HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Articles by Gutzmer, R. Articles by Kapp, A. Search for Related Content PubMed PubMed Citation Articles by Gutzmer, R. Articles by Kapp, A. Agricola Articles by Gutzmer, R. Articles by Kapp, A.

Rapid identification and differentiation of fungal DNA in dermatological specimens by LightCycler PCR

Department of Dermatology and Allergology, Hannover Medical University, Ricklinger Str. 5, D-30449 Hannover, Germany

Correspondence Ralf Gutzmer rgutzmer{at}gmx.de

Received June 16, 2004
Accepted August 12, 2004

The aim was to develop a LightCycler PCR method for the rapid detection and differentiation of fungal DNA in dermatological specimens such as skin scales and skin swabs. LightCycler PCR assays were established for seven primer sets specific for fungal DNA. For each primer set LightCycler melting points were defined by amplification of DNA from 21 fungi and sensitivity was determined by amplification of serial dilutions of fungal DNA. A protocol was established that allows detection and differentiation of mould and yeast DNA with one highly sensitive PCR reaction by assessment of LightCycler melting points. Two subsequent LightCycler PCR reactions and one RFLP reaction allowed the differentiation of dermatophytes and non-dermatophyte moulds and the subclassification of yeasts. Analysis of clinical samples from 38 patients with fungal skin diseases provided conclusive new diagnostic information in 9/38 cases (23.7 %) by this PCR protocol that was not equally provided by direct microscopy and mycological culture. Thus the LightCycler PCR protocol established here represents a rapid diagnostic tool that aids in the diagnosis of fungal skin disease in a substantial number of patients.

	INTRODUCTION

TOP INTRODUCTION METHODS RESULTS AND DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
Fungal infections have to be considered in the differential diagnosis of many dermatological problems such as nail dystrophy, paronychia or hair loss. The detection of fungal growth in mycological culture and by direct microscopy of clinical specimens is the gold standard for the diagnosis of fungal infection. Culture regularly requires incubation periods of 2–4 weeks before a definitive result is obtained. For the microscopic identification of fungi, trained personnel are needed, but atypical morphology might still obscure the identification. Moreover, for the classification of some fungi, additional methods are necessary (e.g. the evaluation of biochemical properties in the classification of Candida species).

Molecular characterization of fungal DNA has been employed to detect and differentiate fungi (Gil-Lamaignere et al., 2003). Molecular methods are faster, more sensitive, more stable and less dependent on external factors than morphological methods (Faggi et al., 2001; Graser et al., 1998; Kim et al., 2001).

Our aim was to establish a LightCycler PCR protocol for the detection and differentiation of a broad spectrum of dermatologically relevant fungal DNA that allows the rapid and sensitive investigation of clinical specimens and aids in the diagnosis of fungal skin infection.

	METHODS

TOP INTRODUCTION METHODS RESULTS AND DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
Fungal strains, fungal culture and clinical specimens.

The following fungal strains were purchased from the Deutsche Sammlung von Mikroorganismen und Zellkulturen (DSMZ, Braunschweig, Germany; DSMZ number in parentheses): Aspergillus niger (1988), Fusarium oxysporum (2018), Candida parapsilosis (5784), Candida guilliermondii (6381), Candida tropicalis (11953), Candida krusei (6128), Malassezia furfur (6171), Rhodotorula mucilaginosa (70825), Hyphopichia burtonii (Trichosporon behrendii, 70663), Microsporum canis (10708), Microsporum gypseum (3824), Microsporum audouinii (10649), Epidermophyton floccosum (10709), Trichophyton verrucosum (7380) and Trichophyton tonsurans (12285). The other strains used in this study were cultured from clinical samples and characterized in the mycological laboratory of our department by standard methods. Yeasts were further characterized according to their carbohydrate assimilation pattern using the API-32C kit (bioMérieux).

To compare the results of PCR examination and mycological culture in the assessment of clinical samples, scales or skin swabs were obtained. The scales were immediately divided into three portions for direct microscopic examination, mycological culture and PCR (see below). Alternatively, two skin swabs were obtained in parallel for mycological culture and PCR, respectively.

Selection criteria for patients with fungal skin disease were (a) a clinical presentation consistent with fungal disease, (b) a relief of symptoms by antifungal therapy and (c) no history of antifungal therapy within the last 3 months. Samples were obtained before start of therapy. Patients with clinically clear-cut eczema or psoriasis served as negative controls.

Extraction of fungal DNA and dilution experiments.

DNA from fungal cultures, human peripheral blood mononuclear cells (PBMCs), Staphylococcus aureus and clinical specimens was extracted using the QIAamp DNA isolation kit (Qiagen).

To assess the detection limit of fungal DNA with LightCycler PCR, DNA concentrations from selected samples were adjusted to 1 pg µl^–1 and diluted with water in 10-fold dilutions. These samples were subjected to LightCycler PCR and melting curve analysis as well as agarose gel electrophoresis, and the highest dilution was recorded that resulted in a peak in the melting curve analysis or a visible band on the gel.

Detection of fungal DNA by LightCycler PCR.

For the amplification of fungal DNA, a variety of primer sets was used that had been previously described in the literature. Details with regard to primer sequences, amplified products and references are given in Table 1.

LightCycler PCR was performed on a LightCycler (Roche Diagnostics) using the LightCycler Fast Start DNA Master SYBR Green (Roche Diagnostics) and the conditions summarized in Table 1.

To control for a correct amplification product length, the PCR reaction was also analysed by electrophoresis on 1.5 % agarose gels. Amplicons obtained with primer sets TR1–TR2 and N5–N6 contained restriction sites for HaeIII that were characteristic for dermatophytes, yeasts and moulds (Baek et al., 1998). Therefore, these PCR products were digested with HaeIII (MBI Fermentas) and analysed by agarose gel electrophoresis. The following DNA fragment patterns were expected: TR1–TR2, dermatophytes 442/90/49 bp, yeasts 437/142 bp, moulds 441/59/50/30 bp; N5–N6, dermatophytes 147/90/63 bp, yeasts 163/144 bp, moulds 146/71/59/30 bp.

	RESULTS AND DISCUSSION

TOP INTRODUCTION METHODS RESULTS AND DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
LightCycler PCR reactions were established for seven primer pairs located in the fungal rDNA and internal transcribed spacers (ITSs) regions (Fig. 1) that were specific for fungal DNA as previously characterized by others (Table 1). LightCycler PCR for each primer set was investigated for its ability to differentiate dermatologically relevant fungi via melting curve analysis and its sensitivity to amplify fungal DNA. Based on these data, a system for the evaluation of clinical samples was developed and used to analyse clinical specimens in comparison to results from mycological cultures and direct microscopy.

View larger version (12K): [in this window] [in a new window]   Fig. 1. Location of the primers used in this study within the rDNA and ITS regions. V1–V9 denote the variable regions of the 18S rDNA subunit (V6 exists in prokaryotes only). Adapted from Kappe et al. (1996), Lindsley et al. (2001), Machouart-Dubach et al. (2001) and White et al. (1990).

Determination of LightCycler PCR melting points for characterized fungal species

LightCycler PCR was performed with DNA extracted from 21 characterized fungal cultures. Representative experiments are shown in Fig. 2 and melting points obtained from three independent experiments are given in Table 2. Agarose gel electrophoresis of LightCycler PCR products and of HaeIII-digested PCR products revealed bands of the expected sizes (Table 1, Fig. 2).

View larger version (36K): [in this window] [in a new window]   Fig. 2. Representative LightCycler PCR experiments for selected primer sets. Amplicons were analysed in parallel by melting curves and agarose gel electrophoresis. Lanes: L, 100 bp DNA ladder; 1, T. rubrum; 2, E. floccosum; 3, Microsporum canis; 4, Penicillium sp.; 5, A. niger; 6, C. albicans; 7, PBMCs; 8, Staph. aureus; 9, water.

Primer sets N5–N6, ITS86–ITS4, UF1–EU1 and TR1–TR2 amplified moulds and yeasts, whereas 18SF1–5.8SR1 amplified primarily and S3–TR2 amplified exclusively yeasts and DH2L–DH1R amplified dermatophytes (Table 2). Melting points obtained with primer set N5–N6 allowed the differentiation of moulds (dermatophytes and non-dermatophytes, around 87–88 °C) and yeasts (below 86 °C). Melting curve analysis of UF1–EU1 amplicons showed a slight difference between moulds (around 86 °C) and yeasts (around 85 °C). Differences in melting points obtained with primers TR1–TR2 were too little to allow the determination of fungi at the genus level.

Thus differentiation of dermatophytes and non-dermatophyte moulds was possible in two ways: first, via additional restriction digest of the N5–N6 amplicon with HaeIII, since they have different RFLP patterns (Table 1, Fig. 2); second, via LightCycler melting curve analysis of ITS86–ITS4 amplicons (Table 2).

None of the primer sets was able to subclassify dermatophytes and non-dermatophyte moulds since melting points and product lengths were close together. Also other studies were unable to differentiate dermatophyte and non-dermatophyte mould species by analysis of amplification products derived from the rDNA and ITS regions using agarose gel electrophoresis (Baek et al., 1998; Bock et al., 1994; Machouart-Dubach et al., 2001; Makimura et al., 1998, 1999; Turin et al., 2000) or real-time PCR (Kami et al., 2001; Pham et al., 2003).

For yeasts, the two PCR reactions with ITS86–ITS4 and 18S–5.8S primers allowed the differentiation of Candida albicans versus Candida glabrata versus C. guilliermondii/Rhodotorula mucilaginosa versus C. parapsilosis/C. tropicalis versus C. krusei (Table 2).

The ITS regions and adjacent rDNA were also successfully used in previous studies to detect and differentiate yeasts to the species level with a variety of PCR methods (Ahmad et al., 2002; Chen et al., 2000; Elie et al., 1998; Lindsley et al., 2001; Selvarangan et al., 2002; Shin et al., 1997). Differences in the variability of the rDNA and ITS base pair composition are a likely explanation for the ability to differentiate yeasts but not dermatophytes or non-dermatophyte moulds (Graser et al., 1999; Ninet et al., 2003). This is highlighted by the study of Turenne et al. (1999), which examined the length of the ITS2 region amplified with primers ITS86–ITS4. Candida species covered a range of 268–359 bp, dermatophytes a range of 303–321 bp with the exception of E. floccosum with 366 bp, and moulds (Aspergillus and Penicillium species) 283–292 bp. The variability for Candida species is much higher than for dermatophytes and moulds. However, E. floccosum (which yielded a considerably longer amplification product than other dermatophytes) also had a similar melting peak to other dermatophytes in the LightCycler analysis, which underlines the fact that the LightCycler melting curve depends on the length and the base pair composition of an amplicon.

Assessment of the sensitivity of various PCR primers in the detection of fungi by LightCycler PCR

Serial 10-fold dilutions of DNA from Trichophyton rubrum and C. albicans were used to determine the sensitivity of the seven LightCycler PCR reactions in the detection of fungal DNA (Table 2). The lowest detection level for both T. rubrum and C. albicans DNA was obtained with primer set N5–N6. Agarose gel electrophoresis was as equally sensitive as LightCycler melting curve analysis.

Evaluation of clinical specimens with the LightCycler PCR protocol

Based on the results obtained with characterized fungal DNA, the following procedure to evaluate clinical specimens was established. The most sensitive primer set, N5–N6, was used to screen specimens for the presence of fungal DNA. In the case of mould DNA, discrimination of dermatophyte and non-dermatophyte moulds was attempted either by restriction digest of N5–N6 amplicons and assessment of RFLP by agarose gel electrophoresis or by LightCycler melting point determination with primers ITS86–ITS4. In the case of yeast DNA, subclassification was performed by melting curve analysis of PCR reactions with primer sets 18SF1–5.8SR1 and ITS86–ITS4. In cases where only the (most sensitive) melting point for N5–N6 primers but not the (less sensitive) RFLP pattern or melting points of other PCR reactions was available, the statement ‘fungal DNA’ without further specification was made.

Thirty-eight clinical samples from patients with fungal skin infection (Table 3) and 20 samples from patients with eczema (15 patients) and psoriasis (5 patients) were evaluated. Specimens obtained from the latter 20 patients were all negative by PCR and direct microscopy. In two psoriasis patients culture yielded growth of non-dermatophyte moulds; the other 18 patients were negative by culture.

Among the 38 positive patients, skin scales were investigated by direct microscopy, culture and PCR in patients 1–28 and skin swabs were investigated by culture and PCR in patients 29–38 (Table 3). Thirty-three out of 38 (86.8 %) samples were positive by at least one of the three methods. PCR amplification with primers N5–N6 yielded positive results in 32/38 (84 %) specimens by melting point analysis. After restriction digest with HaeIII, in only 22 of these 32 samples (69 %) were bands visible by agarose gel electrophoresis that allowed further classification into dermatophytes, non-dermatophyte moulds and yeasts (Table 3). Thus melting point detection is much more sensitive than detection of bands of 30–163 bp length on agarose gels after restriction digestion.

LightCycler PCR provided conclusive new diagnostic information (that was not given by microscopy or culture) in 9/38 cases (23.7 %, cases 9, 10, 11, 12, 15, 19, 20, 37, 38) and interesting but not conclusive new information in another seven cases (18.4 %, cases 13, 14, 18, 21, 34, 35, 36). Thus results obtained by our PCR protocol allowed a laboratory diagnosis that was not equally well made by culture and microscopy alone in a substantial number of patients.

In 10 cases with detection of yeasts by both PCR and culture (Table 3), yeast classification was concordant in seven cases (70 %). In case 18, PCR detected C. albicans DNA and culture revealed growth of C. krusei. In cases 33 and 34, PCR demonstrated DNA of C. parapsilosis or tropicalis (which could not be differentiated) and culture detected C. albicans and C. krusei, respectively. A rate of 70 % concordant results comparing molecular to biochemical identification of Candida species corresponds well to a previous study that demonstrated a 100 % correct identification of Candida species by DNA fingerprinting as compared to 77.5 % by the API-32C system, which was also used for the phenotypic characterization of Candida species in our study (Latouche et al., 1997).

In conclusion, we developed a fast and simple LightCycler-based PCR protocol to detect DNA of the dermatologically most relevant fungi and to differentiate dermatophytes, yeasts and non-dermatophyte moulds. This protocol can complement mycological culture and direct microscopy in the diagnosis of fungal skin disease and provides additional diagnostic information in a substantial number of patients. However, it can not replace culture and direct microscopy, since dermatophytes as well as non-dermatophyte moulds can not be differentiated at the species level and no information on viability is obtained.

	ACKNOWLEDGEMENTS

TOP INTRODUCTION METHODS RESULTS AND DISCUSSION ACKNOWLEDGEMENTS REFERENCES

 
We greatly appreciate the technical assistance of the staff of the mycological laboratory of our department, namely Mrs Petra Kopp and Mrs Elke Schulz.

	REFERENCES

TOP INTRODUCTION METHODS RESULTS AND DISCUSSION ACKNOWLEDGEMENTS REFERENCES