This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Renshaw, J. Articles by Pritchard-Jones, K. Search for Related Content PubMed PubMed Citation Articles by Renshaw, J. Articles by Pritchard-Jones, K.

¹ Section of Paediatrics, and ² Cancer Research UK Centre of Cancer Therapeutics, Institute of Cancer Research, Sutton, Surrey, United Kingdom and ³ Isis Pharmaceuticals, Inc., Carlsbad, California

Requests for reprints: Jane Renshaw, Section of Paediatrics, Institute of Cancer Research, 15 Cotswold Road, Belmont, Sutton, Surrey SM2 5NG, United Kingdom. Phone: 44-208-7224327; Fax: 44-208-7224321. E-mail: Jane.Renshaw{at}icr.ac.uk

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Deregulated expression of the Wilms' tumor gene (WT1) has been implicated in the maintenance of a malignant phenotype in leukemias and a wide range of solid tumors through interference with normal signaling in differentiation and apoptotic pathways. Expression of high levels of WT1 is associated with poor prognosis in leukemias and breast cancer. Using real-time (Taqman) reverse transcription-PCR and RNase protection assay, we have shown up-regulation of WT1 expression following cytotoxic treatment of cells exhibiting drug resistance, a phenomenon not seen in sensitive cells. WT1 is subject to alternative splicing involving exon 5 and three amino acids (KTS) at the end of exon 9, producing four major isoforms. Exon 5 splicing was disrupted in all cell lines studied following a cytotoxic insult probably due to increased exon 5 skipping. Disruption of exon 5 splicing may be a proapoptotic signal because specific targeting of WT1 exon 5–containing transcripts using a nuclease-resistant antisense oligonucleotide (ASO) killed HL60 leukemia cells, which were resistant to an ASO targeting all four alternatively spliced transcripts simultaneously. K562 cells were sensitive to both target-specific ASOs. Gene expression profiling following treatment with WT1 exon 5–targeted antisense showed up-regulation of the known WT1 target gene, thrombospondin 1, in HL60 cells, which correlated with cell death. In addition, novel potential WT1 target genes were identified in each cell line. These studies highlight a new layer of complexity in the regulation and function of the WT1 gene product and suggest that antisense directed to WT1 exon 5 might have therapeutic potential.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
The Wilms' tumor gene (WT1) is located at the human chromosome region 11p13 and encodes a developmentally regulated transcription factor that is essential for normal genitourinary development (reviewed in refs. 1, 2). In the adult, WT1 expression is restricted to specific cell types in kidney, gonads, hematopoietic and nervous system, and mesothelium (1). However, inappropriate and/or overexpression of WT1 has been reported in leukemias and a wide range of solid tumors including prostate, breast, and lung as well as thyroid, testicular and ovarian carcinomas, melanoma, and mesothelioma (2). Although an oncogenic role in these tumors has not been proven, experimental evidence suggests that expression of WT1 may contribute to the maintenance of a malignant phenotype through a variety of mechanisms including inhibition of differentiation and apoptosis and increased proliferation (reviewed in ref. 2).

The WT1 protein consists of 10 exons with an activator/repressor domain near the NH₂ terminus and four zinc fingers of the Cys₂-His₂ type at the COOH terminus. Alternative translation initiation site usage (3, 4) and alternative splicing of WT1 mRNA (5) produces multiple WT1 protein isoforms. The two alternative splice regions correspond to the 17 amino acids of exon 5 (present only in mammals) and the last three amino acids (KTS) of exon 9 (conserved in all vertebrates; ref. 6). Insertion of KTS alters the spacing between zinc fingers 3 and 4, disrupting sequence-specific DNA binding and, in transient cotransfection assays, transcriptional regulation (reviewed in ref. 2). In addition, +KTS isoforms have been shown to colocalize with splicing factors in the nucleus, suggesting a role in RNA processing (7). In these assays, the presence or absence of exon 5 has little impact.

The four major isoforms of WT1, designated WT1 (+/+), WT1 (+/–), WT1 (–/+), and WT1 (–/–) to indicate the presence or absence of exon 5/KTS, respectively (see Fig. 1), are in general quoted as being coexpressed in a fixed ratio in normal tissues (5) and this is true for the presence or absence of KTS (the WT1 KTS ratio). Studies of transgenic mice have shown that WT1 isoforms with and without the KTS insert have separate and vital functions in sex determination and gonadal development, consistent with a fixed expression ratio (8, 9). However, we and others have shown that WT1 exon 5 ratios differ according to tissue and differentiation stage as well as between species (6, 10). Differential splicing of exon 5 has been confirmed in human kidney and Wilms' tumor samples through analysis of native WT1 protein (11). Mice lacking the WT1 exon 5 insert develop normally and are not compromised with respect to fertility, embryo viability, or the capacity to lactate, suggesting that exon 5 is redundant in genitourinary development and function at least in mice (12). Nevertheless, the presence of the exon 5 insert and the maintenance of the correct balance between WT1 + exon 5 and WT1 – exon 5 isoforms has been suggested to be essential for the regulation of critical cellular functions such as proliferation, differentiation, and resistance to chemotherapeutic drugs (13–16). Furthermore, disruption of exon 5 splicing has been suggested to affect the regulation of genes downstream in the WT1 pathway, as has been shown in Wilms' tumors that express reduced levels of WT1 exon 5 variants relative to those lacking exon 5 (17).

View larger version (13K): [in this window] [in a new window] [Download PPT slide]   Figure 1. Schematic representation of WT1 alternatively spliced variants. Four main splice variants are represented schematically to show the presence or absence of exon 5 and KTS along with their designations: WT1 (+/+), (+/–), (–/+), and (–/–) and WT1 + and – exon 5.

In our previous study, exon 5–containing WT1 transcripts were shown to be in excess in leukemias, whereas in gonadal tumors they were nearer equivalence or marginally underrepresented. These findings were intriguing because the individual isoforms of WT1 seem to mediate different downstream cell type–specific biological effects that may involve isoform-specific interactions with coregulatory binding proteins such as Par4 and cyclic AMP response element binding protein binding protein (reviewed in ref. 2). For example, Par4 interacts with both exon 5 and the zinc fingers of WT1 resulting in opposing effects on transcription and, in exon 5 interaction, the rescue of 293 cells from lethal UV light treatment (25, 26). It is therefore probable that Par4 may mediate both positive and negative interactions with WT1 dictated, at least in part, by the relative ratio of WT1 + to – exon 5 isoforms.

Numerous reports have described cell type–specific regulation of WT1 expression following induction of differentiation in leukemic, embryonal stem cell, and carcinoma cell lines accompanied by growth arrest and apoptosis in some cell types (27–30). We initiated the present study to determine whether similar alterations in the regulation of WT1 mRNA levels and/or alternative splicing were induced following an apoptotic stimulus such as cytotoxic drug treatment. The possibility of differential regulation of WT1 expression in cell lines sensitive to or with acquired resistance to a cytotoxic agent was investigated using two paired cell lines: the paired ovarian papillary cystadenocarcinoma cell lines, CH1-S and CH1-R (sensitive to and with acquired resistance to cisplatin, respectively), and the corresponding testicular germ cell tumor lines, GCT27-S and GCT27-R. The levels of total WT1 and individual alternatively spliced WT1 variants were measured in these cell lines following cisplatin treatment and, in the erythroleukemia cell line K562, following treatment with doxorubicin. A combination of radioactive reverse transcription-PCR (RT-PCR), quantitative real-time (Taqman) RT-PCR, and a RNase protection assay (RPA) was used.

We report here the novel observation that both WT1 exon 5 splicing and total WT1 mRNA transcript expression were dynamically regulated following cytotoxic treatment. Disruption of exon 5 splicing resulted from the down-regulation of WT1 exon 5–containing transcripts relative to those lacking exon 5, probably by a mechanism of induced exon 5 skipping. Induction of total WT1 expression was seen only in those cell lines displaying a resistant phenotype. The downstream sequelae of disrupted exon 5 splicing were investigated further using the leukemic cell lines K562 and HL60 and a combination of antisense and gene expression profiling technology. Both cell lines express high levels of WT1, and HL60 cells have been shown previously to be resistant to WT1-directed antisense oligonucleotides (ASOs), which kill K562 cells (31). These results provide novel insights into the role of WT1 exon 5 splicing and total WT1 expression in the regulation of cell survival signaling pathways. Antisense targeted to WT1 exon 5 may have broader therapeutic potential than previously described WT1-directed ASOs.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Oligonucleotides
The 20-mer, 2'-O-methoxyethyl chimeric oligonucleotides consisting of a central window of eight 2'-deoxy unmodified sugar residues with flanking 2'-O-methoxyethyl regions and a fully thioated backbone were synthesized by Isis Pharmaceuticals Inc. (Carlsbad, CA) as described previously (32). The two active compounds selected for this study were ISIS 16609, sequence 5'-GCCCTTCTGTCCATTTCACT-3', targeting exon 5 (ASWT1 exon 5) and ISIS 16601, sequence 5'-CACATACACATGCCCTGGCC-3', targeting the 3' untranslated region (UTR) of WT1 (ASWT13'UTR). Two oligonucleotides of the same chemistry were used as controls: ISIS 15770, sequence 5'-ATGCATTCTGCCCCCAAGGA-3', a 5-10-5 gapmer targeting murine c-raf kinase (ASmc-raf) and lacking homology to any known human sequence, and ISIS 105730, sequence 5'-CCATCGACCTGCACCGATCA-3', a scrambled sequence of ASWT13'UTR (ASWT1 scram).

Cell Culture and Drug Treatment
K562 cl.6 cells, a subclone of the parent erythroleukemia (33), were kindly provided by Prof. Adrian Newland and Dr. Xu-Rong Jiang (London Hospital Medical College, London, United Kingdom). HL60 promyelocytic leukemia cells were obtained from the American Type Culture Collection (Manassas, VA). K562 cl.6 and HL60 cells were grown in RPMI 1640 (HEPES buffered) supplemented with 10% FCS (Biowest, Ringmer, East Sussex, United Kingdom) and penicillin (100 units/mL). The in-house paired ovarian cell lines CH1-S and CH1-R (34) and testicular germ cell tumor cell lines GCT27-S and GCT27-R (35) were cultured in DMEM (bicarbonate buffered) supplemented with 10% FCS. Doxorubicin (Sigma-Aldrich Co. Ltd., Gillingham, Dorset, United Kingdom) and cisplatin (Johnson Matthey Technology Centre, Reading, Berkshire, United Kingdom) were dissolved in sterile saline and cells were treated (100 µL drug in saline to 10 mL cultures) either continuously (doxorubicin) or for 2 hours (cisplatin) at a concentration determined previously as the IC₉₉ in clonogenic assays: K562, 0.1 µmol/L doxorubicin; CH1-S, 3.8 µmol/L; CH1-R, 18.5 µmol/L; GCT27-S, 9.3 µmol/L; and GCT27-R, 40 µmol/L cisplatin. Antisense or control oligonucleotides were dissolved in PBS and introduced into K562 or HL60 cells by low-voltage electroporation. Briefly, appropriately diluted ASOs (40 µL) were combined with cell suspension (360 µL) at 2 x 10⁷ cells/mL and cells were electroporated (Bio-Rad Gene Pulser II electroporation system with Pulse Controller Plus capacitance extender accessory module, Bio-Rad Laboratories Ltd., Hemel Hempstead, Hertfordshire, United Kingdom) using 300 V and a capacitor value of 1,000 µF, diluted to 10 mL, and incubated at 37°C for various times.

Clonogenic Cell Survival Assays
Following drug treatment, appropriate aliquots of cells were serially diluted in complete medium. Aliquots (2 mL) of diluted cells were added to polystyrene tubes (Elkay Products Ltd., Basingstoke, Hampshire, United Kingdom) containing medium (3 mL) supplemented with 20% FCS and 0.2% Agar Noble (Difco Laboratories, Detroit, MI) and incubated at 37°C and colonies were counted after 14 days. In several separate experiments, plating efficiencies ranged from 22% to 38%, with 800 to 1,600 cells initially plated.

RNA Extraction and Radioactive RT-PCR
At the appropriate times following drug treatment, cells were harvested and RNA was extracted using Trizol (Life Technologies Ltd., Paisley, United Kingdom). RNA (1 µg) was reverse transcribed with SuperScript II and random hexamer primers (100 pmol, Invitrogen Ltd., Paisley, United Kingdom) in a final volume of 20 µL according to the manufacturer's instructions. RT-PCR of all four WT1 alternatively spliced mRNA transcripts was carried out using 1 µL cDNA, [-³²P]dCTP, reduced cold dCTP, and primer pair 297 and 298 spanning WT1 exons 4 to 10. RT-PCR methodology and method validation have been described in detail previously (10). Four individual reaction mixes for each sample were set up in parallel and amplified for 26, 29, 32, and 35 cycles, respectively, and PCR products 482 (–/–), 491 (–/+), 533 (+/–), and 542 (+/+) bp long were separated on standard denaturing polyacrylamide gels. Levels of [³²P]dCTP incorporated into all four transcripts were visualized and analyzed using a Storm 860 PhosphorImager and ImageQuant software (Amersham Pharmacia Biotech UK Ltd., Little Chalfont, Buckinghamshire, United Kingdom) followed by adjustment for the number of possible sites of incorporation of ³²P in the alternatively spliced PCR products.

RNase Protection Assay
A cDNA sequence spanning the whole of WT1 exon 5 and 70 bp of exon 4 was amplified by RT-PCR from a sample of normal human testis. PCR products were subcloned by standard methods and sequenced (automated fluorescent sequencing) using an ABI PRISM 310 Genetic Analyzer (Applied Biosystems, Applera United Kingdom, Warrington, Cheshire, United Kingdom). High specific activity WT1 antisense riboprobes were generated using 1 µg of purified linearized vector, [-³²P]UTP, and a Maxiscript labeling kit (Ambion Europe Ltd., Huntindon, Cambridgeshire, United Kingdom) according to the manufacturer's instructions but with the addition of single-stranded binding protein (1 µL, Amersham Pharmacia Biotech UK). RPA analyses were carried out according to the manufacturer's instructions using total RNA (10 µg) and the RPA III kit (Ambion Europe). WT1 RNase protected fragments, 121 (+ exon 5) and 70 (– exon 5) bp, along with control actin (Ambion Europe) protected fragments, were separated in 1 hour using 8% standard denaturing polyacrylamide gels. Levels of [³²P]UTP incorporated into the protected fragments were visualized and analyzed using a Storm 860 PhosphorImager as above.

Real-time (Taqman) PCR
The primer pairs and probes for the quantitation of WT1, thrombospondin 1 (THBS1), and glypican 5 (GPC5) levels were designed using the Primer Express program (Applied Biosystems) according to the recommended guidelines: WT1 forward primer 5'-TACCCAGGCTGCAATAAGAGATATTTTAAG-3', reverse primer 5'-CCTTTGGTGTCTTTTGAGCTGGTC-3', and probe 5'-CACTGGTGAGAAACCATACCAGTGTGACTTCAAGGACT-3'; THBS1 forward primer 5'-GACAGCATCCGCAAAGTGACT-3', reverse primer 5'-GACAGTGACACTCAGTGCAGCTATC-3', and probe 5'-TGAGCTGAGGCGGCCTCCCCTA-3'; and GPC5 forward primer 5'-GGGCTGCCGGATTCG-3', reverse primer 5'-CTGGTGCAACATGTAGGCTTTT-3', and probe 5'-CGCGGGCAGGACCTGATCTTCA-3'. The primers were designed to amplify across exon/exon boundaries and were confirmed to be lacking significant homology with any known DNA sequences by a search of the HGMP database. All other genes were analyzed using Assays-on-Demand Gene Expression Products (Applied Biosystems). Taqman analysis was carried out according to the manufacturer's instructions using an Applied Biosystems 7700 Sequence Detector. Each assay sample was analyzed in triplicate and multiplexed to facilitate the measurement of gene expression levels relative to 18s rRNA expression (rRNA control reagents, Applied Biosystems) using the standard curve method.

Protein Extraction and Estimation Using Western Immunoblotting
At the appropriate times following drug treatment, 5 x 10⁶ cells were harvested, washed in PBS, and lysed using 2% SDS containing 10% v/v protease inhibitor cocktail (Sigma-Aldrich). Sample protein concentrations were estimated using the Bio-Rad detergent-compatible microtiter plate protein assay according to the manufacturer's instructions. Prior to electrophoresis, an aliquot of each sample was treated with RNase-free DNase (2 units, Ambion Europe) in the presence of Mg²⁺ (5 mmol/L), total volume 20 µL, at 37°C for 30 minutes followed by 75°C for 5 minutes to minimize protein band distortion. Electrophoretic separation of the + and – exon 5 WT1 isoforms (20 µg protein loaded) was achieved using NuPAGE 10% Bis-Tris precast gels (Invitrogen, Groningen, Netherlands) run with NuPAGE MOPS and SDS denaturing running buffer. WT1 proteins were analyzed using a 1:1,000 dilution of the rabbit polyclonal WT(C-19) (Santa Cruz Biotechnology, Santa Cruz, CA). Horseradish peroxidase–conjugated anti-rabbit F(ab')₂ fragment secondary antibodies (Amersham Bioscience UK Ltd.) were used at 1:2,000 dilution, detected using the Enhanced Chemiluminescence Plus system (Amersham Bioscience UK Ltd.), and visualized and analyzed using a STORM PhosphorImager 860 and ImageQuant software. For WT1 estimation, a nonspecific band, running slightly faster than WT1 and shown previously to be minimally affected by antisense treatment, was used to adjust for differences in loading and sample protein concentration.

PolyA + mRNA Isolation and cDNA Microarray Analysis
Multiple (10x) aliquots of K562 and HL60 cells were treated with ASO (10 µmol/L) as described above. After 24 hours, total RNA was extracted and the samples were pooled. PolyA+ mRNA was isolated and purified from total RNA using an Oligotex kit (Qiagen Ltd., Crawley, West Sussex, United Kingdom.) and concentrated using Microcon YM-30 centrifugal filter devices (Millipore, Watford, Hertfordshire, United Kingdom). cDNA microarray analyses of each sample were done in duplicate. The preparation of the cDNA microarray slides, and fluorescent labeling of polyA+ mRNA samples using Superscript II and Cy5-labeled or Cy3-labeled dCTP, was carried out as described previously (36). The microarray hybridization was carried out according to a published protocol (37). The so-called Institute of Cancer Research gene set, consisting of 5,603 IMAGE cDNA clones, is also described in detail in this latter publication along with the image collection and data analysis procedures. Briefly, array images were acquired with an Axon GenePix 4000 scanner (Axon Instruments, Foster City, CA) and analyzed with the Axon GenePix Pro 3 software package (Axon Instruments). Data were filtered for quality by automated spot flagging and manual inspection. Fluorescence intensity ratios (I = Cy5/Cy3) were calculated after background subtraction and normalized to the median expression ratio of all high quality spots. Intensity ratios were transferred to a Microsoft Access database for IMAGE clone to gene assignments, data filtering, and group queries. Gene assignments were checked and updated using National Center for Biotechnology Information UniGene, formatted for local database import by a Perl script.⁴

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Dynamic Alteration of WT1 exon 5 Alternative Splicing and Total WT1 Expression following Treatment of Gonadal Cell Lines with Cisplatin
The paired human ovarian cell lines CH1-S and CH1-R as well as the corresponding testicular germ cell tumor cell lines GCT27-S and GCT27-R have been shown previously to undergo apoptosis following equitoxic concentrations of cisplatin (38, 39). Following treatment with cisplatin at IC₉₉ concentrations (concentration of drug reducing cell survival by 99% of control levels in clonogenic assays), the relative levels of all four WT1 alternatively spliced transcripts were analyzed using radioactive RT-PCR (Fig. 2). WT1 exon 5 ratios were rapidly reduced in all four cell lines, reaching a nadir at 8 hours post-treatment, whereas the WT1 KTS ratios remained relatively constant throughout (data not shown). Control 0-hour exon 5 ratios and ratios in samples processed at various times during the 24-hour period were not significantly different to those of untreated, logarithmically growing cells. In both paired cell lines, the extent of disruption of WT1 exon 5 ratios was greater in the resistant line than in the parent sensitive line, the mean ± range of WT1 exon 5 ratios when expressed as a percentage of control ratios being 65.7 ± 10% versus 48.5 ± 11.5% for the CH1-S and CH1-R cell lines, respectively, and 76.0 ± 14.6% versus 32.4 ± 8% for the GCT27-S and GCT27-R cell lines, respectively, at 8 hours post-treatment.

View larger version (35K): [in this window] [in a new window] [Download PPT slide]   Figure 2. Dynamic regulation of WT1 exon 5 transcript ratios following treatment with cisplatin. WT1 exon 5 ratios (see Fig. 1) were determined using radioactive RT-PCR (as illustrated in Fig. 3) at various time points over a 24-hour period following treatment of (A) CH1-S (*, P < 0.04), (B) CH1-R (**, P = 0.009), (C) GCT27-S (*, P = 0.08), and (D) GCT27-R (**, P = 0.002) cells with IC99 concentration of cisplatin. WT1 exon 5 transcript ratios are expressed as a percentage of ratios at 0 hour.

View larger version (26K): [in this window] [in a new window] [Download PPT slide]   Figure 3. Radioactive RT-PCR analysis of WT1 + and – exon 5 transcript levels following IC99 cisplatin treatment of CH1-S and CH1-R cells. Levels of WT1 amplification products with and without exon 5 generated in (A) CH1-S and (B) CH1-R cells were determined using radioactive RT-PCR following 26 cycles (logarithmic phase) of amplification. Aa and Ba, typical gel images (inset) of all four alternatively spliced RT-PCR products from selected time points along with superimposed line graphs generated from these images. Ab and Bb, normalized (adjusted for length) levels of + exon 5 (open columns) and – exon 5 (closed columns) WT1 transcripts over the 24-hour time course. C and D, confirmation of total WT1 transcript levels in CH1-S (C) and CH1-R (D) cells using real-time (Taqman) analysis.

View larger version (27K): [in this window] [in a new window] [Download PPT slide]   Figure 4. Radioactive RT-PCR analysis of WT1 + and – exon 5 transcript levels following IC99 cisplatin treatment of GCT27-S and GCT27-R cells. Levels of WT1 amplification products with and without exon 5 generated in (A) GCT27-S and (B) GCT27-R cells were determined using radioactive RT-PCR following 26 cycles (logarithmic phase) of amplification. Aa and Ba, typical gel images (inset) of all four alternatively spliced RT-PCR products from selected time points along with superimposed line graphs generated from these images. Ab and Bb, normalized (adjusted for length) levels of + exon 5 (open columns) and – exon 5 (closed columns) WT1 transcripts over the 24-hour time course. C and D, confirmation of total WT1 transcript levels in (C) GCT27-S and (D) GCT27-R cells using real-time (Taqman) analysis.

View larger version (35K): [in this window] [in a new window] [Download PPT slide]   Figure 5. RPA analysis of WT1 + and – exon 5 transcript levels and ratios following IC99 doxorubicin treatment of K562 cells. Levels of WT1 + and – exon 5 transcripts in K562 cells were determined using RPA analysis at various time points over a 24-hour time course following treatment with IC99 doxorubicin. A, WT1 exon 5 transcript ratios in K562 cells following doxorubicin treatment expressed as a percentage of ratios at 0 hour. B, total WT1 mRNA levels normalized to actin. C, typical gel images of RNase protected fragments (inset) along with line graphs generated from these images. D, normalized (adjusted for length) levels of + exon 5 (open columns) and – exon 5 (closed columns) WT1 transcripts over 24 hours following doxorubicin treatment.

Validation of Antisense Activity
Initial experiments using RPA analysis examined total WT1 transcript levels 5 hours following electroporation of K562 cells with 10 µmol/L ASWT13'UTR and ASWT1 exon 5 as compared with electroporation in the presence of PBS alone or the control ASO, ASmc-raf (Fig. 6A). Total WT1 transcript levels were equivalent in PBS-treated and ASmc-raf-treated cells and reduced to 55% and 30% of PBS control levels following treatment with ASWT13'UTR and ASWT1 exon 5, respectively. The WT1 exon 5 ratios did not alter as total WT1 levels were reduced following treatment with ASWT13'UTR. Following treatment with ASWT1 exon 5, WT1 exon 5–containing transcripts were reduced to <10% of control levels, whereas those lacking exon 5 were only minimally affected (Fig. 6B). These data show that ASWT1 exon 5 targets WT1 exon 5–containing transcripts selectively at early time points and also confirm postspliced exon 5–containing transcripts as the primary target. At 24 hours following treatment, WT1 levels were reduced to 26% and 22% of control levels by ASWT13'UTR and ASWT1 exon 5, respectively (Fig. 6C). At this time point, both WT1 + and – exon 5 transcripts were reduced by ASWT1 exon 5. However, selective targeting of WT1 + exon 5 transcripts was still apparent, the WT1 exon 5 ratios being 0.7 as compared with 1.8 following ASWT13'UTR (Fig. 6D).

View larger version (27K): [in this window] [in a new window] [Download PPT slide]   Figure 6. Efficient targeting WT1 mRNA transcripts by ASWT13'UTR and ASWT1 exon 5. RPA analysis of total WT1 levels in K562 cells 5 (A) and 24 (C) hours following treatment with PBS, ASmc-raf, ASWT13'UTR (AS-3'UTR), and ASWT1 exon 5 (AS-exon 5) expressed as a percentage of PBS levels. B and D, same samples expressed as counts (PhosphorImager units) and plotted individually as WT1 + exon 5 transcript levels (open columns) and WT1 – exon 5 transcript levels (closed columns). E, typical gel images of WT1 fragments generated by radioactive RT-PCR analysis (top) and RPA analysis (bottom) of ASO (10 µmol/L)–treated K562 cells. F, radioactive RT-PCR analysis of WT1 exon 5 ratios in K562 cells 4 hours following treatment with 0.5–10 µmol/L ASWT13'UTR (open columns) and ASWT1 exon 5 (closed columns) expressed as a percentage of ASmc-raf (10 µmol/L)–treated levels. G, time course of WT1 exon 5 ratios in K562 cells over the first 4 hours following treatment with 10 µmol/L ASWT13'UTR (open columns) and ASWT1 exon 5 (closed columns) expressed as a percentage of ASmc-raf (10 µmol/L)–treated levels and determined by radioactive RT-PCR. H, RPA analysis of total WT1 transcript levels 4 hours following treatment with 0.5–10 µmol/L ASWT13'UTR (open columns) and ASWT1 exon 5 (closed columns) expressed as a percentage of ASmc-raf (10 µmol/L)–treated levels.

Because we had clearly shown efficient targeting of WT1 expression in K562 cells, we used a simple nonradioactive RT-PCR screen to compare the activity of 10 µmol/L ASWT1 exon 5 and ASWT13'UTR in K562 and HL60 cells 24 hours following treatment. Equivalent activity was shown in both cell lines using both ASOs (data not shown). Antisense activity was subsequently confirmed in the HL60 cells used for cell survival studies using real-time (Taqman) RT-PCR (see below). Basal WT1 transcript levels in K562 cells were shown to be 2-fold higher than in HL60 cells.

Efficient Targeting of WT1 Protein Expression in Both K562 and HL60 Cells by ASWT1 exon 5 and ASWT13'UTR
A reduction in WT1 isoform levels was shown at 24 hours following WT1-directed ASO treatment of both K562 and HL60 cells lines but not at 4 hours (data not shown). Although adequate separation of the WT1 + and – exon 5 isoforms was achieved using NuPAGE gels, the splitting of the WT1 signal reduced the sensitivity of the Western blots, making visual interpretation of the protein bands difficult. Therefore, for clarity, typical Western blot images are shown in Fig. 7 (insets) along with the line graphs generated from these bands using the ImageQuant software. ASWT1 scram was used as control following confirmation that both WT1 isoform levels and exon 5 ratios were equivalent following PBS, ASWT1mc-raf, and ASWT1 scram treatment at 10 µmol/L concentration (data not shown). WT1 exon 5–containing isoforms were overrepresented in both leukemia cell lines, suggesting a correlation between alternatively spliced mRNA transcript levels and WT1 protein isoform expression. In these experiments, there was a slight reduction of nonspecific band signal in antisense-treated samples compared with control samples but no dose response effect between 5 and 10 µmol/L antisense concentrations. Concentration-dependent reduction of total WT1 isoform levels with maintenance of isoform ratios was seen in both K562 and HL60 cell lines following ASWT13'UTR treatment. Differential targeting of WT1 + exon 5 isoforms by ASWT1 exon 5 was also seen in both cell lines as shown by the concentration-dependent reduction in the WT1 exon 5 isoform ratios (see Table 1). However, targeting of exon 5 isoforms was associated with reduced overall levels of WT1 protein as was seen at the mRNA level at 24 hours following treatment (cf. Fig. 6D). A recent report showed that WT1 exon 5 isoforms mediate specifically the transactivation of an antisense WT1 promoter in the first intron of the WT1 gene probably by interaction with an accessory protein (41). The regulatory antisense WT1 mRNA product has been shown previously to positively regulate WT1 protein levels (42), providing a potential mechanism by which selective targeting of WT1 exon 5–containing variants results in the subsequent down-regulation of WT1 protein expression.

View larger version (44K): [in this window] [in a new window] [Download PPT slide]   Figure 7. Efficient targeting of WT1 protein expression in both K562 and HL60 cells by ASWT13'UTR and ASWT1 exon 5. Western immunoblot analysis of WT1 protein 24 hours following ASWT13'UTR (5 and 10 µmol/L) and ASWT1 scram (10 µmol/L; Con) treatment of K562 (A) and HL60 (C) cells and ASWT1 exon 5 (5 and 10 µmol/L) and ASWT1 scram (10 µmol/L) treatment of K562 (B) and HL60 (D) cells. Typical images of WT1 protein bands (inset) are shown along with the superimposed ImageQuant line graphs generated from these bands using the ImageQuant software.

View larger version (13K): [in this window] [in a new window] [Download PPT slide]   Figure 8. Inhibition of cell survival by ASWT13'UTR and ASWT1 exon 5 in K562 and HL60 cells. Soft agar clonogenic assays of cell survival following ASO treatment of K562 (A) and HL60 (B) cells. Control ASO (K562 cells: ASmc-raf and HL60 cells: ASWT1 scram; ); ASWT13'UTR (); ASWT1 exon 5 (•). Points, mean of quadruplicate determinations; bars, SD. C, confirmation of efficient targeting of WT1 transcripts at 24 hours posttreatment in the residual HL60 cells from the experiment in B using real-time PCR analysis.

cDNA Microarray Screen of WT1 Target Gene Expression in HL60 and K562 Cells Treated with ASWT13'UTR and ASWT1 exon 5
cDNA microarray analysis was used to examine which, if any, of the known WT1 target genes were differentially regulated by ASWT13'UTR and ASWT1 exon 5 in the leukemic cell lines. The differential cytotoxicity profiles of ASWT13'UTR and ASWT1 exon 5 in HL60 and K562 cells also provided an opportunity to investigate whether any of these genes were regulated in a manner that correlated with disruption of isoform ratios and also with toxicity.

Of 42 known putative WT1 target genes, 36 (listed in Table 2) were represented in our in-house gene expression array of 5,603 cDNA clones. Eleven genes were represented by more than one clone. Gene expression profiles compared first-strand cDNA populations from cells treated with 10 µmol/L ASWT13'UTR or ASWT1 exon 5 and cells treated with 10 µmol/L ASWT1 scram as the reference. In K562 cells, no alteration in the expression of any of the putative WT1 target genes was detected following treatment with either active ASO (data not shown). These genes seem, therefore, not to contribute to the cytotoxic activity of either ASO in this cell line. In HL60 cells, which were resistant to ASWT13'UTR treatment, again no alteration in the expression of any of these genes by ASWT13'UTR was noted. However, treatment with ASWT1 exon 5, which decreased cell survival by 60%, resulted in the differential expression of MYC, ornithine decarboxylase (ODC1), and THBS1 (Table 2). The proto-oncogene MYC plays a key role in cell proliferation, differentiation, and apoptosis and is reported to be amplified in HL60 cells (44), whereas THBS1 exhibits numerous biological activities including effects on cell adhesion, migration, proliferation, and angiogenesis (45).

View larger version (27K): [in this window] [in a new window] [Download PPT slide]   Figure 9. Confirmation of disruption of gene expression following WT1 ASO treatment of HL60 and K562 cells using real-time (Taqman) RT-PCR analysis. WT1 (A), MYC (B), and THBS1 (C) transcript levels in HL60 cells 24 and 48 hours following treatment with 10 µmol/L ASWT1 scram (gray columns), ASWT13'UTR (open columns), and ASWT1 exon 5 (black columns), expressed as a percentage of ASWT1 scram levels. D, dose-dependent increase of THBS1 in HL60 cells 24 hours following ASWT1 exon 5 treatment (closed columns) and confirmation of lack of disruption of THBS1 following ASWT13'UTR (open columns). TM4SF9 (E) and GPC5 (F) transcript levels in K562 cells following WT1 ASO treatment as in HL60 cells above.

Identification of Novel Putative WT1-Responsive Genes
In addition to the known WT1 regulated genes, we examined the expression profiles of the remaining genes in the expression array. A further 10 genes were found to be modestly up-regulated (1.6- to 2.6-fold) specifically in HL60 cells following ASWT1 exon 5 treatment, mirroring the pattern of differential regulation displayed by THBS1 (Table 3A). Similarly, a further five genes showed a pattern of down-regulation concordant with MYC and ODC1. Our gene profile database was therefore searched for ASWT1 exon 5–induced disruption of gene expression specifically in K562 cells. Eight genes were found to be modestly up-regulated in this category, whereas only two were found to be down-regulated (Table 3B). Significantly, these represent a different set of genes to those disrupted in HL60 cells by ASWT1 exon 5. A further search for genes, common to both cell lines and disrupted by ASWT1 exon 5, revealed seven that were commonly down-regulated but none that were differentially up-regulated convincingly in both cell lines (Table 3C).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
Does Drug-Induced Induction of WT1 Contribute to Chemoresistance Mechanisms?
As mentioned previously, expression of WT1 is dynamically regulated following induction of differentiation in a variety of cell lines (27–30). However, to our knowledge, this is the first report of induction of WT1 expression following cytotoxic drug treatment of cell lines displaying a resistant phenotype, a phenomenon absent in the sensitive cell lines studied. These findings support the notion that increased WT1 expression may contribute to chemoresistance mechanisms, allowing cells to survive following drug therapy. However, at this stage, we can only speculate on the mechanism(s) of induction and the downstream effects of induced WT1 expression.

Little information exists regarding the regulatory factors controlling WT1 expression. Several ubiquitous and tissue-specific transcription factors have been shown to activate or repress the WT1 promoter in transient transfection assays including Sp1, WT1, GATA-1, Pax-2, Pax-8, and nuclear factor-B; (NF-B; refs. 52–56). Regulation by NF-B may be relevant here, because treatment of cells with DNA-damaging agents or various cytokines and mitogens can result in its translocation from the cytoplasm to the nucleus and the modulation of the appropriate target genes. Ectopic expression of nuclear NF-B has been shown to increase the transcription of endogenous WT1, indicating that members of the nuclear NF-B/Rel family may be involved in a regulatory cascade leading to WT1 activation (57). Further studies are clearly required to determine the mechanisms responsible for, and the significance of, the differential regulation of WT1 expression in sensitive and resistant cell lines following cytotoxic drug treatment.

Because G₁ arrest associated with p53-independent regulation of endogenous p21^CIP1 by WT1 has been shown, it is tempting to hypothesize that induction of WT1 contributes to chemoresistance by inducing cell cycle arrest, thereby allowing DNA repair and avoidance of replication using a damaged template. However, in previous studies using both paired cell lines, G₁ arrest was not observed following cisplatin treatment (38, 58). Common to all four lines was a slowdown in S-phase transit, whereas transient G₂ arrest or G₂-M block at later time points was seen dependent on cell line and dose of cisplatin. In the present studies, no temporal correlation between induction of WT1 in CH1-R or GCT27-R cells and the previously reported cell cycle effects was apparent. Lack of G₁ arrest is not due to a p53-null phenotype because all four cell lines show partial G₁-S arrest, associated with induced p53 and p21, following -irradiation (data not shown).

Drug-Induced Disruption of WT1 exon 5 Splicing May Constitute a Cell Stress Response
Alternative pre-mRNA splicing is a fundamental mechanism of gene expression that can be regulated dependent on sex, development stage, or tissue, and in response to extracellular stimuli such as growth factors, hormones, and cytokines (59–61). Recent studies have identified a composite exonic splice control unit, which combines an exon recognition element with splice silencer elements. The splice control unit seems to govern alternative splicing in a cell type–specific manner and in response to activation of protein kinase C or Ras signaling pathways (61). The discovery of signal-responsive splice elements provides a link between extracellular signals and regulation of exon variant transcription.

In these studies, we have provided the first evidence that the alternative splicing of exon 5 of WT1 mRNA is subject to dynamic regulation in response to drug treatment, whereas KTS splicing is not disrupted. Much interest has focused recently on the activation of stress-activated protein kinase/c-Jun-NH₂-kinase, and p38 mitogen-activated protein kinase cascades following cisplatin treatment (62, 63). Whether WT1 possesses signal-responsive elements that respond to activation of these or other protein kinases is unknown. Nevertheless, in the cell lines studied, the disruption of WT1 exon 5 splicing seemed to be an early stress response to a toxic stimulus independent of the tissue of origin of the cell lines, the relative chemosensitivity, or the drug used. We propose that increased exon 5 skipping is the most likely mechanism.

Does Disruption of WT1 exon 5 Ratios Provide a Proapoptotic or Cell Survival Signal?
In these studies, we have attempted to mimic the disruption of WT1 exon 5 splicing seen following drug treatment using an ASO targeted specifically to exon 5–containing WT1 transcripts. ASWT1 exon 5 induced both disruption of exon 5 splicing and down-regulation of WT1 levels. The latter effect may be due to down-regulation of the exon 5 transactivated regulatory antisense WT1 mRNA product, which has been shown previously to positively regulate WT1 protein levels (42). Cell survival studies showed loss of cell viability in both K562 and HL60 cell lines by ASWT1 exon 5, despite HL60 cells being resistant to the effects of a balanced down-regulation of WT1 isoforms. These data support the view that the drug-induced disruption of exon 5 splicing constitutes a proapoptotic signal. It may be that the up-regulation of WT1expression shown here in resistant cell lines following drug treatment temporarily overrides this proapoptotic signal, providing the cells with a greater opportunity for damage repair and escape from apoptosis. Interference with proapoptotic signaling following cell damage can be expected to provide a survival advantage, although additional resistance mechanisms are likely to contribute to the resistant phenotype of these cell lines.

Differential Regulation of Putative WT1 Target Genes following Disruption of WT1 exon 5 Ratios
Some 42 putative WT1 target genes have been identified mainly as a result of transient transfection studies with reporter constructs (for a recent list, see ref. 2). However, with a few exceptions such as p21^CIP1 and EGFR (13, 23), little convincing evidence of WT1-directed regulation of the endogenous expression of these putative target genes has been forthcoming. Cell context specificity further complicates the search for bona fide WT1 target genes, because interactions between WT1 and coregulatory proteins may influence both the transcriptional regulatory properties of WT1 and target recognition. Consequently, there are few clues as to how the various isoforms of WT1 interact intracellularly to exert their biological effects or indeed how disruption of WT1 isoform ratios might induce altered biological function.

In this study, we have combined the use of antisense and cDNA microarray technologies in an attempt to identify which WT1-responsive genes are differentially regulated following disruption of WT1 isoform ratios in a manner correlating with inhibition of cell survival. As found previously, few changes in the expression levels of the identified WT1 putative target genes were noted and none were seen in K562 cells. Differential regulation of MYC and its reported downstream target ODC1 was observed at 24 hours following ASWT1 exon 5 in HL60 cells, but subsequent and equivalent down-regulation of MYC by ASWT13'UTR at 48 hours revealed a lack of correlation between MYC expression and cytotoxicity.

Unlike MYC expression, the 5-fold up-regulation of THBS1 seemed to correlate with the cytotoxic activity of ASWT1 exon 5 in HL60 cells, although the mechanism of such activity in vitro is not obvious. Thrombospondin is a 450-kDa extracellular matrix–bound trimeric glycoprotein that is expressed and secreted by platelets and a wide variety of cell types. It has been suggested to play an important albeit complex role in controlling cancer cell growth and metastasis in vivo and variously implicated in cancer cell adhesion, migration, invasion, proliferation, and apoptosis-dependent inhibition of angiogenesis (reviewed in ref. 45). Endogenous repression of THSB1 by WT1 has been shown previously in response to overexpression of c-Jun (64). Repression involved a factor secreted by c-Jun-activated cells, which triggered a signal transduction pathway culminating in the binding of WT1 to the THBS1 promoter. By contrast, our present studies suggest a clear correlation between disruption of WT1 exon 5 ratios and endogenous regulation of THBS1, a novel observation. Whether this activity is direct or involves a coregulatory protein remains to be determined.

Identification of Previously Unreported WT1-Responsive Genes
Of the remaining genes on our expression array, 2% were significantly altered (directly or indirectly) by WT1-directed ASO treatment, generally in a cell type–specific and WT1 antisense target-specific manner. Our initial aim was to search for evidence of differential regulation of downstream genes following disruption of WT1 exon 5 ratios; clearly, this occurs in both HL60 and K562 cells. However, of the 31 genes affected by ASWT1 exon 5 treatment and listed in Table 3, only 7 were common to both cell lines, highlighting yet again the cell type specificity of WT1 activity. Overall, the changes in expression were modest at 24 hours. It is possible that more extensive alterations in expression may be evident at later time points, or differential regulation may be lost, as in the case of MYC.

These studies fall short of confirming which, if any, of these genes contribute to the cytotoxic activity of ASWT1 exon 5 but can be used as the basis for further studies. Worthy of comment is the up-regulation of interleukin-1ß (IL-1ß) in HL60 cells. IL-1ß, like tumor necrosis factor, initiates the activation of signal cascades by the recruitment of adapter proteins to its receptor. In many cancer cell lines, both IL-1ß and tumor necrosis factor induce apoptosis, whereas untransformed cell types are not sensitive unless mRNA translation or protein synthesis is blocked (65).

The differential regulation of the genes listed in Table 4A by ASWT13'UTR, specifically in K562 cells, shows that a different set of genes respond to the balanced down-regulation of WT1 isoforms as compared with when down-regulation of WT1 is accompanied by disruption of exon 5 ratios (listed in Table 3B). It is possible that networks of genes and gene cascades may be controlled not only by the regulation of the overall levels of WT1 but also by regulation of exon 5 ratios. Interaction with the various stress/apoptotic/differentiation/growth factor signaling pathways is a possibility that requires further investigation.

Finally, there are some genes that respond to down-regulation of WT1 regardless of whether isoform ratios are disrupted. One of these, GPC5, may represent a bona fide WT1 target gene as predicted by the presence of two WT1 consensus binding sequences in its promoter region. GPC5 constitutes one of the membrane-bound heparan sulfate proteoglycan family of genes, members of which have been proposed to function in cellular growth control and morphogenesis (48). Overexpressed in K562 cells, GPC5 was of particular interest because it is expressed in cells of mesenchymal origin during development and shows a similar pattern of expression to WT1 in the developing kidney and gonads (49, 66, 67). In addition, GPC5 has been suggested to be a candidate gene for at least some of the phenotypic features of 13q syndrome, a developmental disorder with a pattern of defects that shows overlap with both WAGR and DDS and also with WT1 knockout mice (49, 68, 69).

In summary, the results of this study support a role for WT1 in the maintenance of viability and proliferative capacity in cancer cells and as a mediator of survival signals following cytotoxic drug treatment. Downstream signaling seems to involve the orchestrated regulation of WT1 exon 5 splicing and total WT1 expression. Using ASOs directed to both exon 5 and the 3' UTR of WT1, we have shown cell type–specific and antisense target-specific regulation of genes, some of which may prove to be novel WT1 target genes. Disruption of WT1 exon 5 ratios by ASWT1 exon 5 was shown to reduce the cell survival of HL60 cells that are resistant to other WT1-targeted ASOs. ASWT1 exon 5 may therefore have a broader therapeutic potential than previously described WT1 ASOs. In preliminary studies, we have shown significant antitumor activity against K562 cells grown in vivo using both WT1-directed ASOs (70). We are extending these studies to explore the therapeutic potential of these compounds in other WT1-expressing tumors such as prostate cancer, breast cancer, and childhood embryonal cancers.

	Acknowledgments

 
The authors thank Dr. Dan Williamson (Section of Molecular Carcinogenesis, ICR, Sutton, UK) for the design of the GPC5 primers and probe and Kathryn R. Taylor for excellent technical assistance.

	Footnotes

 
Grant support: Children's Cancer Unit Fund, Royal Marsden Hospital NHS Trust, Sutton, UK (J. Renshaw, R.D. Williams) and Cancer Research UK (R.M. Orr, M.I. Walton, R. te Poele, P. Workman, and K. Pritchard-Jones).

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.

⁴ http://www.hgmp.mrc.ac.uk/rdwillia/unigene.html.

Received 3/10/04; revised 8/ 6/04; accepted 9/15/04.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Reddy JC, Licht JD. The WT1 Wilms' tumor suppressor gene: how much do we really know? Biochim Biophys Acta 1996;1287:1–28.[Medline]
2. Scharnhorst V, van der Eb AJ, Jochemsen AG. WT1 proteins: functions in growth and differentiation. Gene 2001;273:141–61.[CrossRef][Medline]
3. Bruening W, Pelletier J. A non-AUG translational initiation event generates novel WT1 isoforms. J Biol Chem 1996;271:8646–54.[Abstract/Free Full Text]
4. Scharnhorst V, Dekker P, van der Eb AJ, Jochemsen AG. Internal translation initiation generates novel WT1 protein isoforms with distinct biological properties. J Biol Chem 1999;274:23456–62.[Abstract/Free Full Text]
5. Haber DA, Sohn RL, Buckler AJ, Pelletier J, Call KM, Housman DE. Alternative splicing and genomic structure of the Wilms tumor gene WT1. Proc Natl Acad Sci U S A 1991;88:9618–22.[Abstract/Free Full Text]
6. Kent J, Coriat AM, Sharpe PT, Hastie ND, van Heyningen V. The evolution of WT1 sequence and expression pattern in the vertebrates. Oncogene 1995;11:1781–92.[Medline]
7. Larsson SH, Charlieu JP, Miyagawa K, et al. Subnuclear localization of WT1 in splicing or transcription factor domains is regulated by alternative splicing. Cell 1995;81:391–401.[CrossRef][Medline]
8. Hammes A, Guo JK, Lutsch G, et al. Two splice variants of the Wilms' tumor 1 gene have distinct functions during sex determination and nephron formation. Cell 2001;106:319–29.[CrossRef][Medline]
9. Hastie ND. Life, sex, and WT1 isoforms—three amino acids can make all the difference. Cell 2001;106:391–4.[CrossRef][Medline]
10. Renshaw J, King-Underwood L, Pritchard-Jones K. Differential splicing of exon 5 of the Wilms tumor (WTI) gene. Genes Chromosomes Cancer 1997;19:256–66.[CrossRef][Medline]
11. Iben S, Royer-Pokora B. Analysis of native WT1 protein from frozen human kidney and Wilms' tumors. Oncogene 1999;18:2533–6.[CrossRef][Medline]
12. Natoli TA, McDonald A, Alberta JA, Taglienti ME, Housman DE, Kreidberg JA. A mammal-specific exon of WT1 is not required for development or fertility. Mol Cell Biol 2002;22:4433–8.[Abstract/Free Full Text]
13. Englert C, Maheswaran S, Garvin AJ, Kreidberg J, Haber DA. Induction of p21 by the Wilms' tumor suppressor gene WT1. Cancer Res 1997;57:1429–34.[Abstract/Free Full Text]
14. Hewitt SM, Saunders GF. Differentially spliced exon 5 of the Wilms' tumor gene WT1 modifies gene function. Anticancer Res 1996;16:621–6.[Medline]
15. Kudoh T, Ishidate T, Moriyama M, Toyoshima K, Akiyama T. G₁ phase arrest induced by Wilms tumor protein WT1 is abrogated by cyclin/CDK complexes. Proc Natl Acad Sci U S A 1995;92:4517–21.[Abstract/Free Full Text]
16. Mayo MW, Wang CY, Drouin SS, et al. WT1 modulates apoptosis by transcriptionally upregulating the Bcl-2 proto-oncogene. EMBO J 1999;18:3990–4003.[CrossRef][Medline]
17. Baudry D, Faussillon M, Cabanis MO, et al. Changes in WT1 splicing are associated with a specific gene expression profile in Wilms' tumor. Oncogene 2002;21:5566–73.[CrossRef][Medline]
18. Bergmann L, Miething C, Maurer U, et al. High levels of Wilms' tumor gene (wt1) mRNA in acute myeloid leukemias are associated with a worse long-term outcome. Blood 1997;90:1217–25.[Abstract/Free Full Text]
19. Inoue K, Sugiyama H, Ogawa H, et al. WT1 as a new prognostic factor and a new marker for the detection of minimal residual disease in acute leukemia. Blood 1994;84:3071–9.[Abstract/Free Full Text]
20. King-Underwood L, Renshaw J, Pritchard-Jones K. Mutations in the Wilms' tumor gene WT1 in leukemias. Blood 1996;87:2171–9.[Abstract/Free Full Text]
21. Trka J, Kalinova M, Hrusak O, et al. Real-time quantitative PCR detection of WT1 gene expression in children with AML: prognostic significance, correlation with disease status and residual disease detection by flow cytometry. Leukemia 2002;16:1381–9.[CrossRef][Medline]
22. Miyoshi Y, Ando A, Egawa C, et al. High expression of Wilms' tumor suppressor gene predicts poor prognosis in breast cancer patients. Clin Cancer Res 2002;8:1167–71.[Abstract/Free Full Text]
23. Carrington D, Algar E. Overexpression of murine WT1 +/+ and –/– isoforms has no effect on chemoresistance but delays differentiation in the K562 leukemia cell line. Leuk Res 2000;24:927–36.[CrossRef]