Non-Random Lentiviral Vector Insertions in Bone Marrow Progenitors from Fanconi Anemia Patients

Improved protocols using lentiviral vectors have been established with minimal cytokine exposure and short transduction times proving more suitable for overcoming the disease-specific challenge in correcting functionally defective hematopoietic stem cells (HSCs) of Fanconi Anemia (FA) patients. Bone marrow (BM) cells from FA patients were transduced ex vivo with lentiviral vectors (LVs) expressing FANCA and/or EGFP using optimized conditions to preserve the repopulating properties of the primitive hematopoietic stem cells (manuscript submitted). In a forward preclinical screening of possible LV-induced side effects we analyzed the insertional inventory in colonies generated by FA BM cells previously transduced with the LVs. We have established and optimized DNA and RNA isolation procedures for minimal cell numbers, suitable for large scale screening of colony forming cell (CFC) derived colonies by linear amplification-mediated PCR (LAM-PCR) and massive parallel pyrosequencing (454 GS Flx system; Roche). This approach is applicable for detecting early indicators of clonal selection, and is based on the analysis of common integration sites (CIS) and non-random distribution of vector insertions in particular genomic loci. From a total of 180 CFC-derived colonies expressing the EGFP LV marker gene, 298 vector insertions could be sequenced and mapped to the human genome. The analysis of vector targeted gene coding regions showed a non-random genomic distribution of LV insertions, with a significant overrepresentation of RefSeq genes that are part of distinct functional categories. Accordingly vector associated genes are predominantly involved in cellular signal cascades regulated by the MAP Kinase family known to be involved in a wide variety of cellular processes such as proliferation, differentiation, transcription regulation and development. Apart from the observed high integration frequency in genes (>80%), partial loss of vector LTR nucleotides was detected in >10% of the integrants (3–25bp). Notably, >20% of the lentiviral insertions were found to be located in CIS of predominantly 2^nd order. Further screening assays of LV transduced CFC-derived colonies will allow a deeper investigation in the functional consequences of such CIS targeting in gene therapy protocols of FA. However our results suggest that the LV transduction of FA BM progenitors leads to a relatively high frequency of insertions in CIS which may be indicative of an insertion based (specific) selection mechanism. We herby show that the ex vivo large scale integration site analyses of CFC-derived colonies from patients considered to undergo gene therapeutic treatments constitutes a robust approach, which combined with mouse preclinical biosafety studies will help to improve the safety of clinical gene therapy protocols. The non-random distribution of LV integrations in CIS associated genes and in genes involved in particular cellular pathways may be indicative for the altered biochemical pathways characteristic of FA stem cells, with reported defects in DNA repair and self-renewal.