| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Dep. of Plant Science, Macdonald Campus of McGill Univ., 21,111 Lakeshore Road, Ste Anne de Bellevue, QC, Canada H9X 3V9
* Corresponding author (Donald.Smith{at}McGill.ca).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Abbreviations: NEB, non-Bradyrhizobium endophytic bacteria • PGPB, plant growth promoting bacteria • RZT, root zone temperature
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Rhizobacteria can promote plant growth directly or indirectly. Indirect effects are related to production of metabolites, such as antibiotics, siderophores, or HCN, that decrease the growth of phytopathogens and other deleterious microorganisms. Direct effects are dependent on production of plant growth regulators, or improvements in plant nutrient uptake (Kloepper, 1993; Glick, 1995). Some rhizobacterial strains promote legume nodulation and nitrogen fixation by producing flavonoid-like compounds and/or stimulating the host legume to produce more flavonoid signal molecules (Parmar and Dadarwal, 1999). Endophytic bacteria probably promote host plant growth through similar mechanisms, although, the intimate nature of their endophytic habitats may allow other mechanisms (Hallmann et al., 1997).
Endophytic bacteria are ubiquitous in plant tissues including those of legumes, and have been isolated from flowers, fruits, leaves, stems, roots, and seeds (Kobayashi and Palumbo, 2000), as well as root nodules of legume crops (Sturz et al., 1997). Endophytic bacteria reside intercellularly or even intracellularly within host tissues and therefore are able to form more intimate relationships with the host plant than most other plant-associated bacteria. By residing within plant tissues, endophytic bacteria may also gain advantages for themselves, by being sheltered from environmental stresses and microbial competition. Although some endophytic isolates may inhibit host plant growth (Sturz et al., 2000), it has been shown that a higher proportion of bacterial endophytes is PGPB than is the case for bacteria found on the rhizoplane or in the rhizosphere (Hallmann et al., 1997).
Bacillus are spore-forming Gram positive rod-shaped bacteria. They are highly tolerant of adverse ecological conditions. Bacillus species comprise one of the most common soil bacteria groups and they are frequently isolated from the rhizospheres of plants. Bacillus species are also common plant endophytes (Liu and Sinclair, 1989; Misaghi and Donndelinger, 1990; Sturz and Christie, 1995; Kobayashi and Palumbo, 2000; Araujo et al., 2001). Because of their spore-forming ability, plant growth promoting Bacillus strains are readily adaptable to commercial formulation and field application (Liu and Sinclair, 1993).
Soybean is one of the most important agricultural legumes for oil and protein production. Soybean needs relatively warm temperatures for development; for example, the optimum temperature for soybean symbiotic nitrogen fixation is 25 to 30°C. However, in some short growing season regions, such as southwestern Quebec, the mean soil temperature at a depth of 10 cm is 10°C in mid-May and 15°C in June (Lynch and Smith, 1993). Suboptimal root zone temperatures (RZTs), such as these, strongly inhibit soybean early growth and, especially, establishment of the soybean-B. japonicum nitrogen fixing symbiosis. As such, they are potentially a major limiting factor for soybean production in short season areas (Whigham and Minor, 1978; Zhang and Smith, 1995). Coinoculation of some PGPB, such as Serratia proteamaculans 1-102 and S. liquefaciens 2-68, enhances soybean nodulation and nitrogen fixation under suboptimal RZTs, under both greenhouse and field conditions (Zhang et al., 1996, 1997; Dashti et al., 1997, 1998).
In 1998, we isolated 14 strains of non-Bradyrhizobium endophytic bacteria (NEB) from inside soybean root nodules of particularly vigorous field-grown soybean plants. On the basis of greenhouse run pouch experiments at 25°C RZT, three of the 14 isolates were selected as potential PGPB and identified as Bacillus strains (Bai et al., 2002a). The objective of this work was to determine whether or not these PGPB strains would be able to improve soybean nodulation, nitrogen fixation, and growth under low RZT in the greenhouse, and under field conditions in a short season area, where spring soil temperatures are low.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Bradyrhizobium japonicum was cultured in flasks on a shaker at 200 rev min-1, 50 to 75 mL in 250-mL flasks or 100 to 120 mL in 500-mL flasks, at 28°C in yeast extract mannitol (YEM) culture medium (Vincent, 1970). The initial culture time in flasks inoculated from cold slants was approximately 7 d. The subculture time was not less than 72 h. The cell density in the culture was determined by spectrophotometry at 620 nm, taking an A620nm reading of 0.08 as approximately 108 cells mL-1 (Bhuvaneswari et al., 1980). The Bacillus strains were cultured on a shaker at 200 rev min-1 in flasks, 80 to 100 mL per 250-mL flask or 150 to 180 mL per 500-mL flask, at 28°C. The medium for Bacillus culture was King's Medium B (Atlas, 1995). The initial culture time in flasks inoculated with cold slants was approximately 72 h. The subculture time was 30 h. After the bacterial subcultures were harvested, the cell concentration was determined at 420 nm (Dashti et al., 1997).
The bacterial cultures were diluted with distilled water. The inoculants were prepared by mixing B. japonicum and one of the three tested Bacillus strains. The cell density in the inoculants was 108 cells mL-1 for both B. japonicum and the coinoculated Bacillus strain. Under greenhouse conditions the inoculants were applied immediately after preparation, while for the fieldwork there was a delay of not more than 24 h.
Greenhouse Experiment
In the greenhouse experiments, the only B. japonicum strain used was 532C (Hume and Shelp, 1990). The greenhouse conditions were air temperature of 25 ± 2°C, additional illumination of 300 µmol m-2 s-1 supplied by high pressure sodium lamps (P. L. Light Systems, Canada) for a photoperiod of 16:8 h (day: night). Soybean seeds were surface sterilized in sodium hypochloride [2% (v/v) solution containing 4 mL Tween20 (polysorbate 20) L–1]. The seeds were then rinsed several times with distilled water. The seeds were first planted in trays containing vermiculite and germinated in the greenhouse. Three- or 4-d-old seedlings at the VE stage (Fehr et al., 1971) were transplanted into pots filled with vermiculite (one seedling per pot) or growth pouches (15 x 16 cm, Mega International, Minneapolis, MN, one seedling per pouch). In the pouch experiment, the RZT was controlled, by water bath systems, at 25, 20 and 15°C, respectively (Zhang et al., 1996). Six days after transplanting the seedlings, they were inoculated with the 532C-NEB mixtures at the rate of 1 mL plant-1. Control plants were inoculated with 532C alone or a mixture of 532C and King's Medium B (without bacteria).
The pot experiment was arranged following a completely randomized design with five replicates. The pouch experiment was organized following a randomized complete block design with six replicates. The levels of RZT and inoculations were combined factorially and were allocated to the blocks in a split-plot fashion. The main plots were RZTs. NEB coinoculation treatments formed the subplots. During the growth process, the plants were watered with modified N-free Hoagland's solution (Hoagland and Arnon, 1950), in which Ca(NO3)2 and KNO3 were replaced with 1 mM CaCL2, 1 mM K2HPO4 and 1 mM KH2PO4, to provide a nitrogen-free solution. The plants were harvested at 55 d after inoculation. After harvesting, data on nodule number, nodule weight, shoot weight and root weight were collected. All the samples were weighed after not less than 48 h of drying at 70 to 80°C. The plant weight in the greenhouse experiment was calculated as shoot weight plus root weight.
Field Experiment
The field experiment was structured following a randomized complete block design with three blocks. The B. japonicum and NEB strains, along with the appropriate controls, were combined factorially, resulting in three blocks of 12 plots each. The tested factors were three bradyrhizobial strains (no inoculant control in which the indigenous B. japonicum community was relied on for nodulation, B. japonicum 532C and B. japonicum USDA110), and four NEB treatments (no NEB as a control, NEB4, NEB5, and NEB17). The experiment was conducted at the Emile A. Lods Research Centre of McGill University in 1999, on a clay-loam type soil (Chateauguay clay loam; fine, mixed, non-acid, frigid, typic Humaquept) where the previous crop was corn (Zea mays L.), and in 2000 on a sandy-loam type soil (Chiot fine sandy loam; fine-silty, mixed, nonacid, frigid Humaquept) where the previous crop was barley (Hordeum vulgare L.). The soybean cultivar was OAC Bayfield. In addition, one plot of nonnodulating Evans soybean, planted in the same way, was included in each block. Emergence of the nonnodulating seed was very poor and these plots had to be reseeded 2 wk after the initial planting. Even at the second seeding, emergence in the nonnodulating plots was less than in the OAC Bayfield plots. Thus, data from the nonnodulating plots were not used to calculate N fixation (difference method) by the OAC Bayfield plots. However, relative differences in total N accumulation by nonnodulating plots were used to compare relative soil N levels at the sites used in each of the two years. Each plot was 5 x 1.6 m with 0.2 m between adjacent plots. The plant population was 400 plants plot-1 (500000 plants ha-1) with 10 cm between plants within the row and 20 cm between rows. The sowing date was 20 May 1999 and 17 May 2000. The soybean seed was sown mechanically. The seeds in the furrows were not covered until the inoculants were added. The inoculants were sprayed into the open furrows by hand, with 60-mL sterilized plastic syringes. The inoculation dose for all inoculants was 1 mL seed-1.
The plants were harvested three times during growing season, at V3, R3, and harvest maturity (R8) stages (Fehr et al., 1971). At the first and second harvests, five plants were randomly taken from each plot. After washing the roots with tap water, data on nodule number, nodule weight, shoot weight and root weight were collected in the same way as for greenhouse samples. At the final harvest, plants in the central 1 m of each of the two center rows (an area of 0.4 m2) of each plot were collected by hand. Plant number was determined, and branch number and pod number were counted for each plant. After the roots were removed the shoots were oven dried at 70 to 80°C for not less than 48 h. The shoot weight, including the seeds, was taken as the total weight, i.e., the biological yield or total aboveground biomass. The shoots were mechanically threshed to remove the seeds. The seed weight and the 100-seed weight were also determined. The seed weight was taken as the economic yield. Seed yield is given at 0% moisture. Stem weight was calculated as the difference between the shoot weight and seed weight. The harvest index was expressed as the ratio of the economic yield (the seed weight) to the biological yield (the total weight or total aboveground biomass). The total number of seeds and the seed number per pod were calculated using the variables seed weight, 100-seed weight and pod number. The nitrogen concentrations (%) of the stem and the seed were determined separately with an Elemental Analyzer (NC2500 Elemental Analyzer, ThermoQuest Italic S.P.A., Italy). The nitrogen yield in stem or seed was calculated by multiplying stem or seed weight by their respective nitrogen concentration; the total nitrogen yield was defined as a sum of stem and seed nitrogen yields.
Data Analysis
For the two field experiments, the homogeneity of variance was assessed by a Bartlett's test. If the data were homogenous, they were pooled across experiments in subsequent ANOVA analyses. All the data collected in greenhouse or field experiments were analyzed by the ANOVA procedure of the SAS system (Littell et al., 1991). When analysis of variance indicated differences among means, comparisons among the treatment means were conducted by an ANOVA protected least significance difference (LSD) test (Steel and Torrie, 1980). In general, differences were considered significant when detected at P 
0.05. However, in some cases differences at 0.05 P 0.1 are discussed in the text. When this occurs, the P value is provided.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
2 = 19.23, 
= 0.00001). There were also no interactions among the applied levels of B. japonicum inoculants and Bacillus NEB strains for any of the measured variables in both 1999 and 2000. This occurred in spite of the different growth conditions, because of different soil conditions (soil types and nutrient levels) and weather conditions (Fig. 1) between the experiments in 1999 and 2000, which contributed to different levels of overall plant growth. The overall growth conditions of 2000 would be considered to be better than in 1999. In 1999, the average total biomass production was 10.08 Mg ha-1 and seed production was 5.33 Mg ha-1, whereas these values were13.95 and 7.83 Mg ha-1 in 2000. Similar differences also existed when the respective within growth season harvests (at V3 and R3 stages) were compared across years. However, the included plots of nonnodulating Evans soybean accumulated more total nitrogen in 1999 (219.0 kg/ha) than in 2000 (153.3 kg/ha). This indicated higher soil nitrogen at the 1999 site than the 2000 site. The higher soil nitrogen in 1999 resulted in fewer nodules and less nodule weight than in 2000. The average nodule number and nodule weight were 18.4 (plant-1) and 0.083 (g plant-1) at the V3 stage and 48.5 and 0.46 at the R3 stages in 1999. The average nodule number and nodule weight were 47.6 and 0.29 at the V3 stage and 73.7 and 0.74 at the R3 stage in 2000.
|
|
|
Field Experiments (B. japonicum Strains)
At the final harvests, there were few differences among the three B. japonicum treatments (no-inoculant, B. japonicum 532C and B. japonicum USDA110) in 2000 (Table 4). In 1999, both the inoculated bradyrhizobial strains increased biological and economic yields relative to the control (no-inoculant) (P = 0.10).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
There were obvious differences, due to weather and soil conditions, between the experiments conducted in 1999 and 2000. Different soil types and different previous crops result in different soil nitrogen levels (Dashti et al., 1997; Pan and Smith, 2000). When compared with the 30-yr average, monthly average temperatures from May to September in 1999 were high and were higher than in 2000. This was especially so from May to July when it was 2.5 to 3.5°C higher in 1999 than in 2000. In 1999, the precipitation in May was low, but in September it was much higher than in 2000 and than in the past 30 yr (Fig. 1). In spite of the very different field conditions in 1999 and 2000, NEB17 caused similar effects on plant nodulation and growth. This corroborates the greater stability for NEB17 observed in the greenhouse experiments.
There were differences between the inoculant and no-inoculant B. japonicum treatments in 1999, but not in 2000. The lack of differences among the three levels of B. japonicum in 2000 indicated the presence of an indigenous B. japonicum population that was sufficient to allow nodulation and nitrogen fixation levels that could not be improved by the application of inoculants containing B. japonicum alone. In these two years of experimentation, the two applied B. japonicum strains, 532C and USDA110, performed similarly, as previously observed (Dashti et al., 1997).
Bacillus NEB 17 had beneficial effects on soybean growth similar to those previously reported for the two Serratia strains, S. proteamaculance 1-102 and S. liquefaciens 2-68 (Zhang et al., 1996, 1997; Dashti et al., 1997, 1998). We have determined that S. proteamaculans 1-102 produces flavonoid inducible activator(s) that cause the positive effects on signal exchange between bradyrhizobia and soybean plants (Bai et al., 2002b). Whether NEB17 and S. proteamaculans 1-102 promote plant growth through similar or different mechanisms remains to be investigated. There was no apparent insect or disease pressure either in the greenhouse or in the field. Based on the experimental results reported here, we know that NEB17 did not exert its plant growth promoting effects through biocontrol of any insect or disease organisms. However, the current data do not allow us to conclude that the NEB strains have no biocontrol potential in all situations. In fact, some Bacillus PGPB strains probably promote plant growth through biocontrol of disease and insect pests (Liu and Sinclair 1989, 1990; Handelsman et al., 1990).
In both the greenhouse and field experiments, no nitrogen fertilizer was applied. These experiments were designed so that N was the limiting nutrient for plant growth. Thus, we think that the increased nodulation and subsequent nitrogen fixation resulted in the measured increases in plant growth and grain yield. Although no direct nitrogen fixation data were collected, the higher total nitrogen in NEB coinoculation treatments than in the no-NEB coinoculation control in field suggests greater nitrogen fixation. Coinoculation with the NEB increased nodule number and nodule weight, and root weight was increased more often than shoot weight. Increased root development means increased nutrient uptake capability and some PGPB are known to exert their plant growth promoting effects via stimulating root growth through production of indole-3-acetic acid (Barbieri et al., 1986; Barbieri and Galli 1993; Dubeikovesky et al., 1993; Srinivasan et al., 1996). The NEB strains tested here might also produce plant growth regulators or other activators that cause promotion of root growth and enhancement of nodulation. This possibility needs to be investigated through additional research.
Bacillus species are among the bacteria most often isolated from plant tissues (Kobayashi and Palumbo, 2000). Philipson and Blair (1957) isolated B. megaterium from clover root tissues. Sturz et al. (1997) isolated 32 endophytic bacteria from red clover, six Bacillus strains were included (B. azotoformans, B. brevis, B. circulans, B. insolitus, B. megaterium, and B. subtilis) and four of these were resident within root nodules. In the inoculation experiments, B. brevis and B. insolius promoted clover nodulation when coinoculated with Rhizobium leguminosarium (Sturz et al., 1997). Bacillus species exist in the soybean seed tissues and could survive common surface sterilization procedures (Tenne and Sinclair, 1977). Bacillus subtilis (Ehrenberg) Cohn, the most constantly soybean seed borne bacterial species, always resulted in soybean seed decay (Schiller et al., 1977; Sinclair, 1993). Oehrle et al. (2000) reported that some soybean seedborne Bacillus spp. inhibited the attachment of B. japonicum to the soybean seedling root surface. Thus, soybean seedborne Bacillus strains often show negative effects on soybean growth and symbiosis establishment. The three Bacillus strains used in this work were spore-forming rods isolated from surface disinfected soybean root nodules (Bai et al., 2002a), not seeds, and showed growth promoting effects on soybean plants. To determine whether these strains are suitable for development into commercial inoculants, further experiments are necessary, especially to verify their plant growth promoting efficacy under a wider range of field conditions and with a greater number of soybean cultivars, and to evaluate relevant safety issues. However, their spore-forming characteristics make them more adaptable to commercial formulations in future inoculant production (Liu and Sinclair, 1993).
In conclusion, all three NEB strains tested showed plant growth promoting effects. The growth promotion provided by these strains was apparently related to improved root development and enhanced nodulation, which resulted in better nutrient uptake capability and increased N supply. One of them, NEB17, was superior in this regard because it provided the best and most consistent effects. NEB17 seems to be suitable for use as a plant growth promoting bacterial strain in soybean production in short season regions.
Received for publication June 11, 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  J.-M. Barea, M. J. Pozo, R. Azcon, and C. Azcon-Aguilar Microbial co-operation in the rhizosphere J. Exp. Bot., July 1, 2005; 56(417): 1761 - 1778. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Vadose Zone Journal | |||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
