Date Published: November 20, 2015
Publisher: Public Library of Science
Author(s): Sang Hu Kim, Shawn T. Clark, Anuradha Surendra, Julia K. Copeland, Pauline W. Wang, Ron Ammar, Cathy Collins, D. Elizabeth Tullis, Corey Nislow, David M. Hwang, David S. Guttman, Leah E. Cowen, Deborah A. Hogan.
The microbiome shapes diverse facets of human biology and disease, with the importance of fungi only beginning to be appreciated. Microbial communities infiltrate diverse anatomical sites as with the respiratory tract of healthy humans and those with diseases such as cystic fibrosis, where chronic colonization and infection lead to clinical decline. Although fungi are frequently recovered from cystic fibrosis patient sputum samples and have been associated with deterioration of lung function, understanding of species and population dynamics remains in its infancy. Here, we coupled high-throughput sequencing of the ribosomal RNA internal transcribed spacer 1 (ITS1) with phenotypic and genotypic analyses of fungi from 89 sputum samples from 28 cystic fibrosis patients. Fungal communities defined by sequencing were concordant with those defined by culture-based analyses of 1,603 isolates from the same samples. Different patients harbored distinct fungal communities. There were detectable trends, however, including colonization with Candida and Aspergillus species, which was not perturbed by clinical exacerbation or treatment. We identified considerable inter- and intra-species phenotypic variation in traits important for host adaptation, including antifungal drug resistance and morphogenesis. While variation in drug resistance was largely between species, striking variation in morphogenesis emerged within Candida species. Filamentation was uncoupled from inducing cues in 28 Candida isolates recovered from six patients. The filamentous isolates were resistant to the filamentation-repressive effects of Pseudomonas aeruginosa, implicating inter-kingdom interactions as the selective force. Genome sequencing revealed that all but one of the filamentous isolates harbored mutations in the transcriptional repressor NRG1; such mutations were necessary and sufficient for the filamentous phenotype. Six independent nrg1 mutations arose in Candida isolates from different patients, providing a poignant example of parallel evolution. Together, this combined clinical-genomic approach provides a high-resolution portrait of the fungal microbiome of cystic fibrosis patient lungs and identifies a genetic basis of pathogen adaptation.
The microbiome has a profound impact on diverse facets of human biology and disease. Anatomical sites such as the respiratory tract that were once thought to be sterile are now recognized to harbor complex microbial communities in healthy individuals as well as those suffering from a multitude of conditions . One respiratory condition for which chronic and complex polymicrobial communities are now appreciated to have a severe impact on prognosis is cystic fibrosis. Patients with this genetic disorder caused by reduced function of the transmembrane conductance regulator CFTR experience thick mucus accumulation in airways, which renders them vulnerable to chronic airway infection and repeated episodes of pulmonary exacerbation . Our understanding of microbiomes of the lung and respiratory tract of cystic fibrosis patients is based largely on studies focused on bacteria [2,3]. Traditionally, opportunistic bacterial pathogens such as Pseudomonas aeruginosa, Burkholderia species, and Staphylococcus aureus have been implicated in pulmonary exacerbations . Molecular approaches have provided additional insight into culprits of respiratory infections in cystic fibrosis patients, as with the implication of a bacterial opportunistic pathogen of the Streptococcus milleri group that had been previously overlooked [2,4]. The resultant episodes of pulmonary exacerbations cause decline in lung function, ultimately leading to patient death.
Here, we provide a portrait of species and population level dynamics in the lung mycobiome of cystic fibrosis patients, and illuminate the power of complementary high-throughput sequencing coupled with phenotypic and genotypic analyses. Utilizing high-throughput sequencing of the ribosomal RNA internal transcribed spacer ITS1 amplified from longitudinal sputum samples from 28 cystic fibrosis patients (Fig 1), we were able to identify fungi that were not isolated through culture-based methods, likely due to their specific growth requirements as with lipophilic Malassezia species  and slow growing E. dermatitidis . Our culture-based analysis of 1,603 fungal isolates from the same sputum samples revealed extensive inter- and intra-species phenotypic diversity in growth rate, drug resistance, and cellular morphology (Figs 2–4), with profound implications for inter-kingdom microbial interactions and pathogen adaptation to the host (Fig 5). The strong correlation between our molecular and culture-based methods to define fungal communities (Fig 1) provides strong validation of our approach to study species and population level dynamics in the host.