No description available.

<p> This thesis proposed to analyse the recombination, complementation and biosynthetic implications of a series of hitherto unstudied carotenoidless mutant strains of Neurospora crassa and to confirm the reports of previous authors through analysis of a number of their mutant strains. A new selective technique permitted the fine structure analysis of the locus. Complementation studies with an extensive range of mutants including several recently discovered phenotypes permitted the resolution of new cistronic limits within the locus. A speculative model of the gene products was proposed to embrace the recombination, complementation and biochemical paradigms.</p> / Thesis / Doctor of Philosophy (PhD)

No description available.

No description available.

Lignin is a highly abundant aromatic biopolymer deposited during the final stages of secondary cell wall formation in plants and it constitutes a substantial proportion of the dry weight of woody plant stems. Lignin contributes structural support to xylem cell walls and hydrophobisity to water-conducting vessels and forms a defence mechanism against pathogen invasion. Although being an essential part of normal plant cell development, lignin content and composition are targets for tree improvement, because residual lignin in paper pulp has negative effects on paper quality and lignin therefore has to be removed using treatments that are expensive and often detrimental to the environment. At present, little is known about the amount of allelic diversity in lignin biosynthetic genes and whether such diversity may be associated with variation in lignin content and composition. However, the identification of alleles associated with desirable lignin phenotypes is dependent on a detailed understanding of the molecular evolution and population genetics of these genes. This M.Sc. study was aimed at analysing nucleotide and allelic diversity in two lignin biosynthetic genes of Eucalyptus trees. Additionally, the study aimed to develop single nucleotide polymorphism (SNP) markers that could be used to assay allelic diversity for these genes in populations of two target species, E. grandis and E. smithii. Orthologues of the tobacco LIM-domain1 (NtLIM1) transcription factor gene involved in the regulation of lignin biosynthesis were isolated from E. grandis and E. smithii. Approximately 3 kb of genomic sequence including the promoter and full-length gene regions were isolated for the two orthologues, respectively labeled EgrLIM1 and EsLIM1. The predicted amino acid sequences of EgrLIM1 and EsLIM1 were 99.4% identical to each other and indicated that LIM1 is a small protein of only 188 residues in eucalypt trees and has a predicted molecular weight of 21.0 kDa. Quantitative, real-time RT-PCR analysis confirmed the expression of LIM1 in wood-forming tissues undergoing lignification. Ten putative cis-regulatory elements were observed in the promoter regions of EgrLIM1 and EsLIM1including a GA-dinucleotide microsatellite that appears to be specific to LIM1 promoters of Eucalyptus tree species. The full-length LIM1 gene sequences could subsequently be used in the assessment of nucleotide and allelic diversity, together with the full-length CAD2 sequences that were already available in the public domain. The level of nucleotide and allelic diversity and the distribution and decay of linkage disequilibrium (LD) were surveyed in 5’ and 3’ derived gene fragments of CAD2 and LIM1 obtained from 20 E. grandis and 20 E. smithii individuals. Each gene displayed a unique genetic diversity profile, but for the most part, nucleotide diversity (π) was estimated at approximately 0.0010 except for the E. grandis LIM1 gene where π lower than 0.0040 was observed. Generally, except for the high amounts of LD observed in the CAD2 gene of E. grandis (> 2.5 kb), LD decayed within 500 bp. A large number (13 to 45) of SNP sites (defined as single nucleotide changes with minor allele frequencies of at least 0.10 in each species) were observed in each gene of each species. The high SNP density (ranging from one per 45 to one per 155 bp) observed in the two genes facilitated the efficient development of SNP markers to be used in future aspects of LD mapping, association genetics and marker-assisted breeding. The allele sequences obtained for the CAD2 and LIM1 genes were used as templates for the development of SNP marker panels (a series of six or seven SNP markers analysed together) for the analysis (tagging) of SNP haplotype diversity in species-wide reference populations (100 E. grandis and 137E. smithii individuals) of the two species. Each tag SNP was assayed using a single base extension assay and capillary gel electrophoresis. High polymorphism information content (average PIC of 0.836) was observed for the SNP marker panels. Four SNPs in the CAD2 and two in the LIM1 genes were found to be polymorphic in E. grandis and E. smithii (i.e. trans-specific SNPs), suggesting a possible ancestral origin for these polymorphisms. Assessment of candidate gene variation in the genomes of forest trees is of importance to ultimately be able to predict the amount and structure of nucleotide diversity available for the future design of SNP assays at the whole-genome level. Such assays will be useful to study differentiation among tree species and populations, to associate nucleotide polymorphisms with desirable phenotypes and to increase the efficiency of tree improvement approaches. / Dissertation (MSc (Genetics))--University of Pretoria, 2009. / Genetics / unrestricted

Identifying biosynthetic gene clusters from genomic data is challenging, with many in-silico tools suffering from a high rediscovery rate due to their dependence on rule-based algorithms. Next generation sequencing has provided an abundance of genomic information, and it has been hypothesized that there are many undiscovered biosynthetic gene clusters within this dataset. Here, we aim to develop a machine learning tool, ML-Miner, that infers patterns that describe a biosynthetic gene cluster in an unbiased manner and, as such, enables the identification of new biosynthetic gene clusters from genomic data. To solve this challenging problem, we define a simpler one to predict the class of a known BGC. Specifically, ML-Miner receives as input the concatenation of sequences that are known or believed to be part of a biosynthetic gene cluster. Its task is to identify which class it belongs, i.e. NPRS, PKS terpene and RiPPs. ML-Miner is a machine learning tool that uses Natural Language Processing, dimensionality reduction, and supervised learning to identify novel biosynthetic gene clusters. BioVec is a biological word embedding that we use to transform protein sequences from the highly curated MIBiG database of characterized biosynthetic gene clusters into their respective continuous distributed vector representations. Because the resulting protein vectors are of high dimensionality, a supervised Uniform Manifold and Approximation algorithm was employed to transform the high dimensional vectors into a robust lower-dimensional representation, as evaluated by Silhouette analysis, Hopkins’ statistic, and trustworthiness analysis. The density-Based Spatial Clustering of Applications and Noise algorithm showed that the clusters identified from the low dimensional datasets mapped to biosynthetic gene cluster types, defined with high accuracy in the MIBiG database. A random forest classifier was then trained and evaluated using the low dimensional vectors. It was shown to classify each biosynthetic gene cluster from the MIBiG database with excellent performance metrics. Finally, the model's ability to generalize was evaluated using biosynthetic gene clusters from the antiSMASH dataset, an uncurated database containing uncharacterized biosynthetic gene clusters. The performance metrics were high, with a balanced accuracy of ~85%. After a hyperparameter search, the balanced accuracy rose to ~90%. This suggests that ML-Miner is a robust machine learning pipeline that can be used to identify novel biosynthetic gene clusters. Future development of a confidence score for classification and a workflow for processing bacterial genomes into gene clusters will significantly improve the utility of this tool.

Thiopeptides are posttranslationally-processed macrocyclic peptide metabolites, characterized by extensive backbone and side chain modifications that include a six-membered nitrogenous ring, thioazol(in)e/oxazol(in)e rings, and dehydrated amino acid residues. Thiostrepton A, produced by Streptomyces laurentii ATCC 31255, is one of the more structurally complex thiopeptides, containing a second macrocycle bearing a quinaldic acid. Thiostrepton A and other thiopeptides are of great interest due to their potent activities against emerging antibiotic-resistant Gram-positive pathogens, in addition to their antimalarial and anticancer properties. The ribosomal origins for thiopeptides have been established, however, few details are known concerning the posttranslational modification steps. Alteration to the primary amino acid sequence of the precursor peptide provides an avenue to probe the substrate specificity of the thiostrepton A posttranslational machinery. The information gathered from current studies can also be used to refine thiostrepton’s structure-activity relationship, providing insight into the key features of its scaffold that impart specificity toward each biological target. A fosmid-dependent biosynthetic engineering platform for thiostrepton A was developed and a series of thiostrepton analogs were successfully produced adapting this method. The seventh residue of thiostrepton A is predicted to be critical for the metabolite’s antibacterial activity. Our results were consistent this hypothesis and demonstrated that substitution of Thr7 in the thiostrepton A precursor peptide disrupts both biological activity and successful biosynthesis of the analogs. The thiostrepton biosynthetic machinery’s tolerances toward structural variation at the second and fourth positions of the TsrA core peptide were probed by the saturation mutagenesis of Ala2 and Ala4, respectively. Eight thiostrepton Ala2 variants were isolated with two analogs truncated at the N-terminus by one amino acid, bearing a shortened quinaldic acid-containing macrocycle. Our results suggested that the identity of the core peptide second residue influences the biosynthesis of a thiostrepton analog, however, not essential for the antibacterial and proteasome inhibitory activities of the full-length variants. Additionally, the quinaldic acid loop size affects thiostrepton’s antibacterial potency, but is not critical for the proteasome inhibitory activity. Sixteen thiostrepton analogs were isolated from Ala4 mutagenesis studies. We demonstrated that the identity of the amino acid residue at the fourth position in the thiostrepton scaffold is not critical to inhibit either the ribosome or the proteasome in vitro.

Methanococcus jannaschii is a thermophilic methane producing archaebacterium. In this organism genes encoding the aspartate transcarbamoylase (ATCase) catalytic (PyrB) and regulatory (PyrI) polypeptides were found. Unlike Escherichia coli where the above genes are expressed from a biscistronic operon the two genes in M. jannaschii are separated by 200-kb stretch of genome. Previous researchers have not been able to show regulation of the M. jannaschii enzyme by the nucleotide effectors ATP, CTP and UTP. In this research project we have genetically manipulated the M. jannaschii pyrI gene and have been able to assemble a 310 kDa E. coli like enzyme. By using the second methionine in the sequence we have shown that the enzyme from this organism can assemble into a 310 kDa enzyme and that this enzyme is activated by ATP, CTP and inhibited by UTP. Thus strongly suggesting that the second methionine is the real start of the gene. The regulation of the biosynthetic pathway in Pseudomoans aeruginosa has previously been impossible to study due to the lack of CTP synthase (pyrG) mutants. By incorporating a functional uridine (cytidine) kinase gene from E. coli it has been possible to isolate a pyrG mutant. In this novel mutant we have been able to independently manipulate the nucleotide pools and study its effects on the enzymes in the biosynthetic pathway. The enzyme asapartate transcarbamoylase was repressed 5-fold when exogenous uridine was high and cytidine was low. The enzyme dihydroorotate was repressed 9-fold when uridine was high. These results suggest that a uridine compound may be the primary repressing metabolite for the enzymes encoded by pyrB and pyrC. This is the first study to be done with the proper necessary mutants in the biosynthetic pathway of P aeruginosa. In the past it has been impossible to vary the internal UTP and CTP pools in this organism.

Plants contain a host of secondary metabolites that may be of dietary use to man. A comprehensive approach to plant-based nutrition would include investigating all aspects of a nutrient, from creation through storage and consumption. Here, experiments address each of these facets for a group of important antioxidant and pigment compounds, the carotenoids. The carotenoid biosynthetic pathway regulatory mechanisms leading to lycopene accumulation are well defined in the model fruit, tomato. Those leading to accumulation of other carotenoids and flesh colors, however, are poorly understood. The variety of flesh colors available in watermelon fruit (red, orange, salmon yellow, and canary yellow) makes it an ideal candidate for investigating the regulation of the full pathway. Carotenoid accumulation was measured in ten watermelon varieties, representing the four flesh colors and three ploidy levels, throughout fruit maturation. It was found that the putative regulatory mechanisms controlling lycopene accumulation in red-fleshed fruit may be applied in a generalized fashion to each flesh color in respect to the major carotenoid accumulated at maturity. Additionally, triploid varieties were generally found to have higher accumulation levels than diploids, and tetraploids were intermediate to both. In addition to total carotenoid content, many factors are important in determining perceived benefit. Several of these factors involve components of the food matrix, cellular and subcellular species-specific characteristics of the food which act as barriers to nutrient release. Cell size, cell wall, and chromoplast (the carotenoid storage organelle) characteristics were observed in nine fruits and vegetables using light and transmission electron microscopy. Watermelon, tomato, and melon have the largest cells. Sweet potato, butternut squash, carrot, and mango have the most fibrous cell walls; mango and papaya additionally had the thickest walls. Chromoplast globular, tubular, crystalline, and membranous substructures were described for each food. These food matrix factors may be related to differences in carotenoid bioaccessibility between food sources. An in vitro digestion experiment was used to determine carotenoid bioaccessibility for each of these foods. Per serving, grapefruit yielded the most lycopene while carrot gave the most ?-carotene, ?-carotene, lutein, and phytoene, and mango proved a good source of violaxanthin.

The appended sugar residues of many natural products from Actinomyces are important for their biological activities. Many of these unusual sugar biosynthetic gene clusters have been isolated and many glycosyltransferases from various antibiotic-producing organisms have been identified. The increasing knowledge about these sugar biosynthetic pathways opens up the possibility of generating novel bioactive glycosylated compounds through combinatorial biosynthesis. The work described in this dissertation focuses on the investigation of the biosynthetic pathway of a rare nitro-containing sugar, D-kijanose, from an antibiotic, kijanimicin, and the glycosyltransferase helper proteins involved in the glycosylation of macrolide antibiotics. D-Kijanose, especially its nitro group, plays an important role in conferring the biological activities of the parent antibiotics. Cloning and sequencing of the kijanimicin biosynthetic gene cluster have allowed the proposal of the biosynthetic pathway of D-kijanose. The functions of the enzymes encoded by each open-reading frame in the cluster were also assigned based on sequence comparison with known enzymes found in other biosynthetic reactions. In this thesis, the functions of KijB1, a TDP-4-keto-6-deoxy-hexose 2,3-dehydratase, and KijD2, a TDP-hexose C-3 aminotransferase, were verified. The TDP-3-amino-4-keto-2,3,6-trideoxyhexose produced as an intermediate in the early stage of D-kijanose biosynthesis was also identified. In the second part of this dissertation, the in vivo protein-protein interaction between D-desosaminyl glycosyltransferase, DesVII, and its auxiliary protein, DesVIII, was established by yeast two-hybrid assay. The complex formation between these two proteins was also demonstrated by in vitro binding assay. Several strategies were tried to overexpress the D-mycaminosyl glycosyltransferase and its auxiliary protein, TylM2 and TylM3, although none of them were successful. A two-plasmid in vivo glycosylation system was also developed to test the competence of various DesVIII homologues to serve as the helper protein for glycosyltransferase DesVII, MycB and NbmD. In summary, the work in this dissertation has provided important information on the biosynthesis of D-kijanose and also significant insight into the function of the helper proteins of macrolide glycosyltransferases. These results could be useful for future studies of natural product biosynthesis and exploitation of glycodiversification. / text