pSa Causes Oncogenic Suppression of Agrobacterium by … · At905 At221(pSa::neo) Carr Chlr Genr Kanr Neor Rifr Sper Chlr This study At906 At221(pUCD3960) Carr Rifr Sper This study

JOURNAL OF BACTERIOLOGY,0021-9193/99/$04.0010

Jan. 1999, p. 186–196 Vol. 181, No. 1

Copyright © 1999, American Society for Microbiology. All Rights Reserved.

pSa Causes Oncogenic Suppression of Agrobacterium byInhibiting VirE2 Protein Export

LAN-YING LEE,1,2 STANTON B. GELVIN,1* AND CLARENCE I. KADO2

Department of Biological Sciences, Purdue University, West Lafayette, Indiana 47907-1392,1 andDavis Crown Gall Group, University of California, Davis, California 956162

Received 27 March 1998/Accepted 9 October 1998

When coresident with the Ti (tumor-inducing) plasmid, the 21-kDa product of the osa gene of the plasmidpSa can suppress crown gall tumorigenesis incited by Agrobacterium tumefaciens. Neither T-DNA processingnor vir (virulence) gene induction is affected by the presence of osa in the bacterium. We used Arabidopsisthaliana root segments and tobacco leaf discs to demonstrate that Osa inhibits A. tumefaciens from transform-ing these plants to the stable phenotypes of tumorigenesis, kanamycin resistance, and stable b-glucuronidase(GUS) expression. When A. tumefaciens contained osa, the lack of expression of transient GUS activity ininfected plant tissues, as well as the lack of systemic viral symptoms following agroinfection of Nicotianabenthamiana by tomato mottle virus, suggested that oncogenic suppression by Osa occurs before T-DNA entersthe plant nucleus. The extracellular complementation of an A. tumefaciens virE2 mutant (the T-DNA donorstrain) by an A. tumefaciens strain lacking T-DNA but containing a wild-type virE2 gene (the VirE2 donorstrain) was blocked when osa was present in the VirE2 donor strain, but not when osa was present in the T-DNAdonor strain. These data indicate that osa inhibits VirE2 protein, but not T-DNA export from A. tumefaciens.These data further suggest that VirE2 protein and T-DNA are separately exported from the bacterium. Thesuccessful infection of Datura stramonium plants and leaf discs of transgenic tobacco plants expressing VirE2protein by an A. tumefaciens virE2 mutant carrying osa confirmed that oncogenic suppression by osa does notoccur by blocking T-DNA transfer. Overexpression of virB9, virB10, and virB11 in A. tumefaciens did notovercome oncogenic suppression by osa. The finding that the expression of the osa gene by itself, rather thanthe formation of a conjugal intermediate with pSa, blocks transformation suggests that the mechanism ofoncogenic suppression by osa may differ from that of the IncQ plasmid RSF1010.

The Ti plasmid resident in virulent strains of Agrobacteriumtumefaciens confers on this soil-borne bacterium the ability tocause crown gall tumors on a wide range of dicotyledonousplants, as well as some monocots and gymnosperms. Generally,when this plasmid coresides with other plasmids, such as RP4and RK2 and their derivatives, the virulence of A. tumefaciensremains unaffected (10, 20, 28). However, plasmids of theincompatibility groups Q (12) and W (41) severely affect tu-morigenicity when coresident with the Ti plasmid in A. tume-faciens. Ward et al. (42) showed that the presence of the IncQRSF1010 derivative pJW323 inhibited tumor formation onKalanchoe daigremontiana leaves. This inhibitory effect couldbe overcome by the coordinate overexpression of the virulencegenes virB9, virB10, and virB11. pJW323 inhibited the ability ofA. tumefaciens to serve as a proficient VirE2 donor cell inextracellular complementation assays (34), and this inhibitioncould not be alleviated by overexpressing VirE2 in strainscontaining pJW323 (2). Although the mechanism of the effectof pJW323 on oncogenic suppression remains unclear, Binns etal. (2) proposed that pJW323 preferentially competes for theputative T-DNA–VirE2 transport apparatus composed ofVirB proteins.

Loper and Kado (28) and Farrand et al. (10) showed that thetumorigenicity of A. tumefaciens is suppressed when it harborsthe IncW plasmid pSa. This plasmid, originally isolated fromShigella flexneri, is a 38-kbp broad-host-range plasmid that con-fers resistance to the antibiotics spectinomycin, kanamycin,

gentamicin, chloramphenicol, and sulfonamide. The presenceof pSa in A. tumefaciens does not cause instability of the Tiplasmid, and oncogenicity is regained when the strain is curedof pSa (10). Moreover, pSa does not prevent conjugative trans-fer of the Ti plasmid, and when the Ti plasmid is transferredfrom a pSa-bearing strain to a recipient strain, oncogenicity isconferred upon the recipient (10).

Previous studies localized the oncogenic suppression gene toa 3.1-kbp region of pSa (7). This region contains the osa (on-cogenic suppressive activity) gene that is the fourth gene of afour-gene operon. The presence of osa alone in A. tumefaciensis sufficient for suppressing oncogenicity (3, 8).

The amino acid sequence of Osa shares 39% identity and60% similarity with the FiwA protein (encoded by the fertilityinhibition gene, fiwA) of the plasmid RP1 (3). This gene isinvolved in the inhibition of conjugative transfer of IncW plas-mids, such as R388 and pSa (11). The similarity of the aminoacid sequences between these two proteins suggested that Osamay inhibit T-DNA transfer from A. tumefaciens to plants, aprocess that is believed to occur by a conjugative mechanism(22, 25, 37).

In this study, we have investigated the mechanism of onco-genic suppression mediated by osa. Using a number of differ-ent assays, we demonstrate that the presence of osa in A.tumefaciens inhibits tumorigenesis not by blocking T-DNAtransfer, but rather by inhibiting VirE2 protein export from thebacterium to the plant cell.

MATERIALS AND METHODS

Bacterial strains and growth conditions. The A. tumefaciens strains and plas-mids used in this study are listed in Table 1. A. tumefaciens strains were culturedat 30°C in YEP complete or AB minimal medium containing 0.5% sucrose (26)

* Corresponding author. Mailing address: Department of BiologicalSciences, Purdue University, West Lafayette, IN 47907-1392. Phone:(765) 494-4939. Fax: (765) 496-1496. E-mail: [email protected].

186

on May 24, 2021 by guest

http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

or in medium 523 (23) supplemented with the appropriate antibiotics (rifampin,10 mg/ml; kanamycin, 25 mg/ml; spectinomycin, 50 mg/ml; tetracycline, 5 mg/ml).Plasmids were introduced into A. tumefaciens by electroporation as described inthe Bio-Rad Gene Pulser instruction manual.

Agroinfection. A. tumefaciens strains containing the binary vector pCGN1547with a dimer of the A or the B component of the tomato mottle virus (ToMoV)genome were grown in 523 medium containing the appropriate antibiotics at30°C overnight. The cells were washed and resuspended in fresh medium to a

TABLE 1. A. tumefaciens strains and plasmids used in this study

Strain or plasmida Relevant characteristics Antibiotic resistanceb Source or reference

StrainsA136 Nononcogenic strain; lacks pTi Rifr 43A208 Oncogenic nopaline-type strain; pTiT37 Rifr 36A348 Oncogenic octopine-type strain; pTiA6 Rifr 14At221 Tn3-HoHo1 insertion in virE2; A348mx358 Carr Rifr 38At789 A348(pBISN1) Kanr Rifr 31At793 A136(pBISN1) Kanr Rifr 31At900 LBA4404(pSa) Chlr Genr Kanr Rifr Sper Sulr This studyAt901 LBA4404(pSa::neo) Chlr Genr Kanr Neor Rifr Sper Sulr This studyAt902 LBA4404(pUCD3960) Carr Rifr Sper This studyAt903 LBA4404(pUCD5533) Carr Rifr Sper This studyAt904 At221(pSa) Carr Chlr Genr Kanr Rifr Sper Sulr This studyAt905 At221(pSa::neo) Carr Chlr Genr Kanr Neor Rifr Sper Chlr This studyAt906 At221(pUCD3960) Carr Rifr Sper This studyAt907 At221(pUCD5533) Carr Rifr Sper This studyAt915 LBA4404(pToMoV-A) Kanr Rifr R. GilbertsonAt916 LBA4404(pToMoV-B) Kanr Rifr R. GilbertsonAt925 A208(pBISN1, pUCD5542, pSa) Chlr Genr Kanr Rifr Sper Sulr Tetr This studyAt926 A208(pBISN1, pUCD5542, pUCD3960) Carr Kanr Rifr Sper Tetr This studyAt927 A208(pBISN1, pUCD5542, pUCD5533) Carr Kanr Rifr Sper Tetr This studyAt928 A208(pBISN1) Kanr Rifr This studyAt929 A208(pBISN1, pSa) Chlr Genr Kanr Rifr Sper Sulr This studyAt931 A208(pBISN1, pUCD3960) Carr Kanr Rifr Sper This studyAt932 A208(pBISN1, pUCD5533) Carr Kanr Rifr Sper This studyAt935 A208(pBISN1, pUCD5542) Kanr Rifr Tetr This studyAt985 At221(pJB31) Carr Sper Rifr This studyAt989 At221(pBISN1) Carr Kanr Rifr This studyAt990 At221(pBISN1, pUCD3960) Carr Kanr Rifr Sper This studyAt991 At221(pBISN1, pUCD5533) Carr Kanr Rifr Sper This studyAt1042 LBA4404(pToMoV-A, pSa) Chlr Genr Kanr Rifr Sper Sulr This studyAt1043 LBA4404(pToMoV-B, pSa) Chlr Genr Kanr Rifr Sper Sulr This studyAt1044 LBA4404(pToMoV-A, pSa::neo) Chlr Genr Kanr Neor Rifr Sper Sulr This studyAt1045 LBA4404(pToMoV-B, pSa::neo) Chlr Genr Kanr Neor Rifr Sper Sulr This studyAt1046 LBA4404(pToMoV-A, pUCD3960) Carr Kanr Rifr Sper This studyAt1047 LBA4404(pToMoV-B, pUCD3960) Carr Kanr Rifr Sper This studyAt1048 LBA4404(pToMoV-A, pUCD5533) Carr Kanr Rifr Sper This studyAt1049 LBA4404(pToMoV-B, pUCD5533) Carr Kanr Rifr Sper This studyAt1050 LBA4301(pToMoV-B, pJK703) Kanr Neor Rifr This studyAt1085 A208(pJB31, pUCD5542) Sper Rifr Tetr This studyLBA4301 rec mutant of A. tumefaciens Ach5; no pTi Rifr 24LBA4404 Disarmed octopine-type strain; lacks T-DNA Rifr 33

PlasmidspBISN1 T-DNA binary vector; nos-nptII and

superpromoter-gusA intron geneKanr 31

pJB31 RSF1010 derivative Sper 1pJK703 pTiC58 virD::Tn5 Neor 35pSa Oncogenic suppressive plasmid Chlr Genr Kanr Sper Sulr 28pSa::neo Neomycin phosphotransferase gene disruption of

the osa gene of pSaChlr Genr Kanr Neor Sper Sulr This study

pToMoV-A Dimer of the A component of the ToMoVgenome cloned into pCGN1547

Kanr R. Gilbertson

pToMoV-B Dimer of the B component of the ToMoVgenome cloned into pCGN1547

Kanr R. Gilbertson

pUCD105 Stable cloning vector for A. tumefaciens; pTARori

Carr Kanr Sper 41

pUCD2001 Low-copy cloning vector for A. tumefaciens Carr Kanr Tetr 13pUCD3960 osa gene under the control of a npt promoter in

pUCD105Carr Sper 3

pUCD5533 Asp718-blunted derivative of pUCD105 Carr Sper This studypUCD5542 virB9, -10, and -11 genes under the control of the

virB promoter in a modified pUCD2001Tetr This study

a Laboratory stocks: A, E. W. Nester; At, S. B. Gelvin; LBA, R. A. Schilperoort; UCD, C. I. Kado.b Car, carbenicillin; Chl, chloramphenicol; Gen, gentamicin; Kan, kanamycin; Neo, neomycin; Rif, rifampin; Spe, spectinomycin; Sul, sulfonamide; Tet, tetracycline.

VOL. 181, 1999 ONCOGENIC SUPPRESSION OF AGROBACTERIUM BY pSa 187


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

concentration of 109 cells/ml. Equal volumes of the two strains carrying ToMoVA and B components were mixed, and 40 ml of cells was applied to the cut surfaceof Nicotiana benthamiana stems. The plants were scored after 2 weeks for theappearance of yellow mottling symptoms on the newly expanding leaves. Squashblot analysis was conducted according to the method of Gilbertson et al. (16, 17).PCR analysis for the detection of viral DNA was conducted according to themethod of Gilbertson et al. (18).

Transformation of Arabidopsis roots. Arabidopsis thaliana ecotype Aa-0 wasgrown in MS medium as described by Nam et al. (30). Roots of 2-week-old plantswere cut into 5-mm segments and placed on MS basal agar medium (per liter:4.32 g of MS salts, 0.5 g of MES [4-morpholineethanesulfonic acid] [pH 5.7],0.1 g of myo-inositol, 1 ml of vitamin stock solution [0.5 mg of nicotinic acid perml, 0.5 mg of pyridoxine per ml, 0.5 mg of thiamine-HCl per ml], 10 g of sucrose,and 0.75% Bactoagar [Difco]). A drop of the appropriate A. tumefaciens strainwas applied to the root segments for 10 min, excess bacteria were removed fromthe plate, and the roots were cocultivated at 25°C for 2 days. For assaying tumorformation, the root segments were rinsed with sterile distilled water containing100 mg of Timentin (clavulanic acid) per ml, blotted dry on filter paper, andtransferred to hormone-free MS basal agar medium containing Timentin (100mg/ml). The development of tumors on the cut surfaces of the roots was scoredafter 4 weeks of incubation at 25°C. To assay transformation to kanamycinresistance, the cocultivated root segments were transferred to callus-inducingmedium (CIM; containing per liter: 4.3 g of MS minimal salts, 0.5 g of MES [pH5.7], 1 ml of vitamin stock, 0.1 g of myo-inositol, 20 g of glucose, 0.5 mg of2,4-dichlorophenoxyacetic acid, 0.3 mg of kinetin, 5 mg of indoleacetic acid, and0.75% Bactoagar) containing Timentin (100 mg/ml) and kanamycin (150 mg/ml).Nonselected calli were grown on CIM containing Timentin but lacking kanamy-cin. To assay b-glucuronidase (GUS) activity qualitatively, root segments werecollected from the CIM plate after various periods of incubation, blotted ontofilter paper, and stained with X-Gluc solution (50 mM NaH2PO4, 10 mM EDTA,300 mM mannitol, 2 mM X-Gluc [5-bromo-4-chloro-3-indolyl glucuronide] [pH7.0]) overnight at 37°C. For quantitative GUS assays, the roots were ground in amicrocentrifuge tube containing GUS extraction buffer (50 mM Na2HPO4, 5 mMdithiothreitol, 1 mM EDTA, 0.1% Sarkosyl, 0.1% Triton X-100 [pH 7.0]), andthe specific GUS activity was determined as described by Jefferson et al. (21).

Tobacco leaf disc infection. Leaves derived from axenically grown N. tabacumcv. Wisconsin 38 or transgenic tobacco plants constitutively expressing the virE2gene (kindly provided by Vitaly Citovsky) were used for the source of leaf discs.Overnight liquid cultures of A. tumefaciens were incubated with leaf discs for 10min. The discs were blotted dry on sterile filter paper and placed on MS basalagar medium for 2 days at 25°C. Depending on the experiment, the leaf discswere transferred to various agar media as follows: (i) on MS basal mediumwithout phytohormones for the generation of tumors; (ii) on CIM containing 100mg of kanamycin per ml for the selection of kanamycin-resistant calli; and (iii) onCIM lacking kanamycin for growth of nonselected calli. When infected leaf discswere assayed for GUS activity, discs were sampled and stained with X-Glucbeginning 2 days after the start of cocultivation.

Tobacco suspension cell infection. Nicotiana tabacum BY-2 cells were prop-agated in Murashige and Skoog medium (Gibco-BRL, Gaithersburg, Md.) con-taining 3% sucrose, 1 mg of thiamine-HCl per liter, 0.2 mg of 2,4-dichlorophe-noxyacetic acid per liter, and 370 mg of KH2PO4 per liter. The suspension cellswere incubated at 25°C under continuous light with shaking (140 rpm). A.tumefaciens strains were grown to a density of 2 3 109 cells/ml in AB-glucosemedium. Bacterial cells were centrifuged and suspended at a concentration of109 cells/ml in induction medium (AB salts, 0.5% glucose, 2 mM sodium phos-phate, 50 mM MES [pH 5.6], 100 mM acetosyringone), and incubated with gentleshaking for 14 to 18 h at 25°C. The cells were washed and concentrated 50-foldin plant culture medium, and 200 ml of bacterial cells was added to 50 ml of BY-2cells and cocultivated at 25°C. Two days later, most bacterial cells were washedfrom the plant cells by repeated centrifugation at 300 rpm (Beckman Sorvall[Newtown, Conn.] GLC-2 clinical centrifuge) for 2 min and resuspension in plantculture medium. The plant cells were finally resuspended in plant culture me-dium containing Timentin (100 mg/ml) and continuously cultured. Starting fromday 2 of cocultivation, plant cells were collected and stained in X-Gluc stainingsolution overnight at 37°C.

Extracellular complementation of tumorigenesis on Kalanchoe plants. Leavesof K. daigremontiana plants were washed with sterile water and wounded with asterile toothpick. Various dilutions of different combinations of A. tumefaciensstrains were inoculated into the wound sites. These sites were examined 2 weekslater for tumor formation.

Inoculation of Datura plants. Stems of young Datura stramonium plants (8 to10 cm in height) were inoculated by poking the stems with a sterile toothpickharboring a paste of bacterial inoculum. Tumor formation was scored by exam-ination of the inoculated sites 2 to 3 weeks later.

RESULTS

Expression of osa in A. tumefaciens inhibits the stable trans-formation of plants. To extend the previous observations thatthe oncogenic suppressive activity of Osa occurs irrespective ofthe recipient plant host (10, 28) and to determine whether osa

could inhibit the transformation of plants to other stable phe-notypes, we infected A. thaliana ecotype Aa-0 root segmentsand tobacco leaf discs with A. tumefaciens strains containing orlacking pSa or the plasmid pUCD3960 containing just the osagene under the control of a heterologous kanamycin promoter.In addition, some of these strains contained the plasmidpBISN1, which is a binary vector (31) that contains within theT-DNA a nos-nptII gene and a gusA intron gene under thecontrol of a strong “superpromoter” (32). The presence ofpBISN1 in these strains allowed us to monitor transformationto the stable phenotypes of kanamycin resistance and GUSexpression, respectively, in addition to the formation of tera-tomas. We used A. tumefaciens A208 to inoculate sterile Ara-bidopsis ecotype Aa-0 root segments, because we had previ-ously determined that this combination of bacterial strain andhost ecotype resulted in reproducible transformation to strongand predictable stable phenotypes (30). Two days after inocu-lation, we transferred the root segments to CIM, either with orwithout kanamycin, or to MS medium lacking hormones. Fig-ure 1 shows that in the absence of the osa gene, A. tumefaciensinduced teratomas approximately 1 month after inoculation(Fig. 1A and D). However, A. tumefaciens strains containingthe osa gene (either in the plasmid pSa or pUCD3960) failedto induce tumors (Fig. 1B and C). We observed the productionof kanamycin-resistant calli only when the inciting A. tumefa-ciens strain lacked pSa or osa (Fig. 2A and D). We did notobtain kanamycin-resistant calli when the osa gene was resi-dent in the A. tumefaciens strains (Fig. 2B and C). However, onCIM lacking kanamycin, calli developed approximately 1month after cocultivation with all strains (data not shown).When stained with the chromogenic GUS substrate X-Gluc,only calli derived from root segments infected with A. tumefa-ciens strains lacking osa showed blue spots (data not shown).These data confirm that the presence of the osa gene in A.tumefaciens not only blocks tumorigenesis, but also blocks thetransformation of plant cells to other stable phenotypes, suchas kanamycin resistance and stable GUS expression. When weused tobacco leaf discs to conduct similar infections, we repro-ducibly obtained the same results (data not shown).

osa inhibits the transient transformation of plants by A.tumefaciens. The data described above indicate that the pres-ence of osa in A. tumefaciens inhibits the stable transformationof plants. It is possible, however, that osa inhibits the integra-tion of T-DNA into the plant genome, but not T-DNA transferto and transient expression within plant nuclei. We thereforeconducted two sets of experiments to test whether osa affectsthe transient expression of genes introduced into plants by A.tumefaciens.

Agroinfection is inhibited by osa. The introduction of mul-timeric forms of both of the components of two-componentgeminiviruses into plants can result in systemic viral infection(19). When these components are cloned within the T-DNAand introduced into plants by A. tumefaciens, they can bereleased by homologous recombination. This release presum-ably requires T-DNA nuclear transport but not T-DNA inte-gration into the plant genome. To determine whether thisagroinfection is inhibited by osa, we coinoculated N. benthami-ana plants with A. tumefaciens strains containing multimers ofthe A and B components of ToMoV. The strains additionallycontained the osa gene (either on pSa or cloned separately intopUCD3960) or control constructions. Table 2 shows the resultsof these experiments. Successful agroinfection resulted in thedevelopment of yellow mottling symptoms on the upper leaves14 days after inoculation. We confirmed the systemic spread ofthe virus by squash blot hybridization and PCR analysis ofDNA isolated from the upper leaves, using hybridization

188 LEE ET AL. J. BACTERIOL.


http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

probes and primers, respectively, that would detect viral DNA.We never detected viral symptoms in plants inoculated with A.tumefaciens strains containing osa. In addition, we could notdetect viral DNA in the upper leaves of these plants. Inocula-tion of plants with an A. tumefaciens strain with a mutation invirD2 (and therefore unable to transfer T-DNA to plants)similarly did not result in viral symptoms or systemic viralspread. However, A. tumefaciens strains containing a disruptedosa gene (pSa::neo) or a vector lacking osa (pUCD5533) ef-fected agroinfection of these plants. We repeated these exper-iments three times using five plants of each inoculation groupeach time. The results of these experiments indicate that (i) osacan completely inhibit agroinfection of N. benthamiana plantsby ToMoV, and (ii) osa blocks Agrobacterium-mediated trans-formation at some step prior to T-DNA integration.

osa prevents expression of transient GUS activity in infectedplant cells. To confirm that osa inhibits the transient expres-sion of T-DNA-encoded genes, we infected Arabidopsis rootsegments, tobacco leaf discs, and tobacco BY-2 suspensioncells with A. tumefaciens A208(pBISN1) either containing orlacking osa. The gusA intron gene in pBISN1 allowed us toassay expression of GUS activity within infected plant cellswithout expression of GUS activity in the bacteria (27). Figure3 shows Arabidopsis root segments (Fig. 3A) and tobacco leafdiscs (Fig. 3B) stained with X-Gluc 4 days after inoculationwith various A. tumefaciens strains. Infection of these plants

with strains lacking osa resulted in the production of GUSactivity, as indicated by sectors of blue-staining tissue. How-ever, when the A. tumefaciens strains contained either pSa orthe osa gene, we never detected GUS activity. We obtainedsimilar results after infecting tobacco BY-2 suspension cellswith the A. tumefaciens strains (i.e., when the bacteria lackedosa, 2 to 5% of the tobacco cell clusters stained blue withX-Gluc). Infection of the cells with A. tumefaciens harboringpSa or the osa gene never resulted in detectable GUS activity(data not shown).

Taken together, the agroinfection and GUS expression dataindicate that osa blocks Agrobacterium-mediated transforma-tion at some step prior to nuclear entry of the T-DNA.

osa blocks VirE2 protein but not T-DNA export from A.tumefaciens. To determine whether osa blocks T-DNA transferfrom A. tumefaciens to plant cells, we initially performed ex-tracellular complementation assays. Otten et al. (34) had pre-viously shown that when two avirulent A. tumefaciens strains,one lacking T-DNA (the VirE2 donor) and the other with amutation in virE2 (the T-DNA donor) are coinoculated intoplant wounds, tumors can develop. Control experiments hadshown that this complementation is most likely extracellular(i.e., not within A. tumefaciens), because it could occur in theabsence of Ti plasmid conjugation between the mutant strains,and that it required bacterial binding to plant cells. These datahave been interpreted to indicate the independent transfer of

FIG. 1. Teratoma formation on Arabidopsis roots. Sterile Arabidopsis root segments were cocultivated with A. tumefaciens for 2 days and then transferred to MSbasal medium lacking phytohormones, but containing antibiotics to prevent the growth of bacterial cells. Teratomas first appeared approximately 2 weeks after infection,and the roots were photographed 1 month after infection. Root segments were infected by A. tumefaciens At928 [A208(pBISN1)] (A), A. tumefaciens At929[A208(pBISN1, pSa)] (B), A. tumefaciens At931 [A208(pBISN1 plus osa)] (C), or A. tumefaciens At932 [A208(pBISN1 plus vector)].



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

VirE2 protein and the T-DNA–VirD2 complex from A. tume-faciens (2, 40).

Figure 4 confirms that when coinoculated into Kalanchoeleaves, A. tumefaciens LBA4404 (the VirE2 donor) and At221(the T-DNA donor) can incite crown gall tumors. Neitherstrain alone was tumorigenic. When A. tumefaciens At221 con-taining either pSa or the osa gene was used as the T-DNAdonor strain, tumors still resulted. Thus, osa does not block

T-DNA transport from the bacterium. When A. tumefaciensLBA4404 containing pSa or the osa gene was used as theVirE2 donor, however, no tumors resulted from coinoculationwith A. tumefaciens At221. Oncogenic suppression dependedupon the presence of an active osa gene within the bacterium.Disruption of the osa gene within pSa by a neomycin phos-photransferase gene (pSa::neo) permitted A. tumefaciensLBA4404 to serve as a VirE2 donor.

FIG. 2. Transformation of Arabidopsis roots to kanamycin resistance. Sterile Arabidopsis root segments were cocultivated with A. tumefaciens for 2 days and thentransferred to CIM containing antibiotics to prevent the growth of bacterial cells and to select for transformed plant cells. The roots were photographed 1 month afterinfection. Root segments were infected by A. tumefaciens At928 [A208(pBISN1)] (A), A. tumefaciens At929 [A208(pBISN1, pSa)] (B), A. tumefaciens At931[A208(pBISN1 plus osa)] (C), or A. tumefaciens At932 [A208(pBISN1 plus vector)] (D).

TABLE 2. Agroinfection of N. benthamiana plants by A. tumefaciens strains containing or lacking osa

A. tumefaciensstrain Strain description

Determination of viral infection (no. of positiveplants/no. inoculated)

Systemic symptoms Squash blot PCR product

At915 1 At916 LBA4404(pToMoV-A) 1 LBA4404(pToMoV-B) 15/15 15/15 15/15

At1042 1 At1043 LBA4404(pToMoV-A, pSa) 1 LBA4404(pToMoV-B, pSa) 0/15 0/15 0/15

At1044 1 At1045 LBA4404(pToMoV-A, pSa::neo) 1 LBA4404(pToMoV-B, pSa::neo) 15/15 15/15 15/15

At1046 1 At1047 LBA4404(pToMoV-A, pUCD3960) 1 LBA4404(pToMoV-B, pUCD3960) 0/15 0/15 0/15

At1048 1 At1049 LBA4404(pToMoV-A, pUCD5533) 1 LBA4404(pToMoV-B, pUCD5533) 15/15 15/15 15/15

At915 1 At1050 LBA4404(pToMoV-A) 1 LBA4301(pToMoV-B, pJK703) 0/15 0/15 0/15



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

These data indicate that osa blocks the transfer of VirE2protein, but not the T-DNA, from A. tumefaciens to plant cells.

osa does not inhibit tumorigenesis on Datura by an A. tume-faciens virE2 mutant strain. Although virE2 mutant A. tume-faciens strains are avirulent on many host plants (38), they mayincite small tumors on highly sensitive hosts such as D. stra-monium. When inoculated onto D. stramonium stems, thevirE2 mutant A. tumefaciens At221 incited small tumors (ap-proximately 2 to 3 mm in diameter) compared to the largetumors (100 to 150 mm in diameter) incited by wild-type A.tumefaciens At789 (Fig. 5). A. tumefaciens virE2 mutant strainscontaining the osa gene (At906) or vector sequences alone(At907) both retained the ability to incite small tumors on D.stramonium stems (Fig. 5). These data support the hypothesisthat osa does not prevent T-DNA export from A. tumefaciens.Binns et al. (2) previously showed that the oncogenic suppres-sive IncQ plasmid pJW323 inhibits VirE2 protein export fromA. tumefaciens, but inhibits T-DNA export only to a limitedextent. Figure 5 confirms one of these observations. A. tume-faciens At221 containing another RSF1010 derivative, pJB31(At985), could still transfer T-strands and incite small tumorson D. stramonium stems.

osa does not inhibit tumorigenesis of a virE2 mutant A.tumefaciens strain when inoculated into VirE2-producingtransgenic tobacco plants. The virE2 mutant strain A. tumefa-ciens At221 cannot incite tumors on wild-type tobacco.Citovsky et al. (5) demonstrated that the same A. tumefaciensstrain could induce tumors on transgenic tobacco that pro-

duced VirE2 protein. We reasoned that if osa blocks VirE2protein but not T-DNA transport from A. tumefaciens, a virE2mutant strain containing osa should retain virulence on VirE2-producing tobacco plants. Figure 6A4 shows that as describedby Citovsky et al. (5), A. tumefaciens At221 could induce tu-mors on VirE2-producing transgenic tobacco leaf discs. A.tumefaciens At221 could not induce tumors on wild-type to-bacco leaf discs (data not shown). The presence of osa (Fig.6A5) or vector sequences (Fig. 6A6) in At221 did not inhibittumorigenesis by the strain on VirE2-producing tobacco leafdiscs. In addition, the RSF1010 derivative pJB31 did not in-hibit tumorigenesis by A. tumefaciens At221 on VirE2-produc-ing tobacco plants (Fig. 6A3). Finally, as shown in Fig. 6B2 toB4, neither the presence of the osa gene nor vector sequencescould inhibit tumorigenesis by the wild-type strain A. tumefa-ciens A208 when inoculated onto VirE2-producing transgenictobacco leaf discs. These data confirm that osa does not inhibitT-strand export from A. tumefaciens and further demonstratethat the IncQ plasmid pJB31 also does not inhibit T-DNAexport from the bacterium.

Overexpression of the proteins VirB9, -10, and -11 cannotovercome oncogenic suppression by osa. Binns et al. (2) pre-viously demonstrated that the oncogenic suppression effect ofRSF1010 derivatives could be overcome by overexpression ofVirB9, -10, and -11 proteins in A. tumefaciens. These authorshypothesized, therefore, that IncQ plasmids effect oncogenicsuppression by competing with VirE2 protein for the virB-encoded protein export apparatus. To determine whether osa

FIG. 3. Transient GUS expression in Arabidopsis roots and tobacco leaf discs. Sterile Arabidopsis root segments (A) or tobacco leaf discs (B) were cocultivated withA. tumefaciens for 2 days and then transferred to medium containing Timentin to prevent the growth of bacterial cells. After 2 additional days, the plant material waswashed with 0.9% NaCl and stained with X-Gluc. Photographs were taken with a dissecting microscope. Plant tissue was infected with A. tumefaciens At928[A208(pBISN1)] (a), A. tumefaciens At929 [A208(pBISN1, pSa)] (b), A. tumefaciens At931 [A208(pBISN1 plus osa)] (c), or A. tumefaciens At932 [A208(pBISN1 plusvector)] (d).



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

likewise competes with VirE2 for the VirB protein exportchannel, we cloned virB9, -10, and -11 under the control of thevirB promoter to create pUCD5542. pUCD5542 could com-plement a polar virB10 mutant A. tumefaciens strain and re-store virulence (data not shown). We introduced pUCD5542into A. tumefaciens A208(pBISN1) (At935) and A208(pBISN1)containing either pSa (At925), the osa gene (At926), or vectorsequences alone (At927) and inoculated these strains ontowounded D. stramonium stems. Figure 7 shows that A. tume-faciens A208(pBISN1) containing pUCD5542 was highly tu-morigenic on D. stramonium. However, the presence of addi-tional copies of virB9, virB10, and virB11 in A. tumefaciensA208(pBISN1) could not overcome oncogenic suppression bypSa or by the osa gene. In addition, pUCD5542 could neitherovercome oncogenic suppression by pSa or the osa gene ontobacco leaf discs or Kalanchoe leaves, nor could it permitGUS expression (directed by pBISN1) in inoculated tobaccoleaf discs (data not shown). However, A. tumefaciens At1085,containing pUCD5542 and pJB31 in the oncogenic strainA208, was also not tumorigenic on Kalanchoe leaves (data notshown). Thus, it is likely that the expression of virB9, virB10,and virB11 with our construction was not strong enough toovercome oncogenic suppression.

DISCUSSION

In this study, we have investigated the mechanism of onco-genic suppression by the osa gene of the IncW plasmid pSa.Previous studies had shown that when coresident with the Tiplasmid, pSa inhibits tumor formation by A. tumefaciens (28).Oncogenic suppression did not result from a permanent alter-ation of the host A. tumefaciens strain. When cured of pSa, thebacteria regained their tumorigenic potential (10). Geneticdissection of pSa revealed that a 3.1-kbp region of the plasmidwas responsible for oncogenic suppression (7). Further analysisindicated that a single gene from this region, termed osa, wassufficient to block tumorigenesis by A. tumefaciens (3).

Previous studies had indicated that the oncogenic suppres-sion effect of pSa upon A. tumefaciens did not result from alack of vir gene induction or T-DNA processing (6). In addi-tion, the presence of pSa did not affect the synthesis or mem-brane localization of VirB2, VirB3, VirB4, VirB9, and VirD4proteins in the bacterium (3). These results suggested that pSaexerted its oncogenic suppression effect at some point at orafter T-DNA transfer. We therefore examined the effect of pSaon later steps of the T-DNA transformation process.

We initially confirmed that pSa or a plasmid containing justthe osa gene inhibited tumorigenesis by A. tumefaciens A208on Arabidopsis root segments. According to these assays, on-cogenic suppression was complete, because we never saw thedevelopment of tumors on infected Arabidopsis roots. Similarresults were reported by Farrand et al. (10). In that study, A.tumefaciens C58 (a nopaline-type strain similar to A. tumefa-ciens A208) harboring pSa failed to induce tumors on carrotdiscs or Kalanchoe leaves. However, Farrand et al. reportedthat A. tumefaciens 1D1 (an octopine-type strain) containingpSa was able to incite, at low efficiency, small and slowly de-veloping tumors on carrot discs, Kalanchoe, sunflower, tomato,and marigold. We also have seen the development of smalltumors on D. stramonium infected with A. tumefaciens A277(an agropine-type supervirulent strain) harboring pSa (24a).This lack of complete oncogenic suppression suggests that pSareduces tumorigenicity by competing with some aspect of thenormal T-DNA transfer process.

Chernin et al. (4) previously reported that the IncW plas-mids pSa and R388 suppressed the oncogenicity of A. tumefa-ciens 1D1 by inhibiting auxin production by the bacterium.Application of the auxin indoleacetic acid to the plants a fewdays before inoculation restored oncogenicity to A. tumefaciens1D1(pSa). Because of this report, and because of the involve-ment of phytohormones in the stable transformation of plantcells to crown gall tumors (15), we investigated the effects ofpSa or the osa gene upon the transformation of plant cells toother stable phenotypes. We determined that osa inhibits

FIG. 4. Extracellular complementation of A. tumefaciens strains harboring pSa or osa. A. tumefaciens At221 (virE2 mutant, a T-DNA donor) and LBA4404 (a VirE2donor lacking T-DNA) containing various plasmids were coinoculated onto Kalanchoe leaves. The leaves were photographed after 1 month. The A. tumefaciens strainsare LBA4404, T-DNA2 VirE2 donor; At221, virE2 mutant T-DNA donor; At900, LBA4404(pSa); At901, LBA4404(pSa::neo); At902, LBA4404 (osa); At903, LBA4404(vector); At904, At221(pSa); At905, At221(pSa::neo); At906, At221 (osa); and At907, At221 (vector).



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

transformation of plant tissues to kanamycin resistance and theability to express GUS activity stably. We therefore concludethat the inhibitory effects of osa are not specific to the tumor-igenesis phenotype, but extend to the inhibition of genetictransformation to other stable phenotypes.

Although osa inhibited the genetic transformation of plantcells to express various phenotypes stably, it was possible thatthe effect of osa was on T-DNA integration and not T-DNAtransfer or nuclear targeting. We therefore investigated theeffect of osa on the transient transformation of plant cells byusing two different approaches. The first approach utilizedagroinfection with both components of the two-componentgeminivirus ToMoV. Although never formally proven, it hasbeen widely assumed that agroinfection of multimers of viralgenomes contained within the T-DNA results in the release ofviral DNA by homologous recombination within the plant nu-cleus. Successful agroinfection requires nuclear transport ofthe T-DNA, but presumably not T-DNA integration into theplant genome. The presence of osa within the inciting A. tu-mefaciens strains completely inhibited agroinfection, as deter-mined by the formation of ToMoV symptoms, and the specificdetection of viral DNA by squash blot and PCR analyses.These results indicate that osa inhibits the genetic transforma-tion of plants by A. tumefaciens at or before the stage ofnuclear translocation of the T-DNA. These results were con-firmed by analyzing the effect of osa upon the transient expres-sion of GUS activity in infected plant cells. We have previously

shown that expression of GUS activity, or the production ofgusA mRNA, does not require T-DNA integration into theplant genome (29, 31). The inhibition of expression of transientGUS activity in plant cells infected with A. tumefaciens strainscontaining osa additionally indicates that osa inhibits the tran-sient genetic transformation of plant cells.

We next wished to determine whether osa inhibited thetransfer of T-DNA from A. tumefaciens to plant cells. To dothis, we utilized the genetic approach of extracellular comple-mentation first described by Otten et al. (34). Our data indi-cated that osa does not inhibit T-DNA transfer from the T-DNA donor strain A. tumefaciens At221 to plant cells.However, when osa was present in the VirE2 protein donorstrain A. tumefaciens LBA4404, extracellular complementationwith A. tumefaciens At221 did not occur. We therefore con-clude that the primary effect of osa is to prevent VirE2 protein,and not T-DNA, transfer from the bacterium to the plant cell.

To confirm that osa primarily inhibits VirE2 protein but notT-DNA transfer, we utilized two additional approaches. First,we infected D. stramonium stems with the virE2 mutant strainA. tumefaciens At221. This A. tumefaciens mutant incited smalltumors on this highly sensitive plant. A. tumefaciens At221containing pSa or the osa gene retained the ability to incitesmall tumors on D. stramonium. This result further indicatesthat osa does not block T-DNA transfer. Second, we infectedleaf discs of wild-type and VirE2-producing transgenic tobaccowith A. tumefaciens At221 lacking or containing osa. Leaf discs

FIG. 5. Effect of osa on tumorigenesis by the virE2 mutant A. tumefaciens At221. Datura stems were inoculated with various A. tumefaciens strains, and the infectedstem sections were photographed 30 days later. Panels: 1, A. tumefaciens At789 (wild-type pTi); 2, A. tumefaciens At793 (no pTi); 3, A. tumefaciens At221 (virE2 mutant);4, A. tumefaciens At906 (At221 plus osa); 5, A. tumefaciens At907 (At221 plus vector); 6, A. tumefaciens At985 (At221 plus pJB31).



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

from wild-type plants did not develop tumors with any virE2mutant A. tumefaciens strain tested (data not shown). As firstshown by Citovsky et al. (5), VirE2-producing transgenic plantsdeveloped tumors when inoculated with the virE2 mutant A.tumefaciens At221. The virE2 mutant A. tumefaciens strain

containing osa incited tumors on these transgenic leaf discs ata frequency comparable to that incited by the strain lackingosa. These results further indicate that we could not detect aneffect, either qualitative or quantitative, of osa on T-DNAtransfer to plant cells. Binns et al. (2) showed that the presence

FIG. 6. Tumorigenesis of various A. tumefaciens strains on VirE2-producing transgenic tobacco leaf discs. Leaf discs of sterile virE2 transgenic tobacco plants wereinfected with various A. tumefaciens strains. After 2 days of cocultivation, the discs were incubated on MS medium containing Timentin and photographed 1 monthlater. (A) Panels: 1, infection with A. tumefaciens At789 (wild-type pTi); 2, A. tumefaciens At793 (no pTi); 3, A. tumefaciens At985 [At221 (pJB31)]; 4, A. tumefaciensAt989 (virE2 mutant); 5, A. tumefaciens At990 (At221 plus osa); 6, A. tumefaciens At991 (At221 plus vector). (B) Panels 1, no inoculation; 2, A. tumefaciens At928[A208(pBISN1)]; 3, A. tumefaciens At931 [A208(pBISN1 plus osa)]; 4, A. tumefaciens At932 [A208(pBISN1 plus vector)].



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

of pJW323, a derivative of the IncQ plasmid RSF1010, in A.tumefaciens greatly inhibited the ability of the bacterium toserve as a VirE2 protein donor, but inhibited its ability totransfer T-DNA by only twofold. They also showed thatpJW323 inhibited tumorigenesis by A. tumefaciens A348 byapproximately 90%, but did not completely suppress oncogen-esis by this strain. Our results with oncogenic suppression byosa differ from those of Binns et al. in several ways. First, weconsistently saw complete oncogenic suppression of A. tume-faciens A348 by pSa or osa (24a). Second, we could not detectan effect of pSa or osa on the ability of A. tumefaciens totransfer T-DNA to plant cells. Third, the expression of only theosa gene was sufficient to effect oncogenic suppression. Wehave recently shown that the osa gene, when incorporated intothe A. tumefaciens chromosome, could still effect oncogenicsuppression (24b). Thus, no conjugal intermediate of pSa wasnecessary. This differs from the IncQ system, in which theformation of a conjugal intermediate is necessary to causeoncogenic suppression (39). Fourth, unlike the situation withthe RSF1010 derivative pJW323 (2, 42), we were not able toovercome the oncogenic suppressive effect of osa by the coor-dinate overexpression of VirB9, VirB10, and VirB11 proteins.However, our virB expression plasmid (pUCD5542) could notreverse oncogenic suppression of RSF1010, as described byWard et al. (42). Therefore, the differences between our in-ability to overcome oncogenic suppression of RSF1010 and thesuccess of Ward et al. may have been more quantitative thanqualitative. Overall, however, pSa may be more efficient incompeting for the virB-encoded VirE2 transport apparatusthan is RSF1010. Recent experiments indicate that pSa andRSF1010 directly compete with each other for the same exportapparatus (24b).

Although we have shown that the molecular mechanism ofoncogenic suppression by osa involves inhibition of VirE2 pro-tein export from A. tumefaciens, we have not yet determinedthe details of how this inhibition is effected. The observationsthat VirE2 is an abundant protein in A. tumefaciens (9) andthat Osa is immunologically difficult to detect unless overex-pressed (24a) suggest that it is unlikely that Osa interferes withVirE2 export by stoichiometrically binding to VirE2. It is pos-sible that Osa interacts with VirE1 protein that is needed forVirE2 protein export from the bacterium (40). It is also pos-

sible that Osa interacts with one of the proteins of the virB-virD4-encoded protein export apparatus.

Finally, our data showing that osa inhibits VirE2 protein butnot T-DNA transfer, as well as our data showing that a virE2mutant strain of A. tumefaciens can still incite small tumors onD. stramonium, lend further evidence to the model of Binns etal. (2) and Sundberg et al. (40) that a T-DNA–VirD2 complexis transported to plant cells separately from VirE2 protein.

ACKNOWLEDGMENTS

We thank Vitaly Citovsky for providing seeds of VirE2-producingtransgenic tobacco plants and Robert Gilbertson for providing strainsand assistance for agroinfection and assistance in squash blot and PCRanalyses.

This work was supported by grants from the U.S. Department ofAgriculture (95-37301-2040) to S.B.G. and NIH grant GM45550 fromthe National Institute of General Medicine to C.I.K.

REFERENCES

1. Beaupre, C. E., J. Bohne, E. M. Dale, and A. N. Binns. 1997. Interactionsbetween VirB9 and VirB10 membrane proteins involved in movement ofDNA from Agrobacterium tumefaciens into plant cells. J. Bacteriol. 179:78–89.

2. Binns, A. N., C. E. Beaupre, and E. M. Dale. 1995. Inhibition of VirB-mediated transfer of diverse substrates from Agrobacterium tumefaciens bythe IncQ plasmid RSF1010. J. Bacteriol. 177:4890–4899.

3. Chen, C.-Y., and C. I. Kado. 1994. Inhibition of Agrobacterium tumefaciensoncogenicity by the osa gene of pSa. J. Bacteriol. 176:5697–5703.

4. Chernin, L. S., E. V. Lobanok, V. V. Fomicheva, and N. A. Kartel. 1984.Crown gall-suppressive IncW R plasmids cause a decrease in auxin produc-tion in Agrobacterium tumefaciens. Mol. Gen. Genet. 195:195–199.

5. Citovsky, V., J. Zupan, D. Warnick, and P. Zambryski. 1992. Nuclear local-ization of Agrobacterium VirE2 protein in plant cells. Science 256:1802–1805.

6. Close, S. M. 1991. Genetic and molecular analysis of plasmid pSa conferredsuppression of Agrobacterium tumefaciens oncogenicity, and identification ofan independent pSa locus encoding a nuclease. Ph.D. thesis. University ofCalifornia, Davis.

7. Close, S. M., and C. I. Kado. 1991. The osa gene of pSa encodes a 21.1-kilodalton protein that suppresses Agrobacterium tumefaciens oncogenicity.J. Bacteriol. 173:5449–5456.

8. Close, S. M., and C. I. Kado. 1992. A gene near the plasmid pSa origin ofreplication encodes a nuclease. Mol. Microbiol. 6:521–527.

9. Engstrom, P., P. Zambryski, M. Van Montagu, and S. Stachel. 1987. Char-acterization of Agrobacterium tumefaciens virulence proteins induced by theplant factor acetosyringone. J. Mol. Biol. 197:635–645.

10. Farrand, S. K., C. I. Kado, and C. R. Ireland. 1981. Suppression of tumor-igenicity by the IncW plasmid pSa in Agrobacterium tumefaciens. Mol. Gen.Genet. 181:44–51.

FIG. 7. Infection of D. stramonium stems with A. tumefaciens strains containing multiple copies of virB9, -10, and -11. D. stramonium seedlings were wounded andinoculated with A. tumefaciens A208 containing pBISN1, pUCD5542, and no additional plasmid (At935) (A); pSa (At925) (B); osa (At926) (C); or vector (At927) (D).The stem sections were photographed 1 month after inoculation.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

11. Fong, S. T., and V. A. Stanisich. 1989. Location and characterization of twofunctions on RP1 that inhibit the fertility of the IncW plasmid R388. J. Gen.Microbiol. 135:499–502.

12. Frey, J., and M. Bagdasarian. 1989. The molecular biology of IncQ plasmids,p. 125–163. In C. M. Thomas (ed.), Promiscuous plasmids of Gram-negativebacteria. Academic Press, New York, N.Y.

13. Gallie, D. R., S. Novak, and C. I. Kado. 1985. Novel high- and low-copystable cosmids for use in Agrobacterium and Rhizobium. Plasmid 14:171–175.

14. Garfinkel, D. J., R. B. Simpson, L. W. Ream, F. F. White, M. P. Gordon, andE. W. Nester. 1981. Genetic analysis of crown gall: fine structure map of theT-DNA by site-directed mutagenesis. Cell 27:143–153.

15. Gelvin, S. B. 1990. Crown gall disease and hairy root disease: a sledgeham-mer and a tackhammer. Plant Physiol. 92:281–285.

16. Gilbertson, R. L., S. H. Hidayat, R. T. Martinez, S. A. Leong, J. C. Faria,F. J. Morales, and D. P. Maxwell. 1991. Differentiation of bean-infectinggeminiviruses by nucleic acid hybridization probes and aspects of beangolden mosaic in Brazil. Plant Dis. 75:336–342.

17. Gilbertson, R. L., S. H. Hidayat, E. J. Paplomatas, M. R. Rojas, Y.-M. Hou,and D. P. Maxwell. 1993. Pseudorecombination between infectious clonedDNA components of tomato mottle and bean dwarf mosaic geminiviruses.J. Gen. Virol. 74:23–31.

18. Gilbertson, R. L., M. R. Rojas, D. R. Russell, and D. P. Maxwell. 1991. Useof the asymmetric polymerase chain reaction and DNA sequencing to de-termine genetic variability of bean golden mosaic geminivirus in the Domin-ican Republic. J. Gen. Virol. 72:2843–2848.

19. Grimsley, N., B. Hohn, T. Hohn, and R. Walden. 1986. “Agroinfection,” analternative route for viral infection of plants by using the Ti plasmid. Proc.Natl. Acad. Sci. USA 83:3282–3286.

20. Hernalsteens, J. P., R. Villarroel-Mandiola, M. Van Montagu, and J. Schell.1977. Transposition of Tn1 to a broad-host-range drug resistance plasmid, p.179–183. In A. I. Bukhari, J. A. Shapiro, and S. L. Adhya (ed.), DNAinsertion elements, plasmids, and episomes. Cold Spring Harbor Laboratory,Cold Spring Harbor, N.Y.

21. Jefferson, R. A., T. A. Kavanagh, and M. W. Bevan. 1987. GUS fusions:b-glucuronidase as a sensitive and versatile gene fusion marker in higherplants. EMBO J. 6:3901–3907.

22. Kado, C. I. 1993. Agrobacterium-mediated transfer and stable incorporationof foreign genes in plants, p. 243–254. In D. B. Clewell (ed.), Bacterialconjugation. Plenum Press, New York, N.Y.

23. Kado, C. I., and M. G. Heskett. 1970. Selective media for isolation ofAgrobacterium, Corynebacterium, Erwinia, Pseudomonas, and Xanthomonas.Phytopathology 60:969–976.

24. Klapwijk, P., P. Van Beelen, and R. A. Schilperoort. 1979. Isolation of arecombination deficient Agrobacterium tumefaciens mutant. Mol. Gen.Genet. 173:171–175.

24a.Lee, L.-Y. Unpublished observations.24b.Lee, L.-Y., et al. Unpublished data.25. Lessl, M., and E. Lanka. 1994. Common mechanisms in bacterial conjuga-

tion and Ti-mediated T-DNA transfer to plant cells. Cell 77:321–324.26. Lichtenstein, C., and J. Draper. 1986. Genetic engineering of plants, p.

67–119. In D. M. Glover (ed.), DNA cloning: a practical approach, vol 2. IRLPress, Washington, D.C.

27. Liu, C.-N., X.-Q. Li, and S. B. Gelvin. 1992. Multiple copies of virG enhancethe transient transformation of celery, carrot, and rice tissues by Agrobacte-rium tumefaciens. Plant Mol. Biol. 20:1071–1087.

28. Loper, J. E., and C. I. Kado. 1979. Host range conferred by the virulence-specifying plasmid of Agrobacterium tumefaciens. J. Bacteriol. 139:591–596.

29. Mysore, K. S., B. Bassuner, X.-B. Deng, N. S. Darbinian, A. Motchoulski, W.Ream, and S. B. Gelvin. Role of the Agrobacterium tumefaciens VirD2protein in T-DNA transfer and integration. Mol. Plant-Microbe Interact.11:668–683.

30. Nam, J., A. G. Matthysse, and S. B. Gelvin. 1997. Differences in susceptibilityof Arabidopsis ecotypes to crown gall disease may result from a deficiency inT-DNA integration. Plant Cell 9:317–333.

31. Narasimhulu, S. B., X.-B. Deng, R. Sarria, and S. B. Gelvin. 1996. Earlytranscription of Agrobacterium T-DNA genes in tobacco and maize. PlantCell 8:873–886.

32. Ni, M., D. Cui, J. Einstein, S. Narasimhulu, C. E. Vergara, and S. B. Gelvin.1995. Strength and tissue specificity of chimeric promoters derived from theoctopine and mannopine synthase genes. Plant J. 7:661–676.

33. Ooms, G., L. Molendijk, and R. A. Schilperoort. 1982. Double infection oftobacco plants by two complementing octopine T-region mutants of Agrobac-terium tumefaciens. Plant Mol. Biol. 1:217–226.

34. Otten, L., H. De Greve, J. Leemans, R. Hain, P. Hooykaas, and J. Schell.1984. Restoration of virulence of vir region mutations of Agrobacteriumtumefaciens strain B6S3 by coinfection with normal and mutant Agrobacte-rium strains. Mol. Gen. Genet. 195:159–163.

35. Rogowsky, P. M., T. J. Close, J. A. Chimera, J. J. Shaw, and C. I. Kado. 1987.Regulation of the vir genes of Agrobacterium tumefaciens plasmid pTiC58. J.Bacteriol. 169:5101–5112.

36. Sciaky, D. A., A. L. Montoya, and M.-D. Chilton. 1978. Fingerprints ofAgrobacterium Ti plasmids. Plasmid 1:238–253.

37. Stachel, S. E., and P. C. Zambryski. 1986. Agrobacterium tumefaciens and thesusceptible plant cell: a novel adaptation of extracellular recognition andDNA conjugation. Cell 47:155–157.

38. Stachel, S. E., and E. W. Nester. 1986. The genetic and transcriptionalorganization of the vir region of the A6 Ti plasmid of Agrobacterium tume-faciens. EMBO J. 5:1445–1454.

39. Stahl, L. E., A. Jacobs, and A. N. Binns. 1998. The conjugal intermediate ofplasmid RSF1010 inhibits Agrobacterium tumefaciens virulence and VirB-dependent export of VirE2. J. Bacteriol. 180:3933–3939.

40. Sundberg, C., L. Meek, K. Carroll, A. Das, and W. Ream. 1996. VirE1protein mediates export of the single-stranded DNA-binding protein VirE2from Agrobacterium tumefaciens into plant cells. J. Bacteriol. 178:1207–1212.

41. Valentine, C. R. I., and C. I. Kado. 1989. Molecular genetics of IncWplasmids, p. 125–163. In C. M. Thomas (ed.), Promiscuous plasmids ofGram-negative bacteria. Academic Press, New York, N.Y.

42. Ward, J. E., E. M. Dale, and A. N. Binns. 1991. Activity of the AgrobacteriumT-DNA transfer machinery is affected by VirB gene products. Proc. Natl.Acad. Sci. USA 88:9350–9354.

43. Watson, B., T. C. Currier, M. P. Gordon, M.-D. Chilton, and E. W. Nester.1975. Plasmid required for virulence of Agrobacterium tumefaciens. J. Bac-teriol. 123:255–264.



http://jb.asm.org/

Dow

nloaded from

http://jb.asm.org/

pSa Causes Oncogenic Suppression of Agrobacterium by … · At905 At221(pSa::neo) Carr Chlr Genr Kanr Neor Rifr Sper Chlr This study At906 At221(pUCD3960) Carr Rifr Sper This study

Documents