WO1989009776A1

WO1989009776A1 - VirE OPERON-ENCODED POLYPEPTIDES THAT BIND SINGLE-STRANDED DNA

Info

Publication number: WO1989009776A1
Application number: PCT/US1989/001129
Authority: WO
Inventors: Anath Das
Original assignee: Regents Of The University Of Minnesota
Priority date: 1988-04-14
Filing date: 1989-03-20
Publication date: 1989-10-19

Abstract

An ssDNA-binding polypeptide is provided which consists essentially of the formula: Ala-Asp-Lys-Tyr-Ser-Arg-Asp-Phe-Val-Arg-Pro-Glu-Pro-Ala-Ser-Arg-Pro-Ile-Ser-Asp-Ser-Arg-Arg-Ile-Tyr-Glu-Ser-Arg, which can be employed for the stabilization of ssDNA comprising exogeneous genetic information, thus enabling its direct introduction into cells.

Description

VirE OPERON-ENCODED POLYPEPTIDES THAT BIND SINGLE-STRANDED DNA

Grant Information

This invention was made with support from the National Institutes of Health under Grant No. GM 37555. The Government has certain rights in the invention.

Background of the Invention

Agrobacterium tumefaciens, a phytopathogenic soil bacterium, causes crown gall tumors on most dicotyledonous plants. During infection, agrobacteria donate a 15-20 kilobase (kb) segment of the Ti plasmidborne DNA to plant cells. The transferred DNA (T-DNA) becomes integrated into the plant nuclear genome and is maintained stably thereafter. See E. Nester et al., Ann. Rev. Plant Physiol., 35, 387 (1984). Two cis-acting border sequences consisting of 24 base pair imperfect direct repeats are required for Agrobacterium-mediated DNA transfer to plant cells. Recent studies demonstrate that under induction conditions, site-specific and strand-specific, cleavages occur within these border sequences. These cleavages lead to the formation of a single-stranded (ss) DNA molecule comprised of the T (-transferred) strand of T-DNA. It is likely that the ssDNA molecules are intermediates in the DNA transfer process.

The Ti plasmid virulence (vir) region is located left of the T-DNA segment and encompasses about 35 kb of DNA. This region contains six complementation groups. Of these two, virA and virG, are regulatory loci and are transcriptionally active in free living bacteria. The other four loci, virB, -C, -D and -E, are strongly induced wnen bacteria and plant cells are cocultivated together. These operons are believed to be involved in the transfer ana integration of T-DNA into the plant nuclear genome. For example, it has recently been shown that one or two of the polypeptides encoded by the virD operon catalyzes site-specific cleavages at the T-DNA borders. (M. Yanofsky et al., Cell, 47, 471 (1987)).

A second operon, virE, has been investigated in some detail. DNA sequence analysis showed that the virE operon can encode two polypeptides of apparent mass of 7,000 and 60,500 daltons, respectively. [s. Winans et al., Nucl. Acids Res., 15, 825 (1987)].

Genetic complementation studies revealed that a strain bearing a mutation in the virE locus can be complemented to form tumors on test plants if coinfected with a second Agrobacterium strain that contains wild-type vir loci but no T-DNA. [L. Otten et al., Mol. Gen. Genet., 195, 159 (1984)]. The complementing strain must contain four other vir loci, viz. virA, virB, virD and virG, in addition to the virE locus. In transient assays, where integration of T-DNA was not required, virE was found to be a nonessential locus [R. Gardner et al., Science, 231, 725 (1986)]. These findings indicate that while virE is not absolutely necessary for DNA transfer, it is essential for T-DNA integration. Recently, C. Gietl et al., PNAS USA, 84, 9006

(December 1987) used synthetic oligodeoxy-nucleotides to probe crude protein extracts of A. tumefaciens comprising mutants of virulence loci. Four complexes involving sequence-nonspecific, ssDNA-binding proteins were identified. Oligonucleotides with 36 and more bases gave a DNA complex with the inducible protein determined by the virE2 region. Based on DNA-sequence homology, C. Gietl et al. identified four regions of the virE2 polypeptide that may play a role in ssDNA binding within the virE2 polypeptide; the furthest toward the carboxyl terminus encompassing Arg-342 to Arg-412.

To date, the introduction and expression of foreign structural genes within plant cells via their direct transformation with recombinant plasmids has met with limited success, as have attempts to directly introduce other types of "naked" foreign DNA into plant cells. When naked DNA is used for delivery to cells, it is often observed that the integrated DNA has a different physical map, indicating that significant DNA rearrangement has occurred prior to or during integration. Consequently, the location of two adjacent pieces of DNA following integration is totally unpredictable. In contrast, in Ti-mediated gene transfer, although large segments of DNA (10-15 kb) are transferred to plant cells, a relatively smaller degree of rearrangement is usually observed. However, mono-cotyledonus plants are insensitive to Agrobacterium, severely limiting the usefulness of this technique to transform crops such as corn, wheat, barley, oats and rye.

Therefore, a need exists for improved reagents and methods for the direct introduction of foreign DNA, e.g., DNA comprising useful foreign structural genes, into cells.

Summary of the Invention The present invention provides a synthetic operon consisting essentially of a strong regulatable promoter in a correct reading frame upstream from a DNA segment corresponding to a portion of the virE2 gene. The virE2 gene or fragment thereof encodes a polypeptide which nonspecifically binds to single-stranded DNA. The synthetic operon is constructed to exclude the virE2 promoter. When inserted into a suitable vector, such as a plasmid or a phage, and employed to transform or transfect a suitable host, overproduction of the virE2 encoded gene product can be achieved. For example, overproduction of a virE2 encoded gene product in Escherichia coli was achieved by construction of an operon fusion of the virE2 gene with the E . coli tryptophan (trp) operon. Following transformation of E. coli with a recombinant plasmid comprising this operon, and induction of the transformant, the virE2 gene product partitioned into the insoluble membrane fraction. Other strong promoters are known to the art, which would be useful in the present operon, e.g., tac, the lambda phage promoter (PL,PR) and the T7 phage promoter.

DNA-protein binding experiments showed that a strong, single-stranded (ss)DNA binding activity was present in solubilized protein fractions containing the virE2 gene product. The binding was highly specific with little or no binding observed with either double stranded DNA or ssRNA. No significant binding to Ti plasmid DNA sequences was observed. Protein blotting studies indicated that the ssDNA binding activity was associated with the 68 kD virE2 polypeptide.

Selective mutations introduced into the virE2 gene lead to the identification of segments of the 68 kD virE2 polypeptide which retain a substantial portion of the ssDNA binding activity of the "native" virE2 polypeptide. One such polypeptide, which is a preferred embodiment of the present invention, is of the formula (I), below:

Ala-Asp-Lys-Tyr-Ser-Arg-Asp-Phe-Val-Arg-Pro- Glu-Pro-Ala-Ser-Arg-Pro-Ile-Ser-Asp-Ser-Arg- Arg-Ile-Tyr-Glu-Ser-Arg (I)

The single letter amino acid code for this polypeptide is indicated in Figure 1. Polypeptide (I) formally represents amino acids 496-524 of the 533 amino acid virE2 protein. Polypeptide (I) or the ssDNA binding fragments thereof can be synthesized using the Merrifield solid phase method. This is the method most commonly used for peptide synthesis and is extensively described by J . M. Stewart and J . D. Young, in Solid Phase Peptide Synthesis, Pierce Chem. Co., pub., Rockford, IL (2d ed., 1984), the disclosure of which is incorporated by reference herein.

The recombinant virE2 protein, as well as ssDNA-binding polypeptides which are formally fragments thereof, such as polypeptide I, can be complexed with a wide variety of ssDNA molecules. Useful ssDNA can in turn be obtained by chemical synthesis, or derived from a wide variety of natural sources, including plant pathogens such as DNA viruses (Cauliflower Mosaic virus (CaMV) or geminiviruses), RNA viruses, and viroids; DNA molecules derived from unstable plant genome components like extrachromosomal DNA elements in organelles (e.g., chloroplasts or mitochondria), or nuclearly encoded controlling elements; and DNA molecules from stable plant genome components (e.g., origins of replication and other DNA sequences which allow introduced DNA to integrate into the organellar or nuclear genomes and to replicate normally, to autonomously replicate, to segregate normally during cell division and sexual reproduction of the plant and to be inherited in succeeding generations of plants).

The ssDNA preferably comprises a foreign structural gene coding for an RNA, protein, polypeptide or a portion thereof. The gene is derived from other species, organisms or strains. The gene conveys an identifiable phenotype to the plant, such as disease resistance, temperature resistance, pest resistance, enzymatic activity, utility as a food ingredient or fiber and the like. Useful plant genes include those which have been characterized, transcribed and translated, after T-DNA-mediated transformation (e.g., genes for APH3'II, nopaline synthase, anthranilate synthase, nitrilase, bean phasolin, Zea maus zein, RuBP-Case small subunit, wheat chlorophyll a/b binding protein, and a soybean heat shock protein). For example, see PCT application WO 82/04181 and European patent application Serial No. 227,264, the disclosures of which are incorporated by reference herein.

The resultant complexes can be directly introduced into animal cells or into plant cells. Preferably, the complexes are introduced into animal cells or plant cell protoplasts, by electroporation, in combination with an effective amount of a carrier DNA, such as salmon sperm DNA. For example, M. Fromm, L. Taylor and V. Walbot, in Proc. Natl. Acad. Sci. USA, 82, 5824 (1985), reported the development of a general method for introducing DNA into plant cells by electroporation which is applicable to. both monocot and dicot protoplasts. A sample of 3×10⁶ protoplasts was suspended in 1 ml of buffered saline with about 10-40 μg/ml DNA. An electrical pulse of about 300-400 V was applied for about 30-250 msec as described by H. Potter et al., Proc. Natl. Acad. Sci. USA, 81, 7161 (1984). The protoplasts were then diluted into MS medium containing 2% sucrose, 0.3 M mannitol and 2,4-dichlorophenoxyacetic acid (0.1 μg/ml) and incubated at 26°C. [T. Murashige et al., Physiol. Plant., 15, 473 (1962)]. Other methods for introducing the complexes include microinjection by mechanical or laser beam methods, microprojectiles, and the use of liposomes. For example, see T. M. Klein et al., Nature, 527 70 (1987), the disclosure of which is incorporated by reference herein.

Once plant cells expressing the foreign structural gene are obtained, plant tissue and whole plants can be regenerated therefrom using methods and techniques well known in the art. The regenerated plant are then reproduced by conventional means and the introduced genes are transferred to other strains and cultivars by conventional plant breeding techniques. Thus, in its broadest sense, the present method is applicable to the introduction of a foreign structural gene into any plant or animal species into which DNA can be introduced.

Brief Description of the Figures Figure 1 depicts a nucleotide seguence of a 2.442 kb interval containing the virE operon. VirE1 extends from nucleotide 388 to 583, virE2 extends from nucleotide 590 and 2190, and a hypothetical ORF extends from nucleotide 2257 to the end of the seguence. The transcription initiation site at bp 199 is shown in a black box, while lines drawn over the DNA sequence indicate putative ribosome binding sites upstream of each initiation codon. The 28 amino acid sequence of the ssDNA binding polypeptide of the present invention is also underlined. Figure 2 is a physical map of the synthetic trpE'-virE2 operon of the present invention, as it is present in plasmid pAD1075. The virE2 coding sequence is shown by a dark box (not to scale). Translation of the trpE' protein terminates within the cloned fragment 4 bp upstream of the virE2 start codon. Arrows indicate the direction of transcription. Location of the amipicillin-resistance gene (amp) as well as the recognition sites of restriction enzymes BamHl and BglII are also indicated. The BglII site is located 0.56 kb downstream of the virE2 start codon.

Detailed Description of the Invention

The present invention provides a method for the overproduction of the virE2 protein or a fragment thereof, preferably in bacteria such as E. coli. The overproduction of virE2 proteins is useful to obtain large quantities for use in the present method. Under normal conditions, the vir genes are induced only by interaction with plant cells or by the chemical inducer acetosyringone. The level of induction is not so high to enable the synthesis of these gene products to be followed using standard protein gels. In the absence of a functional assay, it is almost impossible to identify, purify and study these gene products. To circumvent this problem, it is necessary to overproduce these proteins in a suitable host organism.

For overproduction of virE, a trpE'-virE fusion plasmid, pAD1075 (ATCC 40436, American Type Culture Collection, Rockville, MD), was constructed. The virE2 coding segment was cloned downstream of a truncated trpE gene (trpE') such that, upon induction, translation of trpE' will yield a 37 kD protein and stops immediately upstream of the virE initiator codon. Cells containing plasmid pAD1075, when induced with indoleacrylic acid (to activate the trp promoter), synthesized two new polypeptides of 37 and 68 kD.

While the 37 kD protein is the trpE' gene product, the 68 kD protein is the virE2 gene product.

To unequivocally establish that the 68 kD protein was indeed the virE2 gene product, plasmid pAD1082 was constructed by modifying a unique BglII site located within the virE2 coding region. This modification alters the reading frame of the virE2 gene product and is expected to yield a 22 kD protein. Cells containing pAD1082, upon induction, did not produce the 68 kD protein, but instead synthesized a new polypeptide of apparent mass 22 kD. This is the expected translation product if translation is initiated at the virE2 start codon.

In E. coli cells overproducing the virE2 gene product, the protein was found to localize into the insoluble fraction. The protein was solubilized by treatment with urea and its properties were studied. The virE2 protein is a single-stranded (ss) DNA binding protein. It does not have a sequence specificity for DNA binding and does not bind either double-stranded (ds) DNA or ssRNA. By gel retardation studies, it was observed that all DNA fragments were bound with equal efficiency by an E. coli extract containing the virE2 protein. The binding could be competitively blocked by ssDNA of E. coli origin and by ssTi plasmid DNA. To identify the ssDNA binding protein(s) in the E. coli extract, southwestern blotting procedures were used. In these experiments, the proteins were first separated in sodium dodecyl sulfate-polyacryl¬amide gels and were then blotted onto nitrocellulose filters by electroblotting procedures. The filters were then probed with radiolabelled ssDNA prepared by the random primer method. The results showed that the ssDNA binding activity was associated with the virE2 polypeptide. The virE2 polypeptide produced in E. coli was injected into rabbits to produce an t i- v i r E2 antiserum. Using anti-virE2 antiserum and DNA binαing assays, it was observed that agrobacteria, when induced with plant cells, synthesize a protein that comigrates with the virE2 polypeptide produced in E. coli, binds to single-stranded DNA, and crossreacts with anti-virE2 antibody.

The invention will be further described by reference to the following detailed examples.

Example I. Preparation and Expression of the virE2 Gene in E. coli

A, Plasmids and Transformants

The virE operon of Agrobacterium Ti plasmid pTiA6 is encoded within a 3.2 kb Xhol fragment. Plasmid pSW108 is a pUC7 derivative containing this Xhol fragment cloned into a Sall site of the vector. [S. Winans et al., Nucl. Acids Res., 15, 825 (1987)]. To isolate the virE2 coding segment, plasmid pSW108 was first digested with the enzyme TthIIIl. After filling in with Klenow enzyme in the presence of deoxynucleo-side triphosphates, the DNA sample was ligated with BamHl linkers (dCCGGATCCGG) in the presence of T4 DNA ligase. Following inactivation of DNA ligase by heating at 70°C for 10 min, the reaction mixture was digested with the enzyme BamHl. The DNA was extracted with phenol and chloroform, precipitated with ethyl alcohol, dried, and resuspended in water. The BamHl fragment containing the virE2 coding region was then cloned into the BamHl site of plasmid pATH2 to give plasmid pAD1075. Plasmid vector pATH2, an expression vector, contains the E. coli tryptophan (trp) operon promoter-operator region, and a large segment of the trpE coding sequence, followed by multiple cloning sites. LC. Dieckmann et al., 3. Biol. Chem., 260, 1513 (1985)].

Plasmid pAD1082 was constructed by linearizing plasmid pAD1075 at its unique BglII site followed by a filling in reaction with Klenow enzyme and recircularization with T4 DNA ligase. Plasmids pAD1075, pAD1082 and pATH2, were introduced into E. coli strain M0412 (recA- LE392) by transformation to yield strains AD1075, AD1082 and ATH2, respectively.

Plasmid pAD1012 was constructed by cloning a 476 bp Hpal-Nrul restriction fragment (bp 13,800-14,276 of R. Barker et al., Plant Mol. Biol., 2, 335 (1983)) containing the right border sequences of pTiA6 into the HincII site of vector pUC18 [C. Yaniseh-Peron et al., Gene, 33, 103 (1985)] B. Overproduction of virE2 Gene Product in E. coli

Cells were grown in M9 media containing 0.2% glucose, 0.5% acid-hydrolyzed casein, 20 μg/ml L-tryptophan, and 100 μg/ml ampicillin. To induce transcription from the trp promoter, an overnight culture was grown in medium containing tryptophan. The cells were collected by centrifugation, washed twice with M9 medium, and resuspended in an equal volume of the same medium. One ml of culture was then used to inoculate 100 ml of medium without tryptophan. After growth of cells to an A₆₀₀ of about 0.5-0.6, 100 μl of indole acrylic acid (20 mg/ml stock solution in alcohol) was added to the flask and cells were grown for two additional hours. Uninduced cells were grown in the presence of 20 μg/ml L-trp at all times. Following growth, cells were cooled in ice water, harvested by centrifugation at 5,000 rpm for 10 min in a Beckman JΑ-14 rotor, washed twice with cold 0.8% aqueous sodium chloride and frozen at -70°C until further use.

C. Partial Purification of the virE2 Gene Product

Induced AD1075 cells were thawed and resuspended (2 ml/gm of cells) in Buffer A (20 mM Tris-HCl, pH 8.0; 50 mM NaCl; 10% glycerol; 1 mM DTT; and 50 μM EDTA). Cells were lyseα in a French pressure cell press at 10,000 psi. The lysate was adjusted to 0.5 M NaCl by addition of 5 M NaCl and incubated in ice for 45 min. The cell lysate was centrifuged at 5,000 rpm for 5 min in a JA-14 rotor to pellet unlysed cells. The supernatant was transferred to a new tube and centrifuged at 12,000 rpm for 15 min. The 12K pellet contained all of the virE2 protein.

The pellet was washed twice with buffer A and then resuspended in 1 ml of buffer A. An equal volume of buffer A containing 8 M urea was added and mixed briefly. The mixture was incubated in ice for 15 min with occasional swirling. Following centrifugation at 12,000 rpm for 15 min, the supernatant was collected and dialysed overnight against buffer A with one change of buffer. The sample was centrifuged to clarify it, adjusted to 50% in glycerol and stored at -20°C (Fraction II preparation). Most of the virE2 protein was solubilized in this manner. In control experiments, uninduced AD1075 and induced ATH2 cells were processed in similar fashion.

D. DNA-Protein Binding Assay

The assay mixture (20 μl) contained 20 mM Tris-HCl, pH 8.0; 50 mM NaCl, 50 μM EDTA, pH 8.0; 1 mM DTT, 50 μg/ml BSA, 2-3 ng heat-denatured [³²p]DNA (see below), 100 ng sonicated double-stranded (ds) E. coli DNA, and the protein fraction, as indicated. Following incubation for 15 min at 26°C, 2 μl of a dye mix containing 50% glycerol and bromophenol blue as a marker dye was added to the reactions, mixed, and loaded immediately onto 5% polyacrylamide gel (in 0.5×TBE), following the procedure of T. Maiatis et al., in Molecular Cloning, A Laboratory Manual, Cold Spring Harbor Laboratories, New York, NY (1982). The gels were electrophoresed for about 2 hrs at 400 volts, dried, and autoradiographed.

The radiolabelled DNA used in these assays was prepared by digestion of plasmid pAD1012 DNA with EcoRI, HindIII and PvuII, followed by filling in with α-[³²p]dATP and Klenow enzyme.

E. Sodium Dodecyl Sulfate-Polyacrylamide Gel Electrophoresis (SDS-PAGE) and Protein Blotting

Proteins were- separated on 12.5% SDS-Polyacrylamide gels according to U. Laemlli, Nature, 277, 680 (1970). Where indicated, gels were stained with coomassie brilliant blue R. To identify DNA binding proteins after separation of proteins on SDS-PAGE, the gels were blotted onto nitrocellulose filters by the electroblotting procedure of B. Bowen et al., Nucl. Acids Research, 8, 1-20 (1980). The filters were prehybridized with 5% nonfat dry milk in 10 mM Tris-HCl, pH 8.0 and 1 mM DTT for 1 hr at room temperature. Hybridization was for one hr at room temperature in 0.25% nonfat dry milk, 20 mM Tris-HCl, pH 8.0; 50 mM NaCl, 1 mM DTT, 50 μM EDTA and 10⁵ cpm (Cerenkov) per ml of [³²p] labelled probe (sp. act. 1-4 × 10⁹ cpm/μg) Cw. Miskimins et al., PNAS USA, 82, 6741 (1985)]. DNA probes were prepared by the method of Feinberg and Vogelstein, Anal. Biochem., 132, 6 (1983). Following hybridization, filters were washed twice for 15 min in the hybridization buffer, dried and autoradiographed.

F. Other Methods

Antibody against virE2 polypeptide was raised in rabbits by subcutaneous injection with 100 μg protein followed by a second injection of another 100 μg protein after two weeks. Ten days later, the rabbit was bled and serum was collected. The protein used for injection was isolated by excision of the appropriate band from a SDS-polyacrylamide gel.

RNA probe was prepared by transcription from Sphl digested plasmid pGEMl DNA in the presence of T7 RNA polymerase (Promega Biotec, Madison, WI). Protein concentrations were determined by the procedure of Bradford, Anal. Biochem., 72, 248 (1976). Restriction and modifying enzymes were purchased from New England Biolabs. Radionucleotides were purchased from Amersham Corporation. E. coli strain M0412 was a gift of Marc Orbach, Stanford University, Stanford, CA. G. Results

1. Overproduction of virE2 Gene Product

To facilitate studies on the function of the virE gene products in Ti-mediated gene transfer, a recombinant plasmid pAD1075 was constructed that contains a trpE'-virE2 operon fusion (Figure 2). In pAD1075, the coding region of the virE2 gene was cloned downstream from the strong, regulatable trp promoter in a manner such that the trpE' open reading frame (ORF) terminates translation shortly before the virE2 ORF begins. When induced with indole acrylic acid, cells containing plasmid pAD1075 synthesized two new polypeptides of apparent molecular weight of 68,000 and 38,000 daltons. The 38,000 dalton protein is the truncated trpE' gene product which serves as an internal marker to monitor induction of the trp operon. The 68,000 dalton band is the virE2 gene product. This size agrees well with data obtained previously from maxi cell experiments conducted by S. Winans et al., cited above.

Neither product was present In significant amount in extracts of cells induced for 30 min but was easily visible after 90 min of induction.

To confirm that the 68 kilodalton (kd) band is indeed synthesized from the virE coding segment, plasmid pAD1082 was constructed by filling in a unique BglII restriction enzyme site located within the virE2 coding region (Figure 2). Alteration of the BglII site results in a frame-shift mutation. The truncated virE2 polypeptide (virE') in this mutant should be 196 amino acid residues in length. Cell extracts prepared from induced cells containing plasmid pAD1082 did not synthesize the 68 kD polypeptide product, but instead synthesized a new band of about 22,000 daltons . This is the expected molecular weight o f virE' if translation is initiated at the virE translation initiation site. 2. Localization of virE2 in E. coli

In induced E. coli cells, the virE2 protein was always found associated with the insoluble fraction. None was detectable in the soluble fraction. Attempts to solubilize the protein using nonionic detergents, e.g., deoxycholate, Triton X-100, nonidet P-40, etc., were unsuccessful (data not shown). Incubation of the pellet fraction with 4 M urea at 0°C resulted in solubilization of significant amounts of the virE2 polypeptide. Following removal of urea by dialysis, the virE2 polypeptide remained in the soluble fraction. Antibody was raised against the purified virE2 poypeptide, which in turn was isolated by excision of the protein band from SDS gels. When the anti¬body was used to probe total Agrobacterium proteins in a western blot experiment [H. Towbin et al., PNAS USA, 76, 4350 (1979)], a single protein band that comigrated with the protein produced in E. coli cells containing the cloned virE gene was recognized by the antiserum. The antigen was present only in the cytosolic fraction of a Ti plasmid-containing strain A348 that was induced with plant cells. Thus, it appears that the fractionation of the virE2 polypeptide into the E. coli membrane may be an artifact of overproduction.

3. DNA-Protein Binding Studies

To assess if the induced AD1075 cell extract contained any nucleic acid-binding activity, gel retardation experiments were performed. Fraction II preparations from uninduced AD1075 cells showed no binding to double-stranded (ds) DNA, single-stranded (ss) DNA or to ssRNA. Fraction II from induced cells, however, showed a strong ssDNA binding activity. In the presence of induced AD1075 extract, all radiolabelled DNA fragments bo un d to proteins and migrated very slowly at the top of the gel. When dsDNA was used as substrate, a very low level of binding was observed only in the absence of competitor DNA. This binding was lost in the presence of a 50-fold mass excess of E . coli DNA. In similar experiments, no binding of induced AD1075 extract to ssRNA substrate was observed. When both ssDNA and ssRNA were present as substrates, ssDNA was preferentially bound.

Of the radiolabelled fragments a, b, c and d, fragment b is a 510 base pair restriction fragment containing a 476 residue T-DNA sequence (coordinates 13,800 - 14,276 in R. Barker et al., Plant Mol. Biol., 2 , 335 (1983)). This fragment includes the right border region of the octopine-type Ti plasmid pTiA6, and fragments a, c and d are sequences derived from the plasmid vector pUC18. In the binding experiments, all four fragments bound to proteins, as was apparent from the significant loss in intensity of these bands and the appearance of new slower moving bands. To determine if the ssDNA binding protein present in the induced AD1075 extract had any sequence specificity, different amounts of proteins were used in binding experiments. An increase in binding with increasing amounts of proteins was apparent from the appearance of a ladder-like pattern and the retention of all radioactivity at the top of the gel. In all cases, both Ti plasmid-specific sequences and nonspecific vector sequences were retarded in a similar fashion. No preferential retardation of a fragment(s) was apparent in these studies. Binding to all fragments was found to be very strong as no significant reduction in binding was observed even at high salt concentration (0.3 M NaCl)(data not shown). In control experiments, Fraction II preparations, either from uninduced A01075 cells or from induced ATH2 cells, showed little or no DNA binding activity. Affinity for specific DNA sequences was also assessed by competition experiments using a homologous right border specific DNA fragment and non-specific E. coli DNA, as competitor DNAs in gel retardation experiments. Both T-DNA and E. coli DNA competed effectively as was apparent from the reappearance of four new bands with increasing concentration of competitor DNA. A 2-3-fold mass excess of E. coli DNA was necessary to compete as effectively as the homologous T-DNA fragment.

4. Protein Blotting

To identify the DNA binding protein(s) present in induced AD1075 cells, the protein blotting procedures disclosed above were used. Total proteins, from induced and uninduced cells, were first separated on 12.5% SDS-PAGE gels and then transferred onto nitrocellulose filters. When probed with heat-denatured ssDNA, no binding activity was found to be present in the uninduced cells. Induced cells showeα a major ssDNA binding activity that migrated very closely with the bovine serum albumin standard. This band had previously been identified as the virE2 polypeptide. No binding to this (or other protein) was observed when ssRNA was used as a probe (data not shown). The DNA probe used in these studies contained T-DNA sequences as well as vector DNA sequences.

To determine if the same protein in Fraction II preparations used in DNA-protein binding experiments was binding to ssDNA, and to further assess the specificity of the ssDNA binding activity, duplicate blots containing Fraction II preparations from uninduced AD1075 cells, induced AD1075 cells, and induced ATH2 cells were analyzed by DNA blotting experiments. Both the T-DNA-containing probe and the vector DNA probe bound to the virE2 polypeptide that was present only in the induced AD1075 cells. The two control extracts failed to show any DNA-binding activity in these assays. When visualized with coomassie brilliant blue protein stain, all three preparations showed a similar protein pattern except that induced cells containing plasmid pAD1075 contained an additional virE2 polypeptide band and that containing plasmid pATH2 contained β-lactamase gene products. The latter is due to over-production of the β-lactamase gene product as a result of transcription from the trp promoter under induction conditions; the ampicillin resistance gene of the plasmid is located immediately downstream from the trp sequences (Figure 2). Protein blotting procedures were also used to identify DNA binding activity, if any, present in agrobacteria cocultivated in the presence or absence of plant cells. At least two ssDNA-binding proteins were present in induced agrobacteria. Of these, the smaller polypeptide comigrated with the virE2 polypeptide overproduced in E . coli. The virE2 gene product, isolated either from E. coli or from Agrobacterium, did not bind to dsDNA. The larger polypeptide oound to both dsDNA and ssDNA. The affinity for ssDNA, however, was significantly higher as was apparent from the intensity of the bands. This protein was present in uninduced agrobacteria, and agrobacteria lacking a Ti plasmid. Its synthesis was not induced by cocultivation of bacteria with plant cells. To further insure that the Agrobacterium protein comigrating with the single-stranded DNA-binding activity present in induced AD1075 cells was the virE2 polypeptide, an immunological assay was employed. The filter used in experiments was stripped of radioactivity by high salt (0.5 M NaCl) treatment. It was then probed with anti-virE2 antibody raised against virE2 polypeptide produced in E. coli.

Several E. coli proteins present both in induced and uninduced cells reacted with the serum. These bands were also present when control serum was used (data not shown). With immune serum, an additional band that corresponded to the virE2 polypeptide was present only in the induced cell extract. In Agrobacteria, only a single protein reacted with the antibody. This protein was present only in cells induced with plant cells and was absent in Agrobacteria grown in culture.

Example II. Preparation of virE2 Protein Fragments A. Plasmid pAD1183

Plasmid pAD1075 synthesizes a virE2 gene product containing 533 amino acid residues. To localize the ssDNA binding regions, PD1075 was partially digested with PvuII. The Xhol linker, PCCTCGAGG, was introduced and the DNA fragments were digested with Xhol and recircularized with T4 ligase to yield plasmid PAD1182, which contains a Xhol site.

Plasmid pAD1182 was digested with Xhol, treated with S₁ nuclease, and recircularized to delete the Xhol site to yield plasmid pAD1183. This plasmid was used to transform E. coli as described hereinabove. The induced transformants synthesized a truncated virE2 protein of 497 amino acid residues which did not bind ssDNA. This strongly suggests that the polypeptide seguence containing the amino acid residues around or past residue 496 is necessary for the virE2 protein to bind ssDNA.

B. Plasmid pAD1089 Plasmid pAD1075 was digested with Sall and recircularized with T4 ligase to yield plasmid pAD1089. E. coli transformed with this plasmid synthesized a virE2 protein which contains a mutation past amino acid residue 525. Specifically, the carboxyl terminus of the virE2 protein, -Arg(526)-Ser-Gln-Ser-Val-Asn-Ser-Phe-stop has been replaced with -Ala-Ala-Gln-Ala-Tyr-Arg-stop. Since this mutant protein binds ssDNA, it is believed that polypeptide (I), which represents amino acid residues 496-524 of the 533 virE2 protein, contains part, if not all, of the domain responsible for binding ssDNA. This region is located much further toward the carboxyl terminus of the protein than the domains which C. Gietl et al., in PNAS USA, 84, 9006 (1987), speculated might be necessary to bind ssDNA.

Example III. DNA Transfer to Protoplasts

Nicotiana tabacum protoplasts are first prepared from NT culture cells by treatment with the enzymes cellulase and pectolyase. One ml of protoplast solution (10⁶/ml) is mixed with about 10 μg/ml DNA. In samples containing a virE2 protein, the protein is first incubated with ssDNA under DNA-binding conditions and then mixed with plant protoplasts and carrier DNA for electroporation. Plasmid pAD1053 DNA containing a chimeric gene for kanamycin resistance that expresses in plant cells is used in these studies. Double-stranded (ds) circular DNA, ds linear DNA, ss circular DNA with and without virE protein, and ss linear DNA with and without virE protein is employed in these electroporation experiments. Where indicated, the plasmid is linearized by cleavage at its unigue HindIII restriction enzyme site.

The mixture is placed in a cuvette and electroporated at 200 volts and 170 μF. The protoplasts are allowed to recover in ice for 10. min before plating. Following recovery of protoplasts after electroporation, they are plated in Murashige and Skoog (MS) media containing 0.4 M mannitol at a final concentration of 5×10⁴ cells/ml. The protoplasts are kept in the dark at 25°C for a week. The cells are then transferred to the same media containing 0.2 M mannitol for a week and then plated on solid media containing 100 μg/ml kanamycin. Resistant colonies are identified and plants regenerated therefrom as disclosed by J . Paszkowski et al., EMBO J., 3, 2717 (1964).

Discussion

Agrobacterium tumefaciens transfers its T-DNA during the formation of crown gall tumor disease. The precise mechanism of this DNA transfer process is not clearly understood. The process requires participation of twenty or more gene products encoded in six vir operons. Most of these products are synthesized only when agrobacteria recognize a signal molecule(s) from plant cells. To understand the mechanism of DNA transfer, it is necessary to know the role of these vir polypeptides in the overall process. These examples demonstrate that the virE2 gene product is a protein which binds to single-stranαed DNA.

The present invention is directed to a trpE ' - virE operon fusion which results in the overproduction of the virE2 polypeptide in E . coli (Figure 2). A protein fraction preparation containing the virE2 gene product strongly binds single-stranded DNA both in gel retardation and in protein blotting experiments. No significant affinity of this polypeptide for either double-stranded DNA or single-stranded RNA was observed in these studies. Using single-stranded DNA probes specific for right border sequences, left border sequences, internal T-DNA sequences and vector sequences, no apparent specificity for T-DNA sequences was observed. Agrobacterium induced with plant cells synthesized a protein that bound to single-stranded DNA and reacted with anti-virE2 antibody.

The non-discriminatory DNA binding property of the virE2 gene product provides indirect support for the hypothesis that T-DNA transfer from Agrobacterium to plant cells occurs via a single-stranded DNA intermediate. If the single-stranded T-DNA is indeed an intermediate in the DNA transfer process, it probably is necessary to protect the DNA molecules from nucleases prior to their integration into the plant nuclear genome. It is believed that, because of its strong affinity for single-stranded DNA, the virE polypeptide serves this function. This protective role of the virE2 gene product can account for earlier observations that virE mutants can transfer DNA into plant cells but fail to form tumors. In gene transfer studies, a low efficiency of DNA transfer will suffice because of a large amplification of the signal (usually a viral seguence that can replicate autonomously in plant cells). In virE mutants, in the absence of an efficient protection function, a small fraction of T-DNA molecules is expected to survive endogenous nuclease action. This will significantly reduce the efficiency of DNA transfer. Tumor formation, a process involving integration of T-DNA into the plant genome, may reguire a higher efficiency of DNA transfer. The latter, therefore, requires the presence of an active virE gene product. In addition, if the virE protein serves additional function in the plant cell, as suggested by Gardner and Knauf in Science, 231 725 (19δ6), the protein itself must be transported into the plant cells. An economic and efficient way of protein transfer across a phylogenetic barrier would be to piggy back during the DNA transfer mechanism which the cell has already established. The invention has been described with reference to various specific and preferred embodiments and techniques. However, it should be understood that many variations and modifications may be made while remaining within the spirit and scope of the invention.

Claims

WHAT IS CLAIMED IS:

1. A synthetic operon consisting essentially of a strong regulatable promoter in correct reading frame upstream from a DNA segment corresponding to at least a portion of the virE2 gene which encodes a polypeptide which nonspecifically binds to single-stranded DNA, wherein said operon does not comprise the virE2 promoter.

2. The synthetic operon of claim 1 wherein the strong promoter is a bacterial promoter.

3. The synthetic operon of claim 2 wherein the strong promoter is the trp promoter.

4. The synthetic operon of claim 3 which further comprises a portion of the trpE' gene between the trp promoter and the virE2 gene.

5. The synthetic operon of claim 1 wherein the portion of the virE2 gene encodes an about 68 kD polypeptide.

6. A vector comprising the synthetic operon of claim 1.

7. The vector of claim 6 which is a plasmid.

8. The vector of claim 7 which is AD1075.

9. A composition consisting essentially of a polypeptide of the formula:

Ala-Asp-Lys-Tyr-Ser-Arg-Asp-Phe-Val-Arg-Pro-Glu- Pro-Ala-Ser-Arg-Pro-Ile-Ser-Asp-Ser-Arg-Arg-Ile- Tyr-Glu-Ser-Arg

10. A method for overproducing a polypeptide encoded by the virE2 gene comprising:

(a) prov iding a recombinant vector incorporating a synthetic operon consisting essentially of a strong regulatable promoter in correct reading frame upstream from a DNA segment corresponding to at least a portion of the virE2 gene which encodes a polypeptide which nonspecifically binds to single-stranded DNA, wherein said operon does not comprise the virE2 promoter;

(b) transforming or transfecting cells of a suitable host organism with said vector;

(c) inducing the synthesis of said polypeptide by said transformed or transfected cells.

11. the method of claim 10 wherein said vector is a plasmid.

12. The method of claim 10 wherein said host cells are E. coli cells.

13. The method of claim 10 wherein said strong regulatable promoter is a bacterial promoter.

14. The method of claim 13 wherein said promoter is the trp promoter.

15. The method of claim 14 wherein the synthesis is induced by contacting said cells with indole arcylic acid.

16. A method for transferring single-stranded DNA into cells comprising:

(a) mixing said cells with an effective amount of carrier DNA and a complex comprising single- stranded DNA having bound thereto a polypeptide of the formula:

Ala-Asp-Lys-Tyr-Ser-Arg-Asp-Phe-Val-Arg-Pro- Glu-Pro-Ala-Ser-Arg-Pro-Ile-Ser-Asp-Ser-Arg- Arg-Ile-Tyr-Glu-Ser-Arg

(b) electroporating said mixture to introduce said complex into said protoplasts; and

(c) regenerating colonies of plant cells from said colonies, wherein said single-stranded DNA comprises a foreign structural gene which is expressed by said plant cells.

17. The method of claim 16 wherein the cells are plant cells.

18. The method of claim 17 further comprising converting said plant cells- into protoplasts prior to step (a).

19. The method of claim 16 wherein said DNA comprises a gene for an exogeneous protein.

20. The method of claim 16 wherein said protein is an enzyme.