GeoChip-based analysis of the microbial community functional structures in simultaneous desulfurization and denitrification process

Hao Yu, Chuan Chen, Jincai Ma, Wenzong Liu, Jizhong Zhou, Duu‐Jong Lee, Nanqi Ren, Aijie WangJournal of Environmental Sciences 2014 · DOI

TL;DR

This study uses GeoChip, a functional gene microarray, to compare two microbial processes for removing sulfur and nitrogen from wastewater: the integrated simultaneous desulfurization and denitrification (ISDD) model and the denitrifying sulfide removal (DSR) model. The DSR model significantly outperformed ISDD in elemental sulfur recovery, and GeoChip analysis revealed distinct microbial communities and metabolic potentials, highlighting the importance of specific nitrate-reducing sulfide-oxidizing bacteria (NR-SOB) in the DSR model's success.

What Problem Does This Paper Address?

Industries often generate wastewater laden with high levels of sulfate and nitrate. Sulfate can be reduced to toxic and corrosive sulfide by sulfate-reducing bacteria (SRB), while nitrate is a common pollutant. Biological simultaneous desulfurization and denitrification (SDD) processes offer an environmentally friendly solution, aiming to convert sulfide to elemental sulfur (S⁰) using nitrate as an electron acceptor, and then reduce nitrate to nitrogen gas.

Two conceptual models for SDD exist: the DSR model, focusing on sulfide and nitrate removal, and the ISDD model, which integrates sulfate reduction and DSR within a single bioreactor. While previous studies showed DSR's effectiveness, the ISDD model struggled with low S⁰ conversion efficiency, and the underlying microbial mechanisms remained unclear. This paper aims to bridge this knowledge gap by comprehensively comparing the functional diversity, structure, and metabolic potential of microbial communities in both ISDD and DSR models using a high-throughput GeoChip array. This understanding is crucial for optimizing SDD processes, particularly for efficient S⁰ recovery.

Methods

The researchers employed a GeoChip-based metagenomic approach to analyze the microbial communities. This involved:

  • Reactor Setup: Two types of expanded granular sludge bed (EGSB) reactors were used: one for the ISDD model and two for the DSR model. These reactors were operated under controlled conditions with varying loading rates of sulfur and nitrogen.
  • Sample Collection: Sludge samples were collected from the reactors at different operational days and loading rate periods.
  • DNA Extraction and Amplification: Genomic DNA was extracted from the sludge samples using a freeze-grinding method. The DNA was then amplified using a modified Templiphi kit.
  • Labeling and Hybridization: Amplified DNA was labeled with Cy5 fluorescent dye and hybridized to the GeoChip 2.0 microarray. This chip contains probes for thousands of functional genes involved in various biogeochemical processes.
  • Data Analysis: Microarray data was scanned and analyzed to quantify the abundance of detected genes. Diversity indices (Simpson's reciprocal index and Shannon-Weaver index) were calculated. Principal Component Analysis (PCA) was used to visualize the differences in community structure between samples. Hierarchical clustering was employed to analyze the abundance of specific gene categories, such as dissimilatory sulfite reduction (dsr) genes and denitrification genes.

The novelty lies in the application of the high-throughput GeoChip array to directly compare the functional gene profiles of these two SDD models, providing a detailed insight into their microbial metabolic capabilities, which conventional methods like DGGE or SSCP could not achieve.

Datasets

The study utilized sludge samples collected from two distinct bioreactor models (ISDD and DSR) operated under controlled laboratory conditions. These samples served as the primary "datasets" for the GeoChip analysis. No external or publicly available datasets were used for training or evaluation. The paper does not introduce a new benchmark dataset.

Key Insights and Ideas

  • DSR Model Dominance: The DSR model significantly outperforms the ISDD model in S⁰ conversion efficiency, suggesting a more optimized microbial community structure for this specific process.
  • Distinct Microbial Fingerprints: GeoChip analysis revealed clear differences in the functional gene diversity, community structure, and abundance of genes between the ISDD and DSR models, indicating that the operational model is a primary driver of microbial community composition.
  • NR-SOB as Key Players: Specific functional genes targeting nitrate-reducing sulfide-oxidizing bacteria (NR-SOB), such as Thiobacillus denitrificans, Sulfurimonas denitrificans, and Paracoccus pantotrophus, were found to be more abundant in the DSR model and strongly correlated with its higher S⁰ conversion rates. This highlights their crucial role in efficient elemental sulfur recovery.
  • SRB Dynamics: While SRB (indicated by dsr genes) were more abundant in the ISDD model, their activity might be limited or inhibited, contributing to the low S⁰ conversion. In the DSR model, SRB abundance was lower, possibly due to the direct feeding of sulfide and competition for resources.
  • Loading Rate Impact: Functional diversity decreased in the ISDD model with increasing loading rates but increased in the DSR model, suggesting different sensitivities and adaptive capacities of the microbial communities to varying environmental conditions.

Key Results

  • S⁰ Conversion Efficiency: The DSR model achieved a significantly higher S⁰ conversion rate of 91.1% under high loading rates (0.9 kg S/(m³·day) for sulfide and 0.4 kg N/(m³·day) for nitrate), compared to the ISDD model's 25.6%.
  • Functional Diversity: Simpson's diversity index (1/D) and Shannon-Weaver index (H') showed that functional diversity decreased in the ISDD model (I1 → I2) but increased in the DSR model (D1 → D3) with elevated loading rates.
  • Gene Abundance: The abundance of dsr genes (involved in sulfate reduction) was lower in the DSR model compared to the ISDD model. Conversely, genes from NR-SOB were more abundant in the DSR model.
  • PCA Separation: Principal Component Analysis clearly separated the microbial communities of the ISDD and DSR models, indicating distinct functional community structures.

Limitations and Future Work

  • Correlation vs. Causation: While the study shows strong correlations between the abundance of certain NR-SOB genes and S⁰ conversion rates, it doesn't definitively prove causation. Further research could involve targeted manipulation of these microbial groups.
  • Inhibition Mechanisms: The paper suggests potential inhibition of SRB in the ISDD model due to sulfide accumulation or nitrate competition, but further investigation into these specific inhibitory mechanisms would be beneficial.
  • Enrichment Strategies: The authors suggest that enriching NR-SOB might be a more effective method to improve S⁰ recovery efficiency in the ISDD model. Future work could focus on developing strategies to achieve this enrichment.
  • Broader Application: The findings provide valuable insights for optimizing SDD processes. Future research could explore the application of these findings to full-scale industrial wastewater treatment systems.

Summary updated

BibTeX
@article{yu2014geochip,
  author = {Hao Yu and Chuan Chen and Jincai Ma and Wenzong Liu and Jizhong Zhou and Duu‐Jong Lee and Nanqi Ren and Aijie Wang},
  title = {{GeoChip-based analysis of the microbial community functional structures in simultaneous desulfurization and denitrification process}},
  journal = {Journal of Environmental Sciences},
  year = {2014},
  doi = {10.1016/j.jes.2014.05.001},
}
Figure