The Wilms’ tumor suppressor gene, WT1, encodes a transcription factor of the Cys2-His2 zinc finger type. Loss of WT1 gene function has been implicated in the development of malignancies including Wilms’ tumor and acute leukemias. We have shown previously that ectopic expression of WT1 +KTS isoforms in murine M1 leukemic cells spontaneously induces monocytic differentiation without the requirement for external differentiation-inducing stimuli. To determine whether these observed effects in vitro corresponded to a reduction in tumorigenicity in vivo, parental M1, control M1.Neo, and M1.WT1 +KTS cells were transplanted into C.B-17 scid/scid mice, and the growth and metastatic behavior of the cell lines were monitored for a period of 20 weeks. Mice inoculated either s.c. on the flank or directly into the peritoneal cavity, with M1 cells stably expressing WT1 +KTS isoforms exhibited a marked decrease in tumor formation compared with control groups. Moreover, tumors arising in mice after the injection of M1.WT1 +KTS cells exhibited a loss in ectopic WT1 protein expression. Confirmation that the tumors arose from M1.WT1 +KTS cells was achieved by the amplification of the introduced transgene from tumor samples and indicates that the tumorigenicity of leukemic M1 cells in these animals correlates with a loss in WT1 expression. This investigation is the first to demonstrate the tumor-suppressive effects of WT1 expression in a leukemic cell line, further advancing the notion that WT1 acts as a differentiation-promoting gene during hematopoiesis and that loss of functional WT1 expression may contribute to leukemogenesis in vivo.

Implicated in the onset of WT,3a pediatric nephroblastoma, the WT1 gene is a critical regulator of cellular proliferation and differentiation within the genitourinary system (reviewed in Ref. 1). The WT1 gene encodes a Cys2-His2 ZF transcription factor, and two alternative splice sites within the gene produce the four major WT1 polypeptides: ±17 amino acids encoded by the presence or absence of exon 5 and inserted NH2-terminal of the four contiguous ZF DNA-binding domains; and ±3 amino acids (KTS)encoded by the presence or absence of 9 bp at the end of exon 9 and inserted between the third and fourth ZF DNA-binding domains (2). In the developing kidney, WT1 is expressed in the early metanephric stem cells as they condense around the ureteric bud and undergo differentiation into renal epithelial structures, with expression eventually restricted to the podocyte layer of mature glomeruli (3). WT1 transcripts can also be found in the primitive genital ridge tissue that gives rise to mature gonadal structures (3). Accordingly, homozygous inactivation of WT1 in mice results in the failure of both kidney and gonadal formation (4). The necessity for functional WT1 expression during normal genitourinary development is best reflected at a clinical level, with gross deletions and intragenic mutations detectable at the WT1 11p13 locus in at least 5% of WT cases and >95% of patients with Denys-Drash syndrome (WT in association with genital abnormalities and nephropathy;reviewed in Ref. 5). The expression of WT1 in both normal and neoplastic hematopoietic cells (6, 7) has led to speculation that this gene may play a similar role in regulating cellular proliferation and differentiation during blood cell development. Mutations within the WT1 gene have been detected in 12% of acute leukemia patient samples primarily of biphenotypic (lymphoid/myeloid) or early myeloid origin (8, 9). Furthermore, expression and/or mutation of WT1 in acute leukemias may indicate the presence of minimal residual disease and predict a poor prognosis for these patients (8, 9, 10).

The observations from the clinical situation, combined with its function as a transcriptional repressor, have resulted in the classification of WT1 as a tumor suppressor gene, whereby loss of functional WT1 gene expression is sufficient for initiation of tumorigenesis. The paradigm of the tumor-suppressive effect of WT1 within genitourinary development has been well defined. Indeed, ectopic expression of normal WT1transcripts in various malignant kidney cell lines has primarily resulted in a marked suppression of the tumorigenic phenotype of the cells upon s.c. transplant into animal recipients (11, 12, 13). To date, however, no studies have been conducted to ascertain whether WT1 has the potential to perform a similar tumor-suppressive function in leukemic cell lines. The murine myeloblastic leukemia cell line M1 forms myelosarcomas upon syngeneic transplantation into SL mice (14, 15). Induction of cellular differentiation in M1 cells by exposure to differentiation inducers leads to a marked suppression of leukemogenicity (16, 17). Using these cells as an in vitro model for macrophage differentiation, we have shown previously that parental M1 cells, which lack endogenous expression of WT1 in an undifferentiated state, express high levels of WT1 after induction of macrophage differentiation by LIF (18). Furthermore, it was demonstrated that the ectopic expression of WT1+KTS isoforms in M1 cells induced monocytic differentiation without the requirement for external differentiation-inducing stimuli (18). In this investigation, we now demonstrate that M1.WT1 +KTS cells exhibit a marked reduction in tumorigenicity compared with parental and control M1 cells when allogeneically transplanted into immunodeficient SCID mice. This study is the first to demonstrate that WT1 can abrogate the tumorigenic potential of a leukemic cell line and therefore suggests that this gene may act, as it does in the developing kidney, as a classical tumor suppressor gene during blood cell development, with the functional loss of WT1 contributing to the onset of leukemogenesis in vivo.

Cells and Cell Culture.

The myeloblastic leukemia cell line M1 was established from a spontaneous myeloid leukemia that arose in the SL strain of mice (14). The generation, maintenance, and in vitrocharacterization of M1 cells stably transfected with expression vectors containing WT1 +KTS cDNAs or a control neomycin-resistance cDNA have been described in detail elsewhere (18).

Clonogenicity Assay.

Clonogenic potential was assessed by colony formation in soft agar. After washing, 200 cells in 1 ml of Iscove’s modified Dulbecco’s medium (Life Technologies, Inc., Grand Island, NY) supplemented with 10% FCS (HyClone Laboratories, Inc., Logan, UT) and 0.3% Bacto-Agar(Difco Laboratories, Detroit, MI) were plated in duplicate into 35-mm Petri dishes. Recombinant murine LIF (a gift from AMRAD Biotech,Melbourne, Australia) was added to experimental cultures at a concentration of 1 ng/ml. Plates were kept at 37°C in a humidified atmosphere with 5% CO2, and after 7 days,colonies were scored as clones containing >50 cells, whereas abortive clones containing <50 cells were scored as clusters.

Tumorigenicity Assay.

Animal studies were performed with appropriate institutional ethics committee approval and in accordance with National Health and Medical Research Council guidelines. Specific pathogen-free SCID mice were obtained from the Animal Resources Center (Perth, Australia) and housed in air-filtered cages under sterile conditions. M1, M1.Neo,M1.WT1.1(+/+), and M1.WT1.2(−/+) cell lines for in vivotransplantation were harvested, washed, and resuspended in PBS. For s.c. injections, adult female SCID mice were inoculated with 1 × 106 cells in 100 μl of PBS on the right dorsal flank. For i.p. injections, age-matched 6-week-old female SCID mice were inoculated with 1 × 106 cells in 100 μl of PBS. Control mice not receiving injections received PBS alone. For each site tested, six mice/group received injections, and the experiments were performed twice at separate intervals. The results were then combined to give a total number of 12 mice/group for each site. Tumor formation and tumor-related mortality were monitored twice weekly for a period of 20 weeks, after which time any mice showing an absence of tumor incidence were considered to be tumor free.

Histology.

For histological examination, mice were autopsied upon humane sacrifice, and any tumorous material was removed. Tissues were fixed in PBS-buffered 10% formalin and embedded in paraffin, and 5-μm sections were stained with Mayer’s H&E using standard techniques.

Protein Extraction and Western Blotting.

Analysis of WT1 expression in tumors derived from injection of M1.WT1+KTS cells was facilitated by the extraction of protein from paraffin-embedded tumor tissue (19). Briefly, 50μ m sections were deparaffinized using standard techniques, followed by homogenization in radioimmunoprecipitation assay buffer containing 2% SDS [2% SDS, 1% sodium deoxycholate, 1%Triton X-100 (Sigma Chemical Co., St. Louis, MO), 150 mmNaCl, 1 mm EDTA, and 10 mm Tris-HCl, pH 7.4]and the Complete Protease Inhibitor Cocktail (Boehringer Mannheim,Mannheim, Germany). Samples were incubated at 100°C for 20 min,followed by 2 h at 60°C. Protein extraction from cultured cell lines and tumor ascites was performed as described previously (18). After centrifugation, total protein (25 μg) was then separated on an SDS-10% polyacrylamide gel and subjected to immunoblotting with the WT (C-19) antibody (Santa Cruz Biotechnology,Inc., Santa Cruz, CA), also as described previously (18). Protein recovery was monitored with a cross-specific rabbit anti-actin antibody (Sigma Chemical Co.) using a similar immunoblotting procedure.

Genomic DNA Extraction and PCR.

The detection of the WT1 transgene in tumors derived from injection of M1.WT1 +KTS cells was facilitated by the extraction of genomic DNA from paraffin-embedded tumor tissue (20). Briefly, 50-μm sections were deparaffinized using standard techniques, followed by incubation in TNE buffer containing 1% SDS(1% SDS, 5 mm EDTA, 150 mmNaCl, and 50 mm Tris-HCl, pH 8.0) and 500 μg/ml proteinase K (Merck, Darmstadt, Germany) for 48 h at 45°C. An additional 250 μg/ml of proteinase K was added to the samples on each successive day of incubation. DNA extraction from cultured cell lines and tumor ascites was performed as described previously (18). After organic extraction and ethanol precipitation,500 ng of genomic DNA was used in a PCR incorporating a 5′ primer directed to exon 7 of the WT1 transgene 5′-CCCAGGCTGCAATAAGAGATA-3′, and a 3′ primer directed to the retroviral expression vector, 5′-AGCCTGGACCACTGATATCC-3′. Conditions for WT1 amplification, in addition to amplification of the murine β-actin gene to monitor DNA integrity were also as described previously (18). PCR products (5 μl; 10 μl for β-actin PCR on genomic DNA from paraffin-embedded tumor tissue) were separated on 1% agarose gels in 1× TAE buffer (40 mm Tris-acetate, 2 mm EDTA,pH 8.5).

Statistical Analysis.

The significance of differences in the proportion of mice that developed tumors or survived injections between the control and experimental groups was determined using the nonparametric Mann-Whitney U test. The control groups of mice receiving injections,inoculated with either M1 or M1.Neo cells, were combined for the purposes of this analysis. P < 0.05 was considered significant.

Expression of the WT1 + KTS Isoforms Leads to a Reduction in Clonogenicity of M1 Cells in Soft Agar.

As a measure of clonogenic potential, the ability of M1 cells stably transfected with the +KTS isoforms of WT1 to form colonies in soft agar was compared with parental M1 cells and control M1.Neo cells. Colony formation was assessed in either the presence or absence of LIF. In the absence of LIF, M1 and M1.Neo cells exhibited overall plating efficiencies of 50 and 49%, respectively (Fig. 1,A). For M1 cells, compact, undifferentiated colonies represented 88% of cell growth in these cultures, and 95% of M1.Neo growth also exhibited this phenotype (Fig. 1). Despite the observation that M1.WT1.1(+/+) and M1.WT1.2(−/+) cells exhibited a reduction in plating efficiency compared with control cells (34 and 16%,respectively; Fig. 1,A), the proportion of colonies in these cultures, a mixture of compact and diffuse-type colonies (Fig. 1,B), was still significant at 76% for M1.WT1.1(+/+) cells and 63% for M1.WT1.2(−/+) cells. Although the decreased ability of M1.WT1 +KTS cells to grow in soft agar correlates with an increased differentiative phenotype in these cells (Fig. 1 B), it is important to note that sequential replating of differentiated colonies from these cultures resulted in the regeneration of secondary colonies that exhibited both nondifferentiated and differentiated morphologies(data not shown). Furthermore, we had observed previously that the growth of M1.WT1 +KTS cells in liquid culture did not significantly differ from control M1 cells. Unlike M1 cells induced for macrophage differentiation by LIF, these cells do not exhibit clonal exhaustion associated with terminal macrophage differentiation (18). These results would suggest that although ectopic WT1 +KTS isoform expression promotes the differentiation of M1 cells along the monocyte/macrophage pathway, it still enables the maintenance of viable, self-renewing stem cells within this population, thus facilitating their persistence in culture as a continuously proliferating cell line.

Because of the identification of an enduring blast population within M1.WT1 +KTS cells, combined with observations in other cell lines where enforced expression of WT1 rendered them less sensitive to monocyte/macrophage differentiation induction (21, 22), it was important to confirm the responsiveness of these cells to LIF. the addition of LIF to M1 and M1.Neo cells led to a slight decrease in total plating efficiency (46 and 35%, respectively) but a marked decrease in the proportion of colonies in these cultures (26% for M1;22% for M1.Neo) because of LIF-induced induction of terminal differentiation and clonal suppression in these cells (Fig. 1 A). The response of M1.WT1 +KTS cells to LIF exposure was dramatic, resulting in almost a complete abolition of growth in soft agar. Total plating efficiencies of 5 and 2% were observed for M1.WT1.1(+/+) and M1.WT1.2(−/+) cells, respectively. The proportion of colonies within these cultures was also greatly suppressed. For M1.WT1.1(+/+) cells, colonies represented only 22% of the reduced cell growth, whereas only abortive clones and clusters were formed by M1.WT1.2(−/+) cells. This is in agreement with our previous work,which had demonstrated that M1.WT1 +KTS cells were very sensitive to exposure to LIF in liquid culture compared with control M1 cells,resulting in a rapid suppression of clonogenicity because of the immediate induction of terminal macrophage differentiation coupled to apoptotic cell death in these cells (18). These results would suggest that ectopic WT1 +KTS isoform expression does not abrogate the ability of M1 cells to respond to LIF and, in fact, primes their responsiveness because of an increase in their basal level of differentiation when compared with control M1 cells.

Expression of WT1 +KTS Isoforms Leads to a Reduction in Tumorigenicity of M1 Cells upon s.c. Injection into C.B-17 scid/scid Mice.

To determine the effect of WT1 +KTS isoform expression on the in vivo growth and metastatic properties of M1 cells, the various M1 cell-derived populations were injected s.c. into SCID mice. Whereas 11 of 12 mice (92%) and 12 of 12 mice (100%) receiving injections with parental M1 and control M1.Neo cells, respectively, formed tumors, the incidence of tumor formation in mice injected with M1.WT1 +KTS cells was greatly reduced (Table 1). Tumors arose in only 3 of 12 mice (25%) in the group transplanted with M1.WT1.1(+/+) cells, a reduction in tumor formation of up to 75%over control groups. Moreover, in the cohort injected with M1.WT1.2(−/+) cells, tumor formation was completely abolished. No mice in the PBS control group were affected (Table 1).

Enumeration of the median latency period (Table 1) revealed a marked delay in the time elapsed before tumors began to present between the control groups, injected with parental M1 cells (21 days) or control M1.Neo cells (30 days), and the group injected with M1.WT1.1(+/+) cells(59 days). Although no mice receiving injections of M1.WT1.2(−/+)cells developed tumors throughout the duration of the monitoring period(20 weeks), 1 mouse of the 12 initially receiving injections of M1.WT1.2(−/+) cells died of an unrelated respiratory illness and was excluded from the analysis. The survival of SCID mice receiving injections at this site with either M1.WT1.1(+/+) or M1.WT1.2(−/+)cells was significantly enhanced when compared with mice inoculated with control cells, and the cumulative survival of the groups over time is displayed in Fig. 2 A.

As expected, nearly all of the myelosarcomas formed by M1 and M1.Neo cells presented as s.c. nodules on the dorsal flank, and a representative s.c. tumor derived from the injection of these cells is shown in Fig. 3,A. These tumors were well vascularized, consisting of a solid mass of uniform cells that possessed a high nuclear:cytoplasmic ratio,basophilic cytoplasm, and prominent nucleoli. This typical blast morphology is consistent with the phenotype of the injected M1 and M1.Neo cells (18). These cells proliferated within the s.c. layer, leading to the destruction of the underlying skeletal muscle structures and infiltration of the adipose tissue. Somewhat surprisingly, however, in one animal from the control group receiving injections s.c. of M1.Neo cells, the myelosarcoma arose as viscerally disseminated tumors of various organs within the peritoneal cavity, in addition to ascites production. These vascularized tumors occurred within the mesenteric membrane surrounding the upper intestinal tract,resulting in the involvement of neighboring gastrointestinal elements,such as the stomach and duodenum, through the focal neoplastic invasion of the intestinal serosa (Fig. 3,B). As for the control M1 cell populations, tumors elicited by the s.c. injection of M1.WT1.1(+/+) cells primarily presented as well-vascularized s.c. nodules (Fig. 3,C), which contained mixed cell types reminiscent of cultured M1.WT1.1(+/+) cells (18). Interestingly, one of the three animals to succumb to tumor formation upon receiving an injection at this site again formed disseminated peritoneal myelosarcomas involving the gut mesentery and associated organs, resulting in the neoplastic infiltration of the intestinal serosa (Fig. 3 D) and ascites formation.

Expression of WT1 +KTS Isoform Leads to a Reduction in Tumorigenicity of M1 Cells upon i.p. Injection into C.B-17 scid/scid Mice.

Historically, an i.p. route of injection has often been used to assess the leukemogenicity of M1 cell derivatives (15, 17). The observation that cells injected s.c. on the dorsal flank could give rise to myelosarcomas within the peritoneal cavity of affected animals may suggest that M1 cell populations prefer to grow i.p. To confirm that the decrease in tumorigenicity observed at the s.c. site of injection was not attributable to either an inability of M1.WT1 +KTS cells to thrive at this site or its lack of proximity to the peritoneal cavity, an additional i.p. route of inoculation was also examined. As for the s.c. injection, the effect of WT1 +KTS isoform expression in suppressing the tumorigenicity of M1 cells was striking. Whereas 12 of 12 mice (100%) and 10 of 12 mice (83%) receiving injections i.p. with M1 and M1.Neo cells, respectively, formed peritoneal myelosarcomas,only 2 of 12 mice (17%) in groups receiving injections of either M1.WT1.1(+/+) or M1.WT1.2(−/+) cells succumbed to tumor formation at this site (Table 1), representing a significant reduction in tumorigenesis of up to 83% over groups receiving injections of parental and control cells at this site. This substantiates the results observed at the s.c. dorsal site, where both WT1 +KTS isoforms also had the ability to uniformly suppress the tumorigenicity of M1 cells in vivo. No mice in the control PBS group were affected(Table 1).

However, evaluation of the median latency period for onset of tumorigenesis at this site did not reveal the marked differences observed at the s.c. site between the control and experimental groups(Table 1), with all four groups giving comparable mean times of tumor onset (49–60 days). The cumulative survival of SCID mice injected i.p. with the M1 cell populations is presented in Fig. 2 B and further emphasizes the overall increase in survival of mice inoculated with M1.WT1 +KTS cells.

The presentation of myelosarcomas within the peritoneal cavity after the direct i.p. injection of cells closely resembled that of myelosarcomas that had formed at this location after injection of cells at the s.c. site, with all cell populations preferentially lodging within the mesenteric membrane, often leading to the invasion of adjacent organs such as the kidney and duodenum (data not shown). However, additional peritoneal tumors were also observed within the pancreas and the adipose tissue surrounding the uterine horns.

Loss of Ectopic WT1 Expression in Tumors Arising from the Injection of M1.WT1 +KTS Cells.

Because tumors were still observed upon injection of the transfected population of M1.WT1 +KTS cells, it was of interest to determine whether these tumors were in fact derived from clones of cells that had altered levels of expression of the WT1 transgene compared with the original cell line injected. This was assessed by immunoblotting of protein and PCR amplification of genomic DNA isolated from the injected M1.WT1 +KTS cell lines, an ascitic tumor that arose in a mouse receiving an injection of M1.WT1.1(+/+) cells at the s.c. dorsal site, and solid tumors that arose in mice receiving injections of either M1.WT1.1(+/+) or M1.WT1.2(−/+) cells at the i.p. site. In all tumors examined, ectopic WT1 protein expression was either not detectable (in M1.WT1.1(+/+) tumors from both the s.c. and i.p. injections) or dramatically reduced [in an M1.WT1.2(−/+) tumor from the i.p. injection] in comparison with the level of ectopic WT1 expression (visualized as Mr54,000/52,000 protein products) in the original cell lines at the time of injection (Fig. 4,A). The level of endogenous actin expression (a Mr 42,000 protein product) is also shown to indicate the efficiency of protein recovery from each sample(Fig. 4,A). To confirm that these tumors did indeed arise from M1.WT1 +KTS cells, PCR amplification of genomic DNA isolated from these tumors was performed to detect the presence of the integrated transgene containing the WT1 cDNAs (visualized as a 540-bp PCR product;Fig. 4,B), thereby indicating that tumors in mice receiving injections of M1.WT1 +KTS cells arose from M1 cell clones that showed inadequate levels of WT1 +KTS isoform expression. Amplification of β-actin (a 1.2-kb PCR product) is also shown as an indicator of genomic DNA recovery from each sample (Fig. 4 B).

We had shown previously that the Wilms’ tumor suppressor gene, WT1, is a key mediator in the molecular control of LIF-induced M1 cell monocyte/macrophage differentiation in vitro(18). In concordance with the induction of WT1 gene expression during M1 cell macrophage differentiation upon exposure to LIF, ectopic expression of the +KTS isoforms of WT1 in parental M1 cells leads to the spontaneous differentiation of M1 cell blasts along the monocytic lineage in the absence of external differentiation-inducing stimuli. In this study, we now demonstrate that although the differentiation of M1 cells induced by WT1 +KTS isoform expression is associated with a reduction in clonogenicity in soft agar, a renewable subpopulation of blast cells is preserved in these cultures, thereby ensuring their persistence as stable cell lines. However, WT1 +KTS isoform expression in these cells does not abrogate their ability to respond to LIF and, moreover, can suppress the M1 cell leukemic phenotype in vivo, as determined by transplantation of these cells into immunodeficient SCID mice. These observations suggest that WT1 may play a dual role in the maintenance of stem cell populations, in addition to the induction of normal monocyte/macrophage differentiation and suppression of leukemogenicity.

Because of the plethora of data implicating WT1 in the etiology of WT,previous studies on the ability of specific WT1 isoforms to suppress cellular tumorigenicity have focused on malignant kidney-derived cell lines (11, 12, 13), although the ability of WT1 to inhibit tumor growth of rasH-transformed NIH3T3 cells has also been investigated (23). This present study is the first to extend this assessment to include a tumor cell line of hematopoietic origin, and a comparison of WT1 isoform tumor-suppression activity in cell lines examined to date is shown in Table 2. In summary, all WT1 isoforms have exhibited the capacity to suppress tumor formation in at least one cell type; however, the degree of protection afforded by each isoform varies according to the cell line under investigation. Across both renal and hematopoietic cell types,the WT1.2(−/+) isoform appears to be able to uniformly suppress tumor cell growth quite markedly, or even, as in the case of our current study in M1 cells, completely at one site of injection. The WT1.1(+/+)isoform can also function as a tumor suppressor in both cell types tested, although the effect is not as pronounced as with the WT1.2(−/+) isoform. The ability of the WT1.3(+/−) isoform to suppress tumor cell growth in vivo has only been examined in kidney-derived cell lines and is reasonably effective in abrogating tumorigenesis in these cells. The WT1 isoform that lacks both splice inserts, WT1.4(−/−), has demonstrated the ability to circumvent oncogenic transformation of NIH3T3 cells by Harvey ras. However, in adenovirus-transformed kidney cells, this isoform seemingly promotes tumor cell growth in vivo. This result is quite curious, given the growth-suppressive effects that have been observed across a range of cell types forced to express WT1 −KTS isoforms(reviewed in Ref. 24). Clarification as to the role of this isoform in either preventing or promoting tumor cell growth can only come through a more extensive analysis in additional cell lines.

In vitro studies into the role of WT1 in hematopoietic tumors have yielded no clear evidence as to its function in these cells. Endogenous expression of WT1 in the leukemic cell lines K562 and HL60 is down-regulated upon induction of cellular differentiation by chemical agents (25, 26), and furthermore, exposure of K562 cells to WT1 antisense oligonucleotides results in the cessation of cellular proliferation and onset of apoptosis in these cells (27, 28). Also, the enforced expression of WT1 in U937 and HL60 cells renders these lines resistant to induction of monocyte/macrophage differentiation (21, 22). Although these studies are somewhat difficult to interpret, they tend to imply an oncogenic, rather than tumor-suppressive, function for WT1 in these systems. The behavior of WT1 in the myeloblastic M1 cell line, however, has favored a role in either the induction of growth suppression and apoptosis (−KTS isoforms; Ref. 29) or normal monocytic differentiation and a nontumorigenic phenotype in vivo (+KTS isoforms; Ref. 18; this study) and is consistent with imparting a tumor-suppressive effect in these cells. Perceived differences as to the role of WT1 in leukemogenesis may be in part attributable to differences in the in vitro cell systems used to study its function, including the presence of WT1-interacting protein partners,and whether the genetic lesion responsible for initial leukemic transformation events in these cells lies upstream or downstream of WT1.

In summary, this investigation is the first to demonstrate the tumor-suppressive effects of WT1 expression in a leukemic cell line. As assessed by both in vitro and in vivo assays,ectopic expression of WT1 +KTS isoforms has the potential to revert the leukemic phenotype of myeloblastic M1 cells. This study further advances the notion that, as in the developing kidney, WT1acts as a tumor suppressor gene during hematopoiesis, whereby loss of functional gene expression may contribute to leukemogenesis in vivo.

Fig. 1.

A, colony formation in soft agar by M1 cells is markedly reduced by the expression of WT1 +KTS isoforms,particularly in the presence of LIF. Cells were plated in the presence or absence of LIF, and after 7 days, colonies were scored as clones containing >50 cells, whereas abortive clones containing <50 cells were scored as clusters. ▪, colonies; , clusters. B, photomicrographs showing the morphology of the various M1 cell populations upon soft agar culture in the absence of LIF. Colonies shown are representative of those observed throughout the culture. Unstained. ×40.

Fig. 1.

A, colony formation in soft agar by M1 cells is markedly reduced by the expression of WT1 +KTS isoforms,particularly in the presence of LIF. Cells were plated in the presence or absence of LIF, and after 7 days, colonies were scored as clones containing >50 cells, whereas abortive clones containing <50 cells were scored as clusters. ▪, colonies; , clusters. B, photomicrographs showing the morphology of the various M1 cell populations upon soft agar culture in the absence of LIF. Colonies shown are representative of those observed throughout the culture. Unstained. ×40.

Close modal
Fig. 2.

The survival of SCID mice receiving injections s.c. on the dorsal flank (A) or i.p. (B) with M1 cells is significantly enhanced by the expression of WT1 +KTS isoforms. Mice were injected with the various M1 cell populations, as described in “Materials and Methods,” and monitored for tumor-related mortality for a period of 20 weeks. , PBS; ▪, M1; •, M1.Neo;□, M1.WT1.1(+/+); ○, M1.WT1.2(−/+).

Fig. 2.

The survival of SCID mice receiving injections s.c. on the dorsal flank (A) or i.p. (B) with M1 cells is significantly enhanced by the expression of WT1 +KTS isoforms. Mice were injected with the various M1 cell populations, as described in “Materials and Methods,” and monitored for tumor-related mortality for a period of 20 weeks. , PBS; ▪, M1; •, M1.Neo;□, M1.WT1.1(+/+); ○, M1.WT1.2(−/+).

Close modal
Fig. 3.

Photomicrographs showing the histology of tumors resected from SCID mice receiving injections s.c. on the dorsal flank with the various M1 cell populations. M1.Neo and M1.WT1.1(+/+) cells formed myelosarcomas either localized to the site of injection(A and C, respectively) or viscerally disseminated within mesenteric membranes of the peritoneal cavity,leading to the neoplastic infiltration of associated organs such as the duodenum and stomach (B and D,respectively). H&E stain. ×100.

Fig. 3.

Photomicrographs showing the histology of tumors resected from SCID mice receiving injections s.c. on the dorsal flank with the various M1 cell populations. M1.Neo and M1.WT1.1(+/+) cells formed myelosarcomas either localized to the site of injection(A and C, respectively) or viscerally disseminated within mesenteric membranes of the peritoneal cavity,leading to the neoplastic infiltration of associated organs such as the duodenum and stomach (B and D,respectively). H&E stain. ×100.

Close modal
Fig. 4.

Western and PCR analysis of the M1.WT1 +KTS cell lines and tumors subsequently derived from their injection into SCID mice demonstrate a loss of ectopic WT1 expression in the tumor samples,despite the presence of M1.WT1 +KTS cells. Total protein was extracted from cell lines, tumor ascites, or solid tumors, and duplicate blots were hybridized with antibodies to WT1 (A, upper panel)to detect ectopic WT1 expression or actin (A, lower panel) to demonstrate equivalent protein recovery and loading between samples. Genomic DNA isolated from identical samples was subjected to PCR using primers directed toward the integrated WT1 transgene (B, upper panel) to detect the presence of M1.WT1 +KTS cells in the tumors or β-actin (B, lower panel) to indicate the quality of DNA recovered from the samples.

Fig. 4.

Western and PCR analysis of the M1.WT1 +KTS cell lines and tumors subsequently derived from their injection into SCID mice demonstrate a loss of ectopic WT1 expression in the tumor samples,despite the presence of M1.WT1 +KTS cells. Total protein was extracted from cell lines, tumor ascites, or solid tumors, and duplicate blots were hybridized with antibodies to WT1 (A, upper panel)to detect ectopic WT1 expression or actin (A, lower panel) to demonstrate equivalent protein recovery and loading between samples. Genomic DNA isolated from identical samples was subjected to PCR using primers directed toward the integrated WT1 transgene (B, upper panel) to detect the presence of M1.WT1 +KTS cells in the tumors or β-actin (B, lower panel) to indicate the quality of DNA recovered from the samples.

Close modal

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

Supported by the Queensland Cancer Fund, the Leukaemia Foundation of Queensland, and the National Health and Medical Research Council of Australia.

3

The abbreviations used are: WT, Wilms’ tumor;ZF, zinc finger; KTS, lysine-threonine-serine; LIF, leukemia inhibitory factor; SCID, C.B-17 scid/scid.

Table 1

Tumorigenicity of M1, M1.Neo, M1.WT1.1(+/+), and M1.WT1.2(−/+) cells inoculated into C.B-17 scid/scid mice

Cell lineTumor incidencea (%)P              bRange latency periodc days (median)
s.c.i.p.s.c.i.p.s.c.i.p.
PBS 0/12 (0) 0 /12 (0) NAd NA NA NA 
M1 11/12 (92) 12 /12 (100) NA NA 14–35 (21) 28–70 (49) 
M1.Neo 12/12 (100) 10 /12 (83) NA NA 14–94 (30) 42–77 (54) 
M1.WT1.1(+/+) 3/12 (25) 2 /12 (17) 0.000032 0.000132 31–77 (59) 42–70 (56) 
M1.WT1.2(−/+) 0/11c (0) 2 /12 (17) 0.000004 0.000080 NA 49–70 (60) 
Cell lineTumor incidencea (%)P              bRange latency periodc days (median)
s.c.i.p.s.c.i.p.s.c.i.p.
PBS 0/12 (0) 0 /12 (0) NAd NA NA NA 
M1 11/12 (92) 12 /12 (100) NA NA 14–35 (21) 28–70 (49) 
M1.Neo 12/12 (100) 10 /12 (83) NA NA 14–94 (30) 42–77 (54) 
M1.WT1.1(+/+) 3/12 (25) 2 /12 (17) 0.000032 0.000132 31–77 (59) 42–70 (56) 
M1.WT1.2(−/+) 0/11c (0) 2 /12 (17) 0.000004 0.000080 NA 49–70 (60) 
a

Tumor incidence was defined as the number of mice developing tumors proportionate to the total number of mice initially receiving injections in the group.

b

P was determined by a nonparametric Mann-Whitney U test.

c

Latency period was defined as the number of days between inoculation and first detection of the tumor.

d

NA, not applicable.

e

Of the 12 mice initially receiving injections of M1.WT1.2(−/+) cells at this site, one animal died from an unrelated respiratory illness and was excluded from the study.

Table 2

Effect of WT1 isoform expression on the tumorigenicity of malignant cell lines upon transplantation in vivo

Cell lineDerivationEffecta on tumorigenicity after expression of WT1 isoformReference
WT1.1(+/+)WT1.2(−/+)WT1.3(+/−)WT1.4(−/−)
RM1 Human WT cells ↓↓ NTb ↓ NT (11) 
G401 Human wt cells NT ↓↓ ↓ NT (12) 
Ad-BRK Rat adenovirus-transformed kidney cells ←→ ↓↓ ←→ ↑↑ (13) 
rasH-NIH3T3 Murine Harvey ras-transformed fibroblasts NT NT NT ↓↓ (23) 
M1 Murine myeloblastic leukemia cells ↓ ↓↓ NT NT This study 
Cell lineDerivationEffecta on tumorigenicity after expression of WT1 isoformReference
WT1.1(+/+)WT1.2(−/+)WT1.3(+/−)WT1.4(−/−)
RM1 Human WT cells ↓↓ NTb ↓ NT (11) 
G401 Human wt cells NT ↓↓ ↓ NT (12) 
Ad-BRK Rat adenovirus-transformed kidney cells ←→ ↓↓ ←→ ↑↑ (13) 
rasH-NIH3T3 Murine Harvey ras-transformed fibroblasts NT NT NT ↓↓ (23) 
M1 Murine myeloblastic leukemia cells ↓ ↓↓ NT NT This study 
a

Effect on tumorigenicity after WT1 isoform expression was scored as: ↓↓, markedly suppressed; ↓, moderately suppressed;←→, no effect; ↑↑, markedly enhanced.

b

NT, not tested.

We thank Dr. Elizabeth Algar, Dr. Melissa Little, and Prof. David Hume for helpful discussion and Dr. Gu Zhu for assistance with the statistical analysis. The excellent histology work performed by Estelle Schoch and Michael Walsh, and the veterinary pathology advice provided by Dr. John Finnie, are also gratefully acknowledged.

1
Hastie N. D. The genetics of Wilms’ tumor–a case of disrupted development.
Annu. Rev. Genet.
,
28
:
523
-558,  
1994
.
2
Haber D. A., Sohn R. L., Buckler A. J., Pelletier J., Call K. M., Housman D. E. Alternative splicing and genomic structure of the Wilms tumor gene WT1.
Proc. Natl. Acad. Sci. USA
,
88
:
9618
-9622,  
1991
.
3
Pritchard-Jones K., Fleming S., Davidson D., Bickmore W., Porteous D., Gosden C., Bard J., Buckler A., Pelletier J., Housman D., van Heyningen V., Hastie N. The candidate Wilms’ tumour gene is involved in genitourinary development.
Nature (Lond.)
,
346
:
194
-197,  
1990
.
4
Kreidberg J. A., Sariola H., Loring J. M., Maeda M., Pelletier J., Housman D., Jaenisch R. WT-1 is required for early kidney development.
Cell
,
74
:
679
-691,  
1993
.
5
Little M., Wells C. A clinical overview of WT1 gene mutations.
Hum. Mutat.
,
9
:
209
-225,  
1997
.
6
Inoue K., Ogawa H., Sonoda Y., Kimura T., Sakabe H., Oka Y., Miyake S., Tamaki H., Oji Y., Yamagami T., Tatekawa T., Soma T., Kishimoto T., Sugiyama H. Aberrant overexpression of the Wilms tumor gene (WT1) in human leukemia.
Blood
,
89
:
1405
-1412,  
1997
.
7
Miwa H., Beran M., Saunders G. F. Expression of the Wilms’ tumor gene (WT1) in human leukemias.
Leukemia (Baltimore)
,
6
:
405
-409,  
1992
.
8
King-Underwood L., Renshaw J., Pritchard-Jones K. Mutations in the Wilms’ tumor gene WT1 in leukemias.
Blood
,
87
:
2171
-2179,  
1996
.
9
King-Underwood L., Pritchard Jones K. Wilms’ tumor (WT1) gene mutations occur mainly in acute myeloid leukemia and may confer drug resistance.
Blood
,
91
:
2961
-2968,  
1998
.
10
Inoue K., Sugiyama H., Ogawa H., Nakagawa M., Yamagami T., Miwa H., Kita K., Hiraoka A., Masaoka T., Nasu K., Kyo T., Dohy H., Nakauchi H., Ishidate T., Akiyama T., Kishimoto T. WT1 as a new prognostic factor and a new marker for the detection of minimal residual disease in acute leukemia.
Blood
,
84
:
3071
-3079,  
1994
.
11
Haber D. A., Park S., Maheswaran S., Englert C., Re G. G., Hazen Martin D. J., Sens D. A., Garvin A. J. WT1-mediated growth suppression of Wilms tumor cells expressing a WT1 splicing variant.
Science (Washington DC)
,
262
:
2057
-2059,  
1993
.
12
McMaster M. L., Gessler M., Stanbridge E. J., Weissman B. E. WT1 expression alters tumorigenicity of the G401 kidney-derived cell line.
Cell Growth Differ.
,
6
:
1609
-1617,  
1995
.
13
Menke A. L., Riteco N., van Ham R. C., de Bruyne C., Rauscher F. J., III, van der Eb A. J., Jochemsen A. G. Wilms’ tumor 1 splice variants have opposite effects on the tumorigenicity of adenovirus-transformed baby rat kidney cells.
Oncogene
,
12
:
537
-546,  
1996
.
14
Ichikawa Y. Differentiation of a cell line of myeloid leukemia.
J. Cell. Physiol.
,
74
:
223
-234,  
1969
.
15
Honma Y., Kasukabe T., Hozumi M. Relationship between leukemogenicity and in vivo inducibility of normal differentiation in mouse myeloid leukemia cells.
J. Natl. Cancer Inst.
,
61
:
837
-841,  
1978
.
16
Ichikawa Y. Further studies on the differentiation of a cell line of myeloid leukemia.
J. Cell. Physiol.
,
76
:
175
-184,  
1970
.
17
Honma Y., Kasukabe T., Okabe J., Hozumi M. Prolongation of survival time of mice inoculated with myeloid leukemia cells by inducers of normal differentiation.
Cancer Res.
,
39
:
3167
-3171,  
1979
.
18
Smith S. I., Weil D., Johnson G. R., Boyd A. W., Li C. L. Expression of the Wilms’ tumor suppressor gene, WT1, is upregulated by leukemia inhibitory factor and induces monocytic differentiation in M1 leukemic cells.
Blood
,
91
:
764
-773,  
1998
.
19
Ikeda K., Monden T., Kanoh T., Tsujie M., Izawa H., Haba A., Ohnishi T., Sekimoto M., Tomita N., Shiozaki H., Monden M. Extraction and analysis of diagnostically useful proteins from formalin-fixed, paraffin-embedded tissue sections.
J. Histochem. Cytochem.
,
46
:
397
-403,  
1998
.
20
Mies C., Houldsworth J., Chaganti R. S. Extraction of DNA from paraffin blocks for Southern blot analysis.
Am. J. Surg. Pathol.
,
15
:
169
-174,  
1991
.
21
Svedberg H., Chylicki K., Baldetorp B., Rauscher F. J. R., Gullberg U. Constitutive expression of the Wilms’ tumor gene (WT1) in the leukemic cell line U937 blocks parts of the differentiation program.
Oncogene
,
16
:
925
-932,  
1998
.
22
Deuel T. F., Guan L. S., Wang Z. Y. Wilms’ tumor gene product WT1 arrests macrophage differentiation of HL- 60 cells through its zinc-finger domain.
Biochem. Biophys. Res. Commun.
,
254
:
192
-196,  
1999
.
23
Luo X. N., Reddy J. C., Yeyati P. L., Idris A. H., Hosono S., Haber D. A., Licht J. D., Atweh G. F. The tumor suppressor gene WT1 inhibits ras-mediated transformation.
Oncogene
,
11
:
743
-750,  
1995
.
24
Menke A. L., van der Eb A. J., Jochemsen A. G. The Wilms’ tumor 1 gene: oncogene or tumor suppressor gene?.
Int. Rev. Cytol.
,
181
:
151
-212,  
1998
.
25
Phelan S. A., Lindberg C., Call K. M. Wilms’ tumor gene, WT1, mRNA is down-regulated during induction of erythroid and megakaryocytic differentiation of K562 cells.
Cell Growth Differ.
,
5
:
677
-686,  
1994
.
26
Sekiya M., Adachi M., Hinoda Y., Imai K., Yachi A. Downregulation of Wilms’ tumor gene (wt1) during myelomonocytic differentiation in HL60 cells.
Blood
,
83
:
1876
-1882,  
1994
.
27
Algar E. M., Khromykh T., Smith S. I., Blackburn D. M., Bryson G. J., Smith P. J. A WT1 antisense oligonucleotide inhibits proliferation and induces apoptosis in myeloid leukaemia cell lines.
Oncogene
,
12
:
1005
-1014,  
1996
.
28
Yamagami T., Sugiyama H., Inoue K., Ogawa H., Tatekawa T., Hirata M., Kudoh T., Akiyama T., Murakami A., Maekawa T. Growth inhibition of human leukemic cells by WT1 (Wilms tumor gene) antisense oligodeoxynucleotides: implications for the involvement of WT1 in leukemogenesis.
Blood
,
87
:
2878
-2884,  
1996
.
29
Murata Y., Kudoh T., Sugiyama H., Toyoshima K., Akiyama T. The Wilms tumor suppressor gene WT1 induces G1 arrest and apoptosis in myeloblastic leukemia M1 cells.
FEBS Lett.
,
409
:
41
-45,  
1997
.