Detection and quantification of nitrifying bacteria from South African biological nutrient removal plants

Abstract

Nitrification is a crucial step in biological nutrient removal (BNR) processes, mostly carried out by a group of nitrifying bacteria which includes ammonia-oxidising bacteria (AOB) and nitrite-oxidising bacteria (NOB). Nitrification failure has proven to be a common operational problem in full-scale wastewater treatment plants (WWTP) since nitrifying bacteria are very sensitive to sudden changes in environmental or plant operating conditions. The current investigation was carried out to advance our understanding of the distribution of nitrifying bacterial populations and their performance at three different BNR plants in KwaZulu-Natal, South Africa. The latest molecular techniques such as fluorescent in situ hybridisation (FISH)-confocal scanning laser microscopy (CSLM), polymerase chain reaction (PCR) and real-time quantitative PCR (Q-PCR) were applied to detect and quantify nitrifying bacteria. When using FISH to target the nitrifying population, it necessitated optimising pre-treatment protocols of the samples to improve accuracy during quantification. Sonication was found to be the superior method of dispersion based on the least disruption of nitrifier cell integrity, irrespective of the sludge type. The effect of plant configurations and wastewater characteristics on the distribution of the nitrifying bacterial population and subsequently on the nitrification performance was evaluated using FISH and PCR. FISH results revealed the dominance of Nitrosomonas (AOB), Nitrobacter (NOB) and Nitrospira (NOB) for all BNR plants. The 16S rRNA analysis of PCR products using genus-specific primers, revealed the presence of more than one species of the same group at these plants. Nitrosomonas spp. including Nitrosomonas halophila, Nitrosomonas eutropha, Nitrosomonas europaea, Nitrosomonas aestuarii and an unidentified Nitrosomonas spp. were found to dominate among the AOB and Nitrobacter vulgaris, Nitrobacter alkalicus, Nitrobacter hamburgensis and an unidentified Nitrobacter spp. were the dominant species for NOB. Among these species, Nitrosomonas aestuarii, Nitrosomonas europaea, Nitrobacter hamburgensis were detected only from the industrial wastewater samples. The efficiency of two commonly used techniques viz., FISH and Q-PCR for the detection of nitrifiers from WWTP were also studied and compared, specifically targeting Nitrobacter sp. Even though there were slight variations in the quantification results, changes in the Nitrobacter community at these plants were consistent for both FISH and Q-PCR results. Both techniques have their own limitations and advantages. This study has helped to add to the platform of understanding the distribution and activity of nitrifying bacteria by correlating population dynamics with the operational parameters at full-scale level. The observations made in this study will assist researchers and engineers to minimise future nitrification failure at full-scale BNR plants. This study also confirmed the highly complex activities of wastewater treatment processes, which is dependant on a number of factors. Specific AOB or NOB predominant in wastewater rather suggests that the wastewater type and characteristics may contribute to significantly different microbial environments. Among the AOB, Nitrosomonas dominated at all BNR plants throughout the study period and for NOB both Nitrobacter and Nitrospira were found in significant numbers but their dominance varied across the plants. These dissimilar, distinct distribution patterns could be attributed to their environment which in turn impacted on the nitrification performance of the system. It was also noted that the co-existence of more than one group of these communities at the same plant could help the plant escape complete functional failures such as nitrification, due to sudden changes in temperature and substrate concentrations, as this function can be performed by different groups. Although it would have been meritorious to conduct a nitrogen balance in this study, this was not possible since the research focused on full-scale systems.