Elucidating the Impact of Biosolids-Derived Antimicrobials on Denitrifying Microbial Community Function and Structure in Agricultural Soil

Thumbnail Image




Gunsch, Claudia K.

Journal Title

Journal ISSN

Volume Title

Repository Usage Stats



More than 50% of wastewater biosolids are applied to agricultural fields as fertilizer in the U.S. This technique has been used for decades as a widely accepted beneficial reclamation method for biosolids, which meet the established regulatory levels for nutrients, metals, and pathogens. A major drawback to land application is the potential environmental release of non-regulated organic contaminants, which accumulate in biosolids during the wastewater treatment process. Recent studies have been performed to identify and quantify the presence of emerging contaminants in biosolids, and others have investigated the effects of compounds already identified as `priority pollutants' and whose use is waning. However, there is limited research on the effect of emerging organic contaminants on soil microbial ecology and nutrient cycling. Because many of the compounds found in biosolids are specifically designed to elicit biological modifications (e.g., antimicrobials), there is a risk that these compounds will disrupt microbial soil functions, decrease soil productivity, and ultimately affect the long term viability of these ecosystems, resulting in unforeseen economic and social costs. Therefore, there is a clear need to characterize the effects of novel contaminants on soil health.

This dissertation was divided into three distinct parts examining the impacts of emerging organic contaminants on soil microbial ecology with increasing complexity to better reflect environmental conditions. To assess the ecological impacts, the functional endpoint of denitrification was selected because it provides a vital indication of soil health. Denitrifying bacteria play a critical role in this process, and thus, were used as indicator organisms for determining contaminant ecotoxicological potential. Furthermore, antimicrobial agents (a.k.a., bactericides or biocides) were selected as model contaminants because they are designed specifically to deactivate microorganisms, are heavily used in the U.S with over $1 billion in yearly sales, and have been measured in biosolids.

Overall, the objectives of this dissertation were to: 1) develop a rapid, high-throughput functional assay that measured denitrification inhibition for screening potential ecological impacts of biosolids-derived antimicrobial agents, 2) determine the potential effects of common and emerging biosolids-derived antimicrobial agents on denitrification by a model soil denitrifier, Paracoccus denitrificans PD1222, 3) examine the impacts of the most commonly used antimicrobial, triclosan (TCS), on wastewater treatment efficiency in bench scale sequencing batch reactors (SBRs) coupled with anaerobic digesters, 4) examine the impacts of biosolids aged and spiked with TCS on denitrification under simulated agricultural soil conditions, and 5) evaluate potential impacts of TCS in `traditional' biosolids on denitrification in agricultural soil under field conditions.

The first phase of research pertaining to Objectives 1 and 2 examined the baseline interactions between biosolids-derived antimicrobial agents and soil microbial ecology. However, to isolate the effect of an individual contaminant from the myriad of contaminants found in biosolids, there was a need for developing a rapid, high-throughput method to evaluate general ecotoxicity. In the first part of this dissertation, we developed a novel assay that measured denitrification inhibition in a model soil denitrifier, Paracoccus denitrificans Pd1222. Two common (TCS and triclocarban) and four emerging (2,4,5 trichlorophenol, 2-benzyl-4-chlorophenol, 2-chloro-4-phenylphenol, and bis(5-chloro-2-hydroxyphenyl)methane) antimicrobial agents found in biosolids were analyzed as model contaminants. Overall, the assay was reproducible and measured impacts on denitrification over three orders of magnitude exposure. The lowest observable adverse effect concentrations (LOAECs) were 1.04 μM for TCS, 3.17 μM for triclocarban, 0.372 μM for bis-(5-chloro-2-hydroxyphenyl)methane, 4.89 μM for 2-chloro-4-phenyl phenol, 45.7 μM for 2-benzyl-4-chorophenol, and 50.6 μM for 2,4,5-trichlorophenol. Compared with gene expression and cell viability based methods, the denitrification assay was more sensitive and resulted in lower LOAECs. Of the six compounds examined, four resulted in LOAECs that were below or within an order of magnitude of concentrations that were measured in the environment, indicating potential ecological impacts.

In the second part of the dissertation, the impacts of emerging contaminants were examined first under laboratory conditions mimicking wastewater treatment processes (Objective 3) and then agricultural fields (Objective 4). For this phase, TCS, which is the most widely used antimicrobial agent and identified in the first phase for potential ecological impacts, was used as the model contaminant. To mimic wastewater treatment processes, bench scale SBRs coupled with anaerobic digesters were set up and operated. The SBRS and digesters were seeded with activated and anaerobically digested sludge from the North Durham Water Reclamation Facility (NDWRF, Durham, NC). Reactors were fed synthetic wastewater with or without 0.73 &muM of TCS. Samples were taken periodically to monitor chemical oxygen demand (COD), ammonium (NH4+), nitrate (NO3-), nitrite (NO2-), total suspended solids (TSS), volatile suspended solids (VSS), dissolved oxygen (DO), and phosphate (PO43-) and pH. In addition, biomass samples were collected for DNA extraction and microbial community analysis using terminal restriction fragment length polymorphism (T-RFLP) of 16S SSU rDNA. Methane production was also monitored for the anaerobic digesters. In addition, the final digested biosolids that were generated from the SBRs fed with and without TCS were analyzed for TCS concentration, TSS, VSS, TKN, phosphorus (as P2O5), potassium (as K2O), and pH. Overall, biological processes associated with nitrogen removal (nitrification and denitrification), were impacted by TCS entering the SBRs regardless of the starting microbial community. Both of the SBRs that were not receiving TCS reached steady-state at greater than 92% NH4+, removal within the first week of operation, whereas the SBRs receiving TCS took 42 and 63 days to reach steady-state removal at that level. However, while NH4+ removal was temporarily inhibited, elevated levels of NO3- and NO2- in the effluent of the TCS fed SBRs, suggested longer-term impacts on nitrite oxidizing bacteria (NOB) and denitrifiers. After Day 58, the NO3- effluent concentration for the SBRs receiving TCS was 3.9 ± 0.16 mg/L, which was 2.4 times greater than the NO3- effluent of the SBRs not receiving TCS (1.7 ± 0.08 mg/L). Similarly, after Day 58, the NO2- effluent of the SBRs receiving TCS reached a steady-state concentration of 8.7 ± 0.75 mg/L. The mean NO2- concentration in the controls after Day 58 was 7.7 times lower at 1.1 ± 0.78 mg/L, but was still trending towards 0 when the reactors were stopped. No inhibition was observed for COD and PO43- removal. In addition, non-metric multidimensional scaling (NMS) ordination analysis showed that the microbial communities between SBRS fed with and without TCS were similar on Day 0, but increased in difference to Day 41, around when the major changes in nitrification were observed. After a slight increase in similarity between the control and TCS SBR microbial communities on Day 41, the communities increased in difference to Day 63.

To mimic agricultural field conditions, containers of soil were amended with the biosolids generated from the SBRs. The containers were maintained in a growth-chamber to simulate field lighting and watering conditions. Three biosolids treatments were examined: 1) biosolids generated from the SBRs not fed TCS, but that still had low backgrounds of TCS (a.k.a., Control Biosolids); 2) biosolids generated from the SBRs fed with TCS (a.k.a., Aged TCS Biosolids); and 3) biosolids that were generated by the SBRs not fed TCS, but spiked with TCS 24 h before application (a.k.a., Spiked TCS Biosolids). Alfalfa was planted in half of the containers receiving the Control and Aged TCS Biosolids to assess differences due to vegetation. To assess the overall ecotoxicity of biosolids aged and spiked with TCS, the function, abundance, and diversity of the soil denitrifying communities were examined. The impacts on total bacteria abundance and diversity were also examined for comparison. Specifically, the denitrifying enzyme activity (DEA) assay was used to measure functional impacts, quantitative polymerase chain reaction (qPCR) was used to measure impacts on abundance, and T-RFLP was used to measure impacts on diversity. Correlations between these methods were also examined for possible interactions between denitrifier function and community structure and to provide insight into targets of inhibition. Lastly, a denitrification inhibition score was developed to quantify global impacts of TCS on denitrification. The containers with plants that received biosolids aged with and spiked with TCS showed potential long-term inhibition based on measurement of soil denitrification at 26.9 ± 4.6 μg/kg and 68.6 ± 26.9 μg/kg of TCS, respectively. Denitrifier abundance and diversity, however, were more sensitive to TCS in biosolids and inhibition was observed throughout the experiment, with maximum inhibition on Days 7 and 28. Inhibition of denitrifier abundance and diversity was observed at TCS concentrations as low as 17.9 ± 1.93 μg/L, which was about 10 to 3000 times lower than concentrations reported by other studies that showed impacts on other functional endpoints (i.e., respiration, phosphatase activity, NO3- and NO2- production, and Cy17 stress biomarker abundance), even after taking pH into account. Five significant correlations were developed, three of which related qPCR and the DEA assay, or abundance and activity. However, the analyses that were correlated did not yield the same results as far as significant inhibition in the presence of TCS. Thus, while the results suggested some relatedness between activity, abundance, and diversity, the results generally support the use of multiple methods to determine the ecotoxicity of biosolids-derived organic contaminants. As a result, a denitrification inhibition score was developed that took into account all three methods to determine the overall ecotoxicity of TCS in biosolids. Overall, the denitrification inhibition score showed that denitrification was inhibited by both biosolids that were aged and spiked with TCS over the extent of the 84 day experiment, but maximum inhibition occurred after a week to about a month. While the denitrification inhibition score indicated that the TCS in the biosolids aged with TCS was less bioavailable than in the spiked biosolids, the impacts of the aged and spiked biosolids could have also been due to differences in TCS concentrations.

Objective 5 consisted of a long-term soil sampling campaign on four agricultural fields receiving Class B municipal biosolids. Soil samples were taken before and after biosolids application and were analyzed to elucidate potential impacts of TCS in the biosolids on denitrification. Again, to assess the overall impacts of TCS on the soil denitrifying community, the DEA assay, qPCR, and T-RFLP were used to measure impacts on function, abundance, and diversity, respectively. Similar to Objective 4, the analysis included an examination of potential correlations between denitrifying community structure and function, and quantification of global impacts using the denitrification inhibition score. As expected, the results in this pilot-study reflected the complexity of the system that was analyzed and many more samples, which account for variables including, but not limited to soil characteristics, biosolids characteristics, biosolids application rates, and chemical composition and quantities, would be needed to show any statistically significant differences. Nevertheless, several key results were obtained. Again potential long-term inhibition of denitrification was observed using the DEA assay, however the effects of exhaustion of resources, such as NO3-, or significant changes in the local environment were suspected, but could not be verified. Inhibition was also observed for denitrifier abundance, but little to no inhibition was observed when examining the relative number of denitrifying species. Thus, while the abundance of denitrifiers was reduced, and denitrification was eventually depressed, the number of species in the soil remained constant. When looking at the denitrification inhibition score, which took all three measurements into account, increased inhibition over time was observed with the exception of the measurements on Days 30 and 103, which indicated overall, but weak inhibition of denitrification by the application of biosolids. NMS ordinations showed no correlation between the shift in denitrifying microbial community and TCS. Because of the complexity of the soil and biosolids and because of the myriad of contaminants likely in the biosolids, the results may not be significant and a more in-depth study was recommended.

Overall, the results presented in this dissertation provide a systematic evaluation of the effects of biosolids-derived TCS on agricultural soil microbial ecology. First, it was demonstrated that statistically significant inhibition of denitrification could be used as a potential indicator of biosolids-derived emerging organic contaminant ecotoxicity. The denitrification assay that was developed was then used to analyze ecotoxicological potential of six emerging biosolids-derived antimicrobial agents, and found inhibition of denitrification at environmentally relevant concentrations. The most widely used antimicrobial agent, TCS, was further shown to inhibit wastewater treatment processes, as well as, denitrification in simulated agricultural conditions after being aged with and spiked into biosolids. In addition, evidence showing potential inhibition of denitrification by TCS in `traditional' biosolids under field conditions was also obtained. Based on these results, this dissertation asserts that biosolids-derived emerging organic contaminants pose a potential risk to agricultural soil microbial ecology and overall soil health. Future studies, however, are needed to examine the impacts of other contaminants that might be flagged with the assay developed in this dissertation under more complex conditions mimicking the environment. Furthermore, other research is needed to examine the role microbial communities play in the bioavailability of emerging contaminants, especially TCS, and a more extensive, in-depth study is needed to characterize the individual impacts of emerging contaminants on soil microbial communities under field conditions.





Holzem, Ryan Michael (2014). Elucidating the Impact of Biosolids-Derived Antimicrobials on Denitrifying Microbial Community Function and Structure in Agricultural Soil. Dissertation, Duke University. Retrieved from https://hdl.handle.net/10161/8674.


Dukes student scholarship is made available to the public using a Creative Commons Attribution / Non-commercial / No derivative (CC-BY-NC-ND) license.