EP1326992A1

EP1326992A1 - Method for circularizing adenoviral nucleic acid via homologous recombination

Info

Publication number: EP1326992A1
Application number: EP01977917A
Authority: EP
Inventors: Rima Youil
Original assignee: Merck and Co Inc
Current assignee: Merck and Co Inc
Priority date: 2000-10-10
Filing date: 2001-10-05
Publication date: 2003-07-16
Also published as: WO2002031170A1; CA2424323A1; JP2004511233A

Abstract

Applicants have identified a process which exploits the bacterial homologous recombination system to convert double-stranded linear adenovirus genome into circularized plasmid form (See Figure 4). The system functions via adenoviral terminal fragments present on the plasmid that are less than 500 basepairs each. The result is a plasmid which is more readily analyzed by restriction digestion, PCR, DNA sequencing or used in transient transfection studies. The adenovirus plasmids that are generated can be rescued back into virus form. The entire procedure takes 4 days or less instead of the weeks required of plaque purification or dilution cloning isolation techniques. An additional plus of the instant invention is that the disclosed method does not require the use of tissue culture materials or facilities. The disclosed method allows for a more extensive and thorough examination of a viral preparation, in that it allows for the detection of variants incapable of propagation without the assistance of co-infecting intact adenoviral genomes. Under standard conditions of plaque purification, these variant genomes are not detected. It is predicted that far more variant genomes will be observed using the rapid method than would otherwise be detected by standard plaque purification methods.

Description

TITLE OF THE INVENTION

METHOD FOR CIRCULARIZLNG ADENOVIRAL NUCLEIC ACID VIA

HOMOLOGOUS RECOMBINATION

BACKGROUND OF THE INVENTION

Recombinant adenoviruses are widely used in gene therapy research. In fact, much work has been performed to remove specific components of the wild type adenovirus (e.g., El genes) in order to make them safer for human use. There are numerous concerns surrounding the utilization of adenovirus, just as there are in employing most other live viruses. For one, it is vitally important to ensure that the production of recombinant adenoviruses meets safety guidelines not only with regard to the absence of replication competent adenoviruses but for other variant species as well that may be present in a viral preparation.

Standard methods for the isolation of adenovirus (e.g., El-deleted adenovirus) usually require cloning by plaque purification on agar overlays of complementation cells (e.g., 293 cells). This procedure can take up to two weeks. To study the genome of the isolated virus, the plaque needs to be amplified through a series of passages which can introduce further genetic variants. Depending on the scale of amplification and the amount of viral DNA purified, radioactive labeling of restriction digestion products can be required. Furthermore, only a subset of viral genomes that are capable of forming a plaque are analyzed. Any viral genomes that have undergone extensive rearrangement such that they are no longer capable of growth without the assistance of nearby intact viral genomes are necessarily excluded in this approach. This method, thus, further prevents an accurate picture of genetic heterogeneity present in a preparation.

Alternative systems of isolating and manipulating adenovirus avoiding some of these steps are certainly preferred. Chartier et al. (U.S. Patent No. 6,110,735), for instance, teaches a method of homologous recombination within prokaryotic cells wherein a segment carrying a desired gene is imported into an adenovirus-carrying plasmid backbone. This system employs a plasmid carrying two short adenoviral segments which undergo homologous recombination with an adenoviral genome also introduced into the cell. This results in a means to import mutations or desired genes into the adenoviral sequences within the plasmid, and avoids the wildtype background inherent in the previous plaque purification methods. This method, further, curtails the timely and costly nature of adenoviral isolation. Homologous recombination, according to this method, is believed to rely on sequences other than the 5TTR, 3' ITR and encapsidation regions.

It would be advantageous to be able to utilize shorter segments of the adenoviral terminal sequences on a vector, for instance, terminal segments of under 500 basepairs, and still be able to incorporate adenovirus present within the cell into the vector. Just purely in terms of isolation of mutant adenovirus, which is only one possible use of such a method, a process employing shorter terminal segments narrowing into the 5' and 3' ITR regions would identify, with greater frequency, adenovirus bearing mutations in the regions outside of the terminal segments used on the plasmid, for instance, in the packaging region.

SUMMARY OF THE INVENTION

A method is provided for the circularization of adenovirus via homologous recombination. This method involves the importation of adenoviral ends comprising the 5' and 3' ITRs and the encapsidation (or packaging) signal of adenovirus into a vector (e.g., a plasmid) backbone. Homologous recombination between the adenoviral ends and a wild-type, mutant, or replication-deficient adenovirus present in the cell results in the wild-type, mutant, or replication-deficient adenovirus being circularized, or incorporated into the vector backbone. Adenovirus in this form is more readily manipulated and easier to screen and, rather than the weeks timeframe required of plaque purification or dilution cloning techniques, this process takes roughly four days or less. Infectious virus is, furthermore, readily rescued from the vector backbone.

DEFINITIONS

"replication-defective" or "first generation" vectors typically have a deleted or inactivated El gene region, and preferably have a deleted or inactivated E3 gene region as well.

"first adenoviral sequence" or "free adenoviral sequence" refers to any wild-type, mutant, or replication-defective adenovirus which possesses a region with at least 50 basepairs of homology with an area within the regions corresponding to the rightmost and leftmost 500 basepairs of a wildtype adenoviral sequence.

"mutant adenovirus" refers to an adenovirus different in sequence or conformation from wildtype adenovirus, which difference does not effect the replication of same.

9 - "second adenoviral sequence" refers to the 5' terminal adenoviral sequence present within the vectors of the instant invention that comprise at least the adenovirus 5TTR.

"non-adenoviral segment" or "non-adenoviral sequence", with respect to the plasmid or vector sequence, refers to nucleic acid sequence present on the plasmid or vector which is not adenoviral sequence; whether it be native to the plasmid or exogenous thereto.

"third adenoviral sequence" refers to the 3' terminal adenoviral sequence present within the vectors of the instant invention comprising at least the adenovirus 3' ITR.

BRIEF DESCRIPTION OF THE DRAWINGS

FIGURE 1 illustrates the recovery of full adenoviral genomes in a bacterial homologous recombination system. FIGURES 2A and 2B illustrate a restriction digestion analysis of isolated clones. Figure 2A shows EstEII analysis of eighteen individual adenovirus clones isolated by bacterial homologous recombination from a recombinant adenovirus passaged to P19. Figure 2B show Pad and HindΩl double digestion analysis of eighteen individual adenovirus clones isolated by bacterial homologous recombination from recombinant adenovirus passaged to P17.

FIGUREs 3A and 3B illustrate how a longer segment of adenoviral terminal regions can overlook a mutation in the early regions of the adenoviral genome. Figure 3A shows the recombination events that may occur if long terminal regions are used to create the "rescue" plasmid. The 5' end cross-over can override any variations present with a viral mixture. The top panel shows an example of a duplication of the packaging region within a population. The cross-over at the 5' end will not effect the wild-type (normal) species, however, the species containing either one or two duplications of the packaging region may go undetected if the cross-over event between the "rescue" plasmid and the variant genome occurs downstream or (3') of the last packaging region duplication. Figure 3B shows that by restricting the terminal homology ends of the "rescue" plasmid, the cross-over event is forced to occur 5' of the packaging region. In this way, all variants downstream (3') of the region of homology will be detected.

FIGURE 4 illustrates schematically the complete process from potential starting materials, a "rescue" plasmid and a viral DNA "pool". DETAILED DESCRIPTION OF THE INVENTION

Applicants have identified and herein disclose a method for incorporating adenoviral nucleic acid sequences present within a cell population into vector backbones. This method comprises introducing into a host cell an adenoviral nucleic acid sequence encoding a wild-type, mutant, or replication-defective adenovirus (herein referred to as a "first" adenoviral sequence) along with a linear vector comprising, at its ends, terminal adenoviral nucleic acid sequences comprising adenoviral 5' and 3' ITRs (herein, referred to as "second" and "third" adenoviral nucleic acid sequences, respectively). The goal and ultimate outcome of the disclosed method is the importation and isolation of individual adenoviral genomes into circularized plasmids. This process exploits the action of homologous recombination that occurs between two linear nucleic acid molecules present within a cell.

The benefit of such a procedure is that it eliminates the need for plaque purification of individual clones, amplification of the clones in tissue culture, CsCl banding of the virus, and dialysis of the purified virus. A large number of samples can be handled and processed easily without the need for tissue culture materials or facilities. This amounts to a significant time and cost advantage. In addition, large amounts of each adenoviral genome may be produced by cloning in E. coli, so that analysis is made by simple ethidium bromide staining after gel electrophoresis rather than the more tedious radiolabeling techniques used to visualize viral DNA.

Overall, this method allows us to isolate and analyze variants that may be present within a viral preparation, including those which are incapable of forming plaques. During propagation, such variant adenoviruses are capable of growth by utilizing necessary proteins provided in trans by intact adenoviruses within the mixed virus population. In such cases, these adenoviruses would escape detection unless extensive PCR technology is employed. Such an analysis of the viral pool from any preparation, particularly large scale, high passage number, or high multiplicity of infection propagations, will no doubt identify far more variant genomes than has so far been identified by routine plaque purification procedures. This procedure also enables production of plasmid preparations that can be subsequently used in transient transfection studies to analyze transgene expression. The isolated clones may also be used in rescue experiments to identify whether or not these genomes are capable of forming infectious virions. A significant advantage of the instant invention over previous methods such as that of Chartier et al. (U.S. Patent No. 6,110,735) is that the adenoviral terminal portions utilized on the plasmid encompass much shorter regions of the adenoviral ends than previous vectors. The shortened terminal ends afford many advantages to the process.

In addition to relieving the carry constraints of the importing plasmid, the shorter terminal ends provide another unexpected advantage. The range of mutants capable of isolation through homologous recombination is broadened. This was particularly noted for mutants created during the passaging of adenovirus. It has been observed that, during passage, rearrangements can take place in the packaging region of recombinant adenoviruses (Recombinant Advisory Committee meeting, Dec. 1999). To ensure that all rearrangements are capable of being identified, it is important to limit the regions of the terminal homology of the "rescue" plasmid (the plasmid carrying the two terminal segments which undergo homologous recombination with the adenovirus sequence) to what will still provide an acceptable level of genome recovery. If the region is too long then the cross over events during homologous recombination can take place downstream of the rearranged regions; see Figures 3A and 3B. If this occurs then the rearrangement would not be observed. Applicants have identified packaging region mutations by employing left and right adenoviral terminal regions of 195 and 477, respectively. Although the rescue plasmid can be prepared to contain almost any length of terminal homology with the adenovirus to be recovered, it is important to have a stretch of homology that allows for efficient homologous recombination. It is preferable that at least 50 bp of the extreme left ITR and 50 bp of the extreme right ITR segments of the adenovirus are used. Regions of this description can be readily cloned into most cloning vectors, e.g. pUC19.

The "rescue" plasmid, preferably, comprises basepairs 1-50 (preferably, 1-80, and most preferably, 1-100) of a wildtype adenovirus sequence not extending past basepair 500 of a wildtype adenovirus sequence, and basepairs 35,885- 35,935 (more preferably, 35,835-35,935) of a wildtype adenovirus sequence. More preferably, the portion representing the left terminal end of the adenovirus sequence consists essentially of basepairs 1-375 of a wildtype adenovirus sequence or, even more preferably, basepairs 1-195 of a wildtype adenovirus sequence. The right terminal end, most preferably, consists essentially of basepairs 35,458-35,935. The basepair numbers provided throughout the specification are based on the sequence of wildtype adenovirus serotype 5. This sequence is known and described in the art; see Chroboczek et al, 1992 J. Virology 186:280; which is hereby incorporated by reference. Accordingly, a particular embodiment of the instant invention is a vector wherein the sequences described above are of adenovirus type 5. One of skill in the art can readily identify the above regions in other adenovirus serotypes (e.g., serotypes 2, 4, 6, 12, 16, 17, 24, 31, 33, and 42), regions defined by basepairs corresponding to the above basepair positions given for adenovirus serotype 5. Accordingly, the instant invention encompasses all vectors comprising adenoviral terminal ends corresponding to basepairs 1-50 (preferably, 1-100), not to extend past basepair 500, and 35,885-35,935 (preferably, 35,835-35,935) of a wildtype adenovirus serotype 5 (Ad5) nucleic acid sequence. Particularly preferred embodiments of the instant invention are those derived from adenoviruses, like Ad5, which are classified in subgroup C (e.g., Ad2). Nucleic acid variants of said sequences are also contemplated for use in the instant invention. Nucleic acid variants (wherein one or more nucleotides differ from the wild-type or naturally occurring adenovirus sequence) may be produced using site directed mutagenesis or PCR amplification wherein the primer(s) have desired point mutations. Chemical mutagenesis may also be used, as can alternative methods known and available to one of skill in the art. The goal is to obtain sufficient homology (presumably with 50 basepairs or more) to allow for intermolecular homologous recombination to occur between the adenoviral terminal ends present on the vector and the free adenovirus present within the cell.

It is important when constructing a "rescue" plasmid that a unique restriction site exists between the left and right ITR sequences so that the rescue plasmid can be readily linearized between the two sets of sequences in order to prime the homologous recombination reaction. In addition, the "rescue" plasmid should also have an infrequent cutting restriction site (e.g. Pac I) at the junctions where the vector and ITR sequences meet, such that the vector backbone can be suitably removed from the entire adenovirus genomic sequence of the final adenovector plasmid. The Pac I site, in the above example, provides the option to rescue these isolated adenovectors into virus if needed. In both instances, it is recommended that the restriction enzyme site employed be absent from adenoviral sequences both in the terminal regions and in the first (or recombinant) adenoviral sequence; if not, at least underrepresented by comparison. The adenoviral terminal sequences can be obtained in any number of ways familiar to one of ordinary skill in the art. The terminal fragments can be generated by PCR, excised from any existing vector, or alternatively, produced synthetically by any standard method known in the art. A specific preferred embodiment of the instant invention, for instance, employs a "rescue" plasmid with terminal sequences recovered from an El- and E3 -deleted adenovirus.

Incorporation of the terminal sequences into the plasmid is also well within the realm of skills possessed by one of ordinary skill in the art. One method which Applicants have found useful involves first amplifying the terminal ends, with primers designed to contain a restriction site at the extreme ends in common with a restriction site contained within the vector's multiple cloning site (e.g. pUC19). This allows for the terminal ends of the PCR amplified product of each terminal end to ligate to the vector upon digestion with the enzyme. Immediately adjacent to the restriction site used for cloning into the vector of choice, the primers are designed to contain an appropriate restriction site that will be used to ultimately release the adenovector from the vector backbone (e.g. Pacl). Most preferably, the primers are also designed to contain a restriction site at their innermost ends in order that a unique restriction site is created. This unique restriction site is positioned between the two terminal PCR products upon cloning into the vector. This "rescue" plasmid is then linearized at this unique restriction site prior to undergoing homologous recombination.

The method found most useful to Applicants is illustrated by the following: a first terminal fragment can be amplified with a forward primer possessing an EcoRI restriction site, a Pad restriction site and basepairs 1-20 of a wildtype adenovirus, and a reverse primer possessing a BamHI site, a unique site (Ascl) and nucleotides 80-100 of a wildtype adenovirus. The second terminal fragment can be amplified by a first primer possessing BamHI and nucleotides 35,835-35,855, and a reverse primer possessing a HindlU site, a Pad site, and nucleotides 35,915-35,935. A vector, e.g., Pucl9, possessing the restriction sites found on the outside of the fragments (in this example, EcoRI, BamHI and Hindlll, in that order) can then be digested and the fragments inserted. The vector would then possess a Pad restriction site (for adenovector removal), the 5' terminal fragment, an Ascl site for vector linearization, the 3' terminal fragment and the Pad restriction site (for adenovector removal), in that order. Note, this example is meant solely to further illustrate a very particular embodiment of a number of methods understood and available to one of ordinary skill in the art to carry out an equivalent step and achieve an equivalent final result. Any of a number of means considered conventional in the art can be employed to create a vector possessing adenoviral terminal fragments capable of undergoing homologous recombination with free adenoviral genomes present in the same cell.

The vector nucleic acid, all except for any adenoviral sequences, will be referred to as the segment of non-adenoviral sequence on the vector. Preferably, the vector will comprise a bacterial origin of replication. The vector may also contain a selectable marker gene such as an antibiotic (e.g., ampicillin) resistance gene. One of ordinary skill in the art is aware that any of a variety of vectors can be used to effect intermolecular homologous recombination between the adenoviral terminal ends present on the vector and free adenovirus within the cell. Particularly preferred expression vectors include pUC18 and its derivatives, pUC19 and its derivatives, pBR322 and its derivatives, the pBluescript series, the pGEM series (PromegaTM), pET series (PromegaTM), and pESP-1 (StratageneTM). Generally, the preferred vector will be a general vector with a reasonably high copy number in the appropriate host strains used for its amplification. Cosmids, phages, BACs (bacterial artificial chromosomes) and YACs (yeast artificial chromosomes) and other alternative vectors are also of use in the instant invention. In order for efficient homologous recombination to occur, the vector is linearized between the two adenoviral segments described above (herein referred to as the second and third sequences, respectively). The ends of these sequences as well as the first adenoviral sequence introduced may be blunt ended or may contain 5' and 3' overhangs. There is no absolute requirement for phosphatase treatment of the fragments since certain host cells, such as the BJ5183 bacterial cells, are not capable of re-circularizing linear DNA molecules. Therefore, as long as there are no contaminating circular plasmids present, the background of re-circularized "rescue" plasmid backbone is very low. Consequently, a single co-transformation between the linear "rescue" plasmid and the viral DNA pool can yield hundreds of colonies that can be selected and amplified for analysis with a high level of confidence.

The preferred host cell is prokaryotic, and is most preferably bacterial. Particularly preferred are cells of E. coli. Cells and cell lines of particular interest are derived from E. coli BJ5I83. This system relies on the use of the RecF pathway of an E. coli recBC, sbcBC strain. Preferably, the viral nucleic acid to be incorporated into the vector (referred to herein as the first adenoviral sequence) is purified viral nucleic acid. All viral nucleic acid referred to throughout the specification (first, second and third nucleic acid sequences) can be DNA (e.g., PCR product, genomic or cDNA), RNA or any analog thereof and is preferably linear. The purified viral nucleic acid can be an entire adenoviral genomic mixture extracted from purified virus. The virus can, furthermore, be obtained from very high passage number propagation. Alternatively, the virus can be obtained from very low passage number propagation, for instance, in a case where one wishes to create recombinants in El-complementing cell lines and then circularize them into a vector backbone in bacterial cells (e.g., BJ5183). In this case, it is important to limit the number of passages to ensure that wild-type adenovirus does not outcompete the recombinant adenovirus (as will be the case on continued passaging).

Preferably, the first adenoviral sequence is an El-deleted adenovirus, an E3-deleted adenovirus, or an El- and E3- deleted adenovirus.

In particularly preferred embodiments, the first adenoviral sequence comprises an exogenous gene. Genes of use therapeutically or for vaccine purposes are specifically contemplated for use herein. Exogenous genes encoding a protein of interest can exist in the form of an expression cassette. A gene expression cassette preferably comprises (a) a nucleic acid encoding a protein of interest, (b) a heterologous promoter operatively linked to the nucleic acid encoding the protein, and (c) a transcription terminator.

The transcriptional promoter is preferably recognized by an eukaryotic RNA polymerase. In a preferred embodiment, the promoter is a "strong" or "efficient" promoter. An example of a strong promoter is the immediate early human cytomegalovirus promoter (Chapman et al, 1991 Nucl. Acids ResT9:3979-3986, which is incorporated by reference) with or without the intron A sequence (CMV- intA and CMV(no intron), respectively). Those skilled in the art will appreciate that any of a number of other known promoters, such as the strong immunoglobulin, or other eukaryotic gene promoters may also be used, including the EFl alpha promoter, the murine CMV promoter, Rous sarcoma virus (RSV) promoter, SV40 early/late promoters and the beta-actin promoter.

In preferred embodiments, the promoter may also comprise a regulatable sequence such as the Tet operator sequence. This would be extremely useful, for example, in cases where the gene products are effecting a result other than that desired and repression is sought.

Preferred transcription termination sequences present within the gene expression cassette are the bovine growth hormone terminator/polyadenylation signal (bGHpA) and the short synthetic polyA signal (SPA) of 50 nucleotides in length, defined as follows: AATAAAAGATCTTTATTTTCATTAGATCTGTGTGTTGGT- TTTTTGTGTG (SEQ JO NO:JL).

Accordingly, another method provided herein is a method of preparing, within a vector backbone (that described above), recombinant adenovirus carrying a desired gene. This method comprises (a) generating recombinant adenovirus carrying a desired gene in complementation cells by transfecting said cells with a vector comprising the desired gene flanked by adenoviral sequences homologous to a region within an adenoviral genome, also present within the cell, which is targeted for homologous recombination; and (b) extracting recombinant adenoviral nucleic acid produced. This nucleic acid is then introduced into a bacterial (or other prokaryotic) cell along with a linear vector comprising second and third adenoviral sequences in accordance with the above description. The first adenoviral sequence undergoes homologous recombination with the second and third adenoviral sequences in the vector resulting in a vector comprising both the non-adenoviral segment of the vector and the recombinant first adenoviral sequence.

This process utilizes the same process set forth above and follows all the same guidelines drawn out above as to the substrates, conditions, etc. What Applicants have provided here is a source of free adenoviral nucleic acid and, in doing so, provided a complete process for isolating replication-defective vectors containing desired genes.

Any of a number of complementation cells can be used in this method. Particularly preferred complementing cells are of the various El complementing cell lines, including the known cell lines 293 and PER.C6™. Both these cell lines express the adenoviral El gene product. PER.C6™ is described in WO 97/00326 (published January 3, 1997) and issued U.S. Patent No. 6,033,908. It is a primary human retinoblast cell line transduced with an El gene segment that complements the production of replication deficient (FG) adenovirus, but is designed to prevent generation of replication competent adenovirus by homologous recombination. Cells of particular interest have been stably transformed with a transgene that encodes the Ad5ElA and EIB gene, like PER.C6™, from 459-3510 basepairs inclusive. 293 cells are described in Graham et al, 1977 J. Gen. Virol. 36:59-72.

Another method provided which employs the above described process is a method of preparing recombinant adenovirus carrying a desired gene within a vector backbone. This comprises (a) introducing into complementation cells a first adenoviral sequence encoding a wild-type, mutant, or replication-defective adenovirus, said sequence comprising a restriction enzyme site within a region targeted for homologous recombination; (b) generating recombinant adenovirus carrying a desired gene in said complementation cells by transfecting said cells with a vector comprising the desired gene flanked by adenoviral sequences homologous to the region targeted for homologous recombination within the adenoviral genome; (c) digesting resultant cell products with the restriction enzyme specifically capable of digesting the restriction enzyme site of step (a); (d) extracting the resultant adenoviral nucleic acid from the complementation cells and (e) introducing the nucleic acid into bacterial (or prokaryotic) cells along with a linear vector comprising second and third adenoviral nucleic acid sequences as well as a segment of non-adenoviral sequence, the second sequence comprising basepairs 1-100 of a wildtype adenovirus sequence not to extend past basepair 500, and the third sequence comprising basepairs 35,835- 35,935 of a wildtype adenovirus sequence. The first adenoviral sequence then undergoes intermolecular homologous recombination with the second and third adenoviral sequences in the vector of step (e) resulting in a vector comprising both the non-adenoviral segment of the vector of step (e) and the recombinant adenoviral sequence. As with other methods disclosed herein, all of the same preferred embodiments, substrates and guidelines apply. The presence of the restriction enzyme site in step (a) allows for ready identification of sequence insertions. Insertion of a gene into the targeted region containing the restriction site eliminates the restriction site (e.g., Claϊ). Subsequent digestion with the restriction enzyme (e.g., Claϊ) renders those sequences not possessing the desired gene (those sequences susceptible to digestion with the enzyme) incapable of undergoing homologous recombination with both second and third adenoviral sequences in the vector of step (e), and ensures that the sequences that undergo homologous recombination are the ones possessing the desired insertion. This process streamlines the isolation of adenoviral vectors carrying an insertion of interest. EXAMPLES

Example 1

Enzymes and Reagents The restriction enzymes Pac I, Xba I, Avr II, HinάYSl, and Bst EII were purchased from New England Biolabs, Inc. Buffer saturated phenol used for the extraction of DNA from bacterial cell pellets or from virus was purchased from GibcoBRL, Cat# 15513-039. DNA purification kits QIAEXII (Cat#20021), Qiagen plasmid purification columns (Cat #12145) and reagents PI (Cat# 19051), P2 (Cat# 19052) and P3 (Cat# 19053) were all purchased from Qiagen.

Example 2

Extraction of Viral DNA

Recombinant adenovirus was studied from passage 17 and passage 19 preparations. The recombinant adenovirus contained a transgene of 3500 bp in size, inserted into the El region of an El- and E3-deleted Ad5 vector. Viral DNA was extracted from each of these two passages by treating the purified virus with pronase digestion followed by standard phenol extraction and isopropanol precipitation. The purified viral DNA was resuspended in 20 μl dH2O. The DNA concentration was determined by OD measurement at A260nm.

Example 3

Construction of the "rescue" plasmid

The rescue vector was prepared by digesting an El and E3 deleted adenovector pre-plasmid (adenovirus within a plasmid backbone) with Xba I and Avr II. This double digestion resulted in 5 fragments of sizes 12179 bp, 7553 bp, 7103 bp, 2885 bp and 2487 bp. The 2885 bp fragment was gel extracted using QIAEXII kit (Cat# 20021). This fragment contains the vector backbone carrying an origin of replication, the ampicillin resistance gene, 375 bp of the extreme left end of the Ad5 genome and 477 bp of the extreme right end of the Ad5 genome.

An alternative restriction enzyme, BsrGl, which limits the 5' adenovirus region to a stretch of 195 bp was also employed, and shown to be capable of providing as efficient genome recovery as the Xbal restriction enzyme.

The "rescue" plasmid can be constructed to contain any desired length of adenoviral terminal homology. To construct a "rescue" plasmid that contains 100 bp of terminal homology to Ad5, the PCR primers can be designed to amplify the extreme 5' 100 basepairs of the Ad5 wild-type genome and the extreme 3' 100 basepairs of the wild-type Ad5 genome. By using appropriately designed PCR primers that contain suitable restriction sites, the PCR products can be cloned into a suitable cloning vector such as pUC18. The 3' end of the left end adenovirus PCR product and the 5' end of the right end PCR product should contain the same unique restriction site so that cloning of the two fragments together into the cloning vector will create a single unique restriction site that will be used to linearize the "rescue" plasmid. PCR reaction is performed according to conventional methods using wild- type adenovirus 5 DNA as the DNA template. Alternatively, "rescue" plasmids can be created to encompass left end and right end termini of any serotype.

Example 4.

E. coli BJ5183 homologous recombination Approximately, lOOng each of the purified adenoviral DNA and about

100-200 ng of the "rescue" plasmid fragment containing the ITR segments were added to sterile microfuge tubes. The total volume of this DNA mixture was limited to 2-6 μl. Each DNA mixture was then transformed into 100 μl of chemically competent E. coli BJ5183 cells in accordance with that described in Hanahan D., 1983 J. Molec. BioL 166:577-580; which is hereby incorporated by reference. The DNA and cells were then incubated on ice for 20 minutes. The cells were heat- shocked in a 42°C water bath for 45 seconds and then immediately placed on ice for 2 minutes. A 600 μl volume of SOC media (GibcoBRL Cat# 15544-042) at room temperature was added to each tube and incubated at 37°C, 220 rpm for 45 minutes. Following incubation, each transformation culture was plated evenly onto two pre- warmed ampicillin agar plates (BBL™ Cat# 298273). The plates were left to dry at 37°C before being inverted and allowed to incubate overnight. Colonies were randomly selected and inoculated into 2 ml of Terrific broth containing 100 μg/ml ampicillin; according to the procedures described in Sambrook et -?/., 1989 Molecular Cloning. A Laboratory Manual, Second Edition. Cold Spring Harbor Laboratory Press.; which is hereby incorporated by reference. The culture was incubated at 37°C, 220 rpm for approximately 6 hours until cell turbidity was observed. The cells were pelleted by centrifugation at 4,200 rpm at 4°C for 10 minutes. The supernatant was discarded and the tubes briefly drained by inverting onto a hand towel. Example 5

Extraction of Total DNA from E. coli BJ5183 Transformation

The cell pellet was resuspended with 150 μl of Qiagen solution PI (containing RNasel) and transferred to a sterile microfuge tube. 150 μl of Qiagen solution P2 (lysis buffer) was added and each tube was gently inverted 4-6 times. Lysis was allowed to occur for no longer than 5 minutes, then 150 μl of Qiagen solution P3 (neutralization solution) was added to each tube. Mixing was performed by inverting gently 4-6 times. A white precipitate was observed. The tubes were placed on wet ice for 10 minutes, then on dry ice for a further 5 minutes. The precipitate was pelleted by microfuge centrifugation for 10 minutes at 12,000 rpm. The supernatant was transferred to a fresh microfuge tube. After Phenol-chloroform extraction, the DNA was precipitated with isopropanol. To dry the pellet, tubes were placed in a lyophilizer for 5 minutes at high heat. Alternatively, the pellets were air- dried for 15-20 minutes. After all remnants of the isopropanol was removed, the pellet was resuspended in 20 μl of dH2O.

Example 6

E. coli XL-1 Transformation

2 μl of each purified DNA from the BJ5183 cells was added to 50 μl of chemically competent E. coli XL1 blue cells (Stratagene Cat# 200249). After incubation on ice for 20 minutes, the transformation mixture was heat-shocked at 42°C for 45 seconds and then immediately placed on ice for 2 minutes. 300 μl of SOC medium (GibcoBRL Cat# 15544-042) at room temperature was then added to each tube. Before plating out 50 μl and 150 μl of each transformation onto pre- warmed LB + ampicillin containing plates (BBL™ Cat# 298273), the transformation mixture in SOC was incubated at 37°C, 220 rpm for 45 minutes. The plates were allowed to dry before inverting and then were incubated overnight at 37°C. One colony from each BJ5183 transformation was selected and grown overnight in 50 - 100 ml of Luria broth (Sambrook et al., supra) containing 100 μg/ml ampicillin. Qiagen columns (Cat #12145) were used to prepare a mini-prep of the plasmid DNA. The DNA was resuspended in 20 μl of dH2O. A260nm reading of the solutions were used to determine the DNA concentration. Example 7

Diagnostic Restriction Analysis

Suitable restriction enzymes were chosen that allowed identification of any possible rearrangements along the length of the adenoviral genome, as well as in the transgene region. To compare the restriction pattern of the plasmid recovered viral genomes with purified viral DNA, the plasmid DNA samples were first digested with Pac I to release the vector backbone prior to digestion with restriction enzymes of choice. The resulting pattern was then compared with the purified viral DNA from the pooled lot that was originally used to recover individual viral genomes. 1 μg DNA from each sample was digested with the appropriate restriction enzyme and loaded onto a 0.8-1.0% agarose gel. Following electrophoresis, the gel was stained with ethidium bromide and observed on a UN transilluminator.

Example 8 Results

Figures 2A and 2B show the gel pattern of 36 individual viral plasmid clones isolated by bacterial homologous recombination. Figure 2A shows the restriction digestion pattern following Bst Eϋ. digestion of 18 individual clones in comparison with the pre-plasmid DΝA that was used in the original rescue of the virus. As shown, there are various rearrangements identified from this virus at passage 19. In comparison with the control pre-plasmid lane, only one out of the eighteen samples (clone #13) shows a pattern identical to the control pre-plasmid. There are four other types of restriction patterns that are identifiable in this analysis. Eleven out of the eighteen show one type of pattern (clones # 2, 4, 5, 6, 7, 9, 10, 11 14, 15, and 17), four out of eighteen show a different pattern (clones # 1, 8, 16 and 18). Clones #3 and #12 each show a unique pattern that also differs from the parent pre-plasmid.

Figure 2B shows Pac 1 and Hind /// double digestion of a second set of 18 clones isolated from an earlier passage of the same virus (passage 17). The purpose of performing a Pac I digestion is to remove the vector backbone, so that the restriction patterns may be directly compared with the viral DΝA preparation used for the isolation of individual genomes. In this cohort, only two clones (clone #6 and #18) show a pattern identical to the control pre-plasmid. Aside from the normal pattern, there are six types of restriction patterns that are identifiable in this analysis. Clones 1, 3, 8 and 15 show one set of digestion pattern. Different restriction patterns are observed in clones from lanes 2 and 5, clones from lanes 9 and 14 and clones from lanes 7, 10, 11 , 12, 13 and 17. Clones from lanes 4 and 16 each show a unique restriction pattern.

To ensure that the rearrangements identified in this study were not induced by the E. coli itself, a viral preparation known to be free of rearrangements (as analyzed by radioactive restriction analysis), did not result in the observation of any rearrangements following the isolation of 50 genomes using this bacterial homologous recombination system (data not shown).

Claims

WHAT IS CLAIMED IS:

1. A method for incorporating a nucleic acid sequence encoding adenovirus into a vector backbone, which comprises introducing into a host cell: (a) a first adenoviral nucleic acid sequence encoding a wild-type, mutant or replication-defective adenovirus, and

(b) a linear vector comprising second and third adenoviral nucleic acid sequences as well as a segment of non-adenoviral sequence, the second sequence comprising basepairs corresponding to basepairs 1-100 of a wildtype adenovirus sequence not to extend past basepair 500 of a wildtype adenovirus sequence, and the third sequence comprising basepairs corresponding to basepairs 35,835-35,935 of a wildtype adenovirus sequence; wherein the first adenoviral sequence undergoes homologous recombination with the second and third adenoviral sequences in the vector of step (b) resulting in a vector comprising both the non-adenoviral segment of the vector of step (b) and the first adenoviral sequence.

2. A method in accordance with claim 1 wherein the second adenoviral sequence consists of basepairs 1-375 of a wildtype adenovirus sequence.

3. A method in accordance with claim 1 wherein the second adenoviral sequence consists of basepairs 1-195 of a wildtype adenovirus sequence.

4. A method in accordance with claim 1 wherein the third adenoviral sequence consists of basepairs 35,458-35,935 of a wildtype adenovirus sequence.

5. A method in accordance with claim 1 wherein the second adenoviral sequence consists of basepairs 1-375 of a wildtype adenovirus sequence and the third adenoviral sequence consists of basepairs 35,458-35,935 of a wildtype adenovirus sequence.

6. A method in accordance with claim 1 wherein the second adenoviral sequence consists of basepairs 1-195 of a wildtype adenovirus sequence and the third adenoviral sequence consists of basepairs 35,458-35,935 of a wildtype adenovirus sequence.

7. A method in accordance with claim 1 wherein the host cell is prokaryotic.

8. A method in accordance with claim 7 wherein the host cell is bacterial.

9. A method in accordance with claim 8 wherein the host cell is from E. coli.

10. A method in accordance with claim 9 wherein the host cell is an E. coli BJ5183 cell.

11. A method in accordance with claim 1 wherein the nucleic acid is DNA.

12. A method in accordance with claim 1 wherein the nucleic acid is cDNA.

13. A method in accordance with claim 1 wherein the first adenoviral nucleic acid sequence encodes an Εl-deleted adenovirus.

14. A method in accordance with claim 1 wherein the first adenoviral nucleic acid sequence encodes an Ε3-deleted adenovirus.

15. A method in accordance with claim 1 wherein the first adenoviral nucleic acid sequence encodes an El- and E3-deleted adenovirus.

16. A method in accordance with claim 1 wherein the adenoviral nucleic acid sequence comprises an exogenous gene.

17. A method in accordance with claim 1 wherein the vector is a plasmid.

18. A method in accordance with claim 17 wherein the plasmid is pUC19.

19. A method in accordance with claim 17 wherein the plasmid comprises a restriction enzyme site which allows for ready linearization of the plasmid.

20. A method in accordance with claim 19 wherein the restriction enzyme site is placed between the second and third nucleic acid sequences of step (b).

21. A method in accordance with claim 1 wherein the vector comprises a restriction enzyme site between the non-adenoviral segment of the vector of step (b) and the 5' and 3' ITRs to allow for rescue of the encoded adenovirus.

22. A method in accordance with claim 1 wherein the vector comprises an origin of replication.

23. A method in accordance with claim 1 wherein the vector comprises a selectable marker gene.

24. A method in accordance with claim 1 wherein the vector comprises a gene conferring resistance to a particular substance.

25. A method in accordance with claim 1 wherein the vector comprises an ampicillin resistance gene.

26. A method in accordance with claim 1 wherein the adenovirus is of the subgroup C adenoviruses.

27. A method in accordance with claim 16 wherein the adenovirus is wildtype adenovirus type 5.

28. A method of preparing, within a vector backbone, recombinant adenovirus carrying a desired gene, which comprises: (a) generating recombinant adenovirus carrying a desired gene in complementation cells by transfecting said cells with a vector comprising the desired gene flanked by adenoviral sequences homologous to a region within an adenoviral genome also present within the cells which is targeted for homologous recombination;

(b) extracting the recombinant adenoviral nucleic acid; and

(c) introducing into a host cell:

(i) a first adenoviral nucleic acid sequence encoding said recombinant adenoviral sequence, and (ii) a linear vector comprising second and third adenoviral nucleic acid sequences as well as a segment of non-adenoviral sequence, the second sequence comprising basepairs corresponding to basepairs 1-100 of a wildtype adenovirus sequence not to extend past basepair 500 of a wildtype adenovirus sequence, and the third sequence comprising basepairs corresponding to basepairs 35,835-35,935 of a wildtype adenovirus sequence; wherein the first adenoviral sequence undergoes homologous recombination with the second and third adenoviral sequences in the vector of step (c)(ii) resulting in a vector comprising both the non-adenoviral segment of the vector of step (c)(ii) and the recombinant first adenoviral sequence.

29. A method in accordance with claim 28 wherein the second adenoviral sequence consists of basepairs 1-375 of a wildtype adenovirus sequence.

30. A method in accordance with claim 28 wherein the second adenoviral sequence consists of basepairs 1-195 of a wildtype adenovirus sequence.

31. A method in accordance with claim 28 wherein the third adenoviral sequence consists of basepairs 35,458-35,935 of a wildtype adenovirus sequence.

32. A method in accordance with claim 28 wherein the second adenoviral sequence consists of basepairs 1-375 of a wildtype adenovirus sequence and the third adenoviral sequence consists of basepairs 35,458-35,935 of a wildtype adenovirus sequence.

33. A method in accordance with claim 28 wherein the second adenoviral sequence consists of basepairs 1-195 of a wildtype adenovirus sequence and the third adenoviral sequence consists of basepairs 35,458-35,935 of a wildtype adenovirus sequence.

34. A method in accordance with claim 28 wherein the complementation cells are 293 cells.

35. A method in accordance with claim 28 wherein the cells are PER.C6™ cells.

36. A method of preparing, within a vector backbone, recombinant adenovirus carrying a desired gene, which comprises:

(a) introducing into complementation cells a first adenoviral sequence encoding a wild-type, mutant, or replication-defective adenovirus, said sequence comprising a restriction enzyme site within a region targeted for homologous recombination;

(b) generating recombinant adenovirus carrying a desired gene in said complementation cells by transfecting said cells with a vector comprising the desired gene flanked by adenoviral sequences homologous to the region targeted for homologous recombination within the adenoviral genome;

(c) digesting resultant cell products with a restriction enzyme capable of digesting the restriction enzyme site of step (a);

(d) extracting the recombinant adenovirus from the complementation cells; and

(e) introducing into a bacterial cell both the recombinant adenovirus and a vector comprising second and third adenoviral nucleic acid sequences as well as a segment of non-adenoviral sequence, the second sequence comprising basepairs corresponding to basepairs 1-100 of a wildtype adenovirus sequence not to extend past basepair 500 of a wildtype adenovirus sequence, and the third sequence comprising basepairs corresponding to basepairs 35,835-35,935 of a wildtype adenovirus sequence; wherein the first adenoviral sequence undergoes homologous recombination with the second and third adenoviral sequences in the vector of step (e) resulting in a vector comprising both the non-adenoviral segment of the vector of step (e) and the recombinant first adenoviral sequence.

37. A method in accordance with claim 36 wherein the second adenoviral sequence consists of basepairs 1-375 of a wildtype adenovirus sequence.

38. A method in accordance with claim 36 wherein the second adenoviral sequence consists of basepairs 1-195 of a wildtype adenovirus sequence.

39. A method in accordance with claim 36 wherein the third adenoviral sequence consists of basepairs 35,458-35,935 of a wildtype adenovirus sequence.

40. A method in accordance with claim 36 wherein the second adenoviral sequence consists of basepairs 1-375 of a wildtype adenovirus sequence and the third adenoviral sequence consists of basepairs 35,458-35,935 of a wildtype adenovirus sequence.

41. A method in accordance with claim 36 wherein the second adenoviral sequence consists of basepairs 1-195 of a wildtype adenovirus sequence and the third adenoviral sequence consists of basepairs 35,458-35,935 of a wildtype adenovirus sequence.

42. A method in accordance with claim 36 wherein the restriction enzyme site is incorporated into the El region.

43. A method in accordance with claim 36 wherein the restriction enzyme site is a Clal site.

44. A method in accordance with claim 36 wherein the complementation cells are 293 cells.

45. A method in accordance with claim 36 wherein the cells are PER.C6™ cells.

99