Abstract
An on-plate testing method using formic acid was evaluated on the Bruker Biotyper matrix-assisted laser desorption ionization–time of flight (MALDI-TOF) mass spectrometry system using 90 yeast and 78 Corynebacterium species isolates, and 95.6 and 81.1% of yeast and 96.1 and 92.3% of Corynebacterium isolates were correctly identified to the genus and species levels, respectively. The on-plate method using formic acid yielded identification percentages similar to those for the conventional but more laborious tube-based extraction.
TEXT
Matrix-assisted laser desorption ionization–time of flight (MALDI-TOF) mass spectrometry (MS) is a rapid and accurate method for identification of bacteria and fungi (2, 3, 5, 6). This technology employs an ionizing laser to generate isolate-derived spectra, which are compared in real time to a library of reference spectra. Depending on the similarity between the acquired spectra and the library, an identification and a confidence score are generated. Some organisms, such as Gram-negative bacteria, are easily analyzed by directly smearing a colony onto the MS plate. Other types of bacteria and yeasts generally require preparatory tube extraction (2). Using the Bruker Biotyper MALDI-TOF MS system (Billerica, MA) and a preparatory tube extraction method, we have previously shown that most yeast and Corynebacterium species are correctly identified to the genus and species levels (1, 3). The tube extraction process involves immersing the isolate in 70% ethanol in a separate tube and pelleting and drying the organism followed by extraction and spotting onto the MS plate. This is cumbersome and time-consuming for routine clinical use, and it requires additional labeling steps, which may lead to specimen misidentification, compromising patient safety.
To overcome these preanalytical processing limitations, we evaluated a rapid, direct on-plate testing method using formic acid for the identification of yeast and Corynebacterium species, and we compared our results to those from our prior studies of the same isolates tested by the preparatory tube extraction method (1, 3). Furthermore, the on-plate extraction method uses smaller volumes of formic acid and fewer laboratory consumables than tube extraction, and it is an overall more environmentally friendly process for isolate preparation.
(This material was presented, in part, at the 112th American Society for Microbiology General Meeting, San Francisco, CA, 16 to 19 June 2012.)
A total of 90 yeast and 78 Corynebacterium species isolates were cultured as described previously (1, 3) and analyzed using the Bruker Microflex LT/SH Biotyper following direct on-plate extraction. For yeast isolates, on-plate testing was performed by smearing a small amount of the organism from a single colony directly onto a spot on the MALDI-TOF MS steel anchor plate (BigAnchor 96-well plate; Bruker) and overlaying it with 1 μl of 70% formic acid (Fluka [Sigma-Aldrich], St. Louis, MO). The order of addition was reversed (formic acid followed by smearing with organism) for the Corynebacterium species to allow for better release of the organism from the inoculating plastic loop. Each spot was allowed to dry (∼5 min) and subsequently overlaid with 2 μl of matrix (α-cyano-4-hydroxycinnamic acid [HCCA]; Bruker Daltonics, Billerica, MA) dissolved in 50% acetonitrile, 47.5% water, and 2.5% trifluoroacetic acid (Fluka) (7). Following matrix addition, samples were allowed to dry prior to MALDI-TOF MS analysis (∼7 min). The Bacterial Test Standard (BTS; Bruker) was used for instrument calibration, and positive (Staphylococcus aureus and Candida krusei) and negative (formic acid and matrix) controls were analyzed with each run. Direct on-plate testing hands-on time was measured as the time from picking the isolate to the time the plate was placed in the instrument for a batch of 40 isolates.
The MALDI Biotyper library version 3.0 and MALDI Biotyper software version 3.0 were used for this analysis. Based on previous studies from our group, the manufacturer-recommended cutoff scores for identification of yeast and Corynebacterium species were lowered from ≥1.7 to ≥1.5 for genus-level and from ≥2.0 to ≥1.7 for species-level identification (1, 3). A score of <1.5 was considered as resulting in no reliable identification. Results from direct on-plate testing of yeast isolates were compared to those from phenotypic identification, and discrepant results were resolved by sequencing the D2 region of the 28S rRNA gene. For Corynebacterium species, results were compared to those from rpoB sequencing or 16S rRNA gene sequencing in the event that rpoB sequencing failed (1, 3). Statistical analysis was performed using McNemar's test of paired proportions; P values of <0.05 were considered statistically significant.
Of the 90 yeast isolates, direct on-plate testing using formic acid led to identification of 86/90 (95.6%) to the genus level and 73/90 (81.1%) to the species level (Table 1). A single discordant result was observed: Candida inconspicua was reported as Candida norvegensis (score, 1.762). This organism was reanalyzed by MS, yielding the same result and leading to an overall misidentification rate of 1.1%. Use of the manufacturer's recommended cutoff scores led to a decrease in the species-level identification rate of yeast (37/90 [41.1%]), while the genus-level identification rate remained equivalent to the rate found with the lower cutoff (74/90 [82.2%]) (Table 2).
Table 1.
Bruker Biotyper MALDI-TOF MS identification of yeast isolates using the direct on-plate formic acid-based method
| Organism (no. of isolates tested) | No. of isolates by level of identification (cutoff score) |
|||
|---|---|---|---|---|
| Species identification (≥1.7) | Species-level discordant identification (no. of isolates)a | Genus identification (≥1.5) | No identification (<1.5) | |
| Aureobasidium spp. (2) | 1 | 2 | ||
| Blastoschizomyces capitatus (1) | 1 | 1 | ||
| Candida albicans (9) | 7 | 9 | ||
| Candida colliculosa (3) | 1 | 2 | 1 | |
| Candida dubliniensis (3) | 1 | 3 | ||
| Candida famata (4) | 4 | 4 | ||
| Candida glabrata (1) | 1 | 1 | ||
| Candida guilliermondii (1) | 1 | 1 | ||
| Candida inconspicua (3) | 2 | Candida norvegensis (1) | 3 | |
| Candida kefyr (1) | 1 | 1 | ||
| Candida krusei (5) | 5 | 5 | ||
| Candida lambica (2) | 2 | 2 | ||
| Candida lipolytica (3) | 3 | 3 | ||
| Candida lusitaniae (4) | 3 | 4 | ||
| Candida metapsilosis (1) | 1 | 1 | ||
| Candida nivariensis (1) | 1 | |||
| Candida norvegensis (1) | 1 | 1 | ||
| Candida orthopsilosis (2) | 2 | |||
| Candida parapsilosis (7) | 6 | 7 | ||
| Candida pararugosa (1) | 1 | 1 | ||
| Candida pelliculosa (5) | 5 | 5 | ||
| Candida tropicalis (4) | 4 | 4 | ||
| Candida utilis (2) | 2 | 2 | ||
| Candida zeylanoides (1) | 1 | 1 | ||
| Cryptococcus gattii (2) | 2 | 2 | ||
| Cryptococcus laurentii (2) | 2 | 2 | ||
| Cryptococcus neoformans (3) | 1 | 2 | 1 | |
| Geotrichum silvicola (1) | 1 | 1 | ||
| Lodderomyces elongisporus (2) | 2 | 2 | ||
| Rhodotorula minuta (1) | 1 | |||
| Rhodotorula mucilaginosa (2) | 2 | 2 | ||
| Saccharomyces cerevisiae (5) | 5 | 5 | ||
| Trichosporon asahii (3) | 2 | 2 | 1 | |
| Trichosporon mucoides (2) | 2 | 2 | ||
| Total (n = 90) (%) | 73 (81.1) | 1 | 86 (95.6) | 4 (4.4) |
Discordant results were resolved using 28S rRNA gene sequencing.
Table 2.
Comparison of direct on-plate testing using formic acid versus preparatory tube extraction for identification (ID) of yeast and Corynebacterium isolates
| Isolate and method | Total no. of isolates tested | No. (%) with |
|||||
|---|---|---|---|---|---|---|---|
| Laboratory-validated MALDI-TOF cutoff scores |
Manufacturer's recommended MALDI-TOF cutoff scores |
||||||
| ≥1.7 (species level) | ≥1.5 (genus level) | <1.5 (no ID) | ≥2.0 (species level) | ≥1.7 (genus level) | <1.7 (no ID) | ||
| Yeast | 90a | ||||||
| Tube extraction | 77 (85.6) | 79 (87.8) | 11 (12.2) | 62 (68.9) | 77 (85.6) | 13 (14.4) | |
| Direct, on-plate testing | 73 (81.1) | 86 (95.6) | 4 (4.4) | 37 (41.1) | 74 (82.2) | 16 (17.8) | |
| Corynebacterium species | 78b | ||||||
| Tube extraction | 69 (88.5) | 77 (98.7) | 1 (1.3) | 47 (60.3) | 69 (88.5) | 9 (11.5) | |
| Direct, on-plate testing | 72 (92.3) | 75 (96.2) | 3 (3.8) | 46 (59.0) | 72 (92.3) | 6 (7.8) | |
Of the 78 Corynebacterium isolates tested, direct on-plate testing using formic acid led to identification of 75/78 (96.1%) to the genus and 72/78 (92.3%) to the species level (Table 3). Two isolates identified as Corynebacterium aurimucosum by rpoB gene sequencing were identified as Corynebacterium minutissimum by MS (scores, 2.029 and 1.716). The two species are closely related and cannot adequately be differentiated by the rpoB sequencing method used or by 16S rRNA gene sequencing; therefore we did not consider these isolates as misidentified (1). Sodium hippurate hydrolysis and acid production from mannitol were previously used to aid in the differentiation of these species (1). Use of the manufacturer's recommended cutoff scores showed a decrease in the species-level identification rate (46/78 [59.0%]), while genus-level identification remained equivalent (72/78 [92.3%]) (Table 2).
Table 3.
Bruker Biotyper MALDI-TOF MS identification of Corynebacterium isolates using the direct on-plate formic acid-based method
| Organism (no. of isolates tested) | No. of isolates by level of identification (cutoff score) |
||
|---|---|---|---|
| Species identification (≥1.7) | Genus identification (≥1.5) | No identification (<1.5) | |
| C. accolens (2) | 2 | 2 | |
| C. afermentans (5) | 3 | 3 | 2 |
| C. amycolatum (14) | 13 | 14 | |
| C. aurimucosum/C. minutissimum (8) | 6a | 7 | 1 |
| C. confusum (1) | 1 | 1 | |
| C. durum (2) | 2 | 2 | |
| C. glucuronolyticum (3) | 3 | 3 | |
| C. imitans (2) | 2 | 2 | |
| C. jeikeium (8) | 7 | 8 | |
| Corynebacterium group F1 (3) | 3 | 3 | |
| C. mucifaciens (1) | 1 | 1 | |
| C. pseudodiphtheriticum (2) | 2 | 2 | |
| C. riegelli (1) | 1 | 1 | |
| C. singulare (3) | 3 | 3 | |
| C. striatum (14) | 14 | 14 | |
| C. tuberculostearicum (5) | 5 | 5 | |
| C. ulcerans (1) | 1 | 1 | |
| C. urealyticum (3) | 3 | 3 | |
| Total (n = 78) (%) | 72 (92.3) | 75 (96.1) | 3 (3.8) |
C. minutissimum and C. aurimucosum are not well differentiated by rpoB or 16S rRNA gene sequencing.
Finally, we compared our direct on-plate testing results to previously acquired historical data which used the tube extraction method for identification of the same isolates examined in this study (1, 3). Among the yeasts, while species-level identification rates were equivalent regardless of preparatory protocol (P = 0.439), the percentage identified to the genus level was higher using on-plate testing with formic acid (95.6%) than with tube extraction (87.8%; P = 0.020) (Table 2). Direct on-plate testing and tube extraction of Corynebacterium species yielded equivalent identification percentages at both the genus and species levels (Table 2).
The on-plate method using formic acid is less cumbersome than tube extraction, allowing for low-scoring isolates to be readily retested. Since testing takes place entirely on the plate, there is a decreased risk of patient misidentification compared to tube extraction. Furthermore, the total hands-on time for processing 40 isolates is approximately 45 min for the direct versus 1.5 to 2 h for the tube extraction method (3). One limitation of our study is that we compared direct on-plate testing to historically collected tube extraction data (although the same isolates were tested using each method). Also, while we evaluated organisms commonly encountered in our clinical practice, we analyzed only single isolates for some species. Direct on-plate testing using formic acid has not been evaluated for filamentous molds; however, our internal validation studies indicate that on-plate extraction produces results equivalent to those of tube extraction for aerobic Gram-negative bacteria, streptococci, and staphylococci (data not shown).
In summary, we describe a rapid on-plate preparatory testing method for Bruker Biotyper MALDI-TOF MS identification of yeast and Corynebacterium species, which offers identification percentages that are comparable to those found with the more complex tube extraction process. Our data expand upon results from a letter to the editor by Haigh et al. (4), who used a direct smear method with matrix overlay alone for initial isolate evaluation and reflex to direct on-plate formic acid extraction for poorly scoring isolates. In conclusion, direct on-plate testing using formic acid is a straightforward and environmentally safe method, which reduces reagent cost, waste, and processing time for Bruker Biotyper MALDI-TOF MS-based microbial identification.
ACKNOWLEDGMENTS
We thank Charles J. Cazanave for performing rpoB sequencing for identification of Corynebacterium species, Jay Mandrekar for aid with statistical analyses, and Adnan A. Alatoom and Neelam Dhiman for performing mass spectrometry with tube-based extraction.
Footnotes
Published ahead of print 3 July 2012
REFERENCES
- 1. Alatoom AA, Cazanave CJ, Cunningham SA, Ihde SM, Patel R. 2012. Identification of non-diphtheriae Corynebacterium by use of matrix-assisted laser desorption ionization–time of flight mass spectrometry. J. Clin. Microbiol. 50:160–163 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Alatoom AA, Cunningham SA, Ihde SM, Mandrekar J, Patel R. 2011. Comparison of direct colony method versus extraction method for identification of Gram-positive cocci by use of Bruker Biotyper matrix-assisted laser desorption ionization–time of flight mass spectrometry. J. Clin. Microbiol. 49:2868–2873 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Dhiman N, Hall L, Wohlfiel SL, Buckwalter SP, Wengenack NL. 2011. Performance and cost analysis of matrix-assisted laser desorption ionization–time of flight mass spectrometry for routine identification of yeast. J. Clin. Microbiol. 49:1614–1616 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Haigh J, Degun A, Eydmann M, Millar M, Wilks M. 2011. Improved performance of bacterium and yeast identification by a commercial matrix-assisted laser desorption ionization–time of flight mass spectrometry system in the clinical microbiology laboratory. J. Clin. Microbiol. 49:3441. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Saffert RT, et al. 2011. Comparison of Bruker Biotyper matrix-assisted laser desorption ionization–time of flight mass spectrometer to BD Phoenix automated microbiology system for identification of Gram-negative bacilli. J. Clin. Microbiol. 49:887–892 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Theel ES, Hall L, Mandrekar J, Wengenack NL. 2011. Dermatophyte identification using matrix-assisted laser desorption ionization–time of flight mass spectrometry. J. Clin. Microbiol. 49:4067–4071 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Van Herendael BH, et al. 2012. Validation of a modified algorithm for the identification of yeast isolates using matrix-assisted laser desorption/ionisation time-of-flight mass spectrometry (MALDI-TOF MS). Eur. J. Clin. Microbiol. Infect. Dis. 31:841–848 [DOI] [PubMed] [Google Scholar]