Short communication. Influence of micronutrients on biological nitrogen fixation in bean (Phaseolus vulgaris L.) under greenhouse hydroponic culture conditions

Several microelements (Mo, Cu, Zn, Mn, B) had some degree of influence on nodule dry weight, plant dry weightand nitrogen content of Phaseolus vulgaris cv. Canellini inoculated with Rhizobium tropici CIAT899 under greenhousehydroponic culture conditions. Nodule number was not significantly affected. The absence of boron producedthe highest degree of inhibition. The absence of microelements affected uptake and assimilation of mineral nitrogento a lesser extent than biological nitrogen fixation. A similar effect of microelements was observed on Phaseolus vulgaris(cvs. Canellini and Mutin)/Rhizobium etli ISP23 symbiosis.

Palabras clave: Rhizobium, leguminosas, micronutrientes, nodulación, contenido en nitrógeno. 0.04 CuSO 4 .5H 2 O, 0.06 Na 2 MoO 4 .2H 2 O. To prevent the interference of possible trace oligoelements, the nutritive solution was prepared with distilled water with a low electrical conductivity (EC<4 µS cm -1 ). Leonard jars and the substrate were also washed with this water. Occasionally, in addition to inoculated treatments, a non-inoculated control treatment and another non-inoculated treatment supplemented with mineral nitrogen (3 applications of 400 mg ammonium nitrate per container) were included. At the end of the experiment, plants were collected and shoot dry weight, nodule number and nodule dry weight were determined. Shoot nitrogen content was estimated by the Kjeldahl method (Vincent, 1970). Statistical analysis of variance (ANOVA) and the SNK comparison of means test (p<0.05) were done using the Statistix programme (NH Analytical Software, USA). In a f irst experiment, the effect of the complete microelement solution on Phaseolus vulgaris (cv. Canellini)/Rhizobium tropici CIAT899 symbiosis was studied in greenhouse conditions. The results showed that in the absence of microelements, biological nitrogen fixation was strongly inhibited (Table 1). Shoot and nodule dry weights and shoot nitrogen content significantly decreased in the inoculated treatment when the nutritive solution did not contain microelements, although the number of nodules did not vary significantly. Although the absence of microelements also had a negative effect on dry weight and on the nitro-gen contents of the plant in the treatment not inoculated and fertilized with mineral nitrogen, this effect was less than in the inoculated treatment, suggesting that the assimilation of mineral nitrogen was affected to a lesser extent than biological fixation of nitrogen.
A similar inhibitory effect of nitrogen fixation was observed in greenhouse hydroponic culture conditions with the strain R. etli ISP23 and the cultivars Canellini and Mutin ( Table 2), suggesting that this inhibitory effect generally appears for other bean cultivars and other Rhizobium strains, although with different degrees of sensitivity.
The study of the effect of each microelement separately reveals that all the micronutrients affect, although to a different extent, the nodular mass and growth of bean cv. Canellini plants inoculated with Rhizobium tropici CIAT899. The biggest inhibition was caused by the absence of boron (Table 3).
Several works studied the effect of fertilization with micronutrients on vegetative growth, flowering and harvest yields of different legumes (Singh et al., 1992;Yanni, 1992;Singer et al., 1998), although most of them do not focus on the interaction of the microelements with biological fixation of nitrogen.
The data obtained in this work under controlled conditions indicated that boron was the microelement that most affected the bean/Rhizobium symbiotic process. This element seems to be involved in the synthesis and stability of the cell wall (Augsten and Eichorn, 1976)  and in the structure and function of membranes (Parr and Loughman, 1983). Lukaszewski et al. (1992) demonstrated that the symbiotic process rhizobium-legume requires boron both for indeterminate (Bolaños et al., 1994) and determinate nodule formation (Bonilla et al., 1997), and it has more recently been shown to affect nodulation of Frankia (Redondo-Nieto et al., 2000). It also plays an important role in mediating the surface interaction between bacteria and the plant cell and stabilizing the peribacteriodal membranes and the infection tube (Bolaños et al., 1996). The results obtained in this work show that the nodules formed when boron is limiting are smaller and weigh less than those of the control treatment. Ho-wever, we did not observe a significant reduction in the number of nodules formed. Bolaños et al. (1994) and Bonilla et al. (1997) also observed a significant reduction in nodule number but did not present any data on changes in this parameter, only showing those related with the reduced size and weight of nodules. Consequently, we consider that the absence of boron probably does not affect the number of root infection sites but hinders development of the nodules formed.
In spite of the clear results obtained under controlled conditions, in f ield experiments it was not possible to improve biological nitrogen fixation by foliar application of microelements (unpublished data). Although these results must be analyzed with caution, especially due to the different mobility of the oligoelements in the plant, they seem to suggest that limitation of microelements is not the main cause of poor nodulation and biological fixation presented by the Rhizobium/bean association in our field conditions. To elucidate the possible causes of this limitation, additional studies are required to study more in depth the interaction between boron and other nutrients such as calcium and phosphorus.