Distribution of Soybean Lectin in Tissues of Glycine Max (l.) Merr.1

Three different assay procedures have been used to quantitate the levels of soybean (Glycine max IL.l Meff.) lectin in various tissues of soybean plants. The assays used were a standard hemagglutination assay, a radioimmunoassay, and an isotope dilution assay. Most of the lectin in seeds was found In the cotyledons, but lectin was also detected in the embryo axis and the seed coat. Soybean lectin was present in all of the tissues of young seedlngs, but decreased as the plants matured and was not detectable in plants older than 2 to 3 weeks. Soybean lectin isolated from seeds of several soybean varieties were identical when compared by several methods. Symbiosis between bacteria of the genus Rhizobium and legume plants involves nodulation of the plant roots and biological fixation of atmospheric N2. The host range for each species of Rhizobium is usually narrow, indicating that the Rhizobium-legume root symbiosis is a specific relationship. For example, Rhizobium japonicum, the soybean symbiont, does not infect green bean (Phaseolus) or clover ( Trifolium), and, conversely, the green bean and clover symbionts do not infect soybean. Several authors have suggested that the plant proteins known as lectins (phytohemmagglutinins) may be involved in the specific recognition of symbiotic rhizobia by legumes (2, 3, 10, 12, 30). Lectins have the capacity to bind to, or recognize, specific carbohydrate structures, and various species of legumes produce lectins with different binding specificities (18). Complex polysaccharides (e.g. lipopolysaccharides, exopolysaccharides) are present on the surfaces ofGram-negative bacteria, including Rhizobium. In many cases, the structures of such polysaccharides are known to be strainor species-specific (21). Thus, the recognition ofa symbiotic Rhizobium by a legume host may involve selective binding of the characteristic host plant lectin to a distinctive carbohydrate structure on the surface of the symbiont. Bohlool and Schmidt (3) found that SBL5 bound to 22 of 25 strains of the soybean symbiont, R. japonicum, but not to any strains ofother rhizobial species. These results have been generally confirmed and extended (2). A similar pattern of lectin binding to 'Supported in part by National Science Foundation Grants BMS 7517710 and BMS 76-00913. 2Present address: Department of Biology, University of Missouri, St. Louis, 8001 Natural Bridge Road, St. Louis, Missouri 63121. 3To whom reprint requests should be sent. 4Present address: Department of Botany, University of Wisconsin, Madison, Wisconsin 53706. 'Abbreviations: SBL: soybean lectin; PBS: phosphate-buffered saline. symbiotic rhizobia, but not to nonsymbionts, has been observed for a lectin from white clover (Trifolium repens) (10). However, concanavalin A, a lectin from jackbean (Canavalia ensiformis), was reported to bind to all strains of rhizobia tested, irrespective of their ability to infect and nodulate jackbean (9). If lectins are involved in recognition of rhizobial symbionts, the lectins must be present in or on the roots of the host plant. Yet there is little direct evidence that lectins are indeed present in these locations. The most common source of legume lectins is seeds (4, 29), although hemagglutinating activity, presumably due to lectins, has been detected in roots of soybean (3), and various tissues of several other legume species (6, 16). Howard et al. (14) and Rouge (26) dissected lentil (Lens culinaris) and pea (Pisum sativum) seeds, respectively, into seed coats, embryo axis, and cotyledons. High hemagglutinating activity was found in the cotyledon and embryo axis, but very little was detected in seed coats. Hemagglutination and immunodiffusion techniques have detected low levels of the seed agglutinins in roots, stems, and leaves ofyoung lentil and pea seedlings, but not in nonseed tissues of older plants (14, 25, 26). Dazzo and Brill (8) have recently reported that Rhizobium trifolii-binding lectin can be eluted from the surface of intact white clover roots with a particular sugar hapten. This lectin appears to be localized at or near the tips of the root hairs. As part of our investigation of the involvement of lectins in plant-microorganism interactions, we have developed two techniques to measure specific lectins in various plant extracts quantitatively. We now report on the use of these techniques to determine the quantity of soybean seed lectin in various tissues of soybean seedlings of different ages. MATERIALS AND METHODS Preparation of Flour from Whole Seeds and Seed Tissues. Flours were prepared from whole seeds of Glycine max (L.) Merr. vars. Acme, Beeson, and Wayne. Seed samples were ground to 40 mesh with a Wiley mill and defatted with petroleum ether. Commercial soybean flour (Soya Fluff200W, Central Soya Chemurgy, Chicago, Ill.) was defatted as above, and routinely used as a source of SBL. To procure material for analyzing the distribution of SBL in soybean seeds (var. Beeson), seeds were soaked for 5 min in warm water. The swollen seed coats were then removed, and the embryos were dissected into the cotyledons and embryo axis. The pooled samples of embryo axis, of seed coats, and of cotyledons were dried in a forced air oven at 70 C. This drying procedure does not measurably inactivate SBL. Cotyledons and seed coats were ground to 40 mesh and defatted as above. Embryo samples, because of their small size, were defatted directly, and not passed through the Wiley mill. Defatted samples of embryo axis tissue 779 www.plantphysiol.org on July 22, 2017 Published by Downloaded from Copyright © 1978 American Society of Plant Biologists. All rights reserved. Plant Physiol. Vol. 61, 1978 were pulverized with a mortar and pestle just before analysis. The contribution of each seed tissue to seed weight was estimated by separating two lots of 50 seeds each into seed coats, cotyledons, and embryo axes. The tissues were dried at 70 C for 96 hr, and dry weights were recorded. Growth of Soybean Seedlings. For isotope dilution assays, undamaged soybean seeds (var. Beeson) were planted in pans (10 x 12 x 2 cm deep) containing Vermiculite, irrigated with Hoagland (13) nutrient solution, and maintained in a growth chamber at 25 C and about 7,500 lux from fluorescent and incandescent light sources (16-hr photoperiod). Plants assayed for lectin by hemagglutination and radioimmunoassays were grown under similar conditions except the temperature was maintained at 32 C and 27 C during the day and night cycles, respectively. Under these conditions the seedlings emerged on the 3rd day after planting. The Vermiculite was kept moist by addition of distilled H20 as needed after planting. Healthy seedlings were harvested on various days after planting by carefully removing their roots from the Vermiculite and washing them in a gentle stream of water. Roots, stems, and cotyledons were then excised, blotted dry, and weighed immediately. Tissue samples were stored frozen until used. Purification of SBL. SBL was purified by affinity chromatography using the procedure described by Allen and Neuberger (1). The method of Lowry et al. (20) was used to assay for protein content with BSA (Sigma) as a standard. The homogeneity of SBL preparations was determined by electrophoresis using 7.5% polyacrylamide disc gels (0.5 x 7.5 cm) and the acidic buffer system of Reisfeld et al. (23). Gels were electrophoresed at 3 mamp/gel until the tracking dye was less than I cm from the end of the tubes. Staining was done with Coomassie brilliant blue R250. Radloabeing of SBL. Tritium-labeled SBL for isotope dilution experiments was prepared by oxidation with sodium periodate followed by reduction with potassium 3H-borohydride (24.4 mCi/mmol, Schwarz/Mann) according to the procedure of Lotan et al. (19). 3H-SBL was purified by gel filtration chromatography on a column of Bio-Gel A-0.5m (2.5 x 30 cm) with 0.19% NaCl as eluant and then by affinity chromatography as described above. Recovery of unaggregated 3H-SBL was 82%. Samples were counted with a Nuclear-Chicago liquid scintillation counter using Aquasol 2 (New England Nuclear) as a scintillation fluid. Counting efficiency was about 43%, and specific radioactivity of the 3H-SBL was 1.32 x 107 cpm/mg. Prior to use in isotope dilution experiments, the 3H-SBL was diluted with unlabeled SBL to a more appropriate specific radioactivity. Iodine-labeled SBL (125I-SBL) for radioimmunoassays was prepared using the chloramine-T procedure as described by Sela et al. (27). The iodinated SBL was purified twice by affinity chromatography before use in the radioimmunoassay. The specific radioactivity was 2 x 108 cpm/mg. Quantitation of SBL in Plant Materials by Isotope Dilution. The amounts of SBL in root and cotyledon samples from soybean seedlings were determined as follows. The tissue sample was added to PBS (0.43 g ofKH2PO4, 1.48 g ofNa2HPO4, 7.20 g ofNaCl/liter, pH 7.2) containing 0.05 M ascorbate in a Waring Blendor. 3H-SBL (45-90 ,jg, specific radioactivity 2.2 x 106 cpm/mg) was introduced, and several drops of octanol were added to retard foaming. The sample was ground to a fine suspension and centrifuged at 9,200g for 15 min. The supernatant solution was filtered through glass fiber paper and stirred at 4 C with 9 ml of affinity beads. After 2 hr, the affinity beads were collected in a sintered glass funnel, rinsed carefully with PBS, and poured into a column (0.9 x 15 cm). The column was eluted with PBS, and the eluant was monitored by A at 280 nm (Ameo). When the Am.0 stabilized at a base-line value, elution with 100 mm galactose or 1 mm N-acetylD-galactosamine in PBS was initiated. The eluted SBL was monitored as an Amso peak and was collected as a single fraction. The specific radioactivity of the recovered SBL from each sample was then measured by protein determination and scintillation counting, and an aliquot of each sample was electrophoresed to monitor the identity and homogeneity ofthe purified protein. The recovery of rmdiolabel varied from 55 to 100%. Flours from seeds (50 g/sample) and seed tissues (2 g/sample) were suspended in PBS (20 ml/g) and 3H-SBL was added. After stirring for 1 hr at room temperature, the suspensions were centrifuged and analyzed as for the tissue sample extracts above. The extracts from 2 g of flour were incubated with 9 ml of affinity beads, whereas extracts from 50 g of flour were incubated with 50 ml of beads. Quantitation of SBL Levels Using Radloi unoassay. Antiserum against SBL was prepared in rabbits by intramuscular injections of purified SBL in complete Freund's adjuvant. Antiserum against rabbit IgG was prepared in a goat by injection of purified rabbit IgG in complete Freund's adjuvant. Several series of experiments were carried out to determine the appropriate dilutions of 125I-SBL and rabbit antiserum to be used in the assay, as well as the optimum time for each incubation. The 125I-SBL was diluted in PBS containing 0.25% BSA to a final concentration of40,000 cpm/ml. The 50-SOl aliquot of this solution used in the assay contained approximately 20 ng of SBL. The same batch of rabbit antiserum was used for all of the experiments described here. This antiserum was diluted in PBS such that the aliquot of antiserum used would precipitate approximately 70 to 90% of the '25I-SBL. This was typically a dilution of 35,000-fold. A typical radioimmunoassay consisted of 250 yIA of properly diluted rabbit anti-SBL serum, 50 ,tl of 125I-SBL, and 100 ,ul of a sample containing 0.1 to 1.0 ,ug of SBL. This mixture was incubated for 1 hr at 4 C, followed by the addition of20 IL of undiluted goat anti-rabbit IgG. This mixture was incubated 24 hr at 5 C. The immunoprecipitate was collected by centrifugation for 20 min at 3000 rpm in a nonrefrigerated table top centrifuge. The amount of 1251-SBL in the supernatant solution and the precipitate was determined with a Nuclear-Chicago gamma counter. The percentage of `2I-SBL in the immunoprecipitate was calculated from these numbers. Each sample was assayed in duplicate. Weighed tissue samples to be extracted for radioimmunoassays were cooled to dry ice temperatures and ground into a fine powder in a mortar kept on dry ice. This powder was extracted for 1 to 2 min with 10 volumes of PBS containing 0.5% galactose. The resulting mixture was filtered through a glass fiber filter. Phenylmethane sulfonyl fluoride was added to each filtrate to a final concentration of I mg/ml in order to inhibit protease activity. The filtrate was stored frozen until it was assayed for protein content and SBL content. For every sample, a control for nonspecific binding was performed using PBS instead of the rabbit antiserum. The percentage of `2I-SBL bound in this control (less than 5% if freshly prepared 125I-SBL was used) was subtracted from the experimental sample. A standard curve was prepared for each experiment using increasing dilutions of purified SBL. One such standard curve is shown in Figure 1. The standard curve from each experiment was used to convert the value for per cent "2I bound for an unknown sample into ,ug of SBL/ml. Quantitation of SBL Levels Using Hemagglutinatlon. Hemagglutination assays were carried out with trypsinized rabbit red blood cells (Gibco) using the microtiter plate assay (1 1). In some cases hemagglutination was detected by observation of the red blood cells under a microscope. This assay gave the same results as the microtiter plate assay but was less sensitive and less convenient. With every series of unknown samples, a standard curve was also determined. This allowed a conversion from hemagglutination titer (the reciprocal of the last dilution which gave hemagglutination) to the concentration of lectin. This assay is very sensitive and can detect levels of lectin as low as 10 ng/ml. There is, however, difficulty in reproducibly determining the titer of a 780 PUEPPKE ET AL. www.plantphysiol.org on July 22, 2017 Published by Downloaded from Copyright © 1978 American Society of Plant Biologists. All rights reserved. DISTRIBUTION OF SOYBEAN LECTIN