Adenoviral vectors

Abstract

The invention relates to adenoviral vectors, cells for use in generating adenoviral vectors, methods for generating adenoviral vectors, and therapeutic uses of adenoviral vectors in gene therapy, tumour therapy and as vaccines.

Claims

1. A method for cloning an adenoviral sequence, wherein the adenoviral sequence is a full length adenoviral genome sequence, comprising: a) providing a first linear nucleic acid molecule which comprises the full length adenoviral genome sequence; b) providing a linearized medium copy plasmid which shares at least two regions of sequence homology with the first linear nucleic acid molecule; and c) bringing the first linear nucleic acid molecule and the linearized medium copy plasmid into contact in the presence of a 5 to 3 exonuclease and an annealing protein such that the first linear nucleic acid molecule and the linearized medium copy plasmid recombine to form a circular plasmid containing the full length adenoviral genome sequence; wherein the 5 to 3 exonuclease is full length RecE of SEQ ID NO:1412, or a protein with at least 95% sequence identity to SEQ ID NO:1412, and the annealing protein is RecT, and wherein a first region of the at least two regions of sequence homology is a region of sequence homology with the 5 ITR of the adenoviral sequence in the first linear nucleic acid molecule, and a second region of the at least two regions of sequence homology is a region of sequence homology with the 3 ITR of the adenoviral sequence in the first linear nucleic acid molecule, and wherein each of the first region and the second region is 40-80 nucleotides in length.

2. The method of claim 1, wherein the first linear nucleic acid molecule is present in a mixture.

3. The method of claim 1, wherein the linearized medium copy plasmid is a p15A origin-based vector.

4. The method of claim 1, further comprising the generation of an adenoviral vector from the circular medium-copy plasmid comprising the full length adenoviral genome sequence, comprising: a) providing a second linear nucleic acid molecule which shares at least two regions of sequence homology with the circular medium-copy plasmid comprising the full length adenoviral genome sequence, wherein the second linear nucleic acid molecule comprises one or more transgenes of interest situated between two regions of sequence homology; and b) bringing the circular medium-copy plasmid comprising the full length adenoviral genome sequence and the second linear nucleic acid molecule into contact in the presence of a 5 to 3 exonuclease and an annealing protein such that sequences between the regions of homology in the second linear nucleic acid molecule are introduced into the circular medium-copy plasmid; wherein: i) the 5 to 3 exonuclease is RecE and the annealing protein is RecT; or ii) the 5 to 3 exonuclease is Red alpha and the annealing protein is Red beta.

5. The method of claim 4, further comprising a step of releasing the adenoviral vector in linear form from the circular medium copy plasmid.

6. The method of claim 4, wherein the one or more transgenes include one or more reporter genes.

7. The method of claim 6, wherein the one or more reporter genes include one or more genes encoding a fluorescent protein and/or a luciferase gene.

8. The method of claim 1, further comprising a step of creating a library comprising two or more medium copy plasmids each comprising a full length adenoviral genome sequence.

9. The method of claim 8, wherein the cloned adenoviruses in the library are tagged with one or more reporter genes.

10. The method of claim 9, wherein the one or more reporter genes include one or more genes encoding a fluorescent protein and/or a luciferase gene.

Description

BRIEF DESCRIPTION OF THE FIGURES

(1) FIG. 1Schematic outline of wild-type adenovirus amplification and viral genome isolation

(2) a) Wild-type human adenoviruses from clinical isolation were first pre-amplified in individual permissive cell lines (5080% confluence, 2% FBS) using repeated infection circles to achieve 90% cytopathic effect (CPE). Then each virus was amplified in a large scale manner in 10-20 15 cm tissue culture dishes. Crude cell lysates were used to purify viruses by CsCl gradients, followed by viral genome isolation and sequence verification.
b) Genome isolation of different Ad types visualized on an agarose gel.

(3) FIG. 2Direct cloning (HTC) of Ad genomes by LLHR

(4) a) Prototypal strategy to direct clone Ad genomes by LLHR. The shuttle vector p15A-cm-adHA was constructed by co-electroporation of four DNA fragments sharing terminal homologous arms (HAs) into the RecET expressing E. coli strain GBred-gyrA462, which has an R462C mutation in the GyrA subunit of DNA Gyrase that confers resistance to CcdB expression. The final vector p15A-cm-AdV was generated via a second LLHR step in RecET expressing E. coli strain GBred by co-electroporation of the linearised shuttle vector and the linear double-stranded adenoviral DNA (linear-dsDNA). NdeI and BamHI restriction digestion was used to linearize the circle plasmids.
b) Recombineering efficiency for shuttle vector construction using different amounts of PCR products in four experimental settings.
c) Recombineering efficiencies of viral genome cloning using various amounts of viral genomic DNA (0 to 1 g) and 1 g of the linearized shuttle vector. Error bars, s.d.; n=3.
d) Single-PCR based Ad genome cloning. The vector backbone p15A-cm was generated by a single PCR from a modified plasmid p15A-cm-MCs2.0. 50 bp homology arms (HAs) were incorporated into individual Ad genomes via primer design.
e) Direct Ad genome cloning from cell/virus lysate. Instead of purified Ad genome, a DNA mixture isolated from cell/virus lysate, or a clinical sample, was co-electroporated.
f) Left panel: Strategy to rescue engineered Ads. Different molecular forms based on restriction enzyme (RE) digest for different virus rescue strategies are shown. Right panel: Influence of the genome-releasing status on virus rescue efficiency after DNA transfection into permissive cells. Y-axis indicates viral genome copy numbers isolated from cell/virus lysates collected after each passage. X-axis is the passage number after transfection. dd, completed exposed adenoviral genome released by double-digest; sc, linearized adenoviral genome released by single-cutter; p, circular plasmid. Amp, ampicillin; cm, chloramphenicol; ccdB, counter-selection marker.

(5) FIG. 3Ad genome tagging and in vitro/in vivo characterisation

(6) a) Strategy for high-throughput ad genome tagging by linear-circular homologous recombination (LCHR). The E3 region was first deleted by integration of the spect-cassette, which was then replaced by the multicistronic/triple-expression cassette (GLN).

(7) b) In vitro characterisation of tagged Ad. Transgene expression efficiency of different Ad types was compared to the common used Ad5, and indicated as fold-change. Luciferase expression was measured by addition of furimazine substrate and indicated by luminescence units (RLU).
c) Bio-distribution study of tagged Ad in transgenic mice.

(8) FIG. 4Recombineering toolkit for Ad genome tagging including plasmid maps and sequences of pR6K-spect-adapter and pR6K-GLN

(9) a) pR6K-spect-adapter serves as a PCR template to amplify selection marker spectinomycin adenyltransferase (spect), which is flanked by the same homologous arm (HAR and HAL) as the reporter cassette GLN (shown in b). EcoRI restriction enzyme digest releases the PCR template. The R6K backbone is used to avoid plasmid contamination.
b) pR6K-GLN provides the reporter gene cassette GLN which is released by EcoRI restriction enzyme digest. E. coli strain GB05-Red harbouring an arabinose inducible gbaA operon (red, red, redid and recA) at the ybcC locus2 mediates highly efficient LCHR. E. coli strain GBred-gyrA462 is generated from GB05-Red with the Arg462-coding codon CGT been changed into Cys-coding codon TGC, allowing resistance to CcdB expression.sup.37. All plasmid maps used in this study are created with SnapGene software.

(10) FIG. 5Summary of the complete pipeline to study natural Ad diversity and to generate an engineered Ad library including tagging of viral genomes

(11) The process begins with wild-type adenovirus amplification in respective permissive cell lines (e.g. HeLa, HEK, A549). After virion purification the adenovirus genomic DNA is isolated and verified by sequencing and restriction enzyme digest. According to the sequence identification the shuttle vector p15A-cm-adHA is computationally designed, and constructed by co-electroporation of four DNA fragments containing homologous arms (HA) to each other into the RecET expressed E. coli strain, where LLHR takes place. In the next step, the adenoviral genome is incorporated into the linearised shuttle vector containing HA to each ITR end via LLHR. Sequence-verified plasmids harbouring adenoviral genomes are collected together building up an engineered adenoviral library. To prove integrity of the cloned adenovirus genomes rescue experiments were performed. Marker gene GFP/LUC tagging was mediated by linear-circular homologous recombination (LCHR). The E3 region was first deleted by integration of the ccdB-Amp cassette, which was then replaced by a P2A peptide-mediated bicistronic-expression cassette expressing a Turbo Green fluorescent protein and a NanoLuc luciferase (tGFP-Nluc) to generate p15A-cm-AdV-tGFP-Nluc. The tagged adenovirus AdV-delE3-tGFP-Nluc can be reconstituted in its permissive cell line and further evaluated in vivo.

(12) FIG. 6Genome organisation of cloned adenoviruses

(13) FIG. 7 HCAdV-CRISPR/Cas9 pipeline (1)

(14) Cloning of three guide RNAs and the Cas9 coding sequence expressed under the control of a constitutive or an inducible promotor into the shuttle vector. Partial sequences used in cloning are shown in (A) (SEQ ID NOs: 1448 and 1449) and (B) (SEQ ID NOs: 1450 and 1451).

(15) FIG. 8HCAdV-CRISPR/Cas9 pipeline (2)

(16) Cloning of three guide RNAs and the Cas9 coding sequence expressed under the control of a constitutive or an inducible promotor into the high-capacity adenoviral vector.

(17) FIG. 9HCAdV-CRISPR/Cas9 pipeline (3)

(18) Production of high-capacity adenoviral vectors containing the CRISPR/Cas9 machinery.

(19) FIG. 10Recombineering pipeline for construction of adenovirus-Sleeping Beauty transposon hybrid vectors

(20) After cloning the gene of interest into the shuttle vector pHM5-FRT-IR it can be transferred to the high-capacity adenoviral vector by recombineering.

(21) FIG. 11HAdV17 has tropism for endothelial cells in vitro

(22) A. Cell line screening in vitro. HEK293, A549, Hela, Huh7, Jurkat, MMDH3, SKHep, SKOV3 and EA.hy926 cells were seeded in 24 well plates the day before infection. When 90% confluent, cells were transduced with HAdV5GFP and HAdV17GFP at various multiplicities of infection (0.1, 1, 10 or 100 per cell) for 2 h. GFP expression was analysed 24 h post-infection by FACS. Uninfected cells (negative controls) were used to set the background gate at approximately 1%. Percentages of GFP-positive cells are given.
B. Growth curve comparison between wild type HAdV17 and HAdV5. HEK293, A549, Hela and EA.hy926 were seeded in 24 well plates the day before infection. When 80% confluent, cells were transduced with wild type HAdV5 and HAdV17 at a multiplicity of infection (MOI) of 10. Cells were harvested at various time points (2, 10, 24, 48, 72 hours), and quantitative PCR was performed after cellular DNA isolation to detect assay for virus replication.
C. Virus internalization analysis by quantitative PCR. HEK293, A549, Hela and EA.hy926 were infected at MOIs of 0.1, 1, 10, 100 of HAdV5GFP and HAdV17GFP in 24 well plates for 2 h. Afterwards cells were treated with 5% trypsin for 3 min, centrifuged at 500 g for 3 min and washed twice with DPBS to ensure that only internalised viral particles were analysed. Total cellular DNA (including the adenoviral DNA) was extracted, and quantitative real-time PCR was performed to detect viral genome. Data points represent mean standard error based on three independent experiments (n=3).

(23) FIG. 12Transduction of primary human umbilical vein cells (HUVEC)

(24) Transduction of primary human umbilical vein cells (HUVEC). Primary HUVECs were transduced with HAdV5GFP and HAdV17GFP at various multiplicities of infection (0.1, 1, 10 or 100 per cell) for 2 hours. (A) GFP-positive cell numbers and (B) mean fluorescent intensity were analysed 24 h post-infection by FACS. Uninfected cells (negative controls) were used to set the background gate at approximately 1%. (C) Histograms show the surface markers hCAR on HEK293, HeLa, EA.hy926 and HUVEC by flow cytometry. (D) Quantitation of mean fluorescent intensity of CAR expression on cell surfaces. HeLa, EA.hy926 and HUVEC cells were stained with an anti-CAR antibody labelled with FITC and measured by flow cytometry. As negative controls each cell line was also incubated without supplementation of the primary antibody. Data points represent mean standard error (n=3).

(25) FIG. 13Validation of endothelium tropism that is dependent on interaction between CD46 and HAdV17 fiber by gain of function study

(26) (A) Schematic representation of chimeric fiber proteins incorporated into HAdV5 capsid and structure of chimeric fiber genes (knob, shaft and tail). White depicts the fiber derived from HAdV5GFP and black depicts the fiber derived from HAdV17GFP. HAdV5GFP/17knob contained the shaft and tail from HAdV5 and knob from HAdV17. HAdV5GFP/17 fiber contained the tail from HAdV5 and both shaft and knob from HAdV17. (B) HEK293 and EA.hy926 cells were seeded in 24 well plates the day before infection. When 90% confluent, cells were transduced with HAdV5GFP, HAdV5GFP/17fiber, HAdV5GFP/17Knob and HAdV17GFP at various multiplicities of infection of (0.1, 1, 10 or 100 per cell) for 2 h. GFP expression was analysed 24 h post-infection by FACS. Uninfected cells (negative controls) were used to set the background gate at approximately 1%. Percentages indicate percentage of GFP-positive cells. MFI=mean fluorescent intensity. Data points represent mean standard error (n=3).

(27) FIG. 14in vivo biodistribution of HAdV17GFP vectors

(28) Viral genomes were detected by real-time quantitative PCR in various organs (liver, heart, lung, artery, kidney, pancreas, spleen, intestine and brain) harvested 72 hrs after systemic administration of 210.sup.9 transducing units per mouse of HAdV17GFP into both CD46 transgenic mice and wild-type mice. HAdV5GFP was administered to wild-type mice as a control (n=3 mice per group).

(29) FIG. 15Immunofluorescence analysis of liver sections

(30) Mice were sacrificed at 3 days after vector injection, and livers were excised for histology. (A) The top panel shows colocalisation of viral transgene (GFP) expression (green) with endothelial markers (CD31, red) from CD46 transgenic mice injected with HAdV17GFP. Single stained and merged images at 20 magnification are presented. The scale bar of images is 100 m.

(31) The 4.sup.th panel from the top is from wild type mice injected with HAdV17GFP. The 5.sup.th panel from the top is from wild type mice injected with HAdV5GFP. PBS treated transgenic mice (3.sup.rd panel from the top) and PBS treated wild type mice (6.sup.th panel from the top) are controls. The 2.sup.nd panel from the top is from CD46 transgenic mice injected with HAdV17GFP without CD31 primary antibody treatment. (B) The 7.sup.th and 8.sup.1h panels from the top are views at a higher magnification (60) of the first panel. Images are representative of multiple fields of view. The scale bar of images is 25 m.

(32) FIG. 16Neutralising antibody assay for HadV17

(33) (A) Reciprocal dilution of dog serum, immunised with HAdV-5, was incubated with HAdV17GFP and HAdV5GFP. The serum-virus mixture was used to infect HEK293 cells and 24 hours post-infection, GFP expression levels were determined. (B) Preexisting immunity to HAdV17 in patients. Transduction assays were carried out in the presence of serum samples from 19 patients. Samples were considered neutralising if greater than 90% reduction in transduction (below 10% residual transduction) was seen in comparison to a no-serum control.

(34) FIG. 17Schematic structure of different virus vectors used in Example 3

(35) (A) Diagram of ccdB recombineering to construct first generation p15a-HAdV17GFP. (B) Schematic structure of pseudotyping HAdV5GFP with knob or fiber from HAdV17.

(36) FIG. 18Generating stable complementing cell lines based on HEK293 and A549 cells using the plasmid pIRES-neo

(37) (A) E1 cassette from HAdV17 (2808 bp) was amplified by PCR and ligated into multi cloning site of pIRESneo3 (clontech) to construct the vector pCMV-HAdV17E1-IRES-neo. (B) Methylene blue staining was performed to mark the positive cell clones. (C) PCR and RT-PCR were performed to analyse each cell clone. Genomic DNA was isolated from each cell clone and eluted in dH.sub.2O. Total RNA was isolated and resuspended in RNase-free H2O. Reverse transcription was performed, and 5 l of the cDNA was used for PCR. A negative control without reverse transcriptase was performed.

(38) FIG. 19Histograms show the surface markers of hCAR and CD46 expression level in various cell lines by flow cytometry

(39) 0.510.sup.6 cells (HEK293, A549, HeLa, EA.hy926, MMDH3 and HCT116) were counted and washed with PBS supplemented with 1% BSA, centrifuged (1500 g, 3 min), and resuspended in 100 l PBS/BSA and 2.5 l anti-hCAR antibody (Santa Cruz, sc-56892) following an incubation step at 4 C. for 1 hour. Afterwards the cells were washed again with PBS/BSA to remove unbound antibody, resuspended in 100 l PBS/BSA and 0.5 l of an APC labeled goat anti-mouse secondary antibody (Santa Cruz, sc-3818), and incubated for 1.5 hours at 4 C. with continuous shaking. Afterwards cells were again washed with PBS/BSA and finally resuspended in 400 l PBS for flow cytometry using FACS (BD). As negative controls each cell line was also incubated without supplementation of the primary antibody. Data points represent mean standard error (n=3).

(40) FIG. 20High-throughput screening (HTS) of the reporter-tagged human adenovirus library in a panel of human tumour cells

(41) Transgene expression efficiency of different adenovirus types (Ad type number) was tested in a panel of disease-specific cell lines. Levels were compared to the commonly used adenovirus type 5 (Ad5) and indicated as fold change. Luciferase expression was measured by addition of Furimazine substrate and expressed as relative light units (RLU). In all cell lines error bars represent meanSD with the exception that in A549 and MG-63 is meanSEM.

(42) FIG. 21High-throughput screening (HTS) to identify virus candidates with high oncolytic potency in osteosarcoma cells

(43) (A) Transduction of two osteosarcoma cell lines. 200 vp/cell were applied with three independent experiments performed. Error bars denote s.e.m. Data were analysed by two-tailed unpaired t-test. P-values for analysed virus types compared to Ad5 were <0.05 if not otherwise stated. NS, not significant; vp, viral particles.
(B) Virus internalization efficiency in three osteosarcoma cell lines. Cells were infected with individual viruses at 2,000 vp/cell for three hours to determine viral genome copy numbers (VCN), which were quantified by qPCR and expressed as VCN per 10 ng total DNA. Error bars represent meanSD. n=3 per group. Data were analysed by two-tailed unpaired t-test. P-values for analysed virus types compared to Ad5 were >0.05 if not otherwise stated. *, significant (p<0.05).
(C) Visualization of GFP-expression 2 days post infection. Cells were infected at 1,000 viral particle (vp) per cell.
(D) Crystal violet staining of viable cells was used to evaluate oncolytic activity 7 days after infection.

(44) FIG. 22High-throughput screening (HTS) to identify virus candidates with high oncolytic potency

(45) Oncolysis assay performed in A549 cells as oncolytic assay control. Crystal violet staining of viable cells was used to evaluate oncolytic activity 7 days after infection.

(46) FIG. 23Transgene expression efficiency of different adenovirus types (Ad type number) in the cell line RG2 [D74] at 2000 virus particle per cell (vp/c)

(47) Ad5, Ad21 and Ad37 showed high luciferase (26 hs p.i.) and GFP expression (2 ds p.i.), albeit without being significant. All three viruses reached about the same dimension of efficiency. Furthermore Ad20 reached more than 20% of the efficiency of Ad5. Luciferase expression was measured by addition of Furimazine subtract and expressed as relative light units (RLU). Transgene expression levels were compared to the commonly used adenovirus type 5 (Ad5) and indicated as fold change. Error bars represent standard error of the mean (SEM). Data were analysed by two-tailed unpaired t-test. P-values for analyzed virus types compared to Ad5 were <0.05 if not otherwise stated. NS means not significant (p>=0.05).

(48) The invention is further described with reference to the following non-limiting examples:

EXAMPLES

Example 1Cloning Adenoviral Genomes

(49) Clinical isolates of wild type (WT) Ad were amplified in permissive cell lines (e.g. HeLa, HEK, A549) using serial amplification steps. After large-scale amplification Ads were purified by cesium chloride gradients (FIG. 1). To verify the Ad type a multiplex hexon-specific PCR was performed and the DNA sequence was confirmed. Viral genomes of purified Ad particles were isolated (FIG. 1) and LLHR was applied to directly clone isolated viral genomes as schematically shown in FIG. 2. We first established a fast and reliable strategy to generate master clones for direct Ad genome cloning based on annealing of 4 PCR fragments in one step. These PCR fragments with homology arms (HA) for directed cloning contained the 5 and 3 ITR sequences, the previously published selection marker ccdB10 and ampicillin for advanced positive and counter selection and the plasmid backbone with a p15A origin of replication and the chloramphenicol resistance gene (FIG. 1). Different ratios of PCR products were tested, and after homologous recombination (HR) in a RecET expressing E. coli strain this resulted into the novel medium copy plasmid p15A-cm-adHA. Equal amounts of all 4 PCR fragments resulted in highest cloning efficiency (FIG. 2b).

(50) Next, isolated linear Ad genomes were cloned into this plasmid using LLHR. FIG. 2c shows recombineering efficiencies of direct viral genome cloning using various amounts of viral DNA (0 to 1 g) and 1 g of the linearised shuttle vector. The plasmid backbone for direct cloning of large dsDNA viruses containing the p15A origin of replication results in significantly increased cloning efficiencies and is easier to handle compared to commonly-used bacterial artificial chromosomes (BAC). Since numerous viral genomes show high sequence homologies in their ITR sequences, viral genomes could be cloned simultaneously using the same shuttle vector p15A-cm-adHA containing the same HA. Only 50 bp HA are required, enabling fast and reliable direct HTC. The same strategy was used for Ad from species D, B2 and C.

(51) Integrity of all cloned Ad genomes was checked by diagnostic restriction enzyme digests and compared to originally isolated Ad genomes from virions (FIG. 2). NGS was used to determine exact sequences of all Ad containing plasmids. Extensive rescue experiments were performed and the strategy optimised by testing different molecular forms (precise excision of the Ad genome, linearised Ad containing plasmid and circular Ad genome containing plasmid) (FIG. 3). After introduction of viral DNA into Ad permissive cell lines, viral replication was monitored by quantitative PCR (qPCR) over more than 5 serial passaging steps and the precise excision of the Ad DNA molecule from the plasmid was most sufficient in rescuing Ad. This strategy was then applied to all cloned human Ad.

(52) To further explore the cloned human Ad library, viral genomes were tagged with a 2A peptide-mediated multicistronic-expression cassette (GLN, SEQ ID NO:227) providing a TurboGFP fluorescent protein as an in vitro marker, a NanoLuc luciferase for in vivo studies and kanamycin/neomycin as a selection marker. As schematically outlined in FIG. 3, LCHR was applied for marker insertion into the early gene E3 region of the cloned Ad genomes of the library. In a first step, spectomycin acetyltransferase was inserted via HR in E. coli strain GB05-Red into the adenoviral genome, followed by GLN cloning. By using a conserved region in the E3 region, different Ad genomes could be tagged in a high-throughput manner using the identical shuttle vector for different Ad types (FIG. 4). Tagged Ads were rescued in permissive cell lines.

(53) Tagging of Ad enabled in vitro and in vivo characterisation of chosen Ad types. After infection of cell lines originating from different cell types derived from different organs (epithelium, endothelium, muscle, blood, liver) and measurement of luciferase and GFP expression it was found that tagged Ad show a distinct tropism (FIG. 3). The in vivo tropism was analysed by systemically injected tagged HAdV-B3, HAdV-616, HAdV-650, HAdV-05, and HAdV-E4 into 057BI/6 mice. As shown in FIG. 3, viruses also showed a distinct biodistribution as detected by transgene expression levels and viral genome level. A summary of the complete pipeline to study natural Ad diversity and to generate an engineered Ad library including tagging of viral genomes is shown in FIG. 5.

Methods for Example 1 and Other Examples

(54) Cell Culture

(55) Human Hela cells, A549 cells, HEK293, and EA.hy926 cells were grown in high glucose Dulbecco's Modified Eagle's Medium (DMEM, PAN BIOTECH) supplemented with 10% FBS (GE Healthcare), 100 U ml-1 penicillin (PAN-BIOTECH), and 100 g ml-1 streptomycin (PAN BIOTECH). For human hepatocyte Huh7, Non-Essential Amino Acid (NEAA) was added. For Jurkat cells, RPMI-1640 based Medium supplemented with 10% FBS, 100 U ml-1 penicillin and 100 g ml-1 streptomycin was used. For the murine cell line Neuro2a (N2a) cells, Eagle's Minimum Essential Medium (EMEM, GE Healthcare), supplemented with 10% FBS, 100 U ml-1 penicillin and 100 g ml-1 streptomycin was used. For the murine myoblast C2C12 cells, DMEM supplemented with 10% FBS, 100 U ml-1 penicillin and 100 g ml-1 streptomycin was used.

(56) Wild Type Adenoviruses

(57) HAdV-05 (ATCC VR5) strain and HAdV-F41 (ATCC VR930) was obtained from the American Type Culture Collection (ATCC). HAdV-A12, -A18, -A31, -B3, -B16, B21, -B11, -B14, -B35, -C6, -D9, -D10, -D13, -D17, -D20, -D24, -D25, -D26, -D27, -D33, -D37, -D69 and -E4 were clinical isolates obtained from the diagnostic group of the Max von Pettenkofer-Institute (Department of Virology) at the Ludwig-Maximilians-University Munich in Germany. HAdV-B7, -B50, -B34, -C1, -C2, -D8 and -G52 were kindly provided by the Heinrich Pette Institut (HPI) Hamburg, Germany.

(58) Ad Amplification, Purification and Titration

(59) WT human Ad from clinical isolates were first pre-amplified in individual permissive cell lines (5080% confluence), with serial infection circles to achieve 90% cytopathic effect (CPE). Each virus was amplified to large scale in 10-20 15 cm tissue culture dishes. For virus amplification DMEM supplemented with 2% FBS was used. Crude cell lysates were used to purify viruses by a CsCl gradient-based ultracentrifugation method (Beckman Coulter), followed by a desalting step based on disposable PD-10 desalting columns (GE Healthcare). The purified virus was aliquoted and stored at 80 C. for further use. Ad particle concentrations were determined by measuring the optical density at 260 nm and expressed as viral particles (VPs) per milliliter.

(60) Adenoviral genomic DNA isolation from virions

(61) For cloning of viral genomes, viral genomic DNA was extracted from purified particles by the addition of proteinase K, subsequent phenol-chloroform extraction, and ethanol precipitation. A detailed protocol for isolating viral genomic DNA is found in Example 7. To confirm the Ad type on genome level, multiplex PCR and sequencing were performed using the primer pair hexon-fwd (ATGGCCACCCCATCGATGATGC) (SEQ ID NO: 1452) and hexon-rev (TTATGTGGTGGCGTTGCCGGCC) (SEQ ID NO: 1453) amplifying the hexon regions of the viral genomes. To verify the end-sequence of the adenoviral genome for the following homologous recombineering step, primers reading into the ITR region were designed.

(62) Plasmid Construction

(63) p15A-cm-MCS; p15A-amp-ccdB; pR6K-spect-adapter; pR6K-GLN; Linear-linear homologous recombination (LLHR)-mediated adenoviral genome cloning. Linear-circular homologous recombination (LCHR)-mediated adenovirus genome tagging.

(64) PCR

(65) Homology arm (HA)-containing long primer-mediated PCR was performed with Phusion High-Fidelity DNA Polymerase (New England Biolabs, Frankfurt, Germany) according to the manufacture's protocol. Notably, only the primer binding sequence (20 bp) was used for calculating the annealing temperature. The PCR product purified with the Wizard SV Gel and PCR Clean-Up System (Promega, Mannheim, Germany) and eluted in ddH2O was used for electroporation. To check virus reconstitution and amplification, OneTaq 2X Master Mix (New England Biolabs, Frankfurt, Germany) was used according to the manufacture's standard protocol.

(66) NGS and bioinformatics analyses

(67) For sequencing of plasmids, 200 ng purified DNA was subjected to standard Illumina DNA library preparation. In brief, DNA was enzymatically sheared (NEBnext dsDNA Fragmentase, New England Biolabs). After XP bead purification (Beckman Coulter), ends were polished and A-tailed and universal adapters were ligated (Ultra Directional DNA Library Prep Kit, New England Biolabs). For adapter ligation, custom adaptors were used (Adaptor-Oligo 1: 5-ACA-CTC-TTT-CCC-TAC-ACG-ACG-CTC-TTC-CGA-TCT-3 (SEQ ID NO: 1454), Adaptor-Oligo 2: 5-P-GAT-CGG-AAG-AGC-ACA-CGT-CTG-AAC-TCC-AGT-CAC-3 (SEQ ID NO: 1455)). After ligation, adapters were depleted by XP bead purification (Beckman Coulter). Sample indexing was done in the following PCR enrichment (15 cycles). For Illumina flow cell production, samples were equimolar pooled and distributed on two Illumina MiSeq flow cells for 300 bp paired-end sequencing. The Illumina TruSeq adapter and regions of low quality (phred quality <20) were trimmed with cutadapt requiring a minimum length of 50 bp. Trimmed reads were mapped with BWA onto the reference sequence of the vector and reads consisting entirely of vector sequence were discarded whereas reads without or with only partial vector sequence were kept. Each adenovirus dataset was assembled with IVA de novo as well using the respective GenBank sequence as anchor. The better assembly was chosen based on number of sequences, total length and presence of vector sequence at the flanks of the assembled sequence. The remaining vector sequence was identified with BLAT and removed with in-house Perl scripts and the final assembled sequence was orientated according to the respective GenBank reference sequence. Annotation of coding sequences (CDS) was done with Glimmer in two steps. First, known adenovirus CDS from the GenBank were compared to the assembled sequences using exonerate. The resulting alignments served as training set for Glimmer which then predicted the final CDS regions. Functional identification of CDS was based on BLASTP against known adenovirus protein sequences from GenBank. To align and visualize the adenovirus sequences obtained by NGS against their respective GenBank reference sequences, we used the zPicture program (webpage: zpicture.dcode.org/). Sequence alignments between adenovirus sequence and GenBank reference were done with BLASTZ. As cut-off value, we used a minimum sequence identity of 99% which minimizes false-positive alignments but still allows for studying single nucleotide variants between adenovirus sequences and respective GenBank references. Multiple sequence alignments for conserved E3 and ITR sequences were generated using clustal Omega (www.ebi.ac.uk/Tools/msa/clustalo/) with default parameters and for visualized extent of conservation in aligned sequence sets we used WebLogo (webpage: weblogo.berkeley.edu/logo.cgi). Plasmid DNA transfection and virus rescue10 g of the p15A-based adenovirus genome containing plasmids were either digested with a combination of two restriction enzymes (PmeI/Sbfl or PacI/SwaI) releasing the adenovirus backbone, or with the restriction enzyme I-SceI linearizing the p15A-based Ad genome. To purify and concentrate digested DNA, ethanol precipitation was performed for DNA digested with PmeI/SbfI and PacI/SwaI, and for I-SceI digested DNA a phenol-chloroform extraction followed by ethanol precipitation was conducted.

(68) A549 cells were plated in 6-well plates and at 50-80% confluency 3 g of digested viral DNA was transfected using Superfect transfection Reagent (Qiagen) according to the manufacturer's protocol. After 24-48 hrs, when the cells grew to up to >90% confluency, the medium was changed to 2% FBS-supplemented DMEM. The cells were maintained for up to two weeks, until cytopathic effect (CPE) was observed. If no CPE was obtained, the cell/virus lysate was collected and or of the lysate was used to infect a new well of A549 cells at a confluency of 90-95%. To release the virus from infected cells, the crude lysate was subjected to three freeze/cycles in liquid nitrogen and in a 37 C. water bath. A small aliquot of cells was collected for qPCR analysis.

(69) qPCR Analysis

(70) To monitor virus replication during rescue, quantitative real-time PCR (qPCR) was performed using the CFX96 Touch Real-Time PCR Detection System (Bio-Rad). Previously described primer pairs and probes (Damen, M et al, 2008, JoCM) binding to the hexon of Ad were used to determine the copy number of Ad genomes in infected cells. The PCR was based on the following program: pre-incubation/activation at 95 C. for 5 min, amplification and data collection during 40 cycles (95 C. for 15 s and 60 C. for 30 s). The Sso Fast Probes Supermix (Bio-Rad) was used for these PCRs.

(71) Characterization of Tagged Ad In Vitro and In Vivo

(72) All reconstituted GLN-tagged viruses were confirmed by hexon-PCR of isolated adenoviral genomes. Adenovirus particle concentrations were determined by measuring the optical density at 260 nm and expressed as viral particles (vps) per millilitre.

(73) Nano-Glo Luciferase Assay

(74) Individual tested cells were grown to confluence in 96-well tissue culture plate and infected with different viral partials (VPs) per cells. 26 h after infection NLuc activity was measured with the Nano-Glo assay system (Promega), and luminescence was detected with a plate reader (Tecan).

(75) Genome Uptake Measured by Internalization Assay

(76) To quantify the cell entry efficiency, a defined number of Ad particles (vp) was used to infect pre-seeded tumor cells and incubated for 2 hours. Cell monolayers was digested and flushed off with trypsin, followed by extensive washing with PBS. Genomic DNA was extracted by incubation in TE buffer (10 mM Tris-HCl, 10 mM EDTA, pH 8.0) with 0.5% SDS and 0.5 mg/ml proteinase K. Subsequently a phenol-chloroform extraction and ethanol precipitation was performed. To monitor virus genome uptake efficiency, quantitative real-time PCR (qPCR) detecting the transgene (GLN gene cassette) was performed.

(77) Oncolytic Assays with Most Promising Ad Candidates

(78) Oncolytic assay was performed in 24 well plates. A 10-fold dilution series of individual Ads was prepared freshly to infect pre-seeded cancer cells. Cytopathic effect (CPE) was checked daily until at least one of the viruses on one plate at the lowest dosage showed CPE or until maximal 14 days. The cells were first fixed with 3.7% formaldehyde then stained with crystal violet solution.

(79) Statistics

(80) Statistical analyses were conducted with Microsoft Excel. Experimental differences were evaluated by a Student's one-tailed t-test assuming equal variance.

Example 2Gene Annotation

(81) Genes were predicted using Gene Locator and Interpolated Markov ModelER (Glimmer).sup.38. For each sequenced genome, protein sequences of known genes of the respective reference from GenBank were aligned with exonerate.sup.39 to the assembled genome sequence. The coordinates of the best hits were then used to build a Glimmer model which was subsequently used for prediction of location and orientation of genes in the sequenced genome.

(82) Next, protein sequences were compared between virus genomes. The result of this analysis is given in FIG. 6. Shown are genome organisations for 28 sequenced human adenoviruses. Genes and their orientation are shown as gray-shaded shapes on both strands of the genome (solid line). All genomes show the canonical organization of the Mastadenovirus genus.

(83) Shading of a gene reflects its maximum sequence divergence across all 28 viruses determined through an all-vs-all Blast analysis.

Example 3Cell Tropism of Human Adenovirus D17

(84) A new first generation adenovirus based on human adenovirus D17 was constructed and labelled with a green fluorescent protein (GFP) marker using the recombineering technology described in Example 1. The early E1 gene was deleted in the HAdV17 vector, and a corresponding E1-deleted, GFP-labeled HAdV5 vector was constructed for comparison.

(85) Viruses were rescued in complementary E1-expressing stable cell lines, and then screened against a panel of different cell lines by fluorescence activated cell sorting (FACS) analysis and quantitative PCR. HAdV17 was found to have a tropism for endothelial cells, whereas endothelial cells are normally refractory to HAdV5 infection. This finding was further verified using primary human umbilical vein endothelial cells (HUVEC).

(86) Competition assays based on soluble recombinant fiber knob blocking reagents.sup.40, 17 (5knob, 17knob, JO4, Augmab) were used to characterize the receptor interaction with these vectors in vitro. It was found that HAdV17 could utilize both CD46 (a membrane cofactor protein which is expressed on all nucleated cells) and CAR (coxsackievirus and adenovirus receptor) as cell attachment receptors. The endothelial cell tropism was CD46-dependent and could be blocked by the CD46 blocking reagent Augmab.

(87) In vivo biodistribution analyses were performed after intravenous injection of recombinant viruses into both normal and CD46-transgenic mice. These studies showed significantly increased vector genome copies (VCN) in various organs of CD46-transgenic mice compared to normal mice, indicating the involvement of CD46 as a receptor. These results were confirmed by quantitative PCR (qPCR) and immunohistology analysis.

(88) Neutralising antibody assays revealed that there was less seroprevalence with HAdV17 compared to HAdV5 in humans.

(89) Accordingly, HAdV17-based vectors, which can use both hCAR and CD46 as receptors and display an endothelial cell tropism, hold great promise for gene therapy in endothelial disease.

(90) See also FIGS. 11-19.

Example 4Delivery of all Components of the CRISPR/Cas9 System Using High-Capacity Adenoviral Vectors

(91) A new CRISPR/Cas9 shuttle plasmid toolbox was generated, containing the Cas9 nuclease gene, either utilising a constitutive or an inducible promoter, and a gRNA expression unit. The toolbox allows cloning or recombining of all CRISPR/Cas9 components into the HCAdV genome in one step. To use several gRNA expression units for multiplexing the CRISPR/Cas9 system, further gRNA expression units can be easily included. To enable fast assembly of recombinant CRISPR-HCAdV genomes, DNA recombineering was used to introduce all CRISPR/Cas9 expression units into the HCAdV genome contained in the bacterial artificial chromosome pBHCA. For insertion of multiple gRNA expression units into the HCAdV genome, the established pAdV-FTC plasmid was used in concert with homing endonuclease directed cloning. CRISPR-HCAdVs were produced using a shortened amplification and purification procedure.

(92) The toolbox was used to produce several CRISPR-HCAdVs carrying single and multiplex gRNA units specific for different targets including hCCR5, hDMD, and HPV16- and HPV18-E6 genes, yielding sufficient titers within a short time. T7E1 assays.sup.41 were applied to prove CRISPR/Cas9-mediated cleavage of respective targets. Infection of cultured human cells with respective CRISPR-HCAdVs resulted in efficient site-specific gene editing.

(93) In summary, this new platform enables customisation, cloning and production of CRISPR-HCAdV vectors for single or multiplex approaches within a short time. It simplifies the delivery of the CRISPR/Cas9 machinery by only using one single viral vector. Inducible Cas9 expression helps to avoid targeting of the genome of producer cell lines during vector production and may be beneficial for special approaches where constitutive expression is unwanted.

Example 5Enhanced Oncolytic Activity Mediated by a Novel Human Adenovirus Type 6-Based Vector

(94) Most existing oncolytic adenoviruses (AdV) are based on human AdV type 5 (hAdV-5). Clinical efficacy of hAdV-5 based oncolytic viruses is limited by variable expression levels of coxsackie- and adenovirus receptor (CAR) in different tumour cells, and insufficient replication rates. Additionally, high prevalence of neutralising antibodies against hAdV-5, resulting in lower efficiency, makes hAdV-5 a less suitable candidate for systemic application. Recent studies have highlighted human adenovirus type 6 (hAdV-6) as a promising candidate for oncolytic and vaccine vectors. Thus, development of novel oncolytic AdV based on hAdV-6 may help to overcome these limitations. Oncolytic efficacy of the candidate virus can be augmented by expression of RNAi suppressor protein P19, as has been shown previously for hAdV-5.sup.31. In this example, a novel hAdV-6-based, p19-containing oncolytic AdV was evaluated as a candidate for oncolytic applications in different tumour cell lines.

(95) A P19-containing hAdV-6 based virus (hAdV-6FP19) was cloned by a novel seamless recombineering technique (see Example 1). In order to allow P19 expression from the adenoviral vector genome, the P19 cDNA was fused via an internal ribosome entry site (IRES) to the late fiber gene. After release of the respective recombinant adenoviral genomes from plasmids containing the complete DNA molecule, linearised viral DNA was transfected into HEK 293 cells for virus reconstitution. After initial amplification steps which were monitored by virus specific PCRs, upscaling and virus purification using cesium chloride density gradient ultracentrifugation was performed. Rescue and amplification efficiencies were comparable to commonly used hAdV-5 based vectors.

(96) Various cancer cell lines from different origin were used to perform oncolysis assays. These included: A549 (lung carcinoma), HCT 116 (colon carcinoma), HeLa (cervical carcinoma) and Huh7 (hepatocellular carcinoma). Cells were infected with hAdV-6FP19, hAdV-6 and hAdV-5 at various multiplicities of infection (MOI). Two to three days after infection, cells were fixed and stained with methylene blue. Significantly higher cell lysis (up to 100-fold) was observed for hAdV-6FP19-infected cells as compared to hAdV-5 and 6 at identical MOIs. Higher cell lysis rates for hAdV-6FP19 compared to wildtype virus were present in all evaluated cell lines, suggesting significantly enhanced oncolytic potential for hAdV-6FP19. In summary, hAdV6-based vectors hold great promise for oncolytic applications and their oncolytic effectiveness can be further improved by RNAi suppression.

Example 6High Throughput Screening (HTS) of Adenovirus Library as a Novel Resource for Disease-Specific Targeting

(97) To fully explore our cloned Ad library as a resource for developing of novel translational approaches, the library was further tested on a panel of cell lines using an HTS approach. Cell lines originating from different cell types were infected with the reporter-labelled virus types of the Ad library. Transduction efficiencies measured by luciferase expression levels were compared to the commonly used adenoviral vector type 5 (Ad5). Initial screening revealed that species B adenoviruses have high transduction efficiencies in epithelial (A549, HCT 116, ARPE-19) and endothelial (EA.hy926) cells. While in the liver originated cell lines SK-HEP-1 and Huh-7, the common used vector type 5 (Ad5) highest infection efficiencies (FIG. 20). Also included was an osteosarcoma derived cell line (MG-63) and the breast cancer-derived cell line Hs 578T derived form a triple-negative breast cancer (TNBC). In the TNBC 207 cell line, Ad37 was identified as the most efficient Ad type, while other species B types (Ad16, 50 and 35) revealed improved efficiency compared to the commonly used Ad5 (FIG. 20). After performing HTS on osteosarcoma derived MG-63 cells, it was discovered that Ad21 showed highest transduction efficiencies (FIG. 20).

(98) Therefore, Ad21 was further pursued as a potential oncolytic agent to treat osteosarcoma, because this type of cancer is the most frequent primary cancer of bone which predominantly occurring in the second decade of life. Regarding the age group from 15-19 years, osteosarcomas represent >10% of all solid cancers. Therefore, a panel of osteosarcoma cell lines with different grading related features, including Saos-2 and U-2 OS cells, was further examined.

(99) As displayed in FIG. 21a and in concordance with results obtained in MG-63 cells (FIG. 20), Ad21 was identified as a promising alternative Ad type to transduce osteosarcoma cell lines, which was also confirmed by GFP also expressed form the tagged Ads (FIG. 21b) and by virus internalization assays measuring uptake of virus genomes 3 hrs post-infection (FIG. 21c). The oncolytic potency of selected Ad types was analysed in all three osteosarcoma cells and it was found that Ad21 consistently resulted in efficient oncolysis of respective cell lines at different MOIs tested (FIG. 21d). To show that oncolytic potential varied between different cancer cell lines we also tested the tagged Ad library on lung carcinoma derived A549 cells revealing that in contrast to osteosarcoma cell lines, Ad35 and -69 showed highest oncolytic potency (FIG. 22). This proof-of principle experiment demonstrates that a novel virus candidate for disease specific targeting can be identified by HTS allowing development of novel therapeutic agents in further steps.

(100) Using the same methodology, FIG. 23 shows the high infectivity rates of the Ad5, Ad21 and Ad37 adenoviruses in the glioblastoma derived cell-line RG2 [D74]. This indicates that adenoviral vectors derived from these adenovirus types would be effective in targeting glioblastoma cells.

Example 7Protocol to Isolate Adenoviral Genomic DNA for Use in the Cloning Step

(101) 1. Incubate certain volume of purified virus for 2 hours (or overnight) with proteinase K-SDS solution pH 7.5-8 (TE buffer, 0.5% SDS, 100500 g/ml proteinase K) at 56 C., with low speed shake (300 rpm).

(102) 2. Add equal volume mixture of phenol:chloroform:isoamyl alcohol (25:24:1) to the sample from step 1. In doing this, go inside to the mixture, and do not take the surface layer. To increase the recovery rate, use a phase lock gel tube (Phase Lock Gel Heavy 1.5 ml, uk.vwr.com/store/product/826754/phase-lock-gel).
3. Centrifuge for 5 min at full speed (15,000 g) at room temperature in a microcentrifuge and then transfer the aqueous phases to another clean eppendorf tubes.
4. Precipitate viral DNA by adding 1/10 volume of 3 M sodium acetate (pH 5), 2 g glycogen and 2.53 times of precooled EtOH (99.8%; stored at 20 C.). Mix gently by inverting the tube several times). To increase the recovery rate, put the mixture in 20 C. for 30 mins.
5. Centrifuge for 10 min at full speed (15,000 g) at room temperature in a microcentrifuge and discard the supernatant by pipetting.
6. Add 600 l of 70% ethanol and mix gently by inverting the tube several times. After centrifugation at 15,000 g at room temperature for 5 min, remove the supernatant by pipetting.
7. Repeat step 6
8. Air-dry the DNA pellet briefly and resuspend in 2050 l of sterilized dH2O low speed shake (300 rpm) at room temperature for 15 mins.

(103) In this protocol, which is provided by way of example, it is important that the large genomic DNA is never vortexed or vigorously pipetted during isolation.

REFERENCES

(104) .sup.1 Aiuti, A. et al. Lentiviral hematopoietic stem cell gene therapy in patients with Wiskott-Aldrich syndrome. Science 341, 1233151, (2013). .sup.2 Nathwani, A. C. et al. Adenovirus-associated virus vector-mediated gene transfer in hemophilia B. The New England journal of medicine 365, 2357-2365, (2011). .sup.3 Biffi, A. et al. Lentiviral hematopoietic stem cell gene therapy benefits metachromatic leukodystrophy. Science 341, 1233158, (2013). .sup.4 Wiley, J. M. Gene Therapy Clinical Trials Worldwide. J. Gene Med., (2015). .sup.5 Crystal, R. G. Adenovirus: the first effective in vivo gene delivery vector. Human gene therapy 25, 3-11, (2014). .sup.6 Zabner, J. et al. Adenovirus-mediated gene transfer transiently corrects the chloride transport defect in nasal epithelia of patients with cystic fibrosis. Cell 75, 207-216 (1993). .sup.7 Wilson, J. M. Gendicine: the first commercial gene therapy product. Human gene therapy 16, 1014-1015, (2005). .sup.8 Kaufman, H. L., Kohlhapp, F. J. & Zloza, A. Oncolytic viruses: a new class of immunotherapy drugs. Nature reviews. Drug discovery 14, 642-662, (2015). .sup.9 Hage, E. et al. Human Adenovirus type 70: A novel, multiple recombinant species D adenovirus isolated from diarrheal faeces of a haematopoietic stem cell transplantation recipient. J. Gen. Virol. 96, 2734-2742, (2015). .sup.10 Davison, A. J., et al. Genetic content and evolution of adenoviruses. J. Gen. Virol. 84, 2895-2908, (2003). .sup.11 Fu, J. et al. Full-length RecE enhances linear-linear homologous recombination and facilitates direct cloning for bioprospecting. Nature biotech. 30, 440-446, (2012). .sup.12 Wang, H. et al. Improved seamless mutagenesis by recombineering using ccdB for counterselection. Nucleic acids research 42, e37, (2014). .sup.13 Harrach, B. et al. Adenoviridae. In Virus Taxonomy (ed. King, A. M. Q., Adams, M. J., Carstens, E. B. and Lefkowitz, E. J.), 125-141 (Elsevier, 2011). .sup.14 Bradley, R R. et al. Adenovirus serotype 5 neutralizing antibodies target both hexon and fiber following vaccination and natural infection. J. Virol. 86, 625-629, (2012) .sup.15 Zhang, Y. and Bergelson, J. M. Adenovirus receptors. J. Virol. 79, 12125-12131, (2005). .sup.16 Ghebremedhin, B. Human adenovirus: viral pathogen with increasing importance. Eur. J. Microbiol. Immunol. 4, 26-33, (2014). .sup.17 Wang, H. et al. Desmoglein 2 is a receptor for adenovirus serotypes 3, 7, 11 and 14. Nat. Med. 17, 96-105, (2011) .sup.18 Tuve, S. et al. A new group B adenovirus receptor is expressed at high levels on human stem and tumor cells. J. Virol. 80, 12109-12120, (2006). .sup.19 Danthinne, X. and Imperiale, M J. Production of first generation adenovirus vectors: a review. Gene Ther. 7, 1707-1714, (2000). .sup.20 WO 2011/154927 .sup.21 Cong, L. et al. Multiplex Genome Engineering using CRISPR/Cas Systems. Science 15, 819-823, (2013). .sup.22 Li, X. et al. PiggyBac transposase tools for genome engineering. Proc. Nat. Acad. Sci. USA 110, E2279-E2287, (2013). .sup.23 Hausl, M. et al. Development of Adenovirus Hybrid Vectors for Sleeping Beauty Transposition in Large Mammals. Curr. Gene Ther. 11, 363-374, (2011). .sup.24 Penaloza-MacMaster, P. Alternative serotype adenovirus vaccine vectors elicit memory T cells with enhanced anamnestic capacity compared to Ad5 vectors. J. Virol. 87, 1373-1384, (2013). .sup.25 Wold, W. S. M. and Toth, K. Adenovirus vectors for gene therapy, vaccination and cancer gene therapy. Curr. Gene Ther. 13, 421-433 (2013). .sup.26 Barouch, D. H. Novel adenovirus vector-based vaccines for HIV-1. Curr. Opin. HIV AIDS 5, 386-390, (2010) .sup.27 Geisbert, T. W. et al. Recombinant adenovirus serotype 26 (Ad26) and Ad35 vaccine vectors bypass immunity to Ad5 and protect nonhuman primates against Ebolavirus challenge. J. Virol. 85, 4222-4233, (2011). .sup.28 Barnes, E. et al. Novel adenovirus-based vaccines induce broad and sustained T cell responses to HCV in man. Sci. Transl. Med. 4, 115ra1, (2012) .sup.29 Yamamoto, M. and Curiel, D. T. Current issues and future directions of oncolytic adenoviruses. Mol Ther. 18, 243-250, (2010). .sup.30 Cheong, S. C. et al. E1A-expressing adenoviral E3B mutants act synergistically with chemotherapeutics in immunocompetent tumor models. Cancer Gene Ther. 15, 40-50, (2008). .sup.31 Rauschhuber, C. et al. RNAi suppressor P19 can be broadly exploited for enhanced adenovirus replication and microRNA knockdown experiments. Sci. rep. 3, 1363 (2013). .sup.32 Rodriguez, R. et al. Prostate attenuated replication competent adenovirus (ARCA) CN706: a selective cytotoxic for prostate-specific antigen-positive prostate cancer cells. Cancer Res. 57, 2559-2563, (1997). .sup.33 Cody, J. J. and Douglas, J. T. Armed replicating adenoviruses for cancer virotherapy. Cancer Gene Ther. 16, 473-488, (2009). .sup.34 Green, N. K. et al. Extended plasma circulation time and decreased toxicity of polymer-coated adenovirus. Gene Ther. 11, 1256-1263, (2004). .sup.35 Doronin, K. et al. Chemical modification with high molecular weight polyethylene glycol reduces transduction of hepatocytes and increases efficacy of intravenously delivered oncolytic adenoviruses. Hum. Gene Ther. 20, 975-988, (2009). .sup.36 WO 2009/104094 .sup.37 Wang, H. et al. Improved seamless mutagenesis by recombineering using ccdB for counterselection. Nucleic Acids Res. 42, e37, (2014) .sup.38 Salzberg, S. L. et al. Microbial gene identification using interpolated Markov models. Nucleic Acids Research 26, 544-548, (1998) .sup.39 Slater, G. and Birney, E. Automated generation of heuristics for biological sequence comparison. BMC Bioinformatics 6, 31, (2005) .sup.40 Gaggar, A. et al. CD46 is a cellular receptor for group B adenoviruses. Nat. Med. 9, 1408-1412 (2003). .sup.41 Mashal R. D. et al. Detection of mutations by cleavage of DNA heteroduplexes with bacteriophage resolvases. Nat. Genet. 9, 177-183, (1995).