Identification of a high incidence region for retroviral vector integration near exon 1 of the LMO2 locus

Therapeutic retroviral vector integration near the oncogene LMO2 is thought to be a cause of leukemia in X-SCID gene therapy trials. However, no published studies have evaluated the frequency of vector integrations near exon 1 of the LMO2 locus. We identified a high incidence region (HIR) of vector integration using PCR techniques in the upstream region close to the LMO2 transcription start site in the TPA-Mat T cell line. The integration frequency of the HIR was one per 4.46 × 104 cells. This HIR was also found in Jurkat T cells but was absent from HeLa cells. Furthermore, using human cord blood-derived CD34+ cells we identified a HIR in a similar region as the TPA-Mat T cell line. One of the X-linked severe combined immunodeficiency (X-SCID) patients that developed leukemia after gene therapy had a vector integration site in this HIR. Therefore, the descriptions of the location and the integration frequency of the HIR presented here may help us to better understand vector-induced leukemogenesis.


Preparation of MLV vector and infection
To produce recombinant ecotropic MLV viruses containing GFP, Plat-E cells (1 × 10 6 ) were seeded in 10-cm dishes, incubated for 24 hours and transfected with 10 μg of pMXs-GFP by using a calcium-phosphate precipitation method. After 12 hours of incubation, the culture media was exchanged for fresh media. The supernatants were harvested 48 hours after transfection and filtered with 0.45-μm pore-size membrane filters (Millipore, Billerica, MA, USA). Next, TPA-Mat-ecoR or Jurkat-ecoR cells (1 × 10 6 ) were mixed with 0.5 ml of the supernatants and diluted with an equal volume of RPMI medium containing 4 μg/ml of polybrene with or without 200 nM PDBu. The cells were then transferred to 12-well plates and centrifuged at 2,200 rpm for 1 hour at 30°C. After 12 hours of incubation, the culture media was exchanged for fresh media. 48 hours after infection, the cells were harvested, and fixed with 1% paraformaldehyde in PBS.
The infection efficiencies of TPA-Mat-ecoR and Jurkat-ecoR cells were assessed according to the GFP fluorescence by flow cytometry using a FACSCalibur (Becton Dickinson, San Jose, CA, USA). In order to produce recombinant MLV viruses that can infect HeLa or human CD34 + cells, 293T cells (1 × 10 6 ) were transfected with 10 μg each of gagpol-IRES-brs, pMXs-GFP and 5 μg of pcDNA VSV-G by a calcium-phosphate precipitation method.

Cloning of MLV integration sites
To identify MLV vector integration sites near exon 1 of the LMO2 gene or integration hotspots in the TRAF2-and NCK-interacting kinase (TNIK) gene, genomic DNA was isolated from infected TPA-Mat-ecoR cells, Jurkat-ecoR cells, HeLa cells or CD34 + cells using a proteinase K extraction method. The polymerase chain reaction (PCR) conditions were denaturation at 94°C for 2 minutes, followed by 9 cycles of denaturation at 94°C for 20 seconds, annealing at 65°C for 20 seconds graduating by -0.5°C per cycle and extension at 72°C for 4 45 seconds, and 30 cycles of 94°C for 20 seconds, 65°C for 20 seconds and 72°C for 45 seconds. Next, 1/100 of the volume of the first-round PCR product was subjected to a second round of PCR under the same PCR conditions with nested primers. The amplified PCR products were separated in 1.5% agarose gels containing ethidium bromide, cloned into the p3T vector (MoBi Tec, Goettingen, Germany), transformed into DH5α and sequenced in an ABI PRISM 3100 Genteci Analyzer (Applied Biosystems, Foster City, CA, USA).

Mapping of MLV integration sites
MLV vector integration sites were mapped to the human genome as described previously [2]. The raw sequences were trimmed to remove all vector elements, and the remaining portion of the sequences was reviewed. The junction sequences next to the 3' or 5' long terminal repeat (LTR) end sequences were compared with the human genome using the BLAST program (http://blast.ncbi.nlm.nih.gov/Blast.cgi). Authentic integration sites used in the analysis were defined as: (i) containing a sequence contiguous from the nested primer for the LTR (e.g., the end of the 3' or 5' LTR) to the nested primer sequence in the genome (e.g., LMO2 or TNIK locus); (ii) matching a genomic

Luciferase promoter assay
The promoter region of LMO2, corresponding to -3020 to +147 bp (relative to the LMO2 transcription start site, +1), was amplified by PCR using TPA-Mat genomic DNA as template, and cloned into MluI and XhoI sites of pGL3-basic plasmid containing a SV40 promoter (Promega Madison, WI USA), and designated pGL3lmo2(3020) as previously described [7]. To assess the influence of MLV vector insertion in the high incidence region (HIR) of the LMO2 gene, the LTR of and Renilla-luciferase (PRL-null plasmid).

Real-time PCR
To estimate copy number of the integrated MLV vector per genome, we performed quantitative PCR analysis for TPA-Mat-ecoR, HeLa and CD34 + cells.   2) To avoid an overestimation of the integration frequency, we selected 3'L1L2/R5R6 (3' LTR primer direction).
Therefore, the integration frequency derived from one primer set (3'L1L2/R5R6) is half a correct value and was doubled.
Estimation of the integration frequency in CD34 + cells We tried to estimate the integration frequency in CD34 + cells. We first performed real-time PCR to estimate the vector integration copy number in CD34 + cells.
The PCR analyses showed that the number of vector integrations in CD34 + cells (infection efficiency; 14.7% based on GFP fluorescence) was estimated at 0.32 per diploid genome.
We calculated the value of the vector integrations in the HIR (primer set; 13 Therefore, the integration frequency of the HIR in CD34 + cells is estimated to be 2.36 times higher than that of TPA-Mat cells on the basis of real-time PCR. Accordingly, the integration frequency of the HIR in CD34 + cells was estimated to be one per 9.00 × 10 4 integrations (2.125 × 10 5 integrations/2.36) or one per 1.89 × 10 4 cells (4.46 × 10 4 cells/2.36).