Characteristics of Agrobacterium tumefaciens auxotrophic mutant infectivity

Lippincott, Barbara B. (Northwestern University, Evanston, Ill.), and James A. Lippincott. Characteristics of Agrobacterium tumefaciens auxotrophic mutant infectivity. J. Bacteriol. 92:937-945. 166.-Mutants of Agrobacterium tumefaciens auxotrophic for adenine, methionine, or asparagine are less infectious than the wild-type strain B6 from which they were derived and show increased infectivity on pinto bean leaves when the specific compounds required for growth of the mutants are added to the infected leaf. Reversion to a prototrophic form of nutrition is accompanied by increased infectivity. Tumors initiated by these auxotrophic mutants are shown to arise only at large wound sites where nutritional conditions may be less restricting. The data indicate that, after inoculation, the bacteria pass through a phase in which host-supplied nutrients are utilized for the production of one or more factors necessary for successful tumor initiation.     

Bioassay and attributes of a growth factor associated with crown gall tumors

An improved bioassay is described for a factor that promotes tumor growth which was first obtained from extracts of pinto bean leaves with crown gall tumors. Sixteen primary pinto bean leaves per sample are inoculated with sufficient Agrobacterium tumefaciens to initiate about 5 to 10 tumors per leaf and treated with tumor growth factor at day 3 after inoculation. The diameters of 30 to 48 round tumors (no more than 3 randomly selected per leaf) are measured per test sample at day 6. Mean tumor diameter increased linearly with the logarithm of the concentration of tumor growth factor applied. The tumor growth factor was separated by column chromatography from an ultraviolet light-absorbing compound previously reported to be associated with fractions having maximal tumor growth factor activity. Partly purified tumor growth factor showed no activity in a cytokinin bioassay or an auxin bioassay, and negligible activity in gibberellin bioassays. Representatives of these three classes of growth factors did not promote tumor growth. Extracts from crown gall tumors on primary pinto bean leaves, primary castor bean leaves, Bryophyllum leaves, carrot root slices, and tobacco stems showed tumor growth factor activity, whereas extracts from healthy control tissues did not. Extracts from actively growing parts of healthy pinto beans, Bryophyllum, and tobacco, however, showed tumor growth factor activity. Tumor growth factor is proposed to be a normal plant growth factor associated with rapidly growing tissues. Its synthesis may be activated in nongrowing tissues by infection with Agrobacterium sp.     

Tumor Growth Complementation Among Strains of Agrobacterium

The ability of 31 strains of Agrobacterium to initiate the production of a tumor growth factor (TGF) which is associated with crown-gall tumors on primary pinto bean leaves was determined. Extracts from bean leaves inoculated with these bacteria were tested and they showed that 16 of the 19 strains that induced tumors on the leaves also initiated TGF production. The three strains for which no TGF was detected were of low infectivity and included two strains of A. tumefaciens and a strain of A. rhizogenes. Five of the 12 strains that did not induce pinto bean leaf tumors were found to initiate TGF production. Representatives of A. tumefaciens, A. rhizogenes, and A. radiobacter among these 12 strains were present in both categories. Mixed inocula composed of one of the three infectious TGF-negative strains and one of the five nontumorigenic TGF-positive strains resulted in increased growth of tumors induced by the former. These growth changes were not correlated with changes in tumor number. The ability of different strains to show these tumor growth complementation effects corresponded fully with their ability to initiate TGF, as determined by the assay of leaf extracts. The nontumorigenic TGF-positive strains also promoted the growth of tumors initiated by low concentrations of strain B6. These complementation effects were due, therefore, to the same TGF found in extracts of B6 inoculated leaves and of leaves inoculated with most tumorigenic as well as many nontumorigenic strains of Agrobacterium. Heat-inactivated cells of strain B6 failed to initiate sufficient TGF to be detected in extracts, and heat-inactivated cells of several strains failed to show tumor growth complementation, indicating bacterial viability to be one prerequisite for TGF initiation. Heat inactivated cells also inhibited TGF production by viable cells, similar to their ability to inhibit tumor initiation. Consequently, bacteria capable of attaching to the A. tumefaciens infection site may initiate one of four patterns of events: (i) TGF production only, (ii) tumor induction only, (iii) both, or (iv) neither. Suggestive evidence for a second tumor-associated growth factor is presented.     

Evidence for a Tumor-associated Factor Active in the Promotion of Crown-gall Tumor Growth on Primary Pinto Bean Leaves

The growth of crown-gall tumors on primary pinto bean leaves (Phaseolus vulgaris L. cv. “Pinto”) between day 3 and day 6 after inoculation was found to be proportional to the number of tumors on the leaves. Similar differences observed in the growth of tumors induced by adenine, methionine and asparagine requiring mutants of Agrobacterium tumefaciens (Smith and Town.) Conn appear to be due to the same phenomenon. Tumors induced by these auxotrophs thus show no obvious growth differences from those induced by the prototrophic strain despite the lower specific infectivity and the existence of a mutational lesion in these bacteria. A diffusible growth factor(s) produced by the tumor tissue is proposed to account for the relation between tumor number and early tumor growth.     

Bacterial Attachment to a Specific Wound Site as an Essential Stage in Tumor Initiation by Agrobacterium tumefaciens

The number of tumors initiated by Agrobacterium tumefaciens strain B6 on primary pinto bean leaves was decreased when cells of an avirulent strain (IIBNV6) were included in the inoculum. With sufficient B6 cells to initiate ca. 50% of the maximal number of tumors per leaf, inhibition was detected at a 1:1 ratio of B6 to IIBNV6 cells and increased linearly with the logarithm of the number of IIBNV6. Varying the number of B6 in the presence of a constant number of IIBNV6 or varying the number of both, while maintaining a constant ratio of B6 to IIBNV6, showed that the inhibition was a function of the absolute concentration of each cell type. The data fit a one-particle dose response curve, which indicates that a single IIBNV6 cell can prevent tumor initiation by a single B6 cell. Inhibition was obtained with mixed inocula and when the addition of IIBNV6 preceded B6, but not when B6 preceded IIBNV6. Heat-inactivated IIBNV6 inhibited, as did ultraviolet or heat-inactivated B6. Several unrelated bacteria and certain strains of Agrobacterium failed to inhibit, whereas other related strains gave inhibition. Attachment of IIBNV6 to a specific would site, thus excluding B6 from the site, is proposed to account for these data. A specific complementary binding of a virulent bacterium to a host wound site exposed by the inoculation procedure is suggested as an essential early event in the crown-gall tumor initiation process.     

QUANTITATIVE MEASUREMENTS OF INTERACTIONS BETWEEN CROWN-GALL TUMORS AND THE PINTO BEAN HOST

Interactions between crown-gall tumors on the primary pinto bean leaf and the pinto bean seedling (Phaseolus vulgaris L. ‘Pinto‘) were estimated by quantitative measurements of tumor initiation and growth as affected by certain modifications of the host. Effects of the tumors on the host were estimated by measurements of host growth and correlation responses. The presence of crown-gall tumors was found to reduce the growth of the leaf in area but to nearly double the weight of the leaf 9 days after inoculation with Agrobacterium tumefaciens (Smith and Town.) Conn, strain B6. The presence of tumors on only one of the two primary leaves resulted in a decrease in the weight of the leaf without tumors, showing the tumors to be effective mobilization centers. Tumors also delayed the abscission of petiole explants and delayed the growth of the epicotyl bud, both reminiscent of auxin effects. The excision of the cotyledons, the epicotyl bud, or one of the pair of primary leaves at the time of inoculation increased the number of tumors initiated per leaf. Removing the epicotyl bud or one of the primary leaves, or placing a cytokinin on one of the leaves, altered leaf growth but failed to alter tumor growth, indicating that tumor growth is not affected by the changes responsible for the compensatory growth effects induced by these treatments. Tumor growth was shown to be generally correlated with leaf growth from day 2 through 8 after inoculation, suggesting that the factors normally limiting leaf growth in a determinate type leaf are also active in limiting tumor growth. The changes in the plant cell responsible for the enhanced rate of growth seen in crown-gall tumor cells, therefore, appear to occur in regulatory systems other than those normally limiting leaf growth. The regulatory systems that are affected may be identical with those activated in compensatory host growth effects.     

Role of bacterial cellulose fibrils in Agrobacterium tumefaciens infection.

During the attachment of Agrobacterium tumefaciens to carrot tissue culture cells, the bacteria synthesize cellulose fibrils. We examined the role of these cellulose fibrils in the attachment process by determining the properties of bacterial mutants unable to synthesize cellulose. Such cellulose-minus bacteria attached to the carrot cell surface, but, in contrast to the parent strain, with which larger clusters of bacteria were seen on the plant cell, cellulose-minus mutant bacteria were attached individually to the plant cell surface. The wild-type bacteria became surrounded by fibrils within 2 h after attachment. No fibrils were seen with the cellulose-minus mutants. Prolonged incubation of wild-type A. tumefaciens with carrot cells resulted in the formation of large aggregates of bacteria, bacterial fibrils, and carrot cells. No such aggregates were formed after the incubation of carrot cells with cellulose-minus A. tumefaciens. The absence of cellulose fibrils also caused an alteration in the kinetics of bacterial attachment to carrot cells. Cellulose synthesis was not required for bacterial virulence; the cellulose-minus mutants were all virulent. However, the ability of the parent bacterial strain to produce tumors was unaffected by washing the inoculation site with water, whereas the ability of the cellulose-minus mutants to form tumors was much reduced by washing the inoculation site with water. Thus, a major role of the cellulose fibrils synthesized by A. tumefaciens appears to be anchoring the bacteria to the host cells, thereby aiding the production of tumors.     

THE INDUCTION OF LEAF TUMORS BY AGROBACTERIUM TUMEFACIENS

Using carborundum as an abrasive and light rubbing with a culture of Agrobacterium tumefaciens, leaves of various species of bean and tobacco develop tumors on the leaf lamina. The induction of these tumors requires wounding, the presence of a virulent strain of the bacterium and is due to the bacterium, not substances released into the bacterial culture medium during growth. Observations of the histology and cytology of these tumors on the primary leaves of pinto bean show no significant differences from the more commonly studied stem tumors. The tumors on pinto beans first appear as chlorotic nests of dividing cells which gradually accumulate chlorophyll, eventually becoming dark green in color as opposed to the surrounding leaf tissue which is completely chlorotic at this stage. Tumor development is enhanced by a dark period following inoculation while growth of the leaf is essentially stopped. The tumors thus exhibit a pattern of growth and development independent of that of the normal leaf. The number of tumors obtained on pinto bean leaves was found to depend on the concentration of bacteria in the inoculum and on the age of the plants. A sharp peak in response was observed at about 7 days from planting. Best results were obtained by adding the bacterium at the time of wounding. The tumors were shown to differ from IAA-induced leaf proliferations with respect to their point of origin on the leaf, morphology, physiology and development.     

Characterization of nonattaching mutants of Agrobacterium tumefaciens.

The first step in tumor formation by Agrobacterium tumefaciens is the site-specific binding of the bacteria to plant host cells. Transposon mutants of the bacteria which fail to attach to carrot suspension culture cells were isolated. These mutants showed no significant attachment to carrot cells with either microscopic or viable cell count assays of bacterial binding. The nonattaching mutants were all avirulent. When revertants of the mutants were obtained by enriching for bacteria which do bind to carrot cells, the bacteria were found to have regained the ability to bind to carrot cells and virulence simultaneously. These results suggest that the ability of the bacteria to bind to plant cells is required for virulence. Like the parent strain, all of the nonattaching mutants synthesized cellulose, but unlike the parent strain, they failed to aggregate carrot suspension culture cells. The transposon Tn5, which was used to obtain the mutants, was located on a 12-kilobase EcoRI fragment of the bacterial chromosomal DNA in all of the nonattaching mutants from strain C58. That the mutant phenotype was due to the Tn5 insertion was shown by cloning the Tn5-containing DNA fragment from the mutant bacteria and using it to replace the wild-type fragment in the parent strain by marker exchange. The resulting bacteria had the same mutant phenotype as the original Tn5 mutants; they did not attach to carrot cells, they did not cause the aggregation of carrot cells, and they were avirulent. No difference was seen between the parent strain and the nonattaching mutants in hydrophobicity, motility, flagella, fimbriae, beta-2-glucan content, size of lipopolysaccharide, or ability of the lipopolysaccharide to inhibit bacterial attachment to tissue culture cells. Differences were seen between the parent strain and the nonattaching mutants in the polypeptides removed from the bacteria during the preparation of spheroplasts. Three of the mutants were lacking a polypeptide of about 34 kilodaltons (kDa). One mutant was lacking the 34-kDa polypeptide and another polypeptide of about 38 kDa. The fifth mutant was lacking a polypeptide slightly smaller than the 34-kDa polypeptide missing in the other four mutants. These missing polypeptides all reappeared in the revertants of the mutants. Thus, bacterial binding to plant cells appears to require the presence of these polypeptides.     

Extraction, Assay and Partial Purification of a Factor from Tumorous Leaves that Promotes Crown-gall Tumor Growth

A tumor growth factor was extracted from primary pinto bean leaves {Phaseolus vulgaris L. cv. “Pinto”) having crown-gall tumors. Application of these extracts to primary pinto bean leaves at day 3 after inoculation with Agrobacterium tumefaciens (Smith and Town.) Conn resulted in an increase in the diameter of tumors on these leaves after 24 hours that was 25–60 percent greater than that of the controls. No growth promoting activity was detected in comparable extracts from control leaves or from cultures of the bacterium. The tumor growth factor did not affect tumor number when applied in this fashion. The active extracts were fractionated on Sephadex (1–15 columns and an active component obtained which eluted in a volume suggesting a molecular weight in the range of 100–200. A spectrum of the material in this fraction is shown and some of the characteristics of this material are described. Reasons are presented for considering this growth factor to be identical with at least one of the growth factors previously hypothesized to account for the greater tumor growth observed as tumor number is increased in this system.     

Conditioned medium promotes the attachment ofAgrobacterium tumefaciens strain NT 1 to carrot cells

Summary Wild-typeAgrobacterium tumefaciens bind to carrot suspension culture cells. Avirulent strain NT 1 did not bind to carrot cells when they were incubated together in Murashige and Skoog medium. Conditioned medium was prepared by incubatingA. tumefaciens virulent strain C 58 with carrot cells and removing the bacteria and carrot cells using filter sterilization. This conditioned medium promoted the binding of NT 1 to carrot cells. Conditioned medium did not promote the nonspecific attachment ofEscherichia coli to carrot cells. These results suggest that when wild-typeA. tumefaciens are incubated with plant host cells, some substance(s) involved in bacterial attachment are released into the medium. Filter-sterilized medium from the incubation of the nonattachingchvB mutant A 1045 with carrot cells promoted the attachment of strain NT 1 even though A 1045 bacteria did not bind to the carrot cells. However, filter-sterilized medium from the incubation of the non-attachingatt mutant Att-B 123 with carrot cells was unable to promote the binding of strain NT 1. This suggests that nonattaching mutants ofA. tumefaciens can be divided into two groups on the basis of the properties of the substances released into the medium when the bacteria are incubated with carrot cells.Abbreviations MS Murashige and Skoog tissue culture medium Dedicated to the memory of Professor John G. Torrey     

Agrobacterium rhizogenes mutants that fail to bind to plant cells.

Transposon insertion mutants of Agrobacterium rhizogenes were screened to obtain mutant bacteria that failed to bind to carrot suspension culture cells. A light microscope binding assay was used. The bacterial isolates that were reduced in binding to carrot cells were all avirulent on Bryophyllum diagremontiana leaves and on carrot root disks. The mutants did not appear to be altered in cellulose production. The composition of the medium affected the ability of the parent and mutant bacteria to bind to carrot cells. The parent strain bound to carrot cells in greatest numbers in low-ionic-strength media such as 4% sucrose but still showed significant binding in Murashige-Skoog tissue culture medium. All of the mutants showed reduced binding in 4% sucrose after 2 h of incubation with carrot cells. One mutant was delayed in binding in 4% sucrose. This mutant and one other mutant also showed reduced binding to carrot cells in Murashige-Skoog medium. To determine whether the Tn5 insertion was responsible for the mutant phenotype, DNA containing the Tn5 insertion was cloned from the mutant bacteria and used to introduce Tn5 into the parent strain in the same location as in the original mutant by marker exchange. The resulting transconjugants had the same avirulent, nonattaching phenotype as the original mutants, suggesting that the mutant phenotype was due to the Tn5 insertion. The cloned DNA containing the Tn5 insertion was also tested for homology to DNA of known genes that affect attachment of Agrobacterium tumefaciens to plant cells by DNA hybridization. No homology to chv, att, or pscA clones was observed. In addition, cloned chv, att, and pscA genes from A. tumefaciens were unable to complement the attachment-minus A. rhizogenes mutants. Thus, the A. rhizogenes nonattaching mutants appear to be different from the previously described A. tumefaciens mutants.     

Involvement of a vitronectin-like protein in attachment of Agrobacterium tumefaciens to carrot suspension culture cells.

Infections of dicotyledonous plants by Agrobacterium tumefaciens result in the formation of crown gall tumors. Attachment of the bacteria to plant host cells is required for tumor formation. Human vitronectin and antivitronectin antibodies both inhibited the binding of A. tumefaciens to carrot cells. Wild-type bacteria are able to bind radioactive vitronectin; nonattaching mutants showed a reduction in the ability to bind vitronectin. The binding of biotype 1 A. tumefaciens to carrot cells or to radioactive vitronectin was not affected by high ionic strength. Detergent extraction of carrot cells removed the receptor to which the bacteria bind. The extract was found to contain a vitronectin-like protein. These results suggest that A. tumefaciens utilizes a vitronectin-like protein on the plant cell surface as the receptor for its initial attachment to host cells.     

Mechanism of Polymyxin B Resistance in Proteus mirabilis

The lipids from three types of organisms-a Proteus mirabilis wild type highly resistant to polymyxin B, a polymyxin B-sensitive mutant derived from the wild type, and the wild type grown in the presence of sulfadiazine resulting in phenotypic conversion to polymyxin B sensitivity-were examined to determine the nature of polymyxin B resistance. The phospholipid compositions were nearly identical; each organism contained similar small amounts of N-methyl phosphatidylethanolamine in addition to comparable quantities of phosphatidylethanolamine, phosphatidylglycerol, and cardiolipin. the fatty acid compositions were similar in the exponential phase of growth; in the stationary phase, sulfadiazine markedly inhibited the synthesis of cyclopropane fatty acids. Liposomes prepared from the dried lipids of the three types of organisms were extensively and similarly disrupted by the polymyxin. These findings suggest that polymyxin B resistance in P. mirabilis is determined by the cell envelope which prevents access of the antibiotic to the susceptible lipid target sites.     

Tumor induction by agrobacterium involves attachment of the bacterium to a site on the host plant cell wall

Cell wall preparations from primary bean leaves were found to inhibit tumor initiation by Agrobacterium tumefaciens strain B6 when inoculated with the bacteria on bean leaves. Membrane fractions from these same leaves were noninhibitory. The cell walls were effective when applied prior to or with bacteria, but application of cell walls about 15 minutes after bacteria did not affect the number of tumors initiated. Much of the inhibitory activity of the plant cell walls was eliminated by pretreatment with dead site-attaching bacteria or with lipopolysaccharide from these bacteria. Cells and lipopolysaccharide from non-site-attaching agrobacteria had no effect on the activity of the plant cell walls. About 30% inhibition of tumor initiation was obtained with plant cell walls at 50 μg/ml dry weight, and at 10 mg/ml dry weight about 70% inhibition was typical. Both early and late appearing tumors were affected by the cell walls, indicating that they do not exclusively affect tumors arising from either small or large wounds. These data show that plant cell walls but not membranes contain surfaces to which A. tumefaciens adheres and these exhibit the specificity typical of the host site to which virulent agrobacteria must attach to induce tumors. It is concluded that some portion of wound-exposed plant cell wall constitutes the host adherence site in Agrobacterium infections.     

Resistance of Common Bean (Phaseolus vulgaris) to Bacterial Wilt Caused by Curtobacterium flaccumfaciens pv. flaccumfaciens

Bacterial wilt caused by Curtobacterium flaccumfaciens pv. flaccumfaciens is an important new disease of common bean (Phaseolus vulgaris) in western Canada. Both yellow and orange variants of the pathogen were found in the region. A controlled environment study was conducted to assess 124 common bean cultivars and lines from eight market classes for resistance to the yellow and orange variants of the pathogen, using the hilum injury/seed inoculation method. Results of the screening tests showed significant (P < 0.05) differences in resistance to bacterial wilt among the cultivars or lines. The great northern line L02E317, the great northern cultivar Resolute and pinto lines L02B662 and 999S‐2A, were highly resistant to both variants of the pathogen, with disease severity indices of 0 on a rating scale of 0 (no wilt symptoms) to 5 (dead seedling). Resistant cultivars or lines were found among black, great northern, pink, pinto, small red and Flor de Mayo bean market classes. The study concludes that new bacterial wilt‐resistant germplasm exists among Canadian bean cultivars and lines, and constitutes a valuable resource for breeding common beans for resistance to both yellow and orange variants of C. flaccumfaciens pv. flaccumfaciens.     

Effect of gibberellic Acid on crown gall tumor induction in aging primary pinto bean leaves

Gibberellic acid was tested for its effect on tumor induction by Agrobacterium tumefaciens in primary pinto bean (Phaseolus vulgaris) leaves in various stages of development. The hormone was found to promote tumor induction in partially aged leaves but did not effect tumor induction in very young leaves or in fully matured leaves. It is suggested that the natural loss of susceptibility to tumor induction in maturing pinto bean leaves is associated with a concomitant loss of endogenous gibberellins and/or a sensitivity to gibberellins.     

Outer membrane protein H1 of Pseudomonas aeruginosa: involvement in adaptive and mutational resistance to ethylenediaminetetraacetate, polymyxin B, and gentamicin.

It is well established that Pseudomonas aeruginosa cells grown in Mg2+-deficient medium acquire nonmutational resistance to the chelator ethylenediaminetetraacetate and to the cationic antibiotic polymyxin B; this type of resistance can be reversed by transferring the cells to Mg2+-sufficient medium for a few generations. Stable mutants resistant to polymyxin B were isolated and shown to have also gained ethylenediaminetetraacetate resistance. Both the mutants and strains grown on Mg2+-deficient medium had greatly enhanced levels of outer membrane protein H1 when compared with the wild-type strain or with revertants grown in Mg2+-sufficient medium. It was determined that in all strains and at all medium Mg2+ concentrations, the cell envelope Mg2+ concentration varied inversely with the amount of protein H1. In addition, the increase in protein H1 in the mutants was associated with an increase in resistance to another group of cationic antibiotics, the aminoglycosides, e.g., gentamicin. We propose that protein H1 acts by replacing Mg2+ at a site on the lipopolysaccharide which can otherwise be attacked by the cationic antibiotics or ethylenediaminetetraacetate.     

In vitro assembly of microtubules from tubulins of several higher plants

Tubulins were isolated by a combination of affinity (ethyl N-phenylcarbamate-Sepharose 4B) and ion exchange (DEAE-Sephacel) chromatography from several higher plants (mung bean, pea, whole pod bean, zucchini, cucumber seedlings and carrot suspension cultured cells). All these higher plant tubulins readily polymerized to microtubules in a polymerization medium containing GTP, Mg2+, EGTA, leupeptin and DMSO. Tubulins from mung bean, pea and whole pod bean showed identical behaviour on polyacrylamide gel electrophoresis but differed from carrot zucchini and cucumber tubulin. Consequently, tubulin of higher plants seems to have different molecular properties in different plant species.