Realizing the therapeutic potential of human induced pluripotent stem (iPS) cells will require robust, precise and safe strategies for genetic modification, as cell therapies that rely on randomly integrated transgenes pose oncogenic risks. Here we describe a strategy to genetically modify human iPS cells at ‘safe harbor’ sites in the genome, which fulfill five criteria based on their position relative to contiguous coding genes, microRNAs and ultraconserved regions. We demonstrate that ~10% of integrations of a lentivirally encoded β-globin transgene in β-thalassemia-patient iPS cell clones meet our safe harbor criteria and permit high-level β-globin expression upon erythroid differentiation without perturbation of neighboring gene expression. This approach, combining bioinformatics and functional analyses, should be broadly applicable to introducing therapeutic or suicide genes into patient-specific iPS cells for use in cell therapy.

Polymorphous low-grade adenocarcinoma (PLGA) and cribriform adenocarcinoma of minor salivary gland (CAMSG) are low-grade carcinomas arising most often in oral cavity and oropharynx, respectively. Controversy exists as to whether these tumors represent separate entities or variants of one spectrum, as they appear to have significant overlap, but also clinicopathologic differences. As many salivary carcinomas harbor recurrent translocations, paired-end RNA sequencing and FusionSeq data analysis was applied for novel fusion discovery on two CAMSGs and two PLGAs. Validated rearrangements were then screened by fluorescence in situ hybridization (FISH) in 60 cases. Histologic classification was performed without knowledge of fusion status and included: 21 CAMSG, 18 classic PLGA, and 21 with "mixed/indeterminate" features. The RNAseq of 2 CAMSGs showed ARID1A-PRKD1 and DDX3X-PRKD1 fusions, respectively, while no fusion candidates were identified in two PLGAs. FISH for PRKD1 rearrangements identified 11 additional cases (22%), two more showing ARID1A-PRKD1 fusions. As PRKD2 and PRKD3 share similar functions with PRKD1 in the diacylglycerol and protein kinase C signal transduction pathway, we expanded the investigation for these genes by FISH. Six additional cases each showed PRKD2 and PRKD3 rearrangements. Of the 26 (43%) fusion-positive tumors, there were 16 (80%) CAMSGs and 9 (45%) indeterminate cases. A PRKD2 rearrangement was detected in one PLGA (6%). We describe novel and recurrent gene rearrangements in PRKD1-3 primarily in CAMSG, suggesting a possible pathogenetic dichotomy from "classic" PLGA. However, the presence of similar genetic findings in half of the indeterminate cases and a single PLGA suggests a possible shared pathogenesis for these tumor types.

564 Insertional oncogenesis poses a severe hurdle to current gene therapy for blood disorders. The semi-random insertion of retroviral vectors combined with the inability for prolonged ex vivo culture of hematopoietic stem cells (HSCs) prohibits prospective integration site selection. The advent of induced pluripotent stem cells (iPSCs) offers for the first time the possibility of generating patient-specific stem cells that can be extensively manipulated in vitro, thus creating a unique platform for precise genetic engineering. Here, we present a novel strategy for the genetic correction of ß-thalassemia major based on the identification and selection of patient-specific iPSC clones harboring a normal ß-globin gene integrated in genomic “safe harbor” sites that permit therapeutic levels of expression while minimizing the possibility of oncogenic risks. We generated a total of 20 iPSC lines from bone marrow stromal cells or skin fibroblasts from 4 patients with ß-thalassemia major of various genotypes using our traceable lentiviral vector system (Papapetrou et al., PNAS, 2009; Lee et al., Nature, 2009), as well as an excisable single polycistronic vector co-expressing OCT4, SOX2, KLF4 and cMYC. Additional transgene-free thal-iPSC lines were generated following Cre recombinase-mediated excision of the reprogramming vector. Seven thal-iPSC lines were selected for further characterization and were shown to fulfill multiple criteria of pluripotency, including teratoma formation. Four thal-iPSC lines were transduced with a lentiviral vector encoding the human ß-globin gene cis-linked to its hypersensitive site (HS) 2, HS3 and HS4 locus control region elements, derived from the TNS9 vector, previously shown to confer erythroid-specific ß-globin gene expression at therapeutic levels (May et al., Nature, 2000). Thal-iPSC clones harboring single vector copies were selected and vector integration sites were mapped to the human genome. To identify safe harbor sites we adopted a set of 5 criteria: (1) distance of at least 50 kb from the 5' end of any gene, (2) distance of at least 300 kb from any cancer-related gene, (3) distance of at least 300 kb from any miRNA, (4) location outside a transcription unit and (5) location outside ultraconserved regions. A survey of 5840 integration sites of the globin lentiviral vector that we mapped in thal-iPSCs revealed that 17.3% meet all five “safe harbor” criteria, supporting the feasibility of recovering thal-iPSC clones harboring vector integrations in “safe harbors” by screening a relatively small set of single-copy clones. Indeed, 3 “safe harbor” integrations were retrieved amongst 36 sites mapped in thal-iPSC clones. Among 13 clones randomly selected and thoroughly confirmed to harbor a single copy of the globin vector, we found one clone, thal5.10-2, with an integration that meets all five “safe harbor” criteria. Upon erythroid differentiation, 12 of the 13 single copy thal-iPSC clones expressed detectable vector-encoded ß-globin at levels ranging from 9% to 159% (mean 53%) of a normal endogenous ß-globin allele. 9 out of 13 clones expressed the ß-globin transgene at levels higher than 30%. Remarkably, the “safe harbor” clone 5.10-2, expressed 85% of a normal ß-globin allele. Microarray analysis of undifferentiated thal-iPSC clones and their erythroid progeny revealed that 3 out of 5 integrations eliminated by our “safe harbor” criteria result in perturbed expression of neighboring genes at a distance ranging between 9 and 275 kb from the vector insertion. Of note, the “safe harbor' integration site in clone 5.10-2 is in a genomic region with no genes within 300 kb on either side, while no significant differentially expressed genes were found elsewhere in the genome. These data demonstrate that the selection of iPSC clones, wherein therapeutic levels of transgene expression without perturbation of endogenous genes is obtained from selected chromosomal sites is feasible by screening a limited number of single-copy clones and applying five “safe harbor” criteria. This study provides a framework and a strategy combining bioinformatics and functional analyses for identifying “safe harbors” for transgene integration and expression in the human genome. This approach may be broadly applicable to introducing therapeutic, suicide or marker genes into patient-specific iPSCs towards the development of safer stem cell therapies. Disclosures: No relevant conflicts of interest to declare.