WO2017070113A1 - Pharmaceutical composition comprising nanog shrna, and method of using nanog shrna to treat cancer - Google Patents

Abstract

The present description relates to an inhibitory RNA molecule, comprising an oligonucleotide that selectively knocks down expression a NANOG pseudogene expressed in many human cancers, a replicating viral vector capable of encoding such inhibitory RNA molecule, pharmaceutical compositions comprising said vector, and methods of treating cancer by administration of said pharmaceutical composition.

Description

PHARMACEUTICAL COMPOSITION COMPRISING NANOG SHRNA, AND METHOD OF USING NANOG SHRNA TO TREAT CANCER

CROSS REFERENCE TO RELATED APPLICATION

This application claims the benefit of U.S. Patent Application Serial No.

14/886,970, filed October 19, 2015, which is incorporated herein by reference.

SEQUENCE LISTING DATA

The Sequence Listing text file attached hereto, created October 13, 2015, size 2 kilobytes, filename "6137NCI-37-PUS-C1_Sequence listing_ST25" is incorporated herein by

reference in its entirety.

TECHNICAL FIELD

The technical field consists of a pharmaceutical composition comprising an oligonucleotide that selectively knocks down expression of either NANOG or a NANOG pseudogene, and a method of treating cancer by administering said pharmaceutical composition to a patient in need thereof.

BACKGROUND

Colorectal carcinoma (CRC) is the second highest cancer-specific mortality in the U.S. population. There remains a need to develop effect therapeutics to prevent and/or treat CRC.

NANOG was identified to be a core embryonic stem cell gene (Chambers et al., 2003, Cell 113:643-55), and was later determined to be associated with a family of eleven pseudogenes with the original parent gene on chromosome 12 (Booth et al., 2004, Genomics 84:229-38). Other members of the family may influence NANOG's function in carcinoma cells, for example by reactivation.

The cancer stem cell hypothesis postulates that a small pluripotent population of cancer cells generates cancer heterogeneity and resistance to chemotherapy and radiation therapy. Three transcription factors, SOX2, OCT4 and NANOG form a core regulatory network that coordinately determines embryonic stem cell self-renewal and differentiation (Zhang et al., 2006, FEBS J 273: 1723-30). Cultured cancer cells, as well as xenograft and human primary prostate cancer cells express a functional variant of NANOG, NANOGP8, which is enriched in putative cancer stem/progenitor cell populations (Zhang et al.; and Jeter et al. 2009, Stem Cells 27:993-1005). NANOG protein expression is functionally important for tumor development based because knockdown of NANOG expression inhibits growth of xenografts of cancer cells from cancer cell lines Du145 HPCa and LAPC4 (prostate), MCF7 (breast cancer), and

l Colo320 (colon cancer) (Jeter et al.). NANOGP8 is also expressed in human glioblastoma cells (Zbinden et al. 2010, EMBO J 29:2659-74).

The role of NANOG and its pseudogenes in tumorigenesis and progression of metastasis for a number of cancers, including gastric, colorectal (CRC), breast, glioblastoma, cervical cancers is currently unknown.

SUMMARY

One aspect of the description is an inhibitory RNA molecule, comprising an oligonucleotide that knocks down expression of either NANOG or NANOGP8. One embodiment is an oligonucleotide that inhibits NANOG but not NANOGP8 expression. Another embodiment is an oligonucleotide that inhibits NANOGP8 but not NANOG expression. Another embodiment is an oligonucleotide that selectively inhibits another NANOG pseudogene. Another embodiment of the RNA molecule is one in which the oligonucleotide is preferably selected from the group consisting of the shRNA sequences of SEQ ID NOS: 3-44, more preferably selected from the group consisting of SEQ ID NOS: 3, 8, 34, and 37.

Another aspect of the invention is a viral vector, comprising the inhibitory RNA molecule described above. The viral vector may be a lentivirus vector. The viral vector may also be from a replicating virus, including a replicating oncolytic virus, for example, an oncolytic adenovirus. Thus, the viral vector may be an adeno virus vector, for example a type 2 or type 5 adenovirus vector. In one embodiment the viral vector is capable of infecting cancer cells. Another embodiment is a lentivirus vector that is an integrating vector. The viral vector preferably is capable of transducing cancer cells. More preferably, the vector is capable of inducing apoptosis of cancer cells. Another embodiment is a viral vector that is capable of inhibiting cancer cell proliferation or inhibiting tumor mass. The viral vector is preferably packaged in a coat protein that specifically binds to CRC cells. The viral vector preferably is capable of inducing spherogenicity of CRC cells in vitro. The viral vector preferably is capable of overexpressing an RNA that inhibits either NANOGP8 or NANOG expression. An embodiment of the viral vector is one in which the

oligonucleotide is preferably selected from the group consisting of the shRNA sequences of SEQ ID NOS: 3-44, more preferably selected from the group consisting of SEQ ID NOS: 3, 8, 34, and 37. Another embodiment of the invention is one in which the viral vector preferably is a lentivirus vector, more preferably an integrating lentivirus vector. Another embodiment of the invention is one in which the viral vector is a replicating vector, preferably a replicating oncolytic adenovirus vector, including a conditionally replicating adenovirus vector (CRAd). Another embodiment of the invention is one in which the viral vector is preferably produced by a vector transfer cassette and a separate helper plasmid.

Another aspect of the invention is a pharmaceutical composition for treating a cancer, comprising the viral vector described above. One embodiment is a

pharmaceutical composition comprising an inhibitory oligonucleotide that is a double stranded RNA molecule. One embodiment of the pharmaceutical composition is one in which the oligonucleotides is a component of a viral vector capable of infecting cancer cells, more preferably colorectal cancer cells. Another embodiment is one in which the viral vector is a lentivirus. Another embodiment is one in which the viral vector is an adenovirus. Alternatively, the oligonucleotide may be a double stranded siRNA. In either instance, the oligonucleotide is preferably selected from the group consisting of the shRNA sequences of SEQ ID NOS: 3-44, more preferably selected from the group consisting of SEQ ID NOS: 3, 8, 34, and 37.

Another aspect of the invention is a method of treating cancer, comprising administering to a patient in need thereof a pharmaceutical composition that inhibits expression of NANOGP8 without substantially changing NANOG function or NANOG without substantially changing NANOGP8 function. An embodiment of the method of treating cancer is one which preferably inhibits cancer cells having a malignant phenotype, more preferably cancer cells capable of metastasizing. Another preferred embodiment is a method in which colorectal cancer is treated. Another embodiment is a method of treating cancer by administering to a patient in need thereof an effective amount of the pharmaceutical compositions described above.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 shows shRNA vector, pPS-H1-LCS-GFP, and packaging plasmids, pVSV- G and pPACK-GAG/REV. The SBI Clone-it enzyme free system was used according to the manufacturer's protocol to assemble the shRNAs described herein. The shRNAs were cloned into the Ligase Cloning Site (LCS) in the SBI lentivector and sequences confirmed by sequencing. HEK293TN cells were used for packaging according to the protocol from manufacturer.

FIG. 2 shows relative expression of core embryonic genes in CRC spheroids and monolayer cultures. Total RNA isolated from seven-day monolayer or spheroid cultures of CRC cell lines, including HCC2998, MIP-101 , Clone A, CX-1 and LS174T. Expression levels of NANOG, POU5FI (OCT4), and SOX2 were measured by RT-PCR. E expression levels were normalized to the expression of GAPDH in each culture condition, and are shown in units of fold-change.

FIG. 3 shows specific inhibition of NANOGP8 and NANOGP8 affects side population in CRC Cells. Fig Effect of shRNA on single cell spherogenicity of Clone A, CX-1 , and LS 174T. shNP8-1 and shNg-2 are more active than the commercially available shNANOG against all 3 human CRC. Mean %±SD. P values by contingency table analysis with Bonferroni correction. P values after Bonferroni correction of contingency table analysis. # P<0.001 vs Parental CRC. * P<0.01 vs Parental CRC.

FIG. 4 shows the effect of inhibiting core embryonic genes on growth of CRC cell lines. Groups of 5-15 NOD/SCID mice were injected SC with 10⁵ viable cells in PBS that had been stably transduced with lentiviral shRNA to NANOG, OCT4, or SOX2 with the empty vector pLKO and the untreated parental control (None) as controls for Clone A (A), CX-1 (B) and LS 174T (C). Kaplan-Meier plots present data for time to tumor appearance with Mean±SEM days presented in Panel D. Probability is the Wilcoxon log rank significance level for each tumor group in Panels A-C. In Panel E, the one-way analysis of variance is presented for those Clone A tumors 42 days after tumor injection in the None (Parental), pLKO or shNANOG groups.

FIG. 5 shows shNANOG inhibits single cell tumorigenicity. Groups of 5-10

NOD/SCID mice were injected with dilutions of 10⁵-10³ viable CX-1 or Clone A cells subcutaneously in NOD/SCID mice. Parental, pLKO.1 and shNANOG transductants were scored for the appearance of tumors over 70 days after tumor inoculation.

FIG. 6 shows shNANOG inhibits experimental metastasis. When 2*10⁶ CX-1 cells were injected into the spleens of NOD/SCID mice, no shNANOG transduced CX-1 cells formed either gross or microscopic liver experimental metastases whereas 45 to 70% of mice injected with parental or control vector did. Error bars: SD; * P<0.01 vs Parental CRC. P value by contingency table analysis with Bonferroni correction. Experimental metastasis from Parental CX-1 did not generate fibrosis or host inflammatory response.

FIGS. 7A and 7B show shNANOG inhibits tumorigenicity. Clone A (FIG. 7A) or CX-1 (FIG. 7B) cells were transducted with the pLKO.1 , shNANOG, or NANOG.

shNANOG decreases growth by prolonging the median days to appearance of tumors for both Clone A and CX-1 as well as the percentage of mice that are tumor free in mice injected with CX-1 cells. Overexpression of NANOG and NANOGP8 shortened the median number of days to tumor appearance in CX-1 compared to the pLKO.1.

FIG. 8 depicts the design of oncolytic viruses (upper) and the CRAd viral constructs of this disclosure.

FIGS. 9A and 9B show the effects of infection of CRC in 3D Suspension Culture. 500 Clone A (FIG. 9A) or CX-1 (FIG. 9B) cells were incubated in 96 well ultra low attachment (ULLA) serum-free medium (SFM) plates for 24 hr before the indicated numbers of Ad5/3-E1aWT-shNp8 viral particles were added to the cultures. Six days later cells were imaged with a 10x objective on a Nikon Diaphot inverted microscope. FIGS. 10A and 10B demonstrate that adenoviruses inhibit the growth of Clone A (FIG. 10A) and CX-1 (FIG. 10B) in suspension culture at 7 Days. 1000 CRC cells in 96 well ultra low attachment plates in serum-free medium were exposed 16 hours after incubation with the titers of the indicated viruses at the viral particles per cell. After 7 days of infection, the mass areas of culture were measured and means ± SD presented. Separate experiments indicate that area measurements correlate with Alamar Blue post labeling metabolism study. The 30 and 100 vp/cell are at least p<0.01 versus the untreated controls.

FIG. 11 shows the effects of Murine 3T3 Fibroblasts on Infectivity of Ad5/3- E1aWT-shNp8. The indicated numbers of CX-1 , LS 174T and 3T3 cells were incubated for 16 hr in 96 well ULLA plates in SFM and infected with 100 vp/ CRC cell and imaged at days as described in FIGS. 10A and 10B. The 3T3 alone culture was also infected at the same titer as the CRC containing cultures. All cultures performed in triplicate.

FIGS. 12A and 12B show the effects of Murine Fibroblasts and shNp8 on CRC Survival. CX-1 (FIG. 12A) and LS 174T (FIG. 12B) cells were incubated in ULLA microtiter palates in SFM and infected with 100 viral particles (vp) per cell. Plates were incubated with Alomar Blue and then read in a microplate reader at 585 nm. Means ± SD are presented.

FIG. 13 shows that the lysate from CRAd-infected CX-1 infects secondary CX-1 and 293 Cells. CX-1 cells were infected with Ad5/3-E1aNp8-shNp8 (the CRAd) for 48 hr in monolayer culture, harvested, washed and a lysate prepared by repeated freeze thawing. This cell lysate was diluted 1 :10 and then added to fresh monolayers of CX-1 or 293T17 cells at the further dilutions indicated. Phase and epi-fluorescence images were taken with a 4x objective on a Nikon Diaphot inverted microscope 48 hr after secondary infection. 293T17 cells contain E1 a wild type and support the replication of adenovirus.

FIGS. 14A, 14B, and 14C show the transmission between CX-1 cells and Annexin

V in CX-1 suspension culture. CX-1 cells were incubated overnight in ULLA plates in SFM and then infected with dilutions of Ad5/3's for 10 days. Images were taken at 3 and 10 days after infection and the percentages of cells infected and expressing GFP were counted. Results are expressed as Mean ± SD of cells that are GFP+. FIG. 14A is CRAd 5/3, while FIG. 14B is the Ad5/3. FIG. 14C is the percentage of cells that express Annexin

V 6 days after infection of CX-1 cell in suspension culture with the indicated virus at 1000 vp/cell.

FIG. 15 shows that CRAd has activity at high concentration against cells growing as spheroids. CX-1 cells were infected at the vp/cell ratios after cells formed spheroids in serum-free medium in suspension culture. The CRAd is Ad5/3-E1aNp8-shNp8 and indicated gold triangles. Triplicate cultures were analyzed for tumor area 6 days after culture iinfection and means ± SD presented. One way ANOVA was performed and was highly significant at 300 and 1000 vp/cell. The means were compared by Holm-Sidak multiple comparisons correction test. P values compared to the untreated controls. Since a plaque forming unit (pfu) is approximately 30 vp/cell (data not shown), this translates to approximately 10-30 pfu/cell.

FIG. 16 shows CRAd AdNP2 primarily inhibits cells that express NANOGP8. Human cells were cultured for 16 hr at 37°C at 5,000 cells per well in 96 well microtiter plates in complete medium. The cells were incubated with XTT for 2 hours at the end of the incubation and the results for each dilution normalized to wells with the same cells not exposed to virus and expressed as Mean ± SD of the percentage of untreated controls.

FIG. 17 shows that CRAd and Ad5/3's are active in vitro on CSC. Individual human carcinoma or normal cells were cultured in Ultra Low Attachment (ULLA) 96 well microtiter plates (Corning) in serum-free medium by limiting dilution. A day later the indicated virus was added at 100 vp/cell or an equivalent amount of medium and the cultures incubated for an additional 8 days. Spheroids of 50 or more cells were counted in wells with individual cells and the results presented as the mean % of single cells that formed spheroids ±SD. P values were determined when means were significantly different by one-way ANOVA and used the Holm-Sidak correction for multiple

comparisons.

FIG. 18 shows the effect of CRAd (Np8-shNp8) and Ad5/3's after ex vivo infection on Xenograft Growth of CX-1 in NOD/SCI D Mice. CX-1 cells were infected with the indicated virus for 3 hours and then admixed with uninfected CX-1 cells before injection so that a total of 3 x 106 cells were injected into groups of 5 mice each with only 5% or 50% of the cells infected ex vivo. Mice were followed and the results 18 days after transplantation are presented. The CRAd is designated Np8-Np8 for Ad5/3-E1aNp8- shNp8 and the other viruses have the wild type E1 a promoter either without an shRNA (Ad5/3-WT), shNp8 (WT-Np8) or the control shNEG (WT-NEG). Volumes were calculated by the formula width*width*length/width and expressed as mm³. Mean ± SEM are presented. One-way ANOVA indicated that means were significantly different and then the means were compared by the Holm-Sidak multiple comparisons correction test. P values are compared to the untreated controls.

FIGS. 19A and 19B shows that intralesional adenovirus continuously inhibits growth of established subcutaneous human colorectal xenografts in nod/scid mice. 3 x 10⁶ viable CX-1 (FIG. 19A) or LS174T (FIG. 19B) cells were injected subcutaneously into groups of 8 -10 NOD/SCID male mice. Xenografts were injected with 300 viral particles/cell on the day indicated and then followed for growth. Mice were euthanized when the area of the tumor reached 3 cm2 as the product of two largest perpendicular diameters. Results are Mean ± SD.

FIG. 20 shows that intralesional adenoviruses inhibit proliferation. Individual mice in Figure 19 were followed during the course of their disease and tumor areas calculated as the product of the two largest perpendicular diameters. Individual logarithmic regression curves were developed and then intercepts and slopes (the proliferation rate) assessed by one-way ANOVA with Holm-Sidak correction for multiple comparisons. Viral therapy inhibited the rate of both CX-1 and LS174T xenografts. Results are Mean ± SD.

FIG. 21 shows that CRAd inhibits NANOG expression in Clone A. Clone A cells were plated in 6 well plates at 2 x 10⁶ cells/well and then exposed to 100vp/cell 16 hr later. Cells were harvested 3 days later, and qRT-PCR performed as previously described (Zhang et al, Oncogene 2013, 32 (37): 4397-405). Results are mean ± SD that are compared to the untreated control wells. P values are assessed as described in other figures.

DETAILED DESCRIPTION OF ILLUSTRATIVE EMBODIMENTS

What is described herein is an allele-specific shRNA knocks down expression of NANOGP8 while not interfering with expression of NANOG. NANOG mRNA has the following structure, based on a cDNA sequence (SEQ ID NO: 1)

1 attataaatc tagagactcc aggattttaa cgttctgctg gactgagctg gttgcctcat

61 gttattatgc aggcaactca ctttatccca atttcttgat acttttcctt ctggaggtcc

121 tatttctcta acatcttcca gaaaagtctt aaagctgcct taaccttttt tccagtccac

181 ctcttaaatt ttttcctcct cttcctctat actaacatga gtgtggatcc agcttgtccc

241 caaagcttgc cttgctttga agcatccgac tgtaaagaat cttcacctat gcctgtgatt

301 tgtgggcctg aagaaaacta tccatccttg caaatgtctt ctgctgagat gcctcacacg

361 gagactgtct ctcctcttcc ttcctccatg gatctgctta ttcaggacag ccctgattct

421 tccaccagtc ccaaaggcaa acaacccact tctgcagaga agagtgtcgc aaaaaaggaa

481 gacaaggtcc cggtcaagaa acagaagacc agaactgtgt tctcttccac ccagctgtgt

541 gtactcaatg atagatttca gagacagaaa tacctcagcc tccagcagat gcaagaactc

601 tccaacatcc tgaacctcag ctacaaacag gtgaagacct ggttccagaa ccagagaatg

661 aaatctaaga ggtggcagaa aaacaactgg ccgaagaata gcaatggtgt gacgcagaag

721 gcctcagcac ctacctaccc cagcctttac tcttcctacc accagggatg cctggtgaac 781 ccgactggga accttccaat gtggagcaac caatcctgga acaattcaac ctggagcaac 841 cagacccaga acatccagtc ctggagcaac cactcctgga acactcagac ctggtgcacc 901 caatcctgga acaatcaggc ctggaacagt cccttctata actgtggaga ggaatctctg 961 cagtcctgca tgcagttcca gccaaattct cctgccagtg acttggaggc tgccttggaa 1021 gctgctgggg aaggccttaa tgtaatacag cagaccacta ggtattttag tactccacaa 1081 accatggatt tattcctaaa ctactccatg aacatgcaac ctgaagacgt gtgaagatga 1141 gtgaaactga tattactcaa tttcagtctg gacactggct gaatccttcc tctcccctcc 1201 tcccatccct cataggattt ttcttgtttg gaaaccacgt gttctggttt ccatgatgcc 1261 catccagtca atctcatgga gggtggagta tggttggagc ctaatcagcg aggtttcttt 1321 tttttttttt ttcctattgg atcttcctgg agaaaatact tttttttttt ttttttttga 1381 aacggagtct tgctctgtcg cccaggctgg agtgcagtgg cgcggtcttg gctcactgca 1441 agctccgtct cccgggttca cgccattctc ctgcctcagc ctcccgagca gctgggacta 1501 caggcgcccg ccacctcgcc cggctaatat tttgtatttt tagtagagac ggggtttcac 1561 tgtgttagcc aggatggtct cgatctcctg accttgtgat ccacccgcct cggcctccct 1621 aacagctggg atttacaggc gtgagccacc gcgccctgcc tagaaaagac attttaataa 1681 ccttggctgc cgtctctggc tatagataag tagatctaat actagtttgg atatctttag 1741 ggtttagaat ctaacctcaa gaataagaaa tacaagtaca aattggtgat gaagatgtat 1801 tcgtattgtt tgggattggg aggctttgct tattttttaa aaactattga ggtaaagggt 1861 taagctgtaa catacttaat tgatttctta ccgtttttgg ctctgttttg ctatatcccc 1921 taatttgttg gttgtgctaa tctttgtaga aagaggtctc gtatttgctg catcgtaatg 1981 acatgagtac tgctttagtt ggtttaagtt caaatgaatg aaacaactat ttttccttta 2041 gttgatttta ccctgatttc accgagtgtt tcaatgagta aatatacagc ttaaacat

The sequence of NANOGP8, based on the cDNA structure is as follows (SEQ ID NO:2):

1 ccaggatttt aacgttctgc tggactgagc tggttgcctc atgttattat gcaggcaact

61 cactttatcc caatttcttg atacttttcc ttctggaggt cctatttctc taacatcttc

121 cagaaaagtc ttaaagctgc cttaaccttt tttccagtcc acctcttaaa ttttttcctc

181 ctcttcctct atactaacat gagtgtggat ccagcttgtc cccaaagctt gccttgcttt

241 gaagaatccg actgtaaaga atcttcacct atgcctgtga tttgtgggcc tgaagaaaac

301 tatccatcct tgcaaatgtc ttctgctgag atgcctcaca cagagactgt ctctcctctt 361 ccttcctcca tggatctgct tattcaggac agccctgatt cttccaccag tcccaaaggc 421 aaacaaccca cttctgcaga gaatagtgtc gcaaaaaagg aagacaaggt cccggtcaag 481 aaacagaaga ccagaactgt gttctcttcc acccagctgt gtgtactcaa tgatagattt 541 cagagacaga aatacctcag cctccagcag atgcaagaac tctccaacat cctgaacctc 601 agctacaaac aggtgaagac ctggttccag aaccagagaa tgaaatctaa gaggtggcag 661 aaaaacaact ggccgaagaa tagcaatggt gtgacgcaga aggcctcagc acctacctac 721 cccagcctct actcttccta ccaccaggga tgcctggtga acccgactgg gaaccttcca 781 atgtggagca accagacctg gaacaattca acctggagca accagaccca gaacatccag 841 tcctggagca accactcctg gaacactcag acctggtgca cccaatcctg gaacaatcag 901 gcctggaaca gtcccttcta taactgtgga gaggaatctc tgcagtcctg catgcacttc 961 cagccaaatt ctcctgccag tgacttggag gctgccttgg aagctgctgg ggaaggcctt 1021 aatgtaatac agcagaccac taggtatttt agtactccac aaaccatgga tttattccta 1081 aactactcca tgaacatgca acctgaagac gtgtgaagat gagtgaaact gatattactc 1141 aatttcagtc tggacactgg ctgaatcctt cctctcccct cctcccatcc ttcataggat 1201 ttttcttgtt tggaaaccac gtgttctggt ttccatgatg cccatccagt caatctcatg 1261 gagggtggag tatggttgga gcctaatcag cgaggtttct tttttttttt ttttttccta 1321 ttggatcttc ctggagaaaa tacttttttt tttttttttt tttgagacgg agtcttgctc 1381 tgtcgcccag gctggagtgc agtggcgcgg tcttggctca ctgcaagctc cgtctgccgg 1441 gttcacgcca ttctcctgcc tcagcctccc gagcagctgg gactacaggc gcccgccacc 1501 tcgcccggct aatattttgt atttttagta gagacggggt ttcactgtgt tagccaggat 1561 ggtctcgatc tcctgacctt gtgatccgcc cgcctcggcc tccctaacag ctgggattta 1621 caggcgtgag ccaccgcgcc ctgcctagaa aagacatttt aataaccttg gctgccgtct 1681 ctggctatag ataagtagat ctaatacgag tttggatatc tttagggttt agaatctaac 1741 ctcaagaata agaaatacaa gtacaaattg gtgatgaaga tgtattcgta ttgtttggga 1801 ttgggaggct ttgcttattt tttaaaaact attgaggtaa agggttaagc tgtaacatac 1861 ttaattgatt tcttaccgtt tttggctctg ttttgctata tcccctaatt tgttggttgt 1921 gctaatcttt gtagaaagag gtctcgtatt tgctgcatcg taatgacatg agtactactt 1981 tagttggttt aagttcaaat gaatgaaaca actatttttc ctttagttga ttttaccctg 2041 atttcaccga gtgtttcaat gagtaaatat acagcttaaa cat Using an algorithm, an oligonucleotide was selected in which shRNA sequences are selected from the group consisting of SEQ ID NOS:3-44, preferably selected from the group consisting of SEQ ID NOS: 3, 8, 34, and 37 (see Table 3, below).

The RNA can be single stranded, including antisense or shRNA. The RNA can be double stranded, including siRNA. In every instance, the RNA must be capable of hybridizing with NANOGP8 mRNA having the sequence corresponding to cDNA of SEQ ID NO:2.

Lentiviral shRNA Vector Preparation

The RNA can be a component of a viral vector, or a shRNA encoded by a viral vector. In either instance, the viral vector comprises an oligonucleotide that inhibits expression of NANOGP8 while not interfering with expression of NANOG, or encodes a shRNA having such capability.

The sequence of the oligonucleotide is preferably selected from the group consisting of the shRNA sequences of SEQ ID NOS:3-44, more preferably selected from the group consisting of SEQ ID NOS: 3, 8, 34, and 37.

The viral vector may be a lentivirus vector, including an integrating lentivirus vector. The viral vector may be a replicating virus, including a conditional replicating virus.

The viral vector may be an oncolytic adenovirus, including a conditionally replicating adenovirus (CRAd). The adenovirus, including the conditional replicating adenovirus constructs, may be any of the known human adenovirus types (HAdV-1 to 57), in seven species (Human adenovirus A to G). The replicating virus may also be a measles virus.

The replicating virus may also be a virus selected from the group consisting of a modified herpes simplex virus (HSV), modified vaccinia virus, and modified parapoxvirus. The ability of the vector to infect cancer cells can be measured with a variety of cell lines, including primary cultures of tumor cells. The ability of the vector to transduce cancer cells can readily be measured by measuring nuclear DNA integration.

Conditionally Replicative Adenoviruses (CRAds) represent new promising therapeutic agents applied to cancer treatment. In one embodiment, cancer-specific replication of the CRAds result in viral-mediated oncolysis of infected tumor tissues and release of the virus progeny, capable of further propagating in surrounding tumor cells but not in those of normal tissues, which would be refractory to CRAd replication.

"Adenovirus" is a double stranded DNA virus of about 3.6 kilobases. In humans, adenoviruses can replicate and cause disease in the eye and in the respiratory, gastrointestinal and urinary tracts. The term "adenovirus" as used herein is intended to encompass all adenoviruses, including the Mastadenovirus and Aviadenovirus genera.

To date, at least forty-seven human serotypes of adenoviruses have been identified (see, e.g., Fields et al., Virology, volume 2, chapter 67 (3d ed., Lippincoft-Raven Publishers). Preferably, the adenovirus is a serogroup C adenovirus, still more preferably the adenovirus is serotype 2 (Ad2) or serotype 5 (Ad5). The various regions of the adenovirus genome have been mapped and are understood by those skilled in the art (see, e.g., FIELDS et al., VIROLOGY, volume 2, chapters 67 and 68 (3d ed., Lippincoft-Raven Publishers). The genomic sequences of the various Ad serotypes, as well as the nucleotide sequence of the particular coding regions of the Ad genome, are known in the art and may be accessed, e.g., from GenBank and NCBI (See, e.g. , GenBank Accession Nos. J091 , M73260, X73487, AF108105, L19443, NC 003266 and NCBI Accession Nos. NC 001405, NC 001460, NC 002067, NC 00454). Those skilled in the art will appreciate that the inventive adenovirus vectors may be modified or "targeted" as described in Douglas et al., (1996) Nature Biotechnology 14: 1574, as well as in U.S. Pat. Nos.

5,922,315; 5,770,442 and 5,712, 136, each of which is incorporated herein by reference.

One of the beneficial attributes of the Ad vector as an anti-tumor therapy agent is that this virus possesses a lytic life cycle, which can be exploited for oncolysis. In this context, Conditionally Replicative Adenovirus (CRAd) defective in early gene regions have been employed to accomplish tumor-specific oncolysis. This experience established the concept that CRAd systems can accomplish a significant anti-tumor effect.

For both replicative vector-based approaches, ectopic infection of non-tumor targets can elicit tumor-associated toxicities. This issue has been addressed by means of "targeting" the vector exclusively to tumor cells. Methods to this end have attempted to alter viral tropism by fiber genetic modification or by means of restricting a therapeutic gene expression via a tumor/tissue selective promoter (TSP) to achieve targeting by viral replication/oncolysis in the CRAd. The most widely used method to create an Ad-based vector for targeted expression of transgene or a CRAd is to utilize a TSP for transcription of Ad5 E1 gene. As result of this approach, CRAds replicate only in cells with high activity of the promoter, which allow expression of E1 at the levels sufficient to mediate viral replication. Thus, the key issue in this strategy is specificity of the tumor/tissue selective promoter in the context of the Ad genome. Furthermore, low activity in the liver is critical for avoiding adverse effects, as the majority of virus released into the blood stream localizes to the liver. It is noteworthy, however, that many promoters that exhibit specificity in plasmid-based constructs do not show such specificity in adenoviruses. The inventors thus sought promoters usable for CRAd construction with these considerations in mind. In this application conditionally replicating adenovirus constructs were made with a 5/3 fiber (hereinafter "CRAd5/3") by replacing the Ψ and proximal part of the E1 a gene with the proximal 1 Kb of the NANOGP8 promoter. This minimal promoter contains consensus binding sites for a number of transcription binding sites that drive expression of NANOGP8 and drive the rest of the E1 a viral gene to initiate replication of the virus in cells that express NANOGP8. Therefore, a CRAd of this disclosure may only replicate in those cells that actively express NANOGP8.

A variety of biological functions can be measured to demonstrate the efficacy of the virus in vitro and in vivo using animal models, including the ability to induce apoptosis of cancer cells, the ability to inhibit cancer cell proliferation. Separately, the ability of the vector to inhibit tumor mass can be measured, most easily where the tumor mass is a xenograft. It may be important to measure the ability of the viral vector to induce spherogenicity in vitro, and to inhibit formation of malignant tumor cells.

Lentiviral Produced shRNA

A lentiviral vector is infectious lentiviral particle. An infectious lentiviral particle is capable of invading a target host cell, include an envelope and exhibit one or more characteristics of a lentivirus, e.g., containing a lentiviral virion including one or more of the gag structural polypeptides p7, p24 or p17, containing a lentiviral envelope including one or more of the env encoded glycoproteins p41 , p120 or p160, containing a genome including one or more lentivirus cis-acting sequences functioning in replication, proviral integration or transcription, containing a genome encoding a lentiviral protease, reverse transcriptase or integrase, or containing a genome encoding regulatory activities such as Tat or Rev.

A lentiviral vector can exhibit some or all of the natural functions of a lentivirus. A lentiviral vector also can exhibit functions additional to, or different from, a naturally occurring lentivirus. For example, a lentiviral vector can be modified to augment or reduce a lentivirus characteristic or to exhibit characteristics of host cells or heterologous cells. Modifications can include, for example, conferring additional or alternative host cell specificity such as by pseudotyping; augmenting or modulating infectivity by, for example, modifying the binding or fusion functions of an envelope polypeptide; incorporating heterologous, chimeric or multifunctional polypeptides or other membrane components into the envelope; or conferring expression of heterologous genes onto a lentiviral vector genome. Therefore, depending of the design and incorporation of lentivirus constituent components, a lentiviral vector can be, for example, replication-competent, replication incompetent or contain some or all cis elements or trans-acting factors. As such, the term is intended to include enveloped retroviral like particles that derive from or contain at least one functional characteristic of a lentivirus and which is capable of invading a target host cell.

A lentiviral vector genome is a nucleic acid that encodes a lentiviral cis nucleic acid sequence required for genome packaging. A lentiviral vector genome also can encode other cis nucleic acid sequences beneficial for gene delivery, including for example, cis sequences required for reverse transcription, proviral integration or genome transcription. A lentiviral vector genome performs transduction functions of a lentiviral vector. As such, the exact makeup of a vector genome will depend on the genetic material desired to be introduced into a target cell. Therefore, a vector genome can encode, for example, additional polypeptides or functions other than that required for packaging, reverse transcription, integration, or transcription. Such functions generally include coding for cis elements required for expression of a nucleic acid of interest. The lentiviral cis sequences or elements can be derived from a lentivirus genome, a lentiviral vector genome or other virus or vector genome so long as the lentiviral vector genome can be packaged by a lentiviral vector and introduced into a target cell.

A nucleic acid vector is any nucleic acid that functions to carry, harbor or express a nucleic acid of interest. Nucleic acid vectors can have specialized functions such as expression, packaging, pseudotyping, transduction or sequencing, for example. Nucleic acid vectors also can have, for example, manipulatory functions such as a cloning or shuttle vector. The structure of the vector can include any desired form that is feasible to make and desirable for a particular use. Such forms include, for example, circular forms such as plasmids and phagemids, as well as linear or branched forms. A nucleic acid vector can be composed of, for example, DNA or RNA, as well as contain partially or fully, nucleotide derivatives, analogs and mimetics. Such nucleic acid vectors can be obtained from natural sources, produced recombinantly or chemically synthesized.

A packaging construct is a nucleic acid vector that encodes retroviral structural polypeptides sufficient for vector production. When used in reference to a lentiviral nucleic acid vector, the term is intended to refer to lentiviral structural polypeptides sufficient for lentiviral vector production. A lentiviral packaging construct can additionally contain other polypeptides that function in trans to facilitate, augment or supplement the efficiency of vector production or the functional characteristics of the vector particle. Structural polypeptides that function in trans for vector production included, for example, lentiviral polypeptides p6, p7, p17 and p24, which are encoded by gag and reverse transcriptase, protease and integrase, which are encoded by pol. Other trans-acting factors that can function in vector production include, for example, the polypeptides encoded by rev and tat. A packaging construct can be designed to express some or all of such trans-acting factors stably or transiently. Additionally, it should be understood that the term is intended to include construct designs that separate or split the expression of trans-acting factors, or components thereof, onto two or more nucleic acid vectors. Accordingly, a packaging construct can include multiple different nucleic acid vectors which together encode structural polypeptides sufficient for retroviral vector production.

A targeting polypeptide is a polypeptide that contains a binding partner to a molecule expressed on the surface of a targeted cell or tissue, or to a molecule that is otherwise accessible to the vector particle. Expression of a binding partner on a lentiviral vector of the invention allows the vector to be directed to, bind and attach to a

predetermined target cell or tissue type. A targeting polypeptide can consist of, or include, any molecule that exhibits binding affinity forward a cognate binding partner. Therefore, targeting polypeptides can include, for example, ligands, receptors, co-receptors, counter- ligands, counter-receptors, antigens and epitopes, as well as other affinity binders well known to those skilled in the art.

A ligand is a molecule that exhibits selective binding affinity for another molecule. Therefore, the term refers to one component of a bi- or multi-component affinity binding reaction. As one constituent of two or more interacting molecular binding species the term is intended to be neutral with reference to orientation. Therefore, a ligand can refer to all types of affinity ligands well known to those skilled in the art including, for example, ligands, haptens, counter-ligands, receptors and counter-receptors. Ligands include a wide range of molecular species well known to those skilled in the art including, for example, polypeptides, nucleic acid and other macromolecules as well as small inorganic or organic molecules.

A receptor is a molecule that exhibits selective binding affinity for another molecule. A receptor can refer to all types of affinity binding molecules well known to those skilled in the art including, for example, receptors, counter-receptors, counter- ligands, ligands and haptens. Where both or all components of a receptor binding reaction are referred to herein, reference may be made to one component as a receptor and to the cognate component counter-receptor or ligand. However, it is understood that a receptor can be referred to equally as either a receptor or a ligand or by any other nomenclature well known to those skilled in the art which designates a pair or complex of affinity binding components. Binding of a receptor to its partner can be, for example, through non-covalent or covalent interactions and can include, for example, binding affinity to polypeptides, nucleic acid, other macromolecules and small molecules.

Receptors include a wide range of molecular species well known to those skilled in the art including, for example, polypeptides, nucleic acid and other macromolecules as well as small inorganic or organic molecules.

A lentivirus is an icosahedral enveloped virus having a diploid RNA genome that becomes integrated into the host chromosome as a proviral DNA for genome replication. The lentiviral genome contains gag, pol and env genes which encode the structural polypeptides of the virion (p17, p24, p7 and p6); the viral enzymes protease, reverse transcriptase and integrase, and the envelope glycoproteins (gp120 and gp41), respectively. The lentiviral genome also encodes two regulatory polypeptides (Tat and Rev) and four accessory polypeptides that play a role in virulence (Vif, Vpu, Vpr and Nef). Lentiviruses also have the ability to efficiently infect and transduce proliferating cells. Despite the pathogenesis associated with lentiviruses, it is well known to those skilled in the art that the undesirable properties of lentiviruses can be recombinantly separated so that its beneficial characteristics can be harnessed as a delivery vehicle for therapeutic or diagnostic genes. Therefore, lentiviral-based vectors can be produced that are safe, replication-defective and self-inactivating while still maintaining the beneficial ability to transduce non-dividing cells and integrate into the host chromosome for stable expression. A description of the various different modalities of lentiviral vector and packaging systems for vector assembly and gene delivery can be found in, for example, in Naldini et al., Science 272:263-267 (1996); Naldini et al, Proc. Natl. Acad. Sci. USA 93: 11382-1 1388 (1996); Zufferey et al, Nature Bio. 15:871-875 (1997); Dull et al. J. Virol. 72:4638471 (1998); Miyoshi et al, J. Virol. 72:8150-8157 (1998), and Zufferey et al., J. Virol. 72:9873-9880 (1998). Desirable targeting specificities can include highly specific lentiviral targeting vectors as well as broad spectrum targeting vectors. For example, highly specific, or mono-specific, lentiviral targeting vectors of the invention can be generated, for example, by using a heterologous targeting polypeptide that has a corresponding binding partner specific to a particular cell type or tissue. In contrast, broad spectrum targeting vectors can be generated by using a heterologous targeting polypeptide that has a cognate binding partner on multiple different cell types or tissues or by using multiple different heterologous targeting polypeptides that have specificities to different cell types or tissues. Being able to engineer a single specificity to a unique target cell marker, multiple specificities unique to different target cells or a single specificity to a ubiquitous marker provides the lentiviral vectors of the invention with flexible and versatile targeting capabilities because they can be tailored for many different targeting and delivery applications.

Highly specific lentiviral targeting vectors can be useful as delivery vehicles for therapeutic or diagnostic nucleic acids or other compounds when, for example, it is desirable to target a particular tissue, a particular class of cell types, a particular cell type or one or more particular subtypes of any of these categories. Such specific lentiviral vectors can be generated with a heterologous targeting polypeptide that exhibit binding affinity to a unique cognate binding partner on the targeted tissue, class, cell type or subtype for the delivery of a therapeutic or diagnostic molecule. Broad spectrum or ubiquitous targeting vectors can be useful as delivery vehicles for therapeutic molecules when, for example, pathogenesis or the causative agent is pieiotropic in nature.

Ubiquitous targeting vectors also are useful as delivery vehicles for diagnostic molecules to, for example, identify regions of pathogenesis or to ascertain the cause or symptoms of an aberrant condition. As described further below, the therapeutic or diagnostic nucleic acids or compounds include, for example, an encoding nucleic acid such as a transgene, that when expressed in the targeted cell or tissue produces a polypeptide having the desired therapeutic or diagnostic activity.

Engineering a predetermined binding specificity into the vector can be achieved by incorporating into a vector envelope a targeting polypeptide having a desired binding specificity. Because a lentivirus is an enveloped virus, a targeting polypeptide can be incorporated by, for example, normal cellular processes for transmembrane insertion, or for membrane attachment or anchoring, of polypeptides, lipids and other

macromolecules. For example, a targeting polypeptide can contain a transmembrane domain or a membrane attachment domain that incorporates into, or is capable of incorporating into, a lentiviral vector envelope. Alternatively, a targeting polypeptide can contain a membrane anchoring domain that signals attachment to, for example, a lipid anchor. Nucleic acids encoding a particular targeting polypeptide can be generated by recombinant methods or chemical synthesis and then expressed in a vector packaging system for automatic incorporation into a lentiviral vector envelope. Alternatively, a particular targeting polypeptide can be produced in vitro or chemically synthesized and then incorporated into a vector envelope using a cell-free system. Therefore, the targeting polypeptides of the invention include integral membrane polypeptides, peripheral membrane polypeptides as well as those polypeptides anchored by other non-polypeptide molecules or macromolecules.

Other methods known to those skilled in the art also can be used for incorporation or attachment of a targeting polypeptide into a lentiviral vector of the invention. Such other methods include, for example, the noncovalent association of a targeting polypeptide with a envelope associated polypeptide, lipid, carbohydrate or other molecule, or chemical conjugation. Therefore, incorporation can be accomplished using a variety of methods well known to those skilled in the art.

A targeting polypeptide can be, in whole or in part, heterologous or homologous to a lentiviral polypeptide so long as it can incorporate, or be made to incorporate, into a vector envelope of the invention. For example, a heterologous targeting polypeptide can be derived from a species other than lentivirus or derived from a molecule different from the lentivirus gp160 or gp120 envelope polypeptides. Where a targeting polypeptide is heterologous, it is sufficient for at least one portion of the targeting polypeptide to be derived from a non-lentiviral species or non-lentiviral envelope polypeptide.

Targeting polypeptides useful in the lentiviral vectors of the invention include those polypeptides that exhibit binding affinity toward a cognate binding partner on a target of interest. A targeting polypeptide can include, for example, an affinity ligand, a receptor, an antibody, counter-ligand or counter-receptor so long as it exhibits binding affinity toward a cognate binding partner. Specific examples of targeting polypeptides include, for example, transferrin, an apolipoprotein, Rabies G glycoprotein and lentivirus gp120.RNA interference (RNAi) is mediated by double stranded RNA (dsRNA) molecules that have sequence-specific homology to their target nucleic acid sequences (Caplen, N. J., et al, Proc. Natl. Acad. ScL USA 98:9742-9747 (2001)). Biochemical studies in Drosophila cell-free lysates indicate that, in certain embodiments of the present invention, the mediators of RNA-dependent gene silencing are 21-25 nucleotide "small interfering" RNA duplexes (siRNAs). The siRNAs are derived from the processing of dsRNA by an RNase enzyme known as Dicer (Bernstein, E., et al, Nature 409:363366 (2001)). siRNA duplex products are recruited into a multi-protein siRNA complex termed RISC (RNA Induced Silencing Complex). A RISC is then believed to be guided to a target nucleic acid (suitably mRNA), where the siRNA duplex interacts in a sequence-specific way to mediate cleavage in a catalytic fashion (Bernstein, E., et al, Nature 409: 363-366 (2001); Boutla, A., et al, Curr. Biol. 11 : 1776-1780 (2001)). Small interfering RNAs that can be used in accordance with the present invention can be synthesized and used according to procedures that are well known in the art and that will be familiar to the ordinarily skilled artisan. Small interfering RNAs for use in the methods of the present invention suitably comprise between about 0 to about 50 nucleotides (nt). In examples of nonlimiting embodiments, siRNAs can comprise about 5 to about 40 nt, about 5 to about 30 nt, about 10 to about 30 nt, about 15 to about 25 nt, or about 20-25 nucleotides.

An RNAi is an at least partly double-stranded RNA having a structure

characteristic of molecules that are known in the art to mediate inhibition of gene expression through an RNAi mechanism or an RNA strand comprising at least partially complementary portions that hybridize to one another to form such a structure. When an RNA comprises complementary regions that hybridize with each other, the RNA will be said to self-hybridize. An RNAi agent includes a portion that is substantially

complementary to a target gene. An RNAi agent, optionally includes one or more nucleotide analogs or modifications. One of ordinary skill in the art will recognize that RNAi agents that are synthesized in vitro can include ribonucleotides, deoxyribonucleotides, nucleotide analogs, modified nucleotides or backbones, etc., whereas RNAi agents synthesized intracellular^, e.g., encoded by DNA templates, typically consist of RNA, which may be modified following transcription. Of particular interest herein are short RNAi agents, i.e., RNAi agents consisting of one or more strands that hybridize or self-hybridize to form a structure that comprises a duplex portion between about 15-29 nucleotides in length, optionally having one or more mismatched or unpaired nucleotides within the duplex. RNAi agents include short interfering RNAs (siRNAs), short hairpin RNAs (shRNAs), and other RNA species that can be processed intracellular^ to produce shRNAs including, but not limited to, RNA species identical to a naturally occurring miRNA precursor or a designed precursor of an miRNA-like RNA.

The term "short, interfering RNA" (siRNA) refers to a nucleic acid that includes a double-stranded portion between about 15-29 nucleotides in length and optionally further comprises a single-stranded overhang {e.g., 1-6 nucleotides in length) on either or both strands. The double-stranded portion is typically between 17-21 nucleotides in length, e.g., 19 nucleotides in length. The overhangs are typically present on the 3' end of each strand, are usually 2 nucleotides long, and are composed of DNA or nucleotide analogs. An siRNA may be formed from two RNA strands that hybridize together, or may alternatively be generated from a longer double-stranded RNA or from a single RNA strand that includes a self-hybridizing portion, such as a short hairpin RNA. One of ordinary skill in the art will appreciate that one or more, mismatches or unpaired nucleotides can be present in the duplex formed by the two siRNA strands. One strand of an siRNA (the "antisense" or "guide" strand) includes a portion that hybridizes with a target nucleic acid, e.g., an mRNA transcript. Typically, the antisense strand is perfectly complementary to the target over about 15-29 nucleotides, typically between 17-21 nucleotides, e.g., 19 nucleotides, meaning that the siRNA hybridizes to the target transcript without a single mismatch over this length. However, one of ordinary skill in the art will appreciate that one or more mismatches or unpaired nucleotides may be present in a duplex formed between the siRNA strand and the target transcript.

A short hairpin RNA (shRNA) is a nucleic acid molecule comprising at least two complementary portions hybridized or capable of hybridizing to form a duplex structure sufficiently long to mediate RNAi (typically between 15-29 nucleotides in length), and at least one single-stranded portion, typically between approximately 1 and 10 nucleotides in length that forms a loop connecting the ends of the two sequences that form the duplex. The structure may further comprise an overhang. The duplex formed by hybridization of self-complementary portions of the shRNA has similar properties to those of siRNAs and, as described below, shRNAs are processed into siRNAs by the conserved cellular RNAi machinery. Thus shRNAs are precursors of siRNAs and are similarly capable of inhibiting expression of a target transcript. As is the case for siRNA, an shRNA includes a portion that hybridizes with a target nucleic acid, e.g., an mRNA transcript and is usually perfectly complementary to the target over about 15-29 nucleotides, typically between 17-21 nucleotides, e.g., 19 nucleotides. However, one of ordinary skill in the art will appreciate that one or more mismatches or unpaired nucleotides may be present in a duplex formed between the shRNA strand and the target transcript.

An RNAi agent is considered to be "targeted" to a transcript and to the gene that encodes the transcript if (1) the RNAi agent comprises a portion, e.g., a strand, that is at least approximately 80%, approximately 85%, approximately 90%, approximately 91 %, approximately 92%, approximately 93%, approximately 94%, approximately 95%, approximately 96%, approximately 97%, approximately 98%, approximately 99%, or approximately 100% complementary to the transcript over a region about 15-29 nucleotides in length, e.g., a region at least approximately 15, approximately 17, approximately 18, or approximately 19 nucleotides in length; and/or (2) the Tm of a duplex formed by a stretch of 15 nucleotides of one strand of the RNAi agent and a 15 nucleotide portion of the transcript, under conditions (excluding temperature) typically found within the cytoplasm or nucleus of mammalian cells is no more than approximately 15° C. lower or no more than approximately 10° C. lower, than the Tm of a duplex that would be formed by the same 15 nucleotides of the RNAi agent and its exact

complement; and/or (3) the stability of the transcript is reduced in the presence of the RNAi agent as compared with its absence. An RNAi agent targeted to a transcript is also considered targeted to the gene that encodes and directs synthesis of the transcript. A target region is a region of a target transcript that hybridizes with an antisense strand of an RNAi agent. A target transcript is any RNA that is a target for inhibition by RNA interference.

Lentivectors with the Ligase-free Clone-ITTM system in which shRNAs are assembled under the control of the H1 promoter as well as a GFP protein (copGFP) under control of the CMV promoter may be used to produce specific shRNA. The lentivector is created as vector particles that are pseudotyped with the VSV-G envelope protein by transient transfection of 293T cells with co-transfection of three separate packaging plasmids as diagrammed below: SBI's Expression lentivectors together with the pPACK packaging plasmids are third-generation lentiviral expression systems.

This four plasmid co-transfection system is designed to increase the safety of the

HIV-based lentivectors by decreasing the chance of recombination for the potential generation of replication-competent lentivirus (RCL) during production.

The RSV promoter upstream of 5' LTR in the lentivector allows efficient Tat- independent production of viral RNA, reducing the number of genes from HIV-1 that are used in this system that also has eliminated tat.

Number of lentiviral genes necessary for packaging, replication and transduction is reduced to three (gag, pol, rev), and the corresponding proteins are expressed from different plasmids lacking packaging signals. None of the HIV-1 genes (gag, pol, rev) are present in the packaged viral genome, as they are expressed from packaging plasmids lacking a packaging signal.

Pseudoviral particles will carry two copies of the shRNA construct. In addition, a deletion in the enhancer of the U3 region of 3' LTR in the vector transfer plasmid promotes self-inactivation of the lentiviral construct after transduction and integration into genomic DNA of the target cells and is thought to reduce potential insertional

mutagenesis following transduction of the cells of interest.

RNA Delivery by Liposome-Nucleic Acid Particles

A lipid particle is a lipid formulation that can be used to deliver an active agent or therapeutic agent, such as a nucleic acid (e.g., an interfering RNA), to a target site of interest (e.g., cell, tissue, organ, and the like). The lipid particle may be a nucleic acid- lipid particle, which is typically formed from a cationic lipid, a non-cationic lipid, and optionally a conjugated lipid that prevents aggregation of the particle. The nucleic acid may be encapsulated in the lipid portion of the particle, thereby protecting it from enzymatic degradation.

A stable nucleic acid-lipid particle a particle made from lipids (e.g., a cationic lipid, a non-cationic lipid, and optionally a conjugated lipid that prevents aggregation of the particle), wherein the interfering RNA is fully encapsulated within the lipid.

The lipid particles typically have a mean diameter of from about 30 nm to about 150 nm, from about 40 nm to about 150 nm, from about 50 nm to about 150 nm, from about 60 nm to about 130 nm, from about 70 nm to about 1 10 nm, from about 70 nm to about 100 nm, from about 80 nm to about 100 nm, from about 90 nm to about 100 nm, from about 70 to about 90 nm, from about 80 nm to about 90 nm, from about 70 nm to about 80 nm, or about 30 nm, 35 nm, 40 nm, 45 nm, 50 nm, 55 nm, 60 nm, 65 nm, 70 nm, 75 nm, 80 nm, 85 nm, 90 nm, 95 nm, 100 nm, 105 nm, 110 nm, 1 15 nm, 120 nm, 125 nm, 130 nm, 135 nm, 140 nm, 145 nm, or 150 nm, and are substantially non-toxic. In addition, nucleic acids, when present in the lipid particles of the present invention, are resistant in aqueous solution to degradation with a nuclease

A lipid particle provides an interfering RNA with full encapsulation, partial encapsulation, or both. In a preferred embodiment, the nucleic acid is fully encapsulated in the lipid particle to form a nucleic acid-lipid particle.

A conjugated lipid inhibits aggregation of lipid particles, including, PEG-lipid conjugates such as, e.g., PEG coupled to dialkyloxypropyls (e.g., PEG-DAA conjugates), PEG coupled to diacylglycerols (e.g., PEG-DAG conjugates), PEG coupled to cholesterol, PEG coupled to phosphatidylethanolamines, and PEG conjugated to ceramides, cationic PEG lipids, polyoxazoline (POZ)-lipid conjugates (e.g., POZ-DAA conjugates; polyamide oligomers (e.g., ATTA-lipid conjugates), and mixtures thereof. PEG or POZ can be conjugated directly to the lipid or may be linked to the lipid via a linker moiety. Any linker moiety suitable for coupling the PEG or the POZ to a lipid can be used including, e.g., non-ester containing linker moieties and ester-containing linker moieties. In certain preferred embodiments, non-ester containing linker moieties, such as amides or carbamates, are used.

An amphipathic lipid has a hydrophobic portion that orients into a hydrophobic phase, and a hydrophilic portion orients toward the aqueous phase. Hydrophilic characteristics derive from the presence of polar or charged groups such as

carbohydrates, phosphate, carboxylic, sulfato, amino, sulfhydryl, nitro, hydroxyl, and other like groups. Hydrophobicity can be conferred by the inclusion of apolar groups that include, but are not limited to, long-chain saturated and unsaturated aliphatic hydrocarbon groups and such groups substituted by one or more aromatic, cycloaliphatic, or heterocyclic group(s). Examples of amphipathic compounds include, but are not limited to, phospholipids, aminolipids, and sphingolipids.

Representative examples of phospholipids include, but are not limited to, phosphatidylcholine, phosphatidylethanolamine, phosphatidylserine, phosphatidylinositol, phosphatidic acid, palmitoyloleoyl phosphatidylcholine, lysophosphatidylcholine, lysophosphatidylethanolamine, dipalmitoylphosphatidylcholine,

dioleoylphosphatidylcholine, distearoylphosphatidylcholine, and

dilinoleoylphosphatidylcholine. Other compounds lacking in phosphorus, such as sphingolipid, glycosphingolipid families, diacylglycerols, and (3-acyloxyacids, are also within the group designated as amphipathic lipids. Additionally, the amphipathic lipids described above can be mixed with other lipids including triglycerides and sterols.

A neutral lipid exists either in an uncharged or neutral zwitterionic form at a selected pH. At physiological pH, such lipids include, for example,

diacylphosphatidylcholine, diacylphosphatidylethanolamine, ceramide, sphingomyelin, cephalin, cholesterol, cerebrosides, and diacylglycerols.

A non-cationic lipid may be any amphipathic lipid as well as any other neutral lipid or anionic lipid.

An anionic lipid is negatively charged at physiological pH. These lipids include, but are not limited to, phosphatidylglycerols, cardiolipins, diacylphosphatidylserines, diacylphosphatidic acids, N-dodecanoyl phosphatidylethanolamines, N-succinyl phosphatidylethanolamines, N-glutarylphosphatidylethanolamines,

lysylphosphatidylglycerols, palmitoyloleyolphosphatidylglycerol (POPG), and other anionic modifying groups joined to neutral lipids.

A hydrophobic lipid has apolar groups that include, but are not limited to, long- chain saturated and unsaturated aliphatic hydrocarbon groups and such groups optionally substituted by one or more aromatic, cycloaliphatic, or heterocyclic group(s). Suitable examples include, but are not limited to, diacylglycerol, dialkylglycerol, N— N- dialkylamino, 1 ,2-diacyloxy-3-aminopropane, and 1 ,2-dialkyl-3-aminopropane. The nucleic acid-lipid particle comprises: (a) a nucleic acid (e.g., an interfering RNA); (b) a cationic lipid comprising from about 50 mol % to about 65 mol % of the total lipid present in the particle; (c) a non-cationic lipid comprising from about 25 mol % to about 45 mol % of the total lipid present in the particle; and (d) a conjugated lipid that inhibits aggregation of particles comprising from about 5 mol % to about 10 mol % of the total lipid present in the particle.

The nucleic acid-lipid particle comprises: (a) a nucleic acid (e.g., an interfering RNA); (b) a cationic lipid comprising from about 50 mol % to about 60 mol % of the total lipid present in the particle; (c) a mixture of a phospholipid and cholesterol or a derivative thereof comprising from about 35 mol % to about 45 mol % of the total lipid present in the particle; and (d) a PEG-lipid conjugate comprising from about 5 mol % to about 10 mol % of the total lipid present in the particle.

In certain other embodiments, the nucleic acid-lipid particle comprises: (a) a nucleic acid (e.g., an interfering RNA); (b) a cationic lipid comprising from about 55 mol % to about 65 mol % of the total lipid present in the particle; (c) cholesterol or a derivative thereof comprising from about 30 mol % to about 40 mol % of the total lipid present in the particle; and (d) a PEG-lipid conjugate comprising from about 5 mol % to about 10 mol % of the total lipid present in the particle.

Monitoring of the Transduced Cells

Cells transduced with the lentiviral vectors produced using the SBI's system is monitored based on the expression of the green fluorescent protein. Monitoring is performed using quantitative PCR methods based on the detection of the genetic marker Gtag that is a 227 bp non-coding sequence derived from EGFP. Since Gtag is derived from EGFP, the cells transduced with the SBI's LVs are monitored by qPRC using the Gtag primers set and probe.

Generation of Replication Competent Lentivirus (RCL) Cell Lines

The LV shRNA will be incubated with human CRC cell lines that have been found to be free of pathogens including HIV. These include Clone A and CX-1 human CRC lines; LEC-009 and LEC-01 1 cell lines; and a third cell line, LS 174T (ATCC); KM-12c and HCC 2998 (NCI-60 repository through the Developmental Therapeutics Program at the NCI).

Animals

The mice are NOD/SCI D mice that are generally considered a non-permissive host. However, it is clear that local (Matsumoto et al., 2010, BMC Gastroenterology 10:44) or systemic (Shiau et al., 2010, BMC Cancer 10:245) delivery of LV delivered genes or inhibitory RNA can transduce mouse cells with the vector-delivered gene or product. However, since the mice lack HIV LV components (gag, pol and env for instance), they will not be able to support recombination events leading to subsequent viral replication within the mice. Mice will be injected with LV either intratumorally or systemically via tail vein injection. Local treatment causes transduction of target genes locally but does not cause transduction in distant tissues. In contrast, systemic injection of VSV-G vector particles via the tail vein in another report does lead to transduction in liver, lung and other organs (Shiau et al.). Several investigators have demonstrated that intratumoral injection of xenografts in mice with lentiviral shRNA to different genes (Deharvengt et al., 2010, Cancer Gene Ther. 17:325-33; and Frade R, et al., 2008 Cancer Gene Ther. 15:591-604) have caused tumor regression without seeming host toxicity. In addition, Duan et al. (Acta Otolyrngol 130:25-30, 2010) demonstrated that local injection into the cochlea of lentiviral particles remained local without extension into the brain.

Both intratumoral injection of subcutaneous and possibly hepatic CRC xenografts as well as systemic injection of LV shRNA can be performed. shRNA to embryonic genes may be a useful therapy for liver metastases in CRC. First tumor implants in the livers of NOD/SCI D mice are established, and then either intralesional or systemic treatment is performed with LV shRNA. The intralesional treatment of liver lesions cause LV pseudoparticles to be incorporated into the systemic circulation and require that mice with liver tumors will need appropriate, housing and husbandry as for direct systemic injection of LV pseudoparticles.

The lipid particles in which an active agent or therapeutic agent such as an interfering RNA (e.g., siRNA) is entrapped within the lipid portion of the particle and is protected from degradation, can be formed by any method known in the art including, but not limited to, a continuous mixing method, a direct dilution process, and an in-line dilution process.

In particular embodiments, the cationic lipids may comprise one or more of the cationic lipids described herein or salts thereof, alone or in combination with other cationic lipids. In other embodiments, the non-cationic lipids are egg sphingomyelin (ESM), distearoylphosphatidylcholine (DSPC), dioleoylphosphatidylcholine (DOPC), 1-palmitoyl2- oleoyl-phosphatidylcholine (POPC), dipalmitoyl-phosphatidylcholine (DPPC),

monomethyl-phosphatidylethanolamine, dimethyl-phosphatidylethanolamine, 14:0 PE (1 ,2-dimyristoyl-phosphatidylethanolamine (DMPE)), 16:0 PE (1 ,2-dipalmitoyl- phosphatidylethanolamine (DPPE)), 18:0 PE (1 ,2-distearoyl-phosphatidylethanolamine (DSPE)), 18: 1 PE (1 ,2-dioleoyl-phosphatidylethanolamine (DOPE)), 18: 1 trans PE (1 ,2- dielaidoyl-phosphatidylethanolamine (DEPE)), 18:0-18:1 PE (1-stearoyl-2-oleoyl- phosphatidylethanolamine (SOPS)), 16:0-18: 1 PE (1-palmitoyl-2- oleoylphosphatidylethanolamine (POPE)), polyethylene glycolbased polymers (e.g., PEG 2000, PEG 5000, PEG-modified diacylglycerols, or PEG-modified dialkyloxypropyls), cholesterol, derivatives thereof, or combinations thereof.

Nucleic acid-lipid particles may be produced via a continuous mixing method, e.g., a process that includes providing an aqueous solution comprising a nucleic acid (e.g., interfering RNA) in a first reservoir, providing an organic lipid solution in a second reservoir (wherein the lipids present in the organic lipid solution are solubilized in an organic solvent, e.g., a lower alkanol such as ethanol), and mixing the aqueous solution with the organic lipid solution such that the organic lipid solution mixes with the aqueous solution so as to substantially instantaneously produce a lipid vesicle (e.g., liposome) encapsulating the nucleic acid within the lipid vesicle.

The action of continuously introducing lipid and buffer solutions into a mixing environment, such as in a mixing chamber, causes a continuous dilution of the lipid solution with the buffer solution, thereby producing a lipid vesicle substantially

instantaneously upon mixing. The lipid solution is diluted sufficiently rapidly in a hydration process with sufficient force to effectuate vesicle generation. By mixing the aqueous solution comprising a nucleic acid with the organic lipid solution, the organic lipid solution undergoes a continuous stepwise dilution in the presence of the buffer solution (i.e., aqueous solution) to produce a nucleic acid-lipid particle

The nucleic acid-lipid particles formed using the continuous mixing method typically have a size of from about 30 nm to about 150 nm, from about 40 nm to about 150 nm, from about 50 nm to about 150 nm, from about 60 nm to about 130 nm, from about 70 nm to about 110 nm, from about 70 nm to about 100 nm, from about 80 nm to about 100 nm, from about 90 nm to about 100 nm, from about 70 to about 90 nm, from about 80 nm to about 90 nm, from about 70 nm to about 80 nm, less than about 120 nm. 110 nm, 100 nm, 90 nm, or 80 nm, or about 30 nm, 35 nm, 40 nm, 45 nm, 50 nm, 55 nm, 60 nm, 65 nm, 70 nm, 75 nm, 80 nm, 85 nm, 90 nm, 95 nm, 100 nm, 105 nm, 1 10 nm, 115 nm, 120 nm, 125 nm, 130 nm, 135 nm, 140 nm, 145 nm, or 150 nm (or any fraction thereof or range therein). The particles thus formed do not aggregate and are optionally sized to achieve a uniform particle size.

Nucleic acid-lipid particles may be produced via a direct dilution process that includes forming a lipid vesicle (e.g., liposome) solution and immediately and directly introducing the lipid vesicle solution into a collection vessel containing a controlled amount of dilution buffer. In preferred aspects, the collection vessel includes one or more elements configured to stir the contents of the collection vessel to facilitate dilution. In one aspect, the amount of dilution buffer present in the collection vessel is substantially equal to the volume of lipid vesicle solution introduced thereto. As a non-limiting example, a lipid vesicle solution in 45% ethanol when introduced into the collection vessel containing an equal volume of dilution buffer will advantageously yield smaller particles.

The nucleic acid-lipid particles formed using the direct dilution and in-line dilution processes typically have a size of from about 30 nm to about 150 nm, from about 40 nm to about 150 nm, from about 50 nm to about 150 nm, from about 60 nm to about 130 nm, from about 70 nm to about 1 10 nm, from about 70 nm to about 100 nm, from about 80 nm to about 100 nm, from about 90 nm to about 100 nm, from about 70 to about 90 nm, from about 80 nm to about 90 nm, from about 70 nm to about 80 nm, less than about 120 nm, 110 nm, 100 nm, 90 nm, or 80 nm, or about 30 nm, 35 nm, 40 nm, 45 nm, 50 nm, 55 nm, 60 nm, 65 nm, 70 nm, 75 nm, 80 nm, 85 nm, 90 nm, 95 nm, 100 nm, 105 nm, 1 10 nm, 115 nm, 120 nm, 125 nm, 130 nm, 135 nm, 140 nm, 145 nm, or 150 nm (or any fraction thereof or range therein).

Once formed, the lipid particles are useful for the introduction of active agents or therapeutic agents (e.g., nucleic acids such as interfering RNA) into cells. Accordingly, the present invention also provides methods for introducing an active agent or therapeutic agent such as a nucleic acid (e.g., interfering RNA) into a cell. Preferably, the cell is a tumor cell such as, e.g., a cell present in a solid tumor. In certain embodiments, the cell may be a non-tumor cell that produces one or more angiogenic and/or growth factors associated with tumorigenesis or cell transformation. The methods are carried out in vitro or in vivo by first forming the particles as described above and then contacting the particles with the cells (e.g., cells of a solid tumor) for a period of time sufficient for delivery of the active agent or therapeutic agent to the cells to occur. The lipid particles can be adsorbed to almost any cell type with which they are mixed or contacted. Once adsorbed, the particles can either be endocytosed by a portion of the cells, exchange lipids with cell membranes, or fuse with the cells. Transfer or incorporation of the active agent or therapeutic agent (e.g., nucleic acid) portion of the particle can take place via any one of these pathways. In particular, when fusion takes place, the particle membrane is integrated into the cell membrane and the contents of the particle combine with the intracellular fluid.

The lipid particles can be administered either alone or in a mixture with a pharmaceutically acceptable carrier (e.g., physiological saline or phosphate buffer) selected in accordance with the route of administration and standard pharmaceutical practice. Generally, normal buffered saline (e.g., 135-150 mM NaCI) will be employed as the pharmaceutically acceptable carrier. Other suitable carriers include, e.g., water, buffered water, 0.4% saline, 0.3% glycine, and the like, including glycoproteins for enhanced stability, such as albumin, lipoprotein, globulin, etc. Additional suitable carriers are described in, e.g., REMINGTON'S PHARMACEUTICAL SCIENCES, Mack Publishing Company, Philadelphia, Pa., 17th ed.

A carrier includes any and all solvents, dispersion media, vehicles, coatings, diluents, antibacterial and antifungal agents, isotonic and absorption delaying agents, buffers, carrier solutions, suspensions, colloids, and the like. The phrase

"pharmaceutically acceptable" refers to molecular entities and compositions that do not produce an allergic or similar untoward reaction when administered to a human.

The pharmaceutically acceptable carrier is generally added following lipid particle formation. Thus, after the lipid particle is formed, the particle can be diluted into pharmaceutically acceptable carriers such as normal buffered saline.

Systemic delivery for in vivo therapy, e.g., delivery of a therapeutic nucleic acid to a distal target cell via body systems such as the circulation, has been achieved using nucleic acid-lipid particles. For in vivo administration, administration can be in any manner known in the art, e.g., by injection, oral administration, inhalation (e.g., intransal or intratracheal), transdermal application, or rectal administration. Administration can be accomplished via single or divided doses. The pharmaceutical compositions can be administered parenterally, i.e., intraarticularly, intravenously, intraperitoneally,

subcutaneously, or intramuscularly. In some embodiments, the pharmaceutical compositions are administered intravenously or intraperitoneally by a bolus injection. The lipid particles can be administered by direct injection at the site of disease or by injection at a site distal from the site of disease.

Where the lipid particles are administered intravenously, at least about 5%, 10%, 15%, 20%, or 25% of the total injected dose of the particles is present in plasma about 8, 12, 24, 36, or 48 hours after injection. In other embodiments, more than about 20%, 30%, 40% and as much as about 60%, 70% or 80% of the total injected dose of the lipid particles is present in plasma about 8, 12, 24, 36, or 48 hours after injection. In certain instances, more than about 10% of a plurality of the particles is present in the plasma of a mammal about 1 hour after administration. In certain other instances, the presence of the lipid particles is detectable at least about 1 hour after administration of the particle. In certain embodiments, the presence of a therapeutic agent such as a nucleic acid is detectable in cells of a tumor such as a solid tumor at about 8, 12, 24, 36, 48, 60, 72 or 96 hours after administration. In other embodiments, downregulation of expression of a target sequence by an interfering RNA (e.g., siRNA) is detectable at about 8, 12, 24, 36, 48, 60, 72 or 96 hours after administration. In yet other embodiments, downregulation of expression of a target sequence by an interfering RNA (e.g., siRNA) occurs preferentially in tumor cells. In further embodiments, the presence or effect of an interfering RNA (e.g., siRNA) in cells at a site proximal or distal to the site of administration or in cells of a tumor is detectable at about 12, 24, 48, 72, or 96 hours, or at about 6, 8, 10, 12, 14, 16, 18, 19, 20, 22, 24, 26, or 28 days after administration. In additional embodiments, the lipid particles are administered parenterally or intraperitoneally.

The compositions, either alone or in combination with other suitable components, can be made into aerosol formulations (i.e., they can be "nebulized") to be administered via inhalation (e.g., intranasally or intratracheally). Aerosol formulations can be placed into pressurized acceptable propellants, such as dichlorodifluoromethane, propane, nitrogen, and the like.

Vector Generation Considerations

The SBI LV shRNA particles are made in small research batches with kits obtained from SBI in the laboratory. Total yield is 10¹¹ particles with a 1 % TU rate. The shRNA particles are produced in research grade batches in 100 mm Petri dishes with total supernatants of less than 100 ml that are concentrated by high density

ultracentrifugation and will have a 70% TU rate as expected from prior preparations. The total particle yield per lot is 10¹¹ to 10¹² vector particles. Lots are less than 10 ml for either preparation.

Methods of Using the Oligonucleotide or the Viral Vector for Treating Cancer

The compositions described above can be formulated into pharmaceutical compositions for treating a cancer. In every instance these compositions will comprise an oligonucleotide that is capable of inhibiting expression of NANOGP8 while not interfering with expression of NANOG. The efficacy of pharmaceutical compositions will be evaluated in terms of their ability to deliver the oligonucleotide to cancer cells.

Cancer cells of interest preferably include cancers derived from epithelial cells, more preferably gastric, colorectal (CRC), breast, glioblastoma, cervical cancers.

For pharmaceutical compositions comprising double stranded siRNA, a suitable lipid delivery system can be employed.

For pharmaceutical compositions comprising viral vector, the ability of the virus to infect cancer cells, more preferably colorectal cancer cells, can be measured. The virus will contain an oligonucleotide comprising the sequences of SEQ ID NOS:3-44, more preferably selected from the group consisting of SEQ ID NOS: 3, 8, 34, and 37.

Methods of treating cancer in patients in need thereof using the pharmaceutical compositions described above are also readily demonstrated by those of skill in the art. These methods preferably include showing inhibition of cancer cells having a malignant phenotype, more preferably inhibiting cancer cells capable of metastasizing, also preferably inhibition of colorectal cancer cells.

EXAMPLES

Example 1 Spheroid Formation is Associated with Metastatic Potential

The ability of single CRC cells to form spheres, a hallmark of malignant potential, was tested in six cell lines in serum-free medium in ultra-low attachment plates in individual wells. Individual wells were scored to identify those with a single cell within the first 24 hr of plating. Spheroids that contained fifty or more cells were then scored in the single cell wells 7 to 14 days later. All CRC lines formed spheroids at frequencies that ranged from 1 to 56% of single cells plated (FIG. 2). The results for six human CRC lines are tabulated below and correlate with each line's malignant phenotype are summarized in Table 1.

TABLE 1. Characteristics of Human CRC lines

LS 174T + 95% 57%

These results show that spherogenicity was associated with experimental metastatic potential. The formation of spheroids is a measure of the "sternness" of a CRC line in comparison to growth in monolayer culture, the basal condition for that cell line. Example 2 NANOG Expression Increases During Spheroid Formation

Monolayer and spheroid cultures of these six cell lines were analyzed for the expression of embryonic stem cell transcription factors and CD44, CD133 or CD166 transcripts by qRT-PCR. Relative expression of NANOG transcripts was consistently increased in spheroids 10 to 20-fold compared to monolayer cultures except in KM 12c while the relative expression of OCT4 was 10-fold and the SOX2 expression varied between 1 and 63-fold in spheroids compared to monolayer cultures. In contrast, the expression of CD44, CD 133 or CD166 was not significantly increased in spheroid cultures compared to monolayer except in KM 12c, CX-1 and LS 174T. These results show that NANOG is activated as CRC transition to vertical growth to form spheroids. NANOG expression is consistently increased in all spheroid cultures compared to the companion monolayer culture. The expression of OCT4 is also increased in spheroids but generally less than NANOG while SOX2 expression is variable and sometimes much higher than NANOG or OCT4 expression. Another CRC line, KM-12c, expresses a 200- fold increase in NANOG but a 300-fold increase in SOX2 levels.

The results indicate that NANOG gene expression is consistently increased in spheroids compared to the expression in that line's monolayer culture while the expression of the other core embryonic genes OCT4 and SOX2 is more variable. These results suggest that NANOG may be a useful target for therapy.

Example 3 NANOGP8 is the Dominant Form of NANOG Expressed in CRC Lines

To determine which NANOG gene product is expressed in human CRC required identifying a unique NANOGP8 site around one of the 5 single nucleotide changes that is amenable to endonuclease digestion. A search of databases revealed that the only specific commercially available endonuclease that was predicted to cut NANOGP8 cDNA but not NANOG was 143 nt 3' downstream to the ATG by AlwNI which cuts at the following sequence. The restriction endonuclease AlwNI identifies a palindromic hexanucleotidc sequence in NANOGP8 but not NANOG at position 144 relative to the translational start site.

Analyses of cDNA from the CRC lines by digestion with AlwNI show that

NANOGP8 is expressed in both the monolayer and spheroid cultures of the six CRC lines. The results were confirmed by direct sequencing of the cDNA. NANOGP8 was found to be more than 84% of the NANOG expressed in monolayer cultures of CX-1 and Clone A, and more than 94% of NANOG expressed in spheroids of these cells. On this basis, NANOGP8 is the dominant form of NANOG expressed in CRC cell lines.

Example 4 NANOGP8 is the Dominant Form of NANOG Expressed in Patient CRC

To determine whether NANOG and NANOGP8 are expressed in CRC in patients, ten de-identified samples of colorectal carcinoma resected from patients treated at the NIH Clinical Center were analyzed for expression of NANOG and NANOGP8 transcripts and protein. RT-PCR was followed by cutting the cDNA with the restriction endonuclease AlwNI that specifically cuts NANOGP8 but not NANOG. Results of this analysis are presented in Table 2.

TABLE 2. Summary of results in liver samples

Total NP8 Ng Total

Ng+ NP8+ only only Ng- Total

Tumor 8 6 2 1 2 10

Adjacent 4 1 0 3 6 10 normal The results indicate that NANOG transcripts are found in eight of ten tumor specimens and in four adjacent normal liver sections. Six of the eight tumor specimens contained NANOGP8 transcripts. Two of the six specimens only contained NANOGP8, four contained both NANOG and NANOGP8 and one only NANOG. All results were confirmed by Sanger sequencing.

Based on these measurements, approximately 80% of clinical CRC metastases express a NANOG family member with 75% of those expressing NANOGP8.

Example 5 NANOGP8 Promoter Activity Increases During Spherogenicity

As shown above, levels of NANOGP8 transcripts are increased in spheroids of the CRC cell lines Clone A and CX-1 cultured in SFM. The promoter activity of the

NANOGP8 gene was measured as CRC cells transition from a monolayer or two- dimensional culture to a spheroid or three-dimensional culture. The promoter activity of the NANOGP8 gene was compared between monolayer and spheroid CRC cells. A lentiviral NANOGP8 promoter GFP reporter was used to analyze cultures of CX-1 and Clone A. The results showed that monolayer cultures had low levels of GFP reporter expression compared to expression in spheroids of CX-1 and Clone A. In addition, the ubiquitous phosphoglycerate kinase (PGK) promoter was transduced into CX-1 and Clone A as a control, in which monolayer and spheroid transcriptional rates were found to be the same. In contrast, a 6- to 14-fold increase in GFP⁺ cells was found in spheroids compared to monolayers for Clone A and CX-1. This is similar to the relative increase in transcripts assessed by qRT-PCR during transition from 2-D monolayers to 3-D suspension culture.

Accordingly, these results show that the increase in the levels of NANOG transcripts was due to increased NANOG promoter activity.

Example 6 NANOGP8 is Expressed as a Protein in CRC Cells

To investigate if NANOGP8 protein is expressed in CRC cells, the ten clinical samples of metastases and adjacent liver were stained with a commercially available NANOG antibody, which recognizes both NANOG and NANOGP8. Analysis of the immunofluorescent antibody (I FA) staining indicates that staining is cytoplasmic in CRC and present only in tumors or adjacent liver that contained NANOG or NANOGP8 transcripts. In contrast, a control specimen of a human germ cell tumor has the expected intranuclear location for NANOG.

To identify the NANOGP8 protein directly in CRC cell lines, tandem MS/MS was used. Extracts of CRC cell lines were immunoprecipitated, isolated by SDS-PAGE, subjected to in gel tryptic digestion and then MS/MS. Four NANOG-related peptides (KTWFQNQRM, KYLSLQQMQELSNILNLSYKQ, KKEDKVPVKK, and

KGKQPTSAENSVAKK) were identified in extracts from Clone A overexpressing

NANOGP8. The last peptide is unique to NANOGP8, which includes the shift from Lys (K) in NANOG to Asn (N) in NANOGP8 at codon 82. This amino acid change was confirmed by gene sequencing. NANOG/NANOGP8 proteins were not identified in extracts of CRC cell lines without overexpression by transduction presumably because CRC cell lines express low levels of endogenous proteins.

Example 7 Allele-Specific shRNA-Mediated Inhibition of NANOGP8

Several fatal neurodegenerative human diseases are caused by single nucleotide polymorphisms (SNPs) that cause mutant proteins to be formed through alternative splicing as well as insertion of CAG-repeats into critical structural proteins. Thus, neuroscientists have developed strategies for designing inhibitory RNAs that preserve the function of the wild type protein but inhibit the expression of mutant genes that often differ by a single SNP. Using principles derived from these studies (Huang, et al. 2009, Nucl Acids Res 37:7560-69; and Dykxhoorn, et al. 2006, PNAS 103:5953-58), a series of allele-specific shRNAs that target NANOGP8 expression have been designed by us and are here partly evaluated both for specificity of inhibition as well as inhibition of spherogenicity. Allele-specific inhibitory shRNAs described herein are driven by the H1 promoter and contain a CMV driven GFP fluorescent protein that can be used as a reporter in Lentiviral vectors from Systems Biosciences, Inc (SBI). These structures are labeled either NANOGP8- or Ng- for shRNAs that target either NANOGP8 or NANOG respectively. For comparison the commercially available shRNAs from Sigma are also shown that were based on the original designs by members of the RNAi Consortium that is led by investigators primarily from the Broad Institute using their design algorithm.

The shRNAs used to target NANOG or NANOGP8 are listed in Table 3, below. The commercially available shRNAs designated by the prefix TRCN (Sigma-Aldrich) contain shRNA driven by a U6 promoter as well as a puromycin resistance element driven by a CMV promoter. The Np8-, Ng- or Control (CONT) lentivectors shRNA expression vectors of SEQ ID NOS: 1-44 are pPS-H 1-LCS-GFP and pPS-H1-LCS-RFP; and overexpression vector is pPS-EF1-LCS-T2A-RFP (System Biosciences, Inc.).

Packaging plasmids from the Lab of Experimental Carcinogenesis: pVSV-G and pPACK- GAG/REV. HEK293TN cells were used for packaging according to the protocol from SBI. Asterisk* denotes the lenti viral particle that was used to construct shNANOG transduced CRC. shOCT4 and shSOX2 were selected from a commercial source (Sigma). Both the SBI and Sigma lentiviruses are HIV-based lentiviruses that produce replication deficient pseudoparticles.

TABLE 3: Allele-Specific Inhibitory shRNAs to NANOG

SEQ ID

Name/Cat # Sequence NO:

NP8-1 -1 GAGGCAGCAGAGACCGCTGCATGCACTTCCAGCCA 3

NP8-1 -2 TCTGACAGGAAGTGGCTGGAAGTGCATGCAG 4

NP8-1 -3 CTTCCTGTCAGATGGCTGGAAGTGCATGCAGTTTTT 5

NP8-1 -4 CGAACAGAGAGAGACCGAAAAACTGCATGCACTTCCAGCCA 6

NP8-2-1 GAGGCAGCAGAGACCGCTGCATGCACTTCCAGCCG 7

NP8-2-2 TCTGACAGGAAGTGGCTGGAAGTGCATGCAG 8 NP8-2-3 CTTCCTGTCAGATGGCTGGAAGTGCATGCAGTTTTT 9

NP8-2-4 CGAACAGAGAGAGACCGAAAAACTGCATGCACTTCCAGCCG 10

NP8-3-1 GAGGCAGCAGAGACCGTTGTGATCCGCCCGCCTCG 1 1

NP8-3-2 TCTGACAGGAAGCGAGGCGGGCGGATCACAA 12

NP8-3-3 CTTCCTGTCAGACGAGGCGGGCGGATCACAATTTTT 13

NP8-3-4 CGAACAGAGAGAGACCGAAAAATTGTGATCCGCCCGCCTCG 14

NP8-4-1 GAGGCAGCAGAGACCGATCTAATACGAGTTTGGATA 15

NP8-4-2 TCTGACAGGAAGTATCCAAACTCGTATTAGAT 16

NP8-4-3 CTTCCTGTCAGATATCCAAACTCGTATTAGAT TTTTT 17

NP8-4-4 CGAACAGAGAGAGACCGAAAAAATCTAATACGAGTTTGGATA 18

NP8-5-1 GAGGCAGCAGAGACCGATCTAATACGAGTTTGGATG 19

NP8-5-4 CGAACAGAGAGAGACCGAAAAAATCTAATACGAGTTTGGATG 20

NP8-6-1 GAGGCAGCAGAGACCGATGAGTACTACTTTAGTTG 21

NP8-6-2 TCTGACAGGAAGCAACTAAAGTAGTACTCAT 22

NP8-6-3 CTTCCTGTCAGACAACTAAAGTAGTACTCATTTTTT 23

NP8-6-4 CGAACAGAGAGAGACCGAAAAAATGAGTACTACTTTAGTTG 24

NP8-7-1 GAGGCAGCAGAGACCGAACAAAGCACATCTTGCCAGGA 25

NP8-7-2 TCTGACAGGAAGTCCTGGCAAGATGTGCTTTGTT 26

NP8-7-3 CTTCCTGTCAGATCCTGGCAAGATGTGCTTTGTTTTTTT 27

NP8-7-4 CGAACAGAGAGAGACCGAAAAAAACAAAGCACATCTTGCCAGGA 28

NP8-8-1 GAGGCAGCAGAGACCGAACAAAGCACATCTTGCCAGGG 29

NP8-8-2 TCTGACAGGAAGTCCTGGCAAGATGTGCTTTGTT 30

NP8-8-3 CTTCCTGTCAGATCCTGGCAAGATGTGCTTTGTTTTTTT 31

NP8-8-4 CGAACAGAGAGAGACCGAAAAAAACAAAGCACATCTTGCCAGGG 32

Ng-1 -1 GAGGCAGCAGAGACCGCTGCATGCAGTTCCAGCCA 33

Ng-1 -2 TCTGACAGGAAGTGGCTGGAACTGCATGCAG 34

Ng-1 -3 CTTCCTGTCAGATGGCTGGAACTGCATGCAGTTTTT 35

Ng-1 -4 CGAACAGAGAGAGACCGAAAAACTGCATGCAGTTCCAGCCA 36

Ng2-1 GAGGCAGCAGAGACCGCTGCATGCAGTTCCAGCCG 37 Ng2-2 TCTGACAGGAAGTGGCTGGAACTGCATGCAG 38

Ng2-3 CTTCCTGTCAGATGGCTGGAACTGCATGCAGTTTTT 39

Ng2-4 CGAACAGAGAGAGACCGAAAAACTGCATGCAGTTCCAGCCG 40

Cont-1 GAGGCAGCAGAGACCGTAGCGACTAAACACATCAA 41

Cont-2 TCTGACAGGAAGTTGATGTGTTTAGTCGCTA 42

Cont-3 CTTCCTGTCAGATTGATGTGTTTAGTCGCTATTTTT 43

Cont-4 CGAACAGAGAGAGACCGAAAAATAGCGACTAAACACATCAA 44

CCGGGCTGCTAAGGACAACATTGATCTCGAGATCAATGTTGTCCTTA

TRCN004884 45

GCAGCTTTTT

CCGGGCTTTGAAGCATCCGACTGTACTCGAGTACAGTCGGATGCTTC TRCN004885* 46

CCGGCTGTAAAGAATCTTCACCTATCTCGAGATAGGTGAAGATTCTTT TRCN004886 47

ACAGTTTTT

CCGGCCTGGAACAGTCCCTTCTATACTCGAGTATAGAAGGGACTGTT TRCN004887 48

CCAGGTTTTT

CCGGCCTAAACTACTCCATGAACATCTCGAGATGTTCATGGAGTAGT

TRCN004888 49

TTAGGTTTTT

Double stranded siRNA effective in selectively knocking down NANOG parent gene expression include the following sequences.

Ng-siRNA-1 :

Sense:

SEQ ID NO: 50

5'-CTGCATGCAGTTCCAGCCA-3'

Antisense:

SEQ ID NO: 51

5 -TGGCTGGAACTGCATGCAG-3'

Ng-siRNA-2:

Sense:

SEQ ID NO: 52

5'-CTGCATGCAGTTCCAGCCG-3' Antisense:

SEQ ID NO: 53

5 -TGGCTGGAACTGCATGCAG-3'

Double stranded siRNA effective in selectively knocking down NANOGP8 pseudogene expression include the following sequences.

NP8-SIRNA-1 :

Sense:

SEQ ID NO: 54

5'-CTGCATGCACTTCCAGCCA-3'

Antisense:

SEQ ID NO: 55

5 -TGGCTGGAAGTGCATGCAG-3'

NP8-SIRNA-2:

Sense:

SEQ ID NO: 56

5'-CTGCATGCACTTCCAGCCG-3'

Antisense:

SEQ ID NO: 57

5 -TGGCTGGAAGTGCATGCAG-3' Transduction with the herein described allele-specific shRNAs of both monolayer cultures as well as cells in suspension has been achieved with 50 to 90% efficiency as defined by GFP fluorescence.

Example 7 Modulating NANOG Expression Changes Spherogenicity of CRC Cells

If therapy to inhibit NANOG expression by lentiviral-delivered shRNA is to be successful, then it is important to demonstrate that NANOG shRNA can be delivered to cells in 3-D colonies. The main CRC lines growing in suspension can readily be transduced by the vectors described herein.

Screening of constructs shows that Np8-1-1 is an active candidate since it inhibits the spheroid formation of CX-1 , Clone A and LS 174T cells (Table 4).

TABLE 4. Summary of Single Cell Spherogenicity Assays Single Cell/

Cell Line Construct Spheroid Cluster % P

CX-1 shNANOG 10 120 8 0.004

pLKO.1 31 72 30 NS

Ng-1-2 13 10 57 0.04

Np8-2-2 12 20 38 NS

Ng-2-1 1 88 1 O.001

Np8-1-1 1 99 1 O.001

Parental 24 77 24 —

LS 174T shNANOG 12 32 27 0.0031

pLKO.1 15 15 50 NS

Ng-1-2 9 28 24 0.02

Np8-2-2 13 20 39 NS

Ng-2-1 1 31 3 O.001

Np8-1-1 1 65 2 O.001

Parental 37 28 57 —

Clone A shNANOG 1 139 0.7 0.0003

pLKO.1 5 79 6 NS

Ng-1-2 11 9 55 0.001

Np8-2-2 10 12 45 0.008

Ng-2-1 0 47 0 0.06

Np8-1-1 0 56 0 0.026

Parental 9 64 12

CRC cells were plated at 0.6 cells/well in serum-free medium. Wells were scored within 24 hours for single cells. All wells containing a single cell were then re-evaluated at least 9 days after plating for formation of spheroids (≥50 cells in a compact spherical assembly). Spheroid is the number of single cells that formed spheroids; Single

Cell/Cluster is the number of single cells that did not form spheroids. P value is calculated by Fisher S exact test corrected for multiple comparisons by Bonferroni correction. Each shRNA has been tested in at least two separate assays except for the Clone A Ng-1-2 and Np8-2-2 lentivectors. NP8 shRNA does not inhibit NANOG gene expression significantly in a human cell line that expresses NANOG without NANOGP8. NANOGP8-1 (Np8-1-1) causes a ^"50% decrease in NANOG levels in CX-1 cells but no significant decrease in levels in PA-1 cells.

Allele-specific inhibitory shRNA inhibits the expression of its specific target but not that of either the wild type, parental, or a gene which is NANOG. As shown below, Np8-1- 1 inhibits the expression of NANOG by RT-PCR in CX-1 cells that produce a

preponderance of NANOGP8. However, Np8-1-1 does not inhibit expression of parental NANOG by RT-PCR in human PA-1 embryonic carcinoma cells that produce only parental NANOG and no other pseudogenes whose mRNA might be amplified. This confirms that Np8-1-1 is an allele-specific inhibitor.

Whole transcriptome analyses of CX-1 and Clone A in monolayer and spheroid culture shows that NANOGP8 has at most 200 reads even in spheroids, NANOG may have less than 10 in transcriptomes that contain 10-40 million reads. As a result, a gene therapy approach to targeting NANOGP8 will block the progression of CRC once it has metastasized. While these results identify a promising lead candidate to inhibit CX-1 and possibly reduce the growth of the more aggressive LS 174T, other candidates may be more active and/or specific. Inhibition of this low abundance gene will have profound effects in decreasing metastatic potential by both inhibiting cell proliferation and inducing apoptosis.

Total NANOG mRNA expression was inhibited using lentiviral-delivered shRNA technology to examine the functional relevance of NANOG in spherogenicity in CRC cell lines. The efficiency of shRNA targeting of NANOG was confirmed by RT-PCR assay and immunoblotting. Spherogenicity was measured by plating single CRC cells in serum-free medium and then scoring spheroids of fifty or more cells after at least nine days of culture. Inhibiting NANOG transcripts significantly inhibited formation of spheres in both Clone A and CX-1 by 40% to 90% compared to untreated parental cells and cells transduced with the empty vector pLKO.1. A 50% inhibition was also observed in LS 174T cells transduced with shNANOG. The data indicate that inhibition of NANOG consistently inhibits spherogenicity in the three CRC lines (FIG. 3). Single cell spherogenicity, tumorigenicity and in one CRC line metastatic potential have been assessed.

Inhibition of NANOG by shRNA consistently inhibits spherogenicity in each CRC line while shRNA to SOX2 and OCT4 also significantly inhibits the spherogenicity of LS 174T. However, inhibition of either OCT4 or SOX2 increases the spherogenicity of Clone A and CX-1. Also NANOG appears to be the dominant gene because in CX-1 and Clone A inhibition of NANOG inhibits SOX2 and OCT4 but not vice versa. This is consistent with the function of these genes in early implantation embryos. NANOG may inhibit tumor growth through direct inhibition of WEE1 gene and phosphorylation of WEE1 and cdc2^P34 (CDKN1). Alternatively, NANOG may decrease resistance to apoptosis by inhibiting anti- apoptotic genes such as BCL-2 and MLL in CRC.

In both Clone A and CX-1 cells, when NANOG or NANOGP8 are overexpressed there is an increase in the formation of spheroids from single cells compared to parental untreated cells. This is ^"12% for Clone A and about 23-25% for CX-1. This is in contrast to results above that show that spherogenicity is decreased by inhibiting NANOG. Thus, spherogenicity is inhibited by inhibiting NANOG and increase by increasing NANOG or NANOGP8. In every instance, results are based on a single cell spherogenicity assay in serum free medium in which each well is scored for the presence of a single cell, and spheroids are separately assessed 9+ days later. Spheroids are defined as cell masses of fifty or more cells.

In contrast, when Clone A and CX-1 cells were transduced with lentiviral vectors expressing full length NANOG, the increased NANOG expression increased

spherogenicity in Clone A by about 9-fold compared to untreated or vector controls but not in CX-1.

Taken together, these results show that inhibition of NANOG gene expression inhibits growth of human CRC xenografts in vivo. These results indicate that NANOG expression has an important role in the capacity of single CRC cells to form spheroids and also indicates that the suspension cultures in serum-free medium may promote the enrichment of tumor-initiating cells with characteristics of sternness.

Example 8 Inhibition of 3-D Growth by Transduction NANOG/NANOGP8 shRNA

It was important to determine how the LV shRNA modulates apoptosis in 3-D culture. Anoikis— apoptosis caused by suspension culture— is dependent on activation of Caspase 8. Caspase 3, 8 and 9 activities were measured in Clone A and CX-1 treated with LV shRNA at 4 days of suspension culture, 3 days after LV treatment. shRNA to NANOG and NANOGP8 caused an increase in the activity of Caspase 3 in both CX-1 and Clone A cells. shRNA Np8-1 also increased the activity of Caspase 8 or 9 in Clone A or CX-1 , respectively. Cells that were adjacent to transduced cells but not expressing the MD also had similar levels of Caspase 9 activity, pointing to an effect of transduced cells on neighbors. The results confirm that Caspase 9 and the intrinsic pathway that uses mitochondrial amplification in apoptosis is important for apoptosis mediated by shNp8-1 and shNG-1. shRNA to NANOGP8 and NANOG induced apoptosis and 3-D growth by the activation of the intrinsic pathway. This is distinct from the mechanism of apoptosis in these cells induced by anoikis which depends solely on the extrinsic pathway of apoptosis.

Example 9 Inhibition of NANOG Inhibits Tumorigenicity and Metastasis

CRC were injected subcutaneously in NOD/SCI D mice in cell dilutions of 10³-10⁵ cells per mouse. At 10³ and 10⁴ cells per mouse CX-1 cells transduced with shNANOG that inhibits both NANOG and NANOGP8 expression had fewer tumors than did either the parental cell line or the pLKO.1 empty vector control. In contrast, the shNANOG- transduced Clone A cells were less tumorigenic than the pLKO.1 control but were similar to the tumor growth rates in the parental Clone A cells that are more weakly tumorigenic than CX-1 cells (FIG. 4). Different constructs were tested in NOD/SCID mice by subcutaneous injection of 10⁵ cells to assess tumorigenicity at a consistent cell concentration between experiments. The results showed that growth of shNANOG transduced CX-1 and Clone A cells was slower than the pLKO.1 control since the median number of days to palpable tumor is increased by 16% for Clone A and not reached for CX-1 (Tables 5 and 6; and FIG. 7).

TABLE 5

Median (days)

Clone A CX-1 pLKO.1 25 33~ shNANOG 29 >50

NANOG 23 18

TABLE 6

% Tumor free

Clone A CX-1 pLKO.1 0 0~

shNANOG 40 50

NANOG 0 0 In addition, 40% of the mice in the shNANOG Clone A groups remained tumor free during the time of the experiment. Thus, shNANOG decreased growth of CX-1 but had minor effects on Clone A.

Another test of the effect of shRNA to NANOG on the malignant phenotype involves experimental metastasis since liver colonization after intrasplenic injection of viable CRC cells is associated with recurrence in patients operated on for cure. shNANOG transduced CX-1 cells failed to form grossly visible or microscopic hepatic liver colonies compared to either the untreated parental CX-1 cells or CX-1 cells transduced with the control vector pLKO.1 (FIG. 5). These results suggest that inhibition of NANOG function may be important to preventing progression of CRC.

Clone A is less tumorigenic than CX-1 and shNANOG does not decrease the rate of tumor takes compared to the parental line although shNANOG developed fewer tumors than the vector control. However, shNANOG significantly slowed the rate of growth of Clone A tumors compared to either the vector control or the parental line (FIG. 4E). While shSOX2 stimulated the appearance of Clone A tumors, shOCT4 delayed the appearance tumors of Clone A. shNANOG in CX-1 inhibits tumor growth in both the parental and the vector control, whereas both shOCT4 and shSOX2 shorten the time to appearance of tumors in CX-1 (FIG. 4B). LS 175T is considerably more tumorigenic than Clone A or CX- 1 since the parental cell line has a shorter mean time to appearance (21.2 days) than Clone A (25 days) and CX-1 (30.4 days). shNANOG does not inhibit tumor appearance compared to the parental line or vector control. In contrast to their activity in the other CRC, both shOCT4 and shSOX2 inhibited the formation of LS 174T tumors in shSOX2 (FIGS. 4C and 4D). The results indicate that shNANOG significantly decreases the rate of growth of tumors compared to the vector control or the parental cell line.

Example 10 NANOG Inhibition Blocks Metastasis of CRC Cells

One aspect of this description is a therapy to effectively inhibit the formation of liver metastases in patients with CRC. Recurrence in patients operated upon for cure is associated strongly with the ability of their freshly isolated CRC to form liver colonies after intrasplenic injection. Clone A forms too few liver colonies after intrasplenic injection to study. In contrast, NOD/SCID mice receiving shNANOG transduced CX-1 did not develop microscopic or grossly visible liver colonies compared to mice that received CX-1 cells that were either untreated parental cells or cells transduced with the empty vector pLKO (FIG. 7).

These data suggest that targeting NANOG and perhaps OCT4 or SOX2 in appropriate CRC may inhibit growth of CRC even at common sites of metastases.

Treatment might be tailored to individual CRC by typing the ratio of NANOG, SOX2 and OCT4 in the spheroid-forming cells within a CRC. A lentiviral vector was found to be effective as a potential gene therapy to deliver shRNA encoding NANOG. This RNA consistently inhibits tumor growth in vitro, increased the time to tumor appearance in 2 of 3 CRC lines and blocked experimental metastasis in a moderately aggressive CRC. Example 1 1 Designing and Constructing Adenoviral Vectors for Treatment of Tumors

The inventors have assessed a replicating oncolytic viral vector that may continue to infect and kill cancer cells either through direct lysis or through the effects of shRNA or another genetic payload. Different oncolytic adenoviruses are considered useful for these therapeutic RNAi constructs. Adenovirus type 5 is a conditionally replicating adenovirus (CRAd) which may proliferate in tumor cells that have lost function of p53. Early trials demonstrated that the virus replicated in patients with head and neck squamous cell carcinoma but that host immunity limited the utility of systemically administered CRAd. Also, type 5 adenovirus is hepatotoxic. Therefore, the inventors have designed

adenoviruses with a chimeric fiber that combines a serotype 3 knob

with a type 5 fiber to create a virus designated Ad5/3. Such chimeras have reduced hepatotropism (and hepatotoxicity). This Ad5/3 converts the receptor for the virus from the coxsackie adenovirus receptor (CAR gene) to desmoglein 2 (DSG2), a protein in hemidesmosomes expressed in epithelia and endothelial cells. The viruses also contain Hi -driven shRNAs (the control shNEG and the allele-specific shNG-1 and shNp8--1 along with the copGFP reporter driven by a CMV promoter) that were present in lentivirus constructs. These wild type viruses replicate in and lyse human cells but not mouse cells. The viruses with the Wild Type (WT) promoter are Ad5/3-E1 aWT-shNEG, Ad5/3-E1 aWT - shNp8 (also called AdNP8 or NP8) and Ad5/3-E1 aWT (this without shRNA or GFP reporter). The adenovirus construct and the wild type genomes (Ad5/3 WT) are depicted in Figure 8.

To make a conditionally replicating adenovirus with a 5/3 fiber ("CRAd5/3") the inventors replaced the Ψ and proximal part of the E1 a gene with the proximal 1 Kb of the NANOGP8 promoter. This minimal promoter contains consensus binding sites for a number of transcription binding sites that drive expression of NANOGP8 and drive the rest of the E1a viral gene to initiate replication of the virus in cells that express NANOGP8. Therefore, a CRAd may only replicate in those cells that actively express NANOGP8. NANOGP8 is not expressed in normal cells, with the exception of some smooth muscle cells. The inventors' Ad5/3 WT and CRAd5/3 viral constructs infect both mouse and human cells but only replicate in and lyse human cells. The CRAd is Ad5/3-E1 aNp8 - shNp8 (hereinafter termed 'AdNP2' or 'NP2').

Example 12 Testing Adenoviral Anti-Cancer Vectors

The Ad5/3's with wild type promoters were tested first with Ad5/3- E1 aWT and Ad5/3-E1 aWT-shNp8 (hereinafter termed 'AdNP8') in Clone A and CX-1 at various dilutions of viral particles per CRC cell. As expected, increasing the titer of the virus increases the infection rate as represented by the GFP fluorescence (Figures 9A and 9B). Even at rates of only 1 or 3 viral particles per cell the Ad5/3 infects occasional cells that will eventually lead to infection of the majority of CRC cells. Also a transduction unit averages 13 viral particles per cell for all of the constructs made.

Representative results are shown in Figures 10A and 10B demonstrating the utility of the replicating virus. In the inventors' LV experiments, a maximal decrease in CRC cell survival of approximately 50% inhibition was achieved at 3 days. Because the LV is non- replicating, the un-transduced cells will eventually grow even though the regrowth potential of CRC cells treated in 3D cultures is decreased by about 30%. In contrast, exposure to Ad5/3-E1aWT or Ad5/3-E1aWT-shNp8 (AdNP8) decreases cell survival over 10 days to <20% of that of the untreated control cells (Figures 10A and 10B). It should be noted that these cultures are remarkable because they are 3D organoids cultured in aqueous serum-free medium and as such mimic aspects of growth in vivo.

A major advantage of the replicating type of Ad5/3 is that as cells lyse they release up to 10,000 mature viruses that then infect neighboring cells. Wild type Ad5/3 will only replicate in human xenografts in murine preclinical models, but both the wild type Ad5/3's and the CRAd amplified in CRC growing in CX-1 in suspension, indicating the potential for oncolysis over time.

Example 13 Modeling the Effect of Mouse Stroma on Ad5/3 Function

The inventors tested whether murine fibroblasts affected the ability of Ad5/3 viruses to kill CRC cells. 3T3 fibroblasts were used as stromal cells where Ad5/3 attaches but does not replicate. This is confirmed in co-culture experiments with CX-1 and LS 174T in which 500 CRC cells were incubated with 100, 250 or 500 3T3 cells, or without 3T3 cells, or no CRC with only 500 3T3 cells. As shown in Figures 11 A and 11 B, 3T3 cells infected with 100 vp/cell do not express the GFP reporter that signifies viral replication. In contrast, cells infected with Ad5/3-E1 aWT-shNp8 do express the GFP reporter even at a 1 : 1 ratio of tumor cells to murine fibroblasts (Figures 11A and 11 B). These results suggest that while murine fibroblasts act as a sponge for the virus, the virus will infect susceptible cells and multiply. While infectivity of CX-1 was not affected by increasing concentrations of 3T3, it appears that the relative infection rate of LS 174T decreases with increasing 3T3 concentration. This may occur because LS 174T forms tight junctions that may 'hide' the DSG2 as it forms spheroids that exclude 3T3 fibroblasts. This exclusion has been confirmed by confocal imaging (Data Not Shown). It is not clear that this may occur in vivo in preclinical models or in patients.

When the co-cultures were analyzed with the post assay viability dye Alomar Blue, there was an interesting effect of 3T3 on survival of CX-1 and LS 174T (Figures 12A and 12B). The results indicate that in the absence of 3T3 cells both CX-1 and LS 174T are quite sensitive to Ad5/3-E1 aWT-shNp8 but less sensitive to Ad5/3-E1 a that lacks the shRNA (Figures 12A and 12B). In addition, it is clear that the efficacy of the Ad5/3's decreases when the ratio of tumor cells to 3T3 cells drops to 2.5 or 1 tumor cell to 1 3T3 cell. These results suggest that Ad5/3's may be active in epithelial cancers where the stromal fibroblast to tumor cell ratios is 1 :4 or lower. These results also suggest that the shRNA to NANOGP8 ratio may increase the efficacy of the Ad5/3 (Figures 12A and 12B). The other consideration is that human stroma will support Ad5/3 replication, which may further augment tumor control.

Example 14 Development of the CRAd5/3-E1 aNp8-shNp8 (AdNP2)

Development of similar CRAds with the NANOGP8 promoter for the other shRNAs have not been successful, despite multiple attempts. It is likely that because the adenoviral genome is relatively small, the replacement of the wild type promoter with exogenous DNA may be at the limit of the genome's ability to handle foreign DNA.

Nonetheless, CRAd5/3-E1 aNp8-shNp8 is active, replicates in 293T17 cells and CRC and is capable of killing human CRC as shown below. The first test of the ability of this virus to infect CX-1 and then release active viral particles that lyse fresh second cultures of either CX-1 or 293T17 cells is shown (Figure 13). These data suggest that the CRAd amplify in human CRC and then replicate for secondary and subsequent infections.

The inventors further assessed the ability of the CRAd to infect CRC as well as to induce cell death by comparing the CRAd to AD5/3-E1aWT-shNp8 for transmission between cells and induction of Annexin V positivity over the course of 7 days in a CX-1 suspension culture. The NANOGP8 promoter driven CRAD increased its infectivity 6-fold at 30 vp/cell and 2-fold at 100 vp/cell while Wild Type E1 a-driven Ad5/3 increased its infection rate by 1.5-fold at 30 vp/cell and actually decreased its infection rate at 1000 vp/cell (Figures 14A and 14B). The CRAd induced Annexin V positivity 10 days after infection but the level is clearly lower than what is observed with either of the wild type Ad5/3s (Figure 14C).

When the CRAd was compared for inhibition of growth in the suspension culture model, it inhibited Alamar Blue metabolism significantly, but not as much as the two wild type Ad5/3 viruses (Figure 15). These data indicate that lytic Ad5/3 viruses efficiently kill CX-1 cells. The CRAd5/3 was less efficient and active only at the 1000 viral particles per cell (approximately 30 pfu per cell). Thus, in the in vitro suspension culture system, the CRAd5/3 was not as active as the wild type Ad5/3 and the shRNA was not more active than the control. Both results are explained by the different levels of transmission of adenovirus to secondary cells in cultures as well as reduced expression of Annexin V in the CRAd5/3 infected cells. Example 15 Selective Growth of the CRAd in Cells That Do Not Express NANOGP8 The CRAd AdNP2 was designed to respond to growth factors that stimulate expression of NANOGP8, primarily in carcinomas. We tested such growth in a XTT survival assay in which cells plated in standard 96 well microtiter plates were exposed for 7 days to dilutions of viruses. The cells that were tested and infection was determined by epifluorescence for fluorescence from Green Fluorescent Protein (GFP) in AdNP2. Lysis was determined by cytotoxic assay. Lysis is detected when it occurs at 100 viral particles or less per cell. Weak lysis is lysis that is evident only when cells were exposed to more than 1000 viral particles per cell. The expression of NANOG and/or NANOGP8 are provided in Table

Table 7. Cell Reagents Studied

Gene Expression

Cell

Line Source Tissue Carcinoma NANOG NANOGP8 Infected Lysed

3T3 Mouse Fibroblast No No No No No

WI-38 Human Fibroblast No No No Yes No

Clone A Human CRC Yes Yes Yes Yes Yes

CX-1 Human CRC Yes Yes Yes Yes Yes

LS174T Human CRC Yes Yes Yes Yes Yes

RKO Human CRC Yes Yes Yes Yes Yes

SW48 Human CRC Yes Yes Yes Yes Yes

PA-1 Human CRC Yes Yes No Weak Weak

This table also indicates whether cells were infected, and if infected, were lysed as measured by the loss of cells measured by XTT metabolism 6 days after exposure to virus. The inhibitory activity in terms of the number of viral particles per cell (vp/cell) needed to inhibit cell growth is also shown in FIG. 16).

The AdNP2 has little activity when total NANOG transcript expression is low in Clone A and

CX-1 cells in monolayer culture. The WI-38 fibroblasts are negative for NANOG and NANOGP8 gene expression and do not form spheroids in serum- free medium (data not shown). There is relatively weak activity in PA-1 endometrial carcinoma that does not express NANOGP8 but overexpresses NANOG. Interestingly, the NANOGP8

minipromoter contains NANOG consensus binding sequences and the activity of the CRAd may be related to the expression of NANOG as a result. These results indicate that the CRAd AdNP2 grows best in cells that express NANOGP8 in tumors and has minimal effect on cells that express only NANOG (PA-1). Normal diploid cells (WI-38) that lack NANOG or NANOGP8 expression are not lysed. Example 16 Inhibition of Cancer Stem Cells

Assessment of cancer stem cell activity in solid tumors requires a functional test because there are not commonly accepted membrane or cytoplasmic markers that identify 100% of the cell subset that can renew the heterogeneity of a cancer, initiate tumors in immune-incompetent mice, create spheroids from single cells in semi-solid medium or constitute the side population in a dye efflux assay. The inventors have tested the CRAd AdNP2, the wild type AdNP8 and the control nonreplicating, but oncolytic, virus Ad5/3 - GFP for activity in a cancer stem cell assay in which single tumor cells are plated in serum-free medium in individual wells of 96 well ultra low attachment (ULLA) microtiter plates. The medium is the neural stem cell basal medium supplemented with EGF, bFGF, heparin and a proprietary proliferation mix (Neurocult A, Stem Cell Technologies). When viruses were added after plating at a 100 vp/cell and the cells were allowed to proliferate for 9 days, AdNP2 significantly inhibited growth of the human CRC lines Clone A, CX -1 , RKO and SW48 by 53%, 85%,45 %, and 59%, respectively (Figure 17). In contrast, while there was a decrease in the capacity to form spheres by PA -1 it was only 24% in this line that does not express NANOGP8 (Figure 17). In addition, the immortal but otherwise normal human cells 293HEK kidney and WI -38 fetal fibroblast only formed spheroids in three percent or less of the single cells. These results indicate that the ability of single cells to form spheroids in aqueous medium is a useful measure of CSC 'sternness' and that the CRAd is active against CRC that express NANOGP8 and less active against a carcinoma that expresses NANOG but not NANOGP8.

The inventors tested the effect of the CRAd and the Ad5/3's in vivo in CX-1 CRC cells after ex vivo infection. CX-1 cells were infected for 3 hours and then admixed with uninfected CX-1 cells before injection subcutaneously so that each mouse received a total of 3 x 10⁶ cells that contained either 5% or 50% infected cells. This tests whether cells infected with an Ad5/3 or CRAd can infect adjacent uninfected cells to inhibit tumor outgrowth. As expected, both the wild type Ad5/3 with or without shNp8 inhibited the growth of CX-1 cells (Figure 18). Surprisingly, the Ad5/3 wild type promoter with shNEG weakly inhibited CX-1 growth and suggests that the control shRNA may have some effect that enhances tumor growth (Figure 18). Notably, the CRAd inhibited the growth of CX-1 when 50% of the cells were infected and had a weaker, but still significant, effect when only 5% of the cells were infected (Figure 18). These results suggest that

1) the CRAd is active in preclinical models and

2) the shNEG control may interfere with inhibition of CRC in NOD/SCID mice, but do not necessarily show that shNp8 contributes to tumor inhibition because the most active adenovirus is the wild type Ad5/3 without shRNA. The activity of the CRAd in vivo may be related to the expression of the NANOGs that may be greater than even in suspension cultures.

The inventors tested whether the adenoviruses inhibit the growth of subcutaneous human colorectal xenografts CX-1 and LS174T (Figures 19A and 19B). All viruses were active compared to untreated mice since all are oncolytic and this includes the control virus AdNEG. Further evaluation of the growth characteristics of individual mice in these experiments demonstrate that the virus inhibits the rate of growth of tumor cells more than to delay the calculated initiation of growth (Figure 20).

These results demonstrate that oncolytic adenovirus inhibits tumor growth by more than 40% and does so continuously. This supports current clinical trials of similar viruses where the viral therapy is combined with another cytotoxic or targeted therapy. Since the control virus that is oncolytic also inhibited tumor growth, it is not clear from these experiments that the shRNA to NANOGP8 contributes to the effect of the viral therapy. Also the CRAd NP2 is just as active as the viruses with the wild type E1a promoter in vivo.

Example 17 The CRAd AdNP2 Inhibits total NANOG transcript expression

The inventors have demonstrated that the CRAd is active in vivo against

NANOGP8-expressing carcinomas and is weakly inhibitory to NANOG only expressing carcinomas, and is not active against a human fibroblast line that does not express NANOG-related genes. The inventors tested whether the shRNA in AdNP2 or AdNP8 inhibits NANOG gene expression in vitro and the results show that both AdNP2 and AdNP8 do inhibit NANOG expression. First, the inventors assessed the effect of the viruses on NANOG gene expression in our standard Clone A cells. As shown in Figure 21 , AdNP2 inhibits NANOG gene expression while Ad5/3-GFP did not and AdNP8 actually increased NANOG expression.

These results demonstrate that the inventors have developed a CRAd that inhibits NANOG gene expression and is oncolytic. It has limited activity against carcinomas that overexpress NANOG but not NANOGP8, and infects but does not inhibit growth of normal human fibroblast cells. The CRAd inhibits growth of a human CRC that expresses both NANOG and NANOGP8 when infected ex vivo and also continuously inhibits growth of well-established xenografts as a single agent.

The scope of description is not limited to the examples provided above, and is hereby supplemented by the skill of the ordinary artisan. All publications cited above are hereby incorporated by such reference in their entirety into the description of this specification. The contents of all publications cited above are incorporated herein in their entirety.

Claims

What is claimed is:

I . A conditionally replicating viral vector, comprising an inhibitory RNA molecule comprising an oligonucleotide that knocks down expression of NANOGP8 wherein the oligonucleotide is 15 to 29 nucleotides in length and comprises a sequence selected from the group consisting of

SEQ ID NO:3 (5'-GAGGCAGCAGAGACCGCTGCATGCACTTCCAGCCA-3'); SEQ ID NO:37 (5 -GAGGCAGCAGAGACCGCTGCATGCAGTTCCAGCCG-3'); SEQ ID NO:8 (5'- TCTGACAGGAAGTGGCTGGAAGTGCATGCAG -3') and SEQ ID NO:34 (5'- TCTGACAGGAAGTGGCTGGAACTGCATGCAG -3').

2. The viral vector of claim 1 , wherein the vector is capable of inducing apoptosis of cancer cells.

3. The viral vector of claim 1 , wherein the vector is capable of inhibiting cancer cell proliferation.

4. The viral vector of claim 3, wherein the cancer cell proliferation is continuously inhibited.

5. The viral vector of any of claims 2 to 4, wherein the cancer cell is selected from the group consisting of gastric, colorectal, breast, glioblastoma, and cervical cancer cell.

6. The viral vector of claim 1 , wherein the vector is capable of inhibiting tumor mass.

7. The viral vector of claim 6, wherein the tumor is selected from the group consisting of gastric, colorectal, breast, glioblastoma, and cervical cancer.

8. The viral vector of any of claims 1 to 7, wherein the viral vector is packaged in a coat protein that specifically binds to malignant cells that express receptors for Ad5/3 chimeric fibritins.

9. The viral vector of claim 8, wherein the malignant cell is one of in a leukemia, a sarcoma, and a carcinoma.

10. The viral vector of claim 8, wherein the malignant cell is colorectal carcinoma (CRC) cell.

I I . The viral vector of any of claims 1 to 7, wherein the viral vector is a conditionally replicating adenovirus (CRAd).

12. The viral vector of claim 11 , wherein the CRAd is a conditionally replicating adenovirus with a 5/3 fiber (CRAd5/3).

13. The viral vector of claim 1 1 , wherein the conditionally replicating adenovirus is a type 2 or type 5 adenovirus vector.

14. The viral vector of any of claims 1 to 7, wherein the vector is a measles virus or a herpes simplex virus (HSV) vector.

15. The viral vector of any of claims 1 to 14, wherein the viral vector comprises a minimal promoter operatively linked to the inhibitory RNA molecule, wherein the minimal promoter comprises a portion of the adenovirus E1 a gene and a portion of the NANOGP8 promoter.

16. The viral vector of claim 15, wherein the vector only replicates in those cells that actively express NANOGP8.

17. The viral vector of any of claims 1 to 16, wherein the viral vector overexpresses an RNA that inhibits NANOGP8 expression.

18. The viral vector of claim 17, wherein the inhibitory RNA molecule is a double stranded siRNA.

19. The viral vector of claim 17, wherein the RNA is selected from the group consisting of SEQ ID NO:54 (CTGCATGCACTTCCAGCCA), SEQ ID NO:55

(TGGCTGGAAGTGCATGCAG), SEQ ID NO: 56 (CTGCATGCACTTCCAGCCG) and SEQ ID NO: 57 (TGGCTGGAAGTGCATGCAG).

20. The viral vector of claim 18, wherein the double stranded siRNA is SEQ ID NO:54 (CTGCATGCACTTCCAGCCA) and SEQ ID NO:55 (TGGCTGGAAGTGCATGCAG) or SEQ ID NO:56 (CTGCATGCACTTCCAGCCG) and SEQ ID NO: 57

(TGGCTGGAAGTGCATGCAG).

21. A pharmaceutical composition, comprising the viral vector according to any of claims 1 to 20, and a pharmaceutically acceptable carrier.

22. A method of treating cancer in a patient comprising administering to a patient in need of such treatment a therapeutically effective amount of a viral vector, selected from the group consisting of:

a) a conditionally replicating viral vector, comprising an inhibitory RNA molecule comprising an oligonucleotide that knocks down expression of either NANOG or

NANOGP8 wherein the oligonucleotide is 15 to 29 nucleotides in length and comprises a sequence of SEQ ID NO:3 (5 -GAGGCAGCAGAGACCGCTGCATGCACTTCCAGCCA- 3') or SEQ ID NO:37 (5'-GAGGCAGCAGAGACCGCTGCATGCAGTTCCAGCCG-3'), and,

b) a conditionally replicating viral vector viral vector, comprising an inhibitory RNA molecule comprising an oligonucleotide that knocks down expression of NANOGP8 wherein the oligonucleotide is 15 to 29 nucleotides in length and has a sequence of SEQ ID NO:8 (5'- TCTGACAGGAAGTGGCTGGAAGTGCATGCAG -3') or SEQ ID NO:34 (5'- TCTGACAGGAAGTGGCTGGAACTGCATGCAG -3') .

23. The method of claim 22, wherein the administration is via intravenous administration or direct injection at the site of disease.