Mycology
Rapid identification of fungal pathogens in BacT/ALERT, BACTEC, and BBL MGIT media using polymerase chain reaction and DNA sequencing of the internal transcribed spacer regions

https://doi.org/10.1016/j.diagmicrobio.2005.11.002Get rights and content

Abstract

We report a direct polymerase chain reaction/sequence (d-PCRS)-based method for the rapid identification of clinically significant fungi from 5 different types of commercial broth enrichment media inoculated with clinical specimens. Media including BacT/ALERT FA (BioMérieux, Marcy l'Etoile, France) (n = 87), BACTEC Plus Aerobic/F (Becton Dickinson, Microbiology Systems, Sparks, MD) (n = 16), BACTEC Peds Plus/F (Becton Dickinson) (n = 15), BACTEC Lytic/10 Anaerobic/F (Becton Dickinson) (n = 11) bottles, and BBL MGIT (Becton Dickinson) (n = 11) were inoculated with specimens from 138 patients. A universal DNA extraction method was used combining a novel pretreatment step to remove PCR inhibitors with a column-based DNA extraction kit. Target sequences in the noncoding internal transcribed spacer regions of the rRNA gene were amplified by PCR and sequenced using a rapid (24 h) automated capillary electrophoresis system. Using sequence alignment software, fungi were identified by sequence similarity with sequences derived from isolates identified by upper-level reference laboratories or isolates defined as ex-type strains. We identified Candida albicans (n = 14), Candida parapsilosis (n = 8), Candida glabrata (n = 7), Candida krusei (n = 2), Scedosporium prolificans (n = 4), and 1 each of Candida orthopsilosis, Candida dubliniensis, Candida kefyr, Candida tropicalis, Candida guilliermondii, Saccharomyces cerevisiae, Cryptococcus neoformans, Aspergillus fumigatus, Histoplasma capsulatum, and Malassezia pachydermatis by d-PCRS analysis. All d-PCRS identifications from positive broths were in agreement with the final species identification of the isolates grown from subculture. Earlier identification of fungi using d-PCRS may facilitate prompt and more appropriate antifungal therapy.

Introduction

The incidence of invasive fungal infections (IFIs) has increased significantly in the past decade as a direct consequence of increasing patient populations at risk for developing serious fungal infections (Hajjeh et al., 2004, Walsh et al., 2004). Opportunistic fungi other than Candida albicans and Aspergillus fumigatus are being reported with increased frequency from blood stream infections (BSIs) and IFIs (Pfaller and Diekema, 2004). These include non-albicans Candida species, such as Candida glabrata, Candida parapsilosis, Candida tropicalis, and Candida krusei, which account for more than 50% of all BSIs caused by Candida (Rangel-Frausto et al., 1999). Other fungi increasingly encountered include opportunistic yeast-like fungi such as Trichosporon spp., Rhodotorula spp., Geotrichum capitatum, and filamentous fungi such as Scedosporium spp., Fusarium spp., Acremonium spp., and Mucorales (Pfaller and Diekema, 2004, Walsh et al., 2004). In addition, certain species of fungi are associated with a high degree of mortality (Gudlaugsson et al., 2003), and some are inherently less susceptible to standard antifungal therapy (Pfaller and Diekema, 2004). The rapid and reliable detection and subsequent identification of fungi from blood and other important clinical specimens remain critical in deciding whether to initiate antifungal therapy and in the choice of agent used.

Continuously monitored automated blood culture systems have improved the detection and isolation of fungi from patients with BSIs compared with manual methods of culture (Nolte et al., 1993, O'Hara et al., 2003). Furthermore, inoculation of blood culture media with other types of fluid or tissue samples has been widely implemented in clinical microbiology laboratories to improve recovery of some fungi (Thomson and Miller, 2003). Regardless of the type of sample and system used, the laboratory handling of positive cultures is usually the same; the presence of fungi is confirmed by microscopy and the broth is subcultured onto solid mycological media for phenotypic identification. However, many fungi are slow growing, and current methods of fungal identification are labor-intensive, may lack specificity, and require a wide range of specialized laboratory media and specialized trained staff (Pfaller and Fromtling, 2003).

Numerous molecular methods have been developed to rapidly identify fungi from solid media (Luo and Mitchell, 2002, Shin et al., 1999) and directly from positive blood culture fluids (Borst et al., 2001, Chang et al., 2001, Li et al., 2003, Shin et al., 1997). Polymerase chain reaction (PCR) methods for identifying fungi directly from blood cultures usually target a few species within the genus Candida. Molecular targets for PCR amplification and detection are usually conserved nucleotide sequences within phylogenetically informative genetic regions such as the rRNA gene complex. Ribosomal RNA genes are also targets for sequence-based identification of fungi (Chen et al., 2001, Hall et al., 2003, Hall et al., 2004, Henry et al., 2000, Pryce et al., 2003). PCR amplification using universal primers targeted to conserved regions within the rRNA complex, followed by DNA sequencing of the internal transcribed spacer (ITS) regions, shows promise to identify a broad range of fungi to the species level (Chen et al., 2001, Henry et al., 2000, Iwen et al., 2002, Pryce et al., 2003).

We previously reported a rapid (24 h) sequence-based approach to identify clinically important yeasts and filamentous fungi from the first visible signs of growth on solid media (Pryce et al., 2003). The aim of this investigation was to evaluate a similar approach for the rapid and accurate identification of fungi directly from commercial broth enrichment media. To our knowledge, we report for the first time the use of DNA sequencing for the rapid identification of fungi directly from commercial broth enrichment media. In addition, we report a reliable DNA extraction method to isolate fungal nucleic acids from a range of commercial broth enrichment media commonly used in diagnostic microbiology laboratories.

Section snippets

Clinical samples and controls

A total of 140 samples were collected from 138 patients from the Royal Perth Hospital and the Princess Margaret Hospital for Children, Perth, Australia. Clinical specimens included blood (n = 121), vitreous fluid (n = 3), synovial fluid (n = 2), peritoneal fluid (n = 3), skin biopsy (n = 1), and a variety of other tissues and fluids (n = 10). Commercial broth enrichment media including BacT/ALERT FA (BioMérieux, Marcy l'Etoile, France), BACTEC Plus Aerobic/F (Becton Dickinson, Microbiology

Assessment of the specificity of the PCR method and performance of the DNA extraction methods used

All broth enrichment media determined to be culture-positive for fungi were positive by PCR (n = 45). All isolates grown from subculture were also positive by PCR (n = 46). All positive samples except from patient 5 (Cryptococcus neoformans) were successfully amplified using PCR-A (Table 1). The positive sample from patient 5 was successfully amplified using PCR-B. A single PCR product was observed for all PCR-positive samples (data not shown). PCR products were obtained from a single DNA

Discussion

Given that methods for recovering fungi from blood, fluids, and tissues have improved with continuously monitored blood culture systems (Nolte et al., 1993, O'Hara et al., 2003, Thomson and Miller, 2003), it is reasonable to assume that uncommon fungi are likely to be encountered more frequently in the clinical microbiology laboratory (Groll and Walsh, 2001). The routine identification of some fungi can be time consuming. Isolates from blood cultures are recovered by subculture and are

Acknowledgments

We thank the staff at the West Australian Genome Resource Centre at Royal Perth Hospital for their expertise, technical assistance, and performing the sequencing reactions for this study.

References (39)

  • G.S. de Hoog et al.

    Atlas of clinical fungi

    (2000)
  • D.N. Fredricks et al.

    Improved amplification of microbial DNA from blood cultures by removal of the PCR inhibitor sodium polyanetholesulfonate

    J. Clin. Microbiol.

    (1998)
  • O. Gudlaugsson et al.

    Attributable mortality of candidemia, revisited

    Clin. Infect. Dis.

    (2003)
  • R.A. Hajjeh et al.

    Incidence of bloodstream infections due to Candida species and in vitro susceptibilities of isolates collected from 1998 to 2000 in a population-based active surveillance program

    J. Clin. Microbiol.

    (2004)
  • L. Hall et al.

    Experience with the MicroSeq D2 large subunit ribosomal DNA sequencing kit for identification of commonly encountered, clinically important yeast species

    J. Clin. Microbiol.

    (2003)
  • L. Hall et al.

    Experience with the MicroSeq D2 large-subunit ribosomal DNA sequencing kit for identification of filamentous fungi encountered in the clinical laboratory

    J. Clin. Microbiol.

    (2004)
  • K.C. Hazen et al.

    Candida, Cryptococcus and other yeasts of medical importance

  • C. Hennequin et al.

    Identification of Fusarium species involved in human infections by 28S rRNA gene sequencing

    J. Clin. Microbiol.

    (1999)
  • T. Henry et al.

    Identification of Aspergillus species using internal transcribed spacer regions 1 and 2

    J. Clin. Microbiol.

    (2000)
  • Cited by (0)

    View full text