Elevated temperature differentially affects virulence, VirB protein accumulation, and T-pilus formation in different Agrobacterium tumefaciens and Agrobacterium vitis strains

Abstract

That gene transfer to plant cells is a temperature-sensitive process has been known for more than 50 years. Previous work indicated that this sensitivity results from the inability to assemble a functional T pilus required for T-DNA and protein transfer to recipient cells. The studies reported here extend these observations and more clearly define the molecular basis of this assembly and transfer defect. T-pilus assembly and virulence protein accumulation were monitored in Agrobacterium tumefaciens strain C58 at different temperatures ranging from 20 degrees C to growth-inhibitory 37 degrees C. Incubation at 28 degrees C but not at 26 degrees C strongly inhibited extracellular assembly of the major T-pilus component VirB2 as well as of pilus-associated protein VirB5, and the highest amounts of T pili were detected at 20 degrees C. Analysis of temperature effects on the cell-bound virulence machinery revealed three classes of virulence proteins. Whereas class I proteins (VirB2, VirB7, VirB9, and VirB10) were readily detected at 28 degrees C, class II proteins (VirB1, VirB4, VirB5, VirB6, VirB8, VirB11, VirD2, and VirE2) were only detected after cell growth below 26 degrees C. Significant levels of class III proteins (VirB3 and VirD4) were only detected at 20 degrees C and not at higher temperatures. Shift of virulence-induced agrobacteria from 20 to 28 or 37 degrees C had no immediate effect on cell-bound T pili or on stability of most virulence proteins. However, the temperature shift caused a rapid decrease in the amount of cell-bound VirB3 and VirD4, and VirB4 and VirB11 levels decreased next. To assess whether destabilization of virulence proteins constitutes a general phenomenon, levels of virulence proteins and of extracellular T pili were monitored in different A. tumefaciens and Agrobacterium vitis strains grown at 20 and 28 degrees C. Levels of many virulence proteins were strongly reduced at 28 degrees C compared to 20 degrees C, and T-pilus assembly did not occur in all strains except "temperature-resistant" Ach5 and Chry5. Virulence protein levels correlated well with bacterial virulence at elevated temperature, suggesting that degradation of a limited set of virulence proteins accounts for the temperature sensitivity of gene transfer to plants.

FIG. 1

Growth temperature affects T-pilus production. T pili were isolated from wild-type strain C58 after growth under non-virulence-inducing conditions (−AS) or virulence-inducing conditions (+AS) on AB minimal medium plates at the given temperatures by shearing of cells and high-speed centrifugation. Pilus fractions were subjected to SDS-PAGE followed by silver staining or Western blotting and detection with specific antisera as indicated. Molecular masses of reference proteins are indicated on the right.

FIG. 2

Effect of growth temperature on virulence protein accumulation in strain C58. Wild-type strain C58 was grown under non-virulence-inducing conditions (−AS) or virulence-inducing conditions (+AS) on AB minimal medium plates at the given temperatures, followed by SDS-PAGE, Western blotting and detection with specific antisera as indicated. (A) Class I virulence proteins detected in the cells at elevated temperature. (B) Class II virulence proteins detected in reduced amounts in the cells at elevated temperature. (C) Class III virulence proteins not detected in the cells at elevated temperatures. Molecular masses of reference proteins are indicated on the right.

FIG. 3

Temperature shift does not affect T-pilus stability. Wild-type strain C58 was grown under non-virulence-inducing conditions (−AS) or virulence-inducing conditions (+AS) on AB minimal medium plates at 20°C for 3 days, followed by further incubation at 20°C or a shift to 28 or 37°C for the times indicated. T pili were isolated at the given times, and pilus fractions were subjected to SDS-PAGE followed by Western blotting and detection with VirB2- and VirB5-specific antisera. Molecular masses of reference proteins are indicated on the right.

FIG. 4

Effect of temperature shift on virulence protein levels. Wild-type strain C58 was grown under non-virulence-inducing conditions (−AS) or virulence-inducing conditions (+AS) on AB minimal medium plates at 20°C for 3 days, followed by further incubation at 20°C or a shift to 28 or 37°C for the times indicated. Vir proteins were detected in cell lysates after SDS-PAGE and Western blotting with specific antisera as indicated. (A) Shift-resistant virulence proteins not affected or only modestly degraded after the temperature shift. (B) Shift-sensitive virulence proteins rapidly degraded in the cells after the temperature shift. Molecular masses of reference proteins are indicated on the right.

FIG. 5

Effect of temperature shift on virulence protein levels in different Agrobacterium species. Strains were grown on AB minimal medium plates under non-virulence-inducing conditions (−AS) or virulence-inducing conditions (+AS) at the given temperatures followed by SDS-PAGE, Western blotting, and detection with specific antisera as indicated. (A) Comparison of virulence proteins in lysates from A. tumefaciens cells grown at 20 versus 28°C. (B) Virulence proteins from A. vitis. Arrowheads indicate virulence proteins detected in equal amounts at 28 and 20°C in strains Ach5 and Chry5. Molecular masses of reference proteins are shown on the right.

FIG. 6

T-pilus formation of temperature-resistant strains. Strains were grown under non-virulence-inducing conditions (−AS) or virulence-inducing conditions (+AS) on AB minimal medium plates for 3 days at 20 or 28°C. Proteins from cell lysates and extracellular T-pilus fractions were subjected to SDS-PAGE followed by Western blotting and detection with VirB2- and VirB5-specific antisera. Arrowheads indicate VirB2 and VirB5 in pilus fractions at 28°C. Molecular masses of reference proteins are shown on the right.

FIG. 7

Infection temperature determines tumor formation efficiency. Cells were diluted to the optical densities indicated, followed by infection of wounded K. diagremontiana at 20 or 28°C. (A) Tumor formation was monitored for several weeks in four independent experiments, and representative results are shown (+++, strong tumor formation; ++, intermediate tumor formation; +, weak tumor formation; −, no tumors). (B) Tumor formation after infection with different dilutions of strain C58. (C) Tumor formation after infection at 28°C with the lowest infectious dose.