Linear Amplification-mediated Research Articles

Abstract 2852: A gene activating in vivo screen identifies EBAG9 as novel metastasis regulator in human colorectal cancer

Abstract As colorectal cancer (CRC) mortality remains high due to metastatic spread in patients with advanced disease stages, we here aimed at identifying novel metastasis regulators in CRC and at understanding their mechanism of action. In CRC, tumor progression and metastasis formation are driven by cells that possess tumor-initiating cell (TIC) activity. We therefore employed a collection of primary patient-derived CRC tumor spheroid cultures (TSC) that are enriched for cells with TIC activity to perform a gene activating in vivo screen, intending to identify genes that enhance metastasis formation. To this end, TSC (n=4) were transduced with a lentiviral trapping vector that integrates into the genome and thereby drives the overexpression of genes located in the vicinity of the integration site (IS). Transduced TSC were then serially transplanted under the kidney capsule of immunodeficient mice (NSG) and monitored for the induction of metastasis formation. Analysis of lentiviral IS by linear amplification-mediated (LAM)-PCR revealed an enrichment of an IS within the gene locus of Estrogen Receptor Binding Site Associated Antigen 9 (EBAG9) in metastases. RNA sequencing confirmed the trapping vector mediated upregulation of a truncated version of the EBAG9 gene (exon 5-7), which strongly suggested a contribution of EBAG9 to metastatic spread. To validate these findings, both the truncated (TR) as well as the physiologically occurring full length (FL) EBAG9 gene were overexpressed in TSC (n=4) and serially transplanted into NSG mice. In addition to an augmentation in metastasis upon EBAG9 overexpression compared to controls (FL: 1.4-fold, TR: 2.4-fold), we also detected the occurrence of lymphogenous metastases when EBAG9 levels were increased, indicating that EBAG9 not only contributes to metastatic spread but also modulates tropism. To further characterize the underlying processes, global gene and protein expression analyses were performed in primary TSC as well as in the CRC cell line DLD1. Compared to control transduced cells, EBAG9 overexpression induced expression changes in cell adhesion and microtubule molecules, which led us to investigate cell migration and adhesion as potential contributing mechanisms to metastasis formation. We here observed an increase both in cell migration as well as in adhesion to the extracellular matrix protein collagen I upon EBAG9 overexpression in DLD1. Interestingly, DLD1 adhesion to lymphatic endothelial cells was significantly higher in EBAG9 FL (p=0.008) and EBAG9 TR (p=0.016) overexpressing cells, underscoring the role of EBAG9 in tropism modulation. In summary, we here identified EBAG9 as novel metastasis regulator in CRC which will not only contribute to a better mechanistic understanding of metastatic spread but also to the development of novel future strategies for preventing and targeting advanced disease stages. Citation Format: Karin Laaber, Taronish D. Dubash, Anna Maria Melzer, Tania Christiansen, Nati Ha, Benedikt Brors, Martin Schneider, Friederike Herbst, Hanno Glimm, Claudia R. Ball. A gene activating in vivo screen identifies EBAG9 as novel metastasis regulator in human colorectal cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 2852.

664. Improving AAV Gene Therapy Safety Analysis: Multiplex LAM-PCR Provide New Insights Into AAV Vector Integration

Nowadays, recombinant adeno-associated viruses (rAAV) constitute one of the most successful vectors for human gene therapy. These viral vectors allow a long-term transgene expression mainly due to their ability to form stable concatemeric DNA structures. Interestingly, although it is a rare event, rAAV have been shown to randomly integrate into the host genome.The inverted terminal repeats (ITR) have been shown to be preferred breakage points within the rAAV vectors. Current analysis of AAV integration are based on the linear amplification mediated (LAM)-PCR technology, here the genomic regions adjacent to the ITRs are amplified and sequenced for the determination of the integration sites (IS). However, recent studies have shown that vector breakage may also occur along the whole vector sequence, thus requiring the development of new methods able to identify the integration of internal vector regions.We present a novel method for the detection of rAAV integration based on the LAM-PCR technology by employing simultaneously five primer sets covering the whole AAV vector sequence. This multiplex LAM-PCR was used, in parallel with the standard 5’ LAM-PCR, in liver, spleen and adrenal gland biopsies from monkeys injected with 1×1013 viral genomes (vg)/kg or 5×1013 vg/kg of the AAV2/5-AAT-coPBGD vector (this vector has been used in a phase I clinical trial for acute intermittent porphyria). Subsequent 454 sequencing and data analysis of ≈1 million raw sequences for each tissue allowed the identification of near 800 integration sites (IS). As it has been previously reported, vector integration occurs within different regions of the ITR, although preferentially in the C-region. In addition, multiplex LAM-PCR showed breakage at different positions of the AAV vector genome in all the organs analyzed.Our data show the increased sensitivity of this method and its suitability for the analysis of viral integration occurring at inner regions of the rAAV vector. Thus, this new method will provide a deeper insight into the interaction of AAV vectors with the host genome and will contribute to increase the thoroughness of the safety studies for the increasing number of gene therapy clinical trials using rAAV vectors.

95. Comprehensive wtAAV Integration Site Analysis Reveals Integration Site Formation Throughout the Whole Viral Genome

To establish latency, wildtype adeno-associated virus (wtAAV) is able to integrate its DNA into a specific region on chromosome 19 of the human genome, named AAVS1. This targeted integration mechanism relies on the function of the two larger versions (Rep78/68) of the Rep protein family, expressed by wtAAV. Rep is able to bind a 12 bp long sequence, called Rep binding site (RBS). The RBS can be found both in the viral ITRs and in AAVS1. By the formation of a trimeric complex between wtAAV, Rep and the genomic locus, targeted integration can be achieved. However, the exact mechanisms of Rep-mediated integration still remain unknown. To analyze wtAAV IS distribution genome-wide, large-scale integration analyses using high-throughput next generation sequencing have been performed. These studies revealed that Rep-mediated wtAAV integration preferentially occurs in AAVS1, but that additional integration hotspots associated with RBS can be found as well throughout the genome. Up to this point most integration sites (IS) reported for wtAAV were derived from vector-genome fusion sequences, where the viral sequence fragment was mapped near the ITRs or near the p5 region of the viral genome. Most currently available methods for the analysis of viral IS rely on amplification of the vector-genome junction with primers that bind in the vicinity of the viral ends. Therefore, knowledge about the nature and structure of the internal wtAAV genome remains limited. To address this question, we investigated ~3e6 wtAAV concatemeric structures generated by IS analyses using linear amplification-mediated (LAM) PCR to identify potential alternative breakpoints throughout the wtAAV genome. We discovered that breakpoints occur at any position of the wtAAV genome. Apart from the ITRs and the p5 promoter, there are a number of other preferred breakpoints throughout the wtAAV genome. Using a novel approach involving targeted enrichment of AAV and AAVS1 sequences, we also were able to consider viral IS at any position of the AAV genome. Targeted HiSeq sequencing of wtAAV and AAVS1 determined several hundreds of exact IS and confirmed that wtAAV genome junctions form at any position of its genome. Those cryptic integration events, not picked up by the majority of methods currently available, shed new light onto wtAAV persistence and may further improve current AAV persistence analyses in gene therapy.

Development and Evaluation of Quality Metrics for Bioinformatics Analysis of Viral Insertion Site Data Generated Using High Throughput Sequencing.

Integration of viral vectors into a host genome is associated with insertional mutagenesis and subjects in clinical gene therapy trials must be monitored for this adverse event. Several PCR based methods such as ligase-mediated (LM) PCR, linear-amplification-mediated (LAM) PCR and non-restrictive (nr) LAM PCR were developed to identify sites of vector integration. Coupling the power of next-generation sequencing technologies with various PCR approaches will provide a comprehensive and genome-wide profiling of insertion sites and increase throughput. In this bioinformatics study, we aimed to develop and apply quality metrics to viral insertion data obtained using next-generation sequencing. We developed five simple metrics for assessing next-generation sequencing data from different PCR products and showed how the metrics can be used to objectively compare runs performed with the same methodology as well as data generated using different PCR techniques. The results will help researchers troubleshoot complex methodologies, understand the quality of sequencing data, and provide a starting point for developing standardization of vector insertion site data analysis.

Integration Frequency and Intermolecular Recombination of rAAV Vectors in Non-human Primate Skeletal Muscle and Liver

The comprehensive characterization of recombinant adeno-associated viral (rAAV) integration frequency and persistence for assessing rAAV vector biosafety in gene therapy is severely limited due to the predominance of episomal rAAV vector genomes maintained in vivo. Introducing rAAV insertional standards (rAIS), we show that linear amplification-mediated (LAM)-PCR and deep sequencing can be used for validated measurement of rAAV integration frequencies. Integration of rAAV2/1 or rAAV2/8, following intramuscular (IM) or regional intravenous (RI) administration of therapeutically relevant vector doses in nine adult non-human primates (NHP), occurs at low frequency between 10(-4) and 10(-5) both in NHP liver and muscle, but with no preference for specific genomic loci. High resolution mapping of inverted terminal repeat (ITR) breakpoints in concatemeric and integrated vector genomes reveals distinct vector recombination hotspots, including large deletions of up to 3 kb. Moreover, retrieval of integrated rAAV genomes indicated approximately threefold increase in liver compared to muscle. This molecular analysis of rAAV persistence in NHP provides a promising basis for a reliable genotoxic risk assessment of rAAV in clinical trials.

Assessing the impact of lentiviral vector integration on splicing of cellular genes at the genome-wide level

Oncogenesis induced by insertional mutagenesis with gene therapy vectors occurs mainly by activation of proto-oncogenes found at or nearby the insertion site. This activation often occurs by an enhancer-mediated mechanism or by a process of splicing capture which generates chimeric transcripts comprising portions of vector and cellular mRNAs. Although the activation of oncogenes may be reduced by the use of self-inactivating design and moderate cellular promoters, how to reduce genotoxic splicing capture events and aberrant transcript formation triggered by vector integration is still unclear. We developed a modified Linear Amplification-Mediated (LAM) PCR technique, named cDNA LAM PCR (cLAM-PCR), aimed at retrieving, from the whole transcriptome of LV-transduced cells aberrantly spliced mRNAs that contain lentiviral vector (LV) sequences fused with cellular transcripts in a high-throughput fashion. The sequences of cLAM-PCR products were obtained by 454 pyrosequencing and analyzed by dedicated high-throughput computational pipeline running in a computer cluster that use a dynamic analysis process composed by different steps based on a map-reduce parallelization model. Thus, chimeric LV-genome sequences are recognized, the nucleotide position of the fused sequence is identified (the splice site), and the remaining portion mapped on the appropriate genome assembly by BLAST. We identified several established and previously unknown splice sites within the LV backbone that participate in the aberrant splicing process with variable efficiency. Results obtained with different LV designs show that integrated LVs can perturb the processing of cellular transcripts by interacting with the cellular splicing machinery and fusing with its own splice sites to cellular splice sites both upstream and downstream the integration site. So far, 70 different fusion transcripts could be identified in total, 84% of which were fused to known splice sites of gene exons, 6% were fused to uncharacterized cryptic splice sites located in introns and the remaining 10% were fused to genomic sequences not corresponding to any annotated gene. This analysis allows identifying also several different slice sites within the LV backbone that participated to the aberrabt splicing process. Quantitative PCR on different LV portions within the LV backbone allow measuring the relative contribution to the aberrant splicing process of each splice site identified. Interestingly, the amount of transcription occurring in regions outside the expression cassette reaches the 3% of the entire transgene expression. The cLAM-PCR technique, coupled to high-throughput sequencing and the computational power of our specialized data analysis pipeline allows gaining insights into the biology of vector-mediated splicing alteration. Since this process could induce neoplastic transformation by the generation of aberrant oncogenic protein, its in-depth characterization is instrumental in the development of next-generation LV with a higher safety profile.

In vivo expansion of MDR1‐transduced cells accompanied by a post‐transplantation chemotherapy regimen with mitomycin C and methotrexate

A recurrent breast cancer patient received high-dose chemotherapy, a transplant of multidrug resistance 1 (MDR1)-transduced cells and four different protocols of post-transplantation chemotherapy. We report the analysis of MDR1-transduced cells in this patient. MDR1 transgene levels in the peripheral blood mononuclear cells of the patient were evaluated by polymerase chain reaction (PCR). Retroviral integration sites of the MDR1-transduced cells were identified by linear amplification-mediated (LAM)-PCR. Twelve days after transplantation, approximately 1% of the peripheral blood mononuclear cells were MDR1 transgene-positive. The transgene levels decreased quickly, and were at low levels until day 504. A remarkable increase in MDR1 transgene-positive cells was observed on day 532, during combination chemotherapy with mitomycin C and methotrexate. Using LAM-PCR, 31 MDR1-transduced clones were identified, and eight of these were long-life clones that survived for more than 500 days. Among the 31 clones, ten had a retroviral integration site near genes listed in the Retroviral Tagged Cancer Gene (RTCG) Database. Two long-life clones, N-30 and N-31, had retroviral integration sites within the MDS1-EVI1 locus. Another two long-life clones had integration sites close to PRDM16 or CUEDC1. These results suggest that MDR1-transduced cells were enriched in vivo by an MDR1 substrate, mitomycin C. The possible activation of EVI1 or other RTCGs by retroviral insertion may have affected the survival and persistence of a proportion of the transduced cells.

Stable and Functional Lymphoid Reconstitution in Artemis-deficient Mice Following Lentiviral Artemis Gene Transfer Into Hematopoietic Stem Cells

Patients with mutations in the Artemis gene display a complete absence of T- and B lymphocytes, together with increased cellular radiosensitivity; this leads to a radiosensitive severe combined immunodeficiency (RS-SCID). Allogenic hematopoietic stem-cell (HSC) transplantation is only partially successful in the absence of an human leukocyte antigen-genoidentical donor, and this has prompted a search for alternative therapeutic approaches such as gene therapy. In this study, a self-inactivated lentiviral vector expressing Artemis was used to complement the Artemis knockout mouse (Art(-/-)). Transplantation of Artemis-transduced HSCs into irradiated Art(-/-) mice restored a stable (over a 15-month period of follow-up) and functional T- and cell repertoire that was comparable to that of control mice. The success of secondary transplantations demonstrated that the HSCs had been transduced. One of thirteen mice developed a thymoma 6 months after gene therapy. Although thymic cells were seen to be carrying two lentiviral integration sites, there was no evidence of lentivirus-driven oncogene activation. The Art(-/-) mice were found to be prone to develop T-cell lymphomas, either spontaneously or after irradiation. These data indicate that the observed lymphoproliferation was probably the consequence of the chromosomal instability associated with the Artemis-deficient background. As a whole, our work provides a basis for supporting the gene therapy approach in Artemis-deficient SCID.

High-resolution insertion-site analysis by linear amplification–mediated PCR (LAM-PCR)

Integrating vector systems used in clinical gene therapy have proven their therapeutic potential in the long-term correction of immunodeficiencies. The integration loci of such vectors in the cellular genome represent a molecular marker unique for each transduced cell and its clonal progeny. To gain insight into the physiology of gene-modified hematopoietic repopulation and vector-related influences on clonal contributions, we have previously introduced a technology--linear amplification-mediated (LAM) PCR--for detecting and sequencing unknown DNA flanking sequences down to the single cell level (Supplementary Note online). LAM-PCR analyses have enabled qualitative and quantitative measurements of the clonal kinetics of hematopoietic regeneration in gene transfer studies, and uncovered the clonal derivation of non-leukemogenic and leukemogenic insertional side effects in preclinical and clinical gene therapy studies. The reliability and robustness of this method results from the initial preamplification of the vector-genome junctions preceding nontarget DNA removal via magnetic selection. Subsequent steps are carried out on a semisolid streptavidin phase, including synthesis of double complementary strands, restriction digest, ligation of a linker cassette onto the genomic end of the fragment and exponential PCR(s) with vector- and linker cassette-specific primers. LAM-PCR can be adjusted to all unknown DNA sequences adjacent to a known DNA sequence. Here we describe the use of LAM-PCR analyses to identify 5' long terminal repeat (LTR) retroviral vector adjacent genomic sequences.

Safety of MGMT Gene Transfer into Hematopoietic Progenitors: A Phase I Clinical Trial Using Retroviral Gene Transfer of G156A MGMT To Protect Hematopoiesis during BG Plus Alkylating Agent Treatment of Advanced Malignancies.

The G156A mutation of the DNA repair gene O6-methylguanine DNA-methyltransferase (MGMT) confers hematopoietic cell resistance to O6-benzylguanine (BG) and the DNA-alkylating agents BCNU and temozolomide (TMZ). We conducted a Phase I clinical trial utilizing MGMT-mediated chemoprotection to improve treatment of advanced solid tumors. Here we report the safety and feasibility of this approach. Autologous CD34+ cells were transduced ex vivo with a retroviral vector MFG- G156A MGMT (K. Cornetta, NGVL at Indiana U.), then reinfused into non-myeloablated patients. Eight patients were accrued to the study and 6 patients were infused with a mean of 7.3 × 105 cells/kg obtained post transduction. Of the 6 patients who received cells, all received one cycle of chemotherapy (BG + BCNU or TMZ) 3 days prior to cell infusion and 3 patients received a second cycle of either BG + BCNU or BG + TMZ 4–6 weeks post infusion. No patient experienced grade 2 or greater infusion related toxicity, and one patient went off study prior to receiving cells due to neurological toxicity during cytokine administration. All patients who received G156A MGMT+ cells went off study due to progressive disease, and survival ranged from 2 weeks to 9 months post infusion. No patient showed evidence of replication competent retrovirus (RCR) when assessed for the presence of the viral envelope by quantitative PCR for GALV sequences (performed by the IU Vector Production Facility). The mean vector copy number in post transduction cells (n= 7 patients) was 0.29 copies per genome by quantitative PCR (Q-PCR), with a mean CFU transduction efficiency of 24%, and a mean of 9% CD34 cells expressing the G156A MGMT transgene by flow cytometry. Retroviral insertion sites in pre-infusion CFUs from 2 patients analyzed by linear amplification-mediated (LAM)-PCR were polyclonal, with an average of 1.4 retroviral insertions per positive CFU. Transgene positive cells were detected in vivo in 3 patients. In the first patient, after BG and BCNU at 6 weeks, the transgene was detected in 1 of 100 BM CFU by Q-PCR and one insertion site was detected in this clone by LAM-PCR. The patient went off study at 8 weeks, then received TMZ alone (300 mg/m2/day for 5 days). At 16 weeks, the transgene was detected at low frequency in BM MNCs and granulocytes by Q-PCR and in BM granulocytes as an internal band in LAM-PCR, although no other insertions sites were detectable. In a second patient, the transgene was detected in 2 of 30 BM CFU by PCR and in BM granulocytes and MNCs by LAM PCR at 5 weeks. In a third patient treated with BG + TMZ 4 weeks post cell infusion, 3 of 50 PB CFU were transgene positive by Q-PCR at week 28. These data demonstrate that infusion of retroviral MGMT transduced hematopoietic cells is safe and are the first to show emergence and low level persistence of transduced mutant G156A MGMT cells after nonmyeloablative conditioning in humans. Our future studies include using a lentiviral vector to achieve better levels of gene transfer and reduced risk of insertional mutagenesis.

Malignant Transformation of a PG13 Cell Line Producing an MFG Retroviral Vector Containing Wild-Type Murine Akt.

Over-expression of the pro-survival Akt gene in rat mesenchymal stem cells (MSCs) by retroviral transduction has been shown to improve cardiac repair following MSC transplant into infarcted rat heart (Mangi et al., Nature Medicine, 2003), and may hold great potential for clinical myocardial repair following cardiac damage. In preliminary transplant studies in NOD-SCID mice, we transduced human MSCs with an MFG-based retroviral vector containing genes for wild-type murine Akt and green fluorescent protein (AIG). The PG13-AIG cells used to supply retrovirus were prepared by multiple rounds of transduction of NIH 3T3-derived PG13 retrovirus packaging cells with amphotropic AIG retrovirus produced by transient transfection of the AIG vector plasmid into ϕNX-A producer cells. Primary human MSCs were transduced with PG13-AIG supernatant, were expanded in vitro for 1-3 passages, and surgically implanted after adherence to ceramic cubes into NOD-SCID mice. Three weeks after implantation of 5x106 human MSCs, a tumor mass formed around 1 of 8 ceramic cube implants, characterized histologically as a sarcoma. Retroviral insertion site analysis by linear amplification-mediated (LAM)-PCR revealed multiple insertions present within the tumor DNA. Sequence analysis of several of these vector insertion sites revealed no homology to the human genome. However, two of the insertions gave highly significant sequence homology to mouse chromosomes 15 and 11, inside introns of the mouse Dxd17 putative RNA helicase gene and mouse hypothetical Gm252 gene, respectively. These results indicate that the sarcoma was formed in greater part by Akt-transduced mouse cells rather than by transduced human MSCs. A third insertion site exhibited highly significant sequence homology to the gibbon ape leukemia virus (GALV) envelope gene present within the mouse PG13 producer cells. As GALV-pseudotyped retroviruses produced by PG13 cells do not transduce mouse cells, this argues against the tumor resulting from Akt-transduced NOD-SCID mouse cells due to production of a replication-competent retrovirus by transduced MSCs. PCR analysis detected the presence of the herpes simplex virus thymidine kinase gene (a selectable marker present in PG13 producer cells) in the tumor DNA, which together with the insertion site analysis indicates that the sarcoma was formed at least in part by Akt-transduced PG13 fibroblast cells, which may have been a rare contaminant in the MSC cell implants due to carry-over of producer cells during PG13-Akt viral supernatant transduction of MSCs. No published studies have found a tumorigenic potential for the NIH 3T3-derived PG13 cell line, and transfection of wild-type Akt into NIH 3T3 cells has been shown in several studies to confer little or no transformation or tumor-formation capacity. Regardless of origin, cell transformation by retroviral transduction of wild-type Akt, possibly with insertional mutagenesis, is a potential risk for clinical use of Akt gene transfer for enhanced stem cell survival and repair.

Retrovirus Insertion Site Analysis Following In Vivo Drug Selection in a Phase I Clinical Trial Utilizing G156A MGMT for Enhanced Chemotherapy of Advanced Malignancies.

The G156A mutant of the DNA repair gene O6-methylguanine DNA-methyltransferase (MGMT) confers hematopoietic resistance to O6-benzylguanine (BG) combined with DNA-alkylating agents BCNU or temozolomide, and allows for selective in vivo expansion with drug administration of murine hematopoietic progenitors transduced with G156A MGMT retrovirus. Here we report our latest findings on retroviral vector copy number and insertion site analysis following drug treatment from a Phase I clinical trial utilizing MGMT-mediated chemoprotection for enhanced treatment of advanced solid tumors. Seven patients have entered the trial and 6 have completed the cell infusion process. For all patients, autologous CD34+ cells were transduced ex vivo with an MFG retroviral vector containing the G156A MGMT gene (packaged with PG13 by the National Gene Vector Laboratory, Ken Cornetta, Director) in the presence of the fibronectin fragment CH-296 and the cytokines SCF, Tpo, and Flt-3 ligand for 72 hours with three additions of retroviral supernatant. At 72 hours following patient treatment with BG and BCNU, cells were re-infused. Prior to infusion, the average vector copy number by quantitative real-time PCR analysis for six patients was 0.34 copies per genome, with an average of 24% of CFUs transduced by standard PCR for G156A MGMT, and an average of 9% of CD34+ cells expressing the MGMT transgene by flow cytometry. In one patient with metastatic melanoma we have further analysis of insertions. For this patient, the pre-infusion vector copy number of the bulk CD34+ population was 0.54 copies per genome by real-time PCR, with 27% of CFUs transduced and 8% of CD34+ cells expressing the MGMT transgene prior to infusion. Linear amplification-mediated (LAM)-PCR analysis of retroviral insertion sites in pre-infusion CFUs from this patient confirmed a polyclonal population, with an average of 1.6 retroviral insertions per positive CFU. In this patient, BG (120 mg/m2) and BCNU (33 mg/m2) were administered at 6 weeks post-infusion, and temozolomide (300 mg/day for 5 days) was administered at 13 weeks. Peripheral blood (PB) and bone marrow (BM) granulocyte and mononuclear cells (MNCs) were collected at weeks 5, 11, 15, and 16 for DNA and CFU analysis. Vector copy number at all post-infusion time points was below the limit of detection of SYBR Green probe-based real-time PCR (<100 copies of G156A MGMT per 5000 genomes). LAM-PCR detected the vector in post-treatment samples based on an internal vector control band present in BM MNCs at week 11 and in BM granulocytes at week 16, although specific insertion sites were not detected. Standard PCR revealed 1 out of 100 CFUs from week 11 BM MNCs contained the vector, with 2 out of 30 CFUs from week 15 PB MNCs. LAM-PCR in a subset of week 11 CFUs confirmed a single insertion site present in the same PCR-positive CFU. Sequence analysis of clonal vector insertions pre- and post-infusion is ongoing, and thus far a number of sites have been characterized, adding to the emerging database of clinical retroviral insertions. These are the first data to show emergence of transduced mutant MGMT cells after nonmyeloablative conditioning in humans and suggest that despite a low frequency of vector-marked hematopoietic cells, clinical in vivo drug selection can be observed.

Direct Evidence for Ex Vivo Expansion of Human Hematopoietic Stem Cells.

Ex vivo expansion of hematopoietic stem cells (HSC) is a major challenge for clinical and experimental transplantation protocols. However, no significant clinical benefit has been demonstrated to date. Clonal kinetics of ex vivo-expanded HSCs is one of the basic transplantation biology questions to be addressed before we can optimize ex vivo expansion approaches. To characterize human HSC, xenotransplantation techniques such as the severe combined immunodeficiency (SCID) mouse repopulating cell (SRC) assay have proven the most reliable methods thus far. While SRC quantification by limiting dilution analysis (LDA) is the gold standard for measuring in vitro expansion of human HSC, LDA is a statistical method and does not directly establish that a single HSC has self renewed in vitro. By using lentiviral gene-marking and direct intra-bone marrow injection of cultured CD34+ CB cells, we demonstrate here the first direct evidence for self-renewal of individual SRC clones in vitro.To detect multiplied clones, 5x104 gene-marked CD34+ cells were cultured for 4 days in our ex vivo expansion culture system (Exp Hematol , 27:904–915, 1999), and then divided into 10 lots, each of which was transplanted directly into the bone marrow of a NOD/SCID mouse. We used linear amplification-mediated (LAM)-PCR to detect unique genomic-proviral junctions as clonal markers. Detection of the same clones in different mice would provide direct evidence of ex vivo multiplication of a SRC clone. We identified 20 clone-specific genomic-proviral junction sequences by LAM-PCR on 10 mice. Although 14 clones were detected in only one mouse, six clones were detected in more than 2 mice. In the next experiment, purified CD19+EGFP+ and CD33+EGFP+ cells from each mouse were analyzed for each clone to detect multi-lineage differentiation of amplified SRCs. We identified 15 clonal markers from 6 mice. While 12 clones were present in only one mouse, 3 clones were present in 2 independent mice and reconstituted both CD19+and CD33+cells. Finally, we designed a secondary transplantation experiment to confirm the self-renewal ability of each clone. We identified 39 clonal markers from 10 primary and 10 secondary transplanted mice, 11 of which were detected in multiple mice with secondary transplantable ability. Together, of 74 clones analyzed, 20 clones (27%) divided and repopulated in more than two mice after serum-free and stroma-dependent culture. Some of them were secondary transplantable. Furthermore, we identified new class of stem cells based not on repopulation, or cell surface markers, but on response to cytokine stimulation in vitro. Our data demonstrate that current ex vivo expansion conditions result in reliable stem cell expansion and the clonal tracking we have employed is the only reliable method that can be used in the development of clinically appropriate expansion methods.

5. Differential Expression of PIT1 and PIT2 after G-CSF Mobilization Is Associated with Efficient Gene Transfer Using GALV-Pseudotyped Gammaretroviral Vectors

The optimal stem cell source for stem cell gene therapy has yet to be determined. Most large animal studies have utilized peripheral blood or marrow-derived cells collected after administration of granulocyte colony-stimulating factor (G-SCF) and stem cell factor (SCF), however, SCF is unavailable for clinical use. A recent study of a competitive repopulation assay in the rhesus macaque model showed very inefficient engraftment of transduced G-CSF-mobilized peripheral blood (G-PBSC) CD34+ cells relative to G-CSF- and SCF-mobilized cells using amphotropic pseudotypes (Hematti P et al Blood 101, 2003, 2199|[ndash]|2205). Because G-PBSC would be the preferred target cell population for most clinical stem cell gene therapy applications, we asked whether we could achieve efficient engraftment of transduced G-PBSC in our baboon model using Phoenix-GALV pseudotype vectors. In order to better compare these results to those from previous experiments utilizing G-CSF- and SCF-primed bone marrow (G&S-BM) as a stem cell source, the first two baboons also received G-CSF-primed BM (G-BM) in a competitive repopulation. We transplanted three baboons with G-CSF-mobilized CD34+ cells transduced with GALV pseudotyped retroviral vectors. We observed high-level, persistent engraftment of transduced G-PBSC in all three animals with gene marking in granulocytes in one animal up to 60%. To determine retrovirus integration pattern of clones that contributed to long-term hematopoiesis linear amplification mediated (LAM)-PCR was used for all animal samples studied. In all animals analyzed multiple clones could be detected and in animals that received both G-PBSC and G-BM multiple clones contributed from both transduced cell populations. These in vivo marking results were not in agreement with the report by Hematti et al so we sought to determine whether the different results might be explained by the fact that different combinations of cytokines (G-CSF and SCF) for mobilization modulate receptor expression of the amphotropic (PIT2) and GALV (PIT1). Analysis of receptor expression in baboon CD34+ cells after administration of different combinations of cytokines for mobilization found receptor expression was dramatically affected by different mobilization regimens with G-CSF and SCF increasing PIT2 expression and G-CSF alone increasing PIT1 expression. The low in vivo gene marking in previous reports using amphotropic pseudotyped vectors and G-CSF alone for mobilization could be explained because in animals treated with G-CSF alone, at least in baboons, the amphotropic receptor expression was minimal while including SCF in the mobilization regimen substantially increased amphotropic receptor expression. In contrast to other studies, our high in vivo marking in G-CSF-mobilized baboon peripheral blood CD34+ cells demonstrates the feasibility of efficiently transducing G-PBSC in clinical gene therapy trials and verifies that there is no inherent stem cell deficiency in G-PBSC mobilized cells.

Persistence of eGFP Marked Bone Marrow Cells in Long-Term Hematopoiesis.

To study transplanted unperturbed and mobilized long-term hematopoiesis after selection with an alkylating agent, bone marrow (BM) from 5 C57BL/6J mice was pooled, repeatedly transduced with retroviruses encoding the alkylating agent resistance protein O6-Methylguanine-DNA and enhanced green fluorescent protein (eGFP) as an easily traceable marker. Between 1 to 9x105 transfected BM cells were transplanted into 15 myeloablatively irradiated sex-mismatched C57BL/6J mice. Subsequently, 3 to 4 selection rounds with BCNU/O6-BG were carried out, enriching eGFP marked hematopoiesis in these mice up to 70–90%. Between 1 and 7x107BM cells of different mice were transplanted according to marrow location into groups of 5 sex-matched Bri44[1] mice. Two mice each received BM from the hind limbs, two from the pelvis and one received cells from the spleen, only, respectively. Altogether the study comprised 15 groups divided into 6 female and 9 male groups. Of these, 4 male and 3 female groups received 3 HSC-mobilization courses with G-CSF at intervals of 2 months starting 3 month after transplantation. Hematopoiesis in the other fraction remained unperturbed. During the observation period of 11–14 months in these tertiary recipients, repeated FACS analyses as well as linear amplification mediated (LAM) PCRs were carried out to track the clonal contributions. A decrease in the percentage of eGFP expressing marked hematopoiesis was observed in most cases. However, eGFP expression never disappeared altogether and could still be detected in the different hematopoietic lineages and successfully sorted for further analyses by MoFlo (Dako-Cytomation). Assessment of the clonal status of the Bri44 by LAM-PCR displayed interesting results. In some mice a decline in clone numbers was observed, whereas clone numbers remained stable in others. Tertiary transplantation with long-term follow-up indicates that this observation may be related to the transplantation of limited long-term repopulating clone numbers and progenitor cell exhaustion over time.

Lentiviral vector transduction of NOD/SCID repopulating cells results in multiple vector integrations per transduced cell: risk of insertional mutagenesis

Efficient vector transduction of hematopoietic stem cells is a requirement for successful gene therapy of hematologic disorders. We asked whether human umbilical cord blood CD34(+)CD38(lo) nonobese diabetic/severe combined immunodeficiency (NOD/SCID) repopulating cells (SRCs) could be efficiently transduced using lentiviral vectors, with a particular focus on the average number of vector copies integrating into these primitive progenitor cells. Mouse bone marrow was analyzed by fluorescence-activated cell-sorter scanner and by semiquantitative polymerase chain reaction (PCR) to determine the transduction efficiency into SRCs. Lentiviral vector transduction resulted in an average of 22% (range, 3%-90%) of the human cells expressing green fluorescent protein (GFP), however, multiple vector copies were present in human hematopoietic cells, with an average of 5.6 +/- 3.3 (n = 12) copies per transduced cell. To confirm the ability of lentiviral vectors to integrate multiple vector copies into SRCs, linear amplification mediated (LAM)-PCR was used to analyze the integration site profile of a selected mouse showing low-level engraftment and virtually all human cells expressing GFP. Individually picked granulocyte macrophage colony-forming unit colonies derived from the bone marrow of this mouse were analyzed and shown to have the same 5 vector integrants within each colony. Interestingly, one integration site of the 5 that were sequenced in this mouse was located in a known tumor-suppressor gene, BRCA1. Therefore, these findings demonstrate the ability of lentiviral vectors to transduce multiple copies into a subset of NOD/SCID repopulating cells. While this is efficient in terms of transduction and transgene expression, it may increase the risk of insertional mutagenesis.