Die Frage nach adäquaten mikrobiologischen und molekulargenetischen Methoden, um die Kolonisation des Respirationstrakts von Mukoviszidose-Patienten mit Pseudomonas aeruginosa nachzuweisen und zu charakterisieren, wird kontrovers diskutiert. Von 38 klinisch stabilen Patienten mit cystischer Fibrose (CF) wurden sequentiell im Abstand von 18 Monaten Proben aus Rachenabstrich, Sputum und Bronchiallavage (BAL) entnommen und bezüglich Pseudomonas-Nachweis untersucht. Die Pseudomonas-Stämme wurden mittels Random Amplified Polymorphic DNA (RAPD)-Analyse und Pulsfeld-Gelelektrophorese (PFGE) von DNA-Makrorestriktionsfragmenten typisiert und bezüglich der Frage nach genetisch divergierenden Isolaten innerhalb des selben Individuums sowie nach möglichen longitudinalen genetischen Veränderungen evaluiert. Sensitivität, negative und positive prädiktive Werte und Spezifität, um eine P. aeruginosa-Besiedlung zu erkennen, waren 36%, 74%, 83% und 96% im Falle der Kulturen aus dem Oropharynx von nicht-expektorierenden Patienten und 92%, 94%, 100% und 100% für Sputumkulturen von expektorierenden Probanden. RAPD-Analyse und PFGE waren in der Lage, zwischen unterschiedlichen Pseudomonas-Stämmen zu diskriminieren, wobei nur die DNA-Makrorestriktion zwischen Subtypen unterscheiden konnte. Die Genotypen der Pseudomonas-Isolate aus Rachenabstrich und Sputum divergierten in 55% und 40% zu den Isolaten der BAL. Longitudinale Variationen des Genotyps wurden in 62% der Fälle beobachtet, die Hälfte davon war nur mittels bronchoskopisch gewonnener Proben erkennbar. Zusammengefasst besitzen Sputumproben bezüglich des Pseudomonas-Nachweises dieselbe Wertigkeit wie Kulturen aus der BAL, während Rachenabstriche in einer frühen Krankheitsphase für die Charakterisierung der bakteriellen Flora des unteren Respirationstrakts wenig geeignet sind. Die Methode der DNA-Makrorestriktion kann als zuverlässige Technik für epidemiologische Untersuchungen empfohlen werden. Unterschiedliche Genotypen innerhalb desselben Individuums und longitudinale genetische Alterationen sind häufig, jedoch unter Umständen nur bronchoskopisch nachweisbar. / There is controversy about adequate specimen to detect and characterise colonisation of cystic fibrosis (CF) airways by Pseudomonas aeruginosa. Oropharyngeal, sputum and bronchoalveolar lavage (BAL) samples were evaluated sequentially from 38 stable CF patients for the detection of P. aeruginosa. Pseudomonas strains were typed by random amplified polymorphic DNA (RAPD) analysis and pulsed-field gel electrophoresis (PFGE) of DNA macrorestriction fragments. The occurrence of genetically different isolates within the same host and longitudinal variations in the genotype during repeated examinations was assessed. Sensitivity, negative and positive predictive values and specificity to detect P. aeruginosa were 36%, 74%, 83% and 96% for oropharyngeal cultures in non-expectorating patients and 92%, 94%, 100% and 100% for sputum cultures from expectorating patients, respectively. RAPD analysis and PFGE were suitable to characterize P. aeruginosa CF isolates, although only DNA macrorestriction was able to distinguish between identical and closely related strains. Genotypes of Pseudomonas isolates recovered from oropharyngeal swabs and sputum differed to the strains recovered by bronchoscopy in 55% and 40%, respectively. In 62% longitudinal variations in the genotype occurred. Half of these alterations were only detectable from bronchoscopically obtained samples. In conclusion, sputum samples have the same value as specimens from BAL to detect P. aeruginosa colonisation, whereas cultures from the oropharynx are not suitable for characterising the bacterial conditions in the CF lungs in an early disease state. DNA macrorestriction is recommended as an excellent tool for epidemiological investigations. Different genotypes within the same host and longitudinal genetic alterations are common and may be detectable in the BAL fluid exclusively.

Wong Ka Lok. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2001. / Includes bibliographical references (leaves 121-129). / Abstracts in English and Chinese. / Acknowledgments --- p.i / Abstract --- p.ii / Table A --- p.v / Table B --- p.vi / Table of Contents --- p.vii / Abbreviations --- p.xi / Chapter Chapter 1 --- Molecular authentication of endangered crocodiles and snakes / Chapter 1.1 --- Introduction --- p.1 / Chapter 1.2 --- Traditional method of snake and crocodile identification / Chapter 1.2.1 --- Morphology --- p.7 / Chapter 1.2.2 --- Chemical Analysis --- p.9 / Chapter 1.3 --- Molecular Technology in Authentication / Chapter 1.3.1 --- Polymerase Chain Reactions (PCRs) --- p.11 / Chapter 1.3.2 --- Random-primed amplification reaction --- p.12 / Chapter 1.3.3 --- Sequence Characterized Amplified Region (SCAR) --- p.13 / Chapter 1.3.4 --- PCR-RFLP --- p.13 / Chapter 1.3.5 --- DNA sequencing --- p.14 / Chapter 1.4 --- Objectives and strategies of the study --- p.15 / Chapter Chapter 2 --- Materials and General Methods / Chapter 2.1 --- Reagents and Buffers / Chapter 2.1.1 --- Buffers for Total DNA Extraction --- p.17 / Chapter 2.1.2 --- Reagents for Agarose Gel Electrophoresis --- p.17 / Chapter 2.1.3 --- Reagents for Plasmid DNA Preparation --- p.18 / Chapter 2.1.4 --- Medium for Bacterial Culture --- p.18 / Chapter 2.1.5 --- Reagents for Preparation of Competent Cells --- p.19 / Chapter 2.2 --- DNA Isolation / Chapter 2.2.1 --- Extraction of DNA from meats --- p.20 / Chapter 2.2.2 --- Extraction of DNA from blood --- p.20 / Chapter 2.3 --- Phenol/Chloroform Extraction --- p.21 / Chapter 2.4 --- Ethanol Precipitation --- p.22 / Chapter 2.5 --- DNA Concentration/Purity Estimation --- p.22 / Chapter 2.6 --- Mitochondrial DNA amplification --- p.23 / Chapter 2.7 --- Random-Primed Polymerase Chain Reactions --- p.24 / Chapter 2.8 --- SCAR for Snake samples --- p.24 / Chapter 2.9 --- SCAR for Crocodile samples --- p.25 / Chapter 2.10 --- Restriction fragment length polymorphism analysis --- p.25 / Chapter 2.11 --- Agarose Gel Electrophoresis of DNA --- p.26 / Chapter 2.12 --- Purification of PCR product --- p.26 / Chapter 2.13 --- Preparation of Escherichia coli Competent Cells --- p.27 / Chapter 2.14 --- Ligation and transformation of E. coli --- p.27 / Chapter 2.15 --- Plasmid preparation --- p.28 / Chapter 2.16 --- Screening of Plasmid DNA by Restriction Digestion --- p.29 / Chapter Chapter 3 --- DNA sequencing of snakes & construction of snake database / Chapter 3.1 --- Introduction --- p.30 / Chapter 3.2 --- Materials and methods / Chapter 3.2.1 --- Snake samples --- p.32 / Chapter 3.2.2 --- "DNA Extraction, mitochondrial gene amplification and DNA sequencing" --- p.33 / Chapter 3.2.3 --- Construction of database --- p.33 / Chapter 3.3 --- Results / Chapter 3.3.1 --- Cytochrome b gene amplification and sequencing --- p.34 / Chapter 3.3.2 --- Gene amplification and sequencing of 16S rRNA --- p.42 / Chapter 3.3.3 --- Cytochrome b sequence database --- p.50 / Chapter 3.3.4 --- 16S rRNA sequence database --- p.53 / Chapter 3.4 --- Discussion / Chapter 3.4.1 --- Cytochrome b and 16S rRNA genes of snake species --- p.55 / Chapter 3.4.2 --- Cytochrome b and 16S rRNA databases --- p.55 / Chapter Chapter 4 --- Application of PCR-RFLP and SCAR in snake species identification / Chapter 4.1 --- Introduction --- p.57 / Chapter 4.2 --- Material and Methods / Chapter 4.2.1 --- DNA extraction and PCR-RFLP --- p.58 / Chapter 4.2.2 --- RAPD and SCAR --- p.58 / Chapter 4.3 --- Results / Chapter 4.3.1 --- PCR-RFLP of cytochrome b genes of snakes --- p.59 / Chapter 4.3.2 --- PCR-RFLP of 16S rDNA --- p.61 / Chapter 4.3.3 --- RAPD & SCAR analysis --- p.67 / Chapter 4.4 --- Discussion --- p.72 / Chapter Chapter 5 --- "Application of DNA sequencing, PCR-RFLP and SCAR to identify crocodile species" / Chapter 5.1 --- Introduction --- p.74 / Chapter 5.2 --- Materials and methods / Chapter 5.2.1 --- "Crocodile, human and four animal samples" --- p.75 / Chapter 5.2.2 --- "DNA Extraction, mitochondrial gene amplification and DNA sequencing" --- p.75 / Chapter 5.2.3 --- PCR-RFLP and SCAR --- p.76 / Chapter 5.3 --- Results / Chapter 5.3.1 --- Isolation of crocodiles DNA --- p.77 / Chapter 5.3.2 --- Isolation of DNA from Human and four animal species --- p.78 / Chapter 5.3.3 --- Cytochrome b gene amplification and sequencing --- p.78 / Chapter 5.3.4 --- 16S rRNA gene amplification and sequencing --- p.84 / Chapter 5.3.5 --- PCR-RFLP of cytochrome b --- p.89 / Chapter 5.3.6 --- PCR-RFLP of 16S rRNA --- p.91 / Chapter 5.3.7 --- SCAR primers for four crocodile species --- p.93 / Chapter 5.4 --- Discussion --- p.97 / Chapter Chapter 6 --- A case report - authentication of animal samples using DNA sequencing / Chapter 6.1 --- Introduction --- p.99 / Chapter 6.2 --- Material and methods / Chapter 6.2.1 --- Materials --- p.101 / Chapter 6.2.2 --- DNA Extraction and sequencing --- p.101 / Chapter 6.3 --- Result and discussion / Chapter 6.3.1 --- Cytochrome b gene sequencing --- p.102 / Chapter 6.3.2 --- Sequence homology among samples and meats obtained from the market --- p.111 / Chapter 6.3.3 --- Identity of samples B & D --- p.113 / Chapter Chapter 7 --- General Discussion / Chapter 7.1 --- Advantages and weakness of DNA technology --- p.116 / Chapter 7.2 --- Choosing appropriate molecular markers --- p.118 / Chapter 7.3 --- Further suggested work --- p.119 / Chapter 7.4 --- Conclusion --- p.119 / References --- p.121 / Appendix --- p.130

Black wattle (Acacia mearnsii), which belongs to the genus Acacia, is one of the many species of trees or hardwoods grown commercially in South Africa. Black wattle is a species indigenous to Australia and was introduced into South Africa by the van der Plank brothers in 1864. These trees are grown in South Africa because of its tannin-rich bark, the extract of which is used by the leather tanning industry. Black wattle is also grown for its timber, timber products and pulp. The introduction and cultivation history of black wattle suggests that the South African plantations contain limited genetic variation with relatedness amongst groups estimated to be high, thus implying a narrow genetic base in the South African black wattle population. In this investigation, Random Amplified Polymorphic DNA (RAPD) was used to estimate the genetic variation between seven different black wattle groups. A total number of 34 individuals obtained from different areas in South Africa were examined; Piet Retief (group 47 and 50: half-sibs), Kumbula (group 85: unrelated individuals), Howick (group 400: unrelated individuals) and an unknown area (groups 88, 89, 91: full-sibs). As this investigation was the first of its kind, a DNA isolation method as well as a PCR-RAPD protocol had to be modified. Total genomic DNA was successfully extracted using the CTAB DNA extraction method. This method removed large amounts of tannin present in the cells of the black wattle leaves and extracted high quality DNA to conduct between 50-100 RAPD reactions. The DNA purities ranged from 0.1 to 1.8, with an average of 1.46. A total of fourteen 10-mer RAPD primer sequences were randomly selected from the Operon Technologies primer list A, and tested in this investigation. Of the 14 primers used, only nine primers produced clear, single and repeatable bands. Therefore nine primers were selected for subsequent analyses. Ninety one loci that generated bands ranging from 300-3050 base pairs were produced. Seven to 13 loci per primer were generated. A total of 95.6 % of the loci were polymorphic. The overall expected mean heterozygosity (H = 0.3) obtained in this study was high in comparison to other studies conducted on acacias. The high levels of genetic variation were attributed to mating systems, dissortative mating and geographic distribution. The statistical packages POPGENE and ARLEQUIN were used to analyse the RAPD fingerprints. The genetic measures, Nei's diversity and Shannon's Information Index, showed that there was greater diversity exhibited (Nei's gene diversity = 32.09 % and Shannon's = 48.31 %), in the whole population than in each of the groups (with average of Nei's gene diversity = 20.33 % and Shannon's = 34.64 %). With regards to individual group analyses, low levels of genetic variation was obtained in group 400 (unrelated), from the Howick region, and group 85 (unrelated), from the Kumbula region, (mean 0.14 and 0.17 respectively). The low genetic values were attributed to limited gene exchange occurring in these two areas, bottlenecks and selection pressures. Groups 88, 89 and 91, from the unknown region (full-sib groups), were the most variable in comparison to the other groups, with means of (0.27,0.24 and 0.18 respectively). These high genetic variation values could be due to the fact that gene migration could have occurred between these groups and others in the area. It is thought that most acacias are insect-pollinated and this could have lead to gene migration between groups or populations, thereby explaining the high mean values. The gene flow obtained for the seven groups (FST = 0.174) indicated that great genetic differentiation existed in this population of black wattle studied. This value is higher in comparison to other woody species; however it is similar to other acacia species. UPGMA cluster analysis using Nei's unbiased genetic distance, revealed four distinct clusters of groups corresponding to the distribution areas represented in this study. The Howick (group 400: unrelated) and Kumbula (group 85: unrelated) were more closely related to each other than to the other groups, since both these groups are from Natal. The Piet Retief groups (groups 47 and 50: half-sibs), branched-off together, indicating that they are distinct from the other groups. The pairwise analysis of identity showed that the relationship between the group from Howick (group 400: unrelated) and all the other groups from the other regions was the lowest, ranging from 64 % to 79 %. The relationship between all the groups beside the group from Howick (group 400: unrelated) was reasonably high, ranging from 78 % to 90 %. This distance displayed by group 400 (unrelated) from Howick in relation to the groups, is attributed to the fact that it is frost resistant and the other groups not. Genetic variation was also detected and partitioned, between and within groups, by Analysis of Molecular Variance (AMQVA). Majority of the variation existed within groups (82.65 %) but significant differentiation was recorded between groups (17.44 %). This high level of within group differentiation may be explained by many aspects, such as the species breeding system, genetic drift or genetic isolation of groups or populations. The application of RAPD fingerprinting in black wattle has provided a more in depth understanding of the genetic variation residing in the South African population. The results achieved implementing this technique has shown that significant genetic variation exists within the black wattle population in South Africa. The results obtained in this study are also important since it is contrary to the expectation that the black wattle population in South Africa has low genetic variation. This knowledge is of great value to genetically discriminate between individuals or groups, to improve the selection of superior genotypes and allowing improved quality control in breeding programmes and seed orchard management. / Thesis (M.Sc.)-University of KwaZulu-Natal, Pietermaritzburg, 2005.

A identificação de espécie de Leishmania é crucial no diagnóstico de leishmaniose cutânea (LC), para terapia e prognóstico, em áreas com diferentes espécies endêmicas. A reação em cadeia de polimerase (PCR) foi usada em amostras de biópsias estocadas em parafina e formol de pacientes com LC, do Espírito Santo, Brasil. Usando sequências de mini-exon do gênero Leishmania, 63,2% (36/57) das amostras em parafina e 46,1% (6/13) das amostras em formol foram positivas. Numa abordagem usando sequências de kDNA e RFLP, 100% (58/58) das amostras de parafina testadas foram identificadas como Leishmania (V.) braziliensis. Estas reações podem ser usadas tanto para diagnóstico como para definição da espécie infectante / The identification of Leishmania species, a crucial step in cutaneous leishmaniasis, had therapeutics and prognostics implications, specially at when several agent species are endemic. Polymerase chain reaction (PCR) was used for analysis of paraffin-embedded and formalin stored skin biopsies from patients with parasitologic confirmed cutaneous formalin stored skin biopsies from patients with parasitologic confirmed cutaneous leishmaniasis, from the Espirito Santo State, Brazil. Tested by PCR targeted to mini-exon genus specific primer, 63,2% (36/57) of parafin and 46,1% (6/13) of formalin stored samples were positive, but using a PCR for kDNA primer followed by RFLP analysis, only Leishamnia (V.) braziliensis was identified in 100% (58/58) of parafin biopsies studied. Those reactions allow either diagnosis or species identification in cutaneous leismaniasis

Nas duas últimas décadas, várias espécies de Candida têm surgido como importantes patógenos hospitalares, no mundo e no Brasil. A identificação da origem da infecção tem importância na definição de estratégias de prevenção e controle. As estratégias para a prevenção de candidíase endógena podem focar, parcialmente, em métodos para redução da colonização de mucosas, por exemplo, a restrição ao uso de antibióticos de largo espectro. Entretanto, nos casos nos quais está envolvida uma fonte exógena, um expressivo reforço, na melhoria da qualidade das práticas de assistência à saúde, é prioritário para prevenção da transmissão. O objetivo deste estudo foi avaliar diferentes sítios de colonização por Candida como potenciais fontes de candidemia. O estudo foi desenvolvido em 3 hospitais no Brasil: Instituto Central do Hospital das Clínicas da Faculdade de Medicina de São Paulo, hospital universitário de nível terciário de complexidade, com mil leitos; o Instituto de Infectologia Emílio Ribas, um hospital com 200 leitos, referência para todo o Estado de São Paulo; e o Hospital Geral de Itapecerica da Serra, hospital de cuidados secundários da Grande São Paulo. Foram incluídos no estudo os pacientes com isolamento de Candida em hemocultura obtida de veia periférica após 48 horas de admissão hospitalar. As culturas de vigilância para Candida foram colhidas dos seguintes sítios: urina, reto, cavidade oral, pele (virilha e axila), pele ao redor do cateter e ponta de cateter caso disponível. A tipagem molecular foi realizada quando a mesma espécie de Candida (C. albicans, C. parapsilosis, C. tropicalis and C. glabrata) foi isolada no sangue e nos sítios de vigilância do mesmo paciente. A eletroforese em campo pulsado foi realizada para os isolados de C. albicans, C. parapsilosis e C. glabrata. A amplificação de segmentos polimórficos do DNA foi realizada para C. albicans e C. tropicalis. No total 63 pacientes consecutivos com candidemia foram incluídos no estudo no período de maio de 2004 a outubro de 2005. C. albicans foi isolada em 42% das hemoculturas, C. parapsilosis em 35%, C. tropicalis em 16%, C. guilliermondii, C. krusei, C. glabrata, e C. holmii, em 2% cada uma. Unicamente seis dos 10 isolados de ponta de cateter apresentaram perfil eletroforético idêntico aos isolados de C. parapsilosis do sangue. Os isolados de C. albicans do sangue e de culturas de vigilância do trato gastrintestinal correspondentes, oriundos de 12 pacientes, apresentaram genótipos idênticos. Os resultados sugerem que a colonização do trato gastrintestinal é a provável fonte de candidemia por C. albicans e que a candidemia por C. parasilosis é de origem exógena. / In the last two decades, Candida spp. have emerged as important nosocomial pathogens in the world and in Brazil. The identification of the source of infection is important in approaching prevention and control strategies. Strategies for the prevention of endogenous candidiasis may focus, to a certain extent, on methods for reducing mucosal colonization, for example limitation use of wide-spectrum antibiotics. However, in cases in which an exogenous source is involved, the aggressive reinforcement of adequate healthcare practices is mandatory to prevent transmission. The objective of this study was to evaluate different Candida colonization sites as potential sources for Candida fungemia. The study was done in 3 hospitals in Brazil: the Central Institute of Hospital das Clinicas, a 1000-bed tertiary-care hospital affiliated to the University of São Paulo; the Institute Emilio Ribas, a 200-bed infectious diseases hospital, reference for all the state of São Paulo; and the General Hospital of Itapecerica da Serra, a secondary-care community hospital located in area of the greater São Paulo. The patients with a positive blood culture for Candida, collected from a peripheral vein, were included in the study if they had to be hospitalized for 48 hours or more before candidemia. The following surveillance cultures for Candida were collected from: urine, rectum, oropharynx, skin (groin and axilla), skin around the catheter and catheter tip if available. Molecular typing was performed when the same species of Candida (C. albicans, C. parapsilosis, C. tropicalis and C. glabrata) was isolated from the blood and from surveillance sites of a single patient. Pulsed-field gel electrophoresis was performed for C. albicans, C. parapsilosis and C. glabrata isolates. Randomly amplified polymorphic DNA was performed for C. albicans and C. tropicalis. A total of 63 consecutive patients with candidemia were included in the period from May 2004 to October 2005. C. albicans comprised 42% of the blood isolates, C. parapsilosis 35%, C. tropicalis 16%, C. guilliermondii, C. krusei, C. glabrata, and C. holmii, 2% each. Six of the 10 isolates from catheter tips presented identical electrophoretic profiles to corresponding C. parapsilosis blood cultures and no other surveillance sites were related. C. albicans isolates from blood and from corresponding gastrointestinal surveillance sites from 12 patients presented identical genotypes. In conclusion, our results suggest that tract gastrointestinal colonization is the probable source of C. albicans candidemia and that C. parapsilosis candidemia is not endogenous.

A identificação de espécie de Leishmania é crucial no diagnóstico de leishmaniose cutânea (LC), para terapia e prognóstico, em áreas com diferentes espécies endêmicas. A reação em cadeia de polimerase (PCR) foi usada em amostras de biópsias estocadas em parafina e formol de pacientes com LC, do Espírito Santo, Brasil. Usando sequências de mini-exon do gênero Leishmania, 63,2% (36/57) das amostras em parafina e 46,1% (6/13) das amostras em formol foram positivas. Numa abordagem usando sequências de kDNA e RFLP, 100% (58/58) das amostras de parafina testadas foram identificadas como Leishmania (V.) braziliensis. Estas reações podem ser usadas tanto para diagnóstico como para definição da espécie infectante / The identification of Leishmania species, a crucial step in cutaneous leishmaniasis, had therapeutics and prognostics implications, specially at when several agent species are endemic. Polymerase chain reaction (PCR) was used for analysis of paraffin-embedded and formalin stored skin biopsies from patients with parasitologic confirmed cutaneous formalin stored skin biopsies from patients with parasitologic confirmed cutaneous leishmaniasis, from the Espirito Santo State, Brazil. Tested by PCR targeted to mini-exon genus specific primer, 63,2% (36/57) of parafin and 46,1% (6/13) of formalin stored samples were positive, but using a PCR for kDNA primer followed by RFLP analysis, only Leishamnia (V.) braziliensis was identified in 100% (58/58) of parafin biopsies studied. Those reactions allow either diagnosis or species identification in cutaneous leismaniasis

Nas duas últimas décadas, várias espécies de Candida têm surgido como importantes patógenos hospitalares, no mundo e no Brasil. A identificação da origem da infecção tem importância na definição de estratégias de prevenção e controle. As estratégias para a prevenção de candidíase endógena podem focar, parcialmente, em métodos para redução da colonização de mucosas, por exemplo, a restrição ao uso de antibióticos de largo espectro. Entretanto, nos casos nos quais está envolvida uma fonte exógena, um expressivo reforço, na melhoria da qualidade das práticas de assistência à saúde, é prioritário para prevenção da transmissão. O objetivo deste estudo foi avaliar diferentes sítios de colonização por Candida como potenciais fontes de candidemia. O estudo foi desenvolvido em 3 hospitais no Brasil: Instituto Central do Hospital das Clínicas da Faculdade de Medicina de São Paulo, hospital universitário de nível terciário de complexidade, com mil leitos; o Instituto de Infectologia Emílio Ribas, um hospital com 200 leitos, referência para todo o Estado de São Paulo; e o Hospital Geral de Itapecerica da Serra, hospital de cuidados secundários da Grande São Paulo. Foram incluídos no estudo os pacientes com isolamento de Candida em hemocultura obtida de veia periférica após 48 horas de admissão hospitalar. As culturas de vigilância para Candida foram colhidas dos seguintes sítios: urina, reto, cavidade oral, pele (virilha e axila), pele ao redor do cateter e ponta de cateter caso disponível. A tipagem molecular foi realizada quando a mesma espécie de Candida (C. albicans, C. parapsilosis, C. tropicalis and C. glabrata) foi isolada no sangue e nos sítios de vigilância do mesmo paciente. A eletroforese em campo pulsado foi realizada para os isolados de C. albicans, C. parapsilosis e C. glabrata. A amplificação de segmentos polimórficos do DNA foi realizada para C. albicans e C. tropicalis. No total 63 pacientes consecutivos com candidemia foram incluídos no estudo no período de maio de 2004 a outubro de 2005. C. albicans foi isolada em 42% das hemoculturas, C. parapsilosis em 35%, C. tropicalis em 16%, C. guilliermondii, C. krusei, C. glabrata, e C. holmii, em 2% cada uma. Unicamente seis dos 10 isolados de ponta de cateter apresentaram perfil eletroforético idêntico aos isolados de C. parapsilosis do sangue. Os isolados de C. albicans do sangue e de culturas de vigilância do trato gastrintestinal correspondentes, oriundos de 12 pacientes, apresentaram genótipos idênticos. Os resultados sugerem que a colonização do trato gastrintestinal é a provável fonte de candidemia por C. albicans e que a candidemia por C. parasilosis é de origem exógena. / In the last two decades, Candida spp. have emerged as important nosocomial pathogens in the world and in Brazil. The identification of the source of infection is important in approaching prevention and control strategies. Strategies for the prevention of endogenous candidiasis may focus, to a certain extent, on methods for reducing mucosal colonization, for example limitation use of wide-spectrum antibiotics. However, in cases in which an exogenous source is involved, the aggressive reinforcement of adequate healthcare practices is mandatory to prevent transmission. The objective of this study was to evaluate different Candida colonization sites as potential sources for Candida fungemia. The study was done in 3 hospitals in Brazil: the Central Institute of Hospital das Clinicas, a 1000-bed tertiary-care hospital affiliated to the University of São Paulo; the Institute Emilio Ribas, a 200-bed infectious diseases hospital, reference for all the state of São Paulo; and the General Hospital of Itapecerica da Serra, a secondary-care community hospital located in area of the greater São Paulo. The patients with a positive blood culture for Candida, collected from a peripheral vein, were included in the study if they had to be hospitalized for 48 hours or more before candidemia. The following surveillance cultures for Candida were collected from: urine, rectum, oropharynx, skin (groin and axilla), skin around the catheter and catheter tip if available. Molecular typing was performed when the same species of Candida (C. albicans, C. parapsilosis, C. tropicalis and C. glabrata) was isolated from the blood and from surveillance sites of a single patient. Pulsed-field gel electrophoresis was performed for C. albicans, C. parapsilosis and C. glabrata isolates. Randomly amplified polymorphic DNA was performed for C. albicans and C. tropicalis. A total of 63 consecutive patients with candidemia were included in the period from May 2004 to October 2005. C. albicans comprised 42% of the blood isolates, C. parapsilosis 35%, C. tropicalis 16%, C. guilliermondii, C. krusei, C. glabrata, and C. holmii, 2% each. Six of the 10 isolates from catheter tips presented identical electrophoretic profiles to corresponding C. parapsilosis blood cultures and no other surveillance sites were related. C. albicans isolates from blood and from corresponding gastrointestinal surveillance sites from 12 patients presented identical genotypes. In conclusion, our results suggest that tract gastrointestinal colonization is the probable source of C. albicans candidemia and that C. parapsilosis candidemia is not endogenous.