Development of a new MLST scheme for differentiation of Fusarium solani Species Complex (FSSC) isolates

https://doi.org/10.1016/j.mimet.2010.07.008Get rights and content

Abstract

Fungi belonging to the Fusarium solani Species Complex (FSSC) are well known plant pathogens. In addition to being the causative agent of some superficial infections, FSSC has recently emerged as a group of common opportunistic moulds, mainly in patients with haematological malignancies. Molecular typing methods are essential in order to better understand the epidemiology of such opportunistic agents with the final goal of preventing contamination. A three-locus typing scheme has thus been developed for FSSC; based on polymorphisms in the domains of the ITS, EF-1 alpha, and RPB2 genes. This method is now considered to be a useful reference for phylogenetic and taxonomic studies. In other significant clinical fungi (e.g., Candida sp., Cryptococcus neoformans, and Aspergillus fumigatus), genes coding for metabolic enzymes have been widely used and proven to be very informative for diagnosis and epidemiology. The contribution of these genes has never been evaluated for Fusarium sp. and more specifically for F. solani Species Complex.

Here, we have evaluated the contribution of 25 genes for diagnosis and epidemiological purposes. We then report a new five-locus MLST scheme useful for diagnosis and typing of clinical FSSC isolates. The method has been validated on 51 epidemiologically unrelated strains of FSSC and presents a high power of discrimination calculated at 0.991.

Introduction

Members of the Fusarium solani Species Complex (FSSC) are common saprophytes, frequently isolated from environmental sources such as soil, air and plants. These fungi are also well known as plant, animal and human pathogens and account for approximatively two-thirds of all human and animal fusariosis worldwide (De Hoog et al., 2004, Dignani and Anaissie, 2004). During the last decade, F. solani has been increasingly reported in humans, causing localised cutaneous infections in immunocompetent hosts (e.g., onychomycosis, keratitis or endophtalmitis), usually associated with trauma (Chang et al., 2006). In immunocompromised patients, superficial or subcutaneous lesions are susceptible to disseminate and induce serious invasive mycoses with a high mortality rate, ranging from 50% to 80% (Nucci and Anaissie, 2007). Presently, F. solani has emerged as the second most common opportunistic pathogenic mould (after Aspergillus), mainly in patients with haematological malignancies, in recipients of solid organ and allogenic bone marrow or stem cell transplants (Boutati and Anaissie, 1997, Nucci and Anaissie, 2007). More specifically, neutropenia, lymphopenia, graft-versus host disease, corticosteroid therapy or any other immunosuppressive treatments represent high risk factors for disseminated fusariosis (Guarro and Gene, 1995, Nucci and Anaissie, 2007).

Due to their increasing clinical relevance, several molecular phylogenetic studies have been performed involving sequence typing, restriction fragment length polymorphism and microsatellite analysis (Dyavaiah et al., 2007, Godoy et al., 2004, Mehl and Epstein, 2007, Zhang et al., 2006). Multilocus sequence typing (MLST) is a highly accurate method used to distinguish between isolates of microbial species. MLST was first developed to facilitate studies of epidemiology and population structure in several bacterial populations (Maiden et al., 1998). This method compares nucleotide polymorphisms within five to seven gene regions, traditionally housekeeping genes. The different polymorphisms giving rise to allelic variants are recorded and the resulting combinations correspond to the strain sequence types (ST). Because MLST data can be easily accessible at a global scale through dedicated websites, it has the advantage of allowing multiple users to compare their results. MLST has already been used to investigate populations of human pathogenic fungi, including Candida albicans (Bougnoux et al., 2002), Candida glabrata (Dodgson et al., 2003), Candida tropicalis (Tavanti et al., 2005), Candida krusei (Jacobsen et al., 2007), Cryptococcus neoformans var. gattii (Feng et al., 2008, Meyer et al., 2009), C. neoformans var. grubii (Litvintseva et al., 2006, Meyer et al., 2009) and Aspergillus fumigatus (Bain et al., 2007).

Recently, a three-locus typing scheme has also been developed for the F. solani Species Complex that is based on polymorphisms in portions of the internal transcribed spacer region and domains D1 plus D2 of the nuclear large-subunit rRNA, the translation elongation factor 1 alpha gene (EF-1alpha), and the second largest subunit of the RNA polymerase II gene (RPB2). This multilocus method is particularly useful for phylogenetic and taxonomic studies and has led to a new nomenclature of FSSC containing three main clades. Twenty phylogenetically distinct species have thus been found in the medical FSSC clade 3 (Zhang et al., 2006, O'Donnell et al., 2008). By using this typing method, authors theorise that species-level studies using additional phylogenetically informative data may lead to the discovery of unsuspected cryptic species (O'Donnell et al., 2008). In other clinical fungi, MLST schemes have been developed using metabolic genes, for example glyceraldehyde 3P deshydrogenase (Feng et al., 2008) or isocitrate lyase (Tavanti et al., 2005). Although widely used for fungal MLST, these metabolic genes have never been evaluated for the epidemiology of F. solani.

In the present study, we aim to evaluate the contribution of 25 genes to specifically differentiate clinical FSSC isolates. A new five-locus MLST scheme has thus been described for the typing of these fungi. Tested over a panel of 51 epidemiologically unrelated strains, the method was shown to be reproducible, generates a stable profile and presents a power of discrimination calculated at 0.991. These characteristics made this method suitable for tracing strains in a hospital context.

Section snippets

Isolates

Fifty-one epidemiologically unrelated isolates of F. solani Species Complex were used in this study. These strains originated from clinical, plant and environmental sources and were obtained from our own collection or provided by collaborators. All strains were deposited at the CBS-KNAW Fungal Biodiversity Centre (http://www.cbs.knaw.nl). The corresponding collection numbers and the characteristics of all strains are listed in Table 1.

Strains were stored at − 80 °C in water/glycerol (1/1) and

Selection of genes for F. solani MLST

Amplification of each of the 25 initially selected genes was attempted on 5 isolates. Among them, eight were successfully amplified for F. solani strains and then sequenced (Table 2). Five show high variability whereas three were not very variable. Based on this highest inter-strain discrimination, we selected the five most variable loci to build our MLST scheme: FsACC, FsICL, FsGPD, FsMPD and FsSOD that encode respectively for acetylCoenzyme A carboxylase, isocitrate lyase, glyceraldehyde 3P

Discussion

Typing methods contribute greatly to the understanding of the epidemiology of infections and the evolution of pathogens. They are therefore important for monitoring infectious disease outbreaks. Two are predominant because of their high level of standardisation: microsatellite and MLST. Among them, MLST has previously been shown to be highly resolutive for epidemiological and population structure analysis of several fungi (Meyer et al., 2009). This technique is based on the identification of

Acknowledgments

We are grateful to M. Kombila, M.L. Darde and L. Delhaes for kindly providing isolates included in this study, and to the staff of the CBS collection service for providing and processing isolates. We thank Tiffany Dunham for having checked the English.

References (24)

  • X. Feng et al.

    Genotype and mating type analysis of Cryptococcus neoformans and Cryptococcus gattii isolates from China that mainly originated from non-HIV-infected-patients

    FEMS Yeast Res.

    (2008)
  • P. Godoy et al.

    Genotyping of 44 isolates of Fusarium solani, the main agent of fungal keratitis in Brazil

    J. Clin. Microbiol.

    (2004)
  • Cited by (26)

    • In vitro susceptibility and multilocus sequence typing of Fusarium isolates causing keratitis

      2018, Journal de Mycologie Medicale
      Citation Excerpt :

      Identification of Fusarium species from human samples is important not only from an epidemiological viewpoint, but also for choosing the appropriate antifungal agent for difficult-to- treat infections, such as fungal keratitis. MLST has previously been shown to be highly resolutive for epidemiological and population structure analysis of several fungi including Fusarium [20]. Specific regions of TEF1 and RPB2 have good discriminatory power and are often used in MLST schemes [21,22].

    • Cluster analysis of Scedosporium boydii infections in a single hospital

      2015, International Journal of Medical Microbiology
      Citation Excerpt :

      Hospital-related outbreaks due to fungal pathogens other than Candida or Aspergillus species have been rarely investigated by using molecular methods (Etienne et al., 2011; Repetto et al., 2012; Ruiz-Diez et al., 1997). MLST is described as a method with a high discriminatory power for S. aurantiacum and the Fusarium solani species complex (Debourgogne et al., 2010; Harun et al., 2009). MLST had turned out to be a reproducible and informative tool also for S. apiospermum and S. boydii in a previous report (Bernhardt et al., 2013).

    • Multilocus sequence typing of Scedosporium apiospermum and Pseudallescheria boydii isolates from cystic fibrosis patients

      2013, Journal of Cystic Fibrosis
      Citation Excerpt :

      For epidemiological studies it is necessary to exactly identify strains on a subspecies level. MLST has been proven to give highly reproducible genotyping results that are comparable worldwide [11–14]. Therefore we decided to use MLST in studying S. apiospermum and P. boydii as the most frequently isolated species in the CF context within the S. apiospermum species complex in Germany.

    View all citing articles on Scopus
    View full text