Clinics in Dermatology (2010) 28, 190196

New molecular tools in the diagnosis of superficial fungal infections

Roderick J. Hay, DM, Rachael Morris Jones, PhD
Infectious Disease Clinic, Dermatology Department, Kings College Hospital NHS Trust, Denmark Hill, London SE5 9RS United Kingdom

Abstract Laboratory diagnosis is in a process of major change, with the rapid development of techniques that rely on the application of new scientific skills that provide more accurate and faster confirmation of the presence of infection. This chapter describes the evolution of molecular methods for use in dermatology and their current status. Although many of the techniques that have been published use different detection systems, the evidence shows that there is a growing consensus on the best ways forward. For instance, in dermatophytosis polymerase chain reaction (PCR)-based methods, such as real-time PCR, and detection of changes in the internally transcribed spacer regions are proving to be the most suitable for the detection of different species. Diagnostic application for detection of Candida infection has largely focused on systemic disease, although methods for the diagnosis of vaginal candidosis have been evaluated. It is difficult to separate diagnosis from the other benefits of molecular methodologies in mycology, such as their use in promoting our understanding of the pathogenesis and epidemiology of these diseases. A new gene-based taxonomy has also been necessary. These changes provide a challenge to the clinical dermatologist, because there is still an important task that relies on clinical expertise in relating the presence of molecular fragments of each organism to the presence of infection. 2010 Published by Elsevier Inc.

Introduction
With the possible exception of tinea versicolor (pityriasis versicolor), few superficial fungal infections can be diagnosed accurately from clinical grounds alone. In the case of tinea capitis, for instance, clinical diagnosis generally provides an inaccurate estimate of the numbers of infected individuals. Likewise, onychomycosis is notoriously difficult to diagnose clinically with precision, even though algorithms have been introduced to simplify diagnosis using signs and symptoms alone.
Corresponding author. E-mail address: roderick.hay@ifd.org (R.J. Hay). 0738-081X/$ see front matter 2010 Published by Elsevier Inc. doi:10.1016/j.clindermatol.2010.01.001

For this reason, the gold standard of diagnosis usually depends on a series of well-recognized laboratory procedures. Yet, current laboratory tests are also problematic, because they depend on a number of factors, some of which are difficult to influence, such as the skill of the physician in taking adequate material from the correct site and the skills of the laboratory staff in examining and identifying the presence and nature of the organisms. Technical problems are also inherent in such tests.1 For instance, demonstrating the presence of organisms by microscopy is difficult at some sites such as the nail, where, even in the most experienced hands, direct microscopy usually underestimates the true number of positive cases, unless additional procedures such as rapid periodic-acid Schiff (PAS) staining or the use of a

Molecular tools for fungal infection diagnosis fluorochrome, such as Calcofluor (Polysciences Inc, Warrington, PA) are used. These procedures can increase the microscopic yield by at least 10%. False-negative cultures are also commonplace. Likewise, cultural identification is not a simple process, because it is based on the recognition of specific features, such as macroscopic and microscopic morphology and pigmentation. For some fungi, such as yeast species, rapid biochemical tests or colorimetric growth media are available and so recognition is more readily standardized. Mold fungi remain a challenge, and the outcome of diagnostic tests depends on the skills and experience of the diagnostic laboratory staff. Identification is based on visual recognition of changes, some subtle, in the appearance and growth of organisms on laboratory media; further confusion is caused by the atypical growth forms that occur frequently. The quest for a simplified diagnostic system that provides rapid and accurate answers has long been an aspiration. So where do molecular techniques fit in? In virology, for instance, molecular diagnostic tests are routine in many hospitals and laboratories. Whatever the underlying scientific basis of the molecular assay, which is based on complex scientific theory, the technical aspects of performing the tests can be incorporated into laboratory processes using standard operating procedures. Tests can also be used to screen for a variety of different pathogens from the same sample. At present, costs are comparatively high due to the expense inherent in purchasing automated equipment and the price of reagents, but there is an economy of scale in establishing a molecular diagnostic center. The more the techniques are used, the cheaper they will become.

191 Currently, a set of morphologic criteria are used to determine the identity and origin of organisms. A taxonomy that uses an entirely new set of discriminators, which occurs when molecular tools are used, challenges the conventional division of species and varieties. The difference, for instance, between certain Trichophyton spp was of some value clinically, although the biologic basis for the split was sometimes less certain, and with modern molecular taxonomy, some studies have shown the split is no longer valid. As will be seen, molecular diagnosis can be tuned to various levels of discrimination for different purposes. Differentiation between organisms required for molecular epidemiology, such as investigating the spread of an outbreak of tinea capitis, requires a finer degree of tuning than that required for diagnosis. The use of pan-fungal genetic probes may also be perfectly valid and useful clinically but will not discriminate between different genera. So determining the level of fine tuning of the system is key to the development and application of molecular diagnostic techniques. Although molecular diagnosis may theoretically involve any technique that identifies, through discrimination, at the level of specific molecule(s), the term is generally used to describe a diagnosis that depends on the recognition of fragments of DNA or RNA from nuclear or extranuclear sources, or both. The techniques of detection involve a variety of different indicator methods, from those that are based on the interpretation of gels to the generation of photons (fluorescence). The equipment needed differs widely in capacity and complexity. Most systems dedicated to diagnostic purpose use a form of amplification of genetic fragments, based on the polymerase chain reaction (PCR), but even here, many different techniques are available, from real-time PCR to nested PCR or sequencing. Describing these different methods is beyond the scope of this chapter, which will concentrate on the assays developed and the gene sections targeted.

Molecular diagnosis in medical mycology

The application of molecular genetic diagnostic tests in medical mycology has had a slower start than in other spheres of microbiology, but the sameor similar techniques are now used widely as tools for clinical and epidemiologic research and for studies of immunology and pathogenesis. To date in diagnostics, however, there are few commercially available test systems, a sign that molecular methods have yet to be established in diagnostic practice. In the case of yeasts, such as the Candida spp, the use of molecular tools is well developed in the diagnosis of systemic infections. The speed and accuracy of assays is very useful, although gene-based methods have still not displaced conventional culture techniques. The same methods can be applied in superficial yeast infections; but, at present, there is little incentive to develop a test system for superficial candidosis or vaginal thrush. A major challenge is provided by the dermatophyte fungi, which are slow to grow and often difficult to identify in many laboratories. Dermatophytes are closely related genetically, and the earliest attempts to separate them using crude genetic tools such as DNA homologies showed this clearly.2

Dermatophyte infections
Most of the earlier work on the use of molecular tools focused on different PCR techniques to discriminate between dermatophyte fungi in culture.1,3 For instance, reseachers4 used genomic DNA from Trichophyton spp in an arbitrarily primed PCR and showed that by using a primer (5ACCCGACCTG-3), it was possible to discriminate between three species, T rubrum, T mentagrophytes, and T tonsurans, and also to separate these organisms from controls. The accuracy of such methods was highlighted by a report5 that used a similar approach with a number of different primers, such as AC10 and M13, to probe for differences not only between species but also between different varieties (intraspecific differences). The researchers were able to discriminate between varieties of T mentagrophytes but not between the phenotypically distinct forms of T

192 tonsurans. They also used the same PCR approach, using DNA from the type strains of dermatophytes, to show that it is possible to provide a genetic identification for strains with phenotypic traits that are atypical and that are unidentifiable by morphologic criteria. This provided insight into the ability of the new systems to not only identify cultures but also discriminate between isolates that had previously been difficult to classify. As an example, using a different approach, the diagnosis of a difficult case of black dot ringworm of the scalp was confirmed as T violaceum by amplifying 620-bp genomic DNA fragments of the CHS1 gene from a group of Trichophyton spp that included T tonsurans and T violaceum.6 This work reinforced the view that different analytic techniques produced responses with different sensitivities. For instance, one comparison used different techniques, such as restriction enzyme analysis (REA), hybridization with a poly (dG-dT) DNA probe, randomly amplified polymorphic DNA (RAPD) with PCR, and restriction analysis of a segment of PCR-amplified rDNA with T mentagrophtyes and T rubrum isolates from skin or nails.7 The tests differentiated the isolates of the two species; however, there was variation in the ability of each to detect intraspecies differences. REA, for instance only demonstrated some isolate variation, and RAPD demonstrated variation between strains of T mentagrophytes but not of T rubrum. A study that used a similar approach and different molecular methods, such as PCR fingerprinting and amplified fragment length polymorphism analysis,8 was able to reduce the number of true Trichophyton spp of human origin but included the separation of T interdigitale and T mentagrophytes. Such methods applied to other species also gave rise to similar but surprising results. For instance, a number of less common morphologic varieties of organisms assigned to distinct species by the old methods, such as T megninii, T soudanense, and T yaoundei, were all assigned to T rubrum or T violaceum.9 These findings emphasize that molecular tools can be refined to provide simplification of what appears to be an unduly complex form of classification, and the classification of Trichophyton species, in particular, can now be substantially revised. The use of different methods or probes can also provide subspecies separation to further epidemiologic studies of geographic origin or outbreaks. An example of the latter is provided by a study of two novel tandem repetitive subelements (TRSs), TRS-1 and TRS-2, of the ribosomal DNA nontranscribed-spacer region (NTS) of T rubrum.10 These were used to identify varieties of T rubrum from 101 different isolates, and subsequently to investigate relapse after therapy of onychomycosis and to demonstrate T rubrum spp concordance in most patients with coexistent tinea pedis and manuum.11 Both studies also showed that some patients were infected by more than one strain. A word of caution, however, was sounded by results of a study on T. violaceum12 in which heterogeneity was found between IGS regions within a single

R.J. Hay, R.M. Jones T violaceum genome due to different copy numbers of a 171-bp tandem repeat. Heterogeneity of this sort is not known to affect other dermatophytes,13 but it provides an example of variation that may obscure the interpretation of some methods. An ideal means of establishing a thorough and accurate diagnostic system is through the application of a speciesspecific identifier, such as a short probe. An oligonucleotide sequence from the variable internal transcribed spacer (ITS) 2 regions of ribosomal DNA from T rubrum provides an example of such a specific marker.14 This was shown to be specific, and when labeled, did not hybridize to other dermatophytes, fungi, or human DNA. There was 100% success in hybridization with T rubrum itself. The ITS region has been increasingly targeted by investigators. For instance, one study sequenced the ITS region of 42 dermatophytes belonging to T rubrum, T mentagrophytes, T soudanense, T tonsurans, Epidermophyton floccosum, Microsporum canis, and M gypseum.15 A different detection system, rolling-circle amplification, was used for identification of single nucleotide polymorphisms. Intraspecies genetic variation was also found for T tonsurans, T mentagrophytes, and T soudanense but not for T rubrum. Another group used the ITS regions to identify T concentricum in culture-negative lesions of tinea imbricata.16 When a single repetitive oligonucleotide (GACA) primer was used to identify strains, it could identify many species, such as M canis and E floccosum; however, three distinct profiles occurred with T mentagrophytes.17 Other targets for species identification have included the DNA topoisomerase II gene, which again provides specific identification while also revealing some intraspecific differences.18 Microsatellite markers would provide another target.19 Careful selection of target genes can potentially enhance the identification of difficult organisms or those with atypical features from cultures isolated or from clinical material, as described subsequently. Two examples are provided by the previously cited cases of T violaceum and tinea imbricata.6,16 A third example was an unidentifiable mold grown from a patient with tinea corporis; after considerable delay, it expressed some of the features of T rubrum, including negative urease. Sequencing the ITS region of this isolate provided a more rapid identification of T rubrum.20 Further potential diagnostic use of molecular tests on cultured organisms would be to identify strains with different antifungal sensitivities. As yet, such studies are sparse; however, one investigation of isolates of T rubrum before and after therapy used TRS-1 and TRS-2 of ribosomal DNA to differentiate strains.21 The fungi isolated were also subjected to antifungal sensitivity assays. An increase in the minimum inhibitory concentrations (MIC) to most of the azoles tested was found in cultures obtained after treatment in four patients, and the authors associated this with an increase in a strain designated by genotype as 1-I. Although overt resistance to antifungal agents is not common amongst dermatophytes, rapid identification of

Molecular tools for fungal infection diagnosis organisms with reduced drug sensitivity would be of great potential clinical value. The application of genetic tests to the direct identification of fungi in clinical material such as nails has been an obvious use of the new techniques. An early study showed that PCR could be used to detect dermatophytes in infected but not healthy nail. The primers targeted conserved sequences in the ribosomal subunit 18S-rRNA.22 Using probes for a similar target region along with the adjacent ITS rDNA, another group23 was able to produce a fungal specific assay that would not detect other eukaryote or prokaryote DNA. The test detected as little as 10 pg of fungal DNA or about 25 fungal genome fragments per sample specimen from a variety of fungal organisms. The group applied this to medical samples and showed that it would detect the presence of fungi compared with conventional culture procedures. Providing species-specific probes is also appropriate. Two primers, designated TRIF and TR1R, were found that specifically identified T rubrum, T soudanense, and T. gourvilii.24 No other microorganism was identified. Increasing the sensitivity of nail sampling through molecular methods remains an important target, although again, this may depend on the target gene. Actin gene-based primers were used to detect T rubrum in nail material.25 When this was compared with Calcofluor-stained samples and with culture on routine media, 33% of the samples were positive by Calcofluor compared with 25.3% for the molecular detection of T rubrum. The combination of Calcofluor and genetic detection had a higher yield than culture and Calcofluor. Again, the result raised questions about whether the Calcofluor was detecting fungi not isolated in culture. In a subsequent study using the ITS region, the researchers found that the detection rate for T rubrum was 45.2% by targeting the ITS1 region in a PCR assay, higher than the 22.6% obtained by culture (P b .008).26 Although many of the published studies have focused on the diagnosis of fungal nail disease, molecular methods can equally be applied to infection of other skin sites11; likewise, there is potential to increase the speed of laboratory diagnosis. For instance, by using a two-step DNA extraction method and a multiplex PCR, the identification of fungi in nail samples could be achieved in 5 hours.27 The development of commercial tests for use in the diagnosis of fungal infection of nails and elsewhere is now a real possibility. One such test, called Onychodiag (BioAdvance, Bussy St. Martin, France), was designed to detect dermatophytes by using a PCR enzyme-linked immunosorbent assay in nail samples and was evaluated in 438 patients with suspected onychomycosis and 108 controls.28 The overall sensitivity of Onychodiag in three different laboratories was 83.6% compared with culture, and specificity was 100%; however, the test was positive in 68 samples that were culturally or microscopically positive for nondermatophytes. It is not clear if these were true positives due to the expected variation of culture yield in nail material or to overgrowth of

193 contaminants; however, these initial results are useful. Whether, in the long term, it will be best to use a dermatophyte generic test that identifies the main pathogens of nail without speciation or highly specific tests based on identification of DNA such as ITS region is yet to be seen. Although the use of molecular diagnostic techniques applied to the identification of fungi in fixed and paraffin embedded material is not common, this method has been used for the diagnosis of an unusual deep dermatophyte infection caused by T rubrum. The organism was identified using a nested PCR with Trichophyton-specific primers directed against the ITS1 gene. This allowed the investigator to find the causative organism without resorting to culture.29 Beyond diagnostic use, molecular methods have been of great value in identifying the epidemiologic spread of dermatophyte infections. To provide one example, an outbreak of T tonsurans infection amongst Japanese sportsmen, including wrestlers, was investigated using restriction enzyme analysis of PCR-amplified fragments targeting the NTS region of the ribosomal RNA gene.30 This showed that the NTS type 1 was usually the predominant strain in Japan, yet the heterogeneity of the NTS types suggested that many of these strains had been imported, perhaps as a result of international competitions. Similarly, these techniques have been used with great effect to examine issues in the pathogenesis of infection. A study relating inflammatory response in tinea capitis in children due to T tonsurans involved analyzing isolates of the fungus using 11 sequence variations in the rDNA and ALP1 loci showing a correlation between strain type and severity of inflammation. The closer the strains were to the ancestral genotype, the greater the inflammation.31

Malassezia Infections
The use of molecular methods has revolutionized the study of disease due to Malassezia spp. Because the organisms are fastidious and difficult to identify, the introduction of a new taxonomy in 1996 led to a series of revealing studies of the etiology of diseases linked to Malassezia spp.32 In the case of tinea versicolor, the first studies published by researchers in Spain33,34 showed that M globosa predominated in lesions of tinea versicolor compared with healthy or seborrheic skin. Others confirmed this work with molecular methods; for instance, studies from Japan35 using PCR identification of Malassezia showed that M globosa was the dominant variety in 97% of lesions. Likewise, researchers in Greece36 used DNA-PCR procedures on lesional skin scales of pityriasis versicolor to detect M globosa. A recurring observation in studies of tinea versicolor has been the suggestion that there are atypical clinical forms that, for instance, affect areas of the skin such as the groin or lower part of the abdomen and that in such lesions the dominant morphologic type of yeast is oval rather than the more typical

194 round yeasts and short stubby hyphae (spaghetti and meat balls). Recent work has shown that in some areas M furfur predominates in certain lesions of tinea versicolor.37,38 This introduces the interesting concept that the atypical types of tinea versicolor may be caused by a different species; in this case, the species that bears the same name as the yeast thought to be responsible for tinea versicolor in the times preceding the new taxonomy. These findings, however, have not yet been confirmed by molecular means. Molecular methods have been used to examine the complex relationships between Malassezia spp and the skin in other settings where the organisms are thought to cause human disease. For instance, in seborrheic dermatitis, the pattern of expression of different genes by Malassezia spp provided some explanation for the dependence of these organisms on lipids but also showed similarities with other epidermal pathogens, including Candida.39 So at present, although molecular tools have been used very successfully in the investigation of the pathogenesis and epidemiology of Malassezia infections, they have not been adopted as diagnostic tests. They may well be useful in the future for the diagnosis of some cases.

R.J. Hay, R.M. Jones reported a hybridization assay for rapid detection of Candida isolates from vaginal swabs that used a PCR method and primers targeting the ITS2 region from 28S rRNA. The six probes selected appeared to be specific, and the genetic findings agreed with the culture results in 97% of cases. The test also detected a higher number of Candida spp than culture alone or direct examination of smears; the question of whether this represented true infection or carriage remains unresolved.46 The other clinical area in the superficial infections where molecular methods have been described for Candida spp is in the diagnosis of onychomycosis.47 One study that used PCR-based methods and DNA extracted from nail identified C albicans from two samples, along with dermatophytes in others. The methods are available and of theoretic value, although the difficulty of relating the presence of DNA to real infection remains. In this later study, the two samples positive for Candida were negative by direct microscopy and culture.

Other fungi
The capability of using molecular techniques for the identification of less common molds in skin samples is already available. In general, the techniques have been developed for other diseases where these fungi are more common. For instance, a major cause of visual loss through mycotic keratitis is Fusarium infection. Fusarium spp may also cause superficial infections, including those affecting the feet and onychomycosis. Tests using gene targets such as the ITS regions have already been developed for the diagnosis of mycotic keratitis, and so the incorporation of the relevant probes into nail assays is feasible. One study reported the sequence variation in the ITS region was sufficient to identify 97% of the Fusarium spp encountered in medical microbiology laboratories.48

Candida infections
Molecular tools are well established as putative diagnostic methods in deep infections caused by Candida spp, even though they have still not been incorporated into routine clinical practice. Much of the initial work has involved testing Candida isolates from solid media or blood cultures. Methods used have involved different approaches, such as fluorescence in situ hybridization40,41 and amplification of target genes using electrophoretic methods or hybridization with nucleotide probes.42,43 These methods are somewhat time-consuming but are accurate and hold great promise. Particular focus has on the use of the ribosomal gene complex as a useful target for PCR assays because of high-sequence conservation in certain regions, for example, 18S coupled with high variability of intervening ITS regions. This provides a means of avoiding recurring problems of sensitivity and specificity encountered with other methods.44 The detection of markers of infection directly from blood and serum is a valid diagnostic goal, and its use has been reported in a number of different studies. For instance, a realtime PCR assay based on TaqMan probes (Applied Biosystems, Foster City, CA) capable of detecting Candida spp in serum because of the likelihood of fluconazole susceptibility or resistance (eg, C glabrata and C krusei), provided a high level of sensitivity coupled with the ability to detect as low as 0.22 copies/mL.45 Similar studies have also been published using variations on the technology; however, these methods have largely been used on systemic infections. There are examples, though, of the use of molecular diagnosis in superficial Candida infections. One study

Conclusions
Molecular biology appears to be a double-edged sword for many clinicians. It has resulted in a change and reordering of conventional systems of classification.49 We also stand on the brink of a new approach to diagnosis, where high-speed systems can be applied to the rapid identification of fungal infections affecting the skin, hair, and nails. Although comparatively costly at present, the price of such assays will fall with time, and it will be possible to take a swab sample from a paronychium and use that same specimen to identify the yeast as well as the bacterial flora present in the inflamed nail fold. The coalface of the microbiology laboratory will change. There will no longer be the need for a small corner of the laboratory where a dedicated microbiologist or mycologist

Molecular tools for fungal infection diagnosis will pore over cultures in intellectual isolation. Instead, technical staff and clinical scientists with understanding of the microbial genomes will perform these tests across all pathogenic organisms. At the same time, it will also be possible to refine identification processes so that small variations within the fungal genome can be detected and used to follow the spread of infections, their persistence after treatment, and to identify genetically phenotypic characteristics that are clinically important, such as enhanced virulence or drug resistance. Equally, it is important to interpret results with circumspection, because over-reliance on genetic homogeneity within a single gene region is not necessarily proof of identity, common ancestry perhaps.50 It is important for the future of medical mycology, as the science and practice of the study of fungi pathogenic in man, that both visions are accommodated as molecular methods advance. This may lead to changes in clinical practice because physicians will have to learn to be conversant with the advantages and limitations of the new methods and to take an active role in determining the relevance of the findings to the patient. We need to resolve the old dilemma, ever present in the interpretation of mycologic tests, of whether the presence of genetic evidence signifies infection, tissue penetration, or colonization. The presence of pathogens, such as dermatophytes, is normally assumed to provide evidence of infection; however, with the exception of hair samples where carriage without hair shaft invasion is known to occur, the significance of nondermatophytes or Candida spp, particularly in nail, remains uncertain. With the old technology, microscopy and histology provided some evidence to support the presence of invasion. With the new technology, similar support would come from detection of genes only expressed during the process of tissue invasion. The expression of genes encoding dipeptidyl peptidases IV and V by M canis51 or of T rubrum during growth on protein substrates52 are an example of induced gene expression suggesting tissue invasion. Work on such markers is already well underway, but there is still some way to travel before these techniques become a part of routine laboratory practice. In the meantime, projects such as the creation of a gene database for organisms like T rubrum53 can only help to promote our understanding of this complex, but important field.

195
5. Graser Y, el Fari M, Presber W, et al. Identification of common dermatophytes (Trichophyton, Microsporum, Epidermophyton) using polymerase chain reactions. Br J Dermatol 1998;138:576-82. 6. Okabayashi K, Kano R, Nakamura Y, et al. Molecular confirmation of a Trichophyton violaceum isolate from human black-dot ringworm. Mycopathologia 1999;146:127-30. 7. Howell SA, Barnard RJ, Humphreys F. Application of molecular typing methods to dermatophyte species that cause skin and nail infections. J Med Microbiol 1999;48:33-40. 8. Grser Y, Kuijpers AF, Presber W, et al. Molecular taxonomy of Trichophyton mentagrophytes and T. tonsurans. Med Mycol 1999;37: 315-30. 9. Grser Y, Kuijpers AF, Presber W, et al. Molecular taxonomy of the Trichophyton rubrum complex. J Clin Microbiol 2000;38:3329-36. 10. Jackson CJ, Barton RC, Kelly SL, et al. Strain identification of Trichophyton rubrum by specific amplification of subrepeat elements in the ribosomal DNA nontranscribed spacer. J Clin Microbiol 2000;38: 4527-34. 11. Park BC, Lee SJ, Kim DW, et al. Molecular identification of mycologic correlation of patients with concomitant tinea pedis and tinea manuum infection. Arch Dermatol 2009;145:205-7. 12. Chang JC, Hsu MM, Barton RC, et al. High-frequency intragenomic heterogeneity of the ribosomal DNA intergenic spacer region in Trichophyton violaceum. Eukaryot Cell 2008;7:721-6. 13. Guoling Y, Xiaohong Y, Jingrong L, et al. A study on stability of phenotype and genotype of Trichophyton rubrum. Mycopathologia 2006;161:205-12. 14. El Fari M, Tietz HJ, Presber W, et al. Development of an oligonucleotide probe specific for Trichophyton rubrum. Br J Dermatol 1999;141:240-5. 15. Kong F, Tong Z, Chen X, et al. Rapid identification and differentiation of Trichophyton species, based on sequence polymorphisms of the ribosomal internal transcribed spacer regions, by rolling-circle amplification. J Clin Microbiol 2008;46:1192-9. 16. Pihet M, Bourgeois H, Mazire JY, et al. Isolation of Trichophyton concentricum from chronic cutaneous lesions in patients from the Solomon Islands. Trans R Soc Trop Med Hyg 2008;102:389-93. 17. Faggi E, Pini G, Campisi E, et al. Application of PCR to distinguish common species of dermatophytes. J Clin Microbiol 2001;39:3382-5. 18. Kamiya A, Kikuchi A, Tomita Y, et al. PCR and PCR-RFLP techniques targeting the DNA topoisomerase II gene for rapid clinical diagnosis of the etiologic agent of dermatophytosis. J Dermatol Sci 2004;34:35-48. 19. Graser Y, Frohlich J, Presber W, et al. Microsatellite markers reveal geographic population differentiation in Trichophyton rubrum. J Med Microbiol 2007;56:1058-65. 20. Seyfarth F, Ziemer M, Graser Y, et al. Widespread tinea corporis caused by Trichophyton rubrum with non-typical cultural characteristics diagnosis via PCR. Mycoses 2007;50(suppl 2):26-30. 21. de Assis Santos D, de Carvalho Arajo RA, Kohler LM, et al. Molecular typing and antifungal susceptibility of Trichophyton rubrum isolates from patients with onychomycosis pre- and post-treatment. Int J Antimicrob Agents 2007;29:563-9. 22. Baek SC, Chae HJ, Houh D, et al. Detection and differentiation of causative fungi of onychomycosis using PCR amplification and restriction enzyme analysis. Int J Dermatol 1998;37:682-6. 23. Turin L, Riva F, Galbiati G, et al. Fast, simple and highly sensitive double-rounded polymerase chain reaction assay to detect medically relevant fungi in dermatological specimens. Eur J Clin Invest 2000;30: 511-8. 24. Liu D, Pearce L, Lilley G, et al. PCR identification of dermatophyte fungi Trichophyton rubrum, T. soudanense and T. gourvilii. J Med Microbiol 2007;51:117-22. 25. Gupta AK, Zaman M, Singh J. Diagnosis of Trichophyton rubrum from onychomycotic nail samples using polymerase chain reaction and calcofluor white microscopy. J Am Podiatr Med Assoc 2008;98:224-8. 26. Gupta AK, Zaman M, Singh J. Fast and sensitive detection of Trichophyton rubrum DNA from the nail samples of patients with

References
1. Borelli C, Beifuss B, Borelli S, et al. Conventional and molecular diagnosis of cutaneous mycoses. Hautarzt 2008;59:980-5. 2. Davison FD, Mackenzie DW. DNA homology studies in the taxonomy of dermatophytes. Sabouraudia 1984;22:117-23. 3. Kanbe T. Molecular approaches in the diagnosis of dermatophytosis. Mycopathologia 2008;166:307-17. 4. Liu D, Coloe S, Pedersen J, et al. Arbitrarily primed polymerase chain reaction to differentiate Trichophyton dermatophytes. FEMS Microbiol Lett 1996;136:147-50.

196
onychomycosis by a double-round polymerase chain reaction-based assay. Br J Dermatol 2007;157:698-703. Brillowska-Dabrowska A, Saunte DM, Arendrup MC. Five-hour diagnosis of dermatophyte nail infections with specific detection of Trichophyton rubrum. J Clin Microbiol 2007;45:1200-4. Savin C, Huck S, Rolland C, et al. Multicenter evaluation of a commercial PCR-enzyme-linked immunosorbent assay diagnostic kit (Onychodiag) for diagnosis of dermatophytic onychomycosis. J Clin Microbiol 2007;45:1205-10. Nagao K, Sugita T, Ouchi T, et al. Identification of Trichophyton rubrum by nested PCR analysis from paraffin embedded specimen in trichophytia profunda acuta of the glabrous skin. Nippon Ishinkin Gakkai Zasshi 2005;46:129-32. Mochizuki T, Kawasaki M, Anzawa K, et al. Epidemiology of sporadic (non-epidemic) cases of Trichophyton tonsurans infection in Japan based on PCR-RFLP analysis of non-transcribed spacer region of ribosomal RNA gene. Jpn J Infect Dis 2008;61:219-22. Abdel-Rahman SM, Talib N, Solidar A, et al. Examining Trichophyton tonsurans genotype and biochemical phenotype as determinants of disease severity in tinea capitis. Med Mycol 2008;46:217-23. Guho E, Midgley G, Guillot J. The genus Malassezia with description of four new species. Antonie Van Leeuwenhoek 1996;69:337-55. Crespo-Erchiga V, Ojeda A, Vera A, et al. Aislamiento e identificacion de Malassezia spp en pitiriasis versicolor, dermatitis seborreica y piel sana. Rev Iberoam Micol 1999;16(S):S16-21. Crespo-Erchiga V, Ojeda A, Vera A, et al. Mycology of pityriasis versicolor. J Mycol Med 1999;9:143-8. Nakabayashi A, Sei Y, Guillot J. Identification of Malassezia species isolated from patients with seborrhoeic dermatitis, atopic dermatitis, pityriasis versicolor and normal subjects. Med Mycol 2000;38:337-41. Gaitanis G, Velegraki A, Frangoulis E, et al. Identification of Malassezia species from patient skin scales by PCR-RFLP. Clin Microbiol Infect 2002;8:162-73. Razanakolona I, Rakotozandrindrainy N, Razafimahefa J, et al. Pityriasis versicolor Antananarivo: premire tude sur l'identification d'espces de Malassezia responsables. J Mycol Med 2004;14:152. Miranda KC, Rodrigues de Araujo C, Soares AJ, et al. Identificao de espcies de Malassezia em pacientes com pitirase versicolor em Goiania-GO. Rev Soc Bras Med Trop 2006;39:582-3. Xu J, Saunders CW, Hu P, et al. Dandruff-associated Malassezia genomes reveal convergent and divergent virulence traits shared with plant and human fungal pathogens. Proc Natl Acad Sci U S A 2007;104: 18730-5. Lischewski A, Amann RI, Harmsen D, et al. Specific detection of Candida albicans and Candida tropicalis by fluorescent in situ

R.J. Hay, R.M. Jones

hybridization with an 18S rRNA-targeted oligonucleotide probe. Microbiology 1996;142:2731-40. Rigby S, Procop GW, Haase G, et al. Fluorescence in situ hybridization with peptide nucleic acid probes for rapid identification of Candida albicans directly from blood culture bottles. J Clin Microbiol 2002;40: 2182-6. Playford EG, Kong F, Sun Y, et al. Simultaneous detection and identification of Candida, Aspergillus, and Cryptococcus species by reverse line blot hybridization. J Clin Microbiol 2006;44:876-80. Morace G, Sanguinetti M, Posteraro B, et al. Identification of various medically important Candida species in clinical specimens by PCRrestriction enzyme analysis. J Clin Microbiol 1997;35:667-72. Moreira-Oliveira MS, Mikami Y, Miyaji M, et al. Diagnosis of candidemia by polymerase chain reaction and blood culture: prospective study in a high-risk population and identification of variables associated with development of candidemia. Eur J Clin Microbiol Infect Dis 2005;24:721-6. McMullan R, Metwally L, Coyle PV, et al. A prospective clinical trial of a real-time polymerase chain reaction assay for the diagnosis of candidemia in nonneutropenic, critically ill adults. Clin Infect Dis 2008;46:890-6. Xiang H, Xiong L, Liu X, et al. Rapid simultaneous detection and identification of six species Candida using polymerase chain reaction and reverse line hybridization assay. J Microbiol Methods 2007;69: 282-7. Arca E, Saracli MA, Akar A, et al. Polymerase chain reaction in the diagnosis of onychomycosis. Eur J Dermatol 2004;14:52-5. Alfonso EC. Genotypic identification of Fusarium species from ocular sources: comparison to morphologic classification and antifungal sensitivity testing (an AOS thesis). Trans Am Ophthalmol Soc 2008;106:227-39. Graser Y, Scott J, Summerbell R. The new species concept in dermatophytesa polyphasic approach. Mycopathologia 2008;166: 239-56. Woodgyer A. The curious adventures of Trichophyton equinum in the realm of molecular biology: a modern fairy tale. Med Mycol 2004;42: 397-403. Vermout S, Baldo A, Tabart J, et al. Secreted dipeptidyl peptidases as potential virulence factors for Microsporum canis. FEMS Immunol Med Microbiol 2008;54:299-308. Zaugg C, Monod M, Weber J, et al. Gene expression profiling in the human pathogenic dermatophyte Trichophyton rubrum during growth on proteins. Eukaryot Cell 2009;8:241-50. Yang J, Chen L, Wang L, et al. TrED: the Trichophyton rubrum Expression Database. BMC Genomics 2007;8:250.

27.

41.

28.

42.

29.

43.

44.

30.

31.

45.

32. 33.

46.

34. 35.

47. 48.

36.

49.

37.

50.

38.

51.

39.

52.

40.

53.