| dc.description.abstracteng | Lipolytic enzymes, including lipases and esterases, have a wide spectrum of applications in various industrial fields, due to their functional versatility. Despite an increasing number of lipolytic enzymes that have been reported recently, only a small portion were experimentally verified and studied. Until recently, the demand for new lipolytic enzymes, particularly the extremophilic ones that can survive in extreme environments, is still high. In this thesis, two compost samples (compost55 and compost76) at thermophilic stage were collected for exploring potential extremophilic lipolytic enzymes. Firstly, we investigated the microbial composition in the two composts. The 16S rRNA genes and transcripts data as well as metagenomic data revealed that Actinobacteria, Proteobacteria, Firmicutes, Bacteroidetes and Chloroflexi were dominant in the microbial consortia of both samples. The taxonomic disparity between compost55 and compost76 was mainly attributed to the different feedstock composition and composting conditions. In general, analysis of the metagenome data of compost55 and compost76 showed that both share similar metabolic pattern. Nevertheless, genes involved in lipid transport and metabolism as well as the categories fatty acids, lipids, and isoprenoids were more abundant in the compost55 than in compost76 community, suggesting a higher possibility to identify lipolytic genes in the compost55 metagenome. Four metagenomic libraries were prepared to probe the diversity of LEs from compost microbes by function-based approaches. Overall, 199 and 51 positive clones for compost55 and compost76, respectively, were recovered. The inserts of the recovered plasmids with a confirmed phenotype were sequenced, and putative lipolytic genes were identified. Clustering the amino acid sequences deduced from the corresponding lipolytic genes yielded 115 unique and full-length lipolytic enzymes. Then, the family assignment of these enzymes was conducted by analyzing the phylogenetic relationship and protein sequence similarity network according to an integral classification system. To the best of our knowledge, the system included all the reported lipolytic enzymes so far (46 families in total) to avoid artificial inflation of the number of families during classification. The functional screening- derived lipolytic enzymes were affiliated to 12 lipolytic families. Seven LEs were not assigned to any known families, indicating new branches of lipolytic families. Subsequently, the multiple sequence alignment further showed the catalytic residues and conserved motif in each family. For sequence-based screening, we have developed a searching and subsequent annotation strategy specific for putative lipolytic genes in metagenomes. The profile hidden Markov models-based searching methods was highly sensitive and accurate for lipolytic enzymes. With this sequence-based screening and annotation strategy, 4,157 and 2,234 putative lipolytic proteins (PLPs) were initially identified in the assembled metagenomes of compost55 and compost76, respectively. Among them, 1,234 (compost55) and 759 (compost76) of these were further assigned into 28 and 26 families, respectively. The enrichments were observed in families, such as VIII, hormone-sensitive lipase-like, patatin-like proteins, II, A85-Feruloyl-Esterase, Carb_B_Bacteria and homoserine transacetylase. Analysis of the phylogenetic origin of the assigned PLPs indicated a potential link between microbial taxa and their functional traits. By comparing the lipolytic hits identified by function-driven and sequence-based screening indicated that the activity-directed selection complements sequence-based selection, and vice versa.
In addition, comparative analysis of the distribution of lipolytic genes in metagenomes from various ecological niches were explored using the sequence-based approach developed in this thesis. The lipolytic family assignment (functional profile) and phylogenetic origin (taxonomic profile) of assigned PLPs for each sample were driven by the ecological factor (habitat). Moreover, the habitat also determined the conserved and distinctive microbial groups harboring the putative lipolytic genes. Finally, three lipolytic genes (est1, est2, and est56) belonging to different lipolytic families and showing low sequence identity to known lipolytic enzymes were selected for biochemical characterization. The three genes were heterologously expressed and characterized.
The gene product Est1 (est1) and Est2 (est2) are thermostable enzymes with optimal enzyme activities at 80 and 70 °C, respectively. The two enzymes, particularly Est2, were also proved as organic solvent tolerant. Est2 activity was significantly enhanced (two- to tenfold) in the presence of ethanol, methanol, isopropanol, and 1-propanol over a concentration range ranging from 6 to 30% (v/v). Moreover, Est2 exhibited short-term (2 h of incubation) and long-term (up to 26 days) stability towards various water-miscible organic solvents at different concentrations. Est2 also displayed high stability towards water-immiscible organic solvents of ethyl acetate, diethyl ether, and toluene. All of these features indicated that Est1 and Est2 possess application potential. The other lipolytic enzyme, Est56 (gene product of est56), was halotolerant. It exhibited high activity and stability towards up to 4 M NaCl and KCl. In the presence of NaCl, Est56 also displayed enhanced stability against denaturants including high temperatures (50 and 60 oC) and urea (2, 4, and 6 M). The amino acid composition of recently reported halotolerant lipolytic enzymes (40 in total) and halophilic enzymes was statistically compared to reveal the potential salt resistance mechanism for Est56. The results indicate that the haloadaptation of Est56 was mainly attributed to the high content of acidic residues (Asp and Glu, 12.2 %), the low content of lysine residues (0.7 %), and the excess of surface-exposed acidic residue. | de |