MALDI-TOF MS identification of anaerobic bacteria: assessment of pre-analytical variables and specimen preparation techniques

Abstract

Matrix-assisted laser desorption ionization–time of flight mass spectrometry (MALDI-TOF MS) has emerged as a tool for identifying clinically relevant anaerobes. We evaluated the analytical performance characteristics of the Bruker Microflex with Biotyper 3.0 software system for identification of anaerobes and examined the impact of direct formic acid (FA) treatment and other pre-analytical factors on MALDI-TOF MS performance. A collection of 101 anaerobic bacteria were evaluated, including Clostridium spp., Propionibacterium spp., Fusobacterium spp., Bacteroides spp., and other anaerobic bacterial of clinical relevance. The results of our study indicate that an on-target extraction with 100% FA improves the rate of accurate identification without introducing misidentification (P < 0.05). In addition, we modify the reporting cutoffs for the Biotyper “score” yielding acceptable identification. We found that a score of ≥1.700 can maximize the rate of identification. Of interest, MALDI-TOF MS can correctly identify anaerobes grown in suboptimal conditions, such as on selective culture media and following oxygen exposure. In conclusion, we report on a number of simple and cost-effective pre- and post-analytical modifications could enhance MALDI-TOF MS identification for anaerobic bacteria.

Introduction

Anaerobic bacteria are an important component of the normal flora of the human gut, skin, and mucosal surfaces (Norin, 2011). There is an emerging interest in understanding the microbiome of anaerobic bacteria in these body niches and the relationship of the community of anaerobic microorganisms to various disease states (Bartlett, 2012, Brook, 2010, McCollum and Rodriguez, 2012, Siqueira and Oocas, 2013). In addition to the importance of anaerobic microorganisms in human normal flora, infections due to anaerobic organisms occur in both children and adults; these may be the result of traumatic injuries and wounds. Furthermore, immunosuppressed patients with malignancies or undergoing hematopoietic stem cell transplantation and older adults are more susceptible to these infections, which often arise from translocation of endogenous flora (Chen et al., 2013, El-Sharif et al., 2012, Ngo et al., 2013). Identification of the specific causative agent can be important in optimization of antimicrobial therapy and medical management, especially if the organism is a toxin producer (such as Clostridium perfringens) or is known to have a specific medical association (such as Clostridium septicum, associated with gastrointestinal neoplasm). It has been demonstrated that delayed or inaccurate identification of anaerobic organisms during infection can lead to increased morbidity and mortality in these patients (Brook, 2010). Thus, rapid and accurate methods for identification of anaerobic organisms are necessary in a clinical setting.

Traditional methods for the identification of clinically relevant anaerobes are cumbersome, time-consuming, and expensive. This is largely attributed to the fact that these organisms are frequently slowly growing, and their identification involves a complex workup, including an aerotolerance test and phenotypic organism characterization for identification. This can be challenging due to the fact that in addition to the relatively slow growth rate for these organisms, they are frequently biochemically inert (Nagy et al., 2006).

Several biochemical and molecular techniques have been developed to improve the identification of clinical anaerobes (Nagy et al., 2006). Recent studies have demonstrated that 16S rRNA gene sequencing is useful for identifying clinically relevant anaerobic bacteria and/or resolving discrepant identifications (Drancourt et al., 2000, Petti et al., 2008, Watson et al., 2010). However, this technique has a relatively slow turnaround time, is technically challenging, and costly, making its integration into routine laboratory identification schema impractical. Recently, a growing body of literature illustrates the utility of matrix-assisted laser desorption ionization–time of flight mass spectrometry (MALDI-TOF MS) for the identification of various microbes grown on solid medium and in positive blood culture broth (Bizzini et al., 2010, Christner et al., 2010, Prod’hom et al., 2010). In some studies, it has been suggested that MALDI-TOF MS can be a functional replacement for routine 16S rRNA gene sequencing for identification of anaerobic organisms (Bizzini et al., 2011). MALDI-TOF MS has been demonstrated to be relatively easy to integrate into the routine microbiology laboratory as a result of the rapid results produced by this method and the fact that the technique is relatively easy to perform. However, the analytical performance characteristics of this method are somewhat variable, depending on the instrumentation, database, and technique employed (Fournier et al., 2012, Garner et al., 2013, Justesen et al., 2011, Nagy et al., 2012, Veloo et al., 2011). To date, data on MALDI-TOF MS identification of clinically relevant anaerobes are still somewhat scarce compared to aerobic organisms such as the Gram-positive cocci (GPC) and Enterobacteriaceae (Christensen et al., 2012, Ford and Burnham, 2013, Saffert et al., 2011, Schmitt et al., 2013, Tekippe et al., 2013).

The objective of this study was to evaluate MALDI-TOF MS for identification of clinically relevant anaerobic bacteria and to systematically assess variables impacting the rate and accuracy of identification of these organisms, such as the method of application to the MALDI target plate, growth medium, and incubation environment, as well as identify the optimal reporting parameters for accurate species level identifications.