Viral vector use in gene therapy has highlighted several safety concerns, including genotoxic events. Generally, vector-mediated genotoxicity results from upregulation of cellular proto-oncogenes via promoter insertion, promoter activation, or gene transcript truncation, with enhancer-mediated activation of nearby genes the primary mechanism reported in gene therapy trials. Vector-mediated genotoxicity can be influenced by virus type, integration target site, and target cell type; different vectors have distinct integration profiles which are cell-specific. Non-viral factors, including patient age, disease, and dose can also influence genotoxic potential, thus the choice of test models and clinical trial populations is important to ensure they are indicative of efficacy and safety. Efforts have been made to develop viral vectors with less risk of insertional mutagenesis, including self-inactivating (SIN) vectors, enhancer-blocking insulators, and microRNA targeting of vectors, although insertional mutagenesis is not completely abrogated. Here we provide an overview of the current understanding of viral vector-mediated genotoxicity risk from factors contributing to viral vector-mediated genotoxicity to efforts made to reduce genotoxicity, and testing strategies required to adequately assess the risk of insertional mutagenesis. It is clear that there is not a ‘one size fits all’ approach to vector modification for reducing genotoxicity, and addressing these challenges will be a key step in the development of therapies such as CRISPR-Cas9 and delivery of future gene-editing technologies.

Despite years of research and numerous clinical trials, only 2 gene therapies, Glybera and Strimvelis, are approved for clinical use, largely due to safety concerns of viral vectors. The advent of new therapies, such as Clustered Regularly Interspaced Short Palindromic Repeat (CRISPR)-Cas9 (an RNA-guided gene-editing platform that directs cutting of DNA in a specific gene) and modified RNA, has revived interest in the safety of viral vectors (it should be noted that all DNA editing technology comes with an additional safety concern of off-target cutting). The most commonly reported viruses utilized for gene therapy are adenovirus (AV), adeno-associated virus (AAV), retrovirus (γ-RV), and lentivirus (LV). Integrating vectors (γ-RV and LV) have been preferred as they allow permanent DNA integration, leading to long-term expression of corrective genes (Aiuti et al., 2013), and gene therapy vectors have most commonly been based on the γ-RV murine leukaemia virus (MLV). Safety concerns associated with viral vectors include those that can lead to genotoxic events e.g. inflammation, random insertion disrupting normal genes, activation of proto-oncogenes, and insertional mutagenesis (Baldo et al., 2013). This typically manifests as disruption or dysregulation of gene expression, but can cause clonal expansion and oncogenesis. This phenomenon is referred to as vector-mediated genotoxicity.

VECTOR-MEDIATED GENOTOXICITY

Insertional mutagenesis is a recognized safety concern of viral vector-based gene therapy. The probability of insertion is higher for integrating vectors (RV insertion has been investigated in the most detail), and although these are typically employed for ex vivo gene transfer, which can reduce the risk of insertional oncogenesis since transduced cells can be pre-screened to select only those clones with the desired insertion (Yi et al., 2011), they have also been used for in vivo approaches (Balaggan and Ali, 2012).

Insertional mutagenesis is complex; a single integration can affect several genes, and transcriptional dysregulation can be induced by multiple mechanisms (Sokol et al., 2014; Suerth et al., 2014). The insertion into DNA can be biologically silent, it can interfere with the host cell transcription/post-transcriptional activity of neighbouring genes but be biologically or clinically irrelevant, or combined with additional mutagenic events can lead to cellular transformation (Aiuti et al., 2013). Of most concern are dominant gain-of-function mutations that activate proximal proto-oncogenes, which generally occurs via promoter insertion, promoter activation, gene transcript truncation, or epigenetic gene silencing (Aiuti et al., 2013; Rae and Trobridge, 2013; Suerth, et al., 2014; Touw and Erkeland, 2007). Insertion of the viral promoter upstream of cellular transcription units can result in read-through transcription into adjacent cellular genes, either through the promoter or long terminal repeat (LTR) (Figure 1Ai) (Suerth, et al., 2014). Insertion of the virus into the promoter 5′ of a target gene may activate the gene through LTR promoter sequences (Figure 1Aii) or induce overexpression of aberrant splice variants (Touw and Erkeland, 2007). If the virus integrates near a target gene, gene expression may be increased via endogenous promoter activation by viral enhancer or promoter sequences (Figure 1Bi), or enhanced mRNA stability if the virus integrates in the 3′ untranslated region (UTR) (Figure 1Bii). Intragenic integration of the vector can lead to disruption of mRNA and/or overexpression of truncated mRNA, resulting in transcripts lacking 3′ or 5′ sequences, which may inactivate a gene (Figure 1C). Gene silencing can also be the result of epigenetic changes, where virus-induced gene methylation inhibits the expression of flanking genes due to histone and DNA methylation “spreading” (Figure 1D) (Rae and Trobridge, 2013; Touw and Erkeland, 2007).

FIG. 1

Mechanisms of viral vector-mediated genotoxicity. A, Promoter insertion either (i) upstream of cellular transcription units or (ii) into the promoter 5′ of a target gene. B, Promoter activation either (i) if the virus integrates near a target gene mediated by enhancer sequences in the viral LTR or viral promoter or (ii) if the virus integrates in the 3′ UTR enhancing the stability of the mRNA. C, Gene transcript truncation by intragenic integration of the vector can lead to disruption of the mRNA and/or overexpression of truncated mRNA, resulting in transcripts lacking 3’ or 5’ sequences, which may inactivate a gene. D, Epigenetic changes where virus-induced gene methylation inhibits the expression of flanking genes due to histone and DNA methylation ‘spreading’. Adapted from (Suerth et al., 2014; Touw and Erkeland, 2007).

FIG. 1

Mechanisms of viral vector-mediated genotoxicity. A, Promoter insertion either (i) upstream of cellular transcription units or (ii) into the promoter 5′ of a target gene. B, Promoter activation either (i) if the virus integrates near a target gene mediated by enhancer sequences in the viral LTR or viral promoter or (ii) if the virus integrates in the 3′ UTR enhancing the stability of the mRNA. C, Gene transcript truncation by intragenic integration of the vector can lead to disruption of the mRNA and/or overexpression of truncated mRNA, resulting in transcripts lacking 3’ or 5’ sequences, which may inactivate a gene. D, Epigenetic changes where virus-induced gene methylation inhibits the expression of flanking genes due to histone and DNA methylation ‘spreading’. Adapted from (Suerth et al., 2014; Touw and Erkeland, 2007).

Enhancer-mediated promoter activation has been shown to be the most prominent mechanism for proto-oncogene activation from mouse models and clinical trials, and involves the interaction of short- and long-range viral enhancer sequences with cellular promoters (Cesana et al., 2012; Hacein-Bey-Abina et al., 2010; Kool and Berns, 2009). When enhancer activity is reduced, however, inactivation of tumour suppressor genes leading to oncogenesis has been demonstrated (Cesana et al., 2014).

Gene transcript truncation is frequently observed when using viral vectors with active LTRs. Vector-derived transcripts could use splice donor sequences to fuse with splice acceptors in exons of cellular genes leading to the production of truncated proteins (Cesana et al., 2012), which could lead to loss-of-function mutations in tumour suppressor genes (Aiuti et al., 2013). There is increasing evidence that the potential of inducing aberrant transcripts might constitute a previously unappreciated genotoxicity and safety concern; LV-induced gene transcript truncations of 5′ or 3′ sequences led to removal of regulatory regions of proto-oncogenes and leukaemia in mice (Montini et al., 2009) or downregulation of full-length transcripts of supposedly haplosufficient tumour suppressor genes (Heckl et al., 2012). Furthermore, in a clinical trial using LVs, generation of a truncated HMG2A mRNA rendered it insensitive to let-7 microRNA-mediated degradation, resulting in clonal expansion (Cavazzana-Calvo et al., 2010).

INFLUENCERS OF GENOTOXICITY

Viral Factors

Vector Design

Vector design has been shown to modulate the genotoxic potential of RVs (Montini et al., 2009). The transcriptionally active LTR bearing strong enhancer and promoter sequences has been identified as the major determinant of genotoxicity; once the vector has inserted into the genome, these active promoters and enhancers can trans-activate neighbouring genes (Aiuti et al., 2013; Montini et al., 2009). Moreover, the selection of enhancers/promoters is critical for the genotoxicity/carcinogenicity risk, with increased rates of hepatocellular carcinoma observed following treatment with adeno-associated virus (AAV) vectors containing thyroxine-binding globulin (TBG) and chicken β-actin enhancer promoters, but not others (e.g. human α-1 antitrypsin (hAAT)), therefore trans regulatory sequences influence AAV genotoxicity (Chandler et al., 2015). The presence of splice-donor or -acceptor sites and polyadenylation signals in integrating vectors may favour the generation of alternative transcripts (Cesana et al., 2012; Heckl et al., 2012; Moiani et al., 2012) and there is potential for inherent genetic instability, which is important when evaluating safety (Aiuti et al., 2013).

Integration Site/Insertion Profile

RV integration is not random; each viral integrase has a weak preference for a short genomic sequence. MLV integration sites are strongly clustered in the genome, with ≥50% assigned to tight clusters that cover <2% of the genome (LaFave et al., 2014; Niederer and Bangham, 2014). γ-RVs preferentially integrate near transcriptional start sites (TSS), gene regulatory elements (e.g. promoters/enhancers) (Cattoglio et al., 2010; De Ravin et al., 2014; LaFave et al., 2014; Mitchell et al., 2004; Niederer and Bangham, 2014; Wu et al., 2003), DNase-I hypersensitive sites (HSS) (Lewinski et al., 2006) and CpG islands (Suerth et al., 2014). Moreover, γ-RVs reportedly have an integration bias for “hot spots” (genes associated with cell growth and cancer), which may further increase the risk for insertional mutagenesis (Montini et al., 2009). For example, γ-RVs have been shown to integrate at enhancers in lymphoid tumors independently of the distance to the TSS (Sokol et al., 2014) and the rate and type of tumour formation is controlled by enhancer activity within the U3 region (Bokhoven et al., 2009).

HIV-1 has a strong bias for transcriptional units in actively transcribed genes avoiding TSS (Mitchell et al., 2004; Niederer and Bangham, 2014; Suerth et al., 2014; Wu et al., 2003) with regional hotspots having been identified (Schröder et al., 2002). LVs integration site preference changes according to the activity status of the transduced cells, with different integrations between resting and activated CD4+ T cells (Brady et al., 2009) and dividing and non-dividing rodent cells (Bartholomae et al., 2011). Furthermore, LVs preferentially integrate into the major grooves of outward facing DNA on the nucleosome surface (Wang et al., 2009), and insertion can induce proto-oncogenes (e.g. Evi1 and B-Raf) (Heckl et al., 2012).

Non-Viral Factors

Transgene

The cellular role of the transgene product is important (e.g. involvement in growth regulation increases the risk of oncogenesis (Suerth et al., 2014)). For example, while MLV-based RVs were used in both X-linked severe combined immunodeficiency (SCID-X1) and adenosine deaminase-SCID (ADA-SCID) trials, and LMO2 integrations were observed at a similar frequency in both trials, leukaemia was only observed in the SCID-X1 trial (Aiuti et al., 2007; Niederer and Bangham, 2014). This implies that integration in LMO2 alone is not enough for malignant transformation, and the transgene being delivered may play a role; ADA is a housekeeping enzyme and promotes survival rather than growth, while IL2RG is a potentially oncogenic growth factor receptor (Aiuti et al., 2007; Niederer and Bangham, 2014).

Target cell

Integration of vectors can be cell-specific. For example, MLV transduction of haematopoietic stem cells (HSC) and mature peripheral blood lymphocytes (PBL) resulted in a different integration pattern, which was linked to the cell’s differential gene expression profile and epigenetically controlled overall accessibility of the genome during transduction. Specifically, 3 epigenetic determinants of cell-dependent integration were identified: 1) proximity to HSSs (related to open chromatin state and presence of active DNA binding sites) as integrations in PBLs were on average 2 times closer to HSSs than in HSCs; 2) the presence of histone modifications (insertion sites were preferentially located proximal to histone modifications associated with open chromatin); and 3) H3K27me3 (a histone modification indicative of open chromatin) was disfavoured cell-specifically (Biasco et al., 2011).

The risk of oncogenesis appears to be inversely related to cell or tissue maturity (Aiuti et al., 2013). Biasco et al. (2011) found fewer unique insertions in less-mature HSCs compared with more-mature PBLs (2198 and 1959 respectively, but cell maturity was not the only factor) and the distribution of H3K27me3 changes significantly upon differentiation of HSCs (Cui et al., 2009). In vivo, foetal and neonatal mice showed a particular sensitivity to the effects of integrating vectors, having a higher risk of oncogenesis than older mice (Themis et al., 2005).

When cells are transduced ex vivo, consideration should be given to cell exposure conditions, both in vitro, (eg, culture duration or growth factor exposure), and in vivo post-administration, including ‘stressed’ haematopoiesis, which may generate reactive oxygen species. The genetic background of the transduced cells should also be considered as it has been shown that LV insertion leading to Braf activation caused cancer in Cdkn2a/ and Cdkn2a+/ mice, but not in WT mice (Cesana et al., 2014).

Epigenetics

Epigenetic marks on histones play a role in the integration pattern of viral vectors, including the recognition of specific epigenetic marks by integrase-interacting host factors to tether viral intasomes to chromatin. Viral integration sites for HIV-1 correlate with H3K36me3, an epigenetic marker for active transcription units while acetylated H3 and H4 peptides are enriched near TSSs and proto-oncogenes, which are preferred sites for MLV integration (Kvaratskhelia et al., 2014).

Histone modifications are important epigenetic markers of open/closed chromatin states and their distribution, and correlations between insertion sites and histone methylation have been shown (Biasco et al., 2011; Brady et al., 2009; Wang et al., 2010). An open chromatin state and the presence of active DNA binding sites are associated with MLV vector integrations; DNase-I HSSs are related to an open chromatin state and the presence of active DNA-binding sites, and have been associated with insertions in HeLa and CD34+ cells (Biasco et al., 2011; Lewinski et al., 2006; Wang et al., 2010). Biasco et al. (2011) showed MLV integrations in PBLs were on average 2 times closer to HSSs than in HSCs, with the strongest preference of H3K4me3, which is similar to integration preferences observed in the X-SCID trial (Wang et al., 2010). Histone methylation-associated heterochromatic conformations have been shown to be disfavoured by MLV vector integrations, however, such as H3K27me3 in HSCs (Biasco et al., 2011).

Methylation-dependent gene silencing has been observed in a clinical trial for chronic granulomatous disease (CGD), where retroviral promoter methylation correlated with global methylation at the integration site, suggesting epigenetic events at the viral LTR reflected epigenetic marks at the integration site. Moreover, no methylation was observed at the enhancer of the LTR, which the authors concluded may reflect a selective pressure for vector-mediated activation of EVI1 and MDS1-EVI1 (Stein et al., 2010).

Epigenetics also plays a role in AV cell transformation; integration of the viral genome disrupts the virus’ normal regulation and leads to unregulated expression of critical viral genes that can alter cellular gene expression. This results either from indirect- or direct-epigenetic regulation (cell signalling or transcriptional dysregulation, or targeting of epigenetic co-factors eg, histone deacetylases, respectively) (Milavetz and Balakrishnan, 2015).

Age of recipient

The age of the recipient may alter the potential for vector-mediated insertional mutagenesis. For example, AAV-induced insertional mutagenesis has been reported following gene transfer to neonate (Chandler et al., 2015; Donsante et al., 2001, 2007; Walia et al., 2015) but not adult mice (Li et al., 2011). This effect has been attributed to the differing ages of the mice, as a similar cassette was used for all, but integrations into the Rian locus, associated with the resultant insertional mutagenesis, were only observed in neonates (Chandler et al., 2015; Li et al., 2011). This may reflect the fact that Rian is highly expressed early in the neonatal period in mouse liver (Li et al., 2012).

Disease state

The disease state of the patients may also influence the risk of insertional mutagenesis. In SCID-X1 gene therapy .trials, leukaemias developed as a result of integration of γ-RV, (interestingly this was not observed in immune-deficient animals) and the risk of adverse events from dysregulation of proto-oncogenes is significant in immune-deficient patients (Trobridge, 2011).

VIRAL VECTORS: GENOTOXICITY PREVENTION STRATEGIES

Indications from Clinical Trials

A number of clinical trials have been conducted using viral vectors, largely for gene therapy, and details and outcomes of these are listed in Table 1. Unwanted insertions or activation of proto-oncogenes have been observed in several of these trials, leading to the development of cancer in some patients (Table 1). Learnings have been made, and taken in conjunction with what is known about the integration profile of commonly used viral vectors (Table 2), efforts have been made to develop viral vectors with less risk of insertional mutagenesis.

TABLE 1

Indications from Clinical Trials

Viral vector Disease Cell type Transduction Outcome Number of patients Time evident Mechanism References 
γ-RV X-SCID Autologous CD34+ bone marrow cells ex vivo Acute T-cell lymphoid leukaemia 5 (out of 20; 1 died) 2–6 years post-treatment Transcriptional activation of LMO2 proto-oncogene by viral enhancer/promoter sequences present in vector LTR Aiuti and Roncarolo (2009) ; Gaspar et al. (2011); Hacein-Bey-Abina et al. (2010) 
γ-RV Wiskott-Aldrich Syndrome (WAS) Gene-corrected CD34+ cells ex vivo T cell leukaemia; significant clonal imbalances 10  Insertions in proto-oncogenes (LMO2, MDS/EVI1, PRDM16, CCND2) Aiuti and Roncarolo (2009) 
γ-RV CGD Myeloid cells Ex vivo Myelodysplasia Approximately 5 months post-treatment Clusters of integrations in MECOM. Could disrupt normal centrosomes resulting in genomic instability, monosomy 7, clonal progression towards aberrant expression and myelodysplasia Aiuti and Roncarolo (2009); Aiuti et al. (2012); Stein et al. (2010) 
γ-RV WAS Autologous bone-marrow-derived CD34+ cells Ex vivo Acute T cell lymphoblastic leukaemia 2 (out of 10)  Stable engraftment of WAS-expressing cells, improved immune function, amelioration of clinical manifestations Aiuti et al., (2013); Seymour and Thrasher (2012) 
LV X-linked adreno-leukodystrophy (X-ALD) Autologous CD34+ cells Ex vivo 9–14% of peripheral blood cells expressed ABCD1 protein, neurological function stabilised 2 children 2 years No observed clustering of vector insertions in oncogenes or growth-related genes Cesana et al. (2014) 
LV HIV Autologous CD4+ cells (263 infusions)   65 8 years No adverse events detected in subjects McGarrity et al. (2013) 
LV Metachromatic leukodystrophy (MLD) HSC Ex vivo  3 children (pre-symptomatic) Up to 2 years/7-21 months after predicted disease onset Stable and extensive ARSA replacement; No aberrant clonal behaviour/remained asymptomatic Biffi et al. (2013) 
LV Human β-thalassaemia Autologous bone-marrow-derived CD34+ cells Ex vivo Blood haemoglobin maintained between 9 and 10 g dl−1, of which one-third contained vector-encoded β-globin.   Transcriptional activation of HMGA2 in erythroid cells; further increased expression of a truncated HMGA2 mRNA insensitive to degradation by let-7 microRNAs Cavazzana-Calvo et al. (2010) 
Viral vector Disease Cell type Transduction Outcome Number of patients Time evident Mechanism References 
γ-RV X-SCID Autologous CD34+ bone marrow cells ex vivo Acute T-cell lymphoid leukaemia 5 (out of 20; 1 died) 2–6 years post-treatment Transcriptional activation of LMO2 proto-oncogene by viral enhancer/promoter sequences present in vector LTR Aiuti and Roncarolo (2009) ; Gaspar et al. (2011); Hacein-Bey-Abina et al. (2010) 
γ-RV Wiskott-Aldrich Syndrome (WAS) Gene-corrected CD34+ cells ex vivo T cell leukaemia; significant clonal imbalances 10  Insertions in proto-oncogenes (LMO2, MDS/EVI1, PRDM16, CCND2) Aiuti and Roncarolo (2009) 
γ-RV CGD Myeloid cells Ex vivo Myelodysplasia Approximately 5 months post-treatment Clusters of integrations in MECOM. Could disrupt normal centrosomes resulting in genomic instability, monosomy 7, clonal progression towards aberrant expression and myelodysplasia Aiuti and Roncarolo (2009); Aiuti et al. (2012); Stein et al. (2010) 
γ-RV WAS Autologous bone-marrow-derived CD34+ cells Ex vivo Acute T cell lymphoblastic leukaemia 2 (out of 10)  Stable engraftment of WAS-expressing cells, improved immune function, amelioration of clinical manifestations Aiuti et al., (2013); Seymour and Thrasher (2012) 
LV X-linked adreno-leukodystrophy (X-ALD) Autologous CD34+ cells Ex vivo 9–14% of peripheral blood cells expressed ABCD1 protein, neurological function stabilised 2 children 2 years No observed clustering of vector insertions in oncogenes or growth-related genes Cesana et al. (2014) 
LV HIV Autologous CD4+ cells (263 infusions)   65 8 years No adverse events detected in subjects McGarrity et al. (2013) 
LV Metachromatic leukodystrophy (MLD) HSC Ex vivo  3 children (pre-symptomatic) Up to 2 years/7-21 months after predicted disease onset Stable and extensive ARSA replacement; No aberrant clonal behaviour/remained asymptomatic Biffi et al. (2013) 
LV Human β-thalassaemia Autologous bone-marrow-derived CD34+ cells Ex vivo Blood haemoglobin maintained between 9 and 10 g dl−1, of which one-third contained vector-encoded β-globin.   Transcriptional activation of HMGA2 in erythroid cells; further increased expression of a truncated HMGA2 mRNA insensitive to degradation by let-7 microRNAs Cavazzana-Calvo et al. (2010) 
TABLE 2

Integration Profiles of Commonly Used Viral Vectors

 γ-RV LV AV AAV 
Integration site Strong enhancers/active gene promoters Randomly in active genes (within gene-dense regions of chromosomes) Randomly within actively transcribed genes (low frequency) Active genes 
Integration preference Transcript start sites (risk of activating proto-oncogenes) Transcription units of actively transcribed genes  CpG islands, Ribosomal DNA 
Integration profile Unique for each type; largely independent of target cell type Cell cycle status has modest effect on integration site distribution Integration by heterologous > homologous recombination Cell-type dependent 
Cell type Rapidly dividing cells Dividing and non-dividing cells Non-dividing cells Dividing and non-dividing cells 
 γ-RV LV AV AAV 
Integration site Strong enhancers/active gene promoters Randomly in active genes (within gene-dense regions of chromosomes) Randomly within actively transcribed genes (low frequency) Active genes 
Integration preference Transcript start sites (risk of activating proto-oncogenes) Transcription units of actively transcribed genes  CpG islands, Ribosomal DNA 
Integration profile Unique for each type; largely independent of target cell type Cell cycle status has modest effect on integration site distribution Integration by heterologous > homologous recombination Cell-type dependent 
Cell type Rapidly dividing cells Dividing and non-dividing cells Non-dividing cells Dividing and non-dividing cells 

Vector Design

Self-inactivating (SIN) vectors

SIN vectors (LV and γ-RV) lack viral promoter and enhancer activity in their 3′ LTR, via deletion of the U3 in the plasmid during reverse transcription (Hu and Pathak, 2000), limiting the effects of the viral LTR promoter on flanking genes (Niederer and Bangham, 2014) (Figure 2A). It has been demonstrated that SIN-LVs have a lower risk of insertional mutagenesis that γ-RVs (Cesana et al., 2014) and although polyadenylation signals and splice sites are present (Heckl et al., 2012), SIN LVs have been shown to reduce the generation of splice-variant transcripts of genes around LV integration sites (Cesana et al., 2012; Moiani et al., 2012).

Fig. 2

Prevention of genotoxicity through modulation of vector design by (A) SIN vectors or (B) chromatin insulators.

Fig. 2

Prevention of genotoxicity through modulation of vector design by (A) SIN vectors or (B) chromatin insulators.

SIN vectors are not without their problems, however. Although SIN vectors have decreased genotoxic potential, a shift from promoter activation to gene transcript truncation as the predominant genotoxicity mechanism has been observed (Suerth et al., 2014). SIN LVs contain internal enhancers and promoters in order to express the therapeutic gene (Aiuti et al., 2013; Heckl et al., 2012; Modlich et al., 2006), and these have the potential to affect non-target genes. Those using strong enhancers or promoters can transform cells by insertional activation, although their transforming capacity was found to be significantly reduced compared with corresponding LTR vectors (Modlich et al., 2006). Where tumors have been observed, their induction was proportional to promoter strength (Cesana et al., 2014), advocating the use of weaker internal promoters, and the choice of moderately-active versus strong promoters decreases the risk of insertional mutagenesis (Aiuti et al., 2013). However, upon reduction of enhancer activity via the use of moderate promoters, tumour suppressor gene inactivation has been observed, leading to oncogenesis. Even with weak internal promoters/insulators, aberrant or truncated transcripts can form (Cesana et al., 2012; Suerth et al., 2014) from read-though transcription (reportedly high in SIN LVs) (Yang et al., 2007; Zaiss et al., 2002) originating inside and outside the pro-virus, and vector sequences can contribute to the aberrant splicing process. However, it is ultimately the complex interaction between the presence, relative strength, position, and distance of promoters, and mRNA polyadenylation signals that generates specific aberrant mRNAs (Cesana et al., 2012).

Chromatin insulators

Chromatin insulators block the interaction between an integrating vector and the target cell genome independently of vector transgene, regulatory elements, or the virus. There are 2 classes of chromatin insulators, enhancer-blocking insulators, and barrier insulators (Figure 2B(i) and (ii)). Enhancer-blocking insulators are the most common, and prevent enhancer-mediated transcriptional activation of adjoining heterochromatic or transcriptionally silent regions. Barrier insulators block the encroachment of silencing heterochromatin into transcriptionally active adjoining regions of open chromatin. Enhancer-blocking insulators must be located between the vector enhancer and cellular gene promoter, while barrier insulators are situated between the vector expression cassette and the source of silencing heterochromatin. Barrier insulators help reduce the rate of vector silencing, enabling a lower dose for therapeutic efficacy, thereby reducing the risk of vector-mediated genotoxicity (Emery, 2011). Pre-clinical studies have demonstrated that viral vector genotoxicity can be dose-dependent (although not evident in the clinic), therefore decreasing the viral vector dose may help reduce the potential for genotoxicity (Emery, 2011). Activation of cellular genes has been shown to be a component of vector-mediated genotoxicity, and enhancer-blocking chromatin insulators would block this (Emery, 2011). Therefore the use of a combined chromatin insulator (eg, cHS4) may be the most useful. The chromatin barrier functionality of cHS4 has been shown to improve expression of viral vectors by protecting them from chromosomal position effects (Arumugam et al., 2007; Emery et al., 2000; Uchida et al., 2013), while activation of proximal regulatory elements and promoters was reduced through the enhancer-blocking functionality (Li et al., 2009; Ryu et al., 2007; Uchida et al., 2013).

Use of chromatin insulators has been demonstrated to reduce the risk of vector-mediated genotoxicity in cell culture; cHS4 significantly reduced LTR-mediated activation of LMO-2 (Ryu et al., 2008), a genome-wide assessment of cellular gene dysregulation in γ-RV-transduced cell clones showed that the cHS4 insulator reduced the frequency of dysregulated cellular genes by approximately 6-fold (Li et al., 2009), and a 1.2 kb version of cHS4 reduced the number of dominant clones in γ-RV-transduced Jurkat cells (Evans-Galea et al., 2007). Chromatin insulators have been less-well studied in other viral vectors, but in AV, insulators were shown to prevent inappropriate activation of regulated or tissue-specific internal expression cassettes by viral or other vector enhancers (Martin-Duque et al., 2004; Steinwaerder and Lieber, 2000) and improve vector transgene expression (Ye et al., 2003).

Chromatin insulators may not be sufficiently effective alone, however, since other mechanisms of vector-mediated genotoxicity, including transcriptional read-through and alternative splicing, are not prevented by chromatin insulators (Emery, 2011). Moreover, it has been shown in Drosophila that in certain orientations chromatin insulators can enhance vector-mediated genotoxicity by disrupting native insulator networks (Savitskaya et al., 2006). Chromatin insulators may also impact on vector titres, although the data are conflicting. For γ-RV, both no effect (Ramezani et al., 2006) and a 2–6-fold reduction in titres (van Meerten et al., 2006) have been reported, while for LV, both a 6–20-fold reduction in vector titres depending on vector design (Nielsen et al., 2009) and no effect on titre (Liu et al., 2015; Osti et al., 2006) have been reported. This highlights the need to consider both vector design and the insulator incorporated to minimize the effect on vector titre. Another disadvantage, specifically of the fully active cHS4 element (1.2 kb), is its large size, limiting its use. Recently, compact sequence elements have been identified that function as highly potent insulators (reduced risk of LV-mediated carcinogenesis was shown in a mouse model [Liu et al., 2015]) possibly with greater functionality than cHS4, and as they are short (119–284 bp) do not take up much space in the viral vector (Liu et al., 2015).

Polyadenylation signals

The polyadenylation signal is required for correct transcript termination and is located within the R region of the LTR in γ-RVs and LVs. Incorrect termination may result in read-through of retroviral sequences that have been randomly inserted in cellular genes (Schambach et al., 2007), which could increase the risk of activating oncogenes that are positioned 3′ to the viral integration site (Schambach et al., 2007; Zaiss et al., 2002). RV vectors show a high frequency of transcriptional read-through of the 3′ polyadenylation signal in the 3′ LTR as a result of inefficient polyadenylation of viral genomic mRNA (Higashimoto et al., 2007). This highlights the importance of polyadenylation signals when considering viral vector safety (Schambach et al., 2007; Zaiss et al., 2002). Indeed, improving vector transcript termination efficiency can reduce transcription read-through, and increase the level of vector expression, while also improving viral titres (Higashimoto et al., 2007; Schambach et al., 2007).

Tissue-specific promoter

Although viral vectors that contain tissue-specific promoters have been designed to regulate the expression of a gene of interest, this feature may also help in reducing genotoxicity, as the risk of integration in non-target tissues would be reduced. Moreover, inclusion of chromatin insulators has been suggested to improve the tissue-specificity and/or activity of tissue-specific promoters in AV (Steinwaerder and Lieber, 2000; Ye et al., 2003), potentially further contributing to a reduced genotoxicity risk.

MicroRNA control of vectors

MicroRNAs (miRNA) provide a way to target viral vector delivery. For example, an oncolytic virus containing p53-dependent miRNAs that target AV genes can replicate in p53-negative tumors and exert its oncolytic function, but in p53-positive tissues p53 miRNAs are produced, which target AV genes and suppress viral replication and infection, ensuring the virus is only active in the tumour (Gürlevik et al., 2009). Additionally, oncolytic viruses have been developed that contain miR-122 binding sites, and high expression of miR-122 in the liver inhibits viral replication in the liver (substantially reducing liver toxicity) while still permitting replication in other cells (Cawood et al., 2009; Ylösmäki et al., 2008). These approaches may help to reduce viral dose and the integration of viral vectors in non-target tissues, thereby reducing insertional mutagenesis.

Vector dose

The risk of insertional mutagenesis increases with the total number of integrations or transduced cells infused. Reducing the dose of viral vector required for efficacy would help reduce the potential for insertional mutagenesis. A number of strategies outlined above to reduce insertional mutagenesis also reduce the required vector dose, which likely contributes to the reduced risk of insertional mutagenesis.

Integrating promoterless vectors

Recently, promoterless vectors have been shown to have a favourable safety profile in regard to vector-borne promoter gene activation, as the lack of a vector-borne promoter reduces the chance of neighboring oncogene activation, while still having high gene-targeting rates (Barzel et al., 2015).

CHALLENGES FOR GENOTOXICITY TESTING

The current genotoxicity testing strategy relies largely on detecting effects on DNA (damage or mutation) following a relatively short exposure, and expression period for mutation. These are of limited use when considering vector-mediated insertional mutagenesis, which can take weeks, months or even years to manifest in patients. Moreover, Bokhoven et al. (2009) suggest viral vector-induced genotoxicity testing requires an assay that can quantitate insertional mutagenesis reproducibly, by measuring gain of gene function (Bokhoven et al., 2009).

In vivo experiments in mice to measure RV-induced insertional oncogenesis have been conducted, but these are time consuming and expensive (Bokhoven et al., 2009; Du et al., 2005; Li et al., 2007; Montini et al., 2006; Shou et al., 2006). In vitro cell culture assays have also been developed, including one that measures γ-RV insertion in, and upregulation of, the Evi1 gene in primary mouse bone marrow cells, which are cultured for an extended period (Modlich et al., 2006), and another that measures the frequency of interleukin-3 (IL-3)-independent mutants in a murine cell line that is normally IL-3-dependent (Bokhoven et al., 2009). High-throughput insertional mutagenesis screens in mice have been used for many years for cancer gene discovery (Kool and Berns, 2009), and these could potentially be applied to safety testing of viral vectors.

One caveat with these in vivo and in vitro assays is that all utilized mice or mouse-derived cells respectively, therefore relevance to humans is questioned and it is critical to understand whether the likelihood and position of viral vector insertion would be the same in mice as in humans. Importantly, common integration sites in mice show a significant overlap with human oncogenes and tumour suppressor genes (Kool and Berns, 2009), suggesting the utility of mice or mouse cell assays for detecting viral vector-mediated insertional mutagenesis. Consideration should still be given to the scope for detecting less common insertion sites, however.

Given the slow incubation period of viral vectors, exposure and incubation times must be considered, as direct effects may be delayed, thus detecting them in the standard genetic toxicology assay battery would be difficult. Indeed, a time period of 4–5 weeks was required for insertional mutagenesis to be observed in murine bone marrow cells in vitro (Modlich et al., 2006), far in excess of the expression period for mutation in the mouse lymphoma assay (6–8 days) (OECD, 1997). The timescales of in vivo assays is also a consideration: analyses to determine integration sites and tumour development following introduction of transduced-HSCs to tumour prone Cdkn2a−/− mice were conducted 6 weeks post-transplantation (Montini et al., 2006). Assay improvements are also required for assessment of non-viral vector risks for insertional mutagenesis. Such improvements could include using recipient population-derived cells to account for any disease- and/or age-related effects, and using 3D models that may enhance the in vivo relevance of in vitro models and support longer term culture.

APPLICATIONS FOR CRISPR-CAS9

The advent of new technologies such as CRISPR-Cas9 has sparked investigation into appropriate delivery systems, and several groups have devised and tested many viral vectors as delivery systems, since the small size of the 2 core components of CRISPR-Cas9 make it compatible with almost any viral vector. Viral vector delivery of CRISPR-Cas9 was extensively reviewed by Schmidt and Grimm (2015), who conclude that there will not be a single viral vector system for all applications, but rather the vector will be selected on a case-by-case basis, considering factors such as integrating capability, packaging capacity, vector specificity, and safety.

FUTURE DIRECTIONS

A number of factors are associated with the risk of viral vector-mediated insertional mutagenesis (viral and non-viral) and it is clear that there is not a “one size fits all” approach to vector modification for reducing genotoxicity. It is imperative that the choice of viral vector is carefully considered in relation to factors including its insertion profile, and the inclusion of measures demonstrated to reduce the risk of insertional mutagenesis such as SIN vectors, chromatin insulators, or miRNA-targeted delivery. Although some factors that affect the risk of vector-mediated insertional mutagenesis are not modifiable (eg, disease status of the patient, the nature of the inserted transgene), having an awareness of the risks for genotoxicity associated with these is nevertheless important. The choice of test models is also imperative to ensure they are indicative of efficacy and safety, but the development of effective test systems to detect viral vector-induced genotoxicity, particularly considering the often delayed onset, represents a major challenge. Moreover, designing in vitro and in vivo assays to include assessment of non-viral vector risk factors, such as age and disease state, would improve the pre-clinical safety assessment of viral vectors and go some way towards a patient-centric risk assessment.

REFERENCES

Aiuti
A.
Bacchetta
R.
Seger
R.
Villa
A.
Cavazzana-Calvo
M.
(
2012
).
Gene therapy for primary immunodeficiencies: Part 2
.
Curr. Opin. Immunol.
 
24
,
585
591
.
Aiuti
A.
Cassani
B.
Andolfi
G.
Mirolo
M.
Biasco
L.
Recchia
A.
Urbinati
F.
Valacca
C.
Scaramuzza
S.
Aker
M.
(
2007
).
Multilineage hematopoietic reconstitution without clonal selection in ADA-SCID patients treated with stem cell gene therapy
.
J. Clin. Invest.
 
117
,
2233.
Aiuti
A.
Cossu
G.
de Felipe
P.
Galli
M.
Narayanan
G.
Renner
M.
Axel
S.
Schneider
C. K.
Caroline
V. -G.
(
2013
).
The Committee for Advanced Therapies’ of the European Medicines Agency Reflection Paper on Management of Clinical Risks Deriving from Insertional Mutagenesis
.
Hum. Gene Ther. Clin. Dev
 .
24
,
47
54
.
Aiuti
A.
Roncarolo
M. G.
(
2009
).
Ten years of gene therapy for primary immune deficiencies
.
ASH Educ. Progr. Book
 
2009
,
682
689
.
Arumugam
P. I.
Scholes
J.
Perelman
N.
Xia
P.
Yee
J. K.
Malik
P.
(
2007
).
improved human [beta]-globin expression from self-inactivating lentiviral vectors carrying the chicken hypersensitive site-4 (cHS4) insulator element
.
Mol. Ther.
 
15
,
1863
1871
.
Balaggan
K. S.
Ali
R. R.
(
2012
).
Ocular gene delivery using lentiviral vectors
.
Gene Ther.
 
19
,
145
153
. Available at: http://www.nature.com/gt/journal/v19/n2/suppinfo/gt2011153s1.html.
Baldo
A.
van den Akker
E.
Bergmans
H. E.
Lim
F.
Pauwels
K.
(
2013
).
General considerations on the biosafety of virus-derived vectors used in gene therapy and vaccination
.
Curr. Gene Ther.
 
13
,
385
394
.
Bartholomae
C. C.
Arens
A.
Balaggan
K. S.
Yanez-Munoz
R. J.
Montini
E.
Howe
S. J.
Paruzynski
A.
Korn
B.
Appelt
J. U.
MacNeil
A.
et al.  . (
2011
).
Lentiviral vector integration profiles differ in rodent postmitotic tissues
.
Mol. Ther.
 
19
,
703
710
. Available at: http://www.nature.com/mt/journal/v19/n4/suppinfo/mt201119s1.html.
Barzel
A.
Paulk
N. K.
Shi
Y.
Huang
Y.
Chu
K.
Zhang
F.
Valdmanis
P. N.
Spector
L. P.
Porteus
M. H.
Gaensler
K. M.
et al.  . (
2015
).
Promoterless gene targeting without nucleases ameliorates haemophilia B in mice
.
Nature
 ,
517
,
360
364
.
Biasco
L.
Ambrosi
A.
Pellin
D.
Bartholomae
C.
Brigida
I.
Roncarolo
M. G.
Di Serio
C.
von Kalle
C.
Schmidt
M.
Aiuti
A.
(
2011
).
Integration profile of retroviral vector in gene therapy treated patients is cell-specific according to gene expression and chromatin conformation of target cell
.
EMBO Mol. Med.
 
3
,
89
101
. CrossRef][10.1002/emmm.201000108]
Biffi
A.
Montini
E.
Lorioli
L.
Cesani
M.
Fumagalli
F.
Plati
T.
Baldoli
C.
Martino
S.
Calabria
A.
Canale
S.
et al.  . (
2013
).
Lentiviral hematopoietic stem cell gene therapy benefits metachromatic leukodystrophy
.
Science
 
341
,
Bokhoven
M.
Stephen
S. L.
Knight
S.
Gevers
E. F.
Robinson
I. C.
Takeuchi
Y.
Collins
M. K.
(
2009
).
Insertional gene activation by lentiviral and gammaretroviral vectors
.
J. Virol.
 
83
,
283
294
. [18945765]
Brady
T.
Agosto
L. M.
Malani
N.
Berry
C. C.
O'Doherty
U.
Bushman
F.
(
2009
).
HIV integration site distributions in resting and activated CD4(+) T cells infected in culture
.
AIDS (London, England)
 ,
23
,
1461
1471
.
Cattoglio
C.
Pellin
D.
Rizzi
E.
Maruggi
G.
Corti
G.
Miselli
F.
Sartori
D.
Guffanti
A.
Di Serio
C.
Ambrosi
A.
et al.  . (
2010
).
High-definition mapping of retroviral integration sites identifies active regulatory elements in human multipotent hematopoietic progenitors
.
Blood
 
116
,
5507
5517
.
Cavazzana-Calvo
M.
Payen
E.
Negre
O.
Wang
G.
Hehir
K.
Fusil
F.
Down
J.
Denaro
M.
Brady
T.
Westerman
K.
et al.  . (
2010
).
Transfusion independence and HMGA2 activation after gene therapy of human [bgr]-thalassaemia
.
Nature
 ,
467
,
318
322
. Available at: http://www.nature.com/nature/journal/v467/n7313/abs/nature09328.html#supplementary-information.
Cawood
R.
Chen
H. H.
Carroll
F.
Bazan-Peregrino
M.
van Rooijen
N.
Seymour
L. W.
(
2009
).
Use of tissue-specific microrna to control pathology of wild-type adenovirus without attenuation of its ability to kill cancer cells
.
PLoS Pathog.
 
5
,
e1000440.
Cesana
D.
Ranzani
M.
Volpin
M.
Bartholomae
C.
Duros
C.
Artus
A.
Merella
S.
Benedicenti
F.
Sergi Sergi
L.
Sanvito
F.
et al.  . (
2014
).
Uncovering and dissecting the genotoxicity of self-inactivating lentiviral vectors in vivo
.
Mol. Ther.
 
22
,
774
785
.
Cesana
D.
Sgualdino
J.
Rudilosso
L.
Merella
S.
Naldini
L.
Montini
E.
(
2012
).
Whole transcriptome characterization of aberrant splicing events induced by lentiviral vector integrations
.
J. Clin. Invest.
 
122
,
1667
1676
.
Chandler
R. J.
LaFave
M. C.
Varshney
G. K.
Trivedi
N. S.
Carrillo-Carrasco
N.
Senac
J. S.
Wu
W.
Hoffmann
V.
Elkahloun
A. G.
Burgess
S. M.
et al.  . (
2015
).
Vector design influences hepatic genotoxicity after adeno-associated virus gene therapy
.
J. Clin. Invest.
 
125
,
870
880
.
Cui
K.
Zang
C.
Roh
T. Y.
Schones
D. E.
Childs
R. W.
Peng
W.
Zhao
K.
(
2009
).
Chromatin signatures in multipotent human hematopoietic stem cells indicate the fate of bivalent genes during differentiation
.
Cell Stem Cell
 
4
,
80
93
.
De Ravin
S. S.
Su
L.
Theobald
N.
Choi
U.
Macpherson
J. L.
Poidinger
M.
Symonds
G.
Pond
S. M.
Ferris
A. L.
Hughes
S. H.
et al.  . (
2014
).
Enhancers are major targets for murine leukemia virus vector integration
.
J. Virol.
 
88
,
4504
4513
.
Donsante
A.
Miller
D. G.
Li
Y.
Vogler
C.
Brunt
E. M.
Russell
D. W.
Sands
M. S.
(
2007
).
AAV vector integration sites in mouse hepatocellular carcinoma
.
Science
 
317
,
477
477
.
Donsante
A.
Vogler
C.
Muzyczka
N.
Crawford
J.
Barker
J.
Flotte
T.
Campbell-Thompson
M.
Daly
T.
Sands
M.
(
2001
).
Observed incidence of tumorigenesis in long-term rodent studies of rAAV vectors
.
Gene Ther.
 
8
,
1343
1346
.
Du
Y.
Spence
S. E.
Jenkins
N. A.
Copeland
N. G.
(
2005
).
Cooperating cancer-gene identification through oncogenic-retrovirus–induced insertional mutagenesis
.
Blood
 
106
,
2498
2505
.
Emery
D. W.
(
2011
).
The use of chromatin insulators to improve the expression and safety of integrating gene transfer vectors
.
Hum. Gene Ther.
 
22
,
761
774
.
Emery
D. W.
Yannaki
E.
Tubb
J.
Stamatoyannopoulos
G.
(
2000
).
A chromatin insulator protects retrovirus vectors from chromosomal position effects
.
Proc. Natl. Acad. Sci. USA
 
97
,
9150
9155
.
Evans-Galea
M. V.
Wielgosz
M. M.
Hanawa
H.
Srivastava
D. K.
Nienhuis
A. W.
(
2007
).
Suppression of clonal dominance in cultured human lymphoid cells by Addition of the cHS4 Insulator to a lentiviral vector
.
Mol. Ther.
 
15
,
801
809
.
Gaspar
H. B.
Cooray
S.
Gilmour
K. C.
Parsley
K. L.
Adams
S.
Howe
S. J.
Al Ghonaium
A.
Bayford
J.
Brown
L.
Davies
E. G.
(
2011
).
Long-term persistence of a polyclonal T cell repertoire after gene therapy for X-linked severe combined immunodeficiency
.
Sci. Transl. Med.
 
3
,
97ra79
97ra79
.
Gürlevik
E.
Woller
N.
Schache
P.
Malek
N. P.
Wirth
T. C.
Zender
L.
Manns
M. P.
Kubicka
S.
Kühnel
F.
(
2009
).
p53-dependent antiviral RNA-interference facilitates tumor-selective viral replication
.
Nucleic Acids Res.
 
37
,
e84.
Hacein-Bey-Abina
S.
Hauer
J.
Lim
A.
Picard
C.
Wang
G. P.
Berry
C. C.
Martinache
C.
Rieux-Laucat
F.
Latour
S.
Belohradsky
B. H.
et al.  . (
2010
).
Efficacy of Gene Therapy for X-Linked Severe Combined Immunodeficiency
.
N. Engl. J. Med.
 
363
,
355
364
.
Heckl
D.
Schwarzer
A.
Haemmerle
R.
Steinemann
D.
Rudolph
C.
Skawran
B.
Knoess
S.
Krause
J.
Li
Z.
Schlegelberger
B.
et al.  . (
2012
).
Lentiviral vector induced insertional haploinsufficiency of Ebf1 causes murine leukemia
.
Mol. Ther.
 
20
,
1187
1195
. Available at: http://www.nature.com/mt/journal/v20/n6/suppinfo/mt201259s1.html.
Higashimoto
T.
Urbinati
F.
Perumbeti
A.
Jiang
G.
Zarzuela
A.
Chang
L.
Kohn
D.
Malik
P.
(
2007
).
The woodchuck hepatitis virus post-transcriptional regulatory element reduces readthrough transcription from retroviral vectors
.
Gene Ther.
 
14
,
1298
1304
.
Hu
W. S.
Pathak
V. K.
(
2000
).
Design of retroviral vectors and helper cells for gene therapy
.
Pharmacol. Rev.
 
52
,
493
512
.
Kool
J.
Berns
A.
(
2009
).
High-throughput insertional mutagenesis screens in mice to identify oncogenic networks
.
Nat. Rev. Cancer
 
9
,
389
399
.
Kvaratskhelia
M.
Sharma
A.
Larue
R. C.
Serrao
E.
Engelman
A.
(
2014
).
Molecular mechanisms of retroviral integration site selection
.
Nucleic Acids Res.
 
42
,
10209
10225
.
LaFave
M. C.
Varshney
G. K.
Gildea
D. E.
Wolfsberg
T. G.
Baxevanis
A. D.
Burgess
S. M.
(
2014
).
MLV integration site selection is driven by strong enhancers and active promoters
.
Nucleic Acids Res.
 
42
,
4257
4269
.
Lewinski
M. K.
Yamashita
M.
Emerman
M.
Ciuffi
A.
Marshall
H.
Crawford
G.
Collins
F.
Shinn
P.
Leipzig
J.
Hannenhalli
S.
et al.  . (
2006
).
Retroviral DNA Integration: Viral and Cellular Determinants of Target-Site Selection
.
PLoS Pathog.
 
2
,
e60.
Li
C.
Yu
S.
Zhong
X.
Wu
J.
Li
X.
(
2012
).
Transcriptome comparison between fetal and adult mouse livers: implications for circadian clock mechanisms
.
PLoS One
 ,
7
,
e31292.
Li
C. L.
Xiong
D.
Stamatoyannopoulos
G.
Emery
D. W.
(
2009
).
Genomic and functional assays demonstrate reduced gammaretroviral vector genotoxicity associated with use of the cHS4 chromatin insulator
.
Mol. Ther.
 
17
,
716
724
. Available at: http://www.nature.com/mt/journal/v17/n4/suppinfo/mt20097s1.html.
Li
H.
Malani
N.
Hamilton
S. R.
Schlachterman
A.
Bussadori
G.
Edmonson
S. E.
Shah
R.
Arruda
V. R.
Mingozzi
F.
Fraser Wright
J.
et al.  . (
2011
). Assessing the potential for AAV vector genotoxicity in a murine model.
Li
Z.
Kustikova
O. S.
Kamino
K.
Neumann
T.
Rhein
M.
Grassman
E.
Fehse
B.
Baum
C.
(
2007
).
Insertional mutagenesis by replication‐deficient retroviral vectors encoding the large T oncogene
.
Ann. N. Y. Acad. Sci.
 
1106
,
95
113
.
Liu
M.
Maurano
M. T.
Wang
H.
Qi
H.
Song
C. Z.
Navas
P. A.
Emery
D. W.
Stamatoyannopoulos
J. A.
Stamatoyannopoulos
G.
(
2015
).
Genomic discovery of potent chromatin insulators for human gene therapy
.
Nat. Biotech.
 
33
,
198
203
. Available at: http://www.nature.com/nbt/journal/v33/n2/abs/nbt.3062.html#supplementary-information.
Martin-Duque
P.
Jezzard
S.
Kaftansis
L.
Vassaux
G.
(
2004
).
Direct comparison of the insulating properties of two genetic elements in an adenoviral vector containing two different expression cassettes
.
Hum. Gene Ther.
 
15
,
995
1002
.
McGarrity
G. J.
Hoyah
G.
Winemiller
A.
Andre
K.
Stein
D.
Blick
G.
Greenberg
R. N.
Kinder
C.
Zolopa
A.
Binder‐Scholl
G.
(
2013
).
Patient monitoring and follow‐up in lentiviral clinical trials
.
J. Gene Med.
 
15
,
78
82
.
Milavetz
B. I.
Balakrishnan
L.
(
2015
).
Viral epigenetics
.
Methods Mol. Biol. (Clifton, N.J.)
 
1238
,
569
596
.
Mitchell
R. S.
Beitzel
B. F.
Schroder
A. R. W.
Shinn
P.
Chen
H.
Berry
C. C.
Ecker
J. R.
Bushman
F. D.
(
2004
).
Retroviral DNA integration: ASLV, HIV, and MLV show distinct target site preferences
.
PLoS Biol.
 
2
,
e234.
Modlich
U.
Bohne
J.
Schmidt
M.
von Kalle
C.
Knöss
S.
Schambach
A.
Baum
C.
(
2006
).
Cell-culture assays reveal the importance of retroviral vector design for insertional genotoxicity
.
Blood
 
108
,
2545
2553
. CrossRef][10.1182/blood-2005-08-024976]
Moiani
A.
Paleari
Y.
Sartori
D.
Mezzadra
R.
Miccio
A.
Cattoglio
C.
Cocchiarella
F.
Lidonnici
M. R.
Ferrari
G.
Mavilio
F.
(
2012
).
Lentiviral vector integration in the human genome induces alternative splicing and generates aberrant transcripts
.
J. Clin. Invest.
 
122
,
1653
1666
.
Montini
E.
Cesana
D.
Schmidt
M.
Sanvito
F.
Bartholomae
C. C.
Ranzani
M.
Benedicenti
F.
Sergi
L. S.
Ambrosi
A.
Ponzoni
M.
et al.  . (
2009
).
The genotoxic potential of retroviral vectors is strongly modulated by vector design and integration site selection in a mouse model of HSC gene therapy
.
J. Clin. Invest.
 
119
,
964
975
.
Montini
E.
Cesana
D.
Schmidt
M.
Sanvito
F.
Ponzoni
M.
Bartholomae
C.
Sergi
L. S.
Benedicenti
F.
Ambrosi
A.
Di Serio
C.
et al.  . (
2006
).
Hematopoietic stem cell gene transfer in a tumor-prone mouse model uncovers low genotoxicity of lentiviral vector integration
.
Nat. Biotech.
 
24
,
687
696
. Available at: http://www.nature.com/nbt/journal/v24/n6/suppinfo/nbt1216_S1.html.
Niederer
H.
Bangham
C.
(
2014
).
Integration site and clonal expansion in human chronic retroviral infection and gene therapy
.
Viruses
 
6
,
4140.
Nielsen
T. T.
Jakobsson
J.
Rosenqvist
N.
Lundberg
C.
(
2009
).
Incorporating double copies of a chromatin insulator into lentiviral vectors results in less viral integrants
.
BMC Biotechnol.
 
9
,
1
12
.
OECD
. (
1997
). OECD GUIDELINE FOR THE TESTING OF CHEMICALS: In Vitro Mammalian Cell Gene Mutation Test. In (doi.
Osti
D.
Marras
E.
Ceriani
I.
Grassini
G.
Rubino
T.
Viganò
D.
Parolaro
D.
Perletti
G.
(
2006
).
Comparative analysis of molecular strategies attenuating positional effects in lentiviral vectors carrying multiple genes
.
J. Virol. Methods
 
136
,
93
101
.
Rae
D. T.
Trobridge
G. D.
(
2013
). Retroviral genotoxicity, gene therapy - tools and potential applications. In InTech (D. F. Martin, Ed.) Eds.) doi: 10.5772/52530.
Ramezani
A.
Hawley
T. S.
Hawley
R. G.
(
2006
).
Stable gammaretroviral vector expression during embryonic stem cell-derived in vitro hematopoietic development
.
Mol. Ther.
 
14
,
245
254
.
Ryu
B. Y.
Evans-Galea
M. V.
Gray
J. T.
Bodine
D. M.
Persons
D. A.
Nienhuis
A. W.
(
2008
).
An experimental system for the evaluation of retroviral vector design to diminish the risk for proto-oncogene activation
.
Blood
 
111
,
1866
1875
.
Ryu
B. Y.
Persons
D. A.
Evans-Galea
M. V.
Gray
J. T.
Nienhuis
A. W.
(
2007
).
A chromatin insulator blocks interactions between globin regulatory elements and cellular promoters in erythroid cells
.
Blood Cells Mol. Dis.
 
39
,
221
228
. Available at: http://dx.doi.org/10.1016/j.bcmd.2007.05.003.
Savitskaya
E.
Melnikova
L.
Kostuchenko
M.
Kravchenko
E.
Pomerantseva
E.
Boikova
T.
Chetverina
D.
Parshikov
A.
Zobacheva
P.
Gracheva
E.
(
2006
).
Study of long-distance functional interactions between Su (Hw) insulators that can regulate enhancer-promoter communication in Drosophila melanogaster
.
Mol. Cell. Biol.
 
26
,
754
761
.
Schambach
A.
Galla
M.
Maetzig
T.
Loew
R.
Baum
C.
(
2007
).
Improving transcriptional termination of self-inactivating gamma-retroviral and lentiviral vectors
.
Mol. Ther.
 
15
,
1167
1173
.
Schmidt
F.
Grimm
D.
(
2015
).
CRISPR genome engineering and viral gene delivery: A case of mutual attraction
.
Biotechnol. J.
 
10
,
258
272
.
Schröder
A. R. W.
Shinn
P.
Chen
H.
Berry
C.
Ecker
J. R.
Bushman
F.
(
2002
).
HIV-1 integration in the human genome favors active genes and local hotspots
.
Cell
 
110
,
521
529
. Available at: http://dx.doi.org/10.1016/S0092-8674(02)00864-4.
Seymour
L. W.
Thrasher
A. J.
(
2012
).
Gene therapy matures in the clinic
.
Nat. Biotechnol.
 
30
,
588
593
.
Shou
Y.
Ma
Z.
Lu
T.
Sorrentino
B. P.
(
2006
).
Unique risk factors for insertional mutagenesis in a mouse model of XSCID gene therapy
.
Proc. Natl. Acad. Sci. USA
 
103
,
11730
11735
.
Sokol
M.
Wabl
M.
Ruiz
I. R.
Pedersen
F. S.
(
2014
).
Novel principles of gamma-retroviral insertional transcription activation in murine leukemia virus-induced end-stage tumors
.
Retrovirology
 
11
,
1
20
.
Stein
S.
Ott
M. G.
Schultze-Strasser
S.
Jauch
A.
Burwinkel
B.
Kinner
A.
Schmidt
M.
Kramer
A.
Schwable
J.
Glimm
H.
et al.  . (
2010
).
Genomic instability and myelodysplasia with monosomy 7 consequent to EVI1 activation after gene therapy for chronic granulomatous disease
.
Nat. Med.
 
16
,
198
204
. Available at: http://www.nature.com/nm/journal/v16/n2/suppinfo/nm.2088_S1.html.
Steinwaerder
D.
Lieber
A.
(
2000
).
Insulation from viral transcriptional regulatory elements improves inducible transgene expression from adenovirus vectors in vitro and in vivo
.
Gene Ther.
 
7
,
Suerth
J.
Labenski
V.
Schambach
A.
(
2014
).
Alpharetroviral vectors: From a cancer-causing agent to a useful tool for human gene therapy
.
Viruses
 
6
,
4811.
Themis
M.
Waddington
S. N.
Schmidt
M.
von Kalle
C.
Wang
Y.
Al-Allaf
F.
Gregory
L. G.
Nivsarkar
M.
Themis
M.
Holder
M. V.
et al.  . (
2005
).
Oncogenesis following delivery of a nonprimate lentiviral gene therapy vector to fetal and neonatal mice
.
Mol. Ther.
 
12
,
763
771
. Available at: http://www.nature.com/mt/journal/v12/n4/suppinfo/mt20051355s1.html.
Touw
I. P.
Erkeland
S. J.
(
2007
).
Retroviral insertion mutagenesis in mice as a comparative oncogenomics tool to identify disease genes in human leukemia
.
Mol. Ther.
 
15
,
13
19
.
Trobridge
G. D.
(
2011
).
Genotoxicity of retroviral hematopoietic stem cell gene therapy
.
Exp. Opin. Biol. Ther.
 
11
,
581
593
.
Uchida
N.
Hanawa
H.
Yamamoto
M.
Shimada
T.
(
2013
).
The chicken hypersensitivity site 4 core insulator blocks promoter interference in lentiviral vectors
.
Hum. Gene Ther. Methods
 
24
,
117
124
.
van Meerten
T.
Claessen
M. J.
Hagenbeek
A.
Ebeling
S. B.
(
2006
).
The CD20/αCD20 ‘suicide’system: novel vectors with improved safety and expression profiles and efficient elimination of CD20-transgenic T cells
.
Gene Ther.
 
13
,
789
797
.
Walia
J. S.
Altaleb
N.
Bello
A.
Kruck
C.
LaFave
M. C.
Varshney
G. K.
Burgess
S. M.
Chowdhury
B.
Hurlbut
D.
Hemming
R.
et al.  . (
2015
).
Long-term correction of sandhoff disease following intravenous delivery of rAAV9 to mouse neonates
.
Mol. Ther.
 
23
,
414
422
.
Wang
G. P.
Berry
C. C.
Malani
N.
Leboulch
P.
Fischer
A.
Hacein-Bey-Abina
S.
Cavazzana-Calvo
M.
Bushman
F. D.
(
2010
).
Dynamics of gene-modified progenitor cells analyzed by tracking retroviral integration sites in a human SCID-X1 gene therapy trial
.
Blood
 
115
,
4356
4366
.
Wang
G. P.
Levine
B. L.
Binder
G. K.
Berry
C. C.
Malani
N.
McGarrity
G.
Tebas
P.
June
C. H.
Bushman
F. D.
(
2009
).
Analysis of lentiviral vector integration in HIV+ study subjects receiving autologous infusions of gene modified CD4+ T Cells
.
Mol. Ther.
 
17
,
844
850
.
Wu
X.
Li
Y.
Crise
B.
Burgess
S. M.
(
2003
).
Transcription start regions in the human genome are favored targets for MLV integration
.
Science
 ,
300
,
1749
1751,
Yang
Q.
Lucas
A.
Son
S.
Chang
L. J.
(
2007
).
Overlapping enhancer/promoter and transcriptional termination signals in the lentiviral long terminal repeat
.
Retrovirology
 
4
,
4
4
.
Ye
X.
Liang
M.
Meng
X.
Ren
X.
Chen
H.
Li
Z. Y.
Ni
S.
Lieber
A.
Hu
F.
(
2003
).
Insulation from viral transcriptional regulatory elements enables improvement to hepatoma-specific gene expression from adenovirus vectors
.
Biochem. Biophys. Res. Commun.
 
307
,
759
764
.
Yi
Y.
Jong Noh
M.
Hee Lee
K.
(
2011
).
Current advances in retroviral gene therapy
.
Curr. Gene Ther.
 
11
,
218
228
.
Ylösmäki
E.
Hakkarainen
T.
Hemminki
A.
Visakorpi
T.
Andino
R.
Saksela
K.
(
2008
).
Generation of a conditionally replicating adenovirus based on targeted destruction of E1A mRNA by a cell type-specific microRNA
.
J. Virol.
 
82
,
11009
11015
.
Zaiss
A. K.
Son
S.
Chang
L. J.
(
2002
).
RNA 3′ readthrough of oncoretrovirus and lentivirus: Implications for vector safety and efficacy
.
J. Virol.
 
76
,
7209
7219
.