Figures
Abstract
Purpose
The prognosis of people infected with Fungi especially immunocompromised depends on rapid and accurate diagnosis to capitalize on time administration of specific treatments. However, cultures produce false negative results and nucleic-acid amplification techniques require complex post-amplification procedures to differentiate relevant fungal types. The objective of this work was to develop a new diagnostic strategy based on real-time polymerase-chain reaction high-resolution melting analysis (PCR-HRM) that a) detects yeasts and filamentous Fungi, b) differentiates yeasts from filamentous Fungi, and c) discriminates among relevant species of yeasts.
Methods
PCR-HRM detection limits and specificity were assessed with a) isolated strains; b) human blood samples experimentally infected with Fungi; c) blood experimentally infected with other infectious agents; d) corneal scrapings from patients with suspected fungal keratitis (culture positive and negative) and e) scrapings from patients with suspected bacterial, viral or Acanthamoeba infections. The DNAs were extracted and mixed with primers diluted in the MeltDoctor® HRM Master Mix in 2 tubes, the first for yeasts, containing the forward primer CandUn (5'CATGCCTGTTTGAGCGTC) and the reverse primer FungUn (5'TCCTCCGCTT ATTGATATGCT) and the second for filamentous Fungi, containing the forward primer FilamUn (5'TGCCTGTCCGAGCGTCAT) and FungUn. Molecular probes were not necessary. The yields of DNA extraction and the PCR inhibitors were systematically monitored.
Results
PCR-HRM detected 0.1 Colony Forming Units (CFU)/µl of yeasts and filamentous Fungi, differentiated filamentous Fungi from yeasts and discriminated among relevant species of yeasts. PCR-HRM performances were higher than haemoculture and sensitivity and specificity was 100% for culture positive samples, detecting and characterizing Fungi in 7 out 10 culture negative suspected fungal keratitis.
Citation: Goldschmidt P, Degorge S, Che Sarria P, Benallaoua D, Semoun O, Borderie V, et al. (2012) New Strategy for Rapid Diagnosis and Characterization of Fungal Infections: The Example of Corneal Scrapings. PLoS ONE 7(7): e37660. https://doi.org/10.1371/journal.pone.0037660
Editor: Richard C. Willson, University of Houston, United States of America
Received: November 30, 2011; Accepted: April 23, 2012; Published: July 2, 2012
Copyright: © 2012 Goldschmidt et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was funded by the Laboratoire du Centre Hospitalier National d'Ophtalmologie des Quinze-Vingts, Ministry of Public Health, Paris, France. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
The frequency of fungal infections has been increasing for the last 30 years due to viral or iatrogenic immunodeficiencies, the efficiency in treating bacterial infections, the development of in-dwelling devices, and the massive use of contact lenses.[1]–[8] The incidence of fungal keratitis (keratomycosis) is also on the rise, and filamentous Fungi are the most frequently reported pathogens. [3], [8] From yeasts, Candida albicans is the most frequently associated with disease. However, C. glabrata, C. tropicalis, C. krusei, and C. parapsilosis have gained greater significance. [1], [2], [4], [8].
Fungal infection management requires timely diagnosis for rapid onset of treatments, but approximately one half of the samples remain culture negative and/or negative by fungal antigen detection using immunosorbent assays (ELISA).[9]–[10] Improved detection performances were reported by amplifying fungal genomic regions (polymerase chain reactions, PCRs). [11], [12] However, the classic PCRs do not differentiate filamentous Fungi from yeasts and require post amplification procedures (restriction enzyme digestion and analysis; single-base extension; hybridization probes or molecular sequencing).[11]–[15] The “gold standard” for fungal characterization is DNA sequencing, but this method is laborious, expensive and cannot be performed routinely for daily diagnosis. [15].
The real-time Taqman PCR using fluorogenic labelled Taqman-probes facilitates the detection and partial characterization of Fungi but requires a series of expensive labelled probes (each probe detects a single fungal type or one species per reaction). [12], [16]–[17].
Because the first-line therapy is different for filamentous Fungi and yeasts as well as for different yeasts, rapid and accurate information is required to target the treatments according to natural fungal susceptibilities.[18]–[20].
The availability of improved fluorescent DNA binding dyes with highly predictable saturation properties allows precise assessment of sequence length by High Resolution Melting real-time PCR (PCR-HRM). [21], [22] Recently, a diagnosis test based on PCR-HRM technology was reported for vaginal samples, detecting and identifying 8 Candida at species level. [23] Nevertheless, it was unable to differentiate Candida from filamentous Fungi and did not detect and characterise S. cervisiae and Trichosporon.
The goal of the present work is to develop a new test able to detect in 1 run the equivalent of at least 1 fungal colony forming unit (CFU) per reaction. In addition this molecular approach should simultaneously differentiate yeasts from filamentous Fungi and discriminate among relevant species of yeasts in clinical samples and in blood experimentally infected with fungal suspensions.
Materials and Methods
Investigations were conducted according to the principles expressed in the Declaration of Helsinki (http://www.wma.net/e/policy/) and were approved by the Institutional Review Board of the Centre Hospitalier National des Quinze-Vingts (CHNO), Ministry of Public Health, Paris-France. Written informed consent was obtained from all participants for the use of each sample. Forms with written consent were drafted according to the requirements of the CHNO Review Board and the National Health Authorities were double checked, validated and signed by the physician in charge of the sampling and sent to the laboratory. The preliminary studies were performed with characterized strains isolated from patients presenting corneal ulcers in the National Eye Hospital in Paris (CHNO des Quinze-Vingts) or from strains isolated from blood stream infections (generous gift from Dr Christophe Hennequin’s laboratory, CHU Saint-Antoine, Paris, France).
In red: Candida tropicalis; green: C. parapsilopsis; violet: C. albicans; black: C. glabrata; yellow: C. krusei; blue: Filamentous Fungi (Aspergillus fumigatus and Fusarium solani.).
In red: Aspergillus sp.; black: Fusarium solani; violet, blue and green: Candida sp.
One colony of each fungal species was scraped from the surface after 48 h of culture on Sabouraud’s dextrose agar, suspended in Phosphate Buffer Solution (PBS) and replated. To reduce the over representation of fungal DNA from non viable organisms, one colony was scraped from the second dish 48 hours later, suspended in PBS and tenfold diluted. Each dilution was divided in several aliquots; three were plated on Sabouraud’s dextrose agar to assess the number of colonies (equivalent CFU/ml) and the others kept as calibrators. For each series of experiments the PCR-HRM detection limits were validated with serial dilutions of fungal suspensions diluted in PBS and simultaneously titrated by plating.
Blood samples were collected in 10 ml citrate tubes from vein puncture of healthy subjects and transported to the laboratory within 1 h. After white cell count to assess the cell load of inoculums (white cell counts >12.000/µl were excluded) randomized aliquots were spiked with different titrated fungal suspensions. Negative controls consisted in non infected blood or leukocyte suspensions from the same individuals.
For each fungal strain, haemoculture bottles were inoculated with 10 ml of saline or blood spiked or not with Fungi and incubated up to 12 days before discarded. Fungal isolates were phenotypically characterized by conventional tests [Chromagar Candida BR, ref 257480 (Becton Dickinson, France); API 20AUX ref. 20210 (Biomérieux, France); API Candida ref 10500 (Biomérieux, France) and Lactophenol Blue, ref 363060-0125, RAL Advanced Chromatic, France)]. The confirmation of fungal species was carried out by sequencing the DNA from both ends using flanking vector primers. [12], [15] Bacteria were cultured and characterized with routine diagnosis tests; Acanthamoeba and Herpesviridae were detected by real-time PCR. [24], [25].
Corneal scrapings from 38 patients, 13 with proven fungal culture positive, 10 with suspected fungal keratitis culture negative and 15 with non suspected fungal keratitis (bacterial, viral or Acanthamoeba) were tested masked. Sampling from patients presenting corneal ulcers and requiring microbiological diagnosis was performed by deep corneal scraping by certified ophthalmologists with sterile stainless steel blades after rinsing of fluorescein and topical anaesthetic from the eye surface. [24] Slides with aliquots of scrapings were fixed and stained (Giemsa pH: 7.4) for direct microscopic examination and the presence of Fungi was confirmed by Grocott's methenamine silver reaction. The second aliquots were cultured within 30 minutes after collection up to 30 days before discarded as culture negative and remnants of blades were frozen dry at –80°C for further molecular diagnosis. PCR-HRM was carried out after thawing and addition of 200 µl of sterile Phosphate Buffer Solution (PBS) to the tubes containing the dry blades.
The DNA extraction was carried out in a vertical safety laminar flow cabinet in a dedicated room. To monitor the extraction yields and the absence of PCR inhibitors the internal control (IC) consisting of 5 µl of a whole virus preparation of seal herpes virus (gift from G. J. van Doornum, Dept. of Virology Erasmus MC, Rotterdam, The Netherlands) was added to 200 µl of each suspension (scraping, blood, leukocytes or saline) before extraction (final concentration of 1000 to 2000 viral particles/ml). [24], [25] In order to obtain spheroplasts, each specimen (sample + IC) was mixed with tris-EDTA buffer and 10 U recombinant lyticase (Sigma-Aldrich, France)/100 µl of suspension and incubated at 37°C for 60 min. After incubation, the suspensions were vortexed thoroughly and 100 µl were used for DNA extraction using the MagNA Pure compact nucleic acid isolation kit I® as described by the manufacturer in the MagNA Pure Compact automate® (Roche Diagnostics, Meylan, France) and eluted in 100 µl of elution buffer. To monitor the DNA extraction yields and the PCR inhibitors the seal herpes virus internal control (IC) was amplified in an independent real-time PCR run. [24], [25] The primer sequences were respectively: 5′GGGCGAATCACAGATTGA ATC and 5′GCGGTTCCAAACGTACCAA and VIC-TTTTTATGTGTCCGCCACCATCT GGATC-TAMRA for the probe. Amplification and detection of the IC was carried out in a separate tube containing 18.5 µl of the TaqMan® FAST Universal PCR Mastermix (2X no Amperase® UNG) (Applied Biosystems-France ABI Ref. 4352042), the forward and the reverse primers (0.5 uM each) with or without the fluorophore-labelled TaqMan® probe (0.5 uM). This solution was mixed with 5 µl of the DNA eluted in DNA and RNA-free solution. The PCR cycling program consisted of one cycle at 95°C for 20 sec and 45 cycles at 95°C for 3 sec and 30 sec at 60°C. [24].
Kits (stable at −20°C for at least 12 weeks) for fungal detection consist of 2 tubes, the first for detection, semi quantification and identification of yeasts; the second for detection and semi quantification of filamentous Fungi. Each tube contains 10 µl of MeltDoctor® HRM Master Mix (MDHRM) (ABI Applied Biosystems-France Ref 4415440), and 1 µL of the forward and 1 µl of the reverse primer, each at 300 nM (final concentration).
For the detection, quantification and characterization of yeasts and Filamentous Fungi the primers were selected in a region bracketing significant polymorphisms of multicopy ribosomal genes of the 18 S ribosomal RNA gene. The Primer 1: HRM CandUn1∶5′CATGCCTGTTTGAGCGTC (conserved sequences of yeasts,) and the Primer 2: HRM FungUn: 5′TCCTCCGCTTATTGATATGCT (conserved regions of all Fungi) allow obtaining profiles for the different yeasts according to the sizes of amplicons (alignment of sequences according to EMBL data library). The amplicon sizes (nucleotides bracketed by the primers CandUn + FungUn) are 189 for Candida albicans, 192 for C. dubliniensis (isolate M334a), 270 for C. Glabrata, 199 for Issatchenkia orientalis (C. krusei), 269 for C. nivariensis (isolate VPCI 1293), 166 for C. metapsilosis (strain CBS-2916), 162 for C. parapsilopsis, 179 for C. tropicalis and 225 for C. zeylanoides (strain TJY13a 2).
For filamentous Fungi the selected sequences for HRM are FilamUn: 5′TGCCTGTTCCGAGCGTCAT (forward primer) and HRM FungUn: 5′TCCTCCGCTTAT TGATATGCT. The amplicons sizes are 189 for A. versicolor and A. heteromorphus, 190 for A. sidowi; A. carbonarius and Aspergillus sp., 191 for A. oryzae and A. niger, and 192 for A. brasiliensis; A. flavus; A. toxicariu and A. bombycis. For Fusarium napiforme and for F. solani (isolates FMR 799; FMR 4389; 4391; FMR 7338-42; FMR 7991; FMR 7993; FMR 7989; 7994; FMR isolates 7995 to 8000 and FMR 8013) the amplicons are sizes are ranged between 194 and 196 nucleotides. For F. polyphialidicum; F. redolens; F. proliferatum; F. proliferatum; F. beomiforme and F. fujikuroi amplicon sizes are 206, and for F. proliferatum; F. fujikuroi; F. dlaminii 201 nucleotides.
For PCR-HRM the DNA extracts (10 µl) were introduced in 2 tubes, the first containing CandUn + FungUn in the MDHRM the second FilamUn + FungUn. During PCR-HRM, the amplicons were automatically measured in a closed tube format using integrated cycler/fluorimeter ABI 7500 upgraded equipment and monitored using fluorescent DNA intercalating dyes present in the MDHRM. The PCR program started with a denaturation of 10 min at 95°C, followed by 55 cycles of amplification (15 s at 95°C, 30 s at 60°C and 30 s at 72°C). The PCR-HRM curve was obtained by denaturation at 95°C for 15 sec, cooling to 50°C for 1 min and a temperature increase until 60°C for 15 sec with a 2.2°C/s ramp rate. Samples with fluorescence of less than the 100% of the maximum were excluded from the analysis. Each run contained negative controls with no template and DNA extracts from the reactants. Linearity, sensitivity and detection limit (Equivalent CFU/ml from the Ct versus dilution curves) and reproducibility were assessed by diluting fungal suspensions in distilled water before DNA extraction. The melting temperature (Tm) at which 50% of the DNA is in the double stranded state was assessed by taking the derivative of the melting curve. The melting curves shapes depend on PCR product (amplicon) length. The DNA patterns of the derivative plot (difference plot) were used for amplicon analysis.
Results
Preliminary experiments were performed to assess the best conditions for extraction of DNA from spores: a- heat for 10 min at 94°C; b- proteinase K at 37°C for 60 min and heat at 94°C for 10 min; c- proteinase K at 37°C for 60 min, heat at 94°C for 10 min and extraction with the MagNA Pure compact nucleic acid isolation kit I® as described by the manufacturer in the MagNA Pure Compact® automate (Roche Diagnostics, Meylan, France); d- shaked in presence of beads and extraction with MagNA Pure; e- shaked in presence of beads with or without proteinase K at 37°C for 60 min, heat at 94°C for 10 min and extraction with Magna Pure; f- shaked in presence of beads with or without lyticase at 37°C for 60 min and heat at 94°C for 10 min; g- shaked in presence of beads with lyticase at 37°C for 60 min, heat at 94°C for 10 min and extracted with Magna Pure; or h- lyticase at 37°C for 60 min, heat at 94°C for 10 min and extraction with Magna Pure. The highest fungal DNA extraction rates were obtained using the procedures g or h (results not shown).
The detection limits have been obtained by dilution of fresh titrated fungal suspensions. PCR-HRM with the primers CandUn + FungUn detected 0.1 CFU/µl of Candida albicans, C. krusei, C. glabrata, C. tropicalis, Saccharomyces cervisiae and Trichosporon and 1 CFU/ml of filamentous Fungi suspended in PBS. As shown in Table 1 PCR-HRM detection capacities were repeatedly higher for yeasts (10 to 100 times) using the set CandUn + FungUn, and more than 10 times higher for filamentous Fungi using the set FilamUn + FungUn (Aspergillus nidulans, A. niger, A. versicolor, A. terreus, Penicillium piccum and Fusarium solani). These results suggest that the optimization of fungal detection requires the simultaneous amplification of DNA extracts in 2 tubes, the first with the set CandUn + FungUn and the second with FilamUn + FungUn. Under these conditions, the PCR-HRM coefficient of variation for the interassay reproducibility in the complete linear range of detection [105 to 10−1 colony forming units (CFU)/ml] for 5 runs was less than 10%. The patterns of the first derivative (difference plot) permitted differentiation of yeasts from filamentous Fungi and the divergence between amplicon sizes of closely related species allowed PCR-HRM to easily discriminate among yeasts (Figure 1). According to the amplicon sizes bracketed by the set of primers FilamUn + FungUn (generally ≥194 nucleotides for Fusarium sp. versus ≤192 for most species of Aspergillus) the melting curve shapes were repeatedly different for Fusarium solani (Figure 2).
Sensitivity and specificity of PCR-HRM (while comparing with corneal scraping cultures) was of 100%. PCR-HRM allowed rapid diagnosis of keratomycosis differentiating clinical relevant species of yeasts, and produced negative results for all the samples obtained from patients with non suspected fungal keratitis and for all the negative controls [DNA extracted from 106 CFU/ml of Bacteria, 106 PFU (plaque forming units)/ml of Herpes simplex virus type 1 or 106 PFU/ml of Herpes simplex virus type 2, and 105 Acanthamoeba cysts/ml suspended in saline] (Table 2). Human cells (106 human epithelial cells or fibroblasts) did not interfere with the PCR-HRM performances, confirming the in-silico specificity predictions. In patients with clinically suspected fungal keratitis (samples 1; 3; 5; 6; 8; 10; 12; 13; 17; 18; 20; 23; 24) culture was positive in 55% and images evoking Fungi were detected in 65% of the clinical samples by direct microscopic examination (Table 2). In 4 out 10 patients with clinically suspected fungal keratitis and culture negative (samples 29–38), Fungi were detected by direct microscopic examination of deep corneal scrapings (filaments in 2; budding yeasts in 1 and pseudohypha in 1) (samples 29; 31; 32 and 36). In addition, for these same 10 culture negative patients, PCR-HRM detected and characterized Fungi in 7 out of 10, including the 4 detected by direct microscopic examination. The relative high number of cornea negative cultures could be partially the result of residual eye drop preservatives carried with the samples from the eye surface. PCR-HRM was negative for all the controls carried out with the samples obtained from the air, surfaces, reactants and the blood of healthy donors (total blood or buffy coats) (Table 2).
Table 3 shows the recovery and detection time for haemoculture bottles spiked with Fungi. For fungal inoculums containing 100 CFU/bottle, the time for positivity was ranged between 16 and 36 hours of incubation for yeasts and between 36–48 hours for filamentous Fungi. Positivity was obtained after 24 to 48 hours of incubation of blood spiked with yeasts and after 24 to 72 hours with filamentous Fungi. For inoculums of 10 CFU/bottle the time for positivity was ranged between 24–48 hours for yeasts and between 48 and 72 hours for filamentous Fungi. For yeasts suspended in blood cultures were positive after 24–48 hours and after 24–72 h for filamentous Fungi. Only C. albicans, Trichosporon and Penicillium piccum could be detected for inoculums containing 1 CFU per bottle (after 96 h of culture) in saline. Inoculums containing 1 CFU per bottle or less of 7 different yeasts or 4 filamentous Fungi suspended in blood were negative. PCR-HRM detected 100% of the samples containing the equivalent of 0.1 CFU/ml of yeasts and filamentous Fungi and generated reproducible melt-curves. When challenged against profiles obtained from referenced strains run in parallel the melting profiles obtained with all the samples allowed differentiating yeasts from filamentous Fungi and discriminating among 7 different strains of yeasts. PCR-HRM performances were equivalent for Fungi suspended in saline or blood at concentrations significantly lower (10 times or more) than the detection limits of fungal cultures (Table 3).
Discussion
The automatic melting analysis of fungal sequences amplified with 2 sets of primers diluted in a mix containing a DNA intercalating dye (SYTO9) allowed rapid detection of Fungi. The differences in amplicon sizes between species were suited to fungal differentiation of yeasts from filamentous Fungi and to speciation among yeasts. By adapting the existing real-time PCR instrumentation for data acquisition it was possible to carry out reproducible diagnosis in less than 2.30 h after DNA extraction, with detection limits of at least 0.1 CFU of filamentous Fungi and yeasts per µl of sample with no need for molecular probes (radioactive, enzymatic or fluorogenic) or post amplification procedures (sequencing, amplicon restriction enzyme analysis, etc.). Corneal samples could be readily assayed after DNA extraction without interference from other DNAs found in the specimens.
Fungal culture performances depend on the type of agent, the fungal load and the mass of material that can be processed, the microbiology laboratory capacities and the presence of antiseptic or antifungal in the samples. Moreover, cultures rely on the ability of the organism to grow ex vivo and may require long incubation periods (may become positive late in the course of the infection), are time consuming and prone to contamination. [1], [4], [11].
In 89% of patients with suspected fungal keratitis PCR-HRM was able to detect Fungi differentiating filamentous Fungi and yeasts. Interestingly, in all patients with positive PCR-HRM the corneal infiltrates were dramatically reduced after antifungal treatments, suggesting that PCR-HRM signals represented true positive infections (results not shown). Local and topical antifungal treatments are long lasting, potentially toxic, and first-line therapies have to selectively target yeasts or filamentous Fungi.[4]–[6], [27]–[29] On this matter, yeasts but not filamentous Fungi require Fluconazole and/or 5-Fluorocytosine (5FC). Aspergilli require Voriconazole and/or Caspofungin; Candida krusei are intrinsically resistant to Fluconazole and Candida glabrata and Candida tropicalis have unpredictable susceptibilities to this latter.[4]–[6], [27]–[29] Hence, false negative diagnosis or lack of discrimination among major fungal types may have a direct impact on infection management (morbidity and mortality).
The yeast detection capacities of the haemoculture system used in the present study are equivalent to those reported for other automated systems (inoculums containing the equivalent of 100 and10 CFU/bottle). Nevertheless, all the blood samples containing fungal loads of 1 CFU/bottle were negative. For automated blood culture bottles inoculated with 1000 yeasts it was shown that the BacT detected growth of 90% of Candida, while Bactec 66%. [6], [7] The mean time to growth detection was between 25.62 h and 27.30 h and both detected experimentally infected blood (simulated candidemia) only when additional specialized mycology media were used. [6] For filamentous Fungi the BACTEC plus Aerobic/F and the BACTEC Mycosis-IC/F automated systems followed by subculture in solid medium detected inoculums of A. fumigatus at concentrations of >3 conidia per 10 ml after 21–40 h, with longer incubation periods for A. flavus and A. terreus. [7].
For the detection and identification of Candida and Aspergillus species a variety of real-time PCR assays based on species specific probes were developed targeting the 18 S rDNA [16], [30], the mitochondrial cytochrome b [31], the 18 S and 28 S rDNA [32], the CaMP65 (65-kDa mannoprotein) gene [33], the RNase P RNA gene [34], and the ITS 2 region [35]. These nucleic-acid amplification techniques are independent of microorganisms’ growth and detect stressed, injured, and fastidious Fungi and those suspended in antimicrobials.[11]–[14] Despite this, expectations for routine direct fungal diagnosis are not fulfilled with PCRs because they a) cover either yeasts or filamentous Fungi, b) do not to detect C. lusitaniae and c) may require complex post amplification procedures and/or the synthesis and stability testing of labelled molecular probes.[11]–[17], [26] Therefore, DNA sequencing was almost the only tool for molecular species assessment of PCR-positive samples.[36]–[37].
For diagnosis of candidemia in subjects with haematological malignancies or various forms of immunodeficiency a real-time PCR targeting the 18 S rRNA gene and requiring a series of labelled molecular probes, yielded positive results in 58.3% of blood culture-positive samples and detected in blood the genomes of Candida 3 days earlier than culture. [37] In this series 27% of whole-blood were PCR positive compared to 15% of haemoculture (92% of correlation of positives). Other studies indicate that numerous pairs of primers and labelled probes were required for each sample to identify 72% of species of positive cultures. [38] For vaginal specimens it was reported a test based on PCR-HRM technology, but it was unable to differentiate yeasts from filamentous Fungi. [22] However, the strategy developed here is different because PCR-HRM is able to detect and differentiate yeasts and filamentous Fungi in one run with detection limits of 0.1 CFU/µl or less. In addition to differentiating filamentous Fungi from yeasts, PCR-HRM discriminated among relevant clinical species of yeasts in less than 2.30 hours after DNA extraction. This only required upgrading the available real-time PCR software of the thermocyclers used in the routine microbiology laboratory.
Because PCR-HRM was able to produce consistent results without the need for synthesizing and labelling molecular probes and without post amplification procedures (restriction enzymes, electrophoresis, gel analysis, hybridisation, sequencing reactants) the cost for reactants for testing one DNA extract could be reduced to less than 2 USD (2 primers and HRM mix). Compared to classic PCR (amplification followed by electrophoreses and/or hybridization and/or sequencing) this new PCR-HRM has the additional advantage of minimizing risks for false positive results due to cross contamination, because the targeted sequence amplification, the signal detection, and the DNA melting analyses are carried out in closed tubes. Moreover, PCR-HRM minimizes risks for false negative results because the yields of extraction of the DNA and the potential interference of PCR inhibitors are systematically monitored in each run and for all the samples. According to the results obtained in this study, if the future runs generate reproducible melt-curves over time in different settings, a reference database could be built to store PCR-HRM calculations and shapes of the melting profiles for each family or species to be challenged against profiles. Larger prospective multicentric trials testing different types of samples (clinical and environmental) are necessary to validate PCR-HRM usefulness as a diagnosis tool and for environmental studies.
Acknowledgments
We thank the Centre de Ressources Biologiques du Centre Hospitalier National d’Ophtalmologie des Quinze-Vingts, 75012 Paris, France, for providing biological samples.
Author Contributions
Conceived and designed the experiments: PG SD. Performed the experiments: PG SD PCS DB OS. Analyzed the data: PG SD DB OS CC. Contributed reagents/materials/analysis tools: PG PC VB LL CC. Wrote the paper: PG.
References
- 1. Edmond MB, Wallace S, McClish D, Pfaller MA, Jones RN, et al. (1999) Nosocomial bloodstream infections in United States hospitals: a three-year analysis. Clin Infect Dis 29: 239–44.
- 2. Ascioglu S, Rex JH, de Pauw B, Bennett JE, Bille J, et al. (2002) Invasive Fungal Infections Cooperative Group of the European Organization for Research and Treatment of Cancer, Mycoses Study Group of the National Institute of Allergy and Infectious Diseases. Defining opportunistic invasive fungal infections in immunocompromised patients with cancer and hematopoietic stem cell transplants: an international consensus. Clin Infect Dis 34: 7–14.
- 3. Szczotka-Flynn LB, Pearlman E, Ghannoum M (2010) Microbial contamination of contact lenses, lens care solutions and their accessories: a literature review. Eye Contact Lens 36: 116–29.
- 4. Marr KA, Carter R, Crippa F, Wald A, Corey L (2002) Epidemiology and outcome of mould infections in hematopoietic stem cell transplant recipients. Clin Infect Dis 34: 909–17.
- 5. Pagano L, Caira M, Candoni A, Offidani M, Fianchi L, et al. (2006) The epidemiology of fungal infections in patients with hematologic malignancies: the SEIFEM-2004 study. Haematologica 91: 1068–75.
- 6. Horvath L, George B, Murray C, Harrison L, Hospenthal D (2004) Direct comparison of the BACTEC 9240 and BacT/ALERT 3D automated blood culture systems for Candida growth detection. J Clin Microbiol 42: 115–8.
- 7. Rosa C, Araujo R, Rodrigues A, Pinto-de-Sousa M, Pina-Vaz C (2011) Detection of Aspergillus species in BACTEC blood cultures. J Med Microbiol 60: 1467–71.
- 8. Hall BJ, Jones L (2010) Contact lens cases: the missing link in contact lens safety? Eye Contact Lens 36: 101–5.
- 9. Allan EK, Jordanides NE, McLintock LA, Copland M, Devaney M, et al. (2005) Poor performance of galactomannan and mannan sandwich enzyme-linked immunosorbent assays in the diagnosis of invasive fungal infection. Br J Haematol. 128: 578–9.
- 10. Maertens J, Verhaegen J, Lagrou K, Van Eldere J, Boogaerts M (2001) Screening for circulating galactomannan as a noninvasive diagnostic tool for invasive aspergillosis in prolonged neutropenic patients and stem cell transplantation recipients: a prospective validation. Blood 97: 1604–10.
- 11. Jordanides NE, Allan EK, McLintock LA, Copland M, Devaney M, et al. (2005) A prospective study of real-time panfungal PCR for the early diagnosis of invasive fungal infection in haemato-oncology patients. Bone Marrow Transplant 35: 389–95.
- 12. Basková L, Landlinger C, Preuner S, Lion T (2007) The Pan-AC assay: a single-reaction real-time PCR test for quantitative detection of a broad range of Aspergillus and Candida species. J Med Microbiol 56: 1167–73.
- 13. Ririe K, Rasmussen R, Wittwer C (1997) Product differentiation by analysis of DNA melting curves during the polymerase chain reaction. Anal Biochem 245: 154–60.
- 14. Loeffler J, Hebart H, Magga S, Schmidt D, Klingspor L, et al. (2000) Identification of rare Candida species and other yeasts by polymerase chain reaction and slot blot hybridization. Diagn Microbiol Infect Dis 38: 207–12.
- 15. Rakeman J, Bui U, Lafe K, Chen C, Honeycutt T, et al. (2005) Multilocus DNA sequence comparisons rapidly identify pathogenic molds. J Clin Microbiol 43: 3324–33.
- 16. Klingspor L, Jalal S (2006) Molecular detection and identification of Candida and Aspergillus spp. from clinical samples using real-time PCR. Clin Microbiol Infect 12: 745–53.
- 17. Schabereiter-Gurtner C, Selitsch B, Rotter M, Hirschl A, Willinger B (2007) Development of novel real-time PCR assays for detection and differentiation of eleven medically important Aspergillus and Candida species in clinical specimens. J Clin Microbiol 45: 906–14.
- 18. Ostrosky-Zeichner L, Rex J, Pappas P, Hamill R, Larsen R (2003) Antifungal susceptibility survey of 2,000 bloodstream Candida isolates in the United States. Antimicrob Agents Chemother 47: 3149–54.
- 19. Srinivasan M (2004) Fungal keratitis. Curr Opin Ophthalmol 15: 321–7.
- 20. White PL, Archer A, Barnes R (2005) Comparison of non-culture-based methods for detection of systemic fungal infections, with an emphasis on invasive Candida infections. J Clin Microbiol 43: 2181–7.
- 21. Monis P, Giglio S, Saint C (2005) Comparison of SYT09 and SYBR Green I for real-time polymerase chain reaction and investigation of the effect of dye concentration on amplification and DNA melting curve analysis. Anal Biochem 340: 24–34.
- 22. Reed G, Kent JO, Wittwer C (2007) High-resolution DNA melting analysis for simple and efficient molecular diagnostics. Pharmacogenomics 8: 597–608.
- 23. Mandviwala T, Shinde R, Kalra A, Sobel JD, Akins R (2010) High-throughput identification and quantification of Candida species using high resolution derivative melt analysis of panfungal amplicons. J Mol Diagn 12: 91–101.
- 24. Goldschmidt P, Rostane H, Saint-Jean C, Batellier L, Alouch C, et al. (2006) Effects of topical anaesthetics and fluorescein on the real-time PCR used for the diagnosis of Herpesviruses and Acanthamoeba keratitis. Br J Ophthalmol 90: 1354–6.
- 25. van Doornum G, Guldemeester J, Osterhaus A, Niesters H (2003) Diagnosing herpesvirus infections by real-time amplification and rapid culture. J Clin Microbiol 41: 576–80.
- 26. Selvarangan R, Bui U, Limaye, Cookson B (2003) Rapid identification of commonly encountered Candida species directly from blood culture bottles. J Clin Microbiol 41: 5660–4.
- 27. Vanden Bossche H, Dromer F, Improvisi I, Lozano-Chiu M, Rex H, et al. (1998) Antifungal drug resistance in pathogenic fungi. Med Mycol 36: (suppl)119–28.
- 28. Sullivan D, Haynes K, Bille J, Boerlin P, Rodero L, et al. (1997) Widespread geographic distribution of oral Candida dubliniensis strains in human immunodeficiency virus-infected individuals. J Clin Microbiol 35: 960–4.
- 29. Fidel Jr P, Vazquez J, Sobel J (1999) Candida glabrata: review of epidemiology, pathogenesis, and clinical disease with comparison to C. albicans. Clin Microbiol Rev 12: 80–96.
- 30. Gomez-Lopez A, Martin-Gomez M, Martin-Davila P, Lopez-Onrubia P, Gavalda J, et al. (2006) Detection of fungal DNA by real-time polymerase chain reaction: evaluation of 2 methodologies in experimental pulmonary aspergillosis. Diagn. Microbiol. Infect. Dis 56: 387–393.
- 31. Spiess B, Buchheidt D, Baust C, Skladny H, Seifarth W, et al. (2003) Development of a LightCycler PCR assay for detection and quantification of Aspergillus fumigatus DNA in clinical samples from neutropenic patients. J Clin Microbiol 41: 1811–1818.
- 32. Hsu M, Chen K, Lo H, Chen Y, Liao M, et al. (2003) Species identification of medically important fungi by use of realtime LightCycler PCR. J. Med. Microbiol 52: 1071–1076.
- 33. Arancia S, Carattoli A, La Valle R, Cassone A, De Bernardis F (2006) Use of 65 kDa mannoprotein gene primers in real time PCR identification of Candida albicans in biological samples. Mol. Cell Probes 20: 263–268.
- 34. Innings A, Ullberg M, Johansson A, Rubin C, Noreus N, et al. (2007) Multiplex real-time PCR targeting the RNase P RNA gene for detection and identification of Candida species in blood. J. Clin. Microbiol 45: 874–880.
- 35. Schabereiter-Gurtner C, Selitsch B, Rotter M, Hirschl A, Willinger B (2007) Development of novel real-time PCR assays for detection and differentiation of 11 medically important Aspergillus and Candida species in clinical specimens. J Clin Microbiol 45: 906–914.
- 36. Vollmer T, Störmer M, Kleesiek K, Dreier J (2008) Evaluation of novel broad-range real-time PCR assay for rapid detection of human pathogenic fungi in various clinical specimens. J Clin Microbiol 46: 1919–26.
- 37. Wellinghausen N, Siegel D, Winter J, Gebert S (2009) Rapid diagnosis of candidaemia by real-time PCR detection of Candida DNA in blood samples. J Med Microbiol 58: 1106–11.
- 38. Obara H, Aikawa N, Hasegawa N, Hori S, Ikeda Y, et al. (2011) The role of a real-time PCR technology for rapid detection and identification of bacterial and fungal pathogens in whole-blood samples. J Infect Chemother 3: 327–33.