Comparative survey of the relative impact of mRNA features on local ribosome profiling read density

Ribosome profiling (Ribo-seq), a promising technology for exploring ribosome decoding rates, is characterized by the presence of infrequent high peaks in ribosome footprint density and by long alignment gaps. Here, to reduce the impact of data heterogeneity we introduce a simple normalization method, Ribo-seq Unit Step Transformation (RUST). RUST is robust and outperforms other normalization techniques in the presence of heterogeneous noise. We illustrate how RUST can be used for identifying mRNA sequence features that affect ribosome footprint densities globally. We show that a few parameters extracted with RUST are sufficient for predicting experimental densities with high accuracy. Importantly the application of RUST to 30 publicly available Ribo-seq data sets revealed a substantial variation in sequence determinants of ribosome footprint frequencies, questioning the reliability of Ribo-seq as an accurate representation of local ribosome densities without prior quality control. This emphasizes our incomplete understanding of how protocol parameters affect ribosome footprint densities.

Comparative genome analysis of human bifidobacteria with commercial potential

This thesis describes two newly sequenced B. longum subsp. longum genomes and subsequent comparative analysis with publicly available B. longum subsp. longum, B. longum subsp. infantis and B. longum subsp. suis genomes (Chapter 2). The acquired data revealed a closed pan-genome for this bifidobacterial species and furthermore facilitated the definition of the B. longum core genome. The comparative analysis also highlights differences in the potential metabolic abilities of all three sub-species. Interestingly, phylogenetic analysis of the B. longum core genome indicated the existence of a novel B. longum subspecies. Characterisation of restriction-modification systems from two B. longum subsp. longum strains is described in Chapter 3. These defence mechanisms limit the uptake of genetic material, which was successfully demonstrated for some of the identified systems. When these systems were by-passed by methylation of DNA prior to the transformation procedure, the resulting transformation efficiency of both B. longum subsp. longum strains was increased to a level that allowed for the generation of mutants via homologous recombination. Arabinoxylan metabolism by B. longum subsp. longum NCIMB 8809 was investigated in Chapter 4 of this thesis. Transcriptome analysis allowed the identification of a number of genes involved in the degradation, uptake and utilisation of arabinoxylan. Biochemical analysis revealed that three of the identified genes encode arabinofuranosidase activity. Phenotypic assessment of a number of insertion mutants in genes identified by the transcriptome analysis revealed the essential role of two of these enzymes in arabinoxylan metabolism, and a third enzyme in the metabolism of debranched arabinan. Furthermore, this investigation revealed that B. longum subsp. longum NCIMB 8809 does not completely degrade arabinoxylan, but utilises the arabinose substitutions only, while leaving the xylan backbone untouched.Finally, Chapter 5 outlines that B. longum subsp. longum NCIMB 8809 is capable of removing ferulic and p-coumaric acid substitutions that originate from arabinoxylan. Analysis of the genome sequence led to the identification of a candidate gene for this activity, which was subsequently cloned and expressed in E. coli. Biochemical analysis revealed that the enzyme, designated here as FaeA, is indeed capable of releasing both ferulic and p-coumaric acid from arabinoxylan. Furthermore, it is shown that a derivative of B. longum subsp. longum NCIMB 8809 carrying an insertion mutation in faeA had lost the ability to release ferulic and p-coumaric acid from arabinoxylan, and that growth of this mutant strain is negatively affected when cultivated on growth-limiting levels of arabinoxylan.

Genome-scale analyses of health-promoting bacteria: probiogenomics

The human body is colonized by an enormous population of bacteria (microbiota) that provides the host with coding capacity and metabolic activities. Among the human gut microbiota are health-promoting indigenous species (probiotic bacteria) that are commonly consumed as live dietary supplements. Recent genomics-based studies (probiogenomics) are starting to provide insights into how probiotic bacteria sense and adapt to the gastrointestinal tract environment. In this Review, we discuss the application of probiogenomics in the elucidation of the molecular basis of probiosis using the well-recognized model probiotic bacteria genera Bifidobacterium and Lactobacillus as examples.

Identification of RNA editing in the human exome and development of DARNED DatabaseSF

RNA editing is a biological phenomena that alters nascent RNA transcripts by insertion, deletion and/or substitution of one or a few nucleotides. It is ubiquitous in all kingdoms of life and in viruses. The predominant editing event in organisms with a developed central nervous system is Adenosine to Inosine deamination. Inosine is recognized as Guanosine by the translational machinery and reverse-transcriptase. In primates, RNA editing occurs frequently in transcripts from repetitive regions of the genome. In humans, more than 500,000 editing instances have been identified, by applying computational pipelines on available ESTs and high-throughput sequencing data, and by using chemical methods. However, the functions of only a small number of cases have been studied thoroughly. RNA editing instances have been found to have roles in peptide variants synthesis by non-synonymous codon substitutions, transcript variants by alterations in splicing sites and gene silencing by miRNAs sequence modifications. We established the Database of RNA EDiting (DARNED) to accommo-date the reference genomic coordinates of substitution editing in human, mouse and fly transcripts from published literatures, with additional information on edited genomic coordinates collected from various databases e.g. UCSC, NCBI. DARNED contains mostly Adenosine to Inosine editing and allows searches based on genomic region, gene ID, and user provided sequence. The Database is accessible at http://darned.ucc.ie RNA editing instances in coding region are likely to result in recoding in protein synthesis. This encouraged me to focus my research on the occurrences of RNA editing specific CDS and non-Alu exonic regions. By applying various filters on discrepancies between available ESTs and their corresponding reference genomic sequences, putative RNA editing candidates were identified. High-throughput sequencing was used to validate these candidates. All predicted coordinates appeared to be either SNPs or unedited.

Ruthenibacterium lactatiformans gen. nov., sp.nov., an anaerobic, lactate-producing member of the family Ruminococcaceae isolated from human faeces

Two novel strains of Gram-stain-negative, rod-shaped, obligately anaerobic, non-spore-forming, non-motile bacteria were isolated from the faeces of healthy human subjects. The strains, designated as 585-1T and 668, were characterized by mesophilic fermentative metabolism, production of d-lactic acid, succinic acid and acetic acid as end products of d-glucose fermentation, prevalence of C18 : 1 ω9, C18 : 1 ω9 aldehyde, C16 : 0 and C16 : 1 ω7c fatty acids, presence of glycine, glutamic acid, lysine, alanine and aspartic acid in the petidoglycan peptide moiety and lack of respiratory quinones. Whole genome sequencing revealed the DNA G+C content was 56.4–56.6 mol%. The complete 16S rRNA gene sequences of the two strains shared 91.7/91.6 % similarity with Anaerofilum pentosovorans FaeT, 91.3/91.2 % with Gemmiger formicilis ATCC 27749T and 88.9/88.8 % with Faecalibacterium prausnitzii ATCC 27768T. On the basis of chemotaxonomic and genomic properties it was concluded that the strains represent a novel species in a new genus within the family Ruminococcaceae , for which the name Ruthenibacterium lactatiformans gen. nov., sp. nov. is proposed. The type strain of Ruthenibacterium lactatiformans is 585-1T (=DSM 100348T=VKM B-2901T).

Microbially derived bioactive peptides to improve human health

This thesis describes a study of various methods to produce bioactive peptides. Initially, the generation of anti-Cronobacter spp. peptides by fermentation of milk protein is described. Lactobacillus johnsonii DPC6026 was used to generate two previously described antimicrobial peptides. Phenotypic analysis indicated unsatisfactory casein hydrolysis. The genome of the strain was sequenced and annotated. Results showed a number of unique features present, most notably a large symmetrical inversion of approximately 750kb in comparison with the human isolate L. johnsonii NCC 533. The data suggest significant genetic diversity and intra-species genomic rearrangements within the L. johnsonii spp.. Cronobacter spp. have emerged as pathogens of concern to the powdered infant formula industry. Chapters 3 and 4 of this thesis describe novel methods to generate two antimicrobial peptides, Caseicin A and B. In Chapter 3 a bank of Bacillus strains was generated and investigated for caseicin production. Following casein hydrolysis by specific B. cereus and B. thuringiensis strains the peptides of interest were generated. Chapter 4 describes a sterile enzymatic method to generate peptides from casein. Bioinformatic tools were used to predict enzymes capable of liberating caseicin peptides from casein. Hydrolysates were generated using suitable enzymes, examined and some were found to produce peptides with activity against Cronobacter spp.. This study establishes a potential industrial-grade method to generate antimicrobial peptides. Administration of GLP-1 leads to improved glycaemic control in diabetes patients. Generation of a recombinant lactic acid bacteria capable of producing a GLP-1 analogue is described in Chapter 5. In-vivo analysis confirmed insulinotropic activity. The results illustrate a method using bacteriocin producing cellular machinery to generate bioactive peptides. This thesis describes the generation of bioactive peptides by bacterial fermentation, tailored enzymatic hydrolysis and recombinant bacterial methods. The techniques described contribute to bioactive peptide research with regards novel methods of production and industrial scale-up.