Planta (1984)161:302-307
P l a n t a 9 Springer-Verlag 1984
Determination of pea (Pisum sativum L.) root lectin using an enzyme-linked immunoassay* C.L. Diaz 1, P. Lems-van Kan 2, I.A.M. Van der SchaaP and J.W. Kijne 1 1 Department of Plant Molecular Biology,Research Group of Biological Nitrogen Fixation, Leiden State University, Nonnesteeg 3, NL-2311VJ Leiden, and z Department of Gastroenterology, Leiden University Hospital, Rijnsburgerweg 10, NL-2333 AA Leiden, The Netherlands
Abstract. R o o t lectins are believed to participate in the recognition between Rhizobium and its leguminous host plant. Among other factors, testing this hypothesis is difficult because of the very low amounts in which root lectins are produced. A double-antibody-sandwich enzyme-linked immunoassay, was used to determine nanogram quantities of pea lectin in root slime and salt extracts of root cell-wall material when pea seedlings were 4 and 7 d old. In addition, a critical N O ~ concentration (20 raM) which inhibited nodulation was found, and the lectin present in root slime and salt extracts of root cell walls of 4- and 7-d-old peas supplied with 20 m M N O 3 was comparatively determined. With the enzyme-linked immunoassay, lectin quantities ranging between 20 and 100 nanograms could be determined. The assay is not affected by monomeric mannose and glucose (pealectin haptens). The slime of the 4-d-old roots contained more lectin than the slime of the 7-d-old roots. Salt-extractable, cell-wall-associated lectin accumulated in the older roots. Nitrate affected slime and cell-wall production, and the extractability of cell-wall material in both age groups. The presence of N O 3 increased lectin in the slime, most notably in the younger roots; the relative amount of lectin in the slime was almost doubled. The cellwall-associated, salt-extractable lectin decreased two- to threefold compared with the control group. Key words: Age (pea root) - Lectin (pea root) Nitrate (nodulation) - Nodule - Pisum (root lectin) R o o t lectin. * Dedicated to Professor A. Quispel on the occasion of his retirement Abbreviations: ELISA=enzyme-linked immunoassay; PTN0.01 M phosphate buffer (pH 7.4), containing 0.15 M NaCI, 0.05% Tween-20 and 0.02% NaN 3
Introduction The lectin-recognition theory, which provides an explanation for the specific attachment of Rhizobium to legume root cells, implies the interaction of a plant multivalent sugar-binding (glyco)protein with the carbohydrate sequence of a bacterial surface receptor (Bohlool and Schmidt 1974; Dazzo and Hubbell 1975). The general applicability of this theory is still a matter of discussion (for recent reviews, see Kijne et al. 1983; Quispel et al. 1984). Verification of the theory is, among other factors, dependent on the demonstration of lectins on the cell surface of susceptible legume roots. Most of the lectin-binding studies with Rhizobium and rhizobial receptors have been carried out with the use of seed lectins. R o o t lectins are difficult to extract and characterize, partly because they seem to be present in very small quantities. The root lectin of Pisum sativum has been isolated with the use of affinity chromatography and several authors have partially analyzed it with semi-quantitative methods (Gatehouse and Boulter 1980; Hosselet et al. 1983; Kato et al. 1981; Kijne et al. 1980). Pea root lecfin has been reported to be similar to pea seed lectin, both structurally and immunologically (Hosselet et al. 1983). On the basis of its electrophorefic behaviour, root-slime lectin is identical to pea seed isolectin2 (Kijne et al. 1980; Kijne et al. 1983). With one exception (Gatehouse and Boulter 1980), its sugar-hapten specificity has been found to be similar to pea seed lectin. Knowledge about the regulation of synthesis and distribution of legume lectins during root development could contribute to a better understanding of the role of lectins in plant metabolism and rhizobial recognition. Such a study requires a very sensitive, quantitative assay for lectins. We have
C.L. Diaz et al. : Determination of pea root lectin
applied the double-antibody-sandwich enzymelinked immunosorbent assay (ELISA; see Engvall and Perlmann 1972; Voller et al. 1976; Clark and Adams 1977) for the detection of pea root lectin in nanogram quantities. The method is specific, reproducible, and allows the determination of lectins in crude plant fractions (Borrebaek and Mattiasson 1983). We applied the method to test the hypothesis that the well-known inhibition of rhizobial legume infection by critical concentrations of NO~ is correlated with a decrease in detectable lectin concentrations on the root surface (Dazzo and Brill 1978; Dazzo and Hrabak 1981). Material and methods Pea seed lectin. Pea seed, Pisum sativum L. cv. Finale, was purchased from Cebeco, Rotterdam, The Netherlands. Pea seed lectin was isolated from dry pea seed meal or from 16-h-imbided seeds by affinity chromatography using Sephadex G-75 (Pharmacia, Upsala, Sweden). Chromatofocusing of these lectins yielded pea isolectins 1, 2 and 3, as described by Van der Schaal 1983. Immunization procedure. A solution of 1 mg isolectin 2 in 2 ml sterile water was mixed with 1 ml complete Freund's adjuvant. The mixture was injected subcutaneously in small doses in the dorsal and ventral regions of an anaesthetized rabbit. Boosting was done by injecting 0.5 mg isolectin 2 intravenously. Bleeding after 10 d showed a high antibody titer by double diffusion in agar. The boosting and bleeding (50 ml) series were successfully repeated every six weeks over a period of 18 months. Specific antibody isolation. The immunoglobulin fractions were precipitated according to Hijmans et al. (1969). Specific antiisolectin 2 was isolated as described by Van Driessche et al. (1982). Isolectin 2 was bound to CNBr-activated Sepharose 4B (Pharmacia).
303 ranged from 20 to 100 ng per ml PTN, and 100 gl per well were pipetted. Root lectin from 7-d-old peas was used as an internaI standard. Plants were grown in Raggio and Raggio (1956) rooting medium. Root lectin was extracted as described by Kijne et al. (1980). After dialysis, the extract was taken to 60% (NH4)2SO 4 saturation and the precipitate was extensively dialyzed against water. The lyophilyzed preparation did not dissolve completely, but root lectin could be determined in the soluble portion. Appropiate dilutions of the internal standard and samples were pipetted (100 gl per well) and the plates were incubated for 3 h at 37 ~ C. Afterwards, the plates were emptied and the washing step was repeated. Addition of the enzyme-labeled antibody. Lectin-standard lines responded to the dilution of the enzyme-labeled antibody as shown in Fig. 1. A 5-gl enzyme-antibody conjugate per ml PTN was chosen to perform the assays. Each well received 100 gl of this conjugate dilution. The plates were incubated for 3 h at 37 ~ C and emptied. Then the washing step was carried out as described. The enzymatic reaction was performed in 0.05 M sodiumcarbonate buffer, pH 9.6, containing 0.02% MgC12 andp-nitrophenylphosphate (Sigma) in a I nag ml-1 concentration. The reaction was read after 1 h incubation (100 gl per well), at 37 ~ C, using a Titertek Multiscan Plate Reader (Flow Laboratories, McLean, Virginia, USA) at 405 nm. The effect of sugars in lectin determination with the ELISA was assayed. Lectin was incubated in PTN containing 10 or 100 mM glucose or mannose (known pea-lectin haptens), or galactose and compared with lectin incubated only in PTN. The complexing of immobilized antibodies and lectin present in crude seed extracts was assessed. For this experiment, the lectin contained in l0 g pea seeds was isolated, as described by Van Driessche et al. (1978). This procedure involves extensive extraction with 1 M NaC1, 20 m M MgC12 and 20 mM CaClz, followed by a precipitation step (60% (NH4)2SO 4 saturation), before affinity chromatography. The same quantity of seeds was similarly extracted and precipitated. The resulting pellet was dialyzed against water and lyophilyzed. The lectin present in this extract was determined with the ELISA. The total lectin in the crude pea seed extract was calculated and compared to the yield of the affinity-purified pea seed lectin.
Influence of NO~ in nodulation. To determine NO~ concentraPreparation of the enzyme-labeled antibody. Alkaline phosphatase (Sigma, Munich, F R G ; Type VII) was covalently bound to approx. 1.5 mg specific rabbit antiisolectin 2, following the procedure outlined by Lems-van Kan et al. (1983).
Immunoassay. Coating and fixing: approx. 4 gg specific antibody per ml sodium-bicarbonate buffer (0.1 M, pH 9.6) were used to coat Falcon 3912 Microtest Flexible Assay Plates (Becton Dickinson and Co., Oxnard, Cal., USA); 100 I11 of this solution were pipetted per well and incubated overnight at 4 ~ C. Afterwards, plates were emptied and shaken dry. The coating antibodies were fixed by adding 1% freshly prepared glutaraldehyde (Fluka, Buchs, Switzerland; electron microscopy grade) in 0.01 M phosphate buffer (pH 7.4), containing 0.15 M NaCI (PBS). Plates were incubated for 10 rain at 4 ~ C and then rapidly washed. Washing included three times pipetting 250 gl PTN (PBS containing 0.05% Tween-20 (Merck, Darmstadt, FRG) and 0.02% NAN3), each time incubating the plates for I min and shaking them dry. Addition of lectins. Sephadex-G-75-purified pea seed lectin and chromatofocused isolectins showed immunological identity in Ouchterlony tests (Van der Schaal 1983). Because of its better solubility, affinity-purified lectin (Van Driessche et al. 1978) was chosen to prepare standard lines. Lectin concentrations
tions inhibiting nodulation, surface-sterilized pea seeds were inoculated with Rhizobium leguminosarum RBL1. The seeds were planted in gravel soaked with Raggio and Raggio (1956) rooting medium provided with 0, 2, 5, 10, 20 and 50 m M NO~as K N O 3 and Ca(NO3) 2. Nodules were counted after 21 d growth at 20 ~ C and 70% relative humidity. The light flux at table surface was 20000 lx (Philips T F L 60W/33 fluorescent tubes; Philips, Eindhoven, The Netherlands) during the 16-h light period.
Root lectins. Pea seeds were surface-sterilized and soaked for 18 h in Raggio and Raggio (1956) rooting medium or in the same rooting medium supplied with a N O r concentration inhibiting nodulation (20 mM N O r ) . Swollen seeds were planted in gravel with or without NO3-, at 20 ~ C. Roots were sectioned after 4 or 7 d growth in darkness. Between 1200 1800 roots per treatment and age group were collected. Root slime was obtained by thoroughly washing roots with distilled water. The washings were concentrated with a rotary evaporator and lyophilyzed. The lectin in this preparation is referred to as rootslime lectin. Root cell walls were prepared according to Buchala and Franz (1974). Cell-wall preparations were lyophilyzed twice before lectin extraction. The root-lectin extracting buffer, used
304
C.L. Diaz et al. : Determination of pea root lectin
1.0-
0
[3
1.4-
(19 t2/i
0.8 E
1.0-
C t~
S
f3
~ 0.7,
P l a n t a 9 Springer-Verlag 1984
Determination of pea (Pisum sativum L.) root lectin using an enzyme-linked immunoassay* C.L. Diaz 1, P. Lems-van Kan 2, I.A.M. Van der SchaaP and J.W. Kijne 1 1 Department of Plant Molecular Biology,Research Group of Biological Nitrogen Fixation, Leiden State University, Nonnesteeg 3, NL-2311VJ Leiden, and z Department of Gastroenterology, Leiden University Hospital, Rijnsburgerweg 10, NL-2333 AA Leiden, The Netherlands
Abstract. R o o t lectins are believed to participate in the recognition between Rhizobium and its leguminous host plant. Among other factors, testing this hypothesis is difficult because of the very low amounts in which root lectins are produced. A double-antibody-sandwich enzyme-linked immunoassay, was used to determine nanogram quantities of pea lectin in root slime and salt extracts of root cell-wall material when pea seedlings were 4 and 7 d old. In addition, a critical N O ~ concentration (20 raM) which inhibited nodulation was found, and the lectin present in root slime and salt extracts of root cell walls of 4- and 7-d-old peas supplied with 20 m M N O 3 was comparatively determined. With the enzyme-linked immunoassay, lectin quantities ranging between 20 and 100 nanograms could be determined. The assay is not affected by monomeric mannose and glucose (pealectin haptens). The slime of the 4-d-old roots contained more lectin than the slime of the 7-d-old roots. Salt-extractable, cell-wall-associated lectin accumulated in the older roots. Nitrate affected slime and cell-wall production, and the extractability of cell-wall material in both age groups. The presence of N O 3 increased lectin in the slime, most notably in the younger roots; the relative amount of lectin in the slime was almost doubled. The cellwall-associated, salt-extractable lectin decreased two- to threefold compared with the control group. Key words: Age (pea root) - Lectin (pea root) Nitrate (nodulation) - Nodule - Pisum (root lectin) R o o t lectin. * Dedicated to Professor A. Quispel on the occasion of his retirement Abbreviations: ELISA=enzyme-linked immunoassay; PTN0.01 M phosphate buffer (pH 7.4), containing 0.15 M NaCI, 0.05% Tween-20 and 0.02% NaN 3
Introduction The lectin-recognition theory, which provides an explanation for the specific attachment of Rhizobium to legume root cells, implies the interaction of a plant multivalent sugar-binding (glyco)protein with the carbohydrate sequence of a bacterial surface receptor (Bohlool and Schmidt 1974; Dazzo and Hubbell 1975). The general applicability of this theory is still a matter of discussion (for recent reviews, see Kijne et al. 1983; Quispel et al. 1984). Verification of the theory is, among other factors, dependent on the demonstration of lectins on the cell surface of susceptible legume roots. Most of the lectin-binding studies with Rhizobium and rhizobial receptors have been carried out with the use of seed lectins. R o o t lectins are difficult to extract and characterize, partly because they seem to be present in very small quantities. The root lectin of Pisum sativum has been isolated with the use of affinity chromatography and several authors have partially analyzed it with semi-quantitative methods (Gatehouse and Boulter 1980; Hosselet et al. 1983; Kato et al. 1981; Kijne et al. 1980). Pea root lecfin has been reported to be similar to pea seed lectin, both structurally and immunologically (Hosselet et al. 1983). On the basis of its electrophorefic behaviour, root-slime lectin is identical to pea seed isolectin2 (Kijne et al. 1980; Kijne et al. 1983). With one exception (Gatehouse and Boulter 1980), its sugar-hapten specificity has been found to be similar to pea seed lectin. Knowledge about the regulation of synthesis and distribution of legume lectins during root development could contribute to a better understanding of the role of lectins in plant metabolism and rhizobial recognition. Such a study requires a very sensitive, quantitative assay for lectins. We have
C.L. Diaz et al. : Determination of pea root lectin
applied the double-antibody-sandwich enzymelinked immunosorbent assay (ELISA; see Engvall and Perlmann 1972; Voller et al. 1976; Clark and Adams 1977) for the detection of pea root lectin in nanogram quantities. The method is specific, reproducible, and allows the determination of lectins in crude plant fractions (Borrebaek and Mattiasson 1983). We applied the method to test the hypothesis that the well-known inhibition of rhizobial legume infection by critical concentrations of NO~ is correlated with a decrease in detectable lectin concentrations on the root surface (Dazzo and Brill 1978; Dazzo and Hrabak 1981). Material and methods Pea seed lectin. Pea seed, Pisum sativum L. cv. Finale, was purchased from Cebeco, Rotterdam, The Netherlands. Pea seed lectin was isolated from dry pea seed meal or from 16-h-imbided seeds by affinity chromatography using Sephadex G-75 (Pharmacia, Upsala, Sweden). Chromatofocusing of these lectins yielded pea isolectins 1, 2 and 3, as described by Van der Schaal 1983. Immunization procedure. A solution of 1 mg isolectin 2 in 2 ml sterile water was mixed with 1 ml complete Freund's adjuvant. The mixture was injected subcutaneously in small doses in the dorsal and ventral regions of an anaesthetized rabbit. Boosting was done by injecting 0.5 mg isolectin 2 intravenously. Bleeding after 10 d showed a high antibody titer by double diffusion in agar. The boosting and bleeding (50 ml) series were successfully repeated every six weeks over a period of 18 months. Specific antibody isolation. The immunoglobulin fractions were precipitated according to Hijmans et al. (1969). Specific antiisolectin 2 was isolated as described by Van Driessche et al. (1982). Isolectin 2 was bound to CNBr-activated Sepharose 4B (Pharmacia).
303 ranged from 20 to 100 ng per ml PTN, and 100 gl per well were pipetted. Root lectin from 7-d-old peas was used as an internaI standard. Plants were grown in Raggio and Raggio (1956) rooting medium. Root lectin was extracted as described by Kijne et al. (1980). After dialysis, the extract was taken to 60% (NH4)2SO 4 saturation and the precipitate was extensively dialyzed against water. The lyophilyzed preparation did not dissolve completely, but root lectin could be determined in the soluble portion. Appropiate dilutions of the internal standard and samples were pipetted (100 gl per well) and the plates were incubated for 3 h at 37 ~ C. Afterwards, the plates were emptied and the washing step was repeated. Addition of the enzyme-labeled antibody. Lectin-standard lines responded to the dilution of the enzyme-labeled antibody as shown in Fig. 1. A 5-gl enzyme-antibody conjugate per ml PTN was chosen to perform the assays. Each well received 100 gl of this conjugate dilution. The plates were incubated for 3 h at 37 ~ C and emptied. Then the washing step was carried out as described. The enzymatic reaction was performed in 0.05 M sodiumcarbonate buffer, pH 9.6, containing 0.02% MgC12 andp-nitrophenylphosphate (Sigma) in a I nag ml-1 concentration. The reaction was read after 1 h incubation (100 gl per well), at 37 ~ C, using a Titertek Multiscan Plate Reader (Flow Laboratories, McLean, Virginia, USA) at 405 nm. The effect of sugars in lectin determination with the ELISA was assayed. Lectin was incubated in PTN containing 10 or 100 mM glucose or mannose (known pea-lectin haptens), or galactose and compared with lectin incubated only in PTN. The complexing of immobilized antibodies and lectin present in crude seed extracts was assessed. For this experiment, the lectin contained in l0 g pea seeds was isolated, as described by Van Driessche et al. (1978). This procedure involves extensive extraction with 1 M NaC1, 20 m M MgC12 and 20 mM CaClz, followed by a precipitation step (60% (NH4)2SO 4 saturation), before affinity chromatography. The same quantity of seeds was similarly extracted and precipitated. The resulting pellet was dialyzed against water and lyophilyzed. The lectin present in this extract was determined with the ELISA. The total lectin in the crude pea seed extract was calculated and compared to the yield of the affinity-purified pea seed lectin.
Influence of NO~ in nodulation. To determine NO~ concentraPreparation of the enzyme-labeled antibody. Alkaline phosphatase (Sigma, Munich, F R G ; Type VII) was covalently bound to approx. 1.5 mg specific rabbit antiisolectin 2, following the procedure outlined by Lems-van Kan et al. (1983).
Immunoassay. Coating and fixing: approx. 4 gg specific antibody per ml sodium-bicarbonate buffer (0.1 M, pH 9.6) were used to coat Falcon 3912 Microtest Flexible Assay Plates (Becton Dickinson and Co., Oxnard, Cal., USA); 100 I11 of this solution were pipetted per well and incubated overnight at 4 ~ C. Afterwards, plates were emptied and shaken dry. The coating antibodies were fixed by adding 1% freshly prepared glutaraldehyde (Fluka, Buchs, Switzerland; electron microscopy grade) in 0.01 M phosphate buffer (pH 7.4), containing 0.15 M NaCI (PBS). Plates were incubated for 10 rain at 4 ~ C and then rapidly washed. Washing included three times pipetting 250 gl PTN (PBS containing 0.05% Tween-20 (Merck, Darmstadt, FRG) and 0.02% NAN3), each time incubating the plates for I min and shaking them dry. Addition of lectins. Sephadex-G-75-purified pea seed lectin and chromatofocused isolectins showed immunological identity in Ouchterlony tests (Van der Schaal 1983). Because of its better solubility, affinity-purified lectin (Van Driessche et al. 1978) was chosen to prepare standard lines. Lectin concentrations
tions inhibiting nodulation, surface-sterilized pea seeds were inoculated with Rhizobium leguminosarum RBL1. The seeds were planted in gravel soaked with Raggio and Raggio (1956) rooting medium provided with 0, 2, 5, 10, 20 and 50 m M NO~as K N O 3 and Ca(NO3) 2. Nodules were counted after 21 d growth at 20 ~ C and 70% relative humidity. The light flux at table surface was 20000 lx (Philips T F L 60W/33 fluorescent tubes; Philips, Eindhoven, The Netherlands) during the 16-h light period.
Root lectins. Pea seeds were surface-sterilized and soaked for 18 h in Raggio and Raggio (1956) rooting medium or in the same rooting medium supplied with a N O r concentration inhibiting nodulation (20 mM N O r ) . Swollen seeds were planted in gravel with or without NO3-, at 20 ~ C. Roots were sectioned after 4 or 7 d growth in darkness. Between 1200 1800 roots per treatment and age group were collected. Root slime was obtained by thoroughly washing roots with distilled water. The washings were concentrated with a rotary evaporator and lyophilyzed. The lectin in this preparation is referred to as rootslime lectin. Root cell walls were prepared according to Buchala and Franz (1974). Cell-wall preparations were lyophilyzed twice before lectin extraction. The root-lectin extracting buffer, used
304
C.L. Diaz et al. : Determination of pea root lectin
1.0-
0
[3
1.4-
(19 t2/i
0.8 E
1.0-
C t~
S
f3
~ 0.7,
