced stage of disease. This model of disease progression is built upon an analysis utilizing the community standards for disease definition published in the Bethesda Proposals (Kogan et al., 2002). This is the first use of these standards in a prospective manner. We have shown that hCG-NuMA-RARa TMs have inherently higher levels of promyelocytes in their BM and, furthermore, that these levels increase with the age of the mouse. According to the profile we have established, all TMs between the age of 10 and 14 months will meet the threshold (20% promyelocytes) for the diagnosis of a myeloid leukemia with maturation. Furthermore, we have observed that the phenotype of the PB in our mice correlated with that of the BM, as reflected in our flow-cytometry data presented in Figures
2 and 3. As we show, promyelocytes appear in the PB after B10 months of age. These data can be interpreted in two ways. Either there is a distinct phenotype transition in the PB, due to secondary genetic events; or, the accumulation of promyelocytes in the BM has passed a certain threshold, leading to their appearance in the PB. We favor the second hypothesis, but further studies are required in order to ascertain which of these events has occurred. Five previous APL transgenic models utilizing the hCG regulatory sequences have been previously developed: two of hCG-PML-RARa (Grisolano et al., 1997; He et al., 1997), two of hCG-PLZF-RARa (He et al., 1998; Chen et al., 1999) and one of hCG-NPM-RARa (Chen et al., 1999). There are two differences that distinguish the hCG-NuMA-RARa leukemic phenotype from these previous models. First, unlike the hCGPML-RARa model, significant leukocytosis, anemia, and thrombocytopenia were not observed in NuMARARa TMs. Instead, we observed a mild leukocytosis, no anemia, and a moderate thrombocytopenia (Table 1). This correlates with the characteristics of the patient from whom the fusion gene was cloned (Wells et al., 1996). Furthermore, leukemic mouse BM was hypercellular, with elevated mature neutrophils and an excess of promyelocytes (Figure 2). Granulocytic and blast cell morphology resembled that presented by the human index case (Wells et al., 1996). Although NuMA-RARa has thus far been found in only one patient with APL, the similarity of the PB and BM characteristics of the mouse model to the index patient is intriguing. Strikingly, this phenotype is also similar to that observed in the hCG-PLZF-RARa models (He et al., 1998; Chen et al., 1999), but distinctly different from that of the hCG-PML-RARa models (Grisolano et al., 1997; He et al., 1997), where there was comparatively little evidence of terminal neutrophil differentiation in the
:————————————————————————————————————————————————— Figure 3 (a) TMs exhibit PB leukocyte composition indistinguishable from WT at 4 weeks of age. PB films were prepared and stained according to Materials and methods, and 4100 nucleated white cells were scored as lymphocytes, neutrophils or monocytes. Percentages of neutrophils/monocytes and lymphocytes in the PB of WT (n ¼ 7, dark bars) and transgenic (n ¼ 8 light bars) littermate mice 4 weeks of age. TMs are indistinguishable from WT at this age. (b) Profile of the My/Ly ratio in the PB a of representative WT and TM F1 mice, months #12–21. Data presented are representative of n ¼ 5 WT and n ¼ 6 TM from a single litter of F1 mice. PB films were scored as indicated in Materials and methods; note the increase in the My/Ly ratio of the TM. (c) Defined population of leukemic cells in transgenic PB. CD45 vs side-scatter dot plots are presented for a representative WT and TM pair, aged 21 months. Data are representative of n ¼ 5 WT and n ¼ 5 TM per time point. Lymphoid (Ly), granulocyte (Gr), erythrocyte (Ery), and leukemic blast cell (B, indicated by the red circle) cell populations are marked on the plots; the TM displayed elevated numbers of granulocytes and a distinct population of leukemic cells. (d) PB flow-cytometry profiles for WT and TMs. Flow cytometry on PB samples was carried out as outlined in Materials and methods. Flow-cytometry histograms (y-axis: cell counts; x-axis: mean fluorescence intensity) are presented for mouse PB, from age-matched WT and TM aged 1, 11, and 21 months, showing increasing expression of the markers Gr1, CD11b, and CD117 in the PB of TM Oncogene
NuMA-RARa transgenic mice develop APL MA Sukhai et al
672
BM. As these transgenic models were all developed using the hCG regulatory sequences, and the same strain of mouse (C57BL6), variation in strain and promoter
a
70 60
# Colonies
50 40 30 20 10 0
WT
TM
Viable Aborted
b
70 60
# Colonies
50 40 30 20 10 0 WT
TM
BFU-E CFU-GM CFU-GEMM
c
180
cannot be used to explain these observations. These differences are thus likely caused by the fusion genes themselves. There are structural and functional differences between the fusion partner genes (reviewed in Melnick and Licht, 1999). It has been previously shown that formation of X-RARa disrupts WT fusion partner function. For example, PML is delocalized from nuclear bodies in the presence of PML-RARa (Mu et al., 1994). This is thought to have a series of effects on PML’s roles in the nuclear matrix and in apoptosis (Dyck et al., 1994; Mu et al., 1994; Wang et al., 1998; Zhong et al., 2000). Likewise, the formation of PLZF-RARa leads to a deregulation of PLZF-responsive genes (Shaknovich et al., 1998). However, the majority of these studies were conducted in cell culture systems, and it is unclear how the effects of X-RARa on fusion partner function will modify the phenotype of an APL TM. These effects need to be taken into account in comparing the phenotypes of APL transgenic models. Whether NuMA function is disrupted by the formation of NuMA-RARa is currently under investigation in our laboratory. A comparison of the phenotype of the hCGNuMA-RARa model to the previous APL transgenic models strongly suggests that disruption of NuMA function in transgenic promyelocytes plays a role in the ultimate development of the phenotype observed in our mice. A further difference between the previous models and ours lies in the penetrance and onset of disease. Our results suggest the development of an MPD-like myeloid leukemia after 44.5 months in all mice derived from both founders, which progressed to a myeloid leukemia
160
# CFU-GM
140
:———————————————————————
120 100 80 60 40 20 0 WT
TM
No treatment G-CSF GM-CSF
d 180 160
# CFU-GM
140 120 100 80 60 40 20 0 WT No treatment +GM-CSF +ATRA +ATRA +GM-CSF
Oncogene
TM
Figure 4 Hematopoietic progenitor assays. BM cells were suspended at a concentration of 10 000 cells/ml in methylcellulose medium (StemCell Technologies, Vancouver, BC, Canada), containing erythropoietin, stem cell factor, and interleukin-3, and grown at 371C in a 5% CO2-humidified chamber. Erythroid and granulocyte–monocyte colonies were scored after 7 days; more primitive colonies (CFU-GEMM) were scored after 12 days. All data presented in this figure are representative of experiments conducted on 40 pairs of age-matched WT and TM. (a) A comparison of the total number of colonies (y-axis) produced by representative WT and TM mice, showing a decrease in the number of colonies, and an increase in the number of abortive colonies, produced by transgenic CFU assays. (b) Breakdown of the numbers of BFU-E, CFU-GM, and CFU-GEMM (y-axis) in cultures from representative WT and TM. While the numbers of BFU-E and CFU-GEMM declined, the proportion of CFU-GM increased (from 70–75 to 490%), suggesting that the BM of TMs consisted predominantly of cells undergoing granulocyte/monocyte commitment. (c) CFU-GM (indicated on the y-axis) detected in cultures from representative WT and TM in response to treatment with 5 ng/ml G-CSF or 5 ng/ml GM-CSF. Cultures were supplemented with either G-CSF or GM-CSF prior to plating, and CFUGM scored on Day 7. Transgenic CFU did not proliferate in response to either G-CSF or GM-CSF, consistent with a block in the normal neutrophil differentiation pathway. (d) CFU-GM (indicated on the y-axis) detected in cultures from representative WT and TM in response to treatment with 5 ng/ml GM-CSF or 106 M ATRA or GM-CSF þ ATRA in combination. Transgenic CFU do not proliferate in response to GM-CSF or ATRA, but cotreatment is sufficient to revert the phenotype to WT. Similar results were obtained for G-CSF
NuMA-RARa transgenic mice develop APL MA Sukhai et al
673
a
c
5000
Percentage of cells
Relative long term culture expression of NuMA-RARα
6000
4000 3000 2000 1000
50 40 30 20 10 0
0 WT WT TM TM TM TM TM TM
PRO
NEU
WT TM TM + ATRA
b i
WT
iii
ii
TM
TM + ATRA
Figure 5 (a) Relative expression of NuMA-RARa in representative WT and TM LTMC, demonstrating the expression of NuMA-RARa. (b) Long-term cultures from WT and TM were prepared and treated with ATRA as outlined in Materials and methods. The arrows identify promyelocytes in nonadherent culture fractions. (i) May-Gru¨nwald- and Giemsa-stained cytospin preparations of nonadherent cells from a WT mouse long-term culture. (ii) TM culture. (iii) TM culture plus 106 M ATRA. (c) Quantitation of the number of blasts/promyelocytes and mature neutrophils (mean7s.d., n ¼ 3) in WT and TM LTMCs and transgenic LTMCs þ ATRA. (d) Flow-cytometry profiles (y-axis: cell counts; x-axis: mean fluorescence intensity) of WT vs TM vs transgenic þ ATRA (TM þ RA) LTMCs, showing the expression of Gr-1, CD11b, CD13, CD45, and CD117. Nonadherent fractions from LTMCs were harvested, and flow cytometry performed as outlined in Materials and methods. Transgenic LTMCs demonstrated decreased percentage of Gr-1 þ cells, as well as a decrease in the mean fluorescence intensity of cells expressing Gr-1. Transgenic LTMCs also exhibited increased percentages of cells expressing CD117 and CD13, relative to WT cultures. However, on treatment with ATRA, transgenic LTMC reverted to an immunophenotype indistinguishable from WT
with maturation in mice older than 10 months. These findings resemble most closely the results obtained with the hCG-PLZF-RARa single TMs, which developed an MPD-like myeloid leukemia after 46 months (He et al., 1998), but are in contradistinction to the phenotypes of the hCG-PML-RARa mice (Grisolano et al., 1997; He et al., 1997). In contrast to the conclusions drawn from those models, where it has been suggested that X-RARa is necessary, but not sufficient, for leukemogenesis, we
suggest that NuMA-RARa is sufficient for disease development in our mice. This difference may be a result of differences in analysis of the transgenic models, rather than a result of fundamental differences (i.e., the fusion partners) between the models. It is worthwhile to note that epidemiological analysis suggests that the rate of incidence of APL is constant in the population, regardless of the age of onset of the disease, suggesting that the presence of the X-RARa fusion gene is the only Oncogene
NuMA-RARa transgenic mice develop APL MA Sukhai et al
674 WT
0
10
102 FL1-H
104
10 0
0 100
102 FL1-H
104
89.5%
20
10
0 0
10
TM+ATRA
84.9%
20 Counts
10
WT+ATRA
66.5%
20 Counts
20 Counts
TM
83.5%
Counts
d Gr-1
100
102
104
100
102
104
FL1-H
FL1-H
Mac-1
10
0
0
20
40.0%
10
104
100
102
Counts 102
104
10
102
104
100
FL2-H
102
4.7%
104
4.9%
20
10
104
102 FL2-H
10
0 100
104
0 100
20
10
102 FL2-H
31.5%
FL2-H
0 100
100
20
10
104
61.3%
20
0
104
37.2%
FL2-H
10
102 FL2-H
0
104
4.1%
20
10
Counts
CD13
102 FL2-H
63.2%
0 100
20
90.0%
0 100
Counts
102 FL2-H
10
0
10 0
100
Counts
Counts
20
104
Counts
CD117
102 FL2-H
Counts
100
20
57.4% Counts
10
20
53.0% Counts
Counts
Counts
20
Counts
47.3%
20
0 100
FL2-H
102
104
100
FL2-H
102 FL2-H
104
CD45
0 10
102
104
100
102 FL3-H
104
10 0
100
102 FL3-H
104
100
102
104
FL3-H
Continued
rate-limiting event in APL pathogenesis (Vickers et al., 2000). While not eliminating the need for secondary genetic events, these findings suggest that any secondary lesions that occur do not play a critical role in defining the disease. It has been suggested that such a fundamental difference in APL pathogenesis between the mouse and humans may be attributable to differences between these organisms, but further work needs to be done in this area. A subset (B30%) of APL patients possesses additional, secondary chromosomal aberrations in their leukemic cells. Interestingly, however, the presence of these genetic changes does not seem to correlate with a worse prognosis (de Botton et al., 2000). Furthermore, we do not observe variability in the rate of progression of the phenotype, which would be expected with the acquisition and/or accumulation of additional genetic changes. We hypothesize that accumulation of blasts in the BM of TMs occurs over time and is the root cause of leukemia in our mice. Given sufficient time, all mice will Oncogene
10 0
0 0
FL3-H
Figure 5
10
20
20
Counts
10
Counts
20 Counts
Counts
20
develop leukemia. Thus, all TMs may possess an inherent phenotype that progresses to disease, which we have identified as the genetic myeloproliferation evident in our mice. The alternative hypothesis, in which mice accumulate random secondary genetic lesions with age, is less attractive, as we observed a rapid progression to leukemia in all our mice. In the scenario posited by this alternative hypothesis, we would expect the onset of leukemia to be later in the lifespan of the mice, and at low frequency. Several ‘secondary hit’ APL transgenic models have previously been developed. Double TMs carrying both hMRP8-PML-RARa and MMTV-BCL2 developed a myeloid leukemia without maturation very similar to the single transgenic hMRP8-PML-RARa model, but with shorter latency and higher penetrance (Kogan et al., 2001; Le Beau et al., 2002). Similarly, mice expressing hMRP8-PML-RARa and overexpressing interleukin-3, the IL-3 receptor or the GM-CSF receptor also developed leukemia with shorter latency and higher
NuMA-RARa transgenic mice develop APL MA Sukhai et al
675
penetrance (Le Beau et al., 2002). Likewise, a transgenic/transplant model examining the effects of FLT3ITD on the hCG-PML-RARa model yielded similar results (Kelly et al., 2002). In all these models, the presence of PML-RARa is necessary for the observed block of differentiation. The additional genetic change altered the accumulative potential of the resulting disease phenotype, and accelerated its progression. This was also evident in models expressing both PML-RARa and the reciprocal fusion gene RARa-PML (Pollock et al., 1999). Furthermore, hCG-PML-RARa mice that developed leukemia also displayed additional acquired nonrandom chromosomal abnormalities (Zimonjic et al., 2000), although it is not yet clear if this observation is a consequence of disease, or a causative event. Similar findings have been reported using hMRP8-PML-RARa mice (Le Beau et al., 2002), though the two models do not give rise to similar additional chromosomal abnormalities. It may be that PML-RARa, and, more generally, X-RARa, may cause genetic instability (supported by recent findings that indicate that hCG-PMLRARa mice are more susceptible to chromosomal aberrations in response to radiation (Lane et al., 2002)), but the resulting chromosomal abnormalities are nonspecific in nature. Given this body of findings, it is likely that secondary genetic events, while not necessary for initiation of disease, may accelerate its progression. In this model, any secondary genetic event resulting in increased myelo- or granulopoiesis, or enhanced myeloid cell survival, would serve to accelerate disease progression, thus decreasing the time to onset or increasing the severity of disease, without altering the underlying phenotype. Conclusions We have developed a TM model of APL, expressing NuMA-RARa. Our analysis is the first to apply community standards for diagnosis of murine leukemia (Kogan et al., 2002). We report that our model bears many similarities to the characteristics of the disease, as well as to the patient in whom NuMA-RARa was first identified. Leukemic cells in these mice resemble APL blasts morphologically and immunophenotypically, exhibit impaired neutrophil differentiation, and respond to ATRA treatment in vitro. Our results support the hypothesis that, while disrupted retinoid signaling is an underlying cause of APL, the N-terminal half of X-RARa has a nontrivial role in the development and characteristics of the disease. We have thus established a physiological system of leukemia caused by NuMARARa, which we will use in addressing questions significant to APL biology.
thus, hCG-NuMA-RARa TMs begin to express NuMA-RARa almost exclusively in promyelocytes during hematopoiesis. Two founder lines were generated from these experiments. Founders were backcrossed with C57BL6 mice in order to generate F1 progeny. Analyses were performed on F1 and subsequent generations of progeny descended from both founder mice, and maintained at the Animal Resource Centre of the Ontario Cancer Institute. Animal care, end point of study, leukemia diagnosis criteria, and controls All animals were treated in accordance with the Canadian Council on Animal Care guidelines. WT littermate mice, matched by age and sex, were used as controls for the TM population. Leukemia was diagnosed based on recommendations set out by Kogan et al. (2002). Mice displaying 420% neutrophils in their PB were said to have a MPD. The end point of the study was defined to be the persistent evidence of disease (longer than 2 months) or obvious signs of illness in the mouse, such that loss of appetite, weight, and lethargy interfered with its quality of life. Mouse genotyping DNA for genotyping was isolated from tail snips of mice according to standard techniques, and genotyping was performed by PCR and Southern Blot as follows. Genomic DNA was subjected to PCR with NuMA-RARa breakpointspecific primers (NR-F 50 -TCT AGC TCG CCT GGG TTC TC-30 ; NR-R 50 -CCC CAT AGT GGT AGC CTG AGG-30 ) using AmpliTaq Gold PCR enzyme and reagents (PE Applied Biosystems), according to the manufacturer’s protocols. Amplification was carried out for 35 cycles at 951C, 40 s, 551C, 40 s, and 721C for 1 min, followed by a final extension step at 721C for 7 min. PCR product (251 bp) was run on a 1.0% agarose gel, stained with ethidium bromide, and visualized under ultraviolet transilluminescence. For Southern genotyping, 5 mg mouse genomic DNA was digested with EcoR1. After digest, the samples were run on a 0.7% agarose gel, transferred to a positively charged nylon membrane (Amersham Pharmacia) by capillary transfer, and probed by a 32P-labeled human cathepsin G genomic probe, as previously described (He et al., 1997). Study design A total of 155 mice, comprising three generations of animals descended from both founder lines, were analysed in this study. A subset of 75 mice (30 wild type and 45 transgenic) was phenotyped on a monthly basis by PB examination as outlined below. This allowed the generation of a consistent, detailed, and reproducible PB profile over the lifespan of the mice. The remaining animals were killed at regular intervals for detailed BM and organ phenotype analysis as outlined below. In this second arm of the study, we were able to develop a representation of the BM and organ phenotypes of mice at 2, 4, 6, 8, 10, 12, 14, and 21 months of age. Finally, by killing of mice from the first cohort, we were able to correlate the PB profile of mice with the organ and BM phenotypes at 1, 11, and 21 months of age.
Materials and methods PB analysis Construction of TMs hCG-NuMA-RARa TMs were constructed as previously described (He et al., 1997). The hCG promoter is activated specifically in the myeloid compartment (He et al., 1999), and,
Mice were tail bled on a monthly basis. Blood samples were diluted in phosphate-buffered saline, and aliquoted for automated complete blood count (Laboratory Hematology Department, Princess Margaret Hospital, Toronto, ON, Oncogene
NuMA-RARa transgenic mice develop APL MA Sukhai et al
676 Canada), flow cytometry (see below), and preparation of PB films according to standard hematologic techniques. Blood films were stained with May–Gru¨nwald and Giemsa stains, according to standard techniques. Films were scored manually by a count of 4100 nucleated white cells for the presence of lymphocytes (Ly) and myeloid cells (My). Myeloid cells included neutrophils, monocytes, and blasts/promyelocytes. From these observations, the percentage of myeloid cells (%My) and My/Ly ratio were calculated.
Long-term marrow cultures Further in vitro analysis of transgenic cells was carried out in long-term BM cultures. Cells for long-term culture were plated in MyeloCult M5300 growth medium (StemCell Technologies) supplemented with 106 M hydrocortisone (StemCell; Okubo et al., 2000). Cultures were allowed to develop for 28 days prior to treatment with 1 mM ATRA (120 h). After treatment, the culture supernatants were harvested for RNA, flowcytometry, and morphology analyses.
Flow cytometry
NuMA-RARa expression analysis
Samples were treated to lysed red cells using a commercially available reagent (Sigma Aldrich Canada, ON, Canada) prior to processing. All samples collected for flow cytometry were washed and resuspended in PBS. Samples were blocked with purified CD16/CD32 receptor antibodies (BD Pharmingen), and incubated with antibodies (BD Pharmingen) to cellsurface antigens CD45 (leukocyte common antigen), CD117 (marker for primitive progenitors), Gr-1 (marker for neutrophils), and CD11b (myeloid cell-surface antigen). The immature myeloid marker CD13 was also used for LTMC samples. Incubations were carried out at 41C in a lightprotected environment. After incubation, samples were washed and resuspended in PBS and analysed on a Becton Dickinson FACScalibur flow cytometer. Data acquisition and analysis were accomplished using CellQuest software.
RNA was isolated using TRIZol Reagent (Invitrogen/Life Technologies), and subjected to RT–PCR: 1 mg total RNA was incubated with MuLV RT enzyme (Invitrogen) at 371C in the presence of first-strand cDNA synthesis buffer (Invitrogen), 0.01 M DTT, RNAGuard RNAse inhibitor (Amersham Pharmacia), random hexamer primers, and 0.1 mM dNTPs (20 ml reaction volume, RT step). PCR was carried out on a 5 ml RT product as follows: NuMA-RARa-specific primers spanning the fusion breakpoint were used in 50 ml reaction volumes (primer N2B, sequence 50 -AGA CCT GGG CAA ATT CCA GG-30 ; primer NR1C, sequence 50 -CTT CTC AAT GAG CTC CCC C-30 ) for 35 cycles (951C, 40 s, 621C, 40 s, 721C, 1 min; 721C for 10 min – final extension step). Products were run on 1% agarose gels, stained with ethidium bromide and visualized under UV transilluminescence (product size 1000 bp).
Animal sacrifice, organ collection, and analysis TMs alongside the corresponding WT controls were euthanized in a CO2 chamber and killed for analysis, according to the Canadian Council on Animal Care guidelines. Hind limb bones were collected for BM analysis. The liver, lung, spleen, heart, kidney, intestine, muscle, thymus, and one tibia were collected for histopathology. Samples for histopathology were fixed in 10% buffered formalin (Sigma), paraffinembedded, and sections mounted on slides and stained with H&E, according to standard techniques (Clinical Research Pathology Group, University Health Network, Toronto, ON, Canada).
Quantitative real-time RT–PCR Quantitative real-time polymerase chain reaction (PCR) was carried out for the assessment of relative gene expression using the ABI Prism 7700 Sequence Detection System (PE Applied Biosystems). SYBR Green I dye, a double-strand DNAbinding dye that allows the quantitative detection of products during the PCR was used. Data were quantified and analysed using the Sequence Detection System software (version 1.7) (PE Applied Biosystems). This method was performed as previously described (Reis et al., 2002). Analysis of quantitative real-time PCR data: DDCt method
BM analysis The BM was flushed from hind limb bones and a single cell suspension generated. The samples were aliquoted for cytospin preparation and subsequent staining with May–Gru¨nwald and Giemsa, and for flow-cytometry analysis. A total of 300–500 nucleated cells from BM preparations were scored as neutrophils, myelocytes, promyelocytes, myeloblasts, erythroblasts, or as cells of the megakaryocytic and monocytic lineages. From these data, a collective neutrophilic lineage percentage (% granulopoiesis) was calculated, as was the percentage of mature neutrophils and promyelocytes.
The relative quantification was given by the Ct values, determined for duplicate reactions for each target and internal control gene (GAPDH). Duplicate Ct values were averaged and subtracted to obtain DCt (DCt ¼ Ct (target gene)Ct (GAPDH)). The relative expression level was determined as 2DDCt, where DDCt ¼ DCt (target sample)DCt (reference sample). For the reference sample, DDCt equals zero, and 20 equals one; so the fold change in the reference sample equals one, by definition. For the other samples, evaluation of 2DDCt indicates the fold change in gene expression relative to the reference sample (Livak and Schmittgen, 2001). PCR amplification
Hematopoietic progenitor assays BM cells were suspended at a concentration of 10 000 cells/ml in methylcellulose medium (StemCell Technologies, Vancouver, BC, Canada), containing erythropoietin, stem cell factor, and interleukin-3, and supplemented with or without 5 ng/ml G-CSF, 5 ng/ml GM-CSF, and 106 M ATRA, and grown at 371C in a 5% CO2 humidified chamber. Erythroid (BFU-E) and granulocyte–monocyte (CFU-GM) colonies were scored after 7 days; more primitive colonies (CFU-GEMM) were scored after 12 days. Oncogene
Reaction mixtures contained cDNA reverse transcribed from 2 mg of the total RNA from each sample, 10 mM of each primer, and 12.5 ml of 2 SYBR Green PCR Master Mix (PE Applied Biosystems), which includes the SYBR Green I fluorescent dye, 0.5 U of AmpErase uracyl-N-glycosylase (UNG) enzyme, 1.25 U of Ampli-Taq Gold DNA polymerase, and 200 mM dNTPs. The thermal cycling conditions were 501C for 2 min (for UNG enzyme activity), 951C for 10 min, and 40 cycles at 951C for 15 s, followed by 601C for 1 min. Experiments were performed in duplicate for each sample in the same reaction
NuMA-RARa transgenic mice develop APL MA Sukhai et al
677 plate and repeated when a coefficient of variation higher than 5% was observed. Data analysis and statistics All results were analysed and graphs generated using Microsoft Excel 2000 and SigmaPlot 6.0/2000. Two-population comparisons, where appropriate, were performed using the t-test; multipopulation comparisons were made using the F-test. Statistical significance was taken as Po0.05.
Abbreviations APL, acute promyelocytic leukemia; ATRA, all trans-retinoic acid; BM, bone marrow; CFU, colony-forming unit; LTMC,
long-term marrow culture; My/Ly, myeloid/lymphoid; MPD, myeloproliferative disorder; NPM, nucleophosmin; NuMA, nuclear mitotic apparatus; PB, peripheral blood; PML, promyelocytic leukemia; PLZF, promyelocytic leukemia zinc-finger; RARa, retinoic acid receptor alpha. Acknowledgements We thank Dr S Arora for reading of the manuscript. This work was supported by funding from the National Cancer Institute of Canada (Grant #13087 to SK-R) and the National Institutes of Health (Grant #s R01 CA74031, R01 CA71692, and U01 CA84292 to PPP). MAS is an Ontario Graduate Scholar. We also gratefully acknowledge the disability accommodations funding provided through the University of Toronto, Office of Accessibility Services.
References Arnould C, Philippe C, Bourdon V, Gr goire MJ, Berger R and Jonveaux P. (1999). Hum. Mol. Genet., 8, 1741–1749. Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton DA, Gralnick HR and Sultan C. (1976). Br. J. Haematol., 33, 451–458. Brown D, Kogan S, Lagasse E, Weissman I, Alcalay M, Pelicci PG, Atwater S and Bishop JM. (1997). Proc. Natl. Acad. Sci. USA, 94, 2551–2556. Chen G-X, Xhu X-H, Men X-O, Wang L, Huang Q-H, Jin XL, Xiang S-M, Zhu L, Guo W-H, Chen J-Q, Xu S-F, So E, Chan L-C, Waxman S, Zelent A, Chen G-Q, Dong S, Liu J-X and Chen S-J. (1999). Proc. Natl. Acad. Sci. USA, 96, 6318–6323. Chen Z, Brand NJ, Chen A, Chen S, Tong J, Wang Z, Waxman S and Zelent A. (1993). EMBO J., 12, 1161–1167. Compton DA and Cleveland DW. (1993). J. Cell Biol., 120, 947–957. Compton DA, Szilak I and Cleveland DW. (1992). J. Cell Biol., 116, 1395–1408. David G, Terris B, Marchio A, Lavau C and Dejean A. (1997). Oncogene, 14, 1547–1554. de Botton S, Chevret S, Sanz M, Donbert M, Thomas X, Guerci A, Fey M, Ragon C, Huguet F, Sotto J-J, Gardin C, Makhoul PC, Travade P, Solary E, Fegeux N, Bordessoule D, San Miguel J, Link H, Desablens B, Stamatoullas A, Deconnick E, Geiser K, Hess U, Maloisel F, Castaigne S, Preudhomme C, Chomienne C, Degos L and Fenaux P. (2000). Br. J. Hematol., 111, 801–806. de The´ H, Chomienne C, Lanotte M, Degos L and Dejean A. (1990). Nature, 347, 558. de The´ H, Lavau C, Marchio A, Chomienne C, Degos L and Dejean A. (1991). Cell, 66, 675. Dyck JA, Maul GG, Miller Jr WH, Chen JD, Kakizuka A and Evans RM. (1994). Cell, 76, 333–343. Early E, Moore MAS, Kakizuma K, Nason-Burchenal K, Martin P, Evans RM and Dmitrovsky E. (1996). Proc. Natl. Acad. Sci. USA, 93, 7900–7904. Grignani F, Fagioli M, Alcalay M, Longo L, Pandolfi PP, Donti E, Biondi A, Lo Coco F and Pelicci PG. (1994). Blood, 83, 10–25. Grisolano JL, Wesselschmidt RL, Pelicci PG and Ley TJ. (1997). Blood, 89, 376–387. Harmon DC. (1991). The leukemias. Beck WS (ed) MIT Press: Cambridge, pp. 399–417. He LZ, Guidez F, Tribioli C, Peruzzi D, Ruthardt M, Zelent A and Pandolfi PP. (1998). Nat. Genet., 18, 126–135. He L-Z, Merghoub T and Pandolfi PP. (1999). Oncogene, 18, 5278–5292.
He LZ, Tribioli C, Rivi R, Peruzzi D, Pelicci PG, Soares V, Cattoretti G and Pandolfi PP. (1997). Proc. Natl. Acad. Sci. USA, 94, 5302–5307. Hummel JL, Wells RA, Dube´ ID, Licht JD and Kamel-Reid S. (1999). Oncogene, 18, 633–641. Kelly LM, Kutok JL, Williams IR, Boulton CL, Amaral SM, Curley DP, Ley TJ and Gilliland DG. (2002). Proc. Natl. Acad. Sci. USA, 99, 8283–8288. Kogan SC, Brown DE, Shultz DB, Truong BT, LallemandBreitenbach V, Guillemin MC, Lagasse E, Weissman IL and Bishop JM. (2001). J. Exp. Med., 193, 531–543. Kogan SC, Ward JM, Anver MR, Berman JJ, Brayton C, Cardiff RD, Carter JS, de Coronado S, Downing JR, Fredrickson TN, Haines DC, Harris AW, Harris NL, Hiai H, Jaffe ES, MacLennan ICM, Pandolfi PP, Pattengale PK, Perkins AS, Simpson RM, Tuttle MS, Wong JF and Morse III HC. (2002). Blood, 100, 238–245. Labrecque J, Allan D, Chambon P, Iscove NN, Lohnes D and Hoang T. (1998). Blood, 92, 607–615. Lane AA, Westervelt P and Ley TJ. (2002). Blood, 100, 538a. Le Beau MM, Bitts S, Davis EM and Kogan SC. (2002). Blood, 99, 2985–2991. Livak KJ and Schmittgen TD. (2001). Methods, 25, 402–408. Mark M, Ghyselinck NB, Wendling O, Dupe´ V, Mascrez B, Kastner P and Chambon C. (1999). Proc. Nutr. Soc., 58, 609–613. Melnick A and Licht JD. (1999). Blood, 93, 3167–3215. Mu ZM, Chin KY, Liu JM, Lozano G and Chang KS. (1994). Mol. Cell. Biol., 14, 6858–6867. Okubo T, Matsui N, Yanai N and Obinata M. (2000). Cell Struct. Funct., 25, 133–139. Paietta E, Andersen J, Gallagher R, Bennett J, Yunis J, Cassileth P, Rowe J and Wiernik PH. (1994). Leukemia, 8, 1108–1112. Pollock JL, Westervelt P, Kurichety AK, Pelicci PG, Grisolano JL and Ley TJ. (1999). Proc. Natl. Acad. Sci. USA, 96, 15103–15108. Redner RL, Rush EA, Faas S, Rudert WA and Corey SJ. (1996). Blood, 87, 882–886. Rego EM and Pandolfi PP. (2001). Semin. Hematol., 38, 54–70. Reis PP, Rogatto SR, Kowalski LP, Nishimoto IN, Montovani JC, Corpus G, Squire JA and Kamel-Reid S. (2002). Oncogene, 21, 6480–6487. Rizzatto EG, Garcia AB, Portieres FL, Silva DE, Martins SLR and Falcao FP. (2002). Am. J. Clin. Pathol., 118, 31–37. Rowley JD, Golomb HM and Dougherty C. (1977a). Lancet, 1, 549–550. Oncogene
NuMA-RARa transgenic mice develop APL MA Sukhai et al
678 Rowley JD, Golomb HM, Vardiman J, Fukuhara S, Dougherty C and Potter D. (1977b). Int. J. Cancer, 20, 869–872. Shaknovich R, Yeyati PL, Ivins S, Melnick A, Lempert C, Waxman S, Zelent A and Licht JD. (1998). Mol. Cell. Biol., 18, 5533–5545. Vickers M, Jackson G and Taylor P. (2000). Leukemia, 14, 722–726. Wang ZG, Ruggero D, Ronchetti S, Zhong S, Gaboli S, Rivi R and Pandolfi PP. (1998). Nat. Genet., 20, 266–272. Weir EG and Borowitz MJ. (2001). Semin. Hematol., 38, 124–138.
Oncogene
Wells RA, Catzavelos C and Kamel-Reid S. (1997). Nat. Genet., 17, 109–113. Wells RA, Hummel JL, DeKoven A, Zipursky A, Kirby M, Dube´ I and Kamel-Reid S. (1996). Leukemia, 10, 735–740. Westerveldt P and Ley TJ. (1998). Blood, 92 (Suppl 1), 211a (abstr). Zhong S, Salomoni P, Ronchetti S, Guo A, Ruggero D and Pandolfi PP. (2000). J. Exp. Med., 191, 631–640. Zimonjic DB, Pollock JL, Westervelt P, Popescu NC and Ley TJ. (2000). Proc. Natl. Acad. Sci. USA, 97, 13306–13311.
