No description available.

No description available.

No description available.

Candida albicans colonizes the gastrointestinal tract of up to 75 % healthy individuals. It usually cohabits the gut as an innocuous commensal. But in critically ill patients whose gut barrier, immune system and normal gut microbiota are compromised, C. albicans often transmigrates the gut barrier, transforms into an invasive pathogen and causes fatal systemic infections. The genetic transitions that drive this transformation in C. albicans have been a major focus of research and have led to the identification of key transcription factors that regulate this commensal-to-pathogen transition. However, the current challenge lies in identifying the downstream pathways and effectors that bring this transition into effect. This thesis addressed this challenge by developing 11 new bioinformatics tools, including 6 comprehensive databases, 4 novel software packages and 1 analysis framework. These databases included a comprehensive topological map of the mammalian gut biogeography, a C. albicans microarray database comprising of 3,091 publically available microarray transcript profiles, C. albicans RNA-seq gene expression and small variant databases extracted from 1,177 publically available RNA-seq samples, a C. albicans gene alias database comprising of 113,297 gene aliases representing the 6,735 open reading frames of C. albicans, and a C. albicans gene ontology slim comprising of 1,194 C. albicans-specific gene ontology terms. These databases were accompanied by a robust analysis framework which brought together these resources for quality control, batch correction and weighted gene co-expression network analysis. All these tools were finally employed in a pilot exploration of the C. albicans gut commensal-to-pathogen transition, which demonstrated the effectiveness of these bioinformatics resources. The analysis unveiled known regulators, uncharacterized gene networks, pathways and effectors potentially crucial for the C. albicans gut commensal-topathogen transition. These resources are a step towards a better understanding of this transition and can also be utilized for examining various other aspects of C. albicans biology.

Thesis (Ph. D.)--University of Hong Kong, 2006. / Title proper from title frame. Also available in printed format.

published_or_final_version / Microbiology / Master / Master of Philosophy

Eleven fungal strains were received from the clinical microbiology laboratory collection of Queen Mary Hospital and Pamela Youde Nethersole Eastern Hospital in Hong Kong from 2010-2011. The collection comprised of ten ascomycetes and one zygomycete. They were identified down to the genus level based on the morphological criteria. Internal transcribed spacer (ITS), beta-tubulin, actin and 28S gene sequencing were used for genotypic characterization. The ITS sequences of four of the strains demonstrate <3%-base difference to a single fungal species. They were species of the genus, Acremonium, Aspergillus, Cladophialophora and Ochroconis respectively. Five strains belonging to the genus, Trichophyton, Monascus, Mucor, Arthrinium and Acremonium could not be identified to the species level due to low interspecies heterogeneity. The tubulin gene was used for two of the strains. The tubulin sequence of a strain of Phaeoacremonium was identified to the species level with 0%-base difference. The ITS, partial beta-actin and 28S rDNA genes were sequenced for a strain of Exophiala. They showed a distinct cluster, mostly closely related to, but distinct from, Exophiala xenobiotica, Exophiala jeanselmei and Exophiala oligosperma. The genotypic characteristics suggest the strain to be a novel species of Exophiala. More genotypic and phenotypic characterization are required to described this strain of Exophiala. / published_or_final_version / Microbiology / Master / Master of Research in Medicine

In clinical microbiology laboratories, the conventional method for identification of pathogenic fungi is based on fungal culture and observation of fungal phenotypic characters. However, it is time-consuming, subjective and unreliable due to the long incubation period and variations in fungal colony morphology. Thus, there is a need for a rapid, objective and accurate identification of pathogenic fungal isolates. ITS regions are most commonly used targets for molecular identification of fungal pathogens because of the optimal inter- and intra-species variations and large copies in fungal genome. In this study, twenty-two clinical fungal isolates were identified using the phenotypic method and ITS sequencing. The results showed that there were only thirteen isolates identified to species level by phenotypic method, while others were only differentiated in genus level. Due to the poor differentiation based on the conventional phenotypic approach, misidentification of fungal pathogens occasionally occurred. However, ITS sequencing successfully achieved accurate species-level identification of all fungal isolates. The results were demonstrated in phylogenetic trees with high bootstrap support. In conclusion, ITS sequencing is a rapid and reliable for the identification of pathogenic fungal isolates. / published_or_final_version / Microbiology / Master / Master of Medical Sciences

Three thermotolerant “Lichtheimia-like” isolates from patients suffering from mucormycosis (nasopharyngeal swab of a liver transplant recipient, gastric biopsy of a renal transplant recipient, and skin biopsy of a man with burn, respectively) were characterized. Phenotypic characteristics, including colony morphology, microscopic morphology and thiamine dependency, were determined. Microscopic examination of agar block smear preparations showed that most side branches of the three “Lichtheimia-like” isolates and Lichtheimia ramosa were circinate, with abundant pleomorphic giant cells with fingerlike projections commonly observed. Sequence analysis of four loci (internal transcribed spacer region of the ribosomal DNA, partial elongation factor 1-α gene, partial β-actin gene, and the D1-D2 region of 28S rRNA) showed that the three “Lichtheimia-like” isolates and L. ramosa formed a distinct cluster, and are closely related to, but distinct from, other Lichtheimia sp.. A new species, Lichtheimia hongkongensis (synonym of L. ramosa), was proposed to describe this fungus. It was also hypothesized that a proportion of “A. corymbifera (L. corymbifera)” reported in the literature could be L. hongkongensis. In this study, 13 fungal strains that were reported as “A. corymbifera (L. corymbifera)” in the literature in an 11-year period were collected. Microscopic examination of agar block smear preparations of these 13 strains revealed unique characteristics of L. hongkongensis, including circinate side branches and pleomorphic giant cell with finger-like projections. Phylogenetic analysis also showed that all 13 strains are closely related with L. hongkongensis. It was confirmed that a significant number of reported A. corymbifera (L. corymbifera) infections are L. hongkongensis infections which are of global distribution. In order to confirm the distinct phylogenetic position of L. hongkongensis, the mitochondrial genomes of three L. hongkongensis strains and a strain of L. corymbifera were sequenced. L. hongkongensis HKU21, HKU22 and HKU23 have circular mitochondrial genomes with sizes ranging from 31830 bp to 32167 bp, which contain 39 to 41 genes transcribed by both strands. L. corymbifera HKU25 has a circular mitochondrial genome of size 37262 bp, which contains 41 genes that are transcribed by both strands. All the four mitochondrial genomes contain a complete set of tRNAs, the small and the large rRNAs, as well as the basic 14 protein-coding genes. Analysis of gene order showed that the three L. hongkongensis strains are clustered together. It is also shown that the Lichtheimia group is also closely related to Rhizopus oryzae, which is a member of Mucorales. A mouse model was used to determine if there is a difference in virulence between L. hongkongensis and L. corymbifera. The groups of mice challenged with L. hongkongensis have a higher survival rate than those challenged with L. corymbifera. Intravenous administration of Lichtheimia spores resulted in an infection in livers and spleens as indicated by positive cultures. The number of spores that could be recovered from these organs was significantly lower in mice challenged with L. hongkongensis than those challenged with L. corymbifera, indicating that L. hongkongensis could be less virulent than L. corymbifera. / published_or_final_version / Microbiology / Doctoral / Doctor of Philosophy

Identification of pathogenic fungi isolated from clinical specimens in clinical microbiology laboratories is primarily based on observing fungal phenotypic structures under the microscope and performing biochemical tests for fungal cultures. This conventional method is very time-consuming and labor-dependent. It usually requires several weeks for the fungi to grow sufficiently on culture media, and the identification processes on fungal phenotypic structure rely very much on experienced staff. Therefore, a more accurate and rapid method for pathogenic fungal identification is necessary for clinical laboratories to get abreast of modern development. Gene sequencing and phylogenetic analysis targeting the internal transcribed spacer (ITS) region in the fungal genomes are the most commonly used molecular methods for fungal identification. Because of the optimal inter and intra-species variation property of the ITS region, it can act as the DNA barcode to identify fungi to the species level. In this study, 33 clinical fungal isolates were identified by both phenotypic method and ITS sequencing. The results showed that 23 isolates were successfully identified to thespecies level by both phenotypic and molecular methods. Moreover, five isolates were only identified to the genus level by phenotypic method, but they could be successfully identified to the species level by ITS sequencing. However, five isolates have not been differentiated because there were mismatched results from phenotypic and sequencing methods. It may be due to the limitation of sequencing method on some fungal species. Building up a more comprehensive database or setting up a standard platform to guide the molecular process may help improve the performance of molecular method. To conclude, molecular method is a rapid and reliable way for fungal identification because ITS region acts as the DNA barcode for pathogenic fungi. / published_or_final_version / Medical Sciences / Master / Master of Medical Sciences