Detection and identification of non-Candida albicans species in human oral lichen planus

Authors


Correspondence
Takuichi Sato, Division of Oral Ecology and Biochemistry, Tohoku University Graduate School of Dentistry, Sendai 980-8575, Japan.
Tel: +81 22 717 8295; fax: +81 22 717 8297; e-mail: tak@m.tohoku.ac.jp

ABSTRACT

Candida species were detected and identified in samples from the buccal mucosa, dorsal surface of the tongue and supragingival plaque of subjects with oral lichen planus (OLP). The Candida in the samples were cultured on selection agars, and identified by sequence analyses of 18S, 5.8S and 25/28S rRNA. The isolation frequency of Candida was higher in subjects with OLP than in those with healthy oral mucosa. Non-C. albicans were only isolated from people with OLP. These results support the notion that subjects with OLP are more likely to have oral colonization with Candida, and that non-C. albicans are specifically present in subjects with this condition.

List of Abbreviations: 
C.

Candida

CFU

colony forming units

OLP

oral lichen planus

PCR-RFLP

polymerase chain reaction–restriction fragment length polymorphism

rRNA

ribosomal RNA

Oral lichen planus is a refractory and chronic inflammatory disease which produces white lesions on the oral mucosa and occurs almost exclusively in middle-aged or elderly people. The condition has a long-term, chronic nature, and has been defined by the World Health Organization as a precancerous condition (1, 2). It has been reported that genetic, environmental and lifestyle factors, such as autoimmune diseases, dental materials, medicaments, microbial infection and stress, are associated with the disease (3–5).

In terms of infection with oral microorganisms, Candida species has been reported to be associated with OLP (6, 7). Among the Candida species, C. albicans is the one that has most frequently been identified by culture methods. Recent molecular biological techniques have enabled us to identify Candida species other than C. albicans (non-C. albicans), and have subsequently suggested a relationship between non-C. albicans and various infectious diseases (8–12). However, the relationship between non-C. albicans and OLP is unknown. Thus, the present study aimed to detect Candida species, including non-C. albicans, in subjects with OLP by molecular biological methods in order to elucidate any relationships with OLP.

Samples were obtained from patients with OLP (11 women and 4 men; age, 61.2 ± 10.0 years) who were attending the Clinical Department of Oral Diagnosis, Tohoku University Hospital, Sendai, Japan. OLP was diagnosed on the basis of clinical features and histopathological findings. In all cases, the lesions showed the clinical features of the erosive form of OLP. Volunteers with healthy oral mucosa (seven women; age, 51.7 ± 9.8 years) were also examined as controls (Table 1). None of the subjects had received either antibiotics or steroid ointment during the 6 weeks prior to the study, and none were taking immunosuppressants, anticholinergics or cancer chemotherapy agents. Among the 15 patients with OLP, seven (47%) had systemic diseases, namely, diabetes, hypertension and hyperlipidemia (Table 1), but these conditions were well controlled by their regular physicians, and the patients were not chronically immunosuppressed or medically compromised. Furthermore, they had experienced neither periodontitis nor oral candidiasis during at least the 6 months prior to the study, and had neither salivary gland dysfunction (decrease in saliva secretion) nor smoking habits. Informed consent was obtained from all subjects, and this study was approved by the Research Ethics Committee of Tohoku University Graduate School of Dentistry, Sendai, Japan. Our preliminary calculations using statistical software (Stat Flex Ver. 6, Artic, Osaka, Japan) showed that at least seven subjects were needed for statistical analysis in each group (subjects with OLP and healthy subjects), accordingly 15 patients with OLP and 7 healthy people were selected for the present study.

Table 1.  Clinical features of subjects with oral lichen planus and subjects with healthy oral mucosa in the present study
SubjectAge (years)SexRemovable dental appliancesSystemic diseases
  1. , Removable full (complete) dentures were used by this subject; ‡1, hypertension; ‡2, diabetes; ‡3, hyperlipidemia.

Subjects with oral lichen planus
  167Male+‡1
  275Female++‡2
  377Female
  465Female
  547Female
  652Female
  772Female
  858Female+‡2
  947Female
 1068Male+‡1
 1168Male+‡1
 1259Female+‡2
 1348Male
 1456Female+‡3
 1559Female
Subjects with healthy oral mucosa
 159Female
 243Female
 343Female
 452Female
 546Female
 670Female
 749Female

An area of 1 cm2 on the surface of the buccal mucosa or rear dorsal surface of the tongue which showed OLP lesions was defined by a window made of sterilized plane paper and firmly scraped 10 times with a sterilized spoon, as described previously (13, 14). In addition, after the sampling sites had been isolated by cotton rolls, samples of supragingival plaque from sites proximal to the upper first molars were collected with sterilized toothpicks.

Each sample (of about 1 mg) was suspended in 1 mL of sterilized 40 mM potassium phosphate buffer (pH 7.0) and dispersed with a Teflon homogenizer. Serial 10-fold dilutions (0.1 mL each) were spread onto the surface of CHROMagar Candida (BD, Franklin Lakes, NJ, USA) and Center for Disease Control and Prevention anaerobe 5% sheep blood agar (BD), and incubated at 37°C for 2 days under aerobic conditions, and for 7 days in an anaerobic glove box (Model AZ-Hard, Hirasawa, Tokyo, Japan) containing 80% N2, 10% H2 and 10% CO2, respectively.

Sub-cultured colonies of Candida isolates on the CHROMagar Candida were suspended individually in 1 mL of sterilized water, harvested by centrifugation at 7700 ×g for 5 min, and the supernatants removed. The genomic DNA of the Candida isolates was then extracted from the pellets using the InstaGene Matrix Kit (Bio-Rad Laboratories, Richmond, CA, USA) according to the manufacturer's instructions.

The gene sequences of 18 S, 5.8S and 25/28 S rRNA regions were amplified by PCR using fungus-specific primers, primers 1 and 3 (15) and Taq DNA polymerase (HotStarTaq Master Mix, Qiagen GmbH, Hilden, Germany) according to the manufacturer's instructions. Primer sequences were: primer 1 (5′- GTC AAA CTT GGT CAT TTA -3′); and primer 3 (5′- TTC TTT TCC TCC GCT TAT TGA -3′) (15). Amplification proceeded using a PCR Thermal Cycler MP (TaKaRa Biomedicals, Ohtsu, Shiga, Japan) programmed as follows: 15 min at 95°C for initial heat activation and 35 cycles of 30 s at 94°C for denaturation, 30 s at 50°C for annealing, 1 min at 72°C for extension and 10 min at 72°C for final extension. PCR products of 18 S, 5.8S and 25/28 S rRNA regions were digested with MwoI (New England Biolabs, Ipswich, MA, USA) according to the manufacturer's instructions. Digestion products were separated on 2% agarose gels (High Strength Analytical Grade Agarose, Bio-Rad Laboratories) in Tris-borate EDTA buffer (100 mM Tris, 90 mM Borate, 1 mM EDTA, pH 8.4), stained with ethidium bromide and photographed under ultraviolet light. The molecular size marker was a 100-bp DNA Ladder (Invitrogen, Carlsbad, CA, USA).

Candida isolates were tentatively identified based on RFLP analysis (15) and representative isolates conclusively identified by sequence analysis as follows. PCR products were sequenced at Hokkaido System Science (Sapporo, Japan) using the BigDye Terminator Cycle Sequencing Kit and an automated DNA sequencer (PRISM-3100, Applied Biosystems Japan, Tokyo, Japan). Primer 3, described above, was used to sequence, and the 5.8S rRNA gene sequences were then compared to 5.8S rRNA gene sequence data from the GenBank database using the Blast search program through the web site of the National Center for Biotechnology Information. Phylogenetic analysis of the 5.8S rRNA genes of Candida isolates by the neighbor-joining method was performed using DNASIS Pro V2.6 (Hitachi Software Engineering, Yokohama, Japan).

Fisher's exact probability test, Mann–Whitney's U test, Kruskal–Wallis test, and Tukey's test were used to determine statistical significance. A P-value of < 0.05 was considered to be statistically significant.

Among the 15 subjects with OLP, Candida species were detected in eight samples from the surface of the buccal mucosa (53%), five from the dorsal surface of the tongue (33%) and eight of supragingival plaque (53%). Meanwhile, among the seven subjects with healthy oral mucosa, Candida species were detected in two samples from the surface of the buccal mucosa (29%), one from the dorsal surface of the tongue (14%), and two of supragingival plaque (29%). Overall, the detection frequency of Candida species was higher in subjects with OLP (n= 12; 80%) than in subjects with healthy oral mucosa (n= 2; 29%) (P < 0.05).

The mean number of CFU (logarithm CFU/mL) of Candida isolates in subjects with OLP was 3.6 ± 4.0, 1.7 ± 2.1 and 2.8 ± 3.2 for the surfaces of the buccal mucosa, dorsal surfaces of the tongue and supragingival plaque, respectively, while, the respective values in subjects with healthy oral mucosa were 1.9 ± 2.2, 1.0 ± 1.4 and 2.1 ± 2.4. Thus, the amounts of Candida in subjects with OLP were apparently higher than in subjects with healthy oral mucosa, although the differences were not significant.

On PCR-RFLP analysis targeting the 18 S, 5.8S and 25/28 S rRNA genes, isolates generating restriction DNA fragments of 261, 184 and 141 bp were identified as C. albicans, and those generating 414, 174, 171, 86 and 80 bp were identified as C. glabrata, as described previously (15). Furthermore, unidentified isolates after PCR-RFLP analysis were subsequently identified as C. fukuyamaensis and C. parapsilosis by 5.8SrRNA genes sequence analysis (Table 2). Non-C. albicans, that is, C. glabrata, C. fukuyamaensis and C. parapsilosis, were detected only in four of the subjects with OLP.

Table 2.  Number of isolates of Candida species from subjects with oral lichen planus and subjects with healthy oral mucosa
 Subjects with oral lichen planus Number (percentage)Subjects with healthy oral mucosa Number (percentage)
Buccal mucosa
 C. albicans412 (93.4) 7 (100.0)
 C. glabrata 28 (6.3) 0 (0.0)
 C. fukuyamaensis  1 (0.2) 0 (0.0)
 C. parapsilosis  0 (0.0) 0 (0.0)
 Total441 (100.0) 7 (100.0)
Dorsal surface of tongue
 C. albicans  5 (21.7) 7 (100)
 C. glabrata 12 (52.2) 0 (0.0)
 C. fukuyamaensis  4 (17.4) 0 (0.0)
 C. parapsilosis  2 (8.7) 0 (0.0)
 Total 23 (100.0) 7 (100.0)
Supragingival plaque
 C. albicans 84 (73.0)76 (100)
 C. glabrata 30 (26.0) 0 (0.0)
 C. fukuyamaensis  1 (0.9) 0 (0.0)
 C. parapsilosis  0 (0.0) 0 (0.0)
 Total115 (100.0)76 (100.0)

Seven (47%) of 15 subjects with OLP had systemic diseases such as diabetes (20%), hypertension (20%) or hyperlipidemia (7%) (Table 1); in all cases these conditions were well controlled by their regular physicians. Among the Candida species, non-C. albicans species (78 strains) were more frequently isolated from subjects with OLP and diabetes (51 strains, 65%, P < 0.05) or hypertension (14 strains, 18%).

Based on phylogenetic analysis of C. albicans and non-C. albicans isolates (C. glabrata, C. fukuyamaensis and C. parapsilosis), using the neighbor-joining method, C. fukuyamaensis and C. parapsilosis were genetically similar to C. albicans, while C. glabrata was genetically distinct from C. albicans (data not shown).

The mean logarithm CFU/mL for bacterial isolates in subjects with OLP was 8.6 ± 9.0, 9.0 ± 9.4 and 8.2 ± 8.3 for the surfaces of the buccal mucosa, dorsal surfaces of the tongue and supragingival plaque, respectively. In subjects with healthy oral mucosa, the respective values were 8.2 ± 8.5, 8.1 ± 8.4 and 8.6 ± 8.8. The amounts of bacteria did not differ (P > 0.9) between subjects with OLP and subjects with healthy oral mucosa.

The detection frequency of Candida species was significantly higher in subjects with OLP than in those with healthy oral mucosa, thus supporting an association between Candida species and OLP (6, 7). The detection frequency and amount of Candida species from the dorsal surface of the tongue was apparently lower than from the buccal mucosa and supragingival plaque samples, although this difference was not statistically significant. The environment of the dorsal surface of the tongue is relatively anaerobic because of its anatomical structure, and may therefore support the growth of anaerobes, but not of Candida species.

C. albicans was the most predominant of the Candida species isolated in the present study (Table 2). The relationship between C. albicans and oral mucosal diseases has been reported (7, 16, 17); however, this microorganism is also known to be an opportunistic pathogen and is non-pathogenic to healthy humans in general (9, 11), and was indeed isolated from subjects with healthy oral mucosa in the present study (Table 2). On the other hand, using PCR and sequencing, non-C. albicans, that is, C. glabrata, C. fukuyamaensis and C. parapsilosis, were isolated and identified from subjects with OLP, but not from subjects with healthy oral mucosa (Table 2). To our knowledge, this is the first study to report the specific isolation of these non-C. albicans species from subjects with OLP.

Phylogenetic analysis of Candida isolates using the neighbor-joining method revealed that C. albicans, C. fukuyamaensis and C. parapsilosis are genetically similar to one another, while C. glabrata isolates, found only in specimens from subjects with OLP, are genetically distinct from these species. The phylogenetic relationship, that is, the great distance between C. glabrata and C. albicans, is in agreement with previous studies (10, 18–20), suggesting that C. glabrata occupies a unique genetic and biological position within Candida species.

It has been reported that C. glabrata, one of the non-C. albicans, is detected more frequently when chemotherapy, such as fluconazole, has been administered to immunocompromised or debilitated patients, such as those with autoimmune diseases (9, 21, 22). However, none of the subjects in the present study suffered from immunosuppressive diseases, and none had received antibiotic or antifungal agents during the 6 weeks prior to the study. Therefore, the occurrence of non-C. albicans in the subjects with OLP in the present study supports a possible association between these yeasts and OLP.

C. glabrata has been reported to be less susceptible to killing by human beta-defensins and exhibits various degrees of resistance to the antifungal activity of salivary histatins and mucins (11). In addition, this yeast possesses phospholipase activity, which is capable of promoting destruction of cellular membranes (23, 24). These observations support the possibility that this yeast is pathogenic in OLP, although there is no definite evidence that it can initiate pathogenicity in this condition.

It has been reported that hypoglycemic and antihypertensive agents reduce salivary flow in general (25), and that these medications are associated with a high frequency of isolation of Candida species from subjects with OLP (6). However, the subjects in the present study had neither salivary gland dysfunction nor decrease in saliva secretion. In addition, the frequencies and numbers of C. albicans were not associated with systemic diseases such as diabetes and hypertension, whereas non-C. albicans species were more frequently isolated from subjects with diabetes. These findings suggest that diabetes may be associated with the detection of non-C. albicans in subjects with OLP, although further study is required to verify this possibility.

To our knowledge, there have been no reports on the proportions of Candida species in oral microflora in subjects with OLP. In the present study, the mean logarithm CFU/mL of bacterial and Candida isolates were 8.7 and 3.2, respectively; thus the proportion of Candida species in oral microflora of subjects with OLP was approximately 0.0003%. A similarly small percentage of Candida species has also been reported in saliva and tongue coating microflora of geriatric edentulous subjects (13, 26). However, because the size of Candida cells is estimated to be approximately 1000-times larger than that of bacterial cells, the volume proportion of Candida cells among oral microflora of OLP is approximately 0.3%. This proportionate volume of Candida cells, irrespective of health- or disease-related properties, may contribute to interactions between bacteria and host tissue through some unique biological characteristics of Candida species, including co-aggregation with oral bacteria such as Streptococcus, Actinomyces and Fusobacterium (27–29) and resistance to host antifungal activity (9, 11).

All of the findings of the present study support the notion that subjects with OLP are more likely to be orally colonized with Candida than subjects with healthy oral mucosa, and that non-C. albicans species are specifically isolated in patients with OLP, particularly those with OLP and diabetes. Further studies, including large-scale studies, are clearly required in order to analyze the association between non-C. albicans species and OLP and hence elucidate the potential pathogenic roles of non-C. albicans species.

ACKNOWLEDGMENTS

This study was supported in part by Grants-in-Aid for Scientific Research (Nos. 17591959, 19390539, 20592195, 20592220 and 21659483) from the Japan Society for the Promotion of Science.

Ancillary