Résumé:
Granstar® 75 DF is one of the most used sulfonylurea herbicide on cereal crops; it contains 75
% of tribenuron-methyl (TBM). Assessing the changes on soil microbiota, particularly on the
most abundant bacterial groups, will be a useful approach to determine the impact of
Granstar® herbicide. For this purpose, Actinobacteria were analysed. They are known for their
abundance and aptitude to resist to xenobiotic substances. Using a selective medium for
Actinobacteria, forty two (42) strains were isolated from both untreated and Granstar® treated
soils. The number of isolates recovered from the treated agricultural soil 2 was fewer than that
isolated from the corresponding untreated soil, suggesting a negative effect of Granstar®
herbicide on Actinobacteria community. Among the isolates, eleven (11) resistant strains,
tolerating high doses of Granstar® ranging from 0.3 to 0.6% (v/v), were selected.
Morphological characterization of strains indicated their affiliation to Streptomyces genus.
The phylogenetic analysis, based on 16S rRNA genes sequencing, clustered five strains
(SRK3, SRK4, SRK9, SRK11 and SRK18) to Streptomyces rubrocyanodiastaticus sp. piger
type strain, with high bootstrap rates varying from 69 to 98%. While SRK13 and SRK14 were
closely related to Streptomyces lavendulocolor and SRK17 to Streptomyces africanus, with
high bootstrap values of 92; 89 and 76% respectively. SRK12 and SRK16 strains were
associated to four Streptomyces type strains (S. vinaceusdrappus, S. mutabilis, S. ghanaensis
and S. enissocaesilis) with a percentage of 70% of bootstrap. Strain SRK15 was related to
Streptomyces diastaticus sp. ardesiacus with high bootstrap level of 92%. The two most
resistant strains SRK12 and SRK17 did not grow on TBM as the sole carbon source, so they
were incapable of degrading it under these conditions. However, in the presence of glucose,
both tolerated its presence, but their growth was affected, compared to that on glucose, used
alone. At the final pH of 6.78-6.79, the TBM was already predominantly and chemically
degraded to an average rate of 76.18%. By decreasing the pH to 6.72-6.73, the TBM
dissipation increased significantly to an average of 89.36%. Furthermore, at lower acid values
of 4.38 and 4.23; obtained with a similar medium composition supplemented with glucose,
the disappearance of the TBM was further improved by 2.26 and 5.56% respectively with the
strains SRK12 and SRK17. Therefore, as the pH decreased, the TBM was more degraded
probably by acid hydrolysis. In all the tests carried out, as the TBM disappeared, two
compounds appeared in the medium and seemed to do not be affected by the acidity. On the
basis of these observations, it was suggested that the acid hydrolysis of the herbicide may be
the result of a co-metabolic process, due to the bacterial activity on another easily
metabolizable substrate. Such a process would be interesting to consider in the bioremediation
of alkaline soils polluted by TBM. The products of the TBM transformation were not
degraded chemically nor by the studied strains and had a negative effect on the growth of
these strains. This observation leaded to the need of more investigations on these compounds,
in particular their identification and the determination of their fate in the soil.
