WORK RESEARCH
Performance of the legume-rhizobium symbiosis in Centrosema plumieri inoculated with Bradyrhizobium sp. Field essay
C. J. Bécquer, J. A. Nápoles, Orquidia Álvarez, Yamilka Ramos y L. A. Palmero
Instituto de Investigaciones de Pastos y Forrajes, Estación Experimental Sancti Spíritus, Apdo 2255, Sancti Spíritus, Cuba
E-mail: pastossp@yayabo.inf.cu
ABSTRACT
A field trial, without irrigation, was conducted at the Sancti Spiritus Experimental Station of Pastures and Forages, Cuba, from November, 2001, to April, 2002. The objective was to evaluate the symbiotic performance of Centrosema plumieri, inoculated with Bradyrhizobium sp. strains, previously isolated from forage legumes of Cuban livestock production ecosystems. The experimental design consisted in randomized blocks; 11 treatments inoculated with native strains, a control inoculated with the reference strain 5030 (Bradyrhizobium sp.) and a non-inoculated control were used, with three replications. A variance analysis (ANOVA), as well as correlation and regression analyses, were applied. The values of the inoculated treatments were statistically equal to those of the absolute control, regarding aerial dry weight (ADW) and N yield (NY), except the one inoculated with the strain HA2 in NY. In general, a weak activity of inoculation was shown. Nevertheless, in the nodulation rate (NR) JH2 and HA2 stood out, with values statistically higher (p<0,05) than the others. The degree of dependence of NY and ADW was weak in the inoculated treatments with regards to NR, although it was strong in NY with regards to ADW. It is concluded that the native strains inoculated in C. plumieri, with the exception of HA2 in NY, as well as JH2 and HA3 in NR, did not exert a highly positive effect on this plant. These results should be evaluated for the continuation of studies related to the legume-rhyzobium symbiosis.
Key words: Bradyrhizobium sp., Centrosema plumieri, inoculation.
INTRODUCTION
The legume-rhyzobium symbiosis reciprocal relation between the plants and the bacterium- was first acknowledged more than 100 years ago, and its agricultural value has been proven by the scientific studies conducted throughout the world.
The importance of the biological nitrogen fixation for food security worldwide is really non-debatable, and the use of crops capable of doing a symbiotic N fixation is the primary component in sustainable agriculture (Vance et al., 2000).
On the other hand, in Cuba there are works describing the symbiotic efficiency of Cuban native strains in tropical legumes (Tang et al., 1994; Tang, 1996; López et al., 2002).
The Sancti Spiritus province has a wide range of genera and species of native or naturalized legumes, disseminated in four edaphoclimatic zones (Hernández, 1989), and can be located in ecosystems with stressing environmental characteristics (acidity, alkalinity or high temperatures, among others) for the plants as well as the microorganisms.
Centrosema is a neotropical genus, which contains very important pasture and forage species. It is native from southern tropical America, abundant in the tropics (Skerman et al., 1991) and it is said to be able to fix up to 280 kg N/ha/year in association with grasses (Sylvester-Bradley et al., 1983). It is moderately specific in nodulation, and the soil can lack strains to infect it effectively, for which its inoculation is recommended. In studies conducted by Hernández et al. (1999), in native or naturalized legumes from the Sancti Spiritus province, Centrosema was found to be one of the most adapted genera to the edaphoclimatic conditions of the province. Olivera et al. (2008) also found this genus to be widely disseminated in three Cuban provinces. Such studies allow further work on the use, under those conditions, of natural species; the plants which have originated in a certain country, or have been introduced several years ago, and are completely adapted to the environment, are considered natural (Funes, 2007).
The objective of this work was to evaluate the symbiotic performance of Centrosema plumieri, inoculated with Bradyrhizobium sp. strains under field conditions, in Sancti Spiritus, Cuba.
MATERIALS AND METHODS
The trial was conducted in areas of the Sancti Spiritus Experimental Station of Pastures and Forages, Cuba (21° 53' 00'' latitude north, 79°21'25'' longitude west, and at 40 masl) and lasted for six months (November, 2001-April, 2002).
The soil of the experimental area is typical Brown with Carbonates (Anon, 1979), with deficit of P2O5 and K2O, and moderate organic matter levels (table 1), which is characteristic of this soil type (Hernández, A. et al., 1999).
No irrigation was applied although in the evaluative period there was scarce rainfall-, except in November, planting date of the experiment (table 2), which propitiated the adequate seed germination (this datum is not shown).
The seeds were from the Sancti Spiritus Experimental Station of Pastures and Forages. They were scarified in water at 90°C, during three minutes, before planting.
The native strains used in the trial belong to Bradyrhizobium sp. and were isolated from C. plumieri and Centrosema virginianum, legumes from the central zone of Sancti Spiritus, which were taxonomically characterized and agronomically evaluated under greenhouse conditions (Bécquer, 2002). The reference strain 5030 (Bradyrhizobium sp.), donated by the Cuban Pasture and Forage Research Institute (table 3), was also used.
The strains were preserved in agarized containers with yeast-mannitol solid medium, at 4°C (Vincent, 1970). The inoculants were prepared by growing the strains in 259-mL Erlenmeyer flasks, which contained 100 mL of yeast-mannitol liquid medium, and they were incubated at 29-30°C, from five to eight days, in a rotary shaker (160 rpm) until reaching a CFU of 106-108 cells/mL. The corresponding inoculants were elaborated in solid substratum (peat) and the seeds were inoculated in the conventional way, according to the methodology proposed by Somasegaran and Hoben (1994).
The plot size was 3 m x 15 m. A planting dose of 15 kg/ha was used and seeding was done by spaced drilling, with a distance between rows of 0,75 m. The dry weight was determined with a framework of 1 m², in four samples within the plot.
The measurements were carried out in the second cutting, during the flowering stage and beginning of pod emergence, to guarantee the highest agronomic expression of the crop after establishment.
The experimental design was completely randomized blocks. Thirteen treatments were used: 11 of them inoculated with strains isolated from Centrosema spp., an inoculated control (with the reference strain 5030) and an absolute control (neither inoculated nor fertilized treatment), with four replications. A variance analysis (ANOVA) was applied for inoculation trials. The evaluated variables were: aerial dry weight, ADW (g/plot), nodulation rate (NR) and nitrogen yield, NY (mg/plot). The inoculation effectiveness rate, IER (Davies et al., 2005), was calculated from the ADW value, for which the following formula was used:
IER: [(inoculated treatment non inoculated treatment) / non inoculated treatment] x 100
An only factor (strains) was considered and its effect on the legume as compared to the non-inoculated control, without using the fertilized control (Danielle Prévost, personal communication). The differences among means were determined by Duncan's comparison test (1955). The statistical pack SPSS 8.0 for Windows was used. To determine the dependence character of the variable NY with regards to NR, as well as NY with NR and ADW with NY, a simple linear regression coefficient was used (Ostle, 1984).
The NR was calculated according to the parameters that are shown in table 4 (Danielle Prévost, personal communication) and the data were transformed by ?x (Ostle, 1984).
The total nitrogen was calculated from its digestion in sulfuric acid and the application of the Kjeldahl microanalitical method. With those data the nitrogen yield (NY) was calculated, through the multiplication of the N percentage by the plant ADW.
RESULTS AND DISCUSSION
The remarkable nutrient deficit propitiated higher reliability of the experimental results, because a significant interference was not to be expected due to the high availability of macroelements which could mask the positive effect of the strains, based on their properties as atmospheric nitrogen fixing organisms.
In the variable ADW (table 5) no effective response of the inoculation on the plant was observed. None of the values of the inoculated treatments was statistically higher (p<0,05) than the non-inoculated control (259,73 g/plot). The one inoculated with the native strain JK4 (181,30 g/plot) was statistically lower with regards to the non-inoculated control. There were no differences between the treatments inoculated with the native strains and the one inoculated with the commercial strain 5030 (243,53 g/plot). Similar results were found by Date (1991) when inoculating Macroptilium atropurpureum cv. Siratro with previously selected Bradyrhizobium spp. strains. Related to this variable, the symbiotic effectiveness rates (table 5) had very low values; the treatment inoculated with the strain SP7 showed the highest value (19,15%), and JI1, the lowest (4,58%); likewise, there were treatments with negative values (JH1, JH3, SP10, JK2, JK4, JK3, 5030). This formula corroborated the weak symbiotic effectiveness of the inoculated strains as compared to the non-inoculated control, previously explained in the ADW values.
On the other hand, in the variable NY (table 5) only the treatment inoculated with the native strain HA2 (505,90 mg/plot) was statistically higher than the non-inoculated control (360 mg/plot), as well as those inoculated with the native strains SP10 (376,13 mg/plot), JK2 (283,90 mg/plot) and JK4 (307,07 mg/plot). As in the variable ADW, from these results it could be inferred that the natural population of rhizobia in the rhizosphere of the non-inoculated control exerted a higher effect on the plant regarding the atmospheric dinitrogen fixation, although this did not imply, necessarily, a higher root infection capacity in such strains. The possibility is not discarded either of the plant having utilized more efficiently the mineral N in the non-inoculated control due to the absence of effective autochthonous rhizobia; as it was observed in the variable NR, such control showed the worst root infection values.
In this sense, Collins et al. (2009) and Arresé-Igor et al. (2009) stated that climate stress after planting affects the process of biological nitrogen fixation, with the inactivation of nitrogenase as first response to the increasing water deficit and, thus, the nodular activity is affected before photosynthesis. As it is shown in table 2, in the months after the legume planting there was a noticeable depression of rainfall in the experimental area (December: 34,8 mm; January: 45,7 mm); thus, it is not discarded that this was the reason for which the inoculated strains could not express their highest atmospheric dinitrogen fixation potential. On the other hand, according to Nápoles et al. (2008), symbiotic efficiency can be based not only on edaphoclimatic factors, but also on nodulation (Nod) factors, which may differ among strains.
However, it was not observed that the low nodulation value forcibly implied low nitrogen yield in the treatment inoculated with the strain HA2 (table 5). Gómez et al. (2009) observed, in plants of Cicer arietinum and Vigna unguiculata, under controlled conditions, that the P deficiency significantly decreased nodulation; while the specific nodular activity and root production increased in some treatments.
The variable NR (table 5) showed that all the inoculated treatments were statistically higher (p<0,05) than the non-inoculated control (2,00), which constitutes an evidence of the fact that the root infection capacity of the inoculated strains was higher than that of the natural rhizobia population of the experimental soil.
The treatments inoculated with the native strains JH2 (4,24) and HA3 (4,24) were statistically higher than the others. The results point, undoubtedly, towards nodule formation statistically higher (p<0,05) in the plants inoculated with the strains used in the trial; this could have occurred due to the fact that the soil showed slight acidity values, and most of the strains applied to C. plumieri combine the characteristics of being tolerant to acid pH and having a high symbiotic fixation potential (Bécquer, 2002). Nevertheless, in legume prospections conducted in the Cauto Valley, Cuba, Gómez et al. (2010) found that C. plumieri as well as the other collected legumes showed high values of natural nodulation. This contradicts, somehow, the report by Howieson et al. (2008) about the fact that the main cause of the non-optimal use of the dinitrogen symbiotic fixation is the competition of the introduced strains with the autochthonous ones, because the latter tend to have very high infective capacity, but low fixation capacity. On the other hand, the high nodulation rates shown by the strains from the central region did not always imply higher values in ADW and NY, possibly due to an imbalance of the nutrient exchange between the symbionts. In this regard, Martínez-Romero and Palacios (1990) considered that the root infection capacity of rhizobia, expressed by the common nod genes, does not always imply high symbiotic capacity, which in turn constitutes the expression of other genes, such as nif and fix.
These results were corroborated with the correlation and regression analyses. In the inoculated treatments, the dependence degree and the correlation of the variable NY with regards to NR (table 6) were weak (R2 = 0,04; r = 0,06), as well as the correlation of ADW with regards to NR (R2 = 0,26; r = 0,16). It is obvious that there was not a close relation among these variables in the host plants, for which it is assumed that the biochemical-biological relation between the root infection and the atmospheric nitrogen fixation capacities could vary in the studied strains, perhaps due to external factors such as the edaphoclimatic ones- or internal such as those related to genetics (Martínez-Romero et al., 1990). Fesenko et al. (1995) obtained similar results in the correlation of these variables by inoculating Pisum sativum with Rhizobium leguminosarum.
However, a higher regression and correlation degree (R2 =0,34; r = 0,59) was found between NY and ADW (table 7). These results, especially those of regression, should not be considered high from the statistical point of view (Antonio Sigarroa, personal communication), but according to the opinion of the authors of this work, they tend to corroborate the observations made by Bordeleau et al. (1977) in Medicago sativa; in this species there was a high correlation between nitrogen yield and aerial dry weight, for which they consider, as best strain selection criterion, the results in dry weight. Other authors, such as Hume et al. (1985) and Prévost (1987), reported that the nitrogen yield in Onobrychis viciifolia was highly correlated to plant growth. This phenomenon was also evaluated by Provorov and Tikhonovich (2003), who stated that the intensity of dinitrogen fixation is correlated to the energetic process of the plant, as is the photosynthetic activity, which provides not only the necessary energy for the nitrogenase reaction, but also the carbon structures for the assimilation of its products.
From the results, it is concluded that the native strains inoculated in C. plumieri did not show, in general, high symbiotic efficiency as compared to the non-inoculated control. These strains, with the exception of HA2 in NY, as well as JH2 and HA3 in NR, did not exert a highly positive effect on the plant, maybe because of the intervention of adverse climate factors, such as post-planting water stress. However, such strains have good infective potential, which is highly evaluated as competitiveness factor in the rhizosphere. A weak interrelation was observed between the variables NY and ADW with regards to NR, as well as a higher interrelation between the variable NY with regards to ADW.
These results should be taken into consideration for further studies related to the legume-rhyzobium symbiosis.
