Method for Deletion of a Gene From a Bacteria

In the example in U.S. Ser. No. 688,612 filed Jan. 3, 1985, transformations were done into E. coli strain JC7623, which carries the recBC and sbcB mutant alleles. The recBC and sbcB mutations inactivate exonucleases which degrade linear DNA fragments, as described by P. J. Goldmark & S. Linn, J. Biol. Chem 247:1849-1860 (1972) and S.R. Kushner et al, Proc. Natl. Acad. Sci. U.S.A. 68:824 827 (1971). The sbcB mutation also suppresses the Rec- phenotype of recBC cells, so that homologous recombination functions via an alternative pathway. Because alaS is an essential gene, transformations were done into JC7623(pMJ901). Plasmid pMJ901 has a temperature-sensitive replicon and a wild-type allele of alaS. This plasmid maintains cell viability when the chromosomal copy of alaS has been deleted, but the resulting cells have temperature-sensitive phenotype.

In a recA+ background or its equivalent, integration of plasmid pMJ901 can occur through homologous recombination at residual alaS sequences on the interrupted chromosome. Because recA is adjacent to alaS on the chromosome, as disclosed by B.J. Bachman, Microbiol. Rev. 44:1-56(1980), it was possible to design linear DNA fragments which, when transformed into E. coli, simultaneously deleted a portion of recA and greater than 90% of alaS.

Two deletions, (.DELTA.alaS1) and (.DELTA.alaS2) were constructed from DNA fragments encoding an N-terminal fragment of the recA protein and a C-terminal segment of the alaS protein. For each linear DNA fragment, the recA segment and its flanking 5' sequences are about 1.5 kb pairs; homologous recombination at these sequences destroys the carboxy-terminal coding region of recA protein, thereby inactivating it. The alaS sequences associated with the linear fragment are 1,100 bp (.DELTA.alaS1) and 225 bp (.DELTA.alaS2). The (.DELTA.alaS1) deletion effectively removes the amino-terminal on-third of the 2,625 bp alaS coding region. All but the carboxy terminal 225 nucleotides are removed in the .DELTA.alaS2 deletion.

Linear DNA fragments were derived from PvuI-BamHI-cut plasmid pMJ520 (.DELTA.alaS1) and BamHI-cleaved pMJ525 (alaS2). Cells transformed with the linear DNA fragments were first selected for Kan.sup.r ; these transformants were then screened for the Amp.sup.S and UV and temperature-sensitive phenotypes.

The organization of sequences in strains containing the alaS1 and alaS2 alleles was examined by Southern blot hybridizations. The controls were the parental strains JC7623 and JC7623(pMJ901). A 1.8 kb BamHI-EcoRI hybridization probe was used to differentiate between the various gene arrangements. This probe contains a portion of recA and flanking 5' sequences, and is described by A. Sancar et al in Proc. Natl. Acad. Sci. U.S.A., 76:3144-3148 (1979).

A PstI digest of chromosomal DNA from each of the four strains shows a 7.8 kb band which corresponds to the 5' end of recA, together with several kilobases of upstream sequences; this region is undisturbed in all cases. For JC7623 and JC7623(pMJ901) a 3.5 kb fragment also hybridizes to the probe. This corresponds to the PstI fragment which extends from an internal site within recA to a site within the carboxyl-terminal coding portion of alaS. For the two deletions, this band is missing and is replaced by a 1.1 kb fragment which extends from the internal PstI site within recA to the downstream PstI site in the Kan.sup.r gene.

Deletion of the single chromosomal copy of alaS enables rigorous testing for complementation of gene fragments that encode pieces of the deleted alanyl tRNA synthetase encoding gene. The gene fragments can be introduced via transformation of a pBR322-derived multicopy plasmid, or other vector. Only where the gene deletion removes all of the gene encoding proteins capable of complementing pieces of the protein to restore activity does one have a system that can be used to accurately measure the functional activity of a gene sequence.

The glyS locus was deleted from E. Coli using a procedure based upon site specific recombination with a linear DNA fragment that replaces the target gene, in this case glyS, with a Kan.sup.r marker that is bounded by glyS 5' and 3' adjoining sequences. This recombination is done in E. coli JC7623 (recBC sbcB) where recBC sbcB mutations inactivate exonucleases that degrade linear DNA fragments. When the chromosomal copy of glyS is deleted to give the .DELTA.glyS strain TM101, cell viability is maintained with plasmid pMT 901. Plasmid pMI 901 encodes a HindIII fragment that contains glyS cloned into the HindIII site of plasmid pMT101. ; Plasmid pMT 101, encoding Cm.sup.R and Tc.sup.R and a temperature-sensitive replicon such that plasmid replication is blocked at the restrictive temperature (42.degree. C.), is constructed by inserting the Cm.sup.R locus from pBR325 into pPM103, which encodes TcR and a temperature-sensitive replicon. By virtue of the temperature sensitive replicon, plasmid replication is blocked at the restrictive temperature of 42.degree. C.

Cell survival therefore depends entirely on the ability of a second plasmid to complement the glyS null allele. This second plasmid is introduced by transformation at the permissive temperature and retained at the temperature that is restrictive for the first temperature-sensitive plasmid. The second plasmid contains specific glyS sequences to be tested.

Functional recA activity is required for the site specific recombination of linear DNA fragments. The recA+ phenotype must be converted to recA- to prevent recombination between plasmid-borne sequences and the chromosome. In the present example, it is not possible to inactivate recA simultaneously with the deletion of glyS by a double reciprocal recombination of a single linear DNA fragment. A two-step procedure is required in which the glyS deletion in TM101/MT901 is first created and a gene for a functionally deficient glycine tRNA synthetase, the recA 56 allele in E. coli GW554, moved into this strain by P1 transduction. The resulting temperature-sensitive .DELTA.glyS recA strain is designated TM 102/pMT 901.

The glyS deletion in TM101/pMT901 was created as follows. Plasmid pTK101 contains a 10 kbp PstI fragment which encompasses glyS and several kbp of 5'- and 3'-adjoining sequences. The central 5 kbp HindIII fragment of the PstI segment encoding glyS was replaced with a 3.3 kbp HindIII fragment encoding Kan.sup.r from bacterial transposon Tn5. One microgram of the resulting plasmid, pMT300, was digested with EcoRI and Ndel at unique sites in the pBR322 portion of pMT300, and transformed into E. coli JC7623/pMT901. Selection for Kan.sup.r yielded approximately 70 transformants. These transformants were presumed to have the .DELTA.glyS/Kan.sup.r allele and one of them was designated as TM101/pMT901. After the recA56 allele was moved by P1 transduction into TM101/pMT901 to produce TM102/pMT901, both .DELTA.glyS and recA56 alleles were moved by P1 transduction from TM102/pM901 to a wild-type C600 strain to produce the .DELTA.glyS/Kan.sup.r recA strain TM202/p901. Linkage of Kan.sup.r to xyl-5 in this strain was shown to be similar (90%) to that reported for linkage of glyS to xyl (80%). The genomic arrangement predicted for .DELTA.glyS Kan.sup.r was also verified by Southern blot analysis.

Further details of this example are discussed in the patent application U.S. Ser. No. 833,920 filed Feb. 26, 1986 entitled "Construction and Application of Polyproteins" by Matthew J. Toth and Paul R. Schimmel and in "Internal Structural Features of E. coli Glycyl-tRNA Synthetase Examined by Subunit Polypeptide Chain Fusions" in J. Biol. Chem. 26115), 6643-6646 (1986).

As demonstrated, the glyS gene was not permanently deleted from the E. coli unless the recA gene was also deleted or inactivated. In the presence of recA functional activity, it appears that the first maintenance plasmid encoding the essential gene is integrated into the cell chromosome under the intense selection pressure created by the removal of the plasmid containing the temperature sensitive replicon by growing the organisms at an elevated temperature. In contrast to the first example requiring the "immediate" removal or inactivation of the recA gene, removal occurs simultaneously with deletion of the gene. In the present case, removal occurs over a period of one to two days. The exact time required is a function of a number of factors which may be determined by one of skill in the art. The selection pressure is a result of the interplay of the means available to maintain the system in the absence of the essential gene, the presence of extrachromosomal DNA in the cell which encodes the essential gene, the method used to remove or inactivate the recA gene, the conditions under which the cell is grown (or not grown, i.e., the status quo could be maintained indefinitely by freezing or lyophilization), and other factors known to those of reasonable skill in the art.

The P1 transduction was done under standard conditions, as described by J. Miller in Experiments in Molecular Biology, pages 201-205 (Cold Spring Harbor, New York, 19720. This took approximately two days to test. "Stressing" in the case of the glyS gene deletion was to raise the temperature to a level at which the maintenance plasmid could not replicate. Since only those organisms having a functional gene encoding glycine tRNA synthetase incorporated into the chromosome can survive, there is intense selection for recombination of the plasmid gene with the cell chromosome.

The linear fragments which are useful in the present invention can be constructed using methods known to those skilled in the art. Techniques including subtraction methods can be used to isolate the recA gene and flanking sequences (for example, using RecA- and RecA+ strains). Other sequences can be prepared using an oligonucleotide synthesizer since less than 125 to 150 nucleotides of homologous sequence on either side of the gene to be deleted are required for the double reciprocal recombination. The partial sequences for a number of recA genes and RecA proteins have been published (see, for example, Mol.Gen.Genet. 204, 161-165 (1986); T. Moris et al, Proc.Natl.Acad.Sci. USA 77, 313-317 (1980); A. Sancar et al, Proc.Natl.Acad.Sci. USA 77, 2611-2615 (1980)). In numerous other cases, the protein has been purified and could be sequenced or probes have been isolated which hybridize to either the 3' or 5' ends of the gene.

Modifications and variations of the present invention, a method and means for deleting a gene by double reciprocal recombination in combination with manipulations of recA in a variety of bacteria, will be obvious of those in the art from the foregoing detailed description of the invention. Such modifications and variations are intended to come within the scope of the appended claims.