UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Shuttling of a functional cre recombinase gene into mammalian cells using a retroviral vector Hendry, Jerrod 1995

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1995-0628.pdf [ 8.81MB ]
Metadata
JSON: 831-1.0098997.json
JSON-LD: 831-1.0098997-ld.json
RDF/XML (Pretty): 831-1.0098997-rdf.xml
RDF/JSON: 831-1.0098997-rdf.json
Turtle: 831-1.0098997-turtle.txt
N-Triples: 831-1.0098997-rdf-ntriples.txt
Original Record: 831-1.0098997-source.json
Full Text
831-1.0098997-fulltext.txt
Citation
831-1.0098997.ris

Full Text

SHUTTLING OF A FUNCTIONAL CRE RECOMBLNASE GENE INTO MAMMALIAN CELLS USING A RETROVIRAL VECTOR By: Jerrod Hendry B.Sc, The University of British Columbia, 1992 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES Genetics Programme We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA October 1995 © Jerrod Hendry, 1995 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department The University of British Columbia Vancouver, Canada DE-6 (2/88) ABSTRACT Cre recombinase of bacteriophage PI has become a useful tool for molecular biological studies in vivo as well as in vitro. The Cre protein is an essential part of a system which catalyzes recombination between specific D N A sequence repeats called loxP sites. This recombination system allows for excision or inversion, depending on loxP orientation, Of D N A sequences flanked by loxP sites. In order to utilize the recombinase in mammalian systems, the retroviral vectors, L X S N and L X S H D , have been incorporated into the production of shuttle vectors and retroviral producing cell lines. The produced retroviruses allow for the transfer into and expression of the Cre recombinase in mammalian cells. These newly constructed retroviral particles will be a useful tool for inducing specific gene ablation in infective mammalian cells. T A B L E OF CONTENTS Abstract 11 Table of Contents m-iv List of Tables v List of Figures vi-vn Abbreviations V111-1X Acknowledgment x Chapter I: Chapter JI: INTRODUCTION Cre Recombinase Retroviral Transfer Systems Thesis Objectives MATERIALS A N D METHODS D N A isolation and purification Isolation of High Molecular Weight D N A Isolation of Plasmid D N A Cloning Cloning of Retroviral Vectors Cloning of Target Vector Cell Lines Tissue Culture Production of Amphotropic Retroviral Producing Cell Lines Production of Stable Target Cell Lines Southern Blot Analysis Preparation of Radiolabeled Probes RESULTS Cloning of Retroviral Vectors Transfer of Retroviral Genome Into Infected Cells Viral Titers Copy Number of Target Vectors in lox^Hyg NIH3T3 Cell Line Copy Number of Target Vectors in Flox NIH3T3 Cell Line Initial Indication of Recombination Using Double Selection of Infected Target Cells Southern Blot Analysis of Cre Function Using Target Cells 1) lox 2 Hyg NIH3T3 Target Cells 2) Flox NIH3T3 Target Cells Amount of Recombination in the Total Cell Population Over Time Chapter IV: DISCUSSION AND CONCLUSIONS REFERENCES V Chapter in Table 1 LIST OF TABLES The Names of the Newly Produced Cre Containing Retroviruses Table 2 Expected Band Sizes for Genomic DNA of Retrovirally Infected Cells Digested with EcoRV and Probed with 32p labeled Cre Gene Sequence Table 3 Retroviral Titers Produced by Clonal PA317 Retroviral Producing Cells Table 4 Indication of Cre Recombinase Functioning In Target Cells Using the Difference In Selection Markers Between the Target Vectors and Retroviruses Table 5 Densitometry of the Autoradiography from Figures 9 and 10. Table 6 Densitometry of the Autoradiographs from Figures 11 and 12. vi LIST O F F I G U R E S Chapter II Figure 1 Schematic representations of the LXSN and LXSHD retroviral constructs A schematic representation of the plasmids containing the Cre recombinase genes Figure 3 Schematic representations of the Cre-containing retroviruses Figure 4 Determination of genome transfer for the Lcre-KSN and Lcre-KSHD retroviruses Figure 5 Determination of genome transfer for the LNKcreSN and LNKcreSHD retroviruses Figure 6 Southern Blot analyses of lox2Hyg NIH3T3 genomic DNA from isolated clones Figure 7 Southern Blot Analyses of clonal Flox NIH3T3 genomic DNA Figure 8 A schematic representation of the method used to initially determine if Cre recombinase was functional in the retrovirally infected cells Figure 9 Southern blot analysis of retrovirally infected lox2Hyg clone #1 target cells using SphI and PvuII restriction enzymes Figure 10 Southern blot analysis of retrovirally infected lox2Hyg clone #1 target cells using SphI and Bgll restriction enzymes Figure 11 Southern blot analysis of retrovirally infected Flox clone #1 target cells using Sspl and Hindm restriction enzymes Figure 12 Southern blot analysis of retrovirally infected Flox clone #1 target cells using Sspl and Xhol restriction enzymes Figure 2 Chapter in vii Figure 13 Determination of change in amount of recombination over time in infected lox^ Hyg clone #1 NIH3T3 cells Figure 14 Determination of change in amount of recombination over time in infected Flox clone #1 NIH3T3 cells viii ABBREVIATIONS bp base pair(s) cm centimeter C M V cytomegalovirus CsCl2 Cesium chloride dATP deoxyriboadenosine 5'-triphosphate ddH20 double distilled water D M E M Dulbecco's modified Eagles medium DNA Deoxyribonucleic Acid DTT dithiothreitol EDTA ethylenediaminetetra-acetic acid ES embryonic stem EtBr Ethidium bromide EtOH ethanol F Farad FCS fetal calf serum hr hour(s) Kb Kilobase pair(s) kd kilodalton(s) 1 liter LTR Long Terminal Repeat M mole(s) per liter min minute(s) ml milliliter M - M L V Moloney murine leukemia virus M-MSV Moloney murine sarcoma virus mwt molecular weight ng nanogram PBS phosphate buffered saline 32P radioactive phosphate RNA Ribonucleic Acid rpm revolutions per minute SDS sodium dodecyl sulphate Tris trisodium citrate M-g microgram |il microliter V Volts ACKNOWLEDGEMENTS I would like to thank Jamey Marth for his advice and patience over the past few years and to the members of the Marth lab for their support and friendship during my stay at the B.R.C. I also would like to thank all the other students and staff at the B.R.C. who have helped along the way. I would especially like to acknowledge my parents and relatives without whom I would not be where I am today. Finally, I wish to thank Carmen for being there even when we were separated by so many miles. 1 CHAPTER I INTRODUCTION Site-specific recombination systems have become novel tools for the manipulation of DNA in both in vitro and in vivo studies. Each system involves a highly specific DNA-protein interaction, recognition of a DNA target site and catalysis of cleavage and ligation of DNA at target sites. These systems catalyze inversion or excision, depending on the orientation of the target sites, of DNA intervening the target sites as well as integration of DNA sequences by interaction of the recombinases with target sites on separate molecules of DNA. Most of these systems involve similar components but catalyze reactions of varying complexities (Stark et al. 1992, and Sadowski, 1993). The Cre-lox system from bacteriophage PI and FLP/FRT system from yeast (Broach et al. 1982 and Senecoff et al. 1986) are two well defined site-specific recombination systems which have been utilized in mammalian system studies. The Cre-lox and FLP/FRT systems have become of interest to molecular biologists because of their ability to produce site-specific DNA manipulations simply by the presence of the recombinase protein and small DNA target sites. Both of these systems involve the use of 34 bp DNA target sites and a corresponding recombinase protein. The target sites of both recombinases are comprised of two 13 bp palindromic sequences separated by an 8 bp asymmetric core sequence. The 8 bp core determines the orientation of the target site and is the area where cleavage and religation of the DNA occurs. The 13 bp repeat elements bind the recombinases for the reaction. Both areas of the target sites are required for proper recombination to occur. All of the recombination reactions Of Cre and FLP occur with base-pair precision. This precision allows for the production of active target sites even after a recombination event. Because active target sites are produced after recombination, the recombination reactions have been shown to be reversible. The reversibility means that sequences which are inverted once are equally likely to inverted again but because the proximity of intramolecular target sites required for excision are much greater than intermolecular target sites required for integration the reaction equilibrium causes excision of target-flanked DNA to be much greater than integration. Both of these recombinase proteins are a part of the integrase family of recombinases and contain conserved histidine, arginine, and tyrosine residues at the C-termini of the integrases. These residues are thought to make up the active sites of the proteins (Argos et al. 1986) which are involved in the breaking and rejoining of the DNA backbone during the recombination process. Previous studies have shown that mutations to the arginine residues of these proteins severely affect the catalytic activities of Cre and FLP but do not have much affect on the site-specific binding (Abremski et al. 1992). Besides the few conserved residues, these proteins are very different. The difference in the proteins is thought to explain why the proteins interact with differing target sites. The Cre and FRT proteins have both been shown to be functional in a number of heterologous cell systems (Sauer et al, 1988, Golic et al, 1989, O'Gorman et al. 1991, Maeser et al. 1991, Onouchi et al. 1992) and in many transgenic systems. The Cre protein has been shown to be functional in cultured plant and mammal cells, murine embryonic stem cells and transgenic mammals and plants (Kilby etal.. 1993). The FLP recombinase has only shown to be functional in cultured mammal and plant Cells as well as yeast and insect cells (Kilby et al. 1993). Several unsuccessful attempts to generate FLP producing transgenic mice but Stephen O'Gorman of the Salk Institute has recently stated that the FLP enzyme does function in transgenic mice (Barinaga, 1994). Because FLP previously was not shown to function in transgenic studies, the Cre-lox system has been used slightly more than the FLP/FRT system in today's studies. We have also chosen to use the Cre-lox system in our studies simply because at this point it is the recombinase of choice and is used slightly more often than the FLP/FRT system. CRE RECOMBINASE Cre recombinase is a 38 Kd protein which was discovered and cloned from bacteriophage PI. The Cre recombinase catalyzes a site specific recombination between two 34 bp repeats called loxP site (locus of cross-over) (Abremski et al. 1983, Hoess and Abremski, 1985, and Sternberg and Hamilton, 1981). One of the normal functions of Cre in bacteriophage PI is to resolve PI DNA dimers, formed by homologous recombination between two daughter molecules, into monomers for proper segregation of the phage into daughter cells during cell division (Austin et al. 1981 and Hoess et al. 1985). The second function of Cre in PI phage is to cyclize the linear DNA phage molecule after its injection into bacteria using the loxP sites present in the terminal region of the viral DNA (Hochman et al. 1983). Therefore the original use of Cre was excise tandem repeats of its genome into individual copies for proper replication and segregation of the phage in infected cells. The Cre recombinase acts as a catalyst for the recombination reaction and as with most catalysts, the greater the amount of catalyst, the faster the reaction unless the substrates are limiting the reaction. Therefore since a greater amount of Cre protein would increase the likelihood of interaction with the target sites, devices have been used to increase the efficiency of Cre production in mammalian cells. Translation of the Cre gene has been optimized previously by placing a purine nucleotide at the -3 position of the gene (Fukushige et al. 1992). The presence of the purine nucleotide at the -3 position has previously been shown to optimal for translation efficiency for many genes in eukaryotic cells (Kozak 1986). Also because Cre is required to interact with cellular DNA which located in the nucleus of eukaryotic cells, it was assumed that the addition of a nuclear localization signal may enhance recombination frequencies by allowing Cre to be transported into the nucleus of the cell. The use of a nuclear localization signal has been shown to be of slight benefit to Cre's recombination frequencies in murine ES cells (Gu et al. 1993). These small changes in the Cre gene have been shown to be effective in optimizing the production of Cre protein and its frequency of recombination. In vitro, Cre mediated recombination has been shown to induce integration, inversion or excision depending on the position and orientation of the loxP sites. Inversion occurs when the loxP sites flanking the DNA are in inverse orientations whereas excision is possible by simply flanking DNA with directly repeated target sites. Integration occurs because of Cre's ability to interact with single loxP sites on separate molecules of DNA and because Cre catalyzes a base-pair specific reaction, integration results in the production of two functional loxP sites flanking the integrated sequence in a directly repeated fashion. Therefore because functional, directly repeated loxP sites flank the integrated sequence, this allows for excision of the newly integrated DNA sequence if Cre continues to be present. Cre mediated integration can be accomplished due to Cre's ability to interact with two active sites present on differing plasmid constructs. This type of integration has become useful but because integrated sequences may be excised back out, methods have had to be used to limit the presence of Cre during these reactions. The use of inducible promoters (Golic et al. 1989, Golic, 1991, Chou et al. 1992 and Sauer and Henderson, 1990), expression of the Cre gene from a non-selected plasmid (Dale et al. 1991, O'Gorman et al. 1991, and Gu et al. 1993) and the transient use of purified enzyme (Baubonis and Sauer, 1993) have been used to induce Cre mediated integration without subsequent excision of the integrated DNA. The site-specific integration abilities of Cre have been used in cloning to introduce genes into large viral genomes such as herpes virus (Gage et al. 1992) and baculovirus (Peakman et al. 1992). Targeted integration has also been used in mammalian cells and yeast to direct transgenes into specific chromosomal locations (Fukushige et al. 1992). This type of integration circumvents difficulties with inappropriate and variable expression of transgenes associated with position effects when random integration of exogenous DNA is used. While Cre-mediated integration is an uncommon event, Cre-mediated excision is more frequent. The likelihood of excision is increased because this type of recombination requires loxP sites to be on the same molecule of DNA and not on separate molecules of DNA as for integration. Because excision is more efficient than integration, it is more widely used in experimentation today (Gu et al. 1993, Lakso et al. 1992, Orban et al. 1992, Sauer et al. 1988, Baubonis et al. 1993 and Dale et al. 1991). Excision of a gene of interest in a cell system can be done by inserting directly repeated loxP sites around the gene using conventional homologous recombination techniques. Once the gene is flanked by loxP sites in proper orientation, subsequent expression of the Cre recombinase can be used to delete the loxP flanked sequence. Cre mediated excision has been used to delete the JH-Ej i switch region of an immunoglobulin H gene in ES cells (Gu et al. 1993), as well as delete a number of non-deleterious markers in plant and animal transgenic studies (Lakso et al. 1992, Odell et al. 1990, Dale et al. 1991, Russell et al. 1992, and Bayley et al. 1992). Tissue specific expression of Cre by the lck promoter has lead to tissue specific knockouts of a loxP flanked (3-gal gene in transgenic mice (Orban et al. 1992). Site-specific excision has also been used to promote gene expression. In one example a lens-specific ctA-crystallin promoter was separated from a dormant SV40 T antigen coding sequence by a stop sequence flanked by loxP sites therefore by crossing mice with this transgene with mice expressing the Cre gene, the Cre mediated excision lead to expression of the T antigen only in the lens cells (Lakso et al. 1992). Also Therefore, the Cre-lox system allows excision of loxP flanked sequences specifically in those cells which express the Cre protein. One of the reasons for the interest in the Cre-lox system is because it may alleviate the problems of embryonic lethality in current gene function studies. Previously, studies of gene function in vivo have used gene "knockout" where the gene of interest is genetically mutated or disrupted in embryonic stem cells. These embryonic stem cells are then used to produce transgenic animals but if the "knocked out" gene is involved in embryogenesis, embryo death occurs. Embryo lethality is detrimental where the effect of a gene on a specific system such as the immune system wishes to be studied. This has been shown in the knockout of genes such as Mgat-1 (Metzler et al. 1994). When the Mgat-17Mgat-l~ homozygous mice were produced the embryos did not live beyond day 9.5. This lethality thus inhibited the original goal of studying of the function of complex oligosaccharides on T lymphocytes. This study could be attained by mating transgenic animals expressing the Cre recombinase thymus-specifically with homozygous animals containing the Mgat-1 gene flanked by directly repeated loxP sites. This mating would cause excision of the Mgat-1 gene only in thymocytes thus allowing for the study of the affect of Mgat-1 on the T lymphocytes and circumvent the lethality produced during embryo development. A further application of the Cre-lox system is the engineering of chromosomes using recombination. It is theorized that if the recombinases can act on distantly located target sites then it may be possible to induce chromosomal rearrangements such as large deletions, inversions and translocations by placing single loxP sites on nonhomologous chromosomes and expressing the Cre recombinase. In a recent study, the Cre recombinase has been used to produce a translocation between the non-homologous chromosomes 2 and 13 in transgenic mice (van Deursen et al. 1995). The recombination reaction caused a site-specific translocation between the Dek gene on chromosome 13 and the Can gene on chromosome 2 which is one of the causes of acute myeloid leukemia in humans. This new application of the Cre recombinase may produce chromosomal aberrations in mice which better mimic the causes of some human diseases. RETROVIRAL TRANSFER SYSTEMS Even though the first retrovirus was discovered in the early 1900's, most of the research on retroviruses did not begin until the 1960's when the first mammalian oncogenic retrovirus was found. The earliest known retrovirus was the Rous Sarcoma virus (RSV) discovered in 1911 (Rous, 1911) which was discovered due to its ability to cause tumors in chickens. It was not until 1978 that the src gene was identified and found to be responsible for the tumorigenesis caused by RSV. The presence of the non-viral src gene within the retroviral genome was the first indication that retroviruses have the ability to transfer non-viral genetic material between cells. Retroviruses, by definition, contain a single + strand of RNA within an enveloped nucleocapsid. Once the retrovirus has entered a cell, it proceeds to produce a DNA form of its genome, called a provirus, using the reverse transcriptase contained within its nucleocapsid. This provirus then makes its way to the nucleus where it integrates into the host genome. While in the genome, the virus genome is transcribed and translated efficiently using host mechanisms and + stranded RNA is produced by transcription of the virus genome. Once produced, the + stranded RNA can be packaged into newly produced nucleocapsids and released from the cell as complete retroviral particles. In order to produce a functional retrovirus, there are three structural genes within a replication-competent retrovirus which must be transcribed and translated by host mechanisms. The first gene is the gag or glycosaminoglycan gene, which produces the virion core protein; the second is the pol gene, which encodes for the reverse transcriptase, protease and integrase; and the third is the env gene which encodes for the major envelope glycoprotein required by the retrovirus. Each of these three genes is normally carried in the retroviral genome but it has been shown that they are not required to be contained within the viral genome. New retroviral particles can be produced provided that these structural proteins are expressed within the cell by some other means. Because these gene sequences are not required to be within the retroviral genome, these three structural elements are classified as trans -acting factors in the production of complete retroviral particles. In addition to the necessity for the three structural proteins, the two Long Terminal Repeats (LTR) and the packaging signal are also essential for proper and efficient production of virus. The two LTRs flank the entire retroviral genome and contain viral promoters and enhancers which are required for the production of the RNA version of the virus. The packaging signal is a nucleotide sequence between the two LTRs which is required for efficient packaging of the RNA virus genome into the nucleocapsid (Mann and Baltimore, 1985). Since retroviral production is adversely affected without these elements in the viral genome, these elements are said to participate in a cis -acting manner. Production of retroviruses has long been possible by utilizing the cis- and trans-acting properties of the retroviral genes. The trans- acting properties of the structural genes have been incorporated to produce retroviral packaging cell lines while the cis- acting properties have been used to create retroviral constructs. The retroviral packaging lines are cell lines which express the three retroviral structural genes. The retroviral constructs are plasmids containing LTRs flanking a packaging signal and a multiple cloning area downstream of the 5' LTR promoter. By cloning genes of interest into one of the cloning sites, the inserted gene can be packaged into and expressed by the retrovirus. These materials have allowed for the production of retroviruses simply by transfecting the retroviral constructs into the retroviral packaging lines. Once the retroviral construct is in the retroviral packaging line, the structural gene products within the cells can be used to package the full-length transcripts from the 5' LTR of the retroviral construct into new retroviral particles. One of the major considerations when looking at a retroviral packaging cell line is the host range of the virus. Host range is defined as the type and species of cells which are able to be infected by and express a given retrovirus. The host range of a retrovirus is determined by the env gene of the retrovirus which encodes for the viral envelope protein. In the study of mouse retroviruses, there are three different types of host ranges. The first major type of retrovirus is the ecotropic type (Levy, 1974) which are able to replicate only in the homologous mouse or rat cells. Secondly, there is the xenotropic type (Levy, 1973) which originate in mouse but are able to infect only heterologous species. Lastly, there is the amphotropic type (Hartley and Rowe, 1976) which has characteristics of both in that these viruses are able to infect mouse, rat and many other cell types including human, rabbit, feline, and canine. This ability to change host ranges using different env genes has been used effectively in the production of differing retroviral packaging cell lines. In a normal retrovirus, expression and replication of the retroviral genome are regulated only by the promoter and enhancer in the 5' LTR. In a retroviral construct, it is, at times, desirable to express more than one gene and studies have shown that there are several different ways to accomplish this. The two major means of expressing more than one gene from a retroviral genome are through the use of internal promoters or alternative splicing. Internal promoters have been used extensively in retroviral constructs to express multiple genes and allow for promoter/enhancer combinations other than the viral LTR to drive gene expression (Overell et al. 1988). Alternative splicing has also been used in retroviral vectors (Cepko et al. 1984) but problems may arise due to the foreign sequences which are inserted into the intron of the spliced gene and can disrupt production of the second gene product (Korman et al. 1987). Because retroviral replication requires an RNA intermediate, either strategy for multiple gene expression must allow for the production of a full length RNA transcript of the retroviral genome. Therefore, multiple gene expression is possible in retroviral constructs but consideration must be taken to avoid insertion of a polyadenylation signal which will prevent production of the full length transcript required for replication. A retroviral transfer system has been chosen for this study because they have been used for many years as an effective way of transferring DNA into mammalian cells. Retroviral transfer systems allow for efficient transfer and expression of genetic material in cells but are not detrimental to the infected cells in the process. There are several other reasons for using retroviral transfer systems such as the ability to infect a wide range of cell types from different animal species, the precise integration of genetic material carried by the vector into recipient cells, and the lack of vector spread. The main reason this system was chosen over other methods of DNA transfection was because of the lack of toxicity of these viruses to infected cells and the efficient expression of the inserted genes from viral promoters. THESIS OBJECTIVES Standard methods of gene knockout in transgenic studies presently are unable to deal with the problem of lethality caused by the ablation of genes involved in embryogenesis. The Cre-lox recombination system allows a means of circumventing lethality by causing specific excision of loxP -flanked genes in cells where the Cre recombinase is expressed. By expressing the Cre protein only in those cells containing loxP -flanked genes, these genes can be specifically excised and the cells can be studied in vitro or transplanted into recipients for study in vivo. Therefore, a system by which the Cre gene can be transferred and expressed in specific cells would be useful to molecular biologists. With the use of a retroviral transfer system, a tool has been produced which will allow the transfer and expression of the Cre gene only in cells which are subjected to infection by the retrovirus. This tool can be used in conjunction with transgenic animals to trigger recombination in isolated cells for specific gene ablation. This specific gene ablation will allow for the study of genes which previously were unable to be studied because of their involvement in embryo development. The overall objective of the work described in this thesis is to produce retroviral particles capable of efficiently transferring and production of Cre recombinase protein in mammalian systems. 11 CHAPTER n MATERIALS AND METHODS DNA isolation and purification Isolation of high molecular weight genomic DNA: To isolate and purify genomic DNA from NIH3T3 murine fibroblast cells, approximately 5 x 10^  cells were released from a 10 cm tissue culture plate using 2 ml of 0.25% Trypsin/0.02% EDTA. The Trypsin-EDTA solution containing the cells was centrifuged at 1500 rpm for 15 mins to pellet the cells and 300 uL of 50 mM Tris pH 8.0/50 mM EDTA and 50 uL of 10 mg/ml proteinase K were added to the pellet. This mixture was incubated at 55°C for approximately 18 hrs following which 300 uL of 20 mM Tris pH 8.0/1 mM EDTA was added. In order to isolate the DNA, the genomic DNA was extracted 2 x with 100 mM Tris pH 8.0 equilibrated phenol/chloroform (Ultra pure) and 1 x with chloroform then precipitated with 95% EtOH and centrifuged at 14000 rpm for 10 mins at 4°C. The genomic DNA samples were resuspended in a 20 mM Tris pH 8.0/1 mM EDTA solution and quantitated using a DNA fluorometer. All genomic DNA samples were stored at 4°C. Isolation of plasmid DNA: Plasmid isolation was carried out from small volume cultures using the TELT method. Five ml of 2 x YT media containing 50 |iL of ampicillin (100 (ig/ml) was inoculated and cultured overnight with competent DH5ct E.coli tranfected with a plasmid. The next day, the bacteria was pelleted and resuspended in 400 \iL of TELT buffer (2.5 M Lithium chloride, 50 mM Tris pH 8.0, 4% Triton X-100, 62.5 mM EDTA). The plasmid DNA was extracted with 1 x 100 mM Tris pH 8.0 equilibrated phenol/chloroform and 1 x chloroform. 95% EtOH was used to precipitate the plasmid DNA which was then pelleted by centrifugation at 14000 rpms for 10 mins at 4°C. Plasmid DNAs were resuspended in 20 mM Tris pH 8.0/1 mM EDTA and stored at -20°C. Large scale preparations of plasmid DNA were performed according to the methods in Maniatis et al (1989). One liter bacterial cultures were pelleted by centrifugation at 5000 rpm for 10 min. The pellets were collected and resuspended in 10 ml of lysis buffer (25 mM Tris pH 8.0, 10 mM EDTA, and 50 mM glucose), 20 ml of solution II (0.2 M NaOH and 1% SDS), and 15 ml of potassium acetate. The samples were centrifuged at 8000 rpm for 20 mins at 10°C. The supernatants were filtered using cheesecloth after which isopropanol was added. After standing at room temperature for 15 mins, the solution was spun at 10000 rpm for 15 mins at 10°C and the DNA pellet was air dried. A CsCl2 gradient was used to separate the closed circular DNA from the nicked and linear DNA. After separation, the DNA was butanol extracted to removed the EtBr used in the CsCl2 gradient then precipitated with 95% EtOH and resuspended in 20 mM Tris pH 8.0/1 mM EDTA. All the plasmids were stored at -20°C. Cloning All of the digestions with restriction enzymes involved the use of New England Biolabs (NEB) restriction enzymes and buffers according to the manufacturer's specifications. Ligations of DNA vectors and inserts all involved the use of NEB T4 Ligase and buffer according to the manufacturer's instructions. Any DNA fragments or vectors used in a ligation procedure would be digested with restriction enzyme as required then subjected to agarose slab gel electrophoresis. The electrophoretic buffer used was 1 x TAE (40 mM Tris-acetate 1 mM EDTA). After sufficient electrophoresis of DNA samples, gels were stained in an Ethidium bromide solution (10 mg/ml in dH20), photographed and the required band of DNA would be cut out from the gel. A kit (Gene Clean) was used according to the manufacturer's specifications to isolate the DNA from the agarose. Small aliquots of the isolated fragments were subjected to agarose slab gel electrophoresis, stained with EtBr and photographed in order to determine the purity and relative quantity of the DNA fragment. Cloning of retroviral vectors: LXSN and LXSHD (Miller and Rosman, 1989) viral vectors (Figure 1) were obtained from A. D. Miller (Fred Hutchison Cancer Center, Seattle, Washington). Both LXSN and LXSHD were digested with the restriction enzyme and isolated as previously described. The Cre plasmids (Figure 2), pBS185 (Sauer and Henderson, 1990) from Brian Sauer (DuPont'-Merck Pharmaceutical Co., Wilmington, Delaware) and pMCcre (Gu et al. 1993) from Klaus Rajewsky (Institute for Genetics, University of Cologne, Cologne, Germany), were used as the sources of Cre recombinase sequences. Both Cre gene sequence fragments were digested, isolated and ligated as described previously. Each of the Cre fragments was ligated separately into each of the retroviral vectors. Ligations products were transformed into competent DH5ot E.coli bacteria and plated onto 2 x YT media containing ampicillin which resulted in only those bacteria containing plasmid DNA to produce colonies. Several random colonies were picked and the DNA recovered from the small culture plasmid isolation was used in restriction enzyme analyses to determine the presence and orientation of a DNA insert. Digestion's with restriction enzymes were performed and the products were visualized, as previously described, using agarose gel electrophoresis. One of the clones identified as containing the retroviral vector with a complete Cre gene insert in a sense orientation was grown for large scale plasmid DNA preparation as described. 14 A: LXSN ne__, Hpal(1650) Xhol(1655) EcoRI ( 1 6 4 4 ) ^ / ^ B a m H I (1660) LTR V ^ a g SV40 NEO LTR B: LXSHD Hpal(1650) Xhol(1655) EcoRI ( 1 6 4 4 ) ^ ^ ^ ^ B a m H I (1660) LTR y-tgag SV40 HIS-D LTR Figure 1. Schematic representations of the L X S N and L X S H D retroviral constructs. Panel A and B: LTR represent M-MSV L T R s , y+gag indicates the retroviral extended packaging signal, S V 4 0 represents the S V 4 0 promoter and the thin black line represents pBR322 plasmid sequence. The restriction enzyme sites of the multiple cloning site are indicated. Panel A: A schematic representation of the L X S N retroviral construct. N E O represents the neomycin transferase gene. The total size of the L X S N vector is 5874 bp. Panel B: A s c h e m a t i c of the L X S H D retroviral construct . H IS-D represents the histidinol dehydrogenase gene. The total size of the L X S H D vector is 6374 bp. 15 A: pBS185 Xhol(1721) s t o p ( 2 8 1 3 ) „ , , A T G ( 1 7 8 7 ) \ 'Mlul(2826) 1 3 3 , M I CRE TGTGAAATGTCC +1 pA B: pMCCre M l u l ( 6 9 2 ) A T G ( 8 1 3 ) * Mlul(1742) stop(1729) \ TCG ACC A T G CCCAAGAAGAAGAGGAAGGTG TCC +1 Figure 2. A schematic representation of the plasmids containing the Cre recombinase genes. The enzyme restriction sites indicated were used to release the Cre gene sequence from the plasmid. The positions of the A T G codons are indicated and the A T G codon is enlarged in the expanded sequence. Panel A and Panel B: C R E in each panel represents the Cre gene, the nucleotide which is bolded in the -3 position of the expanded sequence represents the nucleotide mutated to optimize translational efficiencies, pA represents the polyadenylation signal and the thin line represents the p U C based vector backbone. Panel A: A schematic representation of the pBS185 plasmid. C M V represents the C M V promoter. T h e total vector size is approximately 4.3Kb. Panel B: A schematic representation of the pMCCre plasmid. MT represents the metallothionein promoter and the italicized nucleotides represent the sequence added to produce the Nuclear localization signal. The total vector size is approximately 7.1 Kb. Cloning of target vector: A Clal-PvuII restriction enzyme fragment of pSP72 from Janet Rossant (University of Toronto, Toronto, Ontario) containing a complete murine PGK promoter, hygromycin B resistance marker gene, and a polyadenylation signal was blunted and ligated into BamHI site of plox2 (derived from pBS64, Sauer and Henderson, 1988). The resulting vector contained the PGK promoter, hygromycin B resistance marker, and polyadenylation signal flanked by loxP sites and was named plox^Hyg. The vector was analyzed by restriction enzyme analysis and agarose slab gel electrophoresis as previously described. The pFlox vector was produced and donated by Daniel Chui (Biomedical Research Center, Vancouver, British Columbia). 17 A: Flox - 4 4 -H S V T K PGK Neomycin pA B: lox2Hyg PGK Hygromycin pA - 4 z £ * t \ X ^ S S 3 4 -ATMCTTOGTATAGCATACATTATACGAAGTTAT loxP site Figure 3. Schematic representation of the pFlox and plox^Hyg target vectors. The arrowheads indicate the loxP sites and the thin lines represent p U C based vector sequence. Panel A: A schematic representation of the pFlox target vector. The H S V T K cassette is indicated by the hollow arrow and the neomycin resistance cassette is indicated by the black arrow. The total pFlox vector size is 6.4 Kb. Panel B: A schematic representation of the plox^Hyg target vector. The striped arrow represents a P G K promoter, the hollow box represents the hygromycin b resistance marker and the striped box represents the polyadenylation signal. The DNA sequence shown represents the sequence of the 34 bp loxP sites used in both target vectors. The total p lox 2 Hyg vector size is 5.2 Kb. Cell Lines Tissue culture: NIH3T3 cells (Jainhill et al. 1969), PA317 cells (Miller and Buttimore, 1986)and \|/-2 cells (Mann et al. 1983) were all perpetuated in Dulbecco's modified Eagle's medium (DMEM) high glucose media containing 10% fetal calf serum (FCS), L-glutamine and P-mercaptoethanol. Transduced cells were selected according to the particular vector construct as follows: neomycin phosphotransferase (Palmer et al. 1987) was selected with Genticin (Gibco BRL) at 500 |ig/ml final active concentration, histidinol dehydrogenase (Danos and Mulligan, 1988) was selected for using L-histidinol (Sigma) at 500 \ig/m\ final concentration and the hygromycin resistance marker was selected for using hygromycin B (Gibco BRL) at 500 ug/ml final concentration. Each selection was carried out for a minimum of 4 days to ensure complete selection against those cells not carrying the marker of interest. Additionally, PA317 cells were subjected once a month to selection using HAT (30|iM hypoxanthine, l(iM aminopterin, 20|LLM thymidine) media for 5 days and HT (30|tM hypoxanthine, 20fiM thymidine) media for 4 days in order to select for the presence of the retroviral packaging construct contained within the cell line. A 2 ml solution of 0.25% trypsin and 0.02% EDTA was used to release the NIH3T3, PA317 or \\f-2 cells from the tissue culture plates during passaging. Portions of the trypsin-EDTA solution containing cells were used to seed new tissue culture plates. All the cell lines were incubated at 37°C and 5.5% C O 2 in air. All media and selection stocks (L-histidinol, G418, hygromycin B) were stored at 4°C. Routinely, the cell lines were passaged 1 in 10 every 4 days except were noted. Production of amphotropic retroviral producing cell lines: Viral producer cell lines were generated using standard methods (Miller and Rosman, 1989). 10 \ig of each retroviral vector DNA was digested with restriction enzyme XmnI, a unique restriction site in the pBR322 backbone of the retroviral plasmid, to linearize the plasmid for transfection into 2 x 10^ -3 x 10^  \j/-2 cells using the calcium phosphate precipitation procedure (Gorman et al. 1983). After transfection and 24 hours of culturing, the media from the transfected \|/-2 cells was centrifuged at 1500 rpm for 5 min, aliquoted and stored at -70°C or used directly to infect 2 x 10^ -3 x 10^  PA317 cells in exponential growth phase. Prior to adding \|/-2 media for infection, polybrene was added to the PA317 media at a final concentration of 4 ug/ml. 24 hrs after inoculation with retrovirus, the PA317 cells were split, replated at differing dilutions and the transduced cells were selected in L-histidinol or G418. After sufficient culturing, cloning rings were used to isolate individual PA317 colonies and these cells were replated for further expansion. Titering of retrovirus production from retroviral producing cell lines: 0.25 ml to 1.0 ml of retrovirus containing media from the PA317 retroviral producing clones would be used to replaced the normal culture media on a 10 cm plate of 2 x 10^ -3 x 10^  exponentially growing NIH3T3 cells. The cells and retrovirus would be incubated for 24 hours then the NIH3T3 cells would be released from the plates and plated onto new plates using serial dilutions of 1/100 to 1/10,000. The newly infected NIH3T3 cells would be incubated with new media and placed under selection for the corresponding retroviral selection marker. The cells would be incubated for sufficient time to allow individual colonies to be counted. The colonies would be individually counted and the dilution factors would be used to calculate the corresponding retroviral titers. The titering experiments were completed at least three times for each retrovirus and the multiplicity of infection (MOI) was kept below 0.3 retroviral particles per cell to inhibit multiple infections. Production of stable target cell lines: 10 (Xg of plox^Hyg was digested with the restriction enzyme Sspl and pFlox with XmnI, as previously described, in order to linearize the vectors in preparation for electroporation into NIH3T3 cells. Electroporation was accomplished using standard procedures (Chu et al. 1987).The target vectors were placed into the NIH3T3 cells by electroporating 7 x 10^  cells in 0.8 mL with 10 u,g of DNA using a single pulse of 240 V at 960 |±F from a Bio-Rad Gene Pulser. After electroporation, the cells were plated in 10 cm dishes with 10 mL of media and the corresponding selection. The cell numbers were subsequently quantitated and efficiency of plasmid transfer was calculated to be approximately 1 in 4000. The transduced cell lines were maintained under selection and individual clones were isolated for expansion. Four clones were selected to analyze for the presence of single vectors. Of the 4 selected clones only one was used for infection and analysis during the southern blot studies. The presence of the complete target vector in the NIH3T3 cells was assessed using digestion with restriction enzymes and Southern blot analysis. Southern Blot Analysis Genomic DNA was isolated, digested with restriction enzymes and separated by agarose gel electrophoresis as described above. Southern blot analyses were carried out using standard methods (Southern, 1975). After the gel had been stained with EtBr and photographed, the gel was soaked in denaturation solution (0.5 M NaOH/1.5 M NaCl) 2 x for 15 mins then in neutralization solution (0.5 M Tris pH 7.0/1.5 M NaCl) 2 x for 15 mins. DNA was blotted to nitrocellulose membrane using 20 x SSC (1 x SSC: 0.15 M NaCl, 0.015 M sodium citrate [pH 7.0]) and then the membrane baked in an oven at 80°C for 1 hour. The dried membranes were placed in prehybridization solution (33% deionized formamide, 40 mM NaP04, 4 x SSC, 4 x Denhardts solution [50 x: 5 g ficoll, 5 g polyvinylpyrrolidine, 5 g BSA pentax fraction V to a final volume of 500 ml], 0.8% glycine, 125 [ig/ml herring sperm DNA) and rotated at 42°C for 1 hr. Hybridization to radiolabeled probes (see below) was carried out at 42°C in hybridization solution (33 % deionized formamide, 25 mM NaP04,4 x SSC, 10 %.dextran sulphate, 100 ftg/ml herring sperm DNA) for 18 hrs. After hybridization, the membranes were washed 2 x in 0.1 % sodium dodecyl sulphate (SDS)/0.1 x SSC at 55-60°C, air dried then placed on Kodak X-ray film for autoradiography at -70°C for 1-14 days depending on signal intensity. Both the prehybridization and hybridization procedures were carried out in rotating tubes in a hybridization oven (InterScience) at 42°C. Preparation of radiolabeled probe: All sequence to be radioactively labeled was isolated in the same manner as fragments isolated for cloning described previously. 50 ng of DNA suspended in 20 mM Tris/lmM EDTA was placed in boiling water for 10 min then immediately placed on ice. When sufficiently cooled, 18 ul of oligo mix (400 mM Hepes pH 6.6, 100 mM Tris-HCl pH 8.0, 10 mM MgCl2, 40 uM dCTP, 40 uM dGTP, 40 uM dTTP, 0.25 mg/ml random hexamers [Pharmacia]) was added along with 10 units of klenow (NEB) and 50 uCuries of 32padATP. Labeling was allowed to proceed at room temperature for 1 hr and a Sephadex G50 column was used to recollect the labeled DNA sequence. Only labeled probe with a specific activity of 5 x 10^ -1 x 10^  counts per minute(cpm)/u\g of DNA, as analyzed using a scintillation counter, was used. 1 x 10^  cpm of labeled probe was added per ml of hybridization solution during a hybridization procedure. 22 CHAPTER HI RESULTS Cloning of Retroviral Vectors: In each of the LXSN and LXSHD vectors, there is a multiple cloning site into which genes of interest may be placed and expressed by the 5' LTR promoter/enhancer. Each multiple cloning sequence contains unique restriction enzyme sites for EcoRI, Hpal, Xhol and BamHI. Both LXSN and LXSHD were restriction enzyme digested with EcoRI and the Cre genes were cloned into this site because of its proximal position to the LTR. A 1.1 Kb Mlul-Mlul fragment from pMCCre containing the Cre gene sequence, an optimizing translation initiation site and a nuclear localization signal was isolated for cloning. Also, a 1.1 Kb Xhol-Mlul fragment containing the Cre gene sequence and an optimizing translation initiation site was isolated from pBS185. Both Cre gene sequence fragments were digested with restriction enzymes, isolated and ligated as described previously. The resulting retroviral vectors (Figure 3) were analyzed using digestion with restriction enzymes and named according to which retroviral vector and Cre gene insert were present in the final product (Table 1). 23 A: Lcre-KSN (EcoRI/Xhol) (1644) (Mlul)/EcoRI (2749) LTR V+gag c re -K SV40 Neo LTR B: Lcre-KSHD (EcoRI/Xhol) ( 2630) (Mlul)/EcoRI (2749) CMIul)/l —4 W W ^ V ^ I M ^ ^ J -LTR V+Sag c r e - K SV40 HIS-D LTR C: LNKcreSN EcoRI/(Mlul) ( 1644) (Mlul)/EcoRI (2799) •\wwiiiiiiiiiiiiiiiiiiiiiiiiin^"^^« LTR V+gag NKcre SV40 Neo LTR D: LNKcreSHD EcoRI/(Mlul) ( 1644) (Mlul)/EcoRI (2799) ^  6 4) |Mhjl)/l LTR V+gag NKcre SV40 HIS-D LTR Figure 3. Schematic representations of the Cre-containing retroviruses. The area are marked as indicated. LTR represent the M-MSV LTRs, v+gag represents the extended packaging signal, SV40 represents the SV40 promoter and the thin black line represents pBR322 plasmid DNA. Cre-K represents the Cre gene containing an optimized translation initiation site and NKcre represents the Cre gene containing an optimized translation initiation site and nuclear local ization s ignal . The neomycin phosphotransferase gene is represented by Neo and the histidinol dehydrogenase gene is represented HIS-D. The restriction enzyme sites indicated were used in the cloning of the vectors. Panel A: A schematic representation of the Lcre-KSN retroviral vector. Total size of the vector is 6979 bp. Panel B: A schematic representation of the Lcre-KSHD retroviral vector. Total size of the vector is 7479 bp. Panel C: A schematic representation of the LNKcreSN retroviral vector. Total size of the vector is 7029 bp. Panel D: A schematic representation of the LNKcreSHD retroviral vector. Total size of the vector is 7529 bp. 24 Table 1. The Names of the Newly Produced Cre-containing Retroviruses. Retroviral Vector LXSN LXSHD LXSN LXSHD Origin of Cre gene insert p B S 1 8 5 (B. Sauer) p B S 1 8 5 p M C C r e (K. Rajewsky) pMCCre Resulting Product Name L c r e - K S N L c r e - K S H D LNKcreSN LNKcreSHD A representation of how the retroviruses were named according to the Cre gene which was inserted and the retroviral vector used. The retroviral vectors which have been used in this study were supplied by A. D. Miller and have been described. Both the LXSN (Miller et al. 1989) and LXSHD (Stockshlaeder et al. 1991) vectors contain the 5' LTR from Moloney murine sarcoma virus (M-MSV) and the 3' LTR from Moloney murine leukemia virus (M-MLV). The M-MSV LTR was used as the 5' LTR because the M-MLV LTR could not be separated from the gag start codon. The normal gag start codon has been shown to have a severe deleterious effects on the expression of downstream cDNA inserts when it is not removed from between the LTR and inserted gene (Miller et al. 1989). These retroviral vectors also contain a PvuII to Hindlll fragment from SV40 containing the early promoter (Fiers et al. 1978) and the extended packaging signal from M-MSV. Splice donor and acceptor sites are present between the 5' LTR and the cDNA insert in order to splice out the translational start codons present in the extended packaging signal. The only sequence difference between LXSN and LXSHD is that LXSN contains a neomycin phosphotransferase gene whereas LXSHD contains histidinol dehydrogenase gene. Both of the selection marker are 3' of the SV40 sequence. Expression of the genes in the retroviral constructs come from the presence of two transcriptional promoters. The inserted Cre recombinase cDNA is transcribed by the 5' LTR promoter. The transcript initiated at the 5' LTR extends to the polyadenylation signal present in the 3' LTR. The produced mRNA undergoes splicing to excise the 9 translational start codons in the extended packaging signal. After RNA splicing, translation of the Cre protein should initiate at Cre's AUG codon. To increase the efficiency of translational initiation, the Cre sequences we have incorporated contain a purine at the -3 position which has been shown to have a dominant effect on translational initiation. The mRNA which extends from the 5' LTR to the 3' LTR can also be used for translation of the downstream resistance marker gene. In order to increase production of the resistance marker genes, the SV40 early promoter has been placed to the 5' side of the marker genes therefore giving the marker genes their own strong transcription promoters. The mRNA transcript from the SV40 promoter concludes at the 3' LTR and does not undergo any further splicing. With the addition of the internal promoter, the strong LTR promoter controls the expression of the Cre gene while both the LTR promoter and the internal SV40 control the resistance marker gene expression. By separating the trans - and cis - acting components of a normal retrovirus, retroviral vectors and retroviral packaging cell lines have been produced which allow for the production of retroviruses which will carry non-viral genes. Replication-competent retroviruses produce a DNA copy of themselves before integration into the cellular genome. Once integrated the retrovirus uses the cellular transcription and translation machinery to produce the viral proteins required for the production of new viral particles. Full-length transcripts of the retroviral genome, from the 5' LTR promoter to the polyadenylation signal in the 3' LTR, are produced by the host machinery. These transcripts contain packing signals which direct it into the retroviral head to produce new virion. By transfecting DNA vectors which contain an extended packaging signal and retroviral LTRs flanking inserted gene sequences but no other viral sequences into cells which already produce the retroviral proteins, the full-length transcripts which are produced can be packaged into retroviral heads and produce infectious viral particles. When these retroviruses are used to infect cells which do not produce the viral proteins then retroviral genome is integrated then transcribed and translated from the 5' LTR and full-length transcripts are produced. Because these mutated retroviral genomes do not contain viral protein sequences, no new retroviral particles can be produced therefore the nonviral genes which were inserted between the LTR are transferred from one cell to another in a retroviral particle and the LTR promoters allow expression of the genes using host mechanisms. Transfer of Retroviral Genome Into Infected Cells: Retroviral producer lines for the following series of vectors, Lcre-KSN, Lcre-KSHD, LNKcreSN and LNKcreSHD, were generated. Lines were tested for virus generation by infecting NIH3T3 cells with supernatant harvested from confluent producer cell lines and subsequent isolation of genomic DNA from the resistance-selected polyclonal infected cells for analysis by Southern blotting. All of the genomic DNA samples were digested with EcoRV and probed with 32p labeled Cre gene sequence. EcoRV restriction enzyme was used in the analysis because of the presence of restriction enzyme sites in each of the two LTRs and one within the Cre gene sequence. The use of digestion by EcoRV would yield two predictably sized bands for each retrovirus (Table 2). 28 Table 2. Expected Band Sizes for Genomic D N A of Retrovirally Infected Cel ls Digested With E c o R V and Probed with a 3 2 P Labeled Cre Gene . V i r u s Expected Band Sizes (Kb) Lc reKSN 1.85, 1.91 Lc reKSHD 1.85, 2.41 LNKcreSN 1.90, 1.91 LNKcreSHD 1.90, 2.41 The genomic DNA Southern blots of the Lcre-KSN and Lcre-KSHD infected cells produced bands as shown in Figure 4. The cells harboring the Lcre-KSN retrovirus produced bands of 1.90 Kb and 1.85 Kb while the cells carrying the Lcre-KSHD retrovirus produced bands Of 1.85 Kb and 2.40 Kb. Genomic DNA Southern blot analysis of the LNKcreSN and LNKcreSHD infected cells produced autoradiographs indicated in Figure 5. A single more intense band of 1.90 Kb was produced from LNKcreSN infected cells while bands of 1.90 Kb and 2.40 Kb were produced by those cells containing the LNKcreSHD retrovirus. Since each of the retrovirus containing cells produced the correct band sizes upon Southern blot analyses, this indicates that the retroviral particles are able to transfer complete retroviral genomes into the infected cells. 30 A L c r e - K S N EcoRV (497) ' EcoRV (573) EcoRV(2425) I 1 EcoRV(4335) L c r e - K S H D EcoRV (702) I EcoRV (778) . \/ EcoRV(2630) I 1 EcoRV (5040) EcoRV (5115) £ Figure 4. Determination of genome transfer for the L c r e - K S N and L c r e - K S H D retroviruses. Panel A: A schematic representation of the L c r e - K S N and L c r e - K S H D retroviral vectors. The hollow boxes indicate the L T R s , the black arrow represents the S V 4 0 promoter, the striped boxes represent the Cre gene, the spotted black box represents the Neomycin phosphotransferase gene, the spotted white box represents the histidinol dehydrogenase gene and the thick black line represents the retroviral packaging signal. The E c o R V sites and positions are shown on the diagrams. The area of hybridization of the probe is indicated by H . Panel B and Panel C: Southern analysis of E c o R V digested genomic DNA of retroviral infected cells. 5 pg of digested genomic DNA was loaded per lane. The membrane was probed with a 3 2 P labeled Cre gene. The molecular weight sizes are indicated at the side of the blot. Lane 1: Molecular Weight Markers. Panel B: Southern blot analysis of cells infected or producing Lcre-K S N clone #21 retrovirus. Lane 2: PA317 producer clone #21 of Lcre -KSN. Lane 3: lox 2 Hyg NIH3T3 clone #1 cells infected with Lcre-KSN clone #21 retrovirus. The blot was washed at 5 8 ° C and exposed to film for 4 days. Panel C: Southern blot analysis of cells infected with Lcre -KSHD clone #6 retrovirus. Lane 2: NIH3T3 cells infected with Lcre-KSHD clone #6 retrovirus. The blot was washed at 5 5 ° C and exposed to film for 1 day. blot 31 A L N K c r e S N EcoRV (497) 1 EcoRV (573) EcoRV(2425) EcoRV(4335) ^coRV(4410) DRV k LNKcreSHD EcoRV (702) lEcoRV (778) c c o F EcoRV(2630) EcoRV (5040) f E^oRV(5115) Figure 5. Determination of genome transfer for the LNKcreSN and LNKcreSHD retroviruses. Panel A: A schematic representation of the LNKcreSN and LNKcreSHD retroviral vectors. The hollow boxes indicate the LTRs, the black arrow represents the SV40 promoter, the striped boxes represent the Cre gene, the spotted black box represents the Neomycin phosphotransferase gene, the spotted white box represents the histidinol dehydrogenase gene and the thick black line represents the retroviral packaging signal. The EcoRV sites and positions are shown on the diagrams. The area of hybridization of the probe is indicated by H Panel B: A Southern blot analysis of EcoRV digested genomic DNA of retroviral infected cells. 5 ug of digested genomic DNA was loaded per lane. Lane 1 and Lane 5: Molecular Weight Markers. Lane 2: lox2Hyg NIH3T3 clone #1 cells infected with LNKcreSN polyclonal retrovirus. Lane 3: lox2Hyg NIH3T3 clone #1 cells infected with LNKcreSHD polyclonal retrovirus. Lane 4: PA317 cells producing LNKcreSHD polyclonal retrovirus. The membrane was probed with a 32p labeled Cre gene. The molecular weight sizes are indicated at the side of the autoradiograph. The membrane was washed at 55°C and exposed to film for 14 days. Viral Titers: Titering was accomplished as described. Individual clones of Lcre-KSN and Lcre-KSHD viral producers and polyclonal LNKcreSN and LNKcreSHD viral producers were screened to identify those yielding the highest titers. Ten selection-resistant clones were chosen to be tested for each of the Lcre-KSN and Lcre-KSHD retroviral producer cells lines. From the retroviral producing cell lines produced, the titers ranged from 0 to 1.1 x 10 6 cfu per ml. The best retroviral titers which were achieved from the polyclonal or clonal retrovirus producing lines are indicated in Table 3. One of the main considerations during the production of a retrovirus is to attain the highest possible viral titers. One of the ways this is achieved is to "ping-pong" the retrovirus. This means that the original retroviral construct is transfected into one retroviral packaging cell line and the produced retroviral particles are used to infect a second packaging cell line which then produces the final retroviral particles. "Ping-ponging" is used because it has been found that retroviral titers produced from infected cells containing a single copy integrated provirus are as high if not higher than transfected cells containing multiple copies of retroviral vector. Attempts at using the origin of replication and T antigen coding regions from polyoma virus to generate multiple copies of the viral vector in transfected cells to increase viral titers have shown that the viral titers produced are no higher than retrovirus packaging cells containing single integrated proviruses (Korman et al. 1987). Also while transfection results in insertion of multiple copies of vector into recipient cells and to the production of rearranged virus, infection of virus results in single copy integrated virus which can be screened for structure and generally yields unrearranged virus (Bender et al. 1988). 33 Table 3. Retroviral Titers Produced by Clonal PA317 Retrovirus Producing Cel ls . Retroviral Vector Viral Titer ( x 1 0 5 cfu/ml) LXSN 6.5 L c r e - K S N clone #21 6.1 L c r e - K S N clone #22 8.9 L c r e - K S N clone #23 11.0 L N K c r e S N polyclonal 1.0 LXSHD 8.9 L c r e - K S H D clone #6 0.4 L c r e - K S H D clone #10 0.3 L N K c r e S H D polyclonal 2.0 All of the retroviral titers were determined as previously d e s c r i b e d . T h e final retroviral producing cells were all PA317 cells which produce amphotropic retroviral par t ic les . Copy Number of Target Vectors in the lox^Hyg NIH3T3 Cell Line: Once electroporation and initial selection with hygromycin b was completed as previously described, the newly produced lox^Hyg NIH3T3 cell line was split and individual clones were propagated. Southern blot analyses of genomic DNA from the lox^Hyg NIH3T3 clones were used to determine the number of target vectors within the clonal cell lines. It was important to determine copy number in order to minimize possible ambiguities which may arise in the Southern blot analyses of retrovirally infected target cells. It was assumed that by using restriction enzymes which cut singly within the linear, integrated plox^Hyg vector, a single band of unique size would be produced for each integrant when probed with a 32p labeled sequence which is specific to one side of the plox^Hyg vector and which does not overlap the restriction enzyme site. Figure 6 shows the Southern blot analysis results from the cloned lox^Hyg cell line genomic DNAs when digested with Smal or Spel restriction enzymes and probed with 32p labeled sequence. A single band of 7.0 Kb or 6.9 Kb was detected for clone #1 and clone #4 respectively. This single band production indicates the likelihood of single vector integration. In order to confirm this for clone-#l, Southern blot analysis using digestion with Smal was used. This analysis produced a single band of 4.1 Kb therefore confirming that clone #1 contains a single vector integration. Clone #1 was then used for all further analyses regarding infection with retrovirus and Southern blot analysis. k Spel(1144) probe 1 ^ ^ p r o b e 2 B Smal ( 3264) I Figure 6 . Southern Blot analyses of lox2Hyg NIH3T3 genomic DNA from isolated clones. Panel A: A schematic representation of the linearized lox2Hyg vector in the stable lox2Hyg NIH 3T3 cell line with the Spel and Smal restriction enzyme sites shown. The black arrowheads represent the loxP sites, the hollow arrow represents the PGK promoter, the hollow box represents the hygromycin b resistance marker gene, the striped box represents the polyadenylation signal, the thin black line represents the pUC based vector sequence and the thick black line represent genomic DNA. The line below 'probe 1' indicates the area of hybridization by a 3 2 P labeled Sphl-Hindlll fragment of plox2Hyg and the line below 'probe 2' indicates the area hybridized by a 3 2 P labeled EcoRI-EcoRI fragment of plox2Hyg. Panel B: A Southern blot analysis of Spel digested genomic DNA. A 3 2 P labeled Sphl-Hindlll fragment of plox2Hyg was used as a probe (probe 1) and 2.5 ug of digested genomic DNA was loaded per lane. Molecular weight markers are indicated at the side of the blot. Lane 1 contains lox2Hyg NIH3T3 clone #1 DNA. Lane 2 contains lox2Hyg NIH3T3 clone #4 DNA. Lane 3 contains unaltered NIH3T3 DNA. The membrane was washed at 58°C and exposed to film for 14 days. Panel C: A Southern blot analysis of Smal digested genomic DNA using a 3 2 P labeled EcoRI-EcoRI fragment of plox2Hyg as a probe (probe 2). 2.5 ug of digested genomic DNA was loaded per lane and the molecular weight markers are indicated at the side of the blot. Lane 1 contains lox2Hyg clone #1 NIH3T3 DNA. The membrane was washed at 58°C and exposed to film for 2 days. Copy Number of Target Vectors in the Flox NIH3T3 Cell Line: The pFlox vector contains the gene for neomycin phosphotransferase therefore after electroporation of the linearized vector into NIH3T3 cells, the cells were placed under selection with Genticin and cloned to produce stable clonal cell lines which we will call Flox NIH3T3. These cells were kept under selection with Genticin and the genomic DNA of cloned Flox NIH3T3 cells was isolated. Results of southern blot analyses of the Flox NIH3T3 clones, carried out to assess copy number, are shown in Figure 7. Xbal restriction enzyme digestion of Flox NIH3T3 clone #1 genomic DNA probed with a 32p labeled segment of the Herpes Simplex Virus Thymidine Kinase (HSV TK) sequence produced a single band of approximately 4.3 Kb in size indicating single vector integration. In order to confirm the presence of a single copy of the vector, a HindUI restriction enzyme digestion of Flox NIH3T3 clone #1 genomic DNA was used in Southern blot analysis. A single band of approximately 7.5 Kb was produced. This HindUI digestion result confirmed that the Flox NIH3T3 clone #1 cells contain only a single integrated copy of the Flox vector. Therefore, all further studies used the Flox NIH3T3 clone #1 cells where applicable. A Figure 7. Southern Blot Analyses of clonal Flox NIH3T3 genomic DNA. Panel A: A schematic representation of the linearized Flox vector in the stable Flox NIH3T3 clone #1 cell line with the HindUI and Xbal restriction enzyme sites shown. Area of hybridization by the probe is indicated by I—I. The black arrowheads represent the loxP sites, the hollow arrow represents the HSV TK cassette containing a promoter, HSV TK gene and polyadenylation signal, the black arrow represents the Neomycin cassette containing a promoter, neomycin phosphotransferase gene and a polyadenylation signal, the thin black line represents the pUC based vector sequence and the thick black line represent genomic DNA. Panel B and Panel C: Southern blot analysis of digested Flox NIH3T3 clone #1 genomic DNA. A 3 2 P labeled fragment of HSV TK sequence was used as a probe and 2.5 pg of digested genomic DNA was loaded per lane. Molecular weight markers are indicated at the side of the autoradiograph. Panel B: A Southern blot analysis of Xbal digested genomic DNA. Lane 1 contains 3 2 P labeled molecular weight markers. Lane 2 is blank. Lane 3 contains Flox NIH3T3 clone #1 genomic DNA. The membrane was washed at 60°C and exposed to film for 14 days. Panel C: Southern blot analysis of HindUI digested genomic DNA samples. Lane 1 contains Flox NIH3T3 clone #1 genomic DNA. Lane 2 contains unaltered NIH3T3 DNA. The membrane was washed at 58°C and exposed to film for 1 day. Initial Indication of Recombination Using Double Selection of Infected Target Cells: Because different selection markers are utilized in the retroviral vectors and the target vectors, the initial means of quantifying recombination was to negatively select against those retrovirally infected cells which had undergone recombination causing excision of the resistance marker from the target vector. For example, the established lox^Hyg NIH3T3 clone #1 target cells which had been maintained under hygromycin b selection were infected with the retroviral particles containing the histidinol dehydrogenase gene. The infected lox^Hyg NIH3T3 clone #1 cells were kept under L-histidinol selection for a certain number of days then these same cells were split onto two sets of plates at a series of dilutions. Half of the plates were placed under selection for both the presence of the L -histidinol and the hygromycin b selection markers (double selection) and the other half of the plates remained only under L-histidinol selection (single selection) (Figure 8). The double selection would allow only for the survival of cells which had been infected with the retroviral vector and still contained the hygromycin b selection marker of the lox^Hyg vector while the cells under single selection only needed to contain the retroviral selection marker. A simple formula was used to quantitate the recombination [(number of control cells - number of cells after negative selection) / (number of control cells) x 100% = (percentage of cells killed due to loss of selectable marker by recombination)]. Also by using the non-Cre-containing retroviruses in the same experimentation, any loss of the target vector selection marker occurring naturally in infected cells but in the absence of Cre recombinase could be determined. The results were tabulated as the percentage of cells infected with retroviral particles which were killed due to recombination (Table 4). USING T H E SELECTION MARKERS AS A N INDICATION OF RECOMBINATION lox^Hyg NIH 3T3 clone#1 Hygromycin lox^Hyg NIH 3T3 -i-virus LXSN L c r e - k S N LNKcreSN LXSHD Lcre -kSHD LNKcreSHD + Hygromycin + Viral Selection + Viral Selection (G418 or L-his) (G418 or L-his) Control C e l l s Kills cells which have undergone recombinat ion Figure 8. A schematic representation of the method used to initially determine if Cre recombinase was functional in the retrovirally infected cells. Table 4. Indication of Cre Recombinase Functioning In Target Cells Using the Difference In Selection Markers Between the Target Vectors and Retroviruses. Number of Number of Cells after Percentage Mean of Control Cells Negative Selection of Killing Percentages V i r u s LXSN 265 288 -8 -5 407 410 -1 47 50 -6 L c r e - K S N 1 9 9 0 6 100 97 clone #21 250 2 99 55 5 91 L c r e - K S N 202 55 73 89 clone #22 280 4 99 34 2 94 LNKcreSN 500 71 86 88 polyclonal 192 26 86 140 10 93 LXSHD 400 398 1 -4 403 401 .0 123 138 -12 L c r e - K S H D 168 3 98 99 clone #6 402 4 99 57 0 100 L c r e - K S H D 67 65 3 4 clone #10 55 56 -2 10 9 10 LNKcreSHD 302 25 92 95 polyclonal 404 23 94 181 4 98 T h e cells tested were all l o x 2 H y g NIH3T3 clone #1 cells infected with the noted retrovirus then subjected to selection as previously described. All of the infected cells were cultured for approximately 4 weeks after the initial infection before beginning this study except the L N K c r e S N and L N K c r e S H D infected cells which were cultured for only 2 weeks after infection. T h e cells were quantitated by counting the number of individual colonies of cells grown when subjected to the selection. The number of control cells represents the cells which were placed only under selection for the retroviral resistance marker. T h e number of cells after negative selection represents the cells which were placed under selection for the retroviral resistance marker as well as the hygromycin b resistance marker. T h e results from three experiments are indicated for e a c h retrovirus. The lox^Hyg NIH3T3 clone #1 cells infected with the LNKcreSN polyclonal or LNKcreSHD polyclonal retroviral particles were incubated for approximately 14 days under selection for the retroviral resistance marker before testing. When the LNKcreSN polyclonal and LNKcreSHD polyclonal infected lox^Hyg NIH 3T3 clone #1 cell line were placed under the appropriate double selection, 88% and 95% of the cells, respectively, died when compared to controls. Since the L X S N and LXSHD non-Cre-containing retroviruses produced small differences in the amount of killing when the infected cells were placed under the corresponding single or double selection, the percentages of LNKcreSN and LNKcreSHD infected lox 2Hyg NLH3T3 cells which were killed can be said to have lost the hygromycin b selection marker due to recombination which occurred only in the presence of retrovirally produced Cre protein. The Lcre-KSN clone #21, Lcre-KSN clone #22, Lcre-KSHD clone #6 and Lcre-KSHD clone #10 were individually used to infect lox^Hyg NIH3T3 clone #1 cells and incubated for approximately 28 days under selection for the retroviral marker. As noted previously, the L X S N and L X S H D non-Cre-containing retroviruses produced the same number of colonies when placed under single selection for the retroviral marker or under double selection for the retroviral marker and hygromycin b. For the Lcre-KSN clone #21 retrovirus, 97% of the retrovirally infected lox^Hyg NIH3T3 cells underwent recombination. Similarly, 89 % of the Lcre-KSN clone #22 retrovirally infected lox^Hyg NIH3T3 cells underwent recombination. The Lcre-KSHD clone #6 retrovirus caused 99% of the retrovirally infected lox 2Hyg NIH3T3 cells to lose the hygromycin b selection marker due to recombination whereas the Lcre-KSHD clone #10 retrovirus caused recombination in 4% of the infected target cells. From this study, all of the Cre-containing retroviruses except Lcre-KSHD clone #10 were able to induce recombination and excision of the hygromycin b selection marker in a major portion of the retrovirally infected target cells. This was the first indication that the integrated retroviral vectors are able to produce Cre recombinase protein using the retroviral promoters and enhancers. Since as indicated by the controls, there is no natural loss or mutation of the hygromycin b selection marker, the differences in amount of killing between retroviruses must be due to the amount of Cre protein which is being produced or to the amount of time the retrovirally infected cells were incubated previous to the study. From this study, it has been shown that most of the Cre-containing retroviruses are able to excise the hygromycin b selection marker from approximately 88-99% of the target cell population within 2-4 weeks of initial infection. This is a good indication that the Cre gene is being expressed in the retroviral construct and that the Cre protein is functional within these cells. Southern Blot Analysis of Cre Function Using Target Cells: In this study, the lox^Hyg and Flox NIH3T3 clonal cell lines were infected separately with the retroviral particles. Each group of infected cells was split and placed under selection for the presence of the retroviral selection marker. The infected cells were maintained under this selection for 7 days then the genomic DNA of the infected target cells was isolated, as previously described, for southern blot analysis. All four types of retroviral particles were utilized to infect the lox^Hyg NIH3T3 target cells, whereas, only the histidinol dehydrogenase-containing retroviruses were used to infect the Flox NIH3T3 cell line. The histidinol dehydrogenase containing retroviruses were only used to infect the Flox NIH3T3 cells because both the L X S N vector and the Flox target vector contain the neomycin phosphotransferase gene. After isolation of the genomic DNA, the D N A was digested with the appropriate restriction enzymes and analyzed by southern. In the Southern blot analyses, 32p labeled loxP sequence was used as a probe. 1) lox 2Hyg NIH3T3 target cells In the Southern blot analysis of retrovirally infected lox 2Hyg NIH3T3 clonal cell genomic DNA, two different restriction enzyme combinations were used. The restriction enzyme digestions used were, first, SphI and PvuII and, second, SphI and Bgll. Both of the restriction enzyme combinations allowed for the visualization of differing DNA banding patterns before and after recombination using southern blot analysis and probing with loxP sequence. As controls, the L X S N and LXSHD retroviruses were used because they do not contain Cre recombinase gene sequence and therefore can not induce recombination. Also, non-infected NIH3T3 cell genomic DNA was used to control for any non-specific binding of the DNA probe. In the Southern blot analysis of genomic DNA from infected lox 2Hyg clonal cells where SphI and PvuII was used (Figure 9), the L X S N and L X S H D infected cells produced bands of 2.3 Kb and 0.5 Kb. In the cells infected with Lcre-KSN clone #21, LNKcreSN polyclonal, Lcre-KSHD clone #6 and LNKcreSHD polyclonal retroviruses, relatively less intense bands of 2.3 Kb and 0.5 Kb as well as an additional band of 0.7 Kb were produced. Since equal amounts of genomic DNA were used in all the analyses, some comparison of band intensities can be made. The Lcre-KSN clone #21 and LNKcreSN polyclonal retrovirally infected cells seemed to have caused a greater amount of recombination than the Lcre-KSHD clone # 6 and LNKcreSHD polyclonal retroviruses due to greater intensity of the 0.7 Kb bands produced. Also, the genomic DNA from Lcre-KSN clone #21 and LNKcreSN polyclonal infected cells produced 2.3 Kb and 0.5 Kb bands less intense than those produced by the control, L X S N and L X S H D retrovirally infected cells. None the less, all of the retroviruses expressing Cre produced a 0.7 Kb band indicating that recombination had occurred. Such comparisons are complicated however by the fact that polyclonal populations of infected cells are analyzed representing many different integration events and hence potentially widely different levels of Cre expression. 44 A 1 2 3 4 5 6 7 Figure 9. Southern blot analysis of retrovirally infected lox2Hyg clone #1 target cells. Panel A: A schematic representation of the effects of recombination on the restriction enzyme analyses of the lox2Hyg target vector. The area of hybridization of the probe is indicated by I—I. The hollow arrow represents a PGK promoter, the hollow box represents the hygromycin b resistance marker gene, the black box represents the polyadenylation signal, the arrowheads represent the loxP sites, the thin black line represents the pUC based vector backbone and the thick black line represents genomic DNA. The Sphl and Pvull restriction enzyme sites and positions are indicated on the diagram. Panel B: A Southern blot analysis of Sphl-Pvull digested lox2Hyg clone #1 genomic DNA from cells infected with retroviruses. Each lane contains 5ug of digested genomic DNA which was probed with 3 2 P labeled loxP sequence. The cells were infected with the following retroviruses: Lane 1: LXSN, Lane 2: Lcre-KSN clone #21, Lane 3: LNKcreSN polyclonal, Lane 4: LXSHD, Lane 5: Lcre-KSHD clone #6, and Lane 6: LNKcreSHD polyclonal. Lane 7 is unaltered NIH3T3 genomic DNA as a control. The 2.3 Kb band is the Sphl-Pvull band produced by the nonrecombined target vector. The 0.5 Kb band is the Sphl-Sphl band also produced by the nonrecombined target vector. The 0.7 Kb band is produced only in the recombined target vector from the Sphl-Pvull band. The membrane was washed at 60°C and exposed to film for 11 days. Southern blot analyses carried out using SphI and Bgll confirmed the previous results (Figure 10). Retrovirally infected clonal lox 2Hyg NIH3T3 cells produced bands of 3.3 Kb and 0.5 Kb when probed with loxP sequence. When comparing the bands produced from the genomic D N A of Lcre-KSN clone #21 and LNKcreSN polyclonal retrovirally infected cells with the L X S N retrovirus control cells, the Lcre-KSN clone #21 and LNKcreSN polyclonal lanes produced less intense 3.3 KB and 0.5 Kb bands and produced additional 2.0 Kb bands. The LNKcreSHD polyclonal retrovirus infected cells also produced a distinct 2.0 Kb band and slightly less intense 3.3 Kb and 0.5 Kb bands when compared to the LXSHD retrovirus infected cells. The Lcre-KSHD clone #6 infected cells did not produce significantly less intense 3.3 Kb and 0.5 Kb bands than the L X S H D retrovirus control but did still produce a faint 2.0 Kb. These results indicated, by the production of the 2.0 Kb band, that recombination had occurred in some of the retrovirally infected lox 2Hyg NIH3T3 cells. Sphl(1 010) SphlY.496) M BglJ(4505) t BglJ(2453) ^ -3 .3 Kb Recombinant ^-2.0 Kb 0.5 Kb Figure 10. Southern blot analysis of retrovirally infected lox2Hyg clone #1 target cells using SphI and Bgll restriction enzymes. Panel A: A schematic representation of the effects of recombination on the restriction enzyme analyses of the lox2Hyg target vector. The area of hybridization of the probe is indicated by I—I. The hollow arrow represents a PGK promoter, the hollow box represents the hygromycin b resistance marker gene, the black box represents the polyadenylation signal, the arrowheads represent the loxP sites, the thin black line represents the pUC based vector backbone and the thick black line represents genomic DNA. The SphI and Bgll restriction enzyme sites and positions are indicated on the diagram. Panel B: A Southern blot analysis of Sphl-Bgll digested lox2Hyg NIH 3T3 clone #1 genomic DNA from cells infected with retroviruses. Each lane contains 5ug of digested genomic DNA which was probed with loxP sequence. The cells were infected with the following retroviruses: Lane 1: LXSN, Lane 2: Lcre-KSN clone #21, Lane 3: LNKcreSN polyclonal, Lane 4: LXSHD, Lane 5: Lcre-KSHD clone #6, and Lane 6: LNKcreSHD polyclonal. Lane 7 is unaltered NIH3T3 genomic DNA as a control. The 3.3 Kb band is the Bgll-Bgll band produced by the nonrecombined target vector. The 0.5 Kb band is the Sphl-Sphl band also produced by the unrecombined target vector. The 2.0 Kb band is produced only in the recombined target vector from the Sphl-Bgll band. The membrane was washed at 60°C and exposed to film for 11 days. Densitometry of both the autoradiographs in figures 9 and 10 were used to compare band intensities (Table 5). Densitometry was completed on all of the lanes in each figure and with the use of computer software the major peaks in density were labeled and integrated. The densitometry data for both figures showed that the percent area of the 0.5 Kb band in the cells infected with the Cre-containing retroviruses is always lower than in the cells infected with the corresponding control L X S N or L X S H D retrovirus. The data also indicates that the percent area of the 3.3 Kb band in figure 10 decreases approximately the same amount as the 0.5 Kb band in each corresponding lane. These results are consistent with the idea that, with restriction analyses used, the 0.5 Kb and 3.3 Kb or 2.3 Kb bands are lost after recombination reaction resulting in the production of a single 0.7 Kb or 2.0 Kb recombination band. Inconsistent with this theory is the fact that the percent area of the 2.3 Kb band in most cases increased between the Cre-containing and control retrovirus. This may be true because the area of the film corresponding to this band may have been overexposed due to the prolonged exposures of the autoradiographs therefore changes in the band intensities would not be apparent unless the exposures were shortened. The reason the exposures of figure 9 were extended was because of the low intensities of the 0.5 Kb and 0.8 Kb therefore requiring longer exposures to visualize the bands. In this analysis, comparisons of the recombination bands and the 0.5 Kb bands in each lane were used determine relative efficiencies of each retrovirus. By determining differences in relative band intensities between a band from the unrecombined vector to the recombination band, we see differences in the proportion of cells containing recombined vector. According to this analysis, the LNKcreSHD retrovirus produced ratios above 1:1 whereas the Lcre-KSHD retrovirus produced ratios below 1:1. This may indicate that the LNKcreSHD retrovirus has slightly higher Cre recombination frequencies than the Lcre-KSHD retrovirus since the amount of recombinant band to control band is larger for LNKcreSHD than Lcre-KSHD. The LNKcreSN and Lcre-KSN retroviruses in figure 9 produced ratios of less than 1: 1 but in figure 10 produce ratios of greater than 3:1. Also the Lcre-KSN retrovirus in figure 9 produced a lower ratio than the LNKcreSN virus whereas in figure 10 Lcre-KSN virus produces a higher ratio than LNKcreSN virus. As for differences between the neomycin phosphotransferase (Neo) containing and histidinol dehydrogenase (L-his) containing retroviruses, the Neo viruses produced higher ratios in figure 10 while the L-his viruses produced higher ratios in figure 9. This discrepancy does not show any consistent advantage of one set of retroviruses over the others. Table 5. Densitometry of the Autoradiographs from Figures 9 and 10. Densitometry data from Figure 9 Approximation of % of Total Area of Band relative quantities V i rus 0.5 Kb 0. 7 Kb 2.3 Kb 0.7 Kb/0.5 Kb L X S N (Lane 1) 1 9 . 6 0 0 8 0 . 4 0:19.6 L c r e - K S N (Lane 2) 18 .2 0 0 8 1 . 8 0:18.2 L N K c r e S N (Lane 3) 1 0 . 4 3 0 8 6 . 7 1:3.3 L X S H D (Lane 4) 2 3 . 5 0 0 7 6 . 5 0:23.5 L c r e - K S H D (Lane 5) 1 4 . 7 8 9 7 6 . 4 1:1.6 L N K c r e S H D (Lane 6) 6.3 9 6 84.1 1.5:1 Densitometry data from Figure 10 Approximation of % of Total Area of Bands relative quantities Virus 0.5 Kb 2.0 Kb 3.3 Kb 2.0 Kb/0.5 Kb L X S N (Lane 1) 2 2 . 6 0.0 7 7 . 4 0:22.6 L c r e - K S N (Lane 2) 7.5 32.1 6 0 . 4 4.3:1 L N K c r e S N (Lane 3) 8.6 29.1 6 2 . 3 3.4:1 L X S H D (Lane 4) 2 1 . 2 0.0 7 8 . 8 0:21.2 L c r e - K S H D (Lane 5) 1 8 . 6 6.8 7 4 . 7 1:2.7 L N K c r e S H D ( L a n e 6) 1 3 . 0 16 .2 7 0 . 8 1.2:1 Densitometry was measured using a Hoefer Scientific (San Francisco, Calif.) G S 300 Transmittance/Ref lectance Scanning Densitometer. T h e program application G S 370 version 2.3 was used for scanning and analysis. Peaks of density were labeled and integrated automatically using the application program. % of total a rea of bands represents the relative size of the shown bands as calculated by the application software. 2) Flox NIH3T3 target cells The Flox NIH3T3 clone #1 cell line was infected separately with LXSHD, Lcre-KSHD clone #6 and LNKcreSHD polyclonal retroviruses then split and placed under L -histidinol selection for 7 days before isolating the genomic DNA. The restriction enzyme combinations used in the Southern blot analyses in this study were, first, Sspl and Hindlll and, second, Sspl and Xhol. The initial observation of the all the Flox clonal cell line genomic DNA samples was that all the samples produced an unexpected 1.8 Kb band Of approximately equal intensity (Figures 11 and Figure 12). Because it was produced at equal intensities and only in the genomic samples which contain the Flox vector, it was assumed that the 1.8 Kb band was produced from binding of the loxP probe to a fragment of the backbone vector of the Flox vector. Because the backbone of the vector is outside loxP flanked area, the 1.8 Kb fragment will not be altered by recombination and is not relevant to the results. Therefore for the southern blot results involving the Flox clonal cell line, the 1.8 Kb band is disregarded. When southern blot analysis was used to analyze L X S H D infected Flox NIH3T3 clone #1 genomic DNA digested with Sspl and Hindlll restriction enzymes, a single 4.3 Kb band was produced (Figure 11). When genomic DNA from Flox NIH3T3 clonal cells infected with the Lcre-KSHD clone #6 or LNKcreSHD polyclonal retroviruses was analyzed, a less intense 4.3 Kb band and an addition a 0.8 Kb band was produced when compared to the L X S H D infected cells. This indicated that some of the cells which had been infected with the Cre expressing retroviral particles were able to induce recombination but this recombination did not occur in 100% of the infected population as indicated by the presence of a 4.3 Kb band. 51 A Figure 11. Southern blot analysis of retrovirally infected Flox clone #1 target cells using Sspl and Hindlll restriction enzymes. Panel A : A schematic representation of the effect of recombination on the restriction enzyme band sizes of the Flox target vector. Area of hybridization by the probe is indicated by I—I. The HSV TK cassette is indicated by the hollow arrow and the neomycin cassette is indicated by the black arrow. The arrowheads indicate the positions of the loxP sites. The thin black line represents backbone vector sequence and the thick black line represents genomic DNA. Panel B: Southern blot analysis of genomic DNA from Flox clone #1 NIH3T3 cells infected with retroviruses and incubated for 7 days under selection for the retroviral vector. The blot was probed with 3 2 P labeled loxP sequence. Each lane contains 5 ug of Flox clone #1 genomic DNA digested with Sspl and Hindlll restriction enzymes from cells infected with: Lane 1: LXSHD, Lane 2: Lcre-KSHD clone #6, Lane 3 LNKcreSHD polyclonal and Lane 4 is genomic DNA from NIH3T3 cells infected with LNKcreSHD polyclonal retrovirus. The 4.3 Kb band represents the expected Sspl-Hindll l fragment produced by the unrecombined target vector as shown in panel A . The 0.8 Kb fragment is the expected Sspl-Hindlll fragment produced by the recombined target vector. The 1.8 Kb fragment is produced due to hybridization of the loxP probe to sequence 3' of the Hindlll site in the target vector backbone. The blot was washed at 50°C and exposed to film for 14 days. In order to confirm, the restriction enzyme combination of Sspl and Xhol was used (Figure 12). The genomic D N A from the L X S H D infected Flox NIH3T3 clonal cells produced the expected 4.3 Kb band. The genomic DNA from Flox NIH3T3 clonal cells infected with either Lcre-KSHD clone #6 or LNKcreSHD polyclonal retroviruses produced a reduced 4.3 Kb band as well as a 0.8 Kb band. Therefore because a 0.8 Kb band was expected after Cre-mediated recombination, these results confirmed that Cre-mediated recombination did occur in those target cells infected with a Cre expressing retroviral particle. 53 Sspl(207) t Sspl(207) hol(980) Xhol(4495) B 4.3 Kb 1.8 Kb Recombinant 0.8 Kb Figure 12. Southern blot analysis of retrovirally infected Flox clone #1 target cells using Sspl and Xhol restriction enzymes. Panel A: A schematic representation of the effect of recombination on the restriction enzyme band sizes of the Flox target vector. Area of hybridization by the probe is indicated by I—I. The HSV TK cassette is indicated by the hollow arrow and the neomycin cassette is indicated by the black arrow. The arrowheads indicate the positions of the loxP sites. The thin black line is backbone vector sequence and the thick black line is genomic DNA. Panel B: Southern blot analysis of genomic DNA from Flox clone #1 NIH3T3 cells infected with retroviruses and incubated for 7 days under selection for the retroviral vector. The membrane was probed with 3 2 P labeled loxP sequence. Each lane contains 5 pg of Flox clone #1 genomic DNA digested with Sspl and Xhol restriction enzymes from cells infected with: Lane 1: LXSHD, Lane 2: Lcre-KSHD clone #6, Lane 3: LNKcreSHD polyclonal and Lane 4 is genomic DNA from NIH3T3 cells infected with LNKcreSHD polyclonal retrovirus. The 4.3 Kb band represents the expected Sspl-Xhol fragment produced by the unrecombined target vector as shown in panel A. The 0.8 Kb fragment is the expected Sspl-Xhol fragment produced by the recombined target vector. The 1.8 Kb fragment is produced due to hybridization of the loxP probe to sequence 3' of the Xhol site in the target vector backbone. Blot was washed at 50°C and exposed to film for 14 days. Densitometry was again used in figures 11 and 12 to look at changes in the intensities of bands produced by the Cre-containing and control virus infected cells. In the data from figures 11 and 12 it was possible to use the 1.8 Kb band as a control since as explained earlier we expect this band to remain constant since this fragment lies outside of the loxP sites and therefore should remain the same after recombination. In order to determine relative changes in band strengths we have taken the ratio of the percent areas of the 0.8 Kb and the control 1.8 Kb bands. When looking at the ratios, it is assumed that if the density of the 1.8 Kb band remains the same that the ratio of the bands will increase or decrease with the intensity of the 0.8 Kb band. The ratios from figure 11 are about 1: 2 for Lcre-KSHD to 1: 3 for LNKcreSHD compared to 1:2.1 for Lcre-KSHD and 1.1:1 for LNKcreSHD from figure 12. Because differing retroviruses have higher ratios in each of the figures, we can not make a conclusion as to which of the retroviruses produces a higher Cre recombination frequency. In both figures, the percent areas of the 4.3 Kb bands are remain equal or increase between lanes and this is thought to be due to overexposure of the areas of the autoradiograph film corresponding to this band. 55 Table 6. Densitometry of the Autoradiographs from Figures 11 and 12. Densitometry data from Figure 11 Approximation of % of Total Area of Band relative quantities Vi rus 0.8 Kb 1.8 Kb 4.3 Kb 0.8 Kb/1.8 Kb L X S H D (Lane 1) 0 .0 5 .7 9 4 . 3 0:5.7 L c r e - K S H D (Lane 2) 2.1 4 . 3 9 3 . 6 1:2 L N K c r e S H D (Lane 3) 0.9 2 .7 9 6 . 4 1:3 Densitometry data from Figure 12 % of Total Area of Band Approximation of relative quantities Virus 0.8 Kb 1.8 Kb 4.3 Kb 0.8 Kb/1.8 Kb L X S N ( L a n e 1) 0 .0 2 .0 9 8 . 0 0:2 L c r e - K S H D (Lane 2) 0 .7 1.5 9 7 . 8 1:2.1 L N K c r e S H D (Lane 3) 1.6 1.4 9 7 . 0 1.1:1 Densitometry was measured using a Hoefer Scientific (San Francisco, Calif.) G S 300 Transmittance/Ref lectance Scanning Densitometer. T h e program application G S 370 version 2.3 was used for scanning and analysis. Peaks of density were labeled and integrated automatically using the application program. % of total a rea of bands represents the relative size of each integrated peak. Amount of Recombination in the Total Cell Population Over Time: Because the original Southern blot analyses of the retrovirally infected target cell genomic DNAs did indicate that recombination was occurring but that less than 100% of the infected target cells underwent recombination, the next step was to determine if recombination would reach 100% by increasing the amount of incubation time after the infection. In order to test this, infections were completed in the same manner as previously described but instead the incubation times were increased from 7 days to 3 weeks. During the entire incubation period, the infected cells were maintained under selection corresponding to the retrovirus used in the infection. In order to analyze the genomic D N A samples, the same restriction enzymes and 32p labeled D N A probe were used as previously described. In this study, the lox^Hyg NIH3T3 clone #1 cell line was infected with Lcre-KSN clone #21 retrovirus whereas the Flox NIH3T3 clone #1 cell line was infected separately with the Lcre-KSHD clone #6 and LNKcreSHD polyclonal retroviruses. As shown in Figure 13, Southern blot analysis of the L X S N infected lox^Hyg cells produced the expected 2.8 Kb, 2.3 Kb and 0.5 Kb bands when genomic DNA was digested with SphI and PvuIJ. restriction enzymes whether the infected cells were incubated for 7 or 23 days. As expected, the genomic DNA from lox 2Hyg NIH3T3 cells infected with Lcre-KSN clone #21 retrovirus and incubated for 7 days produced the expected 2.8 Kb, 2.3 Kb and 0.5 Kb bands as well as a 0.7 Kb band. This result, again, indicates that recombination had occurred in some but not all of the infected cells. When the same cells were incubated for 23 days before isolating the genomic DNA, only a single 0.7 Kb band is produced. This demonstrates that while only some of the Lcre-KSN clone #21 infected cells underwent recombination after incubation for 7 days, most of the cells infected with the Lcre-KSN clone #21 retrovirus underwent recombination if the cells were incubated for 23 days after the infection. ^ Sphl(1 01 0) Sph^(49 6) PvuII ( 3264) P. t PvuJI ( 1212) Recombination band 0.7 Kb Figure 13. Determination of change in amount of recombination overtime in infected l o x 2 H y g clone #1 NIH3T3 cells. Blots were probed with 3 2 P labeled loxP sequence. Panel A: A schematic representation of recombination of plox2Hyg vector in NIH3T3 genome. SphI and PvuII restriction enzyme sites are indicated. The area of hybridization of the probe is indicated by I—I. The hollow arrow represents a P G K promoter, the hollow box represents the hygromycin b resistance marker gene, the striped box represents the polyadenylation signal, the arrowheads represent the loxP sites, the thin black line represents the p U C based vector backbone and the thick black line represents genomic D N A . Panel B and Panel C: Southern blot analyses of genomic D N A from retrovirally infected lox 2 Hyg NIH3T3 clone #1 cells. 5 pg of digested genomic DNA was loaded per lane and the samples were digested with SphI and PvuII restriction enzymes. The lox 2 Hyg clone #1 cells were infected with the following retroviruses: Lane 1: L X S N and Lane 2: Lcre-KSN clone #21. Panel B: The infected cells were incubated 7 days after infection with retrovirus before isolation of the genomic DNA. The membrane was washed at 5 7 ° C and exposed to film for 12 days. Panel C: The infected cells were incubated 23 days after infection with retrovirus before isolation of the genomic DNA. The 2.3 Kb band is the Sphl-Pvull band produced by the nonrecombined target vector. The 0.5 Kb band is the Sphl-Sphl band also produced by the nonrecombined target vector. The 0.7 Kb band is produced only in the recombined target vector from the Sphl-Pvull band. The membrane was washed at 5 5 ° C and exposed to film for 7 days. Using the Flox NIH3T3 target cell line in the same study, the infected cells were incubated for 7 to 21 days previous to genomic DNA isolation (Figure 14). As expected, Southern blot analysis of L X S H D infected target cells incubated for 7 or 21 days both produced the a single 4.3 Kb band when the genomic DNA was digested with Sspl and Hindlll restriction enzymes. When the Lcre-KSHD clone #6 or LNKcreSHD polyclonal retrovirally infected target cells were incubated for 7 days previous to genomic D N A isolation, 4.3 Kb and 0.8 Kb bands were produced by Southern blot analysis. But when these same Lcre-KSHD clone #6 infected cells were allowed to incubate for 21 days before genomic D N A isolation, only single 0.8 Kb bands were produced. The LNKcreSHD polyclonal infected cells also produced a much stronger 0.8 Kb recombination band and a less intense 4.3 Kb band. This indicates that most of the target cells infected with Cre-expressing retroviral particles underwent recombination within the 21 days of incubation after the infection thus confirming that by increasing the length of incubation, it is possible to increase the number of infected cells which undergo recombination. 59 Figure 14. Determination of change in amount of recombination over time in infected Flox NIH3T3 clone #1 cells. Panel A: A schematic representation of the effect of recombination on the restriction enzyme band sizes of the Flox target vector. Area of hybridization by the probe is indicated by I—I. The HSV TK cassette is indicated by the hollow arrow and the neomycin cassette is indicated by the black arrow. The arrowheads indicate the positions of the loxP sites. The thin black line represents backbone vector sequence and the thick black line represents genomic DNA. Panel B and Panel C: The membranes were probed with 3 2 P labeled loxP sequence. Each lane contains 5 ug of Flox clone #1 genomic DNA digested with Sspl and Hindlll restriction enzymes from cells infected with: Lane 1: LXSHD, Lane 2: Lcre-KSHD clone #6, Lane 3: LNKcreSHD polyclonal and Lane 4 is genomic DNA from NIH3T3 cells infected with LNKcreSHD polyclonal retrovirus. Panel B: Southern blot analysis of infected cells which were incubated 7 days after infection with retrovirus before isolation of the genomic DNA. The blot was washed at 50°C and exposed to film for 14 days. Panel C: Southern blot analysis of infected cells which were incubated 21 days after infection with retrovirus before isolation of the genomic DNA. The membrane was washed at 55°C and exposed to film for 14 days. The 4.3 Kb band represents the expected Sspl-Hindlll fragment produced by the unrecombined target vector as shown in panel A. The 0.8 Kb fragment is the expected Sspl-Hindlll fragment produced by the recombined target vector. The 1.8 Kb fragment is produced due to hybridization of the loxP probe to sequence 3' of the Hindlll site in the target vector backbone. The blot was washed at 58°C and exposed to film for 2 days. 60 CHAPTER IV D I S C U S S I O N A N D C O N C L U S I O N S We have now shown that stable retroviral producing cell lines have been constructed which will generate high titer retroviruses capable of transducing and expressing a Cre recombinase gene and a selectable marker. The retrovirus producing cell lines are able to generate retroviral titers between 1x10^ and 1x10^ retroviral particles per ml of cell culture media and the retroviral particles can efficiently transfer the retroviral genomes into infected cells. Once expressed, the retrovirally produced Cre protein has been shown to specifically excise DNA sequence flanked by loxP sites and cause recombination in approximately 100% of infected clonal target cell populations within approximately 3 weeks of infection. The efficient transfer system of a retrovirus has now been combined with the site specific recombination abilities of the Cre protein to produce a powerful tool for molecular biological studies. Once the retroviral producing cell lines were constructed, the next step was to determine if the viral titers being generated were high enough to make the retroviruses useful in experimentation. During the original construction of the PA317 retroviral packaging cell line, it had been shown that this cell line was capable of making retroviral titers upto 10^ particles per ml of media (Miller and Buttimore, 1986), but when the L X S N retroviral construct was used in combination with PA317 cells, titers of only 10^ particles per ml of media were produced (Miller and Rosman, 1989). The titers of the retroviruses generated in our study ranged from 3 x 10^ to 1 x 10° (Table 3). Because viral titers between 10^-10^ are accepted as being useful in most applications, only the Lcre-KSHD retroviruses may be limited in their applications due to their slightly lower retroviral titer production. During the study of Cre function, stable target cell lines were generated and tissue culture using the differences in selection markers in target cells and retroviruses was used to determine if the retrovirally produced Cre recombinase was able to excise the selection marker from the target vector. As shown in table 4, the different selection markers allowed for negative selection of those cells which had undergone recombination. This was an effective way of determining if recombination was occurring and a means of quantitating the amount of recombination in a certain amount of time after the initial infection. The results indicated, that compared to the controls, the Cre-containing retroviruses were able to excise the target vector's selection marker and cause loss of selection marker resistance in over 90% of the infected cells. All but one of the retroviruses produced a positive result and gave the indication that the retrovirally produced Cre recombinase was able to excise the target vector's selection marker. Densitometry was used to assess the relative abilities of each of the retroviruses as indicated in tables 5 and 6. The ratio data which allow us to compare the amount of recombined vector to nonrecombined vector within the DNA sample used was effective in giving us an indication as to which of the viruses induced greater Cre recombination frequencies. The one problem with this study was the lack of a control band in the lox 2Hyg studies. Without a control band the ratios are still an indication but are not as conclusive because there is no means of controlling for changes in band intensities between lanes. With these difficulties in mind, the data shown did not seem to be an advantage of using one virus over another. Therefore in our study we are not able to conclude that the addition of the nuclear localization signal was of any affect to Cre recombination frequencies. Also, as indicated in Figures 9, 10, 11 and 12, none of the Cre-containing retroviruses was able to catalyze recombination of the target vectors in 100% of the infected population within 7 days of infection. This became important since, in many studies, it would be necessary excise the gene of interest in all of the infected cells before the study may proceed. Granted, this could be accomplished by analyzing clones of infected cells but in the case where primary cells are being used or cell numbers are limited, this may be detrimental. Another way to do this would be to use selection markers, such as HSV T K , which would allow for positive selection of those cells which have undergone recombination but, again, this is not always possible (Gu et al. 1993). Because these methods have not been used in this study and the infected target cell populations may contain multiple integration sites leading to differing gene expression levels, the incubation times of the retrovirally infected cells were extended to determine if the amount of recombination in the total cell population would increase to 100%. Figures 13 and 14 both indicate that if the retrovirally infected cells are incubated for longer than 21 days the amount of recombination in the total population does approach 100%. Therefore even though the use of cloning after infection and the presence of a negatively selectable marker would be useful, it may not be essential to use these methods to gain a population of infected cells where approximately 100% have undergone recombination. When looking at the increase in recombination over time, a comparison of the data in table 4 and figures 13 and 14 can be made. Figures 13 and 14 use Southern blot analysis to show the increase in recombination when incubation times, after infection, are increased from 7 to 21 days and table 4 indicates the amount of recombination which occurred in target cells, 14 to 28 days after infection, using the differences in selection as an indication. By comparing the intensities of the bands on the autoradiographs with the percentages produced by the "double selection" study, we can see that both sets of data indicate that when the target cells were incubated for greater than 14 days after infection, the amount of recombination which occurred was approximately 90% or more. This coincidence in data confirms that both analyses are valid and that the retrovirally produced Cre gene is able to cause recombination in close to 100% of Cre expressing cells within 14 days of infection. Curiously, as previously described (Miller and Rosman, 1989), when the L X S N retroviral construct was used to produce retrovirus in PA317 cells, the viral titers generated were approximately 10^ particles per ml of media. In the same study a similar retroviral construct which contained the neomycin phosphotransferase gene expressed by the L T R instead of an internal SV40 promoter was produced, this construct produced retroviral titers of approximately 10^ particles per ml. In theory, this may be true because the SV40 is not, in certain situations, the best promoter for gene expression (Thomas et al. 1987 and Miller et al. 1989). In that case, only the L X S N type retroviruses which had integrated into highly transcribed areas would be able to produce high enough expression of the neomycin phosphotransferase gene to generate G418 resistance. If this theory is true then it would also mean that this retroviral construct may select for those sites of integration which have the highest amounts of transcription and ultimately, the highest levels of gene expression. Since this added selection would, of course, be favorable when the ultimate goal is to transfer and express a gene, the downfall would be that the usable retroviral titer may not be as high as would be liked. One of the concerns with this retroviral system was the requirement of 2-3 weeks of incubation for recombination to approach 100% in the population of infected cells. This extended incubation period may be detrimental in studies with the use of primary cells which a finite life span. One of the reasons for this seemingly low efficiency of the Cre recombinase may be that the cell lines which were analyzed for recombination were not clonal after infection and therefore contain cells which have differing integration sites of the retrovirus. With differing integration sites of the retrovirus, it is possible that there may be large differences in the expression levels of the Cre recombinase between cells. In order to reduce the incubation times required, it may be necessary to clone the infected cells and analyze the cells for recombination after a shorter period of time. Those clones which contain a retrovirus in an area of high transcription may have increased expression of the Cre recombinase and thus show recombination approaching 100% in the population much faster than a polyclonal cell lines as was analyzed here. Other possible considerations for the low efficiency of the Cre recombinase could be because the Cre gene originated from bacterial it may contain some sequences which are detrimental to the production of viral RNA transcripts. Low efficiency splice sites within the Cre sequence may inhibit the production of Cre by causing unusual splicing where part of the Cre coding sequence is lost and only part of the Cre sequence is translated. The presence of additional polyadenylation signals within the Cre sequence would also account for the low efficiency of the Cre recombinase. Additional polyadenylation signals within the Cre gene would prematurely halt the transcript and inhibit the production of functional Cre protein because the entire protein coding sequence has not been transcribed. Analysis of the sequence of the Cre gene has shown the presence of at least one or more sequences which is within one nucleotide of a consensus GT(A or G)AGT 5' or (C or T)AG 3' splice site and one A A T A A site which may act as a polyadenylation signal. The use of only one of the clonal target cell lines may also have an affect on the rate of recombination with the cell line. It is theorized that the site of integration of the target vector may have a profound affect on the rate of recombination depending on Cre's ability to access the loxP sites for recombination. Therefore since all of the studies utilized the use of only one target vector clone cell line, the integration site of the target vector may affect all of the studies and reduce the recombination frequencies sufficiently to cause week-long incubations of Cre expression in order to show recombination. With the extended length of incubation required for recombination to approach 100%, the possible uses for the Cre-containing retroviruses has been reduced. Long periods of incubation would not be desirable where the use of primary cells are required because these cells only survive a short time on tissue culture plates and may not survive long enough to ensure recombination. Extended incubations would not be a problem in the use of immortal cell lines due to the cells' immortality and less restriction on time. In order to increase the efficiency of the recombination and reduce the requirement for long incubations after infection, it may be possible to use PCR to remove the splice and polyadenylation signals from the Cre gene by changing the codon usage without affecting the protein structure produced. These changes may make the Cre gene more appropriate for mammalian transcription and translation systems. Due to the necessity of the DNA-protein interaction between Cre recombinase and the genomic DNA of a cell, it can be concluded that it is essential for the Cre recombinase to gain access to the nucleus of a cell. Without a nuclear localization signal, it is assumed that the Cre protein would only be able to enter the nucleus after the nuclear membrane has been degraded in cell division. But in the case of static cells, it would be impossible for the Cre protein to enter the nucleus and induce recombination since the cells do not divide and therefore degrade their nuclear envelopes. This would therefore cause a problem if the cells which are to be infected become terminally differentiated and no longer undergo cell division. Even though this has not been proven, this problem has hopefully been alleviated with the production of retroviral particles which express a Cre recombinase gene containing a nuclear localization signal. With the addition of this signal, the Cre protein should be able to gain access to the nucleus without the need for nuclear degradation. This should therefore make these retroviral particles effective even in the event that the infected cells become static and normally would be unable to undergo recombination due to the unavailability of the Cre protein to enter the nucleus. In an attempt to prove the worth of the nuclear localization signal, a comparison of the Cre sequences which did and did not contain a nuclear localization signal was made. The difference between the efficiencies of the retroviruses containing and not containing the nuclear localization signal was not substantial since both types of retroviruses showed recombination in only some of the total population of target cells 7 days after the retroviral infection and both did cause recombination in almost 100% of the infected target cells when incubated for 3 weeks after the retroviral infection. The densitometry data from tables 5 and 6 also gave no definite trends as to which of the retrovirally produced Cre proteins was more efficient. This lack of difference may be due to the difficulties talked about earlier but previous studies have shown that the addition of a nuclear localization signal from the SV40 large T antigen causes the Cre gene to produce protein with a high recombination frequency in ES cells (Gu et al. 1993). Therefore at least when used in actively dividing NIH3T3 fibroblasts, there does not seem to be a significant difference in the efficiency of the Cre recombinase when a nuclear localization signal is added. Other limitations of this system were discovered during the viral production and testing. Firstly, during the original production of the retroviral vectors, three additional vectors were constructed and tested. These vectors differed from L X S N and L X S H D in that the Cre gene was placed under the control of an internal C M V promoter instead of the LTR promoter. When these vectors were transfected into the retroviral packaging lines, minimal numbers of selection resistant cells were produced. Also when these marker resistant cells were tested for a titer of retrovirus, it was found that none of these cells was able to produce any detectable titers of virus. This result was consistent when any of the three separate retroviral vectors was tested. The inability of retroviral particles to be produced when the C M V promoter was placed in the retroviral vector may have been due to the presence of cryptic splice site and polyadenylation signals within the C M V promoter. By sequence analysis, the Cre gene is already known to contain possible low efficiency splice and polyadenylation signals therefore when the C M V promoter and Cre sequence are placed in the same vector the number of cryptic sites may be large enough to severely hinder the retroviral production. Another possible explanation as to why retroviruses containing the C M V promoter may not be produced is the presence of "promoter suppression" (Emerman et al. 1984). Although the mechanism of "promoter suppression" is not known, it has been shown that the presence of a second promoter in a retroviral construct can cause suppression of either the LTR or the internal promoter to occur. In this case it may be possible that the C M V promoter may be suppressing transcription initiation from the L T R promoter and therefore not allowing production of the full-length viral transcript required for virion production. Secondly, because only those viruses which placed the Cre gene under the control of the LTR promoter were successful, this lead to difficulties when infecting embryonic stem cells. When the virus was used to infect embryonic stem cells, it was found that the cells would become resistant to the selection corresponding to the virus but there was no induction of recombination. This result was found to be expected due to literature which shows that LTR transcriptional enhancers are down regulated in embryonic stem cells (Reisman et al. 1989, Linney et al. 1987, Gorman et al. 1985, Takeda et al. 1988, and Wagner et al. 1988) but other promoter systems such as SV40 are unaffected. This repression is thought to be due to the presence of hypermethylation of the viral enhancer areas (Stuhlman et al. 1981 and Stewart et al. 1982) and/or negative regulation by an adenovirus E l A-like trans -acting repressor which is present in undifferentiated cells (La Thangue et al. 1987 and Sassone-Corsi et al. 1987). Therefore, our result coincided with previous information that LTR promoters are shut down in ES cells. Even with these drawbacks one theoretical dividend has been found. As stated previously, it has been shown that most retroviral LTRs are repressed when used for expression in ES cells. The interesting point is that it has been shown that transcriptional initiation can be induced from M - M L V LTRs by inducing differentiate of the infected ES cells (Reisman et al. 1989, and Wagner et al. 1988). Therefore it may be possible to retrovirally infect stem cells with our retrovirus and be able to induce Cre-mediated recombination by inducing stem cell differentiation. This system may have merit in genetic marking of produced cells as well as knock out of genes only in the differentiated cell types. What we have shown here is the production of a retroviral particle that has the ability to transport and express the Cre recombinase protein within a mammalian cell system. The retrovirally produced Cre protein is able to specifically catalyze recombination in a site-specific manner as expected and we believe this retroviral particle will have a number of applications such as specific knockout of loxP -flanked genes within retrovirally infected cells. This retrovirus may be useful in removing resistance markers from transfected vectors by flanking resistance markers with loxP sites previous to transfection then retrovirally infecting the vector-containing cells to specifically remove the markers. It has also been thought that by using homologous recombination to place loxP sites in specific areas of the genome, it may be possible to study chromosome structure by determining changes in Cre's ability to cause recombination in differing areas associated with chromosome structure. Recently, Cre recombinase has been shown to induce non-homologous chromosomal translocations and deletions to mimic human disease in murine cell systems therefore the produced retrovirus may have some applications in this type of chromosomal engineering. Our original goal was to use this retrovirus to specifically remove loxP flanked genes from isolated thymocytes or bone marrow cells which could be infected, recombination induced and the mutated cells would be studied in vitro or transplanted to SCID mice for in vivo studies. We believe this retrovirus will be useful in number situations where by simply infecting cells with the retrovirus will produce constitutive expression of the Cre recombinase and specific recombination between loxP sites. Overall, recombinases have been shown to be useful for a variety of studies because of their ability to cause site specific integration or excision. The integration abilities of the recombinases may be used for gene therapy because of their ability to site-specifically integrate a sequence by simply placing a target site in the cell genome. It has also been theorized that by mutating the recombinase gene sequences without altering the conserved histidine, arginine and tyrosine residues, the site specificity of the recombinase may be changed to recognize conserved sequences of integrated viral genomes such as HIV as target sites and thus enabling it to excise the viral genome from cells expressing the recombinase. Recombinases have also allowed the production of tissue-specific knockout in transgenic mice as well as mimicking human diseases by inducing non-homologous chromosome translocations. Therefore recombinases have been shown to have a number of useful applications and have the potential to become an important tool for molecular biologists in the future. 69 R E F E R E N C E S Abremski, K., Hoess, R., and Sternberg, N. (1983) Cell 3 2 , 1301-1311. Abremski, K. and Hoess, R. (1992) Protein Engineering 5 , 87-91. Argos, P., Landy, A. , Abremski, K., Egan, J. B., Haggard-Ljunquist, E . , Hoess, R. H. , Kahn, M . L . , Kalionis, B, Narayana, S. V. L . , Pierson, L.S., III, Sternberg, N. and Leong, J. M . (1986) EMBO J. 5 , 433-440. Austin, S., Ziese, M . , and Sternberg, N. (1981) Cell 2 5 , 729-736. Baubonis, W. and Sauer, B. (1993) Nucleic Acids Res. 2 1 , 2025-2029. Barinaga, M . (1994) Science 2 6 5 (5168), 26-28. Bayley, C. C , Morgan, M . , Dale, E . C. and Ow, D. W. (1992) Plant Mol. Biol. 1 8 , 353-361. Bender, M . A., Miller, A. D., and Gelinas, R. E. (1988) Mol Cell Biol 8 , 1725-1735. Broach, J. R., Guarascio, V. R., and Jayaram, M . (1982) Cell 2 9 , 227-234. Cepko, C. L . , Roberts, B.E. , and Mulligan, R.C. (1984) Cell 3 7 , 1053-1062. Chou, T-B, and Perrimon, N. (1992) Genetics 1 3 1 , 643-653. Chu, G., Hayakawa, H. , and Berg, P. (1987) Nuc. Acids Res. 1 5 , 1311-1325. Dale, E . C , and Ow, D. W. (1991) Gene 9 1 , 79-85. Danos, O. and Mulligan, R.C.(1988) Proc. Natl. Acad. Sci. USA 8 5 , 6460-6464. Emerman, M . , and Temin, H . M . (1984) Cell 3 9 , 449-467. Fiers, W., Contreras, R., Haegeman, G., Rogiers, R., Van de Voorde, A., Van Heuverswyn, H. , Van Herreweghe, J., Volckaert, G. and Ysebaert, M . (1978) Nature 2 7 3 , 113-120. Fukushige, S. and Sauer, B. (1992) Proc. Natl Acad. Sci. USA 8 9 , 7905-7909. Gage, P. J., Sauer, B., Levine, M . and Glorioso, J. C. (1992) / . Virol. 6 6 , 5509-5515. Golic, K. G. (1991) Science 2 5 2 , 958-961. Golic, K. G., and Lindquist, S. (1989) Cell 5 9 , 499-509. 70 Gorman, C. M . , Padmanabhan, R. and Howard, B.H. (1983) Science 221, 551-553. Gorman, C. M . , Rigby, P. W., and Lane, D. P. (1985) Cell 42, 519-526. Gu, H. , Marth, J. D., Orban, P. C., Mossmann, H. , and Rajewsky, K. (1993) Cell 73, 1155-1164. Hartley, J. W. and Rowe, W. P. (1976) J. Virol. 19, 19-25. Hochman, L . Seger, N., Sternberg, N., and Cohen, G. (1983) Virology 131, 11-17. Hoess, R. and Abremski, K. (1985) J. Mol. Biol. 181, 351-362. Jainhill, J. L . , Aaronson, S. A. and Todaro, G. J. (1969) / . Virol. 5, 549-553. Korman, A.J., Frantz, J. D., Strominger, J. L . , and Mulligan, R. C. (1987) Proc. Natl. Acad. Sci. USA 84, 2150-2154. Lakso, M . , Sauer, B., Mosinger, B., Lee, E . J., Manning R. W., Yu, S.-H., Mulder, K. L . and Westphal, H. (1992) Proc. Natl. Acad. Sci. USA 89, 6232-6236. La Thangue, N. B., and Rigby, P. W. J. (1987) Cell 49, 507-513. Levy, J. A. (1974) Am. J. Clin. Pathol. 621, 258-280. Levy, J. A. (1973) Science 182, 1151-1153. Linney, E . , Neill, S. D., and Prestridge, D. S. (1987) / Virol 61, 3248-3253. Mack, A. , Sauer, B., Abremski, K. and Hoess, R. (1992) Nucleic Acids Res. 20, 4451-4455. Maeser, S. and Kahmann, R. (1991) Mol. Gen. Genet. 230, 170-176. Maniatis, T., Fritsch, E . F., and Sambrook, J. (1989) Molecular Cloning: A Laboratory Manual, Second Edition, Cold Spring Harbour Laboratory, Cold Spring Harbour, N Y . Mann, R. and Baltimore, D. (1985) J. Virol. 54, 401-407. Mann, R., Mulligan, R . C , and Baltimore, D. (1983) Cell 33, 153-159. Miller, A. D., and Buttimore, C. (1986) Mol. Cell. Biol. 6, 2895-2902. Miller, A. D. and Rosman, G. J. (1989) Biotechiques 7, (9), 980-990. O'Gorman, S., Fox, D. T., and Wahl, G. M . (1991) Science 251, 1351-1355. 71 Odell, J., Caimi, P., Sauer, B., and Russell, S. (1990) Mol. Gen. Genet. 223, 369-378. Onouchi, H. , Yokoi, K., Machida, C , Matsuzaki, H., Oshima, Y., Matsuoka, K., Nakamura, K., and Machida, Y. (1992) Nucleic Acids Res. 19, 6373-6378. Orban, P. C , Chui, D., and Marth, J. D. (1992) Proc. Natl. Acad. Sci. USA 89, 6861-6865. Overell, R. W., Weisser, K. E. , and Cosman, D. (1988) Mol. Cell Biol. 8, 1803-1808. Palmer, T. D., Hock, R. A., Osborne, W. R., and Miller, A. D. (1987) Proc. Natl. Acad. Sci. USA 84, 1055-1059. Peakman, T. C , Harris, R. A. and Gewert, D. R. (1992) Nucleic Acids Res. 20, 495-500. Reisman, D., and Rotter, V. (1989) Mol Cell Biol 9, 3571-3575. Rous, P.(1911) J. Exp. Med. 13, 397. Russell, S. H. , Hoopes, J. L . , and Odell, J. T. (1992) Mol. Gen. Genet. 234, 49-59. Sadowski, P. D. (1993) FASEB J. 7, 760-767. Sassone-Corsi, P., Fromental, C. and Chambon, P. (1987) Oncogene Res. 1, 113-119. Sauer, B. and Henderson, N. (1990) New Biologists 2, 441. Sauer, B. and Henderson, N. (1988) Proc. Natl. Acad. Sci. USA 85, 5166-5170. Senecoff, J. F., and Cox, M . M . (1986) / . Biol. Chem. 261, 7380-7386. Southern, E . M . (1975) / . Mol. Biol. 98, 503. Stark, W. M . , Boocock, M . R., and Sherratt, D. J. (1992) Trends Genet. 8, 432-439. Sternberg, N. and Hamilton, D. (1981) J. Mol. Biol. 150, 467-486. Stewart, C. L . , Stuhlman, H. , Hahner, D. and Janeisch, R. (1982) Proc. Natl. Acad. Sci. 79, 4098-4102. Stockshlaeder, M . A., Storb, R., Osborne, W. R. and Miller, A. D. (1991) Hum. Gene Ther. 2, 33-39. Stuhlman, H. , Jahner, D. and Jaenisch, R. (1981) Cell 26, 221-232. Takeda, S., Hamaguchi, Y. , Zong, S. Q., Kuze, K., Honjo, T., Ishimoto, M . , Nakano, T., Kitamura, Y. (1988) Growth Factors 1, 59-66. Thomas, K. R. and Cappecchi, M . R. (1987) Cell 51, 503-512. van Deursen, J., Fornerod, M . , van Rees, B. and Grosveld, G. (1995) Proc. Natl. Acad. Sci. USA 92, 7376-7380. Wagner, E . F., Vanek, M . , and Vennstrom, B. (1985) EMBO J 4, 663-666. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0098997/manifest

Comment

Related Items