DK174253B1 - Vaccinia virus modified with exogenous DNA in vaccinia genome - useful for vaccination of animals to form antibodies to coded antigens - Google Patents

Abstract

Vaccinia virus modified by the presence of exogenous DNA in the vaccinia genome is new. Biologically pure cultures of vaccinia virus VTK-79 (ATCC VR 2031) and VTK-79L are new. Immunisation of a host animal susceptible to vaccinia virus is effected by inoculation with a vaccinia virus having, within the vaccinia genome, DNA exogenous to the genome and coding for an antigen that will induce the animal to develop antibodies against it. Method for the in vivo recombination of a vaccinia virus and donor DNA comprises contact of a cell monolayer with a cell-compatible medium contg. the vaccinia virus and with a similar medium contg. the donor DNA. The donor DNA comprises DNA to be introduced into the vaccinia genome present in a DNA segment otherwise co-linear with DNA present in the genome of the virus in a non-essential region of the genome. Vaccines contg. the modified virus can be used to immunise mammals so that an approp. antibody response is elicited. Lukaryotic cells of biological prods. other than antigens can be produced. Human and animal individuals and populations may have missing genes or genetic material (for modification, replacement or repair of defective genes) introduced.

Description

DK 174253 B1

The present invention relates to a method of producing a biological product wherein a culture of eukaryotic cells is infected with a recombinant cooktop virus and the method of the invention is characterized in that the infecting virus in a non-essential region of the cooktop genome contains a DNA segment. which do not occur naturally in coconut viruses and which are expressed in infected eukaryotic cells or host organisms and encode the biological product, and that the infected cells are cultured under conditions that allow expression of the biological product, after which the biological product is isolated from the cell culture . The invention further relates to a modified recombinant cupric virus (vaccinia virus) for use in the method.

10 Cockroach viruses are the prototypical viruses of the smallpox virus family, and like other members of the smallpox virus group, it is prominent because of its large size and intricate nature. The DNA of the coconut virus is also large and complex. Coconut DNA is approx. 120 megadaltons in size, e.g. compared to a DNA size of only 3.6 megadaltons for anthropoid virus 40 (or simian virus 40 (SV40)). The DNA molecule of the coconut virus has a double strand and is ultimately cross-linked so that a single strand circle is formed by DNA denaturation. Kokoppe DN A is physically drawn in detail using a number of different restriction enzymes, and a number of such cards are presented in an article by Panicali et al., J. Virol. 37,1000-1010 (1981), which reports that there are two important DNA variants of the CO-strain virus WR strain (ATCC No. VR 20 119), which strain has been extensively used in the detection and characterization of small-pox viruses. The two variants are different in that the S ("smaH") variant (ATCC No. VR 2034) has a 6.3 megadalton annihilation that does not occur in the DNA of the L ("large") variant (ATC No. VR 2035). Schematic maps obtained by treating the variants with the restriction enzymes Hind III, Ava, I, Xho I, Sst I and Sma I 25 are presented in the aforementioned article.

Cuckoos, a eukaryotic virus, are completely regenerated within the host cell cytoplasm.

It is a lytic virus, ie a virus whose duplicate in a cell results in cell lysis. The virus is considered to be non-oncogenic. The virus has been used for approx. 200 years in vaccines for vaccines against smallpox and doctors are familiar with the properties of the virus when used in a vaccine. Although smallpox vaccination is not without risk, however, the risks are largely well-known and defined, and the virus is considered relatively benign.

5 Sam. et al., Ann. Virol. (Institut Pasteur) 1981, 132E, 135-150, the descriptive method of genome rescue, wherein propagation of rabbit pox virus (RP virus) is achieved by introducing DNA from RP virus into cells infected with ectromelia virus under conditions with special temperatures. Likewise, genomic "rescue" of RP virus was obtained using cocoon virus ts mutants as the co-infecting virus instead of ectromelia. Since RP, 10 ectromelia and coopers are members of the same class of viral compounds, namely orthopoxviruses, with functionally similar components, it is not unexpected that such coinfection with either ectromelia or coopers could restore replication of the RP virus.

In the studies of Sam. et al. It is seen that the recombination between RP DNA and the co-infected virus is minimal. Further, the nature of the recombination was not determined.

In contrast, the present invention discloses a method by which exogenous DNA can be recombined in a co-op virus and express the gene product of this DNA, where the gene product is not necessarily of viral nature. Thus, the genome of the coconut virus is altered, the coconut virus essentially merely serving as a cloning vector. Recombination is accomplished through a series of cuts, ligations, and amplifications involving a particular segment of the coconut virus, the exogenous DNA of interest, and a cloning plasmid for DNA amplification. The resulting altered coconut virus segment is reintroduced into the coconut virus and can subsequently be introduced into eukaryotic host cells to express the gene product. In addition, the cupping virus used is capable of replication, but it is non-infectious, and therefore the method of the invention 25 does not include virus production but more precisely gene product expression and propagation.

The core of the present invention is the modification of the naturally occurring co-cup genome to produce co-coop mutants by rearrangement of the natural genome, by removal of DNA from the genome, and / or by inserting DNA into the naturally-occurring co-coop genome that caused the breakdown. in the naturally occurring genome ("foreign DNA"). Such a foreign DNA may occur naturally in coopers or it may occur synthetically or naturally in an organism other than coopers. If genetic data is present in this foreign DNA, then there is a potential for insertion of this data into a eukaryote using modified coconut virus.

This discovery has a number of useful consequences, among which are entirely new methods of producing biological products other than eukaryotic cell antigens.

Appropriately modified coco mutants carrying exogenous genes expressed in meadow host as an antigenic determinant that in the host triggers the production of antigens to the antigen denote brand new vaccines that avoid the disadvantages of traditional vaccines using killed or debilitated live organisms. Thus, for example, production of vaccines from killed organisms requires the growth of large quantities of organisms, followed by a treatment that will selectively destroy their infectiousness without affecting their antigenicity. On the other hand, vaccines containing attenuated living organisms always present the possibility of the attenuated organism's return to a pathogenic state. In contrast, when a modified coconut mutant suitably modified with a gene code for an antigenic determinant of a disease-producing organism is used as a vaccine, the possibility of return to a pathogenic organism 20 is avoided since the coconut virus contains only the gene code for it. the antigenic determinant of disease-producing organisms and not the genetic portions of the organism responsible for the reproduction of the pathogen.

The present invention even offers advantages in novel technology employing genetic engineering which results in the production of an antigen by a recombinant prokaryotic organism containing a plasmid expressing a foreign antigenic protein. Such technology requires e.g. production of large amounts of the recombinant prokaryotic cells and subsequent purification of the antigenic protein produced thereby.

In contrast, a modified goblet virus used for grafting in accordance with the present invention is reproduced within the grafted individual to be immunized, thereby enhancing the antigenic determinant in vivo.

A further advantage of using coconut mutants as vectors in eukaryotic cells as vaccines or for the production of biological products other than antigens is the possibility of rearranged modifications of proteins by transcription of exogenous genes introduced into the cell of the virus. Such retrofitted modifications, e.g. glycosylation of proteins is not likely in a prokaryotic system, but possible in eukaryotic cells where increased enzymes needed for such modifications are available.

A further advantage of the use of cocoon mutants for grafting is the enhancement potential of the antagonist response by incorporation into the mutant of double-acting repeats of the antigen gene or of enhanced genetic elements that stimulate the immune response, or by the use of a strong activator in the modified virus. A similar advantage applies in the production of biological products other than antigens.

15 In order to return to a more detailed consideration of the coca genome, the DNA's cross-linked double strands are characterized by inverted end repeats, each of approx. 8.6 megadaltons in length, representing approx. 10 kilobase pairs (kbp). Since all of the smallpox virus central portions of the DNA are similar to each other, while the final portions of the viruses differ more strongly from one another, the central portion is responsible for functions such as replication, which is presumably common to all viruses, whereas the final portions appear to be responsible. for other characteristics such as pathogenicity, host region, etc. If such a genome is to be modified by rearranging or removing DNA fragments therefrom or introducing exogenous DNA fragments into it while producing a stable viable mutant, it is obvious that that portion of the naturally occurring DNA, which is rearranged, removed, or split upon introduction of exogenous DNA into it, must not be of significant importance to the viability and stability of the host, in this case the coconut virus. Such insignificant portions of the genome were found to be present in the coccyx strain of the virus, eg within the region present within the L variant but deleted from the S variant or within the Hind III genome. F-fragment.

The modification of the coconut virus in the incorporation of exogenous genetic data can be illustrated by the modification of the coconut virus WR strain in the Hind ΙΠ F fragment thereof 5 to incorporate a herpes simplex virus type I (HSV) gene responsible for the production of timidine kinase (TK) into that fragment. TK is an enzyme that phosphorylates nucleoside-timidine to form the corresponding monophosphorylated nucleotide, which is later incorporated into the DNA.

The HSV TK gene designates foreign DNA for the Cuckoo virus, which is convenient to insert 10 into cockscapes for several reasons according to the present invention. First, the gene is relatively readily available and present in a herpes simplex virus DNA fragment produced by solution with Bam HI endonuclease, as reported by Colbere-Garapin et al. in Proc. Natl. Acad. Sci. USA 76 3755-3759 (1979). Second, this HSV Bam HI fragment has been introduced into plasmids and into eukaryotic systems in previous practice, e.g.

15 as reported by Colbere-Garapin et al. ext. cit., and by Wigler et al., Cell 11, 223-232 (1977).

Third, experience has shown that if HSV TK can be introduced as an exogenous gene into a eukaryotic system and expressed - which requires unambiguous and reliable translation and transcription of the genetic locus - then other exogenous genes can be introduced and expressed in a similar way .

A better understanding of the present invention can be obtained by considering the attached drawings, in which

Fig. 1 is a map of the aforementioned L and S variants of a particular coco-WR strain, using Hind ΙΠ as a restriction enzyme and showing the deletion of sequences in the final C fragment and the L variant if deletion is outside the genome's final repeat section. The deleted DNA sequences are unique to the L structure, and since the growth of the S and L variants is exactly the same, this deleted region must be immaterial; 6 DK 174253 B1

FIG. 2 shows in greater detail the coconut Hind III F fragment, including two additional restriction sites therein, namely Sst I and Bam HI, of which at least the last site provides a locus on which exogenous DNA can be introduced into coconut Hind III F - the fragment without disrupting any significant coking genes; Figures 3A-C schematically show a method for introducing the HSV TK gene into the coconut Hind III F fragment;

FIG. Figure 4 is a restriction map of certain coco mutants produced in accordance with the present invention, and it shows in detail the position of the HSV TK insertions present in the Hind III F fragment in two such virus mutants, herein designated VP-110 and VP-1. 2;

FIG. 5 is a table summarizing certain techniques useful in detecting possible reuniting viruses to determine the presence or absence of the HSV TK gene therein; and

Figures 6 A-C are restriction maps of the left end portion of the coconut WR genome, showing the ratio of various restriction fragments to the unique L-variant DNA sequence deleted from the corresponding S-variant.

Figures 7 AC schematically show a method of constructing a new plasmid pDP 120 containing a portion of the Coke-Hind III-F fragment associated with pBR 322, but which plasmid has a lower molecular weight than plasmid pDP 3 , shown in FIG. 3B, 20 as shown above.

Figures 8 A-D schematically show the construction of two plasmids pDP 301A and pDP 301B, which allow the incorporation of the DNA sequence of pBR 322 into the coconut virus to produce coconut mutants VP 7 and VP 8.

7 DK 174253 B1

Figures 9A-C schematically depict the construction of a virus mutant VP 9 into the genome from which the influenza hemagglutin gene (HA) has been incorporated using a technique such as that shown in Figures 8A-D.

FIG. Figure 10 shows schematically the construction of a further coconut mutant VP 10, which also contains 5 influenza hemagglutinin gene (HA), but prepared directly by in vivo recombination, using VP 8.

Figures 11AE show the construction of two coconut mutants, called VP 11 and VP 12, each containing the DNA sequence code in their genome for the hepatitis B virus surface antigen incorporated therein by in vivo recombination of coconut virus VTK "79 with 10 and recently constructed plasmids, respectively. pDP 250B and pDP 250A.

Figures 12 AD show schematically the construction of a new plasmid pDP 252 combining pBR 322 with a portion of the hepatitis B virus (HBV) genome, which is completely within the region of the HBV region genome encoding the surface antigen. . The figures show the introduction of the resultant pDP 252 plasmid into the coconut mutant VP 8 with the resulting formation of a further coconut variant identified as VP 13..

Figures 13 AC show the construction of two more plasmids pBL 520A and pBL 520B and their insertion into VP 7 to produce two additional coco mutants VP 16 and VP 14, each containing the DNA sequence of herpes virus I encoding the production of herpes glycoproteins gA + gB, two of the major antigenic proteins of the herpes simplex-20 virus types I and II.

Figures 14 A-C show the construction of two more plasmid pBL 522A and 522B incorporating the 5.1 md Barn HI fragment G shown in Figs. 13A as present in the Eco RI herpes F fragment (De Luca et al., Virology 122,411-423 (1982)).

S

DK 174253 B1

Figures 15 A-F show the construction of a further 10 coconut variant VP 22, in which foreign DNA, namely a herpes Bgl / Bam TK fragment, has been inserted into the coconut genome in a non-essential portion other than the F fragment thereof.

Regarding FIG. 1, if the L and S variants of the coconut virus are exposed to the effect of 5 Hind III, a restriction enzyme well known in the prior art, and which can be commercially available, the virus genomes are cleaved into 15 or 14 segments denoted by the letters A to 0, respectively. the letter A is used to denote the largest fragment and the letter 0 is used to denote the smallest. The electrophoretic rupture of the restriction fragments is described and shown in the aforementioned publication by Panicali et al., J. Virol. 37, 1000-1010 (1981). The F fragment thus obtained either from the L or S variants has a molecular weight of 8.6 megadaltons. The location of the F fragment is shown on the restriction map, presented as Figs. 1 and enclosed with the patent application, and a restriction map of the F fragment is shown in FIG. 2. The restriction enzyme Hind III recognizes the nucleotide sequence -AAGCTT- and cleaves the DNA between the adjacent adenosis groups to yield fragments having "sticky ends" with the AGCT sequence. Since larger quantities of the cocop Hind III F fragment than can be readily obtained by capping the cocop pene genome are required for manipulation according to the present invention, the F fragment is introduced into a cloning plasmid vector for amplification purposes.

*

Namely, the coconut Hind III F fragment produced in this way is readily introduced into plasmid pBR 322, which is cut only once by a number of restriction enzymes, including Hind III. The plasmid pBR 322 was first described by Bolivar et al. in Gene 2.95-113 (1977) and is now commercially available in the United States from several sources.

The site of cleavage of Hind III on the pBR 322 plasmid is indicated in Fig. 3A relative to the cleavage sites of Eco RI and Barn HI, which are other restriction enzymes. If 25 pBR 322 plasmid is cut with Hind III, and the resulting cleaved DNA is mixed with cocoon Hind-III-F fragment and if the fragments are cut off with T4 DNA ligase, as suggested in Figs. 3A, the F fragment is incorporated into the plasmid to produce the all-new 9D 174173 B1 plasmid pDP 3 shown schematically in FIG. 3B having a molecular weight of approx. 11.3 megadaltons. The coconut Hind III-F fragment comprises ca. 13 kilobase pairs compared to the 4.5 kilobase pairs found in pBR 322 portion of pDP 3. T4 DNA ligase is a commercially available enzyme and the conditions for its use in the manner indicated are well known in the art.

The plasmid pDP 3 is now introduced into a microorganism such as Escherichia coli (E. coli for the purpose of converting the Hind III F fragment to recover larger amounts of the F fragment. These techniques for cleavage of a plasmid to produce linear DNA having clipping terminals and inserting exogenous DNA having complementary clamps to produce a replicon (in this case, pBR 322 which contains cow-cup Hind-III-F fragment) are known in the art, as well as the insertion of the replicon into a mk innate organism by conversion (cf. US Patent 4,237,224).

Unmodified pBR 322 plasmid assigns ampicillin resistance (AmpR) and tetracycline resistance (TetR) to its host microorganism, in this case E. coli.

However, as Hind III intersects the pBR 322 plasmid in the TetR gene, the introduction of the 15 coconut Hind-III-F fragment destroys the TetR gene and tetracycline resistance is lost. As a result, E. coli transformants. containing the pDP 3 plasmid is distinguished from untransformed E. coli by the simultaneous presence of resistance to ampicillin and susceptibility to tetracycline. It is this E. coli transformed with pDP 3 that is grown in large quantities and from which large amounts of pDP 3 are recovered.

The conditions under which plasmids can be amplified in E. coli. are well known in the art, e.g. from Clewel's writing, J. Bacteriol. 110,667-676 (1972). The techniques for isolating the amplified plasmid from the E. coli host are also well known in the art and are described e.g. by Clewel et al. in Proc. Natl. Acad. Sci. USA 62, 1159-1166 (1969).

Similarly, the pBR 322 plasmid can be conveniently cleaved by treatment with the restriction enzyme Barn HI and a modified plasmid can be prepared by insertion into the Bam HS V TK fragment, as discussed above. work by Colbere-Gara-pin et al., loc. cit. The modified plasmid containing the Barn Hi fragment comprising the HSV TK gene can again be introduced into E. coli by known methods and the transformed bacteria grown in large quantities for the amplification of the plasmid. The amplified Barn HSV 5 TK-pBR 322 reuniting plasmid was later cleaved with Bam HI to isolate the Bam HI fragment containing the HSV TK gene using the same phage techniques mentioned previously with respect to the amplification of the Hind III F -fragment.

In order to construct a reuniting plasmid having the Bam HI HSV TK fragment included within the coconut Hind III F fragment, the pDP 3 plasmid is then subjected to a partial restriction with Bam HI such that only one of the two Bam HI cleavage sites within the plasmid are cleaved, i.e. either the Bam Hi site within the Hind III F fragment or the Bam Hi site within the p BR 322 portion of the pDP 3 plasmid, as shown in Figs. 3B. The cleaved, now linear, DNA is then linked to the purified Bam HSV TK fragment.

The linear segments are interconnected and bound by treatment with T4 DNA ligase, again using techniques known in the art.

The combination of the Bam HSV TK fragment with the cleaved pDP 3 plasmid is any random or statistical event leading to a possible production of numerous species, formed by different fragment compounds present in the mixture, all of which have exactly identical "sticky ends". Thus, one possibility is the simple reunification of Bam His cleaved 20 ends of pDP 3 plasmids to again form the circular plasmid. Another possibility is the association of two or more Bam HSV TK fragments in any of two orientations. Furthermore, the Bam HSV TK fragment (or multiple thereof) can be combined with the linear DNA of a pDP 3 plasmid cleaved at the Bam HI site within the pBR 322 portion, again in either of two orientations, or one or more Bam HSV TK 25 fragments can be combined, again in any of two orientations, with linear pDP 3 DNA, which has been cleaved at the Bam HI site within the coco Hind-III-F fragment portion of the pDP 3 plasmid.

11 DK 174253 B1

To allow the identification and separation of these various possibilities, the conjugate products are inserted into a single cell microorganism such as E. coli. with techniques such as those previously described and known in the art. E. coli. treated in this way is then grown on a substrate containing ampicillin. The bacteria that contain plasmid are resistant to ampicillin because all such plasmids contain the gene of pBR 322 that confers resistance to ampicillin. Accordingly, all surviving bacteria are transformants which are then further sorted to determine the presence or absence of the possibly present Bam HSV TK fragment.

To complete it, the bacteria containing any TK gene are identified at 10 junction with radio-labeled TK DNA. If the TK gene is present in the bacterium, the radio-labeled TK DNA hybridizes with the portion of the plasmid present in the bacterium. Since the hybrid is radioactive, the colonies containing TK within their plasmids can be determined by autoradiography. The bacteria containing TK can be re-cultured. Finally, the bacteria containing plasmids with TK incorporated before 15 in the pBR 322 portion can be identified and separated from those having the TK fragment in the coconut Hind III F fragment by analysis with restriction endonucleases.

In more detail, the bacteria that survive growth on nutrient agar plates containing ampicillin are partially transferred to a nitrocellulose filter upon contact of the filter with the plate. The bacteria left on the plate are re-cultured and the bacteria transferred to the nitrocellulose filter to reproduce the original plate are then processed to denature their DNA. Denaturation is performed e.g. by treating the transferred bacteria with sodium hydroxide with subsequent neutralization and washing. Next, the now denatured DNA present on the nitrocellulose filter is hybridized by treatment with HSV Barn TK containing radioactive 32 P. The nitrocellulose filter, treated in this way, is then subjected to X-ray film which darkens in the parts in which hybridization with radio-labeled Bam HSV TK has occurred. The exposed darkened X-ray film is then compared with the original plate and the colonies growing on the original plate and the colonies causing darkening of the X-ray film are identified as those containing a plasmid in which Bam HSV TK is present.

Finally, to distinguish between the bacteria that contain a plasmid in which the Bam HSV TK gene has been incorporated within the pBR 322 portion of the plasmid, from those in which 5 Bam HSV TK are present in the F fragment of the plasmid, small cultures are grown. cultures of the bacteria, and the plasmids are isolated therefrom by a mini-light technique known in the art and described in the article by Holmes et al., Anal. Bioch. 19 4 193-197 (1981). Then, the plasmids with the restriction enzyme Hind III are dissolved, which cleave the circular plasmid at the two points of the original F fragment's original association with the pBR 322 DNA chain. The molecular weight of the solution product 10 is then determined by electrophoresis on agarose gels, with the distance of movement in the gels as the tape measure of the molecular weight.

If the Bam HSV TK fragment or multiple thereof is found in the F segment of the resolved plasmid, the gel will show the presence of the pBR 322 fragment plus another fragment having a molecular weight greater than the F fragment following the molecular weight of the Bam HSV 15 TK DNA. segment or segments included therein. Conversely, if Bam HSV TK is present in pBR 322, the electrophoresis will show the presence of an F fragment of the usual molecular weight plus an additional fragment larger than pBR 322 after the molecular weight of the Bam HSV TK fragment or fragments present therein. The bacteria in which modification with Bam HSV TK occurred in the pBR 322 portion of the plasmid were discarded: The 20 remaining bacteria were modified in the plasmid's F fragment portion therein. It is these plasmids that are used for incorporation of the Bam HSV TK fragment into the coop.

As previously mentioned, the joining of DNA fragments to regenerate a plasmid is a random event, according to which a number of different plasmid structures having Bam HSV TK in the F fragment can result.

In order to determine the orientation of the Bam HSV TK fragment within the F fragment, as well as the number of such Bam HSV TK fragments that may be present, the plasmids from any of the bacterial colonies that are known to have a Bam HS V TK fragment present in the F fragment of the plasmid. The mini-light technique, previously mentioned herein, is used for this purpose. The plasmids are then again subjected to restriction analysis, this time using the restriction enzyme Sst I, which is commercially available. Since each Bam HSV TK-5 fragment has an Sst I restriction site therein, and since the F-fragment of the coop also has one single Sst I restriction site therein (cf. the preparation of these fragments, respectively, in Figures 3A and 3B), different fragment numbers with different molecular weights are detected by electrophoresis on agarose gels, the number of segments and molecular weight are dependent on the orientation of the Barn HSV TK fragment within the F fragment and the number of such Barn TK fragments present. Orientation of the Barn TK fragment within the F fragment can be detected due to the asymmetry of the Barn HSV TK fragment with respect to file Sst Instead thereof (cf. Fig. 3B).

For example, in the particular experiments in question, six bacterial colonies, each containing one or more Barn HSV TK fragments present in the F fragment of the plasmid, were found among the E. coli transformants. Following the restriction analysis of the plasmids in these bacteria, in accordance with what is discussed above, two of the reunited plasmids were selected for further study because the orientation of the HS HS TK fragment within the F fragment was in opposite directions.

At this point, we would like to remind the reader that the introduction of the HSV TK gene into the 20 F fragment of the coke, as discussed in detail above, is merely to serve as an example of one of many possible means of modifying the coke genome to produce desirable co pemutanter. In this way, the introduction of the same exogenous gene into a different portion of the coco genome, or the introduction of different genetic material into the coco-pe-F fragment or into another fragment, may all require modification of the plan that became discussed above, and serves as an example, for the identification of reunited organisms.

14 DK 174253 B1

For example, dissolution of the Co-L-variant with Ava I yields a fragment H, complete with the region deleted from the S-variant (cf. Fig. 6A and the discussion thereof below).

This H fragment contains Barn Hi sites that allow entry into the HSV TK gene. The same plan for identifying F-fragment-HS V TK reunifiers can be used to also identify such H-fragment reunifiers.

Indeed, plans for the construction and identification of F-fragment HSV TK recombinants, or reunite, alternatives to what is presented in detail above by way of illustration exist. For example, the Barn Hi site of pBR 322 can be removed by cleavage of the plasmid with Barn HI and treatment with DNA polymerase I to "fill" the 10 "sticky ends". This product is then cut with Hind III, and the linear fragment is treated with alkaline phosphatase to prevent the plasmid reflux by latching. However, foreign DNA, and especially the Hind III F fragment, can be laced to the treated pBR 322, and the resulting plasmid will recycle. It should be noted that treatment with Barn HI has effect only on the cleavage of the plasmid within the coco-F fragment portion thereof.

The following treatment of the cleavage product with alkaline phosphatase and binding to the Barn HI HSV TK fragment will produce high efficiency recombinants so that the recombinants can be sorted by restriction endonuclease cleavage and gel electrophoresis. This technique eliminates time-consuming work in distinguishing recombinants having HSV TK in the pBR 322 portion or in the F fragment and colony crossing.

We now turn ten to ten further discussion of the plasmids produced in the example of co-co-permutation by introducing HSV TK into the co-co-F fragment, and the two recombinant plasmids selected for further study are shown in Figs. 3C, where they are identified, as a first whole new plasmid pDP 132 incorporating one Bam HSV TK fragment within the Coke-Hind III-F portion, and a second whole new plasmid pDP 137, 25 in which two HSV TK fragments united "head to tail" have been incorporated. The single fragment of Barn HSV TK has been incorporated within pDP 132 in the opposite direction in which two Barn TK fragments have been included in tandem in pDP 137. Namely, the TK gene region within the Barn HI fragment encoding the 5 'end of the gene produced by the gene is located between the Sst I cleavage site and the closer of the two Barn HI sites thereto (again, see Fig. 3B). The transcriptional direction of the HSV TK gene on the Barn TK fragment goes from the 5 'end to the 3' end, and in pDP 132 it will be clockwise as shown in Figs. 3C. (cf. Smiley et al., Virology 102, 83-93 (1980)). Conversely, since the Barn TK fragments included in tandem in pDP 1.37 have been included in the opposite direction, the transcription of the HSV TK genes contained therein will be in the opposite direction, namely in a counterclockwise direction. The Bam HSV TK fragment direction of inclusion within the coccyx Hind III F fragment may be of importance in the case of transcription promotion of the HSV TK gene initiated by a promoter site within the F fragment itself. However, 10 HSV promoter sites exist within the Barn HSV TK fragment itself, so that transcription of the HSV TK gene may occur, no matter in which direction the Barn HSV TK fragment gene and the HSV TK gene have been incorporated within the coconut Hind II-F fragment.

The E. coli transformants. containing pDP 132 or pDP 137 is then grown to produce large amounts of plasmid further processing. When a sufficient amount of the plasmid DNA has been isolated, restriction with Hind III yields a modified coconut Hind III F fragment carrying the HSV TK gene therein. This modified Hind-III-F fragment is now introduced into the coconut virus by quite new methods, described more precisely below, to produce an infectious entity.

To provide an overview of past practice, currently the vector is mainly used to insert exogenous DNA into eukaryotic cells, S V40. The DNA of S V40 is circular and can be treated approximately as a plasmid. That is, the circular DNA is cleaved with a restriction enzyme, combined with exogenous DNA and bound. The modified DNA can be introduced into eukaryotic cells, e.g., animal cells, by standard techniques (cf. Hamnier et al., Nature 281.35-40 (1979)). The DNA is infectious and will replicate in the nucleus of the cell and produce viable mutated viruses. In contrast, cuckoos replicate within the cytoplasm of the eukaryotic cell. The purified DNA of this virus is not infectious and cannot be used by itself to produce coco mutants in a cell in the same manner as SV40. Rather, completely new techniques associated with the mutation of wild-type cooks with foreign DNA in vivo within a cell must be used.

An unpublished document by the applicants, together with Eileen Nakario, reports on a demonstration of marker rescue in the smallpox virus. In accordance with these experiments, the portion of the L-variant DNA, which is normally absent in the S-variant, can be reintroduced into the S-variant ("rescued") under proper conditions. Namely, eukaryotic cells are treated with live infectious, S-variant coconut virus, together with non-infectious L-variant DNA fragmentation fragments, which denote DNA "foreign" to the S-variant, of a particular structure. Namely, the portion of the L variant DNA to be rescued must be present within a DNA chain having portions which are colinear to the S fragment DNA chain into which it is to be inserted. That is, the "foreign" DNA to be introduced into the S variant has a region of DNA that is homologous to similar sequences in the S variant at both ends of the DNA chain. These homologous sequences can be considered as "arms" attached to the L-variant DNA region, which must be rescued by the S-variant.

This mechanism of recombination is complex and has not been implemented yet in vitro. Obviously, L-DNA recombination into the S variant results in homologous base pairing in segments surrounding the region deleted from the S variant. Most likely, crossings from one strand of DNA to another result in an in vivo DNA recombination to rescue the deleted portion.

This technique of in vivo recombination can be used to introduce foreign DNA, other than cocoon DNA, into either cocoon S or L variant. In this way, the modified Hind III F fragment incorporating the Bam HS V TK fragment therein as DNA "foreign" to the coop can be introduced into the coop by treating eukaryotic cells with the modified F fragment together with infectious L and / or infectious S variants of the cockroach virus. In this case, the portions of the F fragment flanking the Barn HSV TK fragment function as the aforementioned "arms" which include DNA homologous to DNA present in the L or S variant into which the modified F fragment must be introduced.

17 DK 174253 B1

Again, by in vivo processes within the cell whose mechanisms are not known in detail, the HSV TK-modified F fragment is incorporated into the coconut variants in the cell and is then susceptible to replication and expression under coconut control.

This in vivo recombination technique is broadly applicable to the introduction of yet another "foreign" DNA into the coop, thereby smoothing the path so that the coke genome can be modified to incorporate a large variety of foreign genetic material into it, whether such foreign DNA originates from the coop itself, is synthetic, or comes from organisms other than coopers.

A large variety of cells can be used as host cells in which the above described in vivo in vivo recombination takes place. However, the recombination occurs with divergent efficiency, depending on the cell used. Of the cells examined to date, Syrian baby hamster ovarian cells (BHK-21 (Clone 13) (ATCC No. CCL10)) have been shown to be most effective for the recombination procedure. However, other cells including CV-1 (ATCC No. CCL70), a green monkey kidney cell line, and human (line 143) TK cells, a 5'-15 BUdR, resistant mutant that derives for human cell line 8970-5, have also been infected with this way to generate coconut mutants. ·,

These cells are treated in a suitable manner with cuprins and the foreign DNA to be incorporated into the cuprican, while the cells have a monolayer form for convenience.

For in vivo recombination purposes, the cells must be infected with cooks with subsequent treatment of foreign DNA to be incorporated therein, or they may first be contacted with the foreign DNA with subsequent cooktop infection. As a third alternative, the coop and foreign DNA must be present simultaneously at the time the cells are processed.

Viruses are suitably contacted with the monocell layer while present in a traditional culture fluid such as phosphate buffered saline, Hepes buffered saline, 18 DK 174253 B1

Eagle's Special culture fluid (with or without serum addition), etc., which is compatible with these cells and viruses.

The foreign DNA is used in a convenient way to treat these cells while in a calcium phosphate precipitate form. Such techniques for introducing DNA into cells have been described in previous practice by Graham et al., Virology 52,456-467 (1973). Modifications of the technique have been discussed by Stow et al., J. Gen. Virol. 33,447-458 (1976) and Wigler et al., Proc. Natl. Acad. Sci. USA 76, 1373-1376 (1979). The treatments taught in these writings take place comfortably at room temperature, but temperature conditions can be varied within limits that maintain cell viability, as can the time in which cells are treated with the virus and / or foreign DNA precipitate. , with various in vivo recombination efficiencies. The concentration of the infecting calyx virus and the amount of foreign DNA precipitate used will also affect the recombination rate or degree. Other factors, such as atmosphere and the like, are all chosen to maintain cell viability. Otherwise, as long as the three necessary components (cell, virus and DNA) are present, in vivo recombination will continue, at least to some extent. Optimization of conditions in a particular case is well within the capabilities of a person skilled in the microbiological field.

Following this recombination step, the cupric viruses that have been mutated by in vivo recombination must be identified and separated from unmodified cupric virus.

Coconut virus, mutated by in vivo recombination of foreign DNA therein, can be separated from unmodified coconut virus by at least two methods which are independent of the nature of the foreign DNA or the mutant's ability to express any gene that may be present in the foreign DNA. In this way, the foreign DNA of the mutant genome can first be detected by restriction analysis of the genome to detect the presence of an additional 25 pieces of DNA in the mutated organism. In this method, individual viruses isolated from purified plaques are grown and the DNA is extracted therefrom and subjected to restriction analysis using appropriate restriction enzymes. Again, by detecting the number and molecular weight of the fragments determined, the structure of the genome before the restriction can be deduced. However, due to the necessity of cultivating purified plaques, the number of analyzes to be performed and the possibility that none of the cultivated and analyzed plaques will contain a mutant, this technique is laborious, time consuming and uncertain.

Furthermore, the presence of foreign DNA in the coconut virus can be determined by using a modification of the technique described by Villarreal et al. teaches in Science 196. 183-185 (1977). Infectious virus is transmitted from virus plaques present on an infected monocell layer to a nitrocellulose filter. Conveniently, a mirror image replica of the transmitted virus present on the nitrocellulose filters is made by contacting another such filter with the side of the first nitrocellulose filter to which viruses have been transmitted. One portion of viruses present on the first filter is transferred to the second filter. One or the other of the filters, usually the first filter, is now used for hybridization. The remaining filter is reserved for the recovery of recombinant virus from there, once the recombinant virus locus has been detected using the hybridization technique practiced on the pendant, mirror image filter.

For hybridization purposes, viruses present on the nitrocellulose filter are denatured with sodium hydroxide in a manner known per se. The denatured genetic material is now hybridized with a radio-labeled counterpart of the gene whose presence is sought to be determined. For example, to detect the possible presence of coconut mutants containing the Bam HSV TK fragment, the corresponding radio-labeled Bam HSV TK fragment containing 32P is used in much the same way as discussed earlier herein for the detection of plasmids, modified by the presence of this fragment. Non-hybridized DNA is washed from the nitrocellulose filter and the remaining hybridized DNA which is radioactive is found by autoradiography, i.e. by contacting the filter with X-ray film. Once the mutated viruses have been identified, the corresponding virus plaques present on the second filter containing a mirror image of the viruses transferred to the first filter are found and cultured for the replication purposes of the mutated viruses.

2 ° DK 174253 B1

The two methods described above entail an analysis of the genotype of the organism involved and, as previously mentioned, they can be used, whether or not any gene present within the foreign DNA incorporated into the coconut virus is expressed. However, if the foreign DNA is expressed, then phenotypic analysis can be used for the detection of 5 mutants. For example, if the gene is expressed in the production of a protein to which a antibody exists, the mutants can be detected by a method using the formation of antigen-antibody complexes. See Bieberfeld et al., J. Immunol. Methods 6,249-259 (1975).

That is, plaques from viruses, including the suspicious mutants, are treated with antidote to the protein produced by the mutant coconut genotype. Excess counterpart 10 is washed from plaques, which are then treated with protein A labeled with 125 I. Protein A has the ability to bind the heavy antibody chains and, consequently, will specifically label the remaining antigen-antibody complexes on the monocell layer. After excess radioactive protein A is inhibited, the monolayers are re-assembled by pilaquelifts onto nitrocellulose filters and subjected to radiography to detect the presence of the radio-labeled immune complexes. In this way, the mutated coco viruses that produce the antigenic protein can be identified.

In the particular case, in which the foreign DNA comprises the HSV TK gene, once it is known that the mutated coconut virus expresses the HSV TK gene therein, a much simpler and elegant means of detecting the presence of the gene exists. Indeed, the ease of distinguishing between cupric mutants containing the HSV TK gene and unmodified cuprins free of this gene provides a powerful tool to distinguish the cupric virus mutant containing other exogenous genes that are present only in the cupric genome or present therein in association with the HSV TK gene. These methods are described in more detail hereinafter.

25 Since eukaryotic cells have their own TK gene, and similarly, the coconut virus hardens its own TK gene (used, as previously noted, for incorporating timidine into DNA), the presence and expression of these genes must be distinguished in some way from the presence and expression of the HSV TK gene in cocoon mutants of the type in question. In order to do this, the fact that the HSV TK gene will phosphorylate halogenated desoxycytidine, especially iododesoxycytidine (IDC), a nucleoside, but neither the coconut TK gene nor the cell TK gene will cause such phosphorylation. Once IDC is incorporated into a cell's DNA, it becomes insoluble. On the other hand, unincorporated IDC can be easily washed from cell cultures with an aqueous solution such as physiological buffer. These facts have been used in the following manner to demonstrate the expression of the HSV TK gene in coconut mutants.

That is, the monocell layers are infected with mutated virus under conditions that promote plaque formation, ie. those that promote cell growth and viral replication. When the cells are infected, they are then treated with the commercially available radio-labeled IDC (IDC *); the labeling can be easily carried out with, 25I. If the cells are infected with a virus containing the HSV TK gene and if the HSV TK gene present therein is expressed, the cell will incorporate IDC * into its DNA. If the monocell layers are now washed with a physiological buffer, unincorporated IDC * will be washed out. If the monocell layers are then transferred to a nitrocellulose filter and subjected to X-ray film, darkening of the film signifies the presence of IDC * in the plaques and demonstrates the expression of the HSV TK gene by coking mutants.

Using the aforementioned genotypic and phenotypic assays, the applicants have identified two co-op mutants, designated VP-1 and VP-2. VP-1 (ATCC No. VR 2032) is a recombinant coconut virus, derived from coconut S-variant, modified by in vivo recombination with the plasmid pDP 132. VP-2 (ATCC No. VR 2030) is an S-variant coconut virus. modified by recombination with pDP 137.

FIG. Figure 4A is a Hind III restriction map of the coke genome showing the site of HSV TK gene insertion. Figures 4B and 4C enlarge the Hind III F fragment, contained in VP-1 and VP-2, respectively, to show the orientation of the Barn HI HSV TK fragment therein. Attention should be drawn to the fact that in vivo the recombination of pDP

The S variant (ie, VP-2) deletes one of the Bam HI HS V TK fragments present in tandem in the starting plasmid.

As previously mentioned, the fact that the HSV TK gene is expressed can be used for the rapid and easy detection and identification of mutants containing or free from the HSV TK gene or from a foreign gene present alone or in association with the HSV gene. The test and its basis are described immediately below.

Applicants have isolated, in biologically pure form, a coconut mutant, a S variant in particular, which is free of any naturally-occurring functional TK gene, termed VTK “79 (ATCC no.

VR 2031). Usually, the S and L variants discussed earlier herein have a TI gene in Hind III 10 fragment J thereof. If this mutant, which is free of coco-TK gene action, is used for the production of additional mutated organisms containing the HSV TK gene incorporated into the coco-mutant by the techniques described herein, the HSV TK gene present in such resultant mutants may be the only functional TK gene present in the virus. The presence or absence of such an HSV TK gene can be immediately detected by culturing cells infected with viruses on one or more selected substrates.

Namely, such a selected substrate contains bromodoxyoxyuridine (BUdR), a nucleoside analogous to timidine, but highly mutagenic and toxic to organisms such as a cell or virus when present in DNA contained therein. Such a substrate is known from Kit et al., Exp.

Cell Res. 31, 297-312 (1963). Other selected substrates are hypoxanthine / aminopterin / thimidine (HAT) substrate by Littlefield, Proc. Natl. Acad. Sci. USA 50, 568-573 (1963) and variants thereof, such as MTAGG, described by Davis et al., J. Virol. J_3, 140-145 (1974) or the additional variant of MTAGG, described by Campione-Piccardo et al. in J. Virol. 31, 281-287 (1979). All of these substrates selectively distinguish between organisms containing 25 and expressing a TK gene and those that do not contain or express any TK gene. The selectivity of the substrates is based on the following phenomena.

23 DK 174253 B1

There are two metabolic modes for phosphorylation of timidine. The major metabolic mode is not dependent on thymidine kinase and, while synthesizing phosphorylated thymidine by intermediate mechanisms, it will not phosphorylate BUdR or directly phosphorylate thymidine. The second metabolic mode implicates the action of thymidine kinase and will result in phosphorylation of both 5 thymidine and its analog BUdR. Since BUdR is a toxic and highly mutagenic substance, the presence of TK, such as HSV TK, as it is concerned, in an organism will result in phosphorylation of BUdR and its incorporation into the DNA of the growing organism, resulting in its death. On the other hand, if the TK gene is absent or not expressed and the major metabolic pathway followed results in the synthesis of phosphorylated thymidine, but not in the phosphorylation of BUdR, the metabolizing organism will survive in the presence of BUdR, since this substance is not incorporated into its DNA.

The growth behavior discussed above is summarized in Figs. 5 of the attached drawings which tabulate the growth behavior of organisms expressing TK (TK +) and organisms free of or which do not express the TK gene (TK *) on a normal substrate, on a selective substrate such as HAT blocking it main metabolic mode that does not use TK, and on a substrate containing BUdR. TK + and TK 'organisms will both grow on a normal growth substrate using the most important metabolic mode that does not require TK. Nevertheless, on a selective substrate, such as HAT, which blocks the major metabolic mode that is not dependent on TK, the TK + organism will grow because the enzyme completes the phosphorylation necessary for incorporation of timidine into DNA. On the other hand, TK 'organisms will not survive. In contrast, if the organisms are grown on a substrate containing BUdR, the TK + variants will die as TK phosphorylates BUdR and this toxic material is incorporated into the DN A. In contrast, since BUdR is not phosphorylated by the major metabolic mode, the TK 'variant will grow as BUdR is not incorporated into the DNA.

Thus, if a coconut TK-free virus, such as VTK “79, is used as a coconut virus into which the HSV TK gene is inserted, the presence and expression, or absence, of the HSV TK gene therein can be readily determined therein. by simply growing the recombinants on a selective substrate such as HAT. The viruses that are mutated will survive as they use HSV TK to synthesize DNA.

Applicants have prepared several coconut virus free mutants of coconut TK. These have been referred to as VP-3 (ATCC No. VR 2036), a recombinant of VTK-79 and pDP 132, and 5 VP-4 (ATCC No. VR 2033), a recombinant of VTK ”79 and pDP 137. The latter expresses The HSV gene and can be readily identified using the selective substrate mentioned above.

Two additional recombinant viruses, named VP-5 (ATCC No. VR 2028), and VP-6 (ATCC No. VR

VR 2029), respectively, are recombinants of pDP 132 and pDP 137 with VTK_11 (ATCC No. VR 2027), a known Kokoppe-L variant that does not express the Kokoppe TK gene. Thus, DNA can be introduced in excess of the maximum coking genome length.

The techniques mentioned can be used to introduce the HSV TK gene into different portions of the coca genome for the purpose of identifying insignificant portions of the genome. That is, if the HSV TK gene can be inserted into the coco genome, as it is in the Hind III F fragment thereof, the region of the genome into which it has been introduced is obviously insignificant. Each insignificant site within the genome is a suitable candidate for the insertion of exogenous genes, so that the methods are useful for mapping such insignificant sites in the coco genome.

Furthermore, if the HSV TK gene is coupled to another exogenous gene and the resulting combined DNA material is inserted into a co-pox virus free from co-poke TK gene, such as 20 VTK "79, recombinants that are formed and which contains the foreign gene, express the HSV TK gene and can be readily separated from the TK variants by the sorting technique described immediately above.

A further embodiment involves the preparation of a co-chicken Hind-HI-F fragment containing an exogenous gene therein, and the treatment of cells with the fragment together with a co-chicken mutant which does not express the co-chicken TK gene but has the HSV TK gene. gene 25 DK 174253 B1 incorporated therein by in vivo recombination. As in the marker direction mentioned earlier herein, and in vivo techniques used to incorporate the TK-modified Hind III F fragment into the coco, cross-over and recombination may occur to produce an additional mutant in which the HSV TK modified F fragment is replicated 5 by an F fragment containing another exogenous gene. The resulting coconut mutant in which the HSV TK F fragment has been replicated by an F fragment containing the exogenous gene will be completely free of TK, whereas the predominantly non-mutated parental virus will still be HSV TK +. Likewise, a foreign gene may be inserted into the HSV TK gene present in such a coco mutant, which degrades the innocence of the gene giving the recombinant organism TK 'as compared to the non-mutated TK + parent. In both examples, an instant distinction can be made between the coco mutants containing the foreign gene and those free of some TK by growth on BUdR and / or a particular substrate such as HAT.

Figures 7 A-C can best be understood in connection with Figures 3 A-C. Thus, it can be seen in FIG. 3B, prior to incorporation of any additional DNA therein, plasmid pDP 3 has a molecular weight of 11.3 megadaltons (md). When additional DNA is incorporated therein, such as the herpes Barn TK fragment, shown in FIG. 3B, to produce plasmids pDP 132 and pDP 137, the latter plasmids have an increased molecular weight of 13, 6 and 15.9 months, respectively. Since these molecular weights are approximately at the upper limit of replication of the plasmid, it has been found desirable to create a plasmid containing the co-cup Hind III F fragment, but which plasmid is of a lower molecular weight than pDP 3, shown in FIG. 3B. One method of creating such a lower molecular weight plasmid is shown in Figures 7A-C.

More particularly, Fig. 7A shows plasmid pDP 3 containing the Hind III F fragment of the 8.6 megadalton molecular weight coconut. As shown in the Figure, the plasmid contains 3 sites susceptible to cleavage by the restriction enzyme Pst I. Two of these sites are within the F-fragment portion of the plasmid, while the third is within the portion of the plasmid derived from Parental plasmid pBR 322. As further shown in FIG. 7A, when 26 plasmid pDP 3 is cleaved with Pst I, three fragments are obtained. The fragment, which is exclusively a portion of the coconut Hind III F fragment, has a molecular weight of 3.7 md. There are also two other fragments, each of which unites portions of the parental pBR 322 and the cow-chicken Hind-III-F fragment.

5 The largest, "pure" F fragment can be easily cultured. As shown in FIG. 7B, the fragment can then be inserted into pBR 322 at a Pst I site therein after cleavage of the pBR 322 plasmid with Pst I. The association of the parental fragments with T4 DNA ligase produces the new plasmid pDP 120, shown in FIG. 7C, having a molecular weight of only 6.4 mg. The lower molecular weight of the pDP 120 plasmid, as compared to pDP 3, allows entry into that of 10 longer DNA sequences without approaching the upper limit of replication, such as plasmids pDP 132 and pDP 137 shown in Figs. 3C, do.

Again, a better understanding of Figures 8 A-D can be obtained by looking at Figures 3 A-C.

More particularly, Figures 3B and 3C show the incorporation of a herpes Barn TK fragment into plasmid pDP 3 to form plasmids pDP 132 and 137. As explained in more detail 15 in Example X, this herpes Barn TK fragment is introduced into the coconut virus. by an in vivo recombination technique which results in the simultaneous treatment of suitable cells with the coconut virus and Hind III treated pDP 132 or pDP 137. It will be obvious by an inspection of Figs. 3 C, that treatment of the above plasmids with Hind 111 will excise the portion of the plasmids originally derived from plasmid pBR 322, since the herpes Barn 20 TK fragment to be incorporated into the coconut virus by in vivo recombination was present in a coconut Hind III F fragment, connected to the pBR 322 segment at a Hind III site. Thus, the herpes Barn TK gene is incorporated into the coop without the pBR 322 DNA sequence.

However, due to the numerous restriction sites available in the pBR 322 plasmid, for example including Eco RI, Hind III, Barn HI, Pst 1, etc., the plasmid is particularly advantageous for the insertion of DNA sequences therein. Accordingly, it would be desirable to be able to introduce pBR 322 into a vector, such as coconut virus.

27 DK 174253 B1

Figures 8 A-D show the evolution of two plasmids by which the multifaceted DNA sequence of pBR 322 can be incorporated into cupric virus by in vivo recombination and, in particular, the production of two cupric mutants VP 7 and VP 8 containing the pBR 322 DNA sequence.

5 More particularly, FIG. The 8A coconut Hind III F fragment, also shown in FIG. 3 A of the drawings. The linear segment may be self-ligated to form a circular F fragment, as also shown in FIG. 8 A. The united Hind III terminals are indicated on both the linear and circular fragment, which, respectively, "a" and "d '\ The terminals of a Child Hi site are also shown in Fig. 8 A such as V and "b".

10 As shown in particular in FIG. 8 B, this circularized F fragment can be treated with Bam HI to produce a linear DNA sequence in which the Barn ΗΙ terminals "b" and "c" are shown with respect to Hind III terminals "a" and "d" . This linear sequence should be referred to as an "inverted F fragment".

In case, as further shown in FIG. 8 B, Bam ΗΙ-treated pBR 322 and the linear inverted 15 F fragment sequence of FIG. 8 B is associated with T4 DNA ligase, 2 plasmids are produced, depending on the relative alignment of the reverse F fragment and the parental pBR-322 sequence. These two plasmids are shown in FIG. 8 C as pDP 301B and pDP 301A, each having the same molecular weight of 11.3 months.

The incorporation of plasmid mepDP 301 Aog301 B into the coke by in vivo recombination is shown in FIG. 8 D. Namely, each of these plasmids was incorporated by in vivo recombination into coconut virus VTK ~ 79, to produce coconut mutants VP 7 (ATCC No. VR 2042) and VP 8 (ATCC No. VR 2053, respectively). As shown in this Figure, the pDP 301 plasmids, for this purpose, are each cleaved with Sst I to produce linear DNA sequences whose clamps are homologous to a corresponding DNA sequence present in the F portion of the coconut virus genome. Concomitant treatment of cells. Sstl-treated plasmids and coconut virus results in in vivo recombination with incorporation of the pBR 322 DNA sequence into the viral genome.

The advantage of the presence of the pBR 322 sequence in the co-genome of VP 7 and VP 8 is that in vivo recombination can be easily effected using these variants and pBR 322-5 sequences, modified to have a variety of foreign DNA sequences therein. In this case, the homologous base pair of pBR 322 in the coke genome and in the modified pBR 322 DNA sequence to be inserted facilitates crossover and recombination, as illustrated below, with respect to the construction of further novel coke virus mutants. identified as VP 10, VP 13, VP 14 and VP 16.

Figures 9 and 10 relate to the insertion of an influenza gene into the coop to provide two additional coop coop mutants VP 9 and VP 10.

The influenza genome consists of eight separate pieces of RNA, each of which encodes at least one different protein. One of the most important antigenic proteins is the hemagglutinin protein, and because of this, the HA gene was selected for insertion into the coop. The influenza virus genome contains genes in an RNA sequence and, for incorporation into a plasmid, they must be converted to a DNA copy identified as cDNA. As is well known in the art, the cDNA copy of the HA RNA genome is made using reverse transcriptase, all as described by Bacz et al. in Nucleic Acids Research 8, 5845-5858 (1980).

The influenza virus exists in a number of variants classified according to the nature of the HA-20 gene and another of the eight genes, namely that which encodes neuraminidase. Within the influenza virus family, there are three major types of the HA serotype, termed Η1-H3.

In the construct of coconut virus mutant VP 9 and 10, the influenza virus used was A / PR / 8/34 containing an HI HA gene.

29 DK 174253 B1

FIG. 9 A shows two circular plasmids pJZ 102 A and pJZ 102 B. The plasmids were prepared by incorporating a cDNA copy of the influenza hemagglutinin (HA) gene into pBR 322 at the Hind III site. A and B plasmid variants differ in the orientation of the HA gene therein, which is indicated in Figures 9A with reference to an initiation codon contained 5 within the HA gene and placed inside the gene at its proximity to an Ava I site within gene.

The relative positions of the Ava I site and the initiation codon of pJZ 102 A and B variants are given in Figs. 9 A.

If plasmid pJZ 102 is treated with BamHI and T4 DNA ligands in the presence of an "inverted" F-fragment of coconut virus (the latter is shown in Fig. 8 B), the result is a plasmid added, shown in Figs. 9 B as pJZ 102 A / F, in which the pJZ 102 A parental plasmid is associated with the coccyx F fragment.

As further shown in FIG. 9 B, if the pJZ 102 A / F plasmid shown in FIG. 9 B is incorporated into the VTK-79 strain of coconut virus by in vivo recombination, a coconut mutant VP 9 (ATCC VR No. 2043) is produced which contains and expresses the influenza hemagglutinin antigen (HA) gene and can be used. , as described below, to activate the production of antigens for the antigen in a mammal.

FIG. 9 C is a map of the portion of the VP 9 genome containing the pJZ 102 A / F DNA sequence and the HA gene therein.

What is particularly interesting about the expression of the HA gene at VP 9 is the remote location of the gene 20 within the genome fragment shown in FIG. 9C, from the tentative site of a coconut activator in the Hind III F fragment. More specifically, with respect to the linear diagram of FIG. 9C, there are approximately 4,000 base pairs between the tentative site of the activator in the co-op F fragment shown on the diagram and the approximate location of the initiation codon within the HA gene (near the Ava I site).

30 DK 174253 B1

Figs. 10 A-C show the construction of another cocoput mutant containing the high-magglutinin (HA) gene, which is called coPop mutant here with VP 10 (ATCC no.

VR 2044).

More specifically, the VP 10 mutant is constructed by in vivo recombination of plasmid pDP 301 A DNA from plasmid pJZ 102 B (cf. Fig. 9A) with the pBR 322 DNA sequence found in coconut mutant VP 7. are shown in Figures 8 C and 8 D.

Thus, FIG. 10 A a linear DNA map of pJZ 102 B after treatment of the circular plasmid with Barn HI. Again, the initiation codon within the HA gene is indicated for an Ava I site within the gene, which in turn is contained within the DNA sequence of plasmid pBR 322. FIG. Figure 10 B shows a portion of the genome found within coconut mutant VP 7 as a result of the incorporation of plasmid pDP 301 B into VTK_79 by in vivo recombination. More specifically, FIG. The 10 B presence of the pBR 322 genome, surrounded on each side by portions of the F-fragment of the coconut virus, whereby the presence of its F "arms" initially allowed the incorporation of the pBR 322 DNA sequence into the coconut genome.

Earlier in the patent specification, reference was made to the use of in vivo recombination to make an HSV TK + coconut virus into an HSV TK. This results in replication of the HSV TK gene present in such a virus with an HSV TK gene containing a foreign gene therein that makes the HSV gene TK ". The work in question illustrates such a technique 20 using another DNA, especially pBR 3 22 DNA, which is also exogenous to the coop, i.e., recombination will occur with the coop as long as there are homologous sequences in the transfecting (donor) DNA and in the infecting virus. flanking the foreign gene to be inserted, whether such sequences are or are not endogenous cooktop sequences, yet other DNA sequences can be inserted into the cooktop and later used for in vivo recombination in a similar manner.

31 DK 174253 B1 Due to the presence of homologous base pairs in pBR 322 portions of pJZ 102 B (shown in Fig. 10 A) and the pBR 322 DNA sequence contained within the genome of VP 7 (m.p.

FIG. 10 B), overcrossing may occur during in vivo recombination, which involves the simultaneous treatment of cells with VIP 7 and pJZ 102 B, with the incorporation of the HA-5 fragment into the coco genome to create the coco mutant VP 10, as shown in FIG.

10 C.

VP 10 illustrates that in vivo recombination may occur within the 350 base pairs between the Hind III and Barn HI sites at the right end of the pBR 322 sequences, shown in Figures 10 A-C.

Expression of the HA gene of this virus, e.g. by stimulating antigen production in mammals after infection with coco mutant VP 10 has not yet been demonstrated.

To determine whether or not viruses VP 9 and VP 10 express the HA gene, a variety of tissue cultures were prepared, namely, BHK cells present in a first pair of petri dishes were infected with A / PR / 8/34 influenza virus. CV-1 cells, present in a second pair of petri dishes, were infected with the VP 9 coconut variant and CV-1 cells, screened in a third pair of petri dishes, were infected with the VP 10 coconut variant. After allowing viruses to grow within the cells, the infected cells were present in one of each of the three pairs of petri dishes treated with H1 HA antiserum: the second set of three cell cultures (one BHK and two CV-1 cultures) was treated with H3 HA antiserum. All cell cultures were then washed and then treated with protein A, labeled with] 2SI. After treatment with the labeled protein A, the cell cultures are washed again and then they are radioautographed. If the influenza antigen has been produced by the infected cells, the antigen will react with antagonists contained within the H1 HA or H3 HA antiserum, respectively. These complexes will not be washed from the plates, and the I25I protein A will bind, with the constant portion of residual solid-fat fat chains in the complex, if the complex is present.

32 DK 174253 B1

It was determined that complexes formed in the petri dish, infected with A / PR / 8/34 and treated with Η1 HA antiserum, which was also the case with the C V-1 cells, infected with VP 9.1, in contrast, no antigenic antibody complexes were formed. in any of the cells treated with VP 10, nor was any complex formation detected in the cell 5 cultures treated either with A / PR / 8/34 or VP 9 when these cell cultures were treated with H3 HA antiserum.

First, from this experiment, it can be concluded that the VP 10 coconut variant does not express the H1 HA gene on a detectable level. In contrast, the VP 9 variant expresses this gene. Furthermore, the expression VP 9 is specific for H1 HA, as there is no complex formation in the VP 9-treated cell cultures, which are subsequently contacted with H3 HA antiserum.

Based on the fact that VP 9 expresses the Η1 HA gene in vitro, it was then tested whether the VP 9 coconut mutant would express the H1 HA gene sufficiently to stimulate the formation of antibodies in an animal infected with this coconut mutant.

For this test, rabbits were infected with coconut virus VP 9 by intravenous injection.

After 17.25 and 41 days, blood was drawn from the rabbits and serum was collected. The presence of antibodies to H1 HA within this serum was tested with a series of in vitro experiments, similar to those described previously, which cause infection of cell cultures with A / PR / 8/34 and VP 9. The formation of an immune complex was observed when a 20 cell layer infected with VP 9 was treated with the rabbit antiserum. However, this test simply indicates that the rabbit produced antidotes against the coconut virus: it is not possible to determine whether antibodies were produced specifically for the Η1 HA antigen. However, the formation of a complex between the rabbit serum and a BHK monocell layer, infected with A / PR / 8/34, signified the presence, in the antiserum, of antibodies specific for the H1 HA HA antigen.

33 DK 174253 B1

As a separate criterion for the production of HA antibodies, a hemaglutinin inhibition test was performed. This test uses the property of HA to agglutinate red blood cells into large complexes.

To perform this test, the rabbit antiserum was first sequentially diluted. Each antiserum dilution in the series was tested for reaction with the same fixed amount of hemagglutinin obtained by extracting cells infected with influenza virus. If antagonists are present in the antiserum in an amount similar to or in excess of the amount of hemagglutinin introduced into each dilution in the series, the resulting mixture will inhibit the agglutination of red blood cells, thus mixed, due to the presence of an excess of antagonist with respect to the agglutinating agent HA. Φ In the serial dilution performed (on the 45th antiserum day), all dilutions inhibited up to and including 1: 320 red blood cell leaglutination. This indicated the presence of H1 HA antibodies in the antiserum in an amount in excess of the HA antigen entered therein.

15 These experiments demonstrate two important facts. First, it is possible to create a coconut mutant in accordance with the aforementioned techniques, which mutant, when introduced into an animal model, will stimulate the production, even with only primary infection, of antibody to a protein encoded by a gene within the coop-mutant, which gene is foreign to both, to co-poop as to the animal into which it is introduced. Second, the experiments show that the animal's production of antibodies against the coop itself does not interfere with the simultaneous production of counterparts against the foreign DNA contained within the coop mutant.

The construction of plasmids pDP 250 A and pDP 250 B and their incorporation into VTK 79 79 to give novel coco mutants VP 12 (ATCC No. VR 2046) and 25 VP 11 (ATCC No. VR 2045) are shown in Figures 11 AE. .

34 DK 174253 B1

It is possible to isolate a circular DNA comprising the entire hepatitis B virus (HBV) genome. As shown in FIG. 11A, the genome is represented as comprising a surface coding region antigen, including a front surface antigen region containing an Eco-RI site therein. These surface antigen regions are depicted in FIG. 11A as a "block" contained within the genome, of which the remaining DNA is made by a zigzag line.

When the genome is processed within Eco-RI to cleave it, the linear DNA sequence obtained is interrupted in the forward surface antigen portion thereof, so that a portion of the forward surface antigen region is present at each of the linear DNA sequence clamps. The two clips of the sequence, one on each side of the Eco-RI site of the circular genome, are represented in both the circular genome and the linear DNA fragment by a circle and square, respectively.

If HBV DNA fragment is now incorporated into pBR 322, as shown in FIG. 11B, and the plasmid containing two hepatitis B fragments in tandem is purified, the known plasmid pTHBV 1, shown in FIG. 11 B. This plasmid will contain the reconstructed front surface antigen and surface antigen regions of the original hepatitis B genome therein, which were pointed out by identifying these regions with short dashed lines in the image given by plasmid pTHBV 1 in FIG. 11 B. This whole construct is described by Hirschman et al., Proc. Natl. Acad. Sci. USA 77, 5507-5511 (1980).

This sAg region of pTHBV 1 can be purified by treatment of the plasmid with the restriction enzyme Bgl II, the pBR 322 portion of the pTHBV 1 plasmid contains no Bgl II site, while each of the two HBV DNA fragments, present in tandem in the plasmid, contains three Bgl II sites. Thus, pTHBV 1 contains six Bgl II sites, all in the HBV DNA, but all of which are outside the sAg region. In this way, as shown in Figures 1B and 11C, cleavage of pTHBV 1 with Bgl II will produce a linear DNA fragment containing the sAg region of the hepatitis B virus. (Galibert et al., Nature 281,646-650 (1979)).

This fragment can be incorporated into plasmid pDP 120, the production of which is described previously in Figures 7A-C, although the pDP 120 plasmid contains no Bgl II site.

DK 174253 B1

The recognition sequence for the enzyme Bgl II is -AGATCT- with the cleavage site located between -A and GATCT-. On the other hand, the reclamation site of Barn HI is -GGATCC-, with the splitting site lying between -G and GATCC-. Therefore, if DNA containing one or the other of these recognition sites is intersected with Bgl II 5 or Barn HI, in each case, a -GATC- "sticky end" will be produced, which ends will be equal (but so no longer subject to cleavage of either Bgl II or Bam HI).

The new plasmids pDP 250 A and pDP 250 B are constructed as shown in Figures 11 C and D by partial cleavage of pDP 120 with Bam HI and ligation of the "sticky ends" 10 thus produced, with the corresponding "sticky ends" Of the hepatitis virus genome Bgl II fragment.

As is well known in the art, it is possible to counteract the recycling of Bam ΗΙ cleaved pDP 120 by treating the cleaved plasmid with alkaline phosphatase, an enzyme that removes final phosphoric acid groups from the 5 'cleaved ends of the linear DNA of pDP 120. 15 of the phosphoric acid groups prevent a recirculation reaction of pDP 120, but do not generate reaction of the 3'-OH clips of the linear pDP 120 DNA with the 5'-phosphate ends present on the Bgl II fragments, with the following circulation to produce plasmids , such as pDP 250 A and pDP 250 B. Thus, the statistical probability of the creation of the latter, tetracycline-resistant, plasmids can be increased with the alkaline phosphatase treatment, as described.

Finally, plasmids pDP 250 A and pDP 250 B are identified by restriction analysis using Xho I to determine orientation.

As shown in Figures 11D and E, the virus mutants VP 11 and VP 12 are derived from plasmids pDP 250 B and pDP 250 A, respectively, by in vivo recombination of these plasmids 25 with VTK_79 coconut virus. Cross-over and recombination occur in the long and short "arms" of the co-chicken F fragment present in the plasmids. The portion of the plasmids derived from pBR 322 is not incorporated into the virus.

FIG. 12 A. shows the structure of the pTHBV 1 plasmid, known in the art and shown earlier herein in FIG.

11 B. As shown in the Figure, if the plasmid is treated with the restriction enzyme Hha I, 5 identical fragments will be obtained containing only the region of the HBV genome contained within the plasmid encoding the surface antigen free of any surface surface. -antigenregion. (There are, in fact, many Hha I restriction sites within pTHBV 1, and numerous fragments will be produced following digestion with this restriction enzyme. However, the fragment of interest discussed above is the largest of the numerous fragments obtained and can be readily purified due to of this fact.) A linear DNA map of this Hha I fragment is also shown in Figs. 12 A, with further indication of a Child Hi site for informational purposes.

If the Hha I fragment of FIG. 12 A is first treated with T4 DNA polymerase and then Hind III linkers are added with T4 DNA ligase, the fragment can be provided with Hind III 15 sticky ends. As known in the art, T4 polymerase has both a polymerase effect in the 5 'to 3' direction, as well as exonuclease action in the 3 'to 5' direction. The two opposite effects will result in "attenuation" of 3'-OH ends in the Hha I fragment, shown in FIG. 12A, to an equilibrium state is obtained, with the resultant production of a DNA fragment with a blunt end. The fragment with the blunt end can be treated with Hind II links, known in the art, which are essentially decanucleotides containing therein the Hind III recombination sequence.

As shown on the map of FIG. 12B, the resulting fragment will have Hind III sticky ends and may be introduced, as shown in FIG. 12 B, into pBR 322 by treatment of the latter plasmid with Hind III and T4 DNA ligase. The resulting plasmid, identified in FIG.

25 C, is designated pDP 252.

37 DK 174253 B1

Finally, as shown in FIG. 12 D, this plasmid can be introduced into coconut mutant VP 8 by in vivo recombination to produce a new coconut variant VP 13 (ATCC no.

2047). Expression of the HBV gene at VP 13 has not yet been demonstrated.

Figures 13 AC show the production of two additional virus mutants VP 16 (ATCC No. VR 5 2050) and VP 14 (ATCC No. VR 2048), each containing the herpes virus type I DNA sequence encoding the production of herpes glycoproteins gA + gB .

More specifically, FIG. 13A is a map of the Eco RI fragment F of herpes virus type I strain KOS (Little et al., Virology 112. 686-702 (1981). As can be seen in the Figure, the DNA sequence contains numerous Bam Hi sites (indicated as B) within the sequence, including a 10 DNA region, 5.1 md in length between adjacent Bam Hi sites and representing the largest Bam Hi fragment within the Eco Ri fragment in question here.

Furthermore, as shown in FIG. 13A, this Eco Ri fragment can be introduced into pBR 322 by treatment with Eco RI and T4 DNA ligase to produce two new plasmids, identified as pBL 520 A and 520 B, respectively, distinguished by the orientation of the Eco RI-15 fragment therein (as indicated by the Hpa I sites, used for orientation.)

Finally, as shown in FIG. 13 C, these plasmids can be introduced into the coconut virus mutant VP 7 (containing the pBR 322 genome) by in vivo recombination, analogous to that discussed for Figures 10 and 12, and the production of coconut mutants VP 16 and VP 14, respectively.

Thus, this is a further example of the use of the pBR 322 DNA sequence (rather than the DNA sequence of the co-pox F fragment) to effect in vivo recombination for the production of additional co-pox virus mutants. In addition, coconut variants VP 16 and 14 produced by this method are of interest, as they contain more than 20,000 base pairs of foreign DNA incorporated into the coconut virus genome. This represents a minimum 25 upper limit of foreign DNA insertion into the coop.

38 DK 174253 B1

Figures 14 A-C show the production of two additional virus mutants VP 17 and VP 18 into which a Bam HI segment of herpes virus type I (KOS strain) Eco RI fragment F. has been introduced. The Eco RI fragment F is shown in Figs. 13A as including this 5.1 mega-dalton Bam Hi fragment G.

5 Fi g. 14 A shows this Bam Hi fragment, including the site therein of two Sst I sites that are asymmetric and are used for orientation.

This Bam Hi segment of the herpes virus, which still encodes the production of herpes glycoproteins gA + gB (cf. De Luca et al.), Has been introduced into pDP 120 (cf. Fig. 7C) by partial digestion with Bam HI and T4. DNA ligase, as further shown in FIG. 14 A.

10 As shown in FIG. 14 B, two new plasmids pBL 522 A and pBL 522 B have been obtained, each having a molecular weight of 11.6 megadaltons and each containing the Bam Hi fragment G of the herpes virus genome in one of two different orientations.

Finally, as shown in FIG. At 14 ° C, these plasmids can be introduced into VTK ~ 79 by in vivo recombination to produce coconut mutants VP 17 (ATCC No. 2051) and VP 18 (ATCC 15 No. VR 2052). Expression of the herpes glycoprotein gene by mutants VP 17 and 18 has not yet been determined.

Figures 15 A-F illustrate the construction of a further co-op variant VP 22 (ATCC no.

VR 2054), in which foreign DNA is present in the coking genome in an insignificant region, different from the fragment used for the production of other coking variants described herein.

More specifically, FIG. 15 A coconut virus L variant genome Ava IH fragment. As shown in FIG. 6A, the Ava I fragment is completely, within the region, deleted from the S variant, and is therefore known to be immaterial to the viability of the virus.

39 DK 174253 B1

As shown in FIG. 15A, this Ava I H fragment is joined to a Hind III cleaved pBR 322 piasmid to form a new plasmid pDP 202 shown in FIG. B. The ligation is completed by "ending blunt" both the Ava I H fragment of the co-cup and the clamps of the Hind III cleaved pBR plasmid, using T4 DNA polymerase. When the blunt-ended DNA-5 sequences are combined in the presence of T4 DNA ligase, pDP 202 is formed.

As further suggested in FIG. 15 B, the new plasmid pDP 202 is combined with a herpes Bgl / Bam TK fragment. The latter fragment was obtained from the herpes Bam TK DNA fragment (cf. Fig. 3 B) by treatment with Bgl II. Treatment with Bgl II removes the endogenous herpes activator region contained within the Barn TK fragment.

Since, as previously noted, Bgl II and Barn HI produce the same "sticky ends" of DNA treated therewith, the resulting herpes Bgl / Bam TK fragment can be inserted into a Bam site within pDP 202.

As suggested in FIG. 15 C, such an insert is effected by partial digestion with Bam HI and treatment with T4 DNA ligase.

Since the H-fragment of the coop, present in pDP 202, contains three Bam ΗΙ sites, a total of six plasmids can be produced by insertions into this region, namely two variants for each of the three Bam ΗΙ sites, depending on the orientation of each site of the herpes Bgl / Bam TK DNA sequence. The orientation of the latter can be recognized by the presence therein of a non-symmetric Sst I site near the Bgl II end of the fragment.

20 As shown in FIG. 15 C, two plasmids pDP 202 TK / A and pDP 202 TK / D are obtained when the herpes Bgl / Bam TK fragment is inserted into the first of the three Bam ΗΙ sites present within the H fragment of the coke that are present in pDP 202. Similarly, two other plasmids pDP 202 TK / E and / C are obtained after inserting the Bgl / Bam DNA sequence into the second of the three available sites. Finally, two more plasmids pDP 202 TK / B and / F are obtained

40 DK 174253 B1 after inserting the Bgl / Bam fragment into each of two possible orientations in the third Barn Hi site. Of these plasmids, pDP 202 TK / E has been found to be of particular interest.

The plasmid is shown in more detail in FIG. 15 D, indicating the orientation of the Bgl / Bam fragment.

The plasmid can be incorporated by in vivo recombination into the genome of VTK ~ 79 L. FIG.

15 E is an Ava I card of the left portion of this coconut genome. A map of the modified genome which is the genome of VP 22 is shown in FIG. 15 F.

This coconut variant is of particular interest as it shows a higher level of TK expression than variants VP 2, VP 4 and VP 6, in which the Bam TK fragment is present within the F-fragment of the coconut 10. Furthermore, VP 22 demonstrates the insertion of foreign DNA into insignificant portions of the coca genome, other than the F fragment, which was used for convenience in the constructs of other coca variants reported herein.

Finally, since all of the herpes virus regulatory sequences have been deleted from the Bgl / Bam herpes virus DNA sequence upon treatment with Bgl II, as described earlier herein, the VP 22 co-variant demonstrates in a crucial way that transcription in this recombinant virus is initiated by regulatory signals within the coke genome.

A better understanding of the present invention and its many advantages is obtained by referring to the following specific Examples which are given by way of illustration. The percentages indicated 20 are percentages by weight unless otherwise indicated.

Example I - Purification of Cocopoe Hind III fragments from agarose gels.

Hind III restriction endonuclease was purchased from Boehringer Mannheim Corp. Preparative digestions of DNA were performed in 0.6 ml Hind III buffer containing 10 millimolar (mM).

Tris-HCl (pH7.6), 50 mMNaCl, 10 mMMgCl2.14 mMdithiothreitol (DTT), and 10 micrograms (pg) / ml bovine serum albumin (BSA) in which 10-20 pg DNA and 20-40 Units Hind III (1 unit is the amount of enzyme sufficient to completely cleave 1 µg of lambda DNA for 30 minutes.) 5 Coconut DNA was extracted and purified from virions as follows. Purified virions were lysed at a concentration having an optical density per 50 ml, measured at 260 nanometers (A260) in 10 mM Tris-HCl (pH 7.8), 50 mM beta-mercapto-ethyl alcohol, 100 mM NaCl, 10 mM Na 3 EDTA, 1% Sarkosyl NL-97 and 26% sucrose. Proteinase K was added to 100 µg / ml and the lysate incubated at 37 ° C overnight. DNA was extracted with the addition of an equal volume of phenol-chloroform (1: 1). The organic phase was removed and, the aqueous phase extracted, until the interphase was clear. Two additional chloroform extractions were performed and the aqueous phase was then extensively dialyzed against 10 mM Tris-HCl (pH 7.4) containing 0.1 mMNa 3 EDTA at 4 ° C. DNA was concentrated to approximately 100 µg / ml with Ficoll (a high synthetic copolymer of sucrose 15 and epichlorohydrin).

Digestion of the DNA was for 4 hours at 37 ° C. The reactions were terminated by heating to 65 ° C for 10 minutes, followed by the addition of an aqueous stopper solution containing, 5% agarose, 40% glycerine, 5% sodium dodecyl sulfate (SDS), and 0.25% bromophenol blue (BPB). Samples were deposited at 65 ° C on agarose gel and allowed to solidify prior to electrophoresis.

Electrophoresis was performed in 0.8% agarose gels (0.3 x 14.5 x 30 cm) in electrophoresis buffer containing 36 mM Tris-HCl (pH 7.8), 30 mM NaH 2 PO 4, and 1 mM EDTA. Electrophoresis was at 4 ° C for 42 hours at 50 volts. The gels were stained with ethidium bromide (1 µg / ml in electrophoresis buffer). The restriction fragments were observed with ultraviolet (UV) light, and some fragments were cut from the gel.

42 DK 174253 B1

Fragments were separated from the agarose gel by Vogelstein's et al.'s methods, Proc. Natl.

Acad. Sci. United States 76,615-619 (1979) using glass powder in the following manner. The agarose gel containing a DNA fragment was dissolved in 2.0 ml of a saturated aqueous solution of NaI. 10 mg of glass powder was added per day. pgDNA, calculated to be present. The solution was rotated at 25 ° C overnight to bind the DNA to the glass powder. The DNA glass powder was collected by centrifugation at 2000 rpm for 5 minutes. The DNA glass was then washed with 5 ml of 70% NAI. The DNA glass was again collected by centrifugation and washed in a mixture of 50% buffer (20 mM Tris-HCl (pH 7.2), 200 mM NaCl, 2 mM EDTA) and 50% ethyl alcohol. The DNA glass was collected again by centrifugation and conveniently clamped 10 in 0.5 ml of 20 mM Tris-HCl (pH 7.2), 200 mM NaCl and 2 mM EDTA. The DNA was then eluted from the glass powder at 37 ° C with incubation for 30 minutes. The glass was then removed by centrifugation at 10000 rpm for 15 minutes. DNA was recovered from what swam upstairs with ethyl alcohol precipitation reaction and dissolved in 10 mMTris-HCl (pH7.2) containing 1 mM EDTA.

The 15 F fragment, purified in this way, was used in the following Examples.

Example II - Insertion of Co-Copper Hind 1II-F Fragment into the Hind III Site of pBR 322 f Construction of pDP 3 fpBR 322 Co-Copper Hind III F Recombinant Nlasmidl

Coconut Hind III F fragment was purified from preparative agarose gels as described in Example I. This fragment was inserted into the Hind III site of pBR 322 (Bolivar et al., 20 Gene. 2, 95-113 (1977) )), in the following way.

Approximately 200 nanograms (ng) of pBR 322 were digested with Hind III in 10 ml Tris-HCl (pH 7.6), 50 mM NaCl, 10 mM MgCl hour at 37 ° C. The reaction was stopped by heating to 65 ° C for 10 minutes.

500 ng of purified Hind III coco-F fragment was added and the DNA was coprecipitated with 2 volumes of ethyl alcohol at -70 ° C for 30 minutes. The DNA was then washed with 70% aqueous ethyl alcohol, dried and re-stretched in ligation buffer consisting of 50 mM Tris-10, 10 mMMgCl2, 10 mMDCl2, 10 mMDTT and 1 mM adenosine triphosphate (ATP). About 100 units of T4 DNA ligase (New England Biolabs) were then added and the mixture incubated at 10 ° C overnight. The 1 gaseous DNA was then used to transform E. coliHB101 (Boyer et al., J. Mol. Biol. 4Λ, 459-472 (1969)).

Example III - Transformation of E. coli and selection for recombinant plasmids.

Competent cells were prepared and transformed with plasmids according to the method described by Dagert et al., Gene 6.23-28 (1979). E. coli HB101 cells were made competent by inoculating 50 ml of LB broth (1% bacto-tryptone, 0.5% bacto-yeast extract and 0.5% NaCl supplemented with 0.2% glucose) with 0.3 ml of a one night culture 10 of the cells and allowed to grow at 37 ° C until the culture had an optical density (absorbency) of. 650 nanometers (A ^ q), 0.2 measured with a spectrophotometer. The cells were then cooled on ice for 10 min, compressed into small beads by centrifugation, re-strained in 20 ml of cold 0.1 molar (M) CaCl2 and incubated on ice for 20 min. The cells were then compressed into beads and re-strained in 0.5 ml of cold 0.1 M CaCl 2 and allowed to remain at 4 ° C for 24 hours. The cells were transformed by adding ligated DNA (0.2-0.5 mg in 0.01-0.02 ml ligation buffer) to competent cells (0.1 ml). The cells were then incubated on ice for 10 min and at 37 ° C for 5 min. 2.0 ml of LB broth was then added to the cells and incubated at 37 ° C for one hour with shaking. Aliquots of 10 microliters (μΐ) or 100 μΐ were then smeared on LB agar plates containing ampicillin (Amp) in a 100 µg / ml concentration.

The transformed bacteria were then sorted for recombinant plasmids by transferring ampicillin-resistant (AmpR) colonies to LB agar containing tetracycline (Tet) at 15 pg / ml. The colonies which were both AmpR and tetracycline-sensitive (Tets) (approximately 1%) were screened for intact coconut Hind III F fragment inserted into pBR 322 according to the method of Holmes et al, Anal. Bioch. 114.193-197 (1981) 2.0 ml cultures of transformed E. coli were grown overnight at 37 ° C. The bacteria were compressed into small beads by centrifugation and resuspended in 105 μΐ of an 8% sucrose solution, 5% Triton X-100, 50 mM EDTA and 50 mM Tris-HCl (pH 8.0), followed by the addition of 7.5 μΐ of a recently prepared lysozyme solution (Worthington Biochemicals) (10 mg / ml in 50 mM Tris-HCl (pH 8.01)). The lysates were placed in a boiling water bath for 1 minute and then centrifuged at 1000 rpm for 15 minutes. Whatever swam on top was removed and plasmid DNA precipitated with an equal volume of isopropanol. The plasmid deme was then re-clamped in 40 μΐ Hind III buffer and digested with 1 unit of Hind III for 2 hours. The resulting digestive products were then analyzed on an analytical agarose gel for the appropriate Hind III-F fragment. One such recombinant plasmid containing an intact Hind III F fragment, designated pDP 3, was used for further modification. (See Fig. 3B).

Example IV - Preparative purification of pDP 3.

Large scale purification and purification of plasmid DNA was performed by a modification of the method of Clewel et al, Proc. Natl. Acad. Sci. USA 62, 1159-1166 (1969). 500 ml of LB broth was inoculated with 1.0 ml of one-night E. coli HB 101 culture containing 15 pDP 3. At an optical density (A600) of approximately 0.6, chloramphenicol (100 pg / ml) was added to increase production of plasmids (Clewel, J. Bacteriol. 110. 667-676 (1972)). The bacteria were incubated at 37 ° C for an additional 12-16 hours, at which time they were collected by centrifugation at 5000 rpm for 5 minutes, washed once in 100 ml of TI buffer 10.1 mM Tris-HCl (pH 8.0), 150 mM NaCl, 10 mM EDTA), collected by centrifugation and re-strained into 14 ml of 25% sucrose solution in 0.05 M Tris-HCl (pH 8.0). 4.0 ml of lysozyme solution (5 mg / ml in 0.25 M Tris-HCl (pH 8.0)) was added and mixtures were incubated at room temperature for 30 minutes with subsequent addition of 4.0 ml 0.25 M EDTA ( pH 8.0). The mixture was then placed on ice for 10 minutes. 2.0 ml of pancreatic RNase A (Sigma Chemical Co.) (1 mg / ml in 0.25 M Tris-HCl (pH 8.0)) was added to this mixture, which was then incubated at room temperature for 1 minute. . The cells were then lysed by adding 26 ml of lytic Triton solution (1% Triton X-100.05 M EDTA, 0.05 M Tris-HCl (pH 8.0)). The mixture was incubated at room temperature for 30-60 minutes. The lysate was purified by centrifugation at 17000 rpm for 30 minutes at 4 ° C. What flowed on top was then removed and plasmid DNA was separated from chromosomal DNA on color-buoyancy CsCl increases.

To this end, CsCl-ethidium bromide gradients were prepared by dissolving 22 g of CsCl in 23.7 ml of purified lysate. 1.125 ml of aqueous ethidium bromide (10 mg / ml) was added to the solution. The mixture was then centrifuged in polyallomeric tubes in a Beckman 60 Ti rotor at 44000 rpm for 48-72 hours. The resulting DNA bands in the gradients were observed with ultraviolet light and the lower band (covalently closed plasmid DNA) was removed by puncturing the tube with an 18 gauge needle attached to a syringe. Ethidium bromide was removed from the plasmid by repeated extraction with 2 volumes of chloroform-isoamyl alcohol 10 (24: 1). Plasmids were then extensively dialyzed against 10 mM Tris-HCl (pH 7.4) containing 0.1 mM EDTA to remove CsCl. The plasmid DNA was then concentrated by ethyl alcohol precipitation.

Example V - Construction of pBR 322 / coke / herpes / virus TK recombinant plasmids.

Figures 3B and 3C summarize the steps that must be taken in the construction of the recombinant plasmids used to insert the Bam HS V TK fragment into S or L variant cooks. Approximately 15 µg of covalently sealed pDP 3 was cleaved by partial digestion with Barn HI (Bethesda Research Laboratories) by incubation in Bam HI buffer consisting of 20 mM Tris-HCl (pH 8.0), 7 mM MgC2,100 MM NaCl and 2 mM beta -mercaptoethyl alcohol, using 20 units of Bam HI for 10 minutes at 37 ° C. Since pBR 322 and kokoppe Hind III F each contain a Bam Hi site, partial cleavage results in a mixture of linear plasmids, cleaved either at pBR 322 or the kokoppe-Bam Hi site. These mixed linear plasmids were then separated. The pDP 3 fragments, cleaved both at pBR 322 and the Coke-Bam ΗΙ sites by electrophoresis on agarose gels, and the single cleaved linear plasmids were purified using glass powder, as described in Example I.

46 DK 174253 B1

A recombinant pBR 322 containing the 2.3 megadalton (md) HS V Bam Hi fragment encoding HSV TK, as described by Colbere Garapin et al., Proc. Natl. Acad. Sci.

US 76,3755-3759 (1979) was completely digested with Bam HI and the 2.3 md Bam TK fragment was purified from an agarose gel as described above.

5 pDP 3 Bam TK recombinant plasmids were constructed by ligating around 1 µg of Bam HI linearized pDP 3 to approximately 0.2 µg of purified Bam TF fragment in 20 μΐ ligation buffer containing 100 units of T4 DNA ligase at 10 ° C overnight. . This ligation mixture was then used to transform competent E. coli HB101 cells, as described in Example III.

Example VI - Sorting of transformed cells to identify those containing recombinant plasmids. which has HSV TK insertions.

Transformed cells containing recombinant plasmids were sorted for HSV TK insertions by colony hybridization, essentially as described by Hanahan et al., Gene 10, 63-67 (1980).

A first set of nitrocellulose filters (Schleicher and Schull BA85) were placed on Petri dishes filled with LB agar containing 100 µg / ml ampicillin. Transformed cells were smeared on the filters and the dishes were incubated at 30 ° C overnight or until the colonies were just visible. A replica nitrocellulose filter of each of the first set of filters was made by placing a sterile nitrocellulose filter on top of each of the original filters mentioned above and by pressing the two filters together firmly. Each filter pair was then incised (wedged) with a sterile scalpel blade, separated, and each filter transferred to a fresh LB agar plate containing ampicillin at 100 µg / ml for 4-6 hours. The first set of filters (original filters) were then placed on LB agar plates containing 200 µg / ml chloramphenicol to supplement the plasmid production. The replica filters were stored at 4 ° C.

47 DK 174253 B1

After 24 hours on chloramphenicol, the original nitrocellulose filters were prepared for hybridization as follows. Each nitrocellulose filter was placed on a sheet of Whatman filter paper, saturated with 0.5 N NaOH for 5 minutes, peeled off on dry filter paper for 3 minutes and put back on the NaOH-saturated filter paper for 5 minutes to lys the bacteria thereon and denature their DNA. This sequence was then repeated using Whatman filter paper sheets, saturated with 1.0 M Tris-HCl (pH 8.0) and repeated a third time. filter paper sheets, saturated with 1.0 M Tris-HCl (pH 8.0) containing 1.5 M NaCl for neutralization. The nitrocellulose filters treated in this way were then air dried and baked in vacuo at 80 ° C for 2 hours.

Prior to hybridization, these nitrocellulose filters were then treated for 6-18 hours with incubation at 60 ° C in a pre-hybridization buffer, which is an aqueous mixture of 6 x SSC (1 x SSC = 0.15 M NaCl and 0.015 M Na citrate ( pH 7.2), 1 x Denharts (1 x Denharts = a solution containing 0.2% each of Ficoll, BS A and polyvinylpyrrolidone) and 100-200 µg denatured cut salmon sperm DNA (SS DNA) / ml, 1 mM EDTA and 0.1% SDS.

This treatment will reduce the amount of binding between the filter and non-hybridized probe DNA, which must then be applied to the filters.

To sort for recombinant plasmids containing HSV TK insertions, the transformed colonies attached to the original treated nitrocellulose filters were hybridized with 32P labeled Barn HSV TK fragment by immersing the filters in hybridization buffer containing 2 x SSC (pH 7.2), 1 x Denharts solution, 50 µg SS DNA / ml, 1 mM EDTA, 0.1% SDS, 10% dextran sulfate and 32P Bam TK as the hybridization assay. The radioactivity plan of the solution was approximately 100,000 counts per count. minute (cpm) per minute. milliliter.

The hybridization was carried out at 60 ° C for 18-24 hours (Wahl et al., Proc. Natl. Acad. Sci. USA 25 76, 3683-3687 (1979)).

48 In order to prepare the hybridization study, the 2.3 md Bam TK fragment was labeled with incision translation according to the method of Rigby et al., J. Mol. Biol. 113. 237-251 (1977). More specifically, 0 , 1 ml of a reaction mixture was prepared containing 50 mM Tris-HCl (pH 7.6), 5 mM MgCl2.20 μΜ desoxycytidine triphosphate (dCTP), 20 μΜ des-5 oxyadenosine triphosphate (dATP), 20 μΜ 2 μΜ (alpha-32β) deoxytimidine triphosphate (dTTP) (410 Curies / mole) (Amersham Corporation), 1 ng DNase I, 100 units of DNA polymerase I (Boehringer Mannheim) and 1 μg Barn TK fragment. The reaction mixture was incubated at The reaction was terminated by adding 50 µl 0.5 M EDTA and heating to 65 ° C for 10 minutes. Unincorporated 10 triphosphates were removed by gel filtration of the reaction mixture on Sephadex G50.)

After hybridization, excess sample material was removed from the nitrocellulose filters by washing 5 times in 2 x SSC (pH 7.2) containing 0.1% SDS at room temperature, followed by 3 washes in 0.2 x SSC (pH 7.2). containing 0.1% SDS at 60 ° C, with each wash lasting 30 minutes. The washed filters were then air-dried and used to expose 15 x-ray films (Kodak X-omate R) at -70 ° C for 6-18 hours using a Cronex Lightening Plus gain screen (du Pont) for elevation.

The exposed and developed X-ray film was then used to determine which colonies contained pBR 322 coco-Bam HSV TK recombinant plasmids. These colonies which exposed the X-ray film were found on the corresponding replica nitrocellulose filter.

Such positive colonies were then peeled from the replica filters for further analysis. Of the approximately 1,000 colonies sorted in this way, 65 colonies were experimentally identified as having a Barn TK insertion within pDP 3.

Example VII - Restriction analysis of recombinant plasmids. containing Barn HSV TK

Each of the 65 colonies identified experimentally as containing recombinant plasmids with Bam HSV TK inserts was used to inoculate 2.0 ml cultures of LB broth containing ampicillin at 100 µg / ml. The cultures were then incubated overnight at 37 ° C. Plasmids were extracted from each culture as described in Example III. The plasmids were dissolved in 50 μΐ water after isopropanol precipitation.

To determine whether the plasmids contained an intact 2.3 md Bam HSV TK fragment and 5 to which Barn Hi site within pDP 3, Bam HSV TK was inserted, 25 μΐ of each plasmid preparation was mixed with 25 μΐ of 2 x Hind III. buffer and digested at 37 ° C for 2 hours with 1 unit of Hind III. The resulting fragments were then analyzed by electrophoresis on a 1.0% agarose gel, as previously described.

Of the 65 plasmid specimens analyzed, it was found that 6 contained Barn HSV TK-10 fragments inserted into the Barn Hi site present in the coconut Hind III F portion of the plasmid, i.e., they yielded Hind III restriction fragments with molecular weights corresponding to linear pBR 322 (2.8 md) and fragments with a molecular weight greater than that of the coco-Hind III F fragment (8.6 md).

These 6 plasmids, further analyzed with Sst I (an isoschizomer of Sac 1) to determine the number and orientation of the Barn HSV TK fragments, inserted inside the coke-Hind III F fragment so that Sst I (Sac I) cleaved both for the Barn HSV TK fragment and the Copper-Hind III F fragment asymmetrically. The assays were performed by mixing 25 μΐ of the plasmid with 25 μΐ of 2 x Sst buffer (50 mM Tris-HCl (pH 8.0), 10 mM MgCl2,100 mM NaCl and 10 mM DTT) and digesting with 1 unit of Sst I ( Bethesda Research Laboratories) at 37 ° C for 2 hours. The resulting fragments were analyzed by electrophoresis in 1% agarose gels. Of the 6 plasmids analyzed, 5 yielded two Sst I fragments with molecular weights of 10.1 md and 3.5 md, indicating a single Barn HSV TK insert.

One of these plasmids was selected for further study and designated pDP 132.

The second plasmid yielded three Sst I fragments with molecular weights of 10.8 md, 2.8 md and 2.3 md, indicating tandem Barn HSV TK inserts, head oriented tail and in the opposite orientation compared to pDP 132. This plasmid was designated pDP 137. The plasmids pDP 132 and pDP 137 are plotted in FIG. 3 C.

50 DK 174253 B1

Example VIII - Cultivation of a TK ~ S variant coconut virus

In order to culture a TK-S variant coconut virus mutant, a viral population was subjected to highly selective pressure for such a mutant by culturing the virus in cells in the presence of BUdR that is deadly to organisms carrying the TK gene. More specifically, confluent monolayers of TK human cells (line 143) growing in Eagle's Special Medium in 150 mm Petri dishes were infected with approximately 3 x 10 3 plaque-forming units (pfu) of S-variant coconut virus. Cheers. (20 dishes were used) in the presence of 20 µg BUdR / ml. (Eagle's Special Medium is a commercially available nutrient substrate for the cultivation of most cell lines. Alternative substrates such as Eagle's Minimum Essential Medium, Basal 10 Eagle's Medium, Hams-F10, Medium 199, RPMI-1640, etc. could also is used.) Growth is at 37 ° C in an atmosphere, enriched with CO2. This can conveniently be done using a CO 2 incubator, which supplies air, enriched with CO 2, to have a CO 2 content of about 5 percent.

Ninety-three of the developing plaques were cultured and replicated on TK ~ 15 human cells (line 143), under the conditions mentioned previously, and again in the presence of 20 µg BUdR / ml. A number (5) of large, well-cultivated plaques were selected for further analysis.

The five pure culture plaques were tested for growth on monocell layers under the same conditions used previously and in the presence or absence of 20 µg BUdR / ml. The relative growth of each plaque in the presence and absence of BUdR was noted and compared to the relative growth of similar monolayer cell cultures of the parental S variant virus. The following results were obtained: 51 DK 174253 B1

Pure Plaque -BUdR infii / mn -fBUdR fnfii / mn

No. 70 5.1x10s 4, 10x105

No. 73 1.0x106 Ι, ΟχΙΟ6

No. 76 4.7x10s 4.7x10s 5 No. 79 5.4x105 4.4x105

No. 89 5.9x105 7.0x105 S variant 1.7x1010 ° 9.7x106 The growth of pure plaque No. 79 was further checked in the presence of 0.20 and 40 µg BUdR / ml and compared with the growth of its parental S variant virus. The following results were obtained: *

Growth (pfu / ml)

Virus 0 µg / ml 20 µg / ml 40 pp / ml

No. 79 2.5x10s 4, lxl05 3.2x105 S-Variant l, 2xl09 l, 3xl06 2.0xl05 15 In addition, the above 5 pure plaques and the parental S-variant were checked for growth on TK human cells (line 143) in the presence of MTAGG. MTAGG is an Eagle's Special Medium modified by the presence of: 8x10'7M methotrexate 1.6x10'5M timidine 20 5x10'5M adenosine 5x10'5M guanosine lx10-4 M glycine (cf. Davis et al., Op. Cit. ) and select for thymidine kinase qg against organisms free of the thymidine kinase gene. The results of such an experiment were as follows: 52 DK 174253 B1

Virus Plaque forming units / ml -MTAGG + MTAGG Nr. 70 4.0x105 0

No. 73 5.8x105 0 5 Nr. 76 2.8x105 3.3x103

No. 79 3.6x105 0.

No. 80 4.3x105 4.0x103 S variant 4.8x109 2.6x109

Of the three purified plaques that showed complete growth inhibition in the presence of 10 MTAGG, purified Nr.79 was randomly selected and extracts prepared from cells infected with Nr. The 79 virus was compared with extracts prepared from uninfected cells and from cells infected with the parental virus of the S variant in terms of the ability of the extracts to phosphorylate tritiated (3H) thymidine. The results are tabulated below:

Extract source -H Timidine 15 Phosphorous (cpm / 15 µg protein *)

Uninfected TK_human 0 (line 143)

No. 79 infected cells 90 20 S variant infected cells 66,792 Considering (1) resistance to BUdR, (2) growth inhibition with a substrate containing MTAGG and (3) failure to detect significant phosphorylation of timidine in infected cell extracts are considered purified plaque no. 79 to lack thymidine kinase effect.

The cultivated is designated VTK_79.

53 DK 174253 B1

Example IX - Marker rescue of L-variant coconut DNA by the S-variant

Four L-variant DNAs were prepared for marker rescue studies. The first consisted of purified, intact L-variant coconut DNA. The second consisted of L-variant cup DNA digested with Bst EII, a restriction endonuclease which generates a donor DNA fragment, fragment C comprising the DNA absent from the S variant and uniquely present. in the L variant, and which also has, at both ends of the DNA chain, a region of DNA, homologous to corresponding sequences in the S variant. The third and fourth preparations consisted of L-variant DNA, digested with Ava I and Hind III, respectively, restriction endonucleases, which cleave the coco-genome within the unique L-variant-10 DN A sequence. The marker rescue studies performed with these four preparations, demon *, state that the L-variant DNA fragments containing the deleted region absent from the S-variant can be reintroduced into the S-variant by an invivo recombination technique, provided that the fragment, in addition to the deleted region, contains end regions that are homologous to corresponding sequences in the S variant.

15 A better understanding of the fragments used in these studies will be obtained by referring to Figures 6A-C, each of which is a restriction map of a portion of the left clamp genome's clamp genome. More specifically, each map refers to the genome's left clamp region, comprising about 60 kilobase pairs, as indicated in the Figure. The portion of the cuckoo genome deleted from the S variant is represented on each map as the region between the 20 dotted lines shown in the Figures, a region approximately 10 kilobase pairs in length.

Turning now more particularly to Figs. 6A, it is obvious that fragment H, obtained by digestion with Ava I, is completely within the deleted region, but will have no final DNA fragments homologous to the S variant DNA because Ava I cleavage sites 25 ne falls completely within the deleted region of the S variant.

54 DK 174253 B1

The restriction map of FIG. 6B, relating to Hind III, shows that this restriction enzyme also cannot produce an L-variant DNA fragment which partially covers the deleted region of the S-variant. In this example, sequences homologous to the S-variant were found at the left clamp of the C fragment of Hind III. However, the restriction site at fragment C's right clamp falls within the deleted region, and there is no clamp sequence, homologous to the S sequence DNA sequence.

In contrast, the restriction map of FIG. 6 C, regarding Bst EII, that digestion with this enzyme produces a fragment, fragment C, which includes the deleted region, absent from the S variant and also has clamp portions on both the left and right ends homologous to S the corresponding portions of the variant.

The results of the experiments, discussed in more detail below, indicate that the DNA present in the L variant, but deleted from the S variant, is rescued by the high efficiency S variant from the intact L variant genome. rescued with lower efficiency from the C fragment of Bst EII and cannot be rescued from either L-variant DNA fragments prepared with Ava I and Hind III restriction endonucleases.

The high efficiency with which the deleted sequence is rescued from the intact L variant is due to the fact that a simple crossover between the intact L variant and the S variant is sufficient to produce an L variant genome. On the other hand, to save the deleted portion from the C fragment of Bst E II, an intersection between the 20 fragment and the S variant is required in both the left and right clamp portions of the C fragment to incorporate the deleted region into the S-variant. Finally, since neither digestion with Aval nor Hind III produces DNA fragments that can be incorporated into the S variant at any crossover, no rescue of the deleted portion has been effected.

The marker rescue was performed on CV-1 monolayer using the calcium phosphate technique of Graham et al., Virology, 52,456-467 (1973), as modified by Stow et al. and Wigler 55 DK 174253 B1 et al., both mentioned earlier herein. Confluent CV-1 monolayers were infected with S-variant coconut virus to yield approximately 50 to 200 plaques in each of a number of (5-20) 6 cm diameter petri dishes. To infect the cells, the growth substrate (i.e., Eagle's Special containing 10% calf serum) and a dilution of the virus containing 50-200 pfu / 0.2 5 ml in a cell compatible substrate such as Eagle's Special containing 2% calf serum is aspirated. to the cell monolayer. After incubation for one hour at 37 ° C in a CO 2 incubator to allow the absorption of the virus into the cells, different of the four L-variant DNA preparations mentioned earlier were each separately added to the monolayers as a calcium phosphate. -fat precipitate, containing one microgram per dish of the L-variant DNA preparation.

After 40 minutes, Eagle's Special substrate with 10% calf serum was added one, and four hours after the first addition of the DNA, the cell monolayers were exposed to 1 ml of buffered 25 percent dimethyl sulfoxide for four minutes. This buffer contains 8 gNaCl, 0.37 g KC1.0, 125 gNa2HPO4-2H2O1 dextrose and 5 gN- (2-hydroxydetyl) piperazine, N '- (2-ethanesulfonic acid) (Hepes) per ml. liter, having a final pH of 7.05. The dimethyl sulfoxide was removed and the monolayers were washed and coated with nutrient substrate. After three days, at 37 ° C in a CO 2 incubator, the cells were stained with a nutrient substrate coating containing Neutral Red dye staining the uninfected cells (nutrient agar = Eagle's Special medium containing 10% calf serum and 1% agar). The next day, the substrate coating was removed and the monolayers were transferred to nitrocellulose filters and prepared for insitu hybridization, as described by Villarreal et al., Ioc. cit. As digestion of the L variant genome with Ava I generates a 6.8 kilobase pair fragment, fragment H, which resides completely with the unique DNA sequences deleted in the S variant genome (cf. Fig. 6 A), yields 32P-labeled incision. -translated Ava I H fragment a highly specific sample to detect the rescue of the unique L-variant DNA sequence by the S-variant.

Prehybridization, the nitrocellulose filters were interleaved with Whatman Nr. 1 filter paper circles in 6 cm petri dishes and were pre-hybridized for 6 hours at 60 ° C in pre-hybridization buffer (SSC, Denhardt solution, EDTA and S.S. DNA), as described earlier herein in Example VI. The radioactive sample consisting of 32P-labeled incision-translated L variant Ava 56 DK 174253 B1 I, H fragment having a specific effect of approximately 1 x 108 cpm / pg was used for hybridization in 2 x SSC, 1 x Denhardt , 1 mM EDTA, 0.1 percent SDS, 10 percent dextran sulfate, and 50 pg / ml sonicated SS DNA at about 1 x 105 cpm / ml overnight at 60 ° C. The radioactive sample was prepared according to the method of Rigby 5 et al., J. Mol. Biol. 113,237-251 (1977). The filters were washed repeatedly at room temperature and at 60 ° C using the washing method of Example VI, air dried and radiographed.

The results of the experiments are summarized in Table I on the following page:

Table I

10 Donor L variant Percentage of plauer DNA preparation Containing L variant

genotype

Intact L variant 5

BstE II total digestion product

Ava I total digestive product

Hind III Total Digestive Product 20 A minimum of 5000 plaques were analyzed for each donor DNA preparation.

57 DK 174253 B1

Example X - In vivo recombination effervescent pDP 132 oe pDP 137 to generate cowpox virus mutants VP-1 to VP-6 oe identification using replication filters

A first calcium orthophosphate precipitate of donor DNA was prepared by combining 5 µg pDP 132 Hind III digested DNA in 50 µl water, 4 µg S variant carrier DNA (prepared as 5 in Example I) in 40 µl water and 10 µ 12 5 M CaCl 2, combining the resulting mixture with an equal volume of 2 x Hepes phosphate buffer comprising 280 mM NaCl, 50 mM Hepes and 1.5 mM sodium phosphate (pH = 7.1), and allowing the precipitate to form for a period of time of 30 minutes at room temperature. Another precipitate was prepared in the same manner, in the same amounts, but using pDP 137 Hind I-II digested DNA.

10 (As described in more detail by Stow et al. Loc. Cit. And Wigler et al. Loc. Cit., The modifications of Graham et al .'s precipitation technique, referred to earlier, employ carrier DNA as a high molecular weight substance that enhances efficiency. Carrier DNA used is the DNA of the virus used to infect the monolayered cells in the in vivo recombination technique.) For in vivo recombination, confluent monolayers of C V-1, growing in Eagle's Special Medium containing 10 0 were infected. calf serum, with S variant cupping virus at a man-preparedness of 1 pfu / cell infection. The infection method is as described in Example IX. The virus was allowed to absorb for 60 minutes at 37 ° C in a CO 2 incubator, after which the inoculum was aspirated and the cell monolayer was washed. The precipitated DNA-20 preparations were applied to separate cell monolayers, and, after 40 minutes at 37 ° C in a CO 2 incubator, liquid coating substrate (Eagle's Special containing 10% calf serum) was added. In each case, the virus was harvested after 24 hours at 37 ° C in a CO 2 incubator at 3 freeze / thaw cycles and titrated on CV-1 monolayers. Approximately 15000 plaques were analyzed on recombinant virus CV -1 monolayers using replica filters, prepared as follows.

58 DK 174253 B1

Plaques formed on confluent CV-1 monolayers under a nutrient substrate coating were transferred to a nitrocellulose filter by removing the substrate coating in a clean manner with a scalpel and applying the nitrocellulose filter to the monolayer. Good contact between the filter and the monolayer was effected by applying a Whatman Nr. 3-filter paper, wetted in 50 mM Tris 5 buffer (pH = 7.4) and 0.015 mil NaCl over the nitrocellulose filter and stamped with a rubber stopper until the monolayer transferred to the nitrocellulose shows a uniform color surrounding separate plaques without color. The monolayer was already stained with Neutral Red, which is taken up by viable cells, ie. cells unlabeled by viral infection).

The nitrocellulose filter having the transferred monolayer thereon is now removed from the petri dish 10 and applied with the monolayer facing up. A second nitrocellulose filter, wetted in the above Tris-NaCl solution, is now placed directly over the first nitrocellulose filter and the two filters are contacted by stomping with a rubber stopper after protecting the filters with a dry Whatman Nr. 3 circle. After removing the filter paper, a notch is made in the nitrocellulose filters for orientation and they are separated. The second (replica) nitrocellulose filter now contains a mirror image of the cell monolayer transferred to the first nitrocellulose filter. The second filter is conveniently placed in a clean Petri dish and frozen. The first nitrocellulose filter is subjected to hybridization, using 32P-labeled Barn HSV TK fragment as a sample. The preparation of the sample and the hybridization technique are described earlier herein in Example VI.

About 0.5 percent of plaques analyzed by hybridization were positive, ie. were recombinant viruses containing Bam HSV TK.

Plaques of recombinant viruses, similar to those identified on the first nitrocellulose filter by hybridization, were then isolated from the nitrocellulose replica filter by the following technique for further purification.

Using a sharp plug drill bit having a diameter slightly larger than the plaque to be selected, a requested plaque is punched from the first or original nitrocellulose filter which has been used to identify recombinants by hybridization. The resulting perforated filter is then used as a stencil to identify and purify the corresponding plaque on the replica filter. Namely, the replica filter is placed face up on a sterile surface and covered with a sheet of Saran Wrap (plastic). The perforated first 5 or original nitrocellulose filter is then placed with the monolayer side downward over the second filter, and the orientation incisions present in the filters are aligned to bring the mirror images of the plaques to exactly match. Again, using a plug drill bit, a plug is removed from the replica filter and, after removing the protective Saran Wrap layer, it is placed in a ml of Eagle's Special Medium containing 2% calf serum. The nitrocellulose 10 plug is sonicated in this substrate for 30 seconds on ice to release the virus. 0.2 ml of this virus preparation and 0.2 ml of a 1.10 dilution of the preparation are plated on CV-JL monolayers present in 6 cm petri dishes.

As a plaque purification step, the entire sequence of preparing a first nitrocellulose filter, a replica filter, hybridization and plaque purification from the replica filter was repeated.

A sample of a purified plaque prepared in this way, starting with a calcium orthophosphate precipitate of pDP 132 Hind III digested DNA, was designated with coconut vitus VP-1. Likewise, a plaque containing a recombinant prepared from pDP-137 Hind III digested DNA was designated VP-2. Both samples were grown on suitable cell cultures for further study, including identification by restriction analysis and other techniques.

Similarly, two additional cobweb mutants, designated VP-3 and VP-4, respectively, were prepared by in vivo recombination using VTK "79 (an S-variant TK" cobweb virus, as described in Example VIII) as the rescue virus. and pDP 132 and pDP 137, respectively, as the plasma donor DNA. The precipitates were formed, as described previously herein, except that 5 µg of plasma donor DNA present in 50 µl of water, 4 µg of VTK-79 carrier DNA in 150 µl of water and 50 µl of 2.5 M CaCl DK 174253 B1 was added dropwise to an equal volume of 250 μΐ Hepes phosphate buffer, described previously.

Furthermore, the cells used for infection by the VTK “79 virus carrier were BHK-21 (clone 13) cells instead of CV-1.

Two further co-virus virus mutants, designated VP-5 and VP-6, were prepared using 5 calcium orthophosphate precipitates of pdP 132 and pDP 137, respectively, each prepared for mutants VP-3 and VP-4. However, in the case of VP-S and VP-6, the carrier DNA of a cocoon virus is VTK ~ 11 rather than VTK_79.

Again, BHK-21 (C-13) cell monolayers were infected and the rescue virus in this case is VTK “11.

Example XI - Expression of HSV TK by Coconut Mutant VP-2 and the Use of IDC * for Identification thereof

The virus product, obtained in Example X by the in vivo recombination of the S variant coccup virus and the calcium orthophosphate precipitate of pDP-137 Hind III digested DNA, was plated onto confluent monolayers of CV-1 cells, present on approximately twenty 6 cm petri dishes. at a concentration that gives approximately 150 plaques per Cheers. The plaques were covered with a liquid coating substrate, ie. Eagle's Special Medium, containing 10% calf serum. After 24 to 48 hours of incubation at 37 ° C in a CO 2 incubator, the liquid coating substrate was removed from the dishes and was replenished in each case with 1.5 ml of the same liquid coating substrate containing 1-10 µCi, 25 I iodosoxycytidine 20 (IDC *). The plates were then further incubated overnight, at 37 ° C in an atmosphere, with CO 2 added, after which the cell monolayer, present thereon, was stained with the addition of Neutral Red to form a clear image of the plaques by contrast.

The substrate was then removed by aspiration, the monolayers washed three times with phosphate-buffered saline and the cell monolayer on each of the plates embossed on a corresponding nitrocellulose filter. The latter was exposed to X-ray film for 1 to 3 days and then developed.

The viable plaques containing and expressing the HSV TK gene will phosphorylate IDC * and incorporate it into their DNA, rendering the DNA insoluble. In addition, unphosphorylated and unincorporated IDC * are removed upon washing so that plaques that darken the X-ray film are those that express recombinant HSV TK gene. Neither CV-1 cells nor co-cups, although containing T, will phosphorylate and incorporate IDC * in the selective manner characteristic of HSV TK.

After the recombinant organisms have been identified by radioautography, filter plugs were cut from the nitrocellulose filter, placed in 1 ml of coating substrate (Eagle's Special, 10% calf serum), sonicated and replicated on CV-1 monolayers. The IDC * assay was then repeated to purify the purified viruses. In this way, a virus identical to the VP-2 mutant identified by hybridization in Example X was purified by a technique dependent on the expression of the HSV TK gene present therein.

Again, the results of this Example demonstrate the expression of the HSV TK gene present in the recombinant organisms of the present invention in some of these organisms.

The co-op mutants derived from pDP 137, namely VP-2, VP-4 and VP-6, will all express the HSV TK gene present therein by phosphorylation and incorporation of IDC * in the manner described. above. However, variants VP-1, VP-3 and VP-5, derived from pDP 132, will not express the gene, possibly because the gene's orientation within viruses opposes the alignment of gene transcription.

62 DK 174253 B1

Example XII - Use of a selective substrate for the identification and purification of recombinant viruses containing the HSV TK

Viruses prepared in accordance with Example X by in vivo recombination, in BHK-21. (C-13) cells, of VTK 79 coconut virus and a calcium orthophosphate precipitate of 5 pDP 137 were used to infect human (line 143) TKT cells. . More specifically, cell monolayers, in five petri dishes 6 cm in diameter, were each infected with the virus of Example X by diluting the virus from 10 ° to 0A in the presence of selective MTAGG substrate. The infecting technique was as described previously.

Five well-separated plaques were purified and one was replated on CV-1 monolayers for a second cycle of plaque purification. A further well-separated plaque, purified twice by plaque purification, was selected and analyzed. A well-cultured plaque, thus plaque-purified twice, was selected and analyzed for the presence of HSV TK gene by in situ hybridization using 32P-labeled Bam HSV TK. The hybridization technique was again as described previously. The mutant chickenpox virus, positive for the presence of the HSV TK gene, was designated VP-4.

Example XIII - Construction of pDP 120.

Approximately 20 micrograms of pDP 3 was digested with Pst I and the obtained fragments were separated on an agarose gel in a method analogous to that described in detail in Example I. The Pst I fragment having a molecular weight of 3.7 md, corresponding to co-cup Hind III F fragment, portion portion, purified.

Approximately 500 nanograms of this fragment were then ligated with 250 ng of pBR 322, previously digested with Pst I, in 20 microliters of O'Farrell buffer (OFB) (cf. O'Farrell et al., Molec. Gen. Genetics 179,421-435 (1980)). The buffer comprises 35 mM trisacetate (pH 7.9), 66 mM potassium acetate, 10 mM magnesium acetate, 100 pg / ml bovine serum albumin and 0.5 mM dithiothreitol. For ligation purposes, 1 mM adenosine triphosphate (ATP) and approximately 20 units were present (New England Biolabs). The mixture was maintained at 16 ° C for 16 hours.

The ligation mixture was then used to transform competent E. coli HB 101. Amps \ TetR recombinants were selected on appropriate antibiotic plates, analogous to the procedure described in Example III. Several recombinant plasmids were then analyzed by restriction analysis with Pst I and Barn HI, as in Example VII, to confirm their construction. A colony containing a plasmid of the correct construction was grown to a larger scale and recovered as in Example IV and designated pDP 120.

Example XIV - Construction of pDP 301 A and 301 B: Construction of VP 7 and VP 8., 10 Approximately 10 micrograms of placid pDP 3 was cleaved with Hind III, and the 8.6 md Co-P-F fragment was purified in an amount of 5 micrograms from an agarose gel using an analogous technique; The fragment was self-ligated by incubation for 16 hours at 16 ° C in 1.0 ml of 0'Farrell buffer (OFB) containing 1 mM ATP and 80 units of T4 DNA ligase.

After incubation, the reaction was terminated by heating to 65 ° C for ten minutes and the DN A was then precipitated with ethyl alcohol.

The self-ligated Hind III F fragment was then re-cleaved in 250 μΐ of OFB and digested for four hours with 30 units of Barn HI. The reaction was terminated by heating to 65 ° C for ten minutes and the resulting DNA fragments separated on a one percent agarose gel.

The band corresponding to the 8.6 md Hind III F fragment, which had been inverted around the Barn Hi site, was then cultured using techniques described previously.

This inverted Hind III F fragment was then inserted into the Barn HI site of pBR 322 as follows.

64 DK 174253 B1 pBR 322 was cleaved with Bam III using traditional techniques. The linear plasmid was then treated with calf intestinal alkaline phosphatase (CIAP) to remove the 5'-phosphates, thus counteracting recirculation (Chaconas et al., Methods in Enzymol. 65.75 (1980)). More specifically, 5 µg of linear pBR 322 in 400 μΐ of OFB, adjusted to pH 9.0, was combined with 0.75 5 units of CIAP (Boehringer Mannheim) for 30 minutes at 37 ° C. An additional 0.75 unit of CIAP was added and the mixture was digested for 30 minutes at 60 ° C. The DNA was then deproteinized with phenol extraction, as described in Example I, for the purification of coconut DNA.

Approximately 450 ng of the pBR 322 DNA was treated as above, then approximately 10,400 ng of inverted Hind III F fragment was ligated into 15 μΐ OFB containing 1 mM ATP and 20 h of T4 DNA ligase at 16 ° C. a period of 16 hours. The alloy mixture was then used directly to transform E. coli HB101 cells, as previously described in Example III.

The transformed bacteria were then sorted for AmpR, Tets recombinants by plating on appropriate antibiotic plates, again as described in Example III.

The recombinant plasmids were sorted by Hind III restriction analysis of minilysates (again as described in Example III) to determine which of the plasmids contained an inverted Hind III F fragment in either orientation. The two plasmids containing the fragment in different orientations were designated as pDP 301 A and pDP 301 B., respectively.

These plasmids were used to construct two new recombinant coco viruses, each containing pBR 322 DNA sequences, inserted into the Bam Hi site of the cocoa Hind III F fragment in opposite orientations.

More specifically, pDP 301 A and pDP 301 B were inserted into VTK ~ 79 by in vivo recombination, as described in Example X. However, 10 µg of donor DNA (either pDP 301 25 A or pDP 301B digested with Sst I) was used and 2 µg of VTK "79 carrier DNA to prepare the calcium orthophosphate precipitate used in addition to CV-1 cells infected with VTK" 79 as the rescue virus.

The recombinant viruses were then sorted using the replication filter technique, also presented in Example X, using incision-translated pBR 322 DNA as the exploding material.

The virus in which pBR 322 DNA sequences from pDP 301 A had been recombined in vivo with VTK ~ 79 was designated as VP 7: the recombinant virus containing pBR 322 DNA from pDP 301 5 B was designated as VP 8.

Example XV - Construction of dJZ 102 A / F: Construction of VP 9, 10 A plasmid containing the complete DNA sequence coding for the hemagglutinin (HA) gene of influenza virus A / PR / 8/34 inserted into pBR 322, is one of the plasmids made by Bacz et al. in Nucleic Acids Research 8, 5845-5858 (1980). The plasmid contains the complete nucleotide coding sequence for the HA gene, inserted at the Hind Hi site of pBR 322.

The hemagglutinin sequence in the plasmids was reversed by digesting approximately 500 ng of the original plasmid designated pJZ 102 A, with Hind III in OFB, then relegated, using T4 DNA ligase and ATP, as previously described.

The ligation mixture was then used to transform competent E. coli. and the bacteria were filtered off for AmpR, Tets colonies. Recombinant plasmids from minilysis preparations were then screened for hemagglutinin sequences, present in opposite orientations by Ava I digestion of the plasmids and assay on agarose gels. The plasmids in which the HA sequence was present in a direction, as opposed to that found in pJZ 102 A, were designated pJZ 102 B (cf. Figs.

9 A).

66 DK 174253 B1

Approximately 500 ng pJZ 102 A was linearized by digestion with Bam HI in OFB, as described previously. The linearized pJZ 102 A was equated with approximately 500 ng of inverted Hind III F fragment, the latter being conveniently obtained by Barn HI digestion of pDP 301 A (cf. Example XIV). Ligation occurred in 20 μΐ OFB, containing 1 mM ATP and 5 approximately 20 units of T4 DNA ligase at 16 ° C for a period of 16 hours.

The ligation mixture was used directly to transform competent E. coli RR 1 cells (Bolivar et al., Gene 2, 95-113 (1977)), as previously described.

Transformed cells were plated on ampicillin plates and sorted for recombinants by colony hybridization using incision-translated coconut Hind III F fragment DNA as exploding material, as previously described in Example VI.

Colonies DNA, found to be positive by hybridization, were then analyzed by Hind III restriction analysis and agarose gel electrophoresis.

A plasmid, containing pJZ 102 A, inserted into the Barn ΗΙ site of coconut Hind III F fragment was purified and designated as pJZ 102 A / F.

Using the in vivo recombination technique, described in detail in Example XV, 10 µg of circular donor DNA from pJZ 102 A / F was used for recombination, together with 2 µg of VTK-79 carrier DNA, into VTK "79 coconut virus. The recombinant virus, purified as such, was designated as VP 9 by the replication filter technique using an incision-translated Hind III HA fragment as the exploding material.

Example XVI - Construction of VP 10.

Plasmid pJZ 102 B (m.l. Fig. 10 A) was inserted into VP 7 with in vivo recombination using standard protocol using 10 µg circular donor pJZ 102 B DNA, 2 µg VTK_79 carrier DNA and CV1 cells. Sorting for recombinant viruses containing 67 HA sequences was by the replication filter technique, already described herein, using an incision-translated Hind III HA fragment as the exploding material.

A positive plaque was cultivated, the plaque cleaned and designed as VP 10.

Example XVII - Determination of the expression of the HA gene at VP 9 and VP 10.5 Two 6 cm petri dishes containing BHK.-21 cells in a nutrient substrate were infected with ca.

200 pfus A / PR / 8/34 influenza virus. Another pair of 6 cm petri dishes, containing a monolayer of CV-1 cells in a nutrient substrate, were infected with ca. 200 pfus VP 9 coconut variant, and a third pair of 6 cm petri dishes containing a CV-1 monolayer in a nutrient substrate, again, were infected with ca. 200 pfus VP 10.

Viruses were grown for 48 hours at 37 ° C and stained with Neutral Red for one hour at 37 ° C to form a clear image of the plaques. The nutrient substrate was then aspirated and the cell monolayers were washed three times with phosphate buffered saline (PBS) containing 1 mg / ml, bovine serum albumin (BSA).

1.5 ml of PBS-BSA containing 5 μΐ HI HA rabbit antiserum was then added to one of 15 of the three pairs of petri dishes, and the dishes were incubated for one hour at room temperature.

Another set of three cell cultures (one BHK and two CV-1 cultures) were treated with 1.5 ml of PBS-BSA containing 5 μΐ H3 HA rabbit antiserum and again incubated for one hour at room temperature.

Next, all the cell monolayers were washed three times with PBS-BSA, and then 1.5 ml of PBS-20 BSA, containing approximately 1 µCi, 25I-labeled protein A (New England Nuclear), was added to each of the 6 Petri dishes. The dishes were then incubated for approximately 30 minutes at room temperature and the radioactive material was absorbed. The cell monolayers were washed five times with PBS-BSA. The cell monolayer on each of the six plates was then embossed onto a corresponding nitrocellulose filter, and the latter was exposed to X-ray film for one to three days. The film was then developed.

The radioautographs showed complex formation in the petri dish in which the BHK cell monolayer had been infected with A / PR / 8/34 and treated with H1 HA antiserum. Similarly, exposed film was found for the CV-1 cell monolayer infected with VP 9 and treated with H1 HA antiserum, also indicative of antigen-antigen complex formation for this sample. All the other four samples were negative for complex formation.

Example XVIII - Determination of HA excretion by VP 9 in rabbits.

Two White New Zealand rabbits were each infected with 4 OD (at A260) units of purified 10 VP 9 mutant coconut virus, either intravenously (IV) (rabbit # 1) or intramuscular (IM) (rabbit # 2).

Each rabbit was tapped for blood and antiserum collected before infection (pre-immune serum) and 17 days, 25 days and 41 days after infection. Antiserum from each rabbit was tested for its ability to neutralize coconut virus infection as follows.

Sequential dilutions of the antiserans were prepared in standard virus plaque substrate (Eagle's Special Medium containing 2% fetal bovine serum), then mixed with an equal volume of infectious coconut virus (100-300 pfus). Each mixture was kept at 4 ° C overnight, then used to infect CV-1 monolayers in 60 mm petri dishes, as previously described. Specific coco-neutralizing antibodies present in the serum were indicated by a reduction in the total number of plaques formed. The results of such a test are shown in the following Table I.

69 DK 174253 B1

TABLE I

Measurement of coconut neutralizing antagonists induced by YP 9 in rabbits

About plaque- Final dilution of antiserum reducing (% of control indicated plaque reduction 5 Rabbit # 1 Rabbit # 2 (ΙΥΊ_ (1M) 17 days 50% 1: 64000 1: 16000 90% 1: 10000 1: 2000 10 25 days 50 % 1: 128000 1: 8000 90% 1: 32000 1: 2000 41 days 50% 1: 128000 1: 16000 15 90% 1: 32000 1: 2000

As a control, preimmune antiserum on a 1:20 dilution, or no antiserum, yielded approximately 180 plaques in the above rabbit sample. 1 and 240 plaques for rabbit no. 2nd

One, 25 I protein A sample was performed as described in Example XVII. More specifically, three mono-20 layers of BHK-21 cells infected with A / PR / 8/34 serum were treated with 25 μΐ of either anti-A / PR / 8/34 serum, antibody serum or -VP 9 serum from the 45-day blood draw of rabbit Nr. First

Washing and treatment with 25 I protein A of each monolayer was as described in Example XVII.

Likewise, three monolayers of cells infected with coconut virus (S variant) were treated with the same three antiserum preparations and reacted with 125 I protein A, as previously described.

The results are presented in the following Table, and they indicate that VP 9 may bring from day a rabbit production of antidote to influenza hemagglutinin, expressed by 10 VP 9.

TABLE II

Antiserum BHK cells CV-1 cells infected with coconut virus ______ A / PR / 8 / 34_ (S-variant / 15 Anti-A / PR / 8/34 +

Anti-cooker - +

Anti-VP 9 - +

The production of HA antibodies by a VP 9-infected rabbit was also tested by measuring hemagglutinin inhibition (HI) titers.

Namely, HI tests were performed on the pre-immune, 25-day and 41-day rabbit serum no.

1 (IV), in accordance with standard protocol, described in detail in "Advanced Laboratory Techniques for Influenza Diagnosis, Immunology Series No. 6, Procedural Guide 71 DK 174253 B1 1975", U.S. Dept. HEW, Public Health Service, Center for Disease Control, Bureau of Laboratories, Atlanta, Georgia.

Table III below shows the HI titers determined using 3+ HA units with the various antisera tested. Extracts of BHK cells infected with A / PR / 8/34 influenza virus were used as a III hemagglutinin source.

TABLE III

hemagglutination-inhibition

Antiserum Ten three

Anti-A / PR / 8/34 greater than 1: 320 10 Pre-immune serum less than 1: 10 25 day antiserum 1: 60 41 day antiserum greater than 1: 320

Example XIX - Construction of pDO 250 A and 250 B: construction of VP 11 and YP 12.

A plasmid pTHBV 1 can be constructed by the methods described by Hirschman et al., Proc.

Natl. Acad. Sci. USA 77,5507-5511 (1980), Christman et al., Proc. Natl. Acad. Sci. USA 79, 1815-1819 (1982).

Ca. Twenty micrograms of pTHBV 1 plasmid, containing two HBV genomes (subtype ayw) in a head-to-tail arrangement inserted into the Eco RI site of pBR 322, was cleaved with Bgl II restriction endonuclease.

The fragments were separated on a 1.0 percent agarose gel, and a 1.5 md fragment of HBV DNA sequences encoding the HBV surface antigen and pre-surface antigen was isolated (cf. Galibert et al., Op. Cit.

72 DK 174253 B1

Ca. 500 ng of pDP 120 was partially digested with Bam HI under conditions similar to those described in Example V earlier herein.

The resulting digestive product was then treated with calf intestinal alkaline phosphatase (CIAP), in a manner analogous to that described earlier herein in Example XIV. Finally, the aforementioned fragment of the Bgl II cleaved pTHBV 1 plasmid was similarly digested into the Bam HJ site of the CIAP-treated pDP 120 Bam HI digest product under conditions similar to those described herein herein.

The ligation mixture was used directly to transform competent E. coli RR 1 cells, also as previously described.

The resulting Amps, TetR colonies were sorted for recombinant plasmids by digesting minilysates of possible recombinants with Xho I and Pst I and assay on an agarose gel.

Two recombinant plasmids were cultured, corresponding to insertion into the Bam HI site, present in the coco portion of pDP 120, of the HBV Bgl II fragment in each of two possible directions. The plasmids were designated as pDP 250 A and pDP 250 B (cf. Fig. 11).

Finally, recombinant coco viruses containing the HBV surface antigen and pre-surface antigen sequences were constructed by in vivo recombination (see Example 10) using 20 µg each of either circular pDP 250 A or pDP 250 B and 2 µg VTK ~ 79 carrier DNA, with VTK_79 as the infecting virus, as previously described.

Viruses were screened for recombinants using the replication filter technique with incision-translated pTHBV DNA as a test material.

The resulting recombinant coco viruses, containing the Bgl II fragments of HBV virus in one of two directions, were designated as VP 11 and VP 12, as shown in Figures 1 ID.

73 DK 174253 B1 and E. (The normal direction of HBV transcription is indicated for the plasmids of Figs.

11D.)

Example XX - Construction of pDP 252: construction of VP 13.

20 µg of plasmid pTHBV 1 (cf. Example XIX) was digested with Hha I restriction endonuclease. The largest fragment of Hha I digestion product comprises 1084 base pairs and contains the complete sequence of hepatitis B virus coding for the surface antigen, without the region coding for the pre-surface antigen (cf. Galibert et al., Op. Cit.

This fragment was inserted into pBR 322 at the Hind III site, using Hind III couplings. More specifically, approximately, 400 ngHBV Hha I fragment, purified from a preparative gel, 10 as previously described, was treated with six units of T4 DNA polymerase (P / L Biochemicals) present in 40 μΐ OFB also containing 2 mM each of deoxyadenosine triphosphate (dATP), deoxyguanidine triphosphate (dGTP), deoxycytosine triphosphate (dCTP) and deoxytimin triphosphate (dTTP). The mixture was incubated at 37 ° C for 30 minutes to trim the 3'-end protruding ends of the fragment generated by Hha I restriction endonuclease (cf. O'Farrell 15 et al., Op. Cit. 1

After the reaction period, approximately 500 ng of phosphorylated Hind III (Collaborative Research) couplings, 2.5 μΐ 20 mM adenosine triphosphate, 1 μΐ 100 mM spermidine (Cal Biochem.) And 1 μΐ (approximately 80 units) T4 DNA ligase were added. continued at 10 ° C for sixteen hours.

The reaction was stopped by heating at 65 ° C for ten minutes, and 400 ng of pBR 322 was added thereto.

The Hind Hi couplings and pBR 322 were then cleaved by adding approximately 20 units of Hind III and digesting the mixture at 37 ° C for four hours.

74 DK 174253 B1

Once thereto, the reaction was stopped by heating at 65 ° C for ten minutes and the unlabeled couplings were removed by spermine precipitadone, in accordance with Hoopes et al., Nucleic Acids Research 9, 5493 (1981).

More specifically, 2.5 μΐ 0.2 M spermine in H2O was added to the reaction mixture to make it 5 10 mM in spermine. The reaction mixture was incubated on ice for 15 minutes and the resulting precipitate was collected by centrifugation. Residual semen was removed from the DNA by re-straining the DNA tablet in 75 percent ethyl alcohol, 0.3 M sodium acetate and 10 mM magnesium acetate. This mixture was incubated on ice for 60 minutes. Residual semen is dissolved in the ethyl alcohol, leaving a suspension of DNA which was again compressed into small beads by centrifugation and redissolved in 20 μΐ OFB containing 1 mM ATP and approximately 20 units of T4 DNA ligase. Ligation of pBR 322 and Hind III coupled fragment was performed for 16 hours at 10 ° C.

The ligation mixture was then used directly to transform the competent E. coli RR1 cell as previously described. The transformants were plated on ampicillin plates and separated by colony hybridization, as previously described herein. An incision-translated HBV Hha I fragment was used as research material.

Colonies that were found to be positive by hybridization were analyzed by restriction digestion of minilysates. A plasmid containing the Hha I fragment inserted at the Hind III site was characterized and designated as pDP 252.

A recombinant coconut virus containing the Hha I-IBV fragment coding for the HBV surface antigen was constructed using the standard in vivo recombination protocol as set forth in Example XV using 10 µg circular pDP 252 as the donor DNA with 2 µg VP 8 DNA as the carrier DNA and VP 8 as the infecting virus for CV-1 cells.

75 DK 174253 B1

Viruses were sorted for recombinants using the replication filter technique with a cut-translated Hha I HBV fragment as test material.

The resulting recombinant virus was designated as VP 13.

Example XXI - Construction of pBL 520 A and 520 B: construction of VP 14 and VP 16.

Approximately 20 µg of herpesvirus type I, KOS, DNA strain, extracted as described by Pignatti et al., Virology 93, 260-264 (1979), were digested with Eco RI and the resulting fragments were separated on an agarose. gel. Eco RI fragment F was cultured from the gel using traditional techniques.

Ca. 200 ng of the Eco RI fragment F was ligated with 60 ng of pBR 322, digested with Eco 10 RI and later treated with calf intestinal calcium phosphate in a manner described herein earlier in Example XIV. The CIAP-treated pBR 322 and the Eco RI F fragment were ligated into 20 μΐ OFB containing 1 mM ATP and approximately 80 units of T4 DNA ligase at 16 ° C for a period of 16 hours.

The entire ligation mixture was used to transform competent E. coli RR I cells, as described in previous examples.

The transformed E. coli was grown on ampicillin plates and the AmpR, TetR transformed E. coli were sorted for recombinant plasmids by restriction analysis of minilysates, as previously described. Restriction analysis was done with Hpa I and Eco RI to determine the orientation of insertion of the Eco RI F fragment into the plasmid.

The two plasmids thus obtained which have the HS V Eco Ri F fragment inserted into pBR 322 in each of two opposite orientations were designated pBL 520 A and pBL 520 B, as shown in FIG. 13B.

76 DK 174253 B1

Two new co-copper recombinants VP 14 and VP 16 were constructed by in vivo recombination techniques described in Examples X and XV, using these plasmids and VP 7. More specifically, 20 µg each of pBL 250 A or pBL 250 B, 2 and VP 7 are supported. DNA and VP 7 virus were used to treat CV-1 cells to effect the in vivo recombination. Recombinant viruses were sorted by the replication filter technique, using incision-translated HSV Eco RI F fragment as exploding material.

Example XXII - Construction of pBL 522 A and 522 B: construction of VP 17 and VP 18.

Approximately 20 µg of pBL 520 A (cf. Example XXI) was digested with Barn HI and the resulting fragments were separated on agarose gel. A 5.1 md fragment, corresponding to the Barn 10 HI C fragment of HSV DNA (KOS strain), was purified from the gel using techniques such as those described in Example I.

Plasmid pDP 120 was partially digested with Bam HI to linearize the plasmid, using techniques analogous to those described previously in Example V. The digested plasmid was further treated with calf intestinal calcium phosphatase as in Example XIV to prevent recirculation.

Approximately 100 ngpDP 120 DNA, thus treated, was ligated with 120 ng of the previously described Barn HI G fragment in 20 µl OPB containing 1 mM ATP and approximately 80 units of T4 DNA ligase at 16 ° C for a period of 16 hours.

Then, the ligation mixture was used directly to transform competent E. coli RR I 20 cells, as described in previous Examples.

The transformed E. coli were then sorted for Amps, TetR recombinants by colony hybridization, as previously described, using HSV ECO'RI fragment as exploding material. Colonies positive by hybridization were sorted by restriction analysis of minilysates with Barn HI and Sst I to determine if the complete 77 fragment was inserted and to determine its orientation within the resulting plasmid.

Two recombinant plasmids were found in this way, each containing the HSV Bam HI G fragment in opposite orientations in the parental plasmid pDP 120. The new plasmids 5 were designated pBL 522 A and pBL 522 B.

Again, using the in vivo recombination technique described in detail in Examples X and XV herein, 20 µg of donor pBL 522 A or B - combined with 2 µg of carrier VTK ~ 79 DNA, respectively, was generated to form a calcium orthophosphate precipitate. This and the coconut virus VTK 79 were used to treat CV-1 cells, with the production of two virus mutants, 10 designated as VP 17 and VP 18, respectively.

The coconut mutants were identified by the replication filter technique with HSV Eco RI F fragment as exploding material.

Example XXIII - Construction of an L variant TK ~ coconut virus from the TK ~ S variant.

In wild-type coconut virus, it is known that the coconut TK gene is present in Hind III J fragments (Hruby et al., J. Vitol. 43,403-409 (1982)). Accordingly, the Hind III J fragment of the TK “79 S variant cupping virus of Example VIII must have a mutation in the TK gene which inactivates the gene.

A TK "L variant cobweb virus was derived as follows, using the Hind III J fragment of the TK" 79 S variant.

The Hind III J fragment of TK_79 was inserted into pBR 322 in a manner similar to that of the Hind III F fragment of Example II. The resulting plasmid was used in the standard in vivo recombination protocol, in particular using 10 µg of the plasmid donor DNA, 2 µg of Invariant Coconut DNA carrier and L-variant Coconut Virus-infected CV-1 cells.

78 DK 174253 B1

Offspring virus was used to infect human TK "cells (line 143) (described previously) in the presence of 40 μg BUdR. Growing viruses were plaque-purified in the presence of BUdR, and viruses from a single plaque were selected and signed. with VTK “79 L (ATCC No. 2056) It cannot be determined whether the new virus is a spontaneous mutation, 5 as a recombinant containing the J fragment of the TK ~ 79 S variant: the latter is more likely.

Example XXIV - Construction of pDP 202.

Ca. 34 µg of L-variant coconut virus DNA was completely digested in Ava I buffer (20 mM Tris-HCl (pH 7.4), 30 mM Nacl, 10 mM MgCl2l with Ava I restriction endonuclease, and the resulting fragments were separated on an agarose gel). as previously described, the Ava I H fragment was then purified from the agarose gel.

Approximately 400 ng of pBR 322 in 50 µl of Hind III buffer was completely digested with Hind III. Reaction was terminated by heating at 65 ° C for ten minutes, at which time 45 μΐ (approximately 600 ng) of the pure-grown Copper A va I H fragment was added. Then the whole mixture was precipitated with ethyl alcohol. The resulting DNA tablet was again illuminated in 9.5 µl of 4DNA polymerase buffer (20 mM tris-HCl (pH 7.6), 10 mMMgCl The 5 'ends protruding from the DNA fragments were filled in by adding 1.5 units of T4 DNA polymerase and incubating at 37 ° C. After 30 minutes, 0.5 μΐ of a 0.02 M solution of ATP was added to make the reaction mixture 1 mM, with respect to ATP, together with 1 μ (approximately 80 units) of T4 DNA ligase. Ligation was then carried out at 10 ° C for 20 hours.

The entire ligation mixture was used directly to transform competent E. coli HB 101.

Transformed bacteria were plated on nitrocellulose filters applied to ampicillin plates. Recombinant colonies were sorted, by colony hybridization, using incision-translated 25-cup Ava I H fragment as exploding material.

79 DK 174253 B1

Plasmids cultured from colonies positive by colony hybridization were digested with Hind III and assayed on an agarose gel as previously described. One such plasmid containing an Ava I fragment fragment inserted at the Hind ΙΠ site of pBR 322 was purified and designated pDP 202. Further characterization of this plasmid by restriction analysis with Sst I and Bam III determined the orientation of the fragment before in the plasmid (cf. Fig. 15A).

Example XXV - Construction of plasmids pDP 202 TK / A-F: construction of VP 22.

Plasmids pDP 202 TK / AF were constructed by inserting the Bgl II / Bam HI TK fragment of HSV into each of the three Barn HI sites in the cocoa Ava I H fragment portion of pDP 10 202. The Bgl II / BamHI fragment contains the coding region for the HSV TK gene, but not the associated HSV activator sequence (McNight et al., Cell 25, 385-398 (1981)).

This was accomplished first by culturing a linear pDP 202 pIasmid (7.3 md) that had been linearized at a Barn ΗΙ site by partial digestion of the plasmid with Bam HI. Bgl II / Bam HI TK fragment was prepared by digesting the HSV Bam TK plasmid (m.p.

Example V) with Bgl II and Bam HT. Bgl II digestion cleaves the Bam HI TK fragment at one site, resulting in a 1.8 md fragment containing the coding region of the TK gene and a 0.5 md fragment corresponding to the 5 'end of the Bam HI TIC fragment, containing the HSV activator. The 1.8 md Bgl II / Bam Hi fragment was isolated from an agarose gel.

To construct the plasmids pDP 202 TK / AF, approximately 500 ng Bam HI linear 20 pDP 202, which had been treated with CIAP, was previously ligated with 250 ng of the aforementioned Bgl II / Bam HI TIC fragment in 20 μΐ OFB, containing 1 mM ATP and approximately 100 units of T4 DNA ligase at 16 ° C for 16 hours. The entire ligation mixture was then used to transform competent E. coli RR I cells, as previously described. Transformed cells were plated on ampicillin plates and the colonies were sorted for recombinant plasmids by restriction analysis of minilysates with Bam HI to determine at which Bam ΗΙ site, Bgl II / Bam HI TK fragment was inserted and in which orientation.

80 DK 174253 B1

The orientation site and direction were confirmed by restriction analysis with Sst I. With this method, it was found that the Bgl II / Bam HI TK fragment was inserted into each of the three Bam HI sites of the cocoa Ava I H fragment in both orientations. Each plasmid pDP 202 TK received a letter designation from P, to F (cf. Fig. 15C).

Preparative amounts of plasmids were then grown and purified and used to construct recombinant viruses, using standard in vivo recombination protocol of Examples X and XV. That is, approximately 20 µg of donor DNA from each recombinant plasmid was mixed with 2 µg of VTK-79 DNA as a carrier and added, in the form of a calcium phosphate precipitate, to a monolayer of CV-1 cells infected with VTK ~ 79 L virus. Recombinant viruses 10 were sorted using the replication filter technique described previously using incision-translated HSV Bam TK DNA as exploding material.

A recombinant virus isolated from in vivo recombination of VTKT79 L and pDP 202 TKÄ was cultured and designed with VP 22. This mutant virus was of particular interest because it induced a higher level of HSV TIK activity in infected cells than 15 VP 2. , VP 4 or VP 6 (previously described herein), as measured by a 125 I-iododesoxycytidine (IDC) test.

More specifically, either L-variant cup VP 4 or VP 22 was used to infect monolayers of CV-1 cells at appropriate dilutions to yield 50 to 200 plaques per cell. 60 mm Petri dish.

Each dish was then treated with 125 I IDC and washed, as previously described in Example 20 XI, to compare the levels of HSV TK activity.

Infected cell monolayers were then lifted onto nitrocellulose filters placed on a single sheet of X-ray film to compare the levels of TK action by comparing the relative exposure (darkening) of the film to each filter.

The results of the IDC test indicated that the L variant coke contained no HSV TK-25 effect and therefore did not expose the film. VP 4, shown earlier in Example XI to contain DK 174253 B1

SI

the HSV TK effect caused a slight darkening of the film, VP 22 caused 5-10 times as much exposure as VP 4, indicating a significantly higher level of HSV TK effect.

Claims (5)

A method of producing a biological product, wherein a culture of eukaryotic cells is infected with a recombinant coconut virus, characterized in that the infected virus in a non-essential region of the coconut genome contains a DNA segment that does not occur naturally. in coconut viruses, which are expressed in infected eukaryotic cells or host organisms and encode the biological product, and that the infected cells are cultured under conditions that allow expression of the biological product, after which the biological product is isolated from the cell culture. 2. A method of replicating DNA in the eukaryotic cell, characterized by infecting the cell with a recombinant coconut virus modified to contain the DNA which is exogenous to the coconut virus. Method according to claim 2, characterized in that the exogenous DNA is expressed by the cell, thereby forming a protein. Method according to claim 3, characterized in that the expression ice is controlled by the control of the smallpox virus. Modified recombinant coconut virus for use in the method of claim 1, characterized in that it contains, in a non-essential region of the coconut genome, a DNA segment that does not occur naturally in coconut virus and is expressed in 20 infected eukaryotic cells or host organisms. codes for the biological product.