The hedgehog pathway in cerebellar cancer cells was modulated with siRNA specifically targeted to the shh and gli-I genes. Silencing of the two genes in a medullablastoma cell line transfected with the siRNAs caused significant reduction of mRNA specific for the targeted shh and gli-I genes and a loss of protein expression. The disclosed methods and compositions may be useful for treatment of a range of primitive neuroectodermal tumors (PNET) by shutting down or modulating the expression of gene products associated with the hedgehog pathway.

Description of the Invention

SUMMARY

The invention relates to the discovery that siRNA-mediated silencing of the shh gene in medullobastoma cells appeared to alter the phenotype, growth rate and growth characteristics of these cells. Accordingly, the invention features compositions and methods of modulating the phenotype of a cerebellar cancer cell by modulating expression of gene encoding a gene product involved in the hedgehog pathway.

DETAILED DESCRIPTION

Multiple siRNAs complementary to shh, gli-1 were evaluated to determine optimal methods for measuring siRNA-induced gene suppression. Multiple siRNAs complementary to the .beta.-actin gene were used as a control. Two siRNAs were prepared for each target gene. These siRNA were transfected into a medulloblastoma cell line using Ambion's Silencer.TM. siRNA transfection kit. The effect of in vitro RNAi treatment of medulloblastoma cells on protein expression was measured using flow cytometry and western blot analysis. Alteration of mRNA levels following treatment was assessed using northern blot analysis. The results of these studies show that shh and gli-1 siRNA specifically targeted the mRNA for both shh and gli-1 genes which resulted in a significant decrease (greater than 90% by 96 hours following transfection) in the levels of targeted mRNAs and 85% protein expression as measured by western blot analysis. The loss of protein expression as measured by flow cytometry also showed that there was a significant decrease in the level of protein expression and a reduction in the number of cells expressing the proteins over the 96 hour period. The ability to silence the shh gene using siRNA appeared to alter the phenotype, growth rate and growth characteristics of the tumor cells in vitro. However, despite the silencing of the gli-1 gene, there was no apparent change in cell proliferation, growth characteristics or phenotype of the siRNA-treated tumor cells.

Materials and Methods

PNET cell cultures: Cells were routinely maintained in IMDM supplemented with 10% FBS and 0.6% L-glutamine. Cultures were incubated at 37.degree. C. in T75 cm.sup.2 plastic culture flasks in a humidified atmosphere of 5% CO.sub.2 in air.

Phenotypic analysis of PNET cell cultures: PNET cell cultures were subjected to both flow cytometric analysis and indirect fluorescent antibody assays for the expression of Heat stable antigen (HSA), Vimentin, Synaptophysin, neurofibrillary protein-70 (NFP-70), neurofibrillary protein-250 (NFP-250), Nestin, glutamine synthetase, neuron-specific enolase (NSE) and neuroectodermal antigen (UJ13A), glial fibrillary acidic protein (GFAP) and for S-100. All the antibodies listed above were purchased from Chemcon International (Temecula, Calif.). The antibodies detecting SHh and Glilantigens were purchased from ATCC (Rockville, Md.) and abcam (Cambridge, Mass.) respectively.

Flow cytometry: Cells were labeled as described below (Indirect fluorescent antibody [IFA] assays), placed in sheath fluid and analyzed on a FACSCalibur four-color flow cytometer (Becton Dickinson Immunocytometry Systems, CA) Data analysis was performed by using CellQuest Pro Data Analysing

Software (Becton Dickinson Immunocytometry Systems, CA). The flow cytometer was calibrated prior to each run. Compensation was set up for FITC using single-stained cell populations. All cell analysis was carried out within a low orthogonal light scatter and forward light scatter windows at a rate of more than 2.times.10.sup.3 cells sec-.sup.1

Indirect fluorescent antibody (IFA) assays: All IFA experiments for the detection of both intracellular and extracellular antigens were conducted using CALTAG labs (Burlingame, Calif.) permeabilization kit. The procedure was as directed by the manufactures instructions. Briefly, for each cell sample to be analyzed an appropriate concentration of primary antibody was added to 1.times.10.sup.6 cells. The cells were vortexed and incubated for 15 minutes at room temperature. Following this step, 100 ul of the fixing reagent was added and the cells incubated for a further 15 minutes at room temperature. Following incubation, the cells were washed once in 3 ml of phosphate buffered saline (PBS) supplemented with 5% FBS, centrifuged and the wash fluid removed. To the cell pellet, 100 ul of the permeabilization reagent and 10 ul of the FITC-labeled secondary antibody (anti mouse IgG) was added. The cells were vortexed and incubated at room temperature for 20 minutes. Following incubation the cells were washed as previously described above and either examined for fluorescence using UV microscopy or placed in sheath fluid for flow cytometric analysis.

Synthesis of siRNA for gene silencing: In the development of systems to measure gene silencing in mammalian cells, it appears that the most potent siRNAs are those that contain a 19 nucleotide complementary region between both strands (sense and antisense) plus a 2 nucleotide overhang at the 3' end.

The selection of siRNA target sites on the genes of interest started at the AUG start codon and the transcript scanned downstream for AA di-nucleotide sequences. All the AA di-nucleotide and the 3' adjacent 19 nucleotides were recorded. All the potential target sites were then compared to an appropriate genome database, such as BLAST for the mouse and human, for the elimination of those sequences that have significant homology to other coding sequences. The resulting target sequences were sent to Ambion, Inc. (Austin, Tex.) and complementary pairs of siRNA oligonucleotides with dTdT or UU 3' overhangs were synthesized. SiRNA's were synthesized for the genes encoding SHh and Gli1 and also for the reporter gene encoding beta-actin.

mRNA Isolation: PNET cells used for both total RNA and mRNA were grown as previously described. The isolation of and subsequent purification of PolyA mRNA was carried out using Qiagen's Oligotex Direct mRNA Kit (QIAGEN, Calif.) according to manufacturer's instructions.

Plasmid preparation: Cultures of E. coli containing plasmids (pT7pT3) with either Shh or Gli1 inserts were grown overnight at 37.degree. C. in LB broth (with 50 ug/ml of ampicillin) in an orbital shaker. Purified plasmid preparations were prepared using Qiagen's QIAprep Spin Miniprep Kit (Qiagen, Calif.) according to manufacturer's instructions. Plasmid linearization was achieved with NotI (Promega Corporation).

Probe synthesis: Linearized plasmid DNA from the above digests was used to generate T7 RNA polymerase probes by in vitro transcription using Ambion's Strip-EZ.TM. RNA Kit (Ambion, Austin, Tex.) and following the procedure recommended by the manufacturer (Ambion, Austin, Tex.). Briefly, the reaction was set up in a 1.5 ml microcentrifuge tube at room temperature. The following components were added in order; 12 .mu.l of nuclease-free water, 12 .mu.l of template DNA from restriction digests, 4 .mu.l of 10.times. Transcription Buffer, 2 .mu.l of ATP Solution, 2 .mu.l of Modified CTP Solution, 2 .mu.l of GTP Solution, 2 .mu.l of UTP Solution and 4 .mu.l of T7 Enzyme Mix. Reactions incubated for 90 minutes at 40.degree. C. After incubation, the DNA template was removed by adding 1 .mu.l of DNase 1 and placing the reaction at 37.degree. C. for 15 minutes. The reactions were stopped with 1 .mu.l of 0.5M EDTA (Gibco BRL.RTM.) incubated at 75.degree. C. for five minutes. Probes were then labeled using Ambion's BrightStar.TM. Psoralen-Biotin Kit. 30 .mu.l of each probe was denatured at 100.degree. C. for 10 minutes. The probes were then quick chilled in an ethanol/ice bath and placed in a 96 well plate that sat on an ice bath. 3 .mu.l of Psoralen-Biotin was mixed with each probe and irradiated for 45 minutes under an ultraviolet 365 nm light. Each probe was diluted in 70 .mu.l of TE Buffer. Non-crosslinked psoralen-biotin was removed by butanol extraction. One extraction/probe was done using 200 .mu.of Water Saturated n-butanol followed by centrifugation and removal of the butanol layer. All probes were then stored at -70.degree. C.

Northern blot analysis: The size and abundance of mRNA was determined by northern blot analysis. All procedures were carried out using Ambion's NorthernMax.TM.-Gly Kit (Ambion, Austin Tex.). Detection of signal was determined using Ambion's BrightStar.TM. BioDectect.TM. Kit. Exposure was done for 4-6 hours on Hyperfilm.TM. ECL (Amersham Biosciences).

Characterization of the siRNA induced gene silencing of target genes: Two target sequences per gene from the 5', 3' ends and medial regions were selected based upon the predicted sequence as reported in `Ensembl Human Genome Browser` (GeneView). For transfection, the different populations of PNET tumor cells were grown to between 40-70% confluency in T75 cm.sup.2 tissue culture flasks in normal IMDM growth media. The individual siRNAs, at varying concentrations, including transfection reagent, either siPORT.TM. Amine (a polyamine) or siPORT.TM. Lipid (a mixture of cationic and neutral lipids), and Opti-MEM were mixed and incubated together at room temperature for 15-20 minutes. Following incubation, the siRNA mixture was added to the cell cultures and incubated for up to 96 hours. At varying time intervals, the tumor cells were harvested for analysis of both specific mRNA (northern blot) and protein (western blot and flow cytometry). Targeted cells were also examined for alterations in phenotype, growth characteristics and for in vivo tumorogenicity.

Results

Phenotypic analysis of PNET cell cultures: To establish the primitive phenotype and neural origin of the PNET cell line D283 and cell line HM75, prior to gene targeting, indirect fluorescent antibody tagging was carried out using antibodies as described in table 1 (see Original Patent). To label intracellular antigens for both FACS analysis and IFA, the cells were subjected to a fixation and permeabilization procedure as previously described. Once labeled, the cells were analyzed on either a FACSCalibur four-color flow Cytometer or a UV microscope. The data summarized in Table I and FIGS. 1 & 2 (see Original Patent), show that both cell lines have similar phenotype, expressing most of the neural stem cell markers that were examined but do not express either of the neurofilament proteins (NFP-70 NFP-250) or glutamine synthetase (GS). For IFA, antibody-labeled cells were subjected to cyto-centrifugation (200 rpm for 5 minutes), air dried and placed under PBS-buffered glycerol and a coverslip. Cells were then examined using uv light microscopy and the degree of fluorescence determined and recorded as follows: 80%-100%=(4+); 50%-80%=(3+); 20%-50%=(2+); 5%-20%=(1+) and 0%-5%=(-). Also shown is the normal protein expression of both SHh and Gli1 in both the PNET tumor lines. FIGS. 2a and 2b (see Original Patent) give examples of the FACS analysis obtained in such experiments. The data suggest good correlation between the results obtained by indirect fluorescence and the Flow data. The representative data in FIGS. 2a and 2b (see Original Patent) shows that under normal cultural conditions both cell lines express significant amounts of both the SHh and Gli1 proteins. Both cell lines nearly 100% of the cells express these proteins. Interestingly, the tumor line, D283, has a small population of cells that exhibit significantly higher levels of SHh, the significance of which still needs to be determined. Either by flow analysis or by indirect fluorescent labeling, the phenotype of the PNET cell lines can be expressed as shown in Table 2 (see Original Patent).

Molecular Studies: From the antibody studies mentioned above, it appears that both the PNET cell line D283 and HM75 would be appropriate to use in the in vitro gene silencing studies. Both total RNA and mRNA have been isolated from the respective tumor line and stored at -80.degree. C. However, only mRNA was used in the northern blot analysis.

Synthesis of the probes for Northern blot analysis: The development of the probes used in this study is summarized in FIG. 3 (a,b,c) (see Original Patent). Briefly, A multi-purpose cloning vector (with an ampicillin resistant marker) also containing opposable T3 and T7 promotors that flanked a multiple cloning site were used to clone portions of the human shh and gli1 genes. The genes of interest were cloned into the vector at a NotI and EcoR1 cloning site. Competent bacteria containing the plasmid were grown as colonies on LB agar (containing 50 ug/ml of ampicillin). Individual colonies of bacteria were picked and placed in 5 ml of LB broth (supplemented with 50 ug/ml of ampicillin) and incubated overnight at 37.degree. C. in an orbital shaker. From overnight cultures, plasmid preparations were carried out using Gibco BRL "CONCERT" mini-plasmid-prep system. 1% agarose gels were run to verify the purity of the plasmid preps (FIG. 3a (see Original Patent)). Restriction enzyme analysis (double digests using Not 1 and EcoR1) was also performed to verify the insert size (data not shown). .beta.actin was the reporter gene that was used as control for the gene silencing experiments.

Northern hybridization was the method used to assay for levels of the target mRNA. RNA probes were chosen over DNA probes because they offer 10-fold better sensitivity and were synthesized by random priming using Strip-EZ RNA Kit. Purified plasmid preps were linearized downstream of the insert with EcoR1. This allowed us to transcribe the antisense RNA probe using the T3 RNA polymerase. Following the removal of the DNA template (linearized plasmid), the RNA probes were purified, concentrated by precipitation and stored at -75.degree. C. (FIG. 3b, see Original Patent).

Because the RNA was synthesized using the Strip-EZ RNA Kit, the probes were labeled post synthesis with Psoralen-Biotin. Furthermore, the use of modified CTP in the transcription and synthesis of the antisense RNA probe allows us to degrade and strip the hybridized probe from the northern blots for re-use is subsequent experiments. FIG. 3c (see Original Patent) is an example of a test blot at varying concentrations showing both the shh and gli probe activities.

Determination of the gene sequences for the synthesis of the 19mer antisense RNAi's: SiRNAs were constructed for the genes encoding shh and gli1 and .beta. actin. The coding sequences and the transcript sequences were taken from the data given in the Ensembl Human and Murine Gene Bank (The Wellcome Trust, Sanger Institute). In this study, the identification of siRNA target sites were determined as stated in the experimental methods section. Two SiRNA sequences per target were designed for each of the genes under study (see FIGS. 4a,b and c (see Original Patent)), an SiRNA containing a `scrambled` sequence was synthesized to serve as a negative control.

Characterization of the siRNA induced gene silencing of targeted genes: To test the protocol and the synthesized SiRNAs, the Shh and gli1-specific SiRNAs were transfected using siPort.TM. Lipid. The mRNA was obtained prior to the SiRNA treatment and at 12, 24, 48 and 96 hours following treatment. Northern blot analysis of cells treated with the SiRNAs (see FIG. 5 (see Original Patent)) indicated that the level of mRNA specific for either of the targeted genes (Shh and gli1) was significantly reduced within 24 hours and under experimental conditions used in this study was not detectable at 96 hours following SiRNA treatment. Treatment of the cells with either of the shh or gli1 SiRNAs did not effect the level of .beta. actin-specific mRNA (FIG. 5, see Original Patent). To assess the effect on protein expression treated cells were examined by IFA (Table 3, see Original Patent), western blot analysis and flow cytometry. FIGS. 6a & 6b (see Original Patent) show the loss of protein expression in the PNET cells following the shh-SiRNA treatment. As with the mRNA levels protein expression was reduced significantly and by 96 hours no detectable protein was observed. The western blot analysis (FIG. 7 (see Original Patent)) of the treated cells also shows significant reduction of the protein expression such that in the shh-treated cells protein was barely detectable by 96 hours post treatment. In the gli1-treated cells a reduction in protein expression was observed but there was significantly more gli1 expression by 96 hours when compared to the shh-treated cells. Cells that were treated with the siRNAs were re-plated in 12-well plates and incubated at 37.degree. C. and observed for in vitro growth characteristics. The only significant change in the growth of the cells is shown in FIG. 8 (see Original Patent). Under normal cell growth these cells were predominantly non-adherent proliferating in clusters in the supernatant. Following siRNA-treatment the cells were predominantly adherent and had a significantly slower growth rate. This change was more visible in those cells that were treated with the shh siRNA than those treated with gli1-siRNA.
 

Claim 1 of 5 Claims

1. A method for inhibiting expression of SHh in primitive neuroectodermal (PNET) tumor cells, comprising transfecting the tumor cells with an effective amount of a siRNA primer pair consisting of the antisense shh gene sequences SEQ ID NO:6 and SEQ ID NO:8 wherein the transfected tumor cells change from non-adherent proliferating cells to adherent proliferating cells after transfecting in vitro thereby altering growth, phenotype and growth characteristics of the adherent proliferating cells.

____________________________________________
If you want to learn more about this patent, please go directly to the U.S. Patent and Trademark Office Web site to access the full patent.