Rapid Identification of Cryptococcus neoformans and Cryptococcus gattii by Matrix-Assisted Laser Desorption Ionization–Time of Flight Mass Spectrometry

Abstract

Compared to DNA sequence analysis, matrix-assisted laser desorption ionization–time of flight mass spectrometry (MALDI-TOF MS) correctly identified 100% of Cryptococcus species, distinguishing the notable pathogens Cryptococcus neoformans and C. gattii. Identification was greatly enhanced by supplementing a commercial spectral library with additional entries to account for subspecies variability.

TEXT

Cryptococcosis is typically a severe infection of the central nervous system characterized by meningoencephalitis and other neurological complications, primarily in patients with AIDS or other forms of immune compromise (26). Cryptococcus isolates must be distinguished and characterized at the species or subspecies level due to differences in epidemiology, virulence, and antifungal drug susceptibility. Cryptococcus gattii has a greater propensity to infect immunocompetent people (26), with some strains being more virulent (3, 16, 27) or less susceptible to fluconazole and other triazoles (6, 20, 42). Cryptococcus neoformans consists of two varieties, var. grubii and var. neoformans. While C. neoformans var. grubii causes the majority of clinical infections worldwide, C. neoformans var. neoformans infections are more prevalent in India (18) and certain areas of Europe (11, 24) and are strongly correlated with infections in the elderly, the presence of skin lesions, and corticosteroid use (12). Infections due to other Cryptococcus species are rare, but the incidence is increasing, requiring greater vigilance by clinical laboratories (21).

Currently, identification of Cryptococcus to the genus or subgenus level from clinical specimens relies upon the microscopic examination of yeast cells in conjunction with biochemical tests, differential media, and/or DNA sequence analysis (9, 22, 23, 31, 33, 44–46). These tests may require multiday incubation or labor-intensive and costly protocols that may delay diagnosis.

Recently, matrix-assisted laser desorption ionization–time of flight mass spectrometry (MALDI-TOF MS) has been used to successfully identify various species of bacteria and fungi (17, 28, 34) By identifying species based on characteristic protein spectra extracted from whole cells, MALDI-TOF MS produces a highly discriminatory identification of a pure culture in 5 to 20 min at minimal cost, depending on the sample preparation procedure utilized. In this study, we tested the ability of whole-cell MALDI-TOF MS to identify clinical isolates of Cryptococcus spp. and to further differentiate them at the species or subspecies level.

We analyzed a total of 160 yeast isolates, including 137 Cryptococcus strains and 23 non-Cryptococcus yeast strains. Twenty-five were type and reference strains (Table 1), 55 were isolated from clinical samples submitted to the Mycology Laboratory of the Ontario Public Health Laboratory from 2007 to 2010. Another 51 were obtained from the Fungus Testing Laboratory, University of Texas Health Science Center at San Antonio, and an additional 15 were donated from the British Columbia Centre for Disease Control Public Health and Reference Microbiology Laboratory. Six were purchased from the American Type Culture Collection (Manassas, VA), and eight were obtained from the United States Department of Agriculture Agricultural Research Service (Peoria, IL).

Table 1.

Summary of MALDI Biotyper identification of 160 clinical, type, and reference isolates of Cryptococcus and non-Cryptococcus yeasts

Genus and speciesa	Type or reference strain(s)i	No. of clinical isolates	Total no. of isolates	No. (%) correctly identified			No. (%) incorrectly identified
				Excellentb	Goodc	Total
Cryptococcus		111	137	128 (93.4)	8 (5.8)	136 (99.3)	1 (0.7)
C. neoformans var. grubii	ATCCd MYA-4564, ATCC MYA-4565	80	82	80 (97.6)	2 (2.4)	82 (100)	0 (0)
C. neoformans var. neoformans	ATCC MYA-4567	3	4	3 (75)	0 (0)	3 (75)	1 (25)
C. gattii	CBSe 6289, ATCC MYA-4561, CBS 6955, ATCC MYA-4563	23	27	24 (89)	3 (11)	27 (100)	0 (0)
C. albidus	ATCC 10666T, NRRLf YB-195, NRRL YB-219, NRRL YB-253, NRRL YB-295	0	5	4 (80)	1 (20)	5 (100)	0 (0)
C. laurentii	ATCC 18803T, NRRL YB-443, NRRL YB-449, NRRL YB-481, NRRL YB-482	0	5	5 (100)	0 (0)	5 (100)	0 (0)
C. magnus	NRRL Y-2537T	4	5	5 (100)	0 (0)	5 (100)	0 (0)
C. terreus	CBS 1895T, ATCC 32422, ATCC 34145, ATCC 32046	0	4	3 (75)	1 (25)	4 (100)	0 (0)
C. unigutulattus	JCMg 3685T, ATCC 32048, ATCC 32047, ATCC 66033	1	5	4 (80)	1 (20)	5 (100)	0 (0)
Non-Cryptococcus yeasts		10	23	20 (87)	3 (13)	23 (100)	0 (0)
Candida albicans	NRRL Y-12983T	1	2	2 (100)	0 (0)	2 (100)	0 (0)
Candida dubliniensis	NRRL Y-17841T	0	1	1 (100)	0 (0)	1 (100)	0 (0)
Candida glabrata	NRRL Y-65T	2	3	3 (100)	0 (0)	3 (100)	0 (0)
Candida guilliermondii	NRRL Y-2075T	0	1	1 (100)	0 (0)	1 (100)	0 (0)
Candida krusei	NRRL Y-5396T	1	2	2 (100)	0 (0)	2 (100)	0 (0)
Candida lusitaniae	NRRL Y-11827T	0	1	1 (100)	0 (0)	1 (100)	0 (0)
Candida parapsilosis	DSMh 5784T	1	2	2 (100)	0 (0)	2 (100)	0 (0)
Candida tropicalis	DSM 11953T	1	2	2 (100)	0 (0)	2 (100)	0 (0)
Galactomyces geotrichum	JCM 6359T	0	1	1 (100)	0 (0)	1 (100)	0 (0)
Rhodotorula mucilaginosa	JCM 8115T	1	2	1 (50)	1 (50)	2 (100)	0 (0)
Saccharomyces cerevisiae	JCM 7255T	1	2	2 (100)	0 (0)	2 (100)	0 (0)
Trichosporon asahii	JCM 2466T	2	3	1 (33)	2 (67)	3 (100)	0 (0)
Trichosporon inkin	JCM 9195T	0	1	1 (100)	0 (0)	1 (100)	0 (0)

^aIdentification was based on >99% ITS and/or >98% IGS sequence identity to a type or reference strain.

^bMALDI Biotyper ID score, ≥2.0.

^cMALDI Biotyper ID score, <2.0.

^dATCC, American Type Culture Collection, Manassas, VA.

^eCBS, Centraalbureau voor Schimmelcultures, Utrecht, Netherlands.

^fNRRL, United States Department of Agriculture, Agricultural Research Service, Peoria, IL.

^gJCM, Japanese Collection of Microorganisms, Saitama, Japan.

^hDSM, German Collection of Microorganisms and Cell Cultures, Braunschweig, Germany.

ⁱA superscript T indicates the type strain.

As a reference “gold standard” for this evaluation, all isolates were identified by DNA sequence analysis of the rRNA internal transcribed spacer (ITS) region (7), which differentiated all yeast species except for C. neoformans and C. gattii. Clinical isolates demonstrated >99% similarity to type strains, with a between-species divergence of >5.8%. While multilocus sequence typing or amplified fragment length polymorphism (30) is considered the reference gold standard for molecular subtyping of Cryptococcus species, we have validated the use of a single target, the intergenic spacer (IGS) (9, 29), to distinguish the species and varieties C. neoformans var. grubii, C. neoformans var. neoformans, and C. gattii with good results. Isolates showed >98.9% similarity to a reference strain (30) and >4% divergence between species and subspecies.

Protein spectra for all isolates were generated and analyzed by MALDI-TOF MS using a Microflex LT instrument (Bruker Daltonics, Billerica, MA), Flexcontrol 3.0 software, and the most recent version of the Biotyper 2.0.1 software and database (Bruker Daltonics) according to the manufacturer's instructions. Following cultivation on Inhibitory Mold Agar (BD, Sparks, MD) for 48 h at 27°C, protein extracts were prepared according to the ethanol-formic acid extraction protocol recommended by Bruker Daltonics. Briefly, a few colonies of each isolate were suspended in 300 μl of distilled H₂O in sterile microtubes. Following the addition of 900 μl of 100% ethanol, the microtubes were centrifuged for 2 min at 16,000 × g. The supernatant was decanted, and the pellets were dried using a Savant SpeedVac DNA 120 Concentrator (Thermo Scientific, Waltham, MA) for 5 min at room temperature. Pellets were resuspended in 50 μl of 70% formic acid and 50 μl of acetonitrile. After centrifugation for 2 min at 16,000 × g, 1 μl of the supernatant was transferred onto an MSP 96 polished steel target (Bruker Daltonics) and allowed to air dry. The samples were overlaid with 1 μl of matrix solution consisting of a saturated solution of α-cyano-4-hydroxycinnamic acid in 50% acetonitrile–2.5% trifluoroacetic acid and again allowed to air dry prior to analysis. For each isolate, a spectrum with a mass-to-charge range of 2,200 to 22,000 Da was generated as an average of 240 laser shots in an automatic acquisition mode. When poor spectra (fewer than 10 well-defined peaks above 1,000 arbitrary units) were obtained, analysis was repeated with an extra wash step during the protein extraction procedure, which improved the quality of the spectra.

For identification of all isolates, MALDI-TOF mass spectra were compared to a spectral database using the Biotyper 2.0.1 software (Bruker Daltonics). The software generates identification scores based on three components: the proportion of signal matches in the reference spectrum also detected in the unknown, the proportion of signal matches in the unknown spectrum also detected in the reference, and the correlation of the relative intensities of the signals from the unknown and reference spectra. The product of these three values is multiplied by 1,000. The logarithm of this product is the identification score, which can range from 0 to 3. Because of the paucity of Cryptococcus entries in the Bruker Daltonics Biotyper 2.0.1 library database, we generated our own library entries consisting of the type and reference strains listed in Table 1, as well as one clinical isolate identified as C. neoformans var. grubii by IGS sequence analysis. Each library entry was generated as a composite of 20 to 24 spectra.

Using the Biotyper 2.0.1 software, we challenged the protein spectra from all isolates against the Biotyper 2.0.1 database supplemented with the in-house-generated library entries. MALDI-TOF MS correctly identified 100% of the isolates at the species level (Table 1). C. neoformans and C. gattii were clearly differentiated from each other, as well as from other Cryptococcus and yeast species.

C. neoformans var. grubii and C. neoformans var. neoformans were distinguished at the subspecies level in 98.8% (85/86) of the cases; however, a single isolate of C. neoformans var. neoformans was misidentified as C. neoformans var. grubii (Table 1). This isolate was reanalyzed with the same results. Whole-cell MALDI-TOF MS analysis is a highly discriminatory identification method capable of subspecies level differentiation of certain bacteria, including Streptococcus (19), Staphylococcus (38), Salmonella (10), Francisella (36), Bifidobacterium (35), and Lactococcus (41). Our results show that in addition to differentiating the closely related species C. gattii and C. neoformans, MALDI-TOF MS analysis may also be capable of discriminating certain fungi at the subspecies level, although the number of C. neoformans var. neoformans isolates tested was low.

The majority of identifications (93.1%, 149/160) received a score of >2.0, indicating excellent identification. The remaining 11 isolates (6.9%) scored >1.7 but ≤2.0, signaling good but less reliable identification. No identification received a score of ≤1.7, which is considered unreliable (Table 1). Bruker Daltonics recommends an identification score of >2.0 for species level identification and a score of >1.7 for genus level identification only. Similar to other studies (8, 40), however, our results show accurate species identification with spectral scores of >1.7, suggesting that a threshold of >1.7 is more appropriate for Cryptococcus species identification by Biotyper 2.0.1 software analysis of MALDI-TOF mass spectra.

To test the reproducibility of the method, we analyzed three separate cultivations each of four strains. In all cases, the identifications were identical, with little variability in the identification scores (means ± standard deviations of 2.374 ± 0.079, 2.292 ± 0.026, 2.251 ± 0.118, and 2.032 ± 0.095).

Our ability to identify Cryptococcus using MALDI-TOF MS was greatly enhanced by supplementing the Biotyper 2.0.1 library with additional entries. Not only did we include entries for species not found in the Biotyper 2.0.1 library (C. gattii [4 entries], C. terreus, and C. magnus), but we also included strains of C. laurentii, C. albidus, C. uniguttulatus, C. neoformans var. grubii (3 entries), and C. neoformans var. neoformans. The supplemental entries increased the percentage of Cryptococcus isolates identified to the species level from 58.4% to 100% and significantly increased the identification scores of isolates correctly identified (2.025 ± 0.167 without supplemental entries, 2.266 ± 0.174 with supplemental entries [P < 0.001, t test]). Presumably, the supplemental library entries enhanced identification performance by accounting for variability that MALDI-TOF MS analysis is capable of detecting at the subspecies level (10, 19, 35, 36, 38, 41). The observation that the identification power of the Bruker MALDI-TOF Biotyper system is limited by the robustness of the reference library is not unique to this system but instead is an inherent limitation of all fingerprinting-based identification technologies. However, supplemental library entries for the Biotyper 2.0.1 database are quickly and easily generated, allowing current users seeking to identify Cryptococcus isolates to generate their own supplemental library entries using type or reference strains whose identities have been confirmed through IGS and/or ITS sequencing. Supplemental library entries may also enhance the identification of other taxa where relatively low percentages of isolates were identified by commercial MALDI-TOF MS libraries (2, 5, 13, 37, 43).

The ethanol-formic acid sample preparation procedure and analysis requires ∼20 min per sample, at a cost of about $0.50 per sample for consumables and reagents. Although the capital acquisition cost of the machine is $200,000, it has diverse applications in a clinical laboratory, including the identification of isolates from various genera of bacteria and fungi (17, 28, 34), the identification of bacteria directly from blood vials (15, 25, 32, 39) or urine (14), and the identification of virulence factors or specific resistance-related proteins (1, 4).

In conclusion, MALDI-TOF MS analysis by the Biotyper 2.0.1 software is a rapid, accurate, and cost-effective method for identifying Cryptococcus spp. Utilizing an expanded library, this method identified isolates of C. neoformans var. grubii, C. neoformans var. neoformans, and C. gattii along with other species of Cryptococcus and other yeasts.