In this study, sequencing of microbiota in resected thymoma samples identified two genera, Sphingomonas and Phenylobacterium, in almost all thymoma samples; the bacterial composition and abundance of these genera were markedly high. We separately analyzed Type AB thymoma for type A and type B components and detected Sphingomonas and Phenylobacterium in both components. Although the oral microbiome is likely to affect and contaminate the lung microbiome, thymoma is anatomically unlikely to be affected by the oral microbiome [40, 41]. The composition and abundance of these two genera were significantly higher in the microbiota of thymoma tissues than in the microbiota of pancreatic cancer tissues. Our results suggest that these two genera are thymoma-specific microbiota. In addition, the pancreatic cancer and thymoma tissue samples were analyzed in the same process at the genome analysis center of our hospital during the same period. This analysis suggested that the presence of the two genera was not a result of contamination during the analysis process. In contrast, Sphingomonas and Phenylobacterium have not been detected in lung cancer tissues according to recent reviews on the microbiota in patients with lung cancer [4, 42–44]. In this study, because these two genera were detected in almost all thymoma samples, it was suggested that Sphingomonas and Phenylobacterium may represent differential microbiome function in thymoma development.
Sphingomonas is a bacterial genus that was separated from Pseudomonas approximately 30 years ago. Members of the former are Gram-negative bacteria; however, they do not contain lipopolysaccharides specific to Gram-negative bacteria [45]. Instead, these bacteria contain glycosphingolipids, which are found in eukaryotic cells [45]. They are common microorganisms inhabiting various environments, such as water environments (e.g., freshwater and seawater), soil, and plant root systems. The wide ecological distribution of these bacteria is attributed to their ability to use diverse organic compounds and their strong vitality, allowing them to survive in nutrient-poor environments [46]. Although several bacteria in the genus Sphingomonas were isolated in relatively clean environments, certain bacterial species were isolated in contaminated environments containing toxic organic compounds, such as polychlorinated biphenyl, creosote, and pentachlorophenol [47]. Subsequent studies revealed that these bacteria take up certain organic contaminants and use them as energy sources [48]. On the basis of these findings, progress has been made in elucidating the mechanism through which Sphingomonas metabolizes organic contaminants. Furthermore, several attempts have been made worldwide for applying this mechanism in environmental cleanup (bioremediation). Meanwhile, with respect to the microbiome, Sphingomonas has been reported to be enriched as blood microbiota in the serum of healthy patients and patients with breast cancer who exhibit a favorable prognosis [49, 50].
The genus Phenylobacterium comprises a single species called P. immobile, which is remarkable for its extremely limited nutritional spectrum. All strains isolated and described hitherto grow optimally only on artificial compounds such as chloridazon, antipyrin, and pyramidon [51]. Species within the genus Phenylobacterium, are capable of degrading xenobiotic compounds with a phenyl moiety such as chloridazon, antipyrine, pyramidon, or their analogues [52]. Additionally, these bacteria can degrade polycyclic aromatic hydrocarbons [53]. Phenylobacterium has now been used in the bioremediation of a petroleum-contaminated soil to degrade polycyclic aromatic hydrocarbons and their analogues [54]. Unlike Sphingomonas, there has been no report of the detection of Phenylobacterium as blood microbiota. Future studies are expected to elucidate how Sphingomonas and Phenylobacterium, which are two genera of environmentally indigenous bacteria used for bioremediation, coexist in thymoma and how they are involved in the carcinogenic mechanism of thymoma.
Several indigenous microorganisms exist in the epithelium of the whole human body (e.g., the mouth, ear, nasal cavity, respiratory organs, digestive tract, skin, and reproductive organs), form microbiota, play various roles in the body, and form a symbiotic relationship with humans [1, 2]. In recent years, it has been considered that disturbance in the microbiota composition (dysbiosis) may alter the risk of disease development, and there is a growing number of reports on the association between intestinal microbiota and several diseases, such as allergy, cancer, multiple sclerosis, Parkinson's disease, depression, inflammatory bowel disease, and rheumatism [42]. Furthermore, sterilization and specific-pathogen-free breeding have been reported to alleviate or cure these diseases in pathological mouse models [55]. Improvement of the microbiota may additionally prevent the development of diseases in humans [56]. If one or several species of bacteria cause a disease, they can be potential therapeutic targets. For example, eradication of Helicobacter pylori is the standard of care for the prevention of gastric cancer in infected patients at present [57]. This study clearly identified the microbiota involved in the development of thymoma. Considering the future clinical application of this result, the development of thymoma can be prevented through controlling the bacterial genera Sphingomonas and Phenylobacterium. Patients with myasthenia gravis are at a high risk of developing thymoma [16, 17], and prevention of thymoma is important for their long-term survival. In this study, case 20 involved a patient with thymoma complicated by myasthenia gravis (Figs. 1 and 2); this patient was positive for Sphingomonas and Phenylobacterium, which were abundant. The development of probiotic models for antibiotics, vaccines, and other therapies targeting these genera identified in this study may be important for the prevention of thymoma.
Bacterial diversity tended to be higher in type A thymoma (least aggressive type) than in type B (more aggressive type). A study comparing the microbiota between tumor and normal peritumoral tissues in lung cancer demonstrated that the bacterial diversity was significantly higher in the normal peritumoral tissues [6]. According to these data, cancer aggressiveness and alpha diversity are negatively correlated. Because the cancer microenvironment is more perturbed, dysbiosis might be enhanced; consequently, bacterial diversity might decrease. In addition, because the lymphocyte counts in the tissues are higher in type B thymoma than in type A thymoma, the immunity against these bacteria may fundamentally differ between these types.
Using LEfSe analysis, we identified variation in specific species between type A and B tumors and between tumors with and without GTF2I mutation, indicating the differential microbiome function in the development of each type of tumor [58]. We determined that Firmicutes, Bacilli, and Lactobacillales were common Gram-positive bacteria in type A thymoma and Proteobacteria were common Gram-negative bacteria in type B thymoma. When the P-value based on the Kruskal–Wallis test was increased from 0.05 to 0.1 (Supplementary Fig. 2), 15 of 20 bacteria that were significantly detected in the microbiota of type A thymoma were Gram-positive bacteria, and all four bacteria significantly detected in the microbiota of type B thymoma were Gram-negative bacteria. Although these findings indicated a correlation between the histological types (types A and B) and Gram staining results for the microbiota, the biological significance of this correlation is unknown. Because Gram-negative bacteria are generally more pathogenic than Gram-positive bacteria, the former may be involved in carcinogenesis in type B thymoma, which is a more aggressive phenotype. Additionally, it is unclear from our observational study whether the identified bacterial differences are causally related to lung carcinogenesis or merely reflective of the disease process in thymoma. In the future, detailed studies with a larger sample size may be needed.
We previously reported that GTF2I mutation is a driver mutation in thymoma [39]. In the present study, specific species were identified between tumors with and without GTF2I mutation. While Alphaproteobacteria were detected in significantly more cases without the driver mutation in the GTF2I gene, Alphaproteobacteria are reported to induce natural genetic transformation, a process in which genetic material passes from one bacterium to another, as reported in Neisseria gonorrhoeae and Haemophilus influenzae [59]. Natural genetic transformation is a process involving DNA transfer from one bacterial cell to another through the integration of the donor sequence into the recipient genome. Such an alternative pathway may be involved in the oncological development of thymoma without driver mutation. Additionally, the mechanisms through which the microbiota contributes to carcinogenesis need to be examined in detail using a large sample size in the future.
This study is associated with some limitations. First, the patient cohort was relatively small owing to the rarity of the tumor. Second, patient survival could not be analyzed as no patients have shown recurrence in the cohort. Third, no blood samples were analyzed for microbiota containing the two genera, Sphingomonas and Phenylobacterium. Analysis of blood samples might have elucidated the reasons for the presence of the microbiota in the sterile anterior mediastinal environment. In addition, the higher abundance of Sphingomonas and Phenylobacterium may be related to the impaired immunity of the tumor microenvironment, which may cause proliferation of these bacteria in the blood. Thus, they may be clinically applicable as serum biomarkers for thymoma. In this context, a larger series of studies need to be performed for evaluating the microbiome landscape of thymomas more comprehensively and elucidate associations with clinical parameters through a more exhaustive multivariate analysis. Nevertheless, since the major aim of this preliminary analysis was identification of the thymoma-specific microbiota that should be prioritized for clinical development, the modestly sized samples provide useful insights.