ABSTRACT
When dealing with the isolation of fungi from clinical samples, in a few cases, yeast colonies with a characteristic salmon-pink to coral-red color appear. A classic procedure of many medical mycologists is to assign such isolates to the genus Rhodotorula on the basis of this sole phenotypic attribute. Today, thanks to the improvement of molecular biology techniques (especially DNA sequencing) and molecular phylogenetics, it is well known that yeasts producing pigmented colonies (also called pigmented or red yeasts) may belong to several basidiomycetous genera other than Rhodotorula. Examples of these genera are Rhodosporidium (teleomorphic stage of many Rhodotorula species), Sporobolomyces, Sporidiobolus, Dioszegia, Cysto¡lobasidium, Xanthophyllomyces, and Cryptococcus. The distinctive pink-to-red coloration of their colonies is the result of the intracellular production and accumulation of carotenoid compounds. The composition of carotenoid pigments may vary qualitatively in red yeasts depending on the species. The most important carotenoids found in yeasts are torularhodin, torulene, and β-carotene.1,2
However, among yeast with current medical relevance, the pigmented species known to be most frequently involved in human mycoses in fact belong to Rhodotorula. There are no records of clinical isolates for almost any other pigmented genera. The only exception is Sporobolomyces (and its sexual
stage: Sporidiobolus), which has been found, in a few occasions, involved in human fungemia or has been isolated from clinical specimens.3-8
The genus Rhodotorula is polyphyletic and has 47 different species, constituting a very heterogeneous taxon. Most of Rhodotorula species belong to the subphylum Pucciniomycotina, either to the class Microbotryomycetes (29 species) or to the class Cystobasidiomycetes (15 species). The remaining three species are classi–ed in the subphylum Ustilaginomycotina and do not produce carotenoid pigments. R. glutinis is the type species of the genus Rhodotorula and belongs to the order Sporidiobolales (Class Microbotryomycetes), together with the abundant yeast R. mucilaginosa (previously known as R. rubra). A third clinically important Rhodotorula species is R. minuta, which, unlike R. glutinis and R. mucilaginosa, is ascribed to the order Cystobasidiales (Class Cystobasidiomycetes). The case of R. glutinis is peculiar. Until recently, it was regarded as a ubiquitous yeast, and at least 370 collection strains were considered to belong to this species.9 The classical circumscription of R. glutinis was based mostly on a few physiological and cultural properties. However, several studies have reported the physiological and genetic heterogeneity of
74.1 Introduction ..................................................................................................................................................................... 653 74.1.1 Taxonomy ............................................................................................................................................................ 653 74.1.2 Environmental Distribution ................................................................................................................................. 654 74.1.3 Clinical Features and Pathogenesis ..................................................................................................................... 655 74.1.4 Diagnosis ............................................................................................................................................................. 656
74.1.4.1 Conventional Identi–cation ................................................................................................................... 656 74.1.4.2 Molecular Identi–cation ........................................................................................................................ 656
74.2 Methods ........................................................................................................................................................................... 659 74.2.1 Sample Preparation .............................................................................................................................................. 659
74.2.1.1 Sample Collection and Culture ............................................................................................................. 659 74.2.1.2 DNA Extraction .................................................................................................................................... 659
74.2.2 Identi–cation Procedures ..................................................................................................................................... 659 74.2.2.1 Micro/Minisatellite Primed PCR ......................................................................................................... 659 74.2.2.2 Species-Speci–c PCR ........................................................................................................................... 660 74.2.2.3 Restriction of the Ampli–ed 5.8S-ITS Region of the Ribosomal rRNA Operon ................................. 660 74.2.2.4 rDNA Sequencing Analysis .................................................................................................................. 660
74.3 Conclusions and Future Perspectives............................................................................................................................... 661 References ................................................................................................................................................................................. 662
R. glutinis.10-14 More recently, Sampaio et al.15 and Gadanho and Sampaio16 reevaluated the circumscription of this species using a large set of isolates that had been previously identi-–ed by phenotypic criteria as R. glutinis and found that most of the isolates did not actually belong to this species. The most common were anamorph strains of Rhodosporidium babjevae, but other species of Rhodotorula (R. dairenensis, R. graminis, and Rhodotorula spp.) and Rhodosporidium (Rh. diobovatum, Rh. sphaerocarpum) had also been misidenti–ed as R. glutinis. It has been also shown that from a total of 45 isolates, only 4 had been properly assigned to R. glutinis.16 The species with the closest phenetic and phylogenetic proximity to R. glutinis are R. graminis and Rh. babjevae. Therefore, it is likely that many of the isolates of R. glutinis from humans, which have been reported, may actually have been confused with other yeast species. For the case of medically important yeasts, a reliable identi–cation is needed for the selection of the appropriate antifungal.
