Evaluation of the Bruker Biotyper Matrix-Assisted Laser Desorption Ionization–Time of Flight Mass Spectrometry System for Identification of Blood Isolates of Vibrio Species

Abstract

Among 56 blood isolates of Vibrio species identified by sequencing analysis of 16S rRNA and rpoB genes, the Bruker Biotyper matrix-assisted laser desorption ionization–time of flight mass spectrometry (MALDI-TOF MS) system correctly identified all isolates of Vibrio vulnificus (n = 20), V. parahaemolyticus (n = 2), and V. fluvialis (n = 1) but none of the isolates of serogroup non-O1/O139 (non-serogroup O1, non-O139) V. cholerae (n = 33) to the species level. All of these serogroup non-O1/O139 V. cholerae isolates were correctly identified using the newly created MALDI-TOF MS database.

TEXT

Vibrio cholerae causes an estimated 3 to 5 million cases of cholera, which result in 100,000 to 120,000 deaths each year (1). Vibrio cholerae and non-cholerae Vibrio species, especially V. vulnificus, V. parahaemolyticus, V. fluvialis, and V. alginolyticus, have been reported to also cause severe gastroenteritis, severe skin soft tissue infection (necrotizing fasciitis), biliary tract infections, septicemia, and other serious infections in immunocompromised and immunocompetent hosts (2–9). Given the potential of V. cholerae for rapid epidemic spread and its listing as a biothreat level B agent, it is crucial for clinical microbiology laboratories to be able to correctly identify and differentiate V. cholerae from other bacterial species, particularly Aeromonas species (10).

Identification of human-pathogenic Vibrio species is traditionally based on conventional biochemical reactions and commercial identification systems (11–14), including API (API 20E), Vitek 2 (ID-GN card), and Phoenix (NMIC/ID-72) systems (11–14). However, the ability of these commercially available identification systems to correctly identify Vibrio species is limited because of the high phenotypic diversity of this species (12, 14). Partial sequencing analysis of the 16S rRNA and rpoB genes is a commonly used identification method (15).

A number of studies have shown that matrix-assisted laser desorption ionization–time of flight mass spectrometry (MALDI-TOF MS) can rapidly and accurately identify bacteria and fungi to the species level (16). Our previous study demonstrated that the Bruker Biotyper MALDI-TOF MS system, with implementation of a newly created database, can accurately identify other important human-enteropathogenic aeromonads, such as Aeromonas dhakensis, A. hydrophila, A. veronii, and A. caviae (17). However, few studies have investigated the ability of the Bruker Biotyper MALDI-TOF MS system to accurately identify pathogenic Vibrio species, particularly V. cholerae (10, 18–20). In the present study, we evaluated the ability of the Bruker Biotyper MALDI-TOF MS system to accurately identify genetically confirmed Vibrio species that were recovered from patients with bloodstream infections.

A total of 50 isolates of Vibrio species that were recovered from patients with bloodstream infections at National Taiwan University Hospital (NTUH; northern Taiwan) during the period 2004 to 2013 were obtained from the hospital's microbiology laboratory. An additional 6 isolates of V. cholerae obtained from patients with bloodstream infections at National Cheng Kung University Hospital (NCKUH; southern Taiwan) from 2010 to 2014 were also included in this study. These 56 isolates were obtained from positive blood culture bottles (Bactec 9240; Becton Dickinson Microbiology Systems, Sparks, MD, USA) and initially identified by colony morphology, Gram staining, growth at 37°C, oxidase test results, and oxidation of glucose. Conventional biochemical identification methods and the Phoenix bacterial identification system (NMIC/ID-72 cards; Becton Dickinson Microbiology Systems) were used to identify the isolates (17). Among these isolates, 33 were identified as V. cholerae, 20 as V. vulnificus, 2 as V. parahaemolyticus, and 1 as V. fluvialis. All V. cholerae isolates were identified as non-serogroup O1, non-O139 (non-O1/O139) based on a negative reaction by slide agglutination with polyvalents O1 and O139 (Difco, Becton Dickinson Microbiology System) (5). All isolates were stored at −70°C until use.

The isolates were further confirmed to the species level by partial rpoB gene sequencing and partial 16S rRNA gene sequencing analysis. Partial rpoB gene sequencing was performed as previously described (17). For 16S rRNA gene sequencing, two primers, 8FPL (5′-AGAGTTTGATCCTGGCTCAG-3′) and 1492RPL (5′-GGTTACCTTGTTACGACTT-3′) were used (21). The sequences (1,425 bp) obtained were compared with published sequences in the GenBank database using the BLASTN algorithm (http://www.ncbi.nlm.nih.gov/blast). The identification results of the 56 Vibrio isolates using the two molecular methods and the biochemical identification methods were identical (see Table S1in the supplemental material).

Samples of the 56 isolates of the Vibrio species were prepared for analysis by the Bruker Biotyper MALDI-TOF MS system as previously described (17). All isolates were inoculated onto Trypticase soy agar with 5% sheep blood (blood agar plates [BAP]) (Becton Dickinson Microbiology Systems) and incubated in 5% CO₂ at 37°C for 18 to 24 h. Two to three colonies were transferred to a 1.5-ml screw-cap Eppendorf tube containing 50 μl of 70% formic acid. After incubation for 30 s, 50 μl of acetonitrile (Sigma-Aldrich) was added. The suspension was centrifuged at 13,000 rpm for 2 min, and then 1.0 μl of the supernatant was applied to a 96-spot polished steel target plate (Bruker Daltonik GmbH) and dried. A saturated solution of 1.0 μl of MALDI matrix (alpha-cyano-4-hydroxycinnamic acid matrix solution; Bruker Daltonik GmbH) was applied to each sample and dried. Measurements were performed with the Bruker Biotyper MALDI-TOF MS system using FlexControl software with Compass Flex Series version 1.3 software and a 60-Hz nitrogen laser (337 nm wavelength). Spectra were collected in the linear positive mode in a mass range covering 1,960 to 20,132 m/z. Spectra ranging from 2,000 to 20,000 m/z were analyzed using the MALDI Biotyper system's automation control and the Bruker Biotyper 3.1 software and library (database [DB] 5627 with 5,627 entries). Identification scores of ≥2.000 indicated species-level identification, scores of 1.700 to 1.999 indicated genus-level identification, and scores of <1.700 indicated no reliable identification (17, 21). All V. vulnificus, V. parahaemolyticus, and V. fluvialis isolates had scores of ≥2.000, indicating that all species were correctly identified by the Bruker Biotyper MALDI-TOF MS system (see Table S1 in the supplemental material). The Bruker Biotyper MALDI-TOF MS system failed to identify all 33 V. cholerae isolates to the species level but yielded a top match of V. albensis. Among the 33 V. cholerae isolates, none of the isolates had scores of ≥2.000, 16 (48.5%) had scores ranging from 1.700 to 1.999, and 17 (51.5%) had scores of <1.700.

Clustering analysis of 50 isolates of the four genetically identified Vibrio species collected from NTUH was performed using ClinProTools 3.0 (Bruker Daltonics GmbH, Bremen, Germany) (17, 21). The MALDI spectra of the two isolates of V. parahaemolyticus were closer to those of V. vulnificus than to those of other species in the MALDI-TOF dendrogram, with dividing branches linked at a distance level of 850 (Fig. 1). Characteristic peaks of the spectra of V. vulnificus, V. parahaemolyticus, and V. fluvialis species are shown in Fig. 2.

FIG 1.

Principal component analysis (PCA) dendrogram generated by Bruker Biotyper MALDI-TOF MS system mass spectra for 50 isolates of four Vibrio species identified by gene sequencing analysis. Stars, MALDI spectra of the three V. cholerae isolates selected for MSP (main spectra projection; database entrance) creation using MALDI Biotyper software (Bruker Daltonics).

FIG 2.

Spectra of four Vibrio species generated by the Bruker Biotyper MALDI-TOF MS system. The absolute intensities and masses (m/z) of the ions are shown (axes).

Because V. cholerae was not listed in the current Bruker Biotyper MALDI-TOF MS database, spectra of three isolates, namely, NTUH-593 (isolate 1), NTUH-802 (isolate 6), and NTUH-1862 (isolate 22), were randomly selected from the 27 isolates of V. cholerae obtained from NTUH (Fig. 1) for the creation of main spectra projection (MSP; database entrance) using Bruker Biotyper MALDI-TOF MS software (Bruker Daltonics) as previously described (17, 21). The database generated using the 3 isolates was blindly tested against those of the 33 V. cholerae isolates (27 from NTUH and 6 from NCKUH) and the 23 non-cholerae Vibrio species. The best identification scores were found according to the database created using NTUH-802 (isolate 6); all 27 V. cholerae isolates obtained from NTUH were identified as V. cholerae with identification scores of ≥2.000 (2.350 to 2.623), and identification scores for the 6 V. cholerae isolates from NCKUH ranged from 2.536 to 2.585 according to MSP. The identification scores for the 23 non-cholerae Vibrio isolates were between 0.558 and 1.125. The characteristic MALDI Biotyper spectrum of the NTUH-802 isolate is shown in Fig. 2.

V. cholerae is not included among the 53 Vibrio species listed in the current Bruker Biotyper MALDI-TOF MS library (DB 5627). In contrast, V. cholerae is included among the 16 Vibrio species listed in the updated list of reference spectra for bacterial species in the Vitek MS plus Saramis Knowledge base 2.0 (bioMérieux, Marcy l'Etoile, France).

Rychert et al. (10) evaluated the use of the Vitek MS v2 MALDI-TOF MS platform with Knowledge base 2.0 (bioMérieux) for the identification of V. cholerae and found that the Vitek MS v2 system successfully identified all 42 isolates of V. cholerae serogroups O1 (n = 27) and O139 (n = 15) and 7 of the 9 non-O1/O139 V. cholerae isolates. Of the 9 non-O1/O139 V. cholerae isolates grown on thiosulfate citrate bile salt sucrose agar (Becton Dickinson Microbiology Systems), the Vitek MS v2 system failed to correctly identify 5 isolates and misidentified 2 isolates as V. cholerae/V. mimicus. For isolates grown on MacConkey II agar (Becton Dickinson Microbiology Systems), the Vitek MS v2 system was unable to identify 1 of the isolates.

Using the Bruker Biotyper MALDI-TOF MS system with ClinProTools software and model construction, Dieckmann et al. (19) successfully identified two species-specific biomarker ions that rapidly discriminate among the most important Vibrio spp. from closely related bacterial species such as Aeromonas spp. The VibrioBase, a newly established MALDI-TOF MS database, has been useful for rapid identification of human-pathogenic Vibrio spp. (15). Erler et al. (15) reported that >70% of the spectra in the VibrioBase generated matching scores higher than the best match with the Bruker Biotyper MALDI-TOF MS database, i.e., 79.4% for V. parahaemolyticus, 74.4% for V. vulnificus, and 100% for V. cholerae main spectra. Alignment of V. cholerae strains with the Bruker Biotyper MALDI-TOF MS database consistently resulted in identification of V. albensis, with matching scores ranging from 1.700 to 2.000 in 87% of cases.

In this study, we found that all three of the blood isolates of non-cholerae Vibrio species were correctly identified to the species level using the current Bruker database. However, using the same database, 48.5% of the 33 isolates of V. cholerae were correctly identified only to the genus level as Vibrio species (identification scores between 1.700 and 2.000). Importantly, all non-O1/O139 V. cholerae isolates grown on BAP were correctly identified to the species level using our newly generated database.

In conclusion, all non-cholerae Vibrio species isolates can be successfully identified to the species level by Bruker Biotyper MALDI-TOF MS. Using our newly created database, all non-O1/O139 V. cholerae isolates are correctly identified to the species level. These findings suggest that MALDI-TOF MS is a versatile and robust tool for the rapid identification of Vibrio species.