Characterization and monitoring of gas-phase bioreactors, accessing the black box through molecular biology tools

Resumen

In this dissertation document, the application of molecular biology techniques for the microbiological characterization and monitoring of gas phase bioreactors is studied. First of all, in the forth chapter of this study, the elimination of toluene by means of GPBs type biofilters was investigated. In summary, the presented results showed that biofiltration of toluene at relatively high loads was effective and robust for sustained toluene treatment. Differences arising from the nature of organic packing materials and how these materials are packed in the reactor lead to find different behaviors on the long run of the biofilters. Even if organic materials may naturally provide nutrients for biomass growth, nutrients supply proved to be useful for increasing reactors performance. Bed density linked to the watering regime was found to be a key factor regarding the pressure drop across the biolfiter. In this sense, coconut fiber and compost showed the lowest pressure drop on the long-run, while peat and pine leaves biofilters had episodes of high pressure drops. Additionally, all biofilters recovered quickly their previous toluene treatment capabilities after a shutdown. In general terms, all four packing materials showed a proper performance for toluene treatment showing high elimination capacities. In any case, coconut fiber and compost biofilters exhibited a better performance in terms of elimination capacity, pressure drop and long-term stability. Also, a change in the dominant microorganisms population from bacteria to fungi was observed when nutrients were added, increasing toluene elimination capacities. The microbial community in the coconut fiber reactor was scrutinized by using the cloning and sequencing and the culture-dependent approaches. Using the conventional culturing approach, four species of fungi were isolated on plates and tested positively as toluene degraders under toluene atmosphere. Also, the novel molecular approach was used for the characterization of the fungal and bacterial communities. Thus, the microbial diversity in a biofilter treating toluene was studied by creating two clone libraries potentially representing the most abundant species present in the reactor and some of them putatively involved in the toluene degradation process. To the author's best knowledge, this is the first research to retrieve fungal and bacterial rRNA gene sequences from a biofilter treating toluene and constructing two collections representative of the existing diversity. Around 20% of the bacterial sequences retrieved were affiliated with the toluene degrading bacterial genus Pseudomonas. The rest of bacterial species could not be directly related to toluene degradation. In the case of the fungal library, species of the genus Exophiala, well-known toluene degrading species, represented about 13% of the collection. Near 8% of the clone library was related to Acremonium species, which have been previously related to degradation of crude oil hydrocarbons. Candida oleophila, representing about 25 % of the clone library, has been previously demonstrated to degrade hydrocarbons from an oily sludge. In summary, a variety of microbial species could be involved in toluene or toluene by-products degradation in the gas phase bioreactor. In the fifth chapter of this dissertation document, the cloning and sequencing approach and the FISH technique were successfully used to characterize and monitor the population dynamics in a lab-scale biotrickling filter treating high loads of H2S. During the studied period, almost complete oxidation of H2S was achieved, recovering more than 90% of the sulfur as sulfate. Low pressure drops were observed along the operational period. The cloning and sequencing technique allowed the detection of most of the microbial populations existing in the bioreactor, including the SOB involved in the RSCs oxidation. This tool provided the phylogenetic information for selecting probes to apply the FISH technique to monitor the most relevant SOB. As a result, mixotrophic species of the genera Thiothrix and Thiomonas predominated along the operational period. Also species of Halothiobacillus, Thiobacillus and Sulfurimonas were found to be significant in the sulfur-oxidizing community. With the information provided by these techniques, it was possible to tentatively link the phylogenetic diversity with the operational conditions in the bioreactor. In the chapter 6, the culture-independent tool termed T-RFLP and culturing methods were used to characterize and monitor the populations in the lab-scale biotrickling filter, obtaining additional information about the sulfur-oxidizing community. In summary, four SOB strains and 2 potential SOB strains were isolated by means of a set of isolation methods covering different sulfur reduced compounds. These isolates were successfully identified and submitted to EMBL databases. Also, these isolates were T-RFLP studied to assign the phylogeny of some of the OTUs found in the biotrickling samples profiles. The T-RFLP technique was successfully used to study the community structure existing in the biotrickling filter treating hydrogen sulfide and to monitor it over time. It has been obtained that mixotrophic species like Thiothrix lakustris, Thiothrix sp. CT3 and Thiomonas intermedia have dominated the community of the reactor along the experimental period. In order to know the phylogeny of the most of the OTUs found in the profiles, this study included the in silico restriction of the clones from the two clone libraries previously constructed and the application of the T-RFLP to the most significant clones of the libraries. Moreover, the phylogeny of some of the still unknown OTUs was putatively assigned by means of the analysis of the T-RFs obtained and their comparison to databases. Thus, other species from the genera Alicyclobacillus, Pseudomonas and Xanthomonas were found to be probably part of the SOB community existing in the bioreactor. The T-RFLP technique has been also applied to study the spatial distribution of some populations in the reactor, obtaining that facultative anaerobic species gained in significance at the inlet part of the reactor, where the ratio oxygen to sulfide is lower. In general, the results concerning the evolution of the populations over time agreed with those obtained by the application of FISH technique. However, it has to be highlighted that with the T-RFLP technique, more species have been monitored than with the FISH technique and the community composition has been more elucidated. In the seventh chapter, the communities of three lab-scale bioreactors degrading thiosulfate and two GPBs degrading H2S were studied. In the lab-scale bioreactor degrading thiosulfate, on the one hand, the species Thiomonas sp. were found to be dominant in ones inoculated with compost from a biofilter degrading H2S. On the other hand, Halothiobacillus neapolitanus was encountered found to be predominant in the activated sludge-inoculated bioreactor. The H2S biofilter, operated at acidic pH, contained two main T-RFs, which were putatively identified as Thiomonas and Acidithiobacillus whereas the H2S biotrickling filter, operated at pH 7, contained several T-RFs related to the SOB genera Thiothrix, Thiomonas and Halothiobacillus. While both the H2S biofilter and biotrickling filter were initially inoculated with activated sludge from two local WWTPs and both successfully treated H2S, it was interesting to find such different community structures in the bioreactors. Further, the presence and expression of sqr and soxB genes was investigated in order to use them as biomarkers for RSCs oxidation. A set of degenerate primers, sqrCHF/sqrCHR, was successfully developed. The wide phylogenetic distribution of the primer set is a useful feature because sulfide-oxidizing abilities are widespread among various phylogenetic groups of SOB. Both soxB and sqr were detected in the DNA of the sulfur-oxidizing mixed cultures in the H2S reactors and in the thiosulfate batch reactors. The expression of soxB and sqr was studied using RT-PCR. sqr expression was detected in the H2S biofilter and in the H2S biotrickling filter; however no sqr expression was observed in the thiosulfate batch reactors. These results suggest that sqr is a potential biomarker for studying H2S oxidation. On the other hand, soxB expression was detected in all the reactors except for the activated sludge-inoculated thiosulfate reactor, dominated by Halothiobacillus neapolitanus. This result agrees with the hypothesis that H. neapolitanus uses the tetrathionate pathway, which does not involve the Sox enzymatic system. This result also suggests that Thiomonas was using the Sox system for thiosulfate degradation. Up to date, this is the first work suggesting that Thiomonas could use the PSO pathway for thiosulfate degradation.