Stability and activity of the major nod factor produced by Bradyrhizobium japonicum following purification, sterilization, and storage.
Host-plant responses to LCOs involve calcium as a second messenger via G protein and phospholipases C and D, and include root hair deformation, infection thread formation, cortical cell division, and the formation of pseudonodules in some species (for recent reviews see Broughton et al., 2003; Limpens and Bisseling, 2003; Ferguson and Mathesius 2003). Recently, studies in our laboratory found that LCO NodBj-V (C18:1,MeFuc) stimulated in vivo germination and early growth of a number of economically important legumes and nonlegumes under laboratory, greenhouse, and field conditions (Souleimanov et al., 2002; Prithiviraj et al., 2003).
Enzymatic degradation of LCOs has been studied. LCOs are degraded by chitinases excreted by the host plants (Prithiviraj et al., 2000; Staehelin et al., 2000). The stability of Nod factor depends on sulfation, the length of the chitin backbone, N-deacetylation, and fucosylation (Staehelin et al., 2000; Minic et al., 1998; Ovtsyna et al., 2000). The rate of Nod factor degradation is inversely correlated with their root hair deformation activity (Staehelin et al., 1994). However, there is no available information regarding the stability and activity of LCOs due to handling and storage. Because LCOs must be purified during isolation and may often need to be stored before use in research applications, there is a need for improved understanding of this area. Therefore, the present experiments were conducted to evaluate whether purification, sterilization, and storage procedures affect the quantity and the biological activity of NodBj-V(C18:1,MeFuc) produced by Bradyrhizobium japonicum.
MATERIALS AND METHODS
Bradyrhizobium japonicum strain 532C was obtained from Liphatec (Milwaukee, WI) and was maintained on yeast extract mannitol agar (YEM) (Vincent, 1970). Seeds of soybean cv. OAC Bayfield were obtained from Secan (Montreal, QC). Filters were purchased from Fisher Scientific (Ontario, Canada). Pyrex glass tubes (10 x 75 mm) were used during handling procedures, except where specified.
The LCO NodBj-V(C18:1 MeFuc) was produced from cultures of B. japonicum strain 532C induced by genistein, following methods described by Prithiviraj et al. (2000). The Nod factor peak was identified by comparing the retention time of a standard [NodBj-V(C18:1, MeFuc)--a gift from Prof. G. Stacey, University of Tennessee, Knoxville, TN]. The collected peak used in this work was relatively pure, although the proximity of other LCO peaks on either side, on the HPLC chromatogram, usually resulted in some contamination with adjacent LCOs. Absolute purification is time consuming and expensive; industrial applications will involve relatively pure material of the type used here. The concentration of the collected Nod factor was calculated through comparison of the area under the peak of the standard, at a known concentration. With the exception of the freeze-drying study, all experiments used lyophilized LCO. Each LCO batch was subjected to bioassay (root hair curling and stimulation of seed germination) for activity before utilization. Any batch failing the bioassay was discarded.
Purification by freeze-drying was needed to eliminate the acetonitrile that acted as a part of the eluent during HPLC purification, from the resulting LCO. Because there was a risk that glass containers would be broken during freezing, we evaluated the suitability of polypropylene conical tubes for freeze drying process. LCO lyophilized in polypropylene tubes was redissolved in deionized water to [10.sup.-5] M and used in other experiments.
Sterilization of Nod factors for aseptic experimentation usually involves the use of filter or autoclaving methods. For filter sterilization, experiments were conducted to examine the effect of various types of filters on the LCO content of the resulting solutions. Five types of filter were tested: cellulose acetate (CA, pore size of 0.22 [micro]m--Nalgene, Rochester, NY), mixed cellulose ester (MCE, 0.22 [micro]m--Fisher), nylon (N, 0.45 [micro]m--Corning, Corning, NY), polyestersulfone (PES, 0.22 [micro]m--Millipore, Billerica, MA), and polytetrafluoroethylene (PTFE, 0.22 [micro]m--Fisherbrand). In each case, 5-mL syringes were used to force 2 mL of LCO solution through each filter and into a receptacle test tube. LCO concentration in the filtrate and the source material was measured by HPLC. In addition, the effect of autoclaving duration on LCO was examined. Two milliliters of LCO solution were placed in an aluminum foil-covered glass tube and autoclaved under 103.4 kPa, at 121[degrees]C, for durations ranging from 15 to 30 min. Because the solution volumes were slightly reduced by autoclaving, the autoclaved solutions were returned to their original volumes by addition of water, before HPLC quantification of LCO. LCO concentration was determined in both autoclaved and source materials.
A storage experiment was conducted to study the longer term stability of NodBj-V(C18-1 MeFuc) at two temperatures (room, 23 [+ or -] 2[degrees]C, and refrigerator, 4 [+ or -] 2[degrees]C). Two batches, from different cultures, acted as replicates; extracted LCO was adjusted to 10-s M with sterile deionized water and 2-mL samples (two from each batch) were established, with one being placed at each storage temperature. Solutions of each batch were put into sterile Pyrex glass tubes and sealed with Parafilm (American National Can, Greenwich, CF) before storage at room temperature or in a refrigerator. The tubes (one sample from each batch) were randomized in a rack for each temperature condition. This process was repeated every 2 mo and all batches were assayed at the end of the experiment. Because some evaporation occurred during storage sterile distilled water was added to each sample, returning them to 2-mL volumes, every other month and before analysis.
Following storage, lyophilization, filtration, and autoclaving, the LCO contents of the treated solutions, along with their respective source materials, were quantified by HPLC as described previously (Prithiviraj et al., 2000). Two hundred microliters of solution were loaded onto the HPLC, and the concentration of LCO were quantified by comparing the area under the peak of a 5-[micro]g standard [NodBj-V(C18:1, MeFuc)] from strain USDA110. Data are presented as percentage LCO recovered from the source material after treatments.
Biological assays were conducted for soybean [Glycine max (L.) Merr.] germination and root hair deformation to determine the biological activity of LCO recovered from autoclaving and storage experiments. A single concentration of [10.sup.-7] M was used in root hair deformation and germination studies, based on previous dose-response experiments for root hair deformation (Prithiviraj et al., 2000) and stimulation of soybean seed germination (Prithiviraj et al., 2003). For germination assays, soybean seeds were carefully selected for uniformity of size and any cracked or broken seeds were removed. Seeds were then surface sterilized with 400 mL/L commercial bleach (52.5 g/L sodium hypochlorite) for 2 min and rinsed four times. Only seeds showing early signs of imbibition (wrinkle coat) were used in germination tests. Ten seeds were put on two layers of germination paper (Anchor Paper Co., St. Paul, MN) in each 15- X 90-mm plastic Petri dish containing 7 mL of treatment solution (water or [10.sup.-7] M LCO). There were 10 replications (Petri dishes) of each treatment. Petri dishes were randomized in an incubator set at 25[degrees]C and were kept in the dark. Germination was observed at regular intervals for 72 h. A seed was considered to be germinating when the radicle tip protrude from the structure surrounding the embryo (Bewley, 1997). In our case, at least 1 mm of radicle had appeared from inside the seed coat. Mean germination time (MGT) of each Petri dish was calculated by the following formula:
MGT = ([summation]XiHi)/[summation]Xi
where [H.sub.i] = the time (h) of observation after germination, and Xi = the number of seeds germinated at [H.sub.i] (Scott et al., 1984). Seedling growth was measured from normal seedlings at 7 d after commencement of the germination test. Measurement included length and dry weight of the shoot and the root, and dry weight of the remainder of the seed. Total plant length was obtained by summing shoot and root lengths; whereas total axis weight was obtained by summing shoot and root weights. Dry weight was measured after oven drying at 80[degrees]C for 3 d.
The root hair deformation assay followed that described by Prithiviraj et al. (2000). Root hair deformation was observed in Zone 2, the zone where root hairs are elongating and are most responsive to LCOs (Heidstra et al., 1994; den Hartog et al., 2001). The deformation responses included wiggling, curling, bulging, and branching of the root hairs; however, the proportion of each type of root hair deformation was not determined.
The data were analyzed by the SAS statistical Package (SAS, 1990). Differences among means are only considered to be statistically significant when detected at the 0.05 level of probability. However, differences between 0.1 and 0.05 are sometimes discussed in the text; when this occurs the P values are given.
RESULTS AND DISCUSSION
We tested whether polypropylene containers can be used to replace glass for freeze-drying of LCOs. When LCO samples were freeze-dried in plastic and glass containers, there was no difference in recovery of LCO compared with the source material (approx. 90%) (Fig. 1). These findings alleviate concern that LCO may stick to the walls of the container (largely because of the lipid side chain) when polypropylene is used.
[FIGURE 1 OMITTED]
Sterilization experiments were conducted to evaluate the feasibility of the two common methods, i.e., filter sterilization and autoclaving, for use in sterilizing LCO. Five different types of filters were tested for their utility in filter-sterilizing LCO (Fig. 1). In all cases, filtering reduced the amount of LCO in the solution (P < 0.01). There were differences among various types of filters for amount of LCO removed (P < 0.05). Polyestersulfone was the most suitable filter with the highest LCO recovery (67.7%), followed by cellulose acetate (54.9%), nylon (48.1%), and polytetrafluoroethylene (38.0%). Mixed cellulose ester, with an average recovery of 31.8%, was the worst filter type for sterilizing LCO. The filter sterilization data showed that care is needed in choosing filter type, as substantial amounts (more than 50%) of the LCO can be lost with inappropriate choice of filter. The difference in LCO recovery after filtration might result from differences in the lypophilic nature of filter types. For example, nylon used in this experiment has a pore size of 0.45 [micro]m, twice that of the other materials, and should trap less LCO, but its more lypophilic character makes it less suitable for LCO filtration with less than 50% of the LCO recovered.
Autoclaving may be more convenient than filtration for sterilization of LCO in some cases, especially if LCOs have to be incorporated into a sterile medium. The suitability of autoclaving for LCO sterilization was examined with autoclaving durations of 15 to 30 min (data not shown). There was some apparent LCO degradation during autoclaving; however, there were no differences among durations of autoclaving (P = 0.60) (data not shown). The amount of LCO recovered following autoclaving was 67.3 to 71.7%. This is the first report on the sterilization of Nod factors by autoclaving. Autoclaving is a viable alternative to filter sterilization of LCO as the results showed that a similar proportion (about 70%) of LCO material could be recovered after both autoclaving a 2-mL solution of [10.sup.-5] M LCO at 121[degrees]C, 103.4 kPa for up to 30 min or filtration using polyester sulfone. It should be noted, however, that we have autoclaved purified LCO solutions. We do not know the fate of LCOs autoclaved with potentially reactive compounds, such as phosphate, included in the solution. Since the amount of LCO recovered following a range of durations, 15 to 30 min, of autoclaving, were not different (P = 0.60), the degradation of LCO that does occur may take place during, at most, the first 15 min of autoclaving, suggesting that a subset of the LCO molecules is more labile than the rest.
Preparing stock solutions and storing them is convenient for conducting repeated experiments. We used water as a storage solvent for studying the stability of LCOs during storage, following the approach of Felle et al. (1998). The choice of solvent is important since organic solvents, such as n-butanol, can be problematic during experiments as it can inhibit root hair development and deformation (den Hartog et al., 2001; Ohashi et al., 2003). We chose water since it can cause hydrophobic contacts between the sugar units and the fatty acid which stabilize the molecule, whereas these contacts do not happen in DMSO (dimethylsulfoxide) (Gonzalez et al., 1999).
The ability to store the LCO NodBj-V(C18:1,MeFuc) was examined with storage durations of up to 16 mo under two storage temperature regimes: room temperature (23 [+ or -] 2[degrees]C) and standard refrigeration (4 [+ or -] 1[degrees]C). The amount of recoverable LCO decreased gradually with the time in storage (P < 0.01); the decrease was greater when the LCO was stored at room temperature (P < 0.05) (Fig. 2). At the end of the experiment (after 16 mo), 84.0% of the LCO was recovered from the 4 [+ or -] 1[degrees]C storage regime, whereas 74.3% was recovered from the 23 [+ or -] 2[degrees]C storage regime.
[FIGURE 2 OMITTED]
LCOs recovered from autoclaving at 121[degrees]C for 30 min, or storage at 4 [+ or -] 1 or 23 [+ or -] 2[degrees]C were corrected for loss, and a [10.sup.-7] M concentration was evaluated for biological activities such as root hair deformation (Prithiviraj et al., 2000) and seed germination tests (Prithiviraj et al., 2003). Results of root hair deformation and seed germination assays showed that, applied at [10.sup.-7] M, LCO recovered from autoclaving at 121[degrees]C for 30 min, or storage at 4 [+ or -] 1 or 23 [+ or -] 2[degrees]C for as long as 16 mo, did not affect the biological activity of LCO as compared with freshly prepared LCO when tested at the same concentration. The percentage deformed root hairs on LCO treated root tips ranged from 65.8 to 75.1%, whereas for water-control treated root tips the average value was only 12.1% (data not shown).
Similarly, reduced mean germination time of soybean seeds was observed in all LCO treatments as compared with water control (Fig. 3). The increase in germination percentage was detected from 21 to 36 h; given the ideal conditions for germination in the Petri plates, most seeds of all treatments eventually germinated, so that differences based on germination percentage disappeared by 72 h (at least 94% germination for all treatments at this time). The earlier germination did result in larger seedlings (next paragraph) presumably because of the export of more stored material from the seeds to the seedlings, as indicated by lighter cotyledons of LCO treated seeds at the end of the experiment. Treatment with LCOs advanced mean germination time by 3 to 5 h over the 30.9-h average of the water control and, at the same concentration, there was no difference in mean germination time between autoclaved or stored LCO and freshly prepared LCO (P < 0.01) (Fig. 3).
[FIGURE 3 OMITTED]
Seedlings were harvested 7 d after germination commenced and their size was measured (data not shown). The number of normal seedlings was not affected by LCO treatment, and an average of six to seven normal seedlings were obtained perl0 seeds germinated in each Petri plate. Root length and seedling length were increased by LCO treatments (P < 0.05 for both). At the same concentration, there were no differences among autoclaved, stored and freshly prepared LCOs. Shoot length was not affected (P = 0.88) by LCO treatment. As a result, root weight and axis weight were increased by LCO treatment (P < 0.05 for both), but shoot weight was also increased (P = 0.09). The weight of the seeds, after removal of the embryo axis, was decreased by LCO treatments (P = 0.08).
In conclusion, the best method for sterilizing LCOs is autoclaving, although if sterilization is required losses with mixed cellulose ester are similar to those with autoclaving. Refrigeration is the best storage method tested, although the freeze-drying procedure shows that freezing is also not detrimental.
We greatly appreciate support by the Natural Science and Engineering Research Council of Canada through a grant proposal to DLS. The first author also acknowledges scholarship support from the DUE-Project, University of Bengkulu, Indonesia.
Bewley, J.D. 1997. Seed germination and dormancy. Plant Cell 9: 1055-1066.
Broughton, W.J., F. Zhang, X. Perret, and K. Staehelin. 2003. Signals exchanged between legumes and Rhizobium: Agricultural uses and perspectives. Plant Soil 252:129-137.
den Hartog, M, A. Musgrave, and T. Munnik. 2001. Nod factor-induced phosphatidic acid and diacylglycerol pyrophosphate formation: A role for phospholipase C and D in root hair deformation. Plant J. 25:55-65.
Felle, H.H., E. Kondorosi, A. Kondorosi, and M. Schultze. 1998. The role of ion fluxes in Nod factor signalling in Medicago sativa. Plant J. 13:455-463.
Ferguson, B.J., and U. Mathesius. 2003. Signaling interactions during nodule development. J. Plant Growth Regul. 22:47-72.
Gonzalez, L., M. Bernabe, J.F. Espinosa, P. Tejero-Mateo, A. Gil-Serrano, N. Mantegazza, A. Imberty, H. Driguez, and J. Jimenez-Barbero. 1999. Solvent-dependent conformational behaviour of lipochitoligosaccharides related to Nod factors. Carbohydr. Res. 318:10-19.
Heidstra, R., R. Geurts, H. Franssen, H.P. Spaink, A. van Kammen, and T. Besseling. 1994. Root hair deformation activity of nodulation factors and their fate on Vicia sativa. Plant Physiol. 105:787-797.
Limpens, R., and T. Bisseling. 2003. Signaling in symbiosis. Curr. Opin. Plant Biol. 6:343-350.
Minic, Z., S. Brown, Y. de Kouchkovsky, M. Schultze, and C. Staehelin. 1998. Purification and characterization of a novel chitinase-lysozyme, of another chitinase, both hydrolysing Rhizobium meliloti nod factors, and of a pathogenesis-related protein from Medicago sativa roots. Biochem. J. 332:329-335.
Ohashi, Y., A. Oka, R. Rodrigues-Pousada, M. Possenti, I. Ruberti, G. Morelli and T. Aoyama. 2003. Modulation of phospholipid signaling by GLABRA2 in root-hair pattern formation. Science 300: 1427-1430.
Ovtsyna, A.O., M. Schultze, LA. Tikhonovich, H.P. Spaink, E. Kondorosi, A. Kondorosi, and C. Staehelin. 2000. Nod factors of Rhizobium leguminosarum bv. viciae and their fucosylated derivatives stimulate a nod factor cleaving activity in pea roots and are hydrolyzed in vitro by plant chitinases at different rates. Mol. Plant Microb. Interact. 13:799-807.
Prithiviraj, B., A. Souleimanov, X. Zhou, and D.L. Smith. 2000. Differential response of soybean (Glycine max (L) Merr.) genotypes to lipo-chito-oligosaccharide Nod Bj-V(C18:1 MeFuc). J. Exp. Bot. 51: 2045-2051.
Prithiviraj, B., X. Zhou, A. Souleimanov, M.K. Wajahatullah, and D.L. Smith. 2003. A host specific bacteria-to-plant signal molecule (Nod factor) enhances germination and early growth of diverse crop plants. Planta 216:437-445.
SAS Institute Inc. 1990. SAS users guide, Version 6.12, Cary, NC, USA.
Scott, S.J., R.A. Jones, and W.A. Williams. 1984. Review of data analysis methods for seed germination. Crop Sci. 24:1192-1199.
Souleimanov, A., B. Prithiviraj, and D.L. Smith. 2002. The major Nod factor of Bradyrhizobium japonicum promotes early growth of soybean and corn. J. Exp. Bot. 53:1929-1934.
Spaink, H.P., G.V. Bloemberg, A.A.N. van Brussel, B.J.I. Lugtenberg, K.M.G.M. van der Drift, J. Haverkamp, and J.E. Thomas-Oats. 1995. Host specificity of Rhizobium leguminosarum is determined by the hydrophobicity of highly unsaturated fatty acyl moieties on the nodulation factors. Mol. Plant Microb. Interact. 8:155-164.
Stacey, G., J. Sanjuan, S. Luka, T. Dockendorff, and R.W. Carlson. 1995. Signal exchange in the Bradyrhizobium-Soybean symbiosis. Soil Biol. Biochem. 27:473-483.
Staehelin, C., M. Schultze, K. Tokuyasu, V. Poinsot, J.C. Prome, E. Kondorosi, and A. Kondorosi. 2000. N-deacetylation of Sinorhizobium meliloti Nod factors increases their stability in the Medicago sativa rhizosphere and decreases their biological activity. Mol. Plant Microb. Interact. 13:72-79.
Staehelin, C., J. Granado, J. Muller, A. Wiemeken, R.B. Mellor, G. Felix, M. Regenass, W.J. Broughton, and T. Boiler. 1994. Perception of Rhizobium nodulation factor by tomato cells and inactivation by root chitinases. Proc. Natl. Acad. Sci. USA 91:2196-2200.
Vincent, J.M. 1970. A manual for the practical study of root nodule bacteria. Oxford, Blackwell Sci. Publ. Oxford. UK.
Supanjani, Fazli Mabood, Alfred Souleimanov, Kyung D. Lee, and Donald L. Smith *
Supanjani, Dep. Agronomy, Fac. Agric. Univ., Bengkulu, Indonesia; F. Mabood, A. Souleimanov, K.D. Lee, and Donald L. Smith, Dep. of Plant Science, McGill Univ., Macdonald Campus, 21111 Lakeshore Road, Ste. Anne de Bellevue, QC, Canada H9X 3V9. Received 17 Mar. 2004. * Corresponding author (Donald.Smith@McGill.Ca).
|Printer friendly Cite/link Email Feedback|
|Author:||Supanjani; Mabood, Fazli; Souleimanov, Alfred; Lee, Kyung D.; Smith, Donald L.|
|Date:||Jul 1, 2005|
|Previous Article:||Disease and insect resistance in cultivated barley accessions from the USDA National Small Grains Collection.|
|Next Article:||An empirical model for pollen-mediated gene flow in wheat.|