Comparative Study of Matrix-Assisted Laser Desorption/Ionization Mass Spectrometry and Culture Test for Candida Identification

Background: A new microorganism identification method using matrix-assisted laser desorption/ionization mass spectrometry (MALDI-TOF MS) has been developed, however, reports on its use for delineating Candida spp. are scarce. Objectives: The purpose of this study was to compare the identification accuracy of mixed infection between culture test and MALDI-TOF MS. Materials and Methods: Eighty-nine denture wearers (average 74.0 ± 9 years) were selected. Specimens were immediately inoculated onto selective medium for CHROMagarTM Candida, and were also carried out using MALDI-TOF MS. The distribution frequencies of them were analyzed. Results: The numbers and rates of detection/non-detection by MALDI-TOF MS of genus Candida were 58/31 (65.2%/34.8%), respectively. Infection types were single infection in 34 (38.2%), mixed infection in 24 (27.0%), and non-infection in 31 (34.8%) cases. Concerning the single infection, C. albicans was the most predominant (58.8%), followed by C. parapsilosis (17.6%), C. glabrata (14.7%), C. tropicalis (5.9%), and C. krusei (2.9%). As for the mixed infection, the most frequent combination was C. albicans and C. glabrata (50.0%), followed by C. albicans and C. parapsilosis (29.2%), C. albicans and C. tropicalis (8.3%), C. glabrata and C. tropicalis (4.2%), C. albicans, C. glabrata, and C. parapsilosis (4.2%), and C. albicans, C. parapsilosis, and C. glabrata (4.2%). There were four MALDI-TOF MS positive results that were negative by the culture test. Conversely, there were six MALDI-TOF MS negative results that were positive by the culture test. The sensitivity and specificity of MALDI-TOF MS were 0.929 and 0.840, How to cite this paper: Kubota, Y., Taguchi, C., Saito, M., Shinozaki-Kuwahara, N., Suzuki, T., Suemitsu, M., Nakayama, M., Utsunomiya, T., Endo, H. and Kuyama, K. (2019) Comparative Study of Matrix-Assisted Laser Desorption/Ionization Mass Spectrometry and Culture Test for Candida Identification. Open Journal of Stomatology, 9, 295-306. https://doi.org/10.4236/ojst.2019.912030 Received: October 20, 2019 Accepted: December 14, 2019 Published: December 17, 2019 Copyright © 2019 by author(s) and Scientific Research Publishing Inc. This work is licensed under the Creative Commons Attribution International License (CC BY 4.0). http://creativecommons.org/licenses/by/4.0/ Open Access


Introduction
Candida is a large genus of ascomycetous yeast, consisting of about 150 species, and more than 20 species have clinical importance [1]. Candida spp. are found in the oral cavity of 25% -50% of healthy individuals, including adults and children. When only denture wearers are considered, these frequencies increase to 60% -100% [2]. Moreover, wearing a dental prosthesis such as a removable denture [3] or removable orthodontic appliance [4] enhances oral candidal colonization and predisposes the wearer to oral candidiasis.
Candida albicans is the most frequently isolated species and the causative species of almost 70% of Candida infections [5] [6]. Although C. albicans is a well-known colonizer and pathogen of the oral mucosa, non-C. albicans Candida (NCAC) is increasingly encountered [7] [8] [9] and its emerging role in human infections has gained attention [10] [11]. This rising number in infections caused by NCAC species may reflect both the improvement in diagnostic methodologies and the superior capability of NCAC species to persist in the host compared with C. albicans. Moreover, in human mixed candidiasis, NCAC species demonstrate high antifungal resistance profiles 10). CHROMagar™ Candida is a very useful medium to distinguish Candida spp. such as C. albicans, C. tropicalis, C. krusei, and C. glabrata, which account for almost 90% of all clinical yeastisolates, including dental samples [12] [13].
However, CHROMagar™ Candida cannot definitely distinguish C. albicans from C. dubliniensis [14] [15], which is more often isolated from HIV-positive patients. In addition, the accuracy of CHROMagar™ to distinguish particular Candida spp. from mixed infection is poorly examined. Namely, insufficient evidence is available about the quality of CHROMagar™ Candida testing in patients with various underlying conditions. A new microorganism identification method using matrix-assisted laser desorption/ionization mass spectrometry (MALDI-TOF MS) has been developed, however, reports on its use for delineating Candida spp. are scarce [16] [17] [18]. Therefore, the purposes of this study were to compare the identification accuracy of mixed infection between CHROMagar™ Candida and MALDI-TOF MS in older denture wearers, and to highlight gaps in knowledge and the justification for the present work about candidal distribution.

Subjects
Eighty-nine denture wearers (49 males, 40 females) who had visited Kubota Dental Hospital for denture adjustment/checkup from August 2017 to September 2018 were selected for inclusion in this study. The mean age and standard deviation of subjects were 74.0 ± 9 years (range, 43 -99 years). Inclusion criteria were individuals who: 1) were wearing full/partial dentures; 2) had data for oral examination and microbiological tests; 3) had agreed to the study. Exclusion criteria were individuals who: a) were taking antifungal agents or using antiseptic mouthwashes; b) were taking any medication known to predispose them to oral candidiasis; c) had a medical history that revealed any disease or medical condition (such as diabetes mellitus or anemia) that predisposed them to oral candidiasis or promoted candidal carriage in the 6 weeks before the study; or d) for whom necessary data were lacking. None of the included subjects had signs or symptoms suggestive of candidal paronychia or dermatologic fungal infection.

1) Sample collection
To reduce the bias in sampling procedures, microbiological samples were collected from all patients by swabbing denture mucosal surfaces 10 times with a cotton swab dipped in sterile purified water approximately 2 hours or more after breakfast eating.
2) Culture test using CHROMagar™ Specimens were then immediately inoculated onto selective medium for Candida, CHROMagar™ Candida (KANTO KAGAKU, Tokyo, Japan). After aerobic incubation for 24 -48 hours at 25˚C, Candida spp. were manually measured and enumerated by colony morphology and color. According to the manufacturer's instructions, C. albicans colonies are distinguished by a distinctive greencolor, C. glabrata colonies exhibit a purple to pale pink color, C. tropicalis colonies have a dark blue color with pink edge, C. krusei colonies are rough with pale pink centers and white edges, and C. parapsilosis colonies present a white or pale pink color. After the 48-hour incubation on CHROMagar™, incubation was continued for more than 2 weeks at 37˚C to confirm not to culture.

3) MALDI-TOF MS analysis
Species determination of all samples was also carried out using MALDI-TOF MS (BD Bruker MALDI Biotyper™; BD). Fungal colonies were thinly applied to the target plate using a toothpick, 1 μl of matrix was added, and samples were measured according to the manufacturer's protocol. Application of a sample is one time, and MALDI Biotyper measures the strain several times and calculates average value of each peak of a mass spectrum and the strength. Acquired spectrum were expressed as a score value by phylogenetic tree analysis using a pattern matching spectral library. The case where the score value was 2.0 or more was taken as the identification result at the bacterial species level.

Compliance with Ethical Standards
Informed consent was obtained from all individuals included in the study. All

Subject Characteristics
Characteristics of subjects in the present study were shown in Table 1. Subjects consisted of 49 males and 40 females (average age 74.1 ± 9.1). Denture types included upper full denture (n = 25), lower full denture (n = 2), upper partial denture (n = 47), and lower partial denture (n = 15

Microbiological Distribution
The results of microbiological distribution were listed on the cross tabulation in Table 2.

3) Comparison of culture test and MALDI-TOF MS results
There were four MALDI-TOF MS positive results (C. albicans, n = 2; C. albicans and C. glabrata, n = 1; and C. albicans and Penicillium spp., n = 1) that were negative by the culture test. Conversely, there were six MALDI-TOF MS negative results (sample degeneration) by drying, n = 4; Aspergillus spp., n = 1; and C. albicans and Penicillium spp., n = 1) that were positive by the culture test.

Statistical Results
The sensitivity and specificity of MALDI-TOF MS were 0.929 and 0.840, respectively. The concordance rate (kappa coefficient) of genus Candida was 0.644, indicating substantial agreement [19]. Table 1 shows all concordance rates for single and mixed infections of all Candida spp. and combinations of mixed infection. Concordance rates of single infection were as follows: C. albicans 85.0%, C. glabrata 80.0%, C. parapsilosis 50.0%, C. tropicalis 50.0%, and C. krusei 0.0%. Furthermore, concordance rates of mixed infection were as follows: C. albicans and C. tropicalis 100.0%, C. albicans and C. glabrata 91.7%, C. albicans and C. parapsilosis 28.6%, C. glabrata and C. tropicalis 0.0%, and C. albicans, C. glabrata, and C. parapsilosis 0.0%. In addition, the concordance rate of Candida negative was 80.6%.

Discussion
The elderly are known to be more vulnerable to fungal infections because of underlying conditions, such as chronic diseases, medications, poor oral hygiene, reduced salivary flow, and immune system impairment [20] [21] [22]. Furthermore, fungi are isolated not only from the oral cavity but also from tissue-fitting surfaces and outer surfaces of dentures [21]. Therefore, in this study, we evaluated two Candida spp. detection methods, culture test and MALDI-TOF MS, in older denture wearers (mean age, 74 years) without subjective symptoms. C. albicans is the most frequently isolated Candida spp. as a colonizer and pathogen of the oral mucosa but reports on the appearance of NCAC are increasing [23] [24] [25]. This shift towards NCAC species is mainly due to severe immunosuppression, use of broad-spectrum antibiotics, and empirical use of antifungal drugs [7] [8] [9]. Nevertheless, NCAC infections are usually indistinguishable based on symptoms alone due to similarities in clinical presentations. The present study detected 66.3% and 65.2% of NCAC by culture test and MALDI-TOF MS, respectively, and these values were higher than that of a previous study (58.5%) [26]. CHROMagar™ is a commercial product that was developed for the isolation and presumptive identification of specimens containing mixtures of yeast species [27]. CHROMagar™ Candida is a very useful medium to distinguish Candida spp. with different morphologies and colors, which account for almost 90% of all clinical yeastisolates, including dental samples [12] [13]. The specificity and sensitivity of C. albicans, C. tropicalis, and C. krusei identification exceed 99% [27]. However, changes in the oral environment, such as during severe immunosuppression, increase the number of NCAC species, and the sensitivity of CHROMagar™ Candida in detecting mixed infections with NCAC has not been sufficiently examined. In the previous two decades, C. albicans represented over 80% of all human candidiasisisolates [32]. In this study, C. glabrata and C. parapsilosis proportions were particularly high. C. glabrata and C. parapsilosis have emerged as significant nosocomial pathogens. C. glabrata is associated with immunosuppressive and antimicrobial treatments [10] [33], while C. parapsilosis is associated with invasive procedures [11].  [36].
The kappa coefficient between culture test and MALDI-TOF MS was 0.644, which indicated substantial agreement 19). The highest concordance rate of single infection was C. albicans (85.0%), followed by C. glabrata (80.0%). These high concordance rates may be explained by the ease of distinguishing C. albicans and C. glabrata colonies by color (green and deep purple, respectively). Conversely, C. parapsilosis, C. tropicalis, and C. krusei, which have similar colony colors, presented low concordance rates. In addition, the concordance rate of Candida negative was 80.6%. Furthermore, concerning mixed infections, C. albicans and C. parapsilosis, C. glabrata and C. tropicalis, and C. albicans, C. glabrata, and C. parapsilosis showed low or non-concordance rates. We speculate that the number of fungi on agar plates was large, colonies were fused due to mixed infection, and identification by colony color is subjective. However, these issues do not affect identification by MALDI-TOF MS. Furthermore, our findings suggested that degeneration due to fungal death leads to overdiagnosis by the culture test. The sensitivity and specificity rates of MALDI-TOF MS were high, even in cases of mixed infection. It was speculated that the reason for the very few misidentifications was caused by spectral pattern matching for each Candida species. More than one strain by the same species was registered in the spectral library, so even if diversity with the spectral pattern existed between strains, it should be possible to correctly identification. With the increase in NCAC and multiple infections, the value of MALDI-TOF MS application is recognized for Candida identification. In addition, there is a need to improve the criteria for simple CHROMagar™ culture testing for diagnosis of oral infections.
In conclusion, Candida mixed infection was observed in 27.0% of older Japanese denture wearers. Candida infection is complicated by disease type and oral cavity environment changes due to aging. A rapid microorganism detection method, such as MALDI-TOF MS, will be helpful to quickly determine the causative pathogen in dental infections.

Conclusions
The following could be concluded in the present study: 1) The numbers and rates of detection by MALDI-TOF MS of genus Candida were 65.2% of elder denture wearers.
2) The rates of single and mixed infection were 38.2/27.0%, respectively.