Summary A laboratory study was carried out in order to measure the degradation rate of atrazine in 36 different soils taken from maize (Zea mays L.) fields in Belgium. These soils differed in their alrazme treatment histories. pH. organic matter content and type of organic and mineral fertili‐zation, Half‐lives of less than 10 days were found in more than 60% of the soils sampled. This rapid dissipation could be linked in a significant way to repeated pretreaiments with atrazine (intensive maize cropping) as well as to higher pH values (from neutral to alkaline), A low organic matter content might also be a factor explaining the rapid degradation of atrazine. but to a lesser extent than the first two factors. On the other hand mineral fertilization was shown to slow down atrazine dissipation. It is hypothesized that repeated treatments of atrazine cause a mi‐crobial adaptation to atrazine degradation and that acidic soil conditions impede this adaptation. To date, this is the first time that evidence for widespread accelerated degradation of atrazine has been reported.

The potential of soil microorganisms to mineralize atrazine was studied in soil samples collected from fields with various histories of atrazine application. In contrast to many previous studies, which showed no atrazine mineralization activity, all the tested soils mineralized atrazine regardless of their atrazine application history. However, the delay before mineralization and the variation in the subsequent mineralization rate were in agreement with the initial copy number of the atrazine dechlorinaze gene, and the proliferation rate of the degraders. Soils from corn fields, which had up to 100 copies of the atzA gene per gram of soil, had a lag period of 4-5 days before atrazine mineralization started, and final mineralization percentages ranged from 40% to 54%. However, soils from fields that were never amended with atrazine had much longer lag periods (more than 17 days), which decreased after enrichment of the degrader population with high concentrations of atrazine for 15 days. Generally the mineralization rate and the atzA gene copy number increased after the enrichment period. The atrazine mineralization potential was measured by PCR of genes from the atrazine mineralization pathway. Magnetic capture hybridization was the most efficient of the two tested methods for purifying target DNA of PCR inhibitors, without reducing the copy number of the required fragment. Nested PCR proved to be the most effective method for predicting the exact potential of the soil to mineralize the pollutant even without enrichment of a small population with the target genes. This method can complement microcosm studies and eliminate futile efforts when the potential to mineralize the pollutant does not exist in the soil.

The potential of a microbial consortium for treating waters contaminated with atrazine was considered. In conventional liquid culture, atrazine and its two dealkylated by-products were equally metabolised by the microbial consortium. Transient production of hydroxyatrazine was observed during atrazine catabolism, indicating that the catabolic pathway was similar to the one reported for isolates capable of atrazine mineralisation. This consortium was then inoculated to sediments sampled from an artificial recharge site. These sediments were contaminated by atrazine and diuron and exhibited only a slow endogenous herbicide dissipation. Inoculated microorganisms led to extensive atrazine degradation and survived for more than 10 weeks in the sediments. A rudimentary bioreactor was then setup using a soil core originating from the same recharge site. Degrading microorganisms rapidly colonised the core and expressed their degrading activity. The efficiency of the bioreactor was improved in the presence of spiked environmental surface waters. Atrazine degraders thus possibly benefited from the other organic sources in developing and expressing their activity. The microbial consortium did not initially exhibit the capacity to degrade diuron, which was used as reference compound. No change in this characteristic was detected throughout the study.

No abstract

No abstract