at 254 nanometers for 15 minutes) reduced ice nucleation activity by. Pseudomonas viridiflava strain W-1 as determined by freezing drops of the bacterial ...
Plant Physiol. (1986) 80, 956-960 0032-0889/86/80/0956/05/$0 1.00/0
The Effects of Streptomycin, Desiccation, and UV Radiation on Ice Nucleation by Pseudomonas viridiftava Received for publication August 15, 1985 and in revised form November 28, 1985
JEFFREY A. ANDERSON AND EDWARD N. ASHWORTH*
Appalachian Fruit Research Station, Kearneysville, West Virginia 25430 ABSTRACI
Streptomycin (100 micrograms per milliliter), desiccation (over CaSO4), and ultraviolet rdiation (4500 microwatts per square centimeter at 254 nanometers for 15 minutes) reduced ice nucleation activity by Pseudomonas viridiflava strain W-1 as determined by freezing drops of the bacterial suspensions. Highest residual ice nucleation activity by dead cells was obtained by desiccation, although no freezing above -3.5C was detected. The rate and extent of loss of ice nucleation activity following streptomycin and ultraviolet treatment was affected by preconditioning temperature. At 21°C and above, loss of activity by dead cells was rapid and irreversible.
Many plants do not tolerate freezing and must remain supercooled to survive temperatures below 0°C. Therefore, the ice nucleating agents responsible for initiating freezing in plant tissues are of primary importance. Levitt (8) concluded that supercooling was probably limited to 1 to 3C in nature due to an abundance of ice nuclei. Schnell and Vali (19) reported a naturally occurring ice nucleus associated with decomposing vegetation that was later identified as an epiphytic bacterium (4, 17). The bacterium, Pseudomonas syringae, is widely distributed, both geographically and in host range (7, 12). Subsequent studies have revealed ice nucleation activity
by Pseudomonas viridiflava (18), Pseudomonasfluorescens (16), and 'Erwinia herbicola (13). Frost damage to tomato (1, 2, 10), soybean (2), and corn seedlings (10, 13, 18) was increased by inoculation with ice nucleation active (INA) bacteria. Similarly, leaves and leaf disks from diverse plant taxa froze at higher temperatures when INA bacteria were present (6, 14). These findings have led to the hypothesis that INA bacteria limit supercooling of plants (13, 14, 20). Several strategies for frost control based on reduction of populations of INA bacteria have been proposed (9-1 1). However, researchers have observed plant freezing temperatures warmer than -3C without detectable populations of INA bacteria (1, 3, 5, 6). Either plant material contained nonbacterial ice nuclei active above -3°C or nondetectable levels of INA bacteria were responsible for ice nucleation. It has been suggested that dead bacteria may serve as ice nuclei (9-1 1). Nonviable cells would have been undetectable since methods of detection rely on multiplication of individual cells to macroscopic colonies. The objective of this study was to determine whether dead INA bacterial cells are effective ice nucleators. Bacterial suspensions were treated with streptomycin, UV light, or exposed to a desiccation stress. Streptomycin, used singly or in combination with other antibiotics, has probably been the most extensively used agricultural bactericide. Desiccation and UV radiation treat-
ments were chosen because they are stresses encountered by bacteria in nature. The observation that the freezing efficiency of stationary phase bacterial suspensions could be affected by a preconditioning temperature treatment (2) was exploited to determine whether declines in freezing efficiency of dead cells could be reversed by incubation at an inductive temperature. MATERIALS AND METHODS
Growth of Bacteria. Pseudomonas viridiflava strains W- 1 (from an unidentified weed in a Florida peach orchard), 50, 51, 62, and 66 (all from Florida citrus leaves, kindly supplied by R. E. Stall) were grown for 24 h in a medium containing 2.5% dextrose, 1% Bacto-peptone, and 0.1% casamino acids. Flasks were held on a reciprocal shaker at 125 rpm in incubators at 21, 27, or 33°C. Cultures in the late log phase of growth were centrifuged at 2000g for 10 min at room temperature. Pellets were resuspended in sterile, distilled H20 and adjusted to 0.3 A at 600 nm (about 4 x 108 cells/ml). All experiments were conducted with strain W-l unless specified otherwise. Effect of Streptomycin. An equal volume of 200 ,ug/ml streptomycin sulfate (Sigma, 750 units/mg) in sterile, distilled H20 was added to the bacterial suspension, resulting in a final concentration of 100 ,ug/ml. Bacterial suspensions were held at room temperature for 15 min following addition of the streptomycin solution to allow uptake. Replicates containing only water or streptomycin were included as controls. The median freezing temperature (TM) of 15 10-,ul drops per treatment was determined with a thermoelectric plate as previously described (2). Freezing temperatures were recorded after 0, 3, 6, and 9 h incubation of the resuspended cells at each growth temperature. In addition, replicates were placed in a bath at the inductive temperature of 4°C at 0, 3, and 6 h to be tested after 3 h intervals. Exposure to 4°C was found to optimize ice nucleation activity by P. syringae (2) and P. viridiflava (J. Anderson, unpublished data). Inductive temperature treatments were included to determine whether declines in ice nucleation activity were reversible for both dead and live cells. In another experiment, suspensions of P. viridiflava strains W1, 50, 51, 62, and 66 which had been grown at 21°C were first held at 33°C for 3 h, then shifted to 4°C for an additional 3 h. Median freezing temperatures of suspensions with and without streptomycin were determined initially and after each temperature treatment. Longer term effects of streptomycin on freezing temperature were also examined. Bacteria grown at 21 °C were resuspended in sterile, distilled H20. Median freezing temperatures of bacteria, bacteria plus 100 ;g/ml streptomycin, and controls were determined periodically during 30 d incubation at about 4°C. This experiment was designed to determine how long streptomycin-killed cells retained ice nucleation activity under optimum conditions, i.e. at an inductive temperature. Effect of Desiccation. Six mm diameter disks of filter paper
956
BACTERIAL ICE NUCLEATION (Whatman No. 1) were immersed in a 108 cells/ml suspension of P. viridiflava strain W- 1 or placed in a sealed beaker as controls. After 20 min, half of the bacteria-soaked disks were placed on a wire screen to air dry at about 4°C. After 4 h, a subset of control, dried, and saturated disks were assayed for ice nucleation activity in 60 AI drops of sterile, distilled H20 on the thermoelectric cooling plate. The remainder of the disks were placed in a cold room at 5C and assayed periodically over 41 d for ice nucleation activity. Dried disks were stored in a desiccator over CaSO4, while saturated disks were kept in the bacterial suspension. Effect of UV Radiation. P. viridiflava strain W-1 grown at 21 C was used to determine the effect of a lethal dose of UV radiation on ice nucleation activity. Twenty-five ml samples of a 108 cells/ml aqueous suspension of bacteria were pipetted into sterile Petri plates. Plates, either covered with aluminum foil as a control, or with the lid removed were exposed to a monochromatic (254 nm) UV source for 15 min. A radiation level of 4500 .W/cm2 at 2.5 cm was administered with a mercury-argon lamp (Fisher Scientific, Springfield, NJ). Petri plates were placed on an aluminum block at 4°C during UV exposure to eliminate temperature induced declines in ice nucleation activity. Freezing temperatures of 10 Al drops were determined as above. Additional replicates were tested after 3 h at 33°C, followed by 3 h at 40C. In streptomycin and UV experiments, viable cell counts were determined by dilution plating on Pseudomonas agar F (Difco Laboratories, Detroit, MI). Viable cells were detected following desiccation by placing disks into sterile medium and incubating at 21°C for up to 1 week. Turbid tubes were freeze-tested to rule out contamination.
RESULTS The freezing temperature of stationary phase bacterial suspensions was strongly dependent on incubation temperature. As incubation temperature increased, median freezing temperature decreased. Live cells grown and held at 21°C nucleated ice at about -3C (Fig. 1). Incubation at 40C produced little or no increase in freezing temperature. The median ice nucleation temperature of cells grown at 270C was -7.6 initially; however, continued incubation of the resuspended cells at 27°C resulted in ice nucleation between -4 and -50C. Incubation at 40C raised freezing temperatures to about -3.5C. Bacteria grown at 33°C nucleated ice at -8.8°C. Further incubation of these cells at 33°C resulted in freezing between -8.0 and -8.4°C, while storage at 4°C for 3 h raised freezing temperatures 4 to 50C relative to storage at 330C. Effect of Streptomycin. Bacteria grown and held at 2 1C with streptomycin added nucleated ice at progressively lower temperatures over 9 h (Fig. 1). The rate of decrease varied between repetitions of the experiment. Transferring replicates to an inductive temperature (4°C) did not restore nucleation activity, but prevented a further decrease in freezing temperature. The ice nucleation temperature of bacteria grown and held at 27°C decreased from -8 to -1 0°C over 9 h after streptomycin was added. Incubation at 4°C raised freezing temperatures initially, but had no effect after 3 h. Cells grown at 330C nucleated ice at -9.5°C just after streptomycin was added. Median freezing temperatures were not reached by -1 5°C in two-thirds of the measurements following further incubation at 33°C. Temperature-induced declines in freezing temperature were reversible in the absence of streptomycin. Median ice nucleation temperatures of cells grown at 2 1C, then held at 33°C dropped from -3 to -80C (Fig. 2). The median ice nucleation temperature of controls transferred to an inductive temperature rose to -3°C, but remained at -8.5°C in streptomycin-treated cells. This behavior was not unique to a single strain since strains 50, 51, 62,
957
n
LU ry D
LLJ FLUJ CD
z N LU
LU LL if
z K
LUJ 3
6
HOURS
9
FIG. 1. Effect of streptomycin on ice nucleation activity of P. viridiflava strain W- 1. Twenty-four h cultures were pelletted and resuspended in sterile H20. Ten ul drops of 108 cells/ml suspensions were cooled at 20°C/h on a thermoelectric cooling block. Median freezing temperatures of 15 drops per treatment were determined after 0, 3, 6, and 9 h incubation at the growth temperature following addition of 100Ig/mI streptomycin or water. Additionally, replicates were placed at 4'C (- - -) for 3 h intervals. Values reported are means ± SE of three repetitions of the experiment.
and 66 responded similarly. However, the first three strains did not drop to as low a temperature as strains 66 and W-l following incubation at 33C (data not presented). Viable cell counts of controls did not change during 9 h incubation at 2 1C. The numbers of living cells were reduced by 50 and 60% following 9 h incubation at 27 and 33C, respectively.
The numbers of viable cells in streptomycin treatments ranged from nondetectable to 0.0004% of controls initially, and from nondetectable to 0.0002% of controls after 9 h.
958
ANDERSON AND ASHWORTH
zU ~-33 cN
',-"
4*CCONTROL,
-4
Plant Physiol. Vol. 80, 1986
.
-2
C
, ,o
SATURATED
L,2 LX 3 0y vi. e
--i----------
LL
L--6
lIy
