www.sciencemag.org/cgi/content/full/317/5837/477/dc1 Supporting Online Material for AAV Vector Integration Sites in Mouse Hepatocellular Carcinoma Anthony Donsante, Daniel G. Miller, Yi Li, Carole Vogler, Elizabeth M. Brunt, David W. Russell,* Mark S. Sands* *To whom correspondence should be addressed. E-mail: msands@im.wustl.edu (M.S.S.); drussell@u.washington.edu (D.W.R.) This PDF file includes: Materials and Methods SOM Text Fig. S1 Table S1 References Published 27 July 2007, Science 317, 477 (2007) DOI: 10.1126/science.1142658
SUPPLEMENTAL ONLINE MATERIALS Results The altered expression of the snornas and micrornas adjacent to the AAV integration sites could be a generalized phenomenon related to the cell cycle status of the tumor cells rather than a direct result of the integrated provirus. We believe that this is unlikely. There is a region on mouse chromosome 7 that contains >50 identified or predicted snornas and micrornas, which is syntenic with the human locus on chromosome 15 associated with Prader Willi /Angelman Syndrome, and is very similar to the locus containing vector proviruses in our study. These snornas and micrornas are between the ubiquitin-protein-ligase E3A (Ube3a) and the small-nuclearribonucleoprotein-n (Snrpn) genes and are thought to be expressed from a common transcript (Supplemental References 1, 2). Twelve primer sets assayed on the microarrays were present in this locus and none were significantly over expressed (expression levels were 22 to +1.6 fold changed in the tumor samples). The Snrpn transcript that encompasses the snornas and micrornas was changed 3 and 5 fold based on two separate primer sets. Thus there is not a generalized increase in the expression of snornas or micrornas in the tumor samples, and the 10 to >100-fold increased expression of the small RNA-containing transcripts on chromosome 12 can be explained by the presence of cis-acting elements within the integrated proviral sequences. Methods Animals. MPS VII and normal mice were produced by breeding B6.C-H-2 bm1 /ByBirgus mps /+ mice and were identified by GUSB assays and PCR. The transgenic animals are on the same genetic background as the experimental animals and have been described
previously. All AAV-treated mice received intravenous injections of 1.5 x 10 11 genomecontaining particles of AAV-GUSB or AAV-GUSB Prom as described. AAV vector stocks were produced at the University of Florida Powell Gene Therapy Center and were all serotype 2. Mice undergoing bone marrow transplantation received either 5 x 10 6 unfractionated nucleated bone marrow cells with no radiation or 1 x 10 6 unfractionated nucleated bone marrow cells following 200 rads of total body radiation from a 137 Cs source at birth. Five µm-thick H&E-stained sections of liver and tumor samples from each of the 163 total animals were examined by two independent, board-certified pathologists (E.M.B. and C.V.) that were blinded to genotype and treatment. Statistically significant differences in tumor frequency between groups was determined by Fisher s exact test. Helicobacter hepaticus and bilus are known to cause hepatocellular carcinoma in mice. Representative AAV-treated mice (6 with, and 3 without HCC) were negative for H. hepaticus and bilus, as determined by assays performed at Charles River Laboratories (Wilmington, MA) using a fluorogenic PCR-based method. Cloning of junction fragments by inverse PCR. Two µg of genomic DNA purified from each liver tumor was digested with 8 units of Fat I restriction endonuclease (New England Biolabs) at 37 o C for 2 hours. Nucleic acid pellets were resuspended in 355 µl of water, 40 µl of 10X ligase buffer, and 5 µl of T4 DNA ligase (New England Biolabs, Beverly, MA). Ligation reactions were incubated at 15 o C for 18 hours. The ligase reactions were heat-inactivated and brought to 50 mm NaCl. Sixty units of Dpn I restriction endonuclease were added to each tube and reactions were incubated at 37 o C for 2 hours. DNA was resuspended in 5 µl of 10 mm Tris (ph 8.0), 1 mm EDTA and 1 µl was used as template for PCR amplification with oligonucleotides 369-(5 -
GCAGTACATCAAGTGTATCATATGCCAA) and 370-(5 - GCTATGAACTAATGACCCCGTAATTGAT). DNA fragments were excised from agarose gels and cloned using the TA cloning vector pgem T-easy (Promega). DNA sequences were obtained from these cloned PCR products. PCR quantification of vector:chromosome junctions. Integration site junctions were quantified in genomic DNA samples using a model 7500 Fast Real-Time PCR machine (Applied Biosystems) and SYBR Green JumpStart Taq ReadyMix (Sigma- Aldrich). A single vector specific primer 406-(5 - CTTGGCATATGATACACTTGATGTACT) was used in combination with primer 408- (5 - ACACAATTCCCTTTAGTTTTCAGAGT) for junctions 1 and 2, primer 413-(5 - CACAGCTCTCCTGGATGACTAAG) for junction 3, and primer 412-(5 - TTGCATTTATTCAGTCTGAGTGTTC) for junction 4. Microarray analysis. RNA was isolated, and cdna was synthesized from tumor and adjacent, normal-appearing liver tissue from animals containing AAV integration sites 1, 3 and 4. The cdna was hybridized to the MU430-2 Gene Chip (Affymetrix, Santa Clara, CA). Expression signals from tumor samples were divided by those from adjacent, normal-appearing tissue to establish fold difference in expression levels. Supplemental References 1. M. Runte et al., Hu Mol Genet 10, 2687 (2001). 2. E. Le Meur et al., Dev Biol 286, 587 (2005).
Table 1. Incidence of HCC in AAV-treated mice. Group (treatment and mouse strain) Fraction of all mice with HCC Fraction of mice 13 months old with HCC Age of onset (months) AAV2-BGUS 6/18 6/13.0034) (p=0.008) 13.5-18 MPSVII (p=0 BMT alone MPSVII BMT radiation MPSVII 1/25 1/12 18 0/12 0/8 N.A. AAV2-BGUS 9/16 7/13 001) (p=.0005) 9-18 WT (p=.0 AAV2-BGUS Prom WT Untreated WT Untreated Transgenic 2/6 (p=.105) 2/4 (p=.05) 15.5-18 5/60 4/52 13-24 0/26 0/24 N.A. The p value is a comparison with the combined BMT alone MPSVII and BMT radiation MPSVII groups. The p value is a comparison with the Untreated WT group. N.A.: Not applicable.
Supplemental Figure 1. AAV vectors. Maps of the AAV vectors used are shown with viral ITRs, CMV enhancer, chicken β- actin promoter and first intron (β-act), human β-glucuronidase gene (GUSB), and polyadenyation site (pa) indicated. A promoterless version of the AAV2-BGUS vector was also used in this study, which retains the CMV enhancer but lacks the β-actin promoter (AAV2-BGUS Prom).
5 -ITR AAV-GUSB 3 -ITR CMV β-act Human GUSB pa AAV-GUSBΔProm CMV Human GUSB pa Supplemental Figure 1. AAV vectors. Maps of the AAV vectors used are shown with viral ITRs, CMV enhancer, chicken β-actin promoter and first intron (β-act), human β-glucuronidase gene (GUSB), and polyadenyation site (pa) indicated. A promoterless version of the AAV2-BGUS vector was also used in this study, which retains the CMV enhancer but lacks the β-actin promoter (AAV2- BGUS Prom).