Compatible and Incompatible Rhizobia Alter Membrane

MATERIALS AND METHODS Inoculation with Rhizobium japonicum or R. meliloti reduced the electrical transmembrane potential (Em) of soybean (Glycine max [L.] Merr.) root cells within I day. The response could be attributed to altered diffusion potential (ED). Em values return to control levels by the second day after inoculation, but again were reduced in R. meliloti-inoculated tissue on the seventh day. Increased concentrations of sodium phosphate in the perfusion solution magnified the effects of inoculation on EmNeither heat-killed rhizobia nor living cells of Pseudomonas fluorescens elicited the response. The Em and ED of nodule cells were nearly 20% lower than corresponding values from adjacent cortical cells of the root. Establishment of the legume-Rhizobium symbiosis proceeds through a sequence of developmental stages that begins in the rhizosphere and culminates with the formation of an effective N2-fixing nodule (1, 7, 19). Cardinal steps in this process include interaction ofthe bacteria with the root surface, root hair curling, infection thread formation, cortical cell divisions, and ultimately, appearance of the nodule. Although intensively studied, the nodulation process is not well understood in any legume. Even less is known about the interactions of rhizobia with root cells in incompatible, nonnodulating combinations. Incompatible rhizobia adsorb to soybean roots, for example (18, 28), but markedly curled root hairs are absent, and infection threads do not form (17, 19). Factors that condition incompatibility thus can be expressed early, prior to infection. Many of the initial cellular responses of plants to infection by parasitic bacteria are confined to the plasmalemma (14). The Em' of the plasmalemma, which has both a passive and an active, energy-dependent component (8, 23), is altered at infection sites (16, 27). Such changes in Em are sensitive indicators of the status of membranes and can readily be monitored by electrophysiological techniques employing glass microelectrodes. Here we present the results of experiments designed to detect responses of soybean root cell membranes to nodulating and nonnodulating rhizobia. The overall objectives were (a) to test the hypothesis that Em can be used as a sensitive indicator of plant response to nonnodulating rhizobia and (b) to compare the effects of nodulating and nonnodulating rhizobia on root cell membranes of soybean. ' Supported by funds from the Food for the 21st Century Program, University of Missouri. Permanent address: Plant Protection Institute, Hungarian Academy of Science, P. 0. Box 102, H1525, Budapest. 2 Abbreviations: Em, electrical transmembrane potential; Er., potential due to the electrogenic pump; ED, diffusion potential. Bacteria. Rhizobium meliloti 102F51 was from C. P. Vance, USDA-ARS, University of Minnesota, and R. japonicum USDA 74 was from H. H. Keyser, USDA-ARS, Beltsville, Maryland. The bacteria were maintained at 4°C on yeast extract-mannitol slants (29). Liquid cultures were initiated by transferring cells into 125 ml Erlenmeyer flasks containing 25 ml of gluconatemannitol medium (4). Cultures were incubated at room temperature on a rotary shaker (125 rpm). Exponential phase bacteria were harvested by centrifugation at 7,700g for 10 min, resuspended in 125 mm PBS (4), and turbidimetrically adjusted to a concentration of 5 x 108 cells/ml. Suspensions for dual inoculation experiments were adjusted to contain 5 x 108 cells of each Rhizobium strain per ml. Pseudomonas fluorescens, a soil saprophyte that was used as a control, was grown in nutrient-yeastglucose broth at 25°C as described above. Heat-killed bacteria controls were prepared by boiling cells for 10 min immediately before inoculation. Preparation of Seedlings. Soybean (Glycine max [L.] Merr. cv McCall) seeds were from D. Whited, North Dakota State Univ., Fargo. Seeds were surface sterilized by treatment for 5 min in aqueous 50% ethanol, followed by 5 min in 2.6% aqueous NaOCl. After thorough rinses in deionized H20 the seeds were placed onto water agar and germinated at 25°C in the dark for 40 h. The radicle of each seedling was aseptically dipped into a freshly prepared suspension of rhizobia. Controls included seedlings dipped into PBS, suspensions of heat-killed rhizobia, or suspensions of live P. fluorescens. Groups of four seedlings were placed into autoclaved plastic growth pouches (Northrup King Co., Minneapolis, MN), each of which contained 15 ml of nitrogen-free Jensen's nutrient solution (29). The position of the primary root tip of each seedling was marked on the pouch at the time of inoculation and designated the root tip mark. Plants were grown at room temperature under a 12-h photoperiod (about 500 ,E/m2 s, photosynthetically active radiation). Seedlings were moistened with glass-distilled H20 as needed. Electrophysiological Measurements. Samples were taken at selected times by slicing across the primary root 1.5 cm on each side of the root tip mark as shown in Figure 1. The resulting 3cm segment was mounted horizontally on a Plexiglas holder and washed for 3 h in an aerated solution of mineral salts (1 mM KCI, 1 mm Ca(NO3)2, 0.25 mM MgSO4, and 0.95 mm sodium phosphate buffer, final pH 5.7) (9). This treatment is necessary to equilibrate the cells in the perfusion solution and to minimize the effects of excision. In some experiments, the concentration of the buffer was increased to 66 mm. Micropipettes were pulled with a vertical pipette puller (D. Kopf, Tujunga, CA) from glass fiber microcapillaries (WP Instruments, New Haven, CT) as described previously (15). Electrode tip potentials and resistances were -2 to -15 mV and 6 to 15 mg, respectively. Micropipettes were filled with 3 M KCI, and 2-cm segments of plastic tubing filled with 3 M KCI in 2% agar 1115 www.plant.org on April 30, 2016 Published by www.plantphysiol.org Downloaded from Copyright © 1986 American Society of Plant Biologists. All rights reserved.