Field Trial I Fate and Impact of an Allochtonous GMM

Upon authorization from the national competent authority, the three GM rhizobia tested in microcosms were released as field inoculants in the presence of their pea host plant (Pisum sativum) by liquid inoculation at about 5x106 cfu per seed. The release took place in soil at the agricultural experimental station of the University of Padova.5 Bacteria were monitored in soil, rhizosphere and nodules. The parental strain 1003, from which the GMMs were derived had been isolated in northern Europe and propagated for a considerable time in the laboratory environment. It was therefore allochtonous with respect to the release site. This condition along with a possible fitness reduction due to extended laboratory maintenance, may have contributed to its poor persistence after inoculation. Although both the wild type and the three GMMs had shown good nodulation and nitrogen fixation pheno-type on different cultivars of peas in preliminary pot trials, analysis of their persistence in the field revealed a decrease from 106 to 103 cfu g-1 within the first 30 days after sowing.

Fig.10.2. Survival (A-B-C) and marker stability (D-E-F) of Rhizobium strains in inoculant packages. A, D, liquid medium; B, E, sterile vermiculite; C, F, unsterilized peat. Symbols: ■, strain 1003; O, strain 1110; □, strain 1111; strain 1112. Each point represents the mean of data obtained from three different packages. Standard deviations were within the ranges ±4% to ±41% for liquid, ±8% to ±59% for vermiculite, and ± 21% to ±68% for peat samples (redrawn with modifications from Corich V, Bosco F, Giacomini A et al. J Appl Bacteriol 1996; 81:319-328.).

Fig.10.2. Survival (A-B-C) and marker stability (D-E-F) of Rhizobium strains in inoculant packages. A, D, liquid medium; B, E, sterile vermiculite; C, F, unsterilized peat. Symbols: ■, strain 1003; O, strain 1110; □, strain 1111; strain 1112. Each point represents the mean of data obtained from three different packages. Standard deviations were within the ranges ±4% to ±41% for liquid, ±8% to ±59% for vermiculite, and ± 21% to ±68% for peat samples (redrawn with modifications from Corich V, Bosco F, Giacomini A et al. J Appl Bacteriol 1996; 81:319-328.).

Nodulation was correspondingly low for all introduced strains, less than 1.4% out of a sampled population of 1229 nodules. The detection limit inherent to the markers used was in the order of 102 cfu g-1 soil dry weight.

In parallel, the possible impact on resident biota was measured by plate counts of total aerobic culturable bacteria, spore forming bacteria, microfungi, streptomycetes, fluorescent pseudomonads, and total rhizobia. No significant numerical perturbations were detected. Other parameters were measured, including respiration and denitrification by assaying CO2 and N2O evolution from soil, and VA mycorrhizal infection of pea plants. No differences were found between control and GMM plots. The persistence of released strains was assessed for a period of up to 2.5 years after inoculation by host plant nodulation and PCR amplification to target the marker gene cassettes in total DNA samples extracted from soil. No trace of viable (nodulating) GMMs or of their DNA (PCR) was found.

Was this article helpful?

0 0

Post a comment