Metagenomic analysis of the lung tuberculomas microbiome: paucibacillary bacterial community

Abstract

Introduction. Globally, an estimated 10.0 million (range, 9.0–11.1 million) people fell ill with tuberculosis (TB) in 2018, a number that has been relatively stable in recent years [1]. Despite the obvious differences in the virulence and transmissibility of different genotypes [2], the pathogenesis of tuberculosis, especially in the later stages, is still poorly understood [3]. At the same time, some researchers suggest that the ability of Mycobacterium tuberculosis to form biofilms in vivo [3] is an important aspect of pathogenesis TB that should be used in the development of new generation vaccines [4]. Previously, we suggested that in the long treatment of TB, a specific lung microbiota can develop in the caseous contents of tuberculosis foci, forming a polymicrobial biofilm, resistant to anti-tuberculosis therapy and contributing to the survival of M. tuberculosis [5]. Here, we present results of the metagenomic analysis of 14 tuberculomas and discuss the results obtained. Methods. The study recruited 13 patients undergoing planned surgical treatment at the Irkutsk Regional TB Hospital, Russia. DNA was isolated from caseous contents of 14 tuberculomas obtained during surgery. Two samples were obtained from one patient at two time points separated by six months. Next-generation sequencing of DNA samples was performed at the NextSeq 550 platform (Illumina, USA). The genomic DNA library was prepared by Nextera XT kit (Illumina). High Performance Sequence Analysis was completed on the NextSeq 550 platform (Illumina, USA) with NextSeq 500/550 v2.5 sequencing reagent kits (Illumina, USA). The NGS short sequencing reads (fastq files) were deposited to the NCBI (PRJNA659860). The assembly of short reads pertaining to mycobacterial DNA from the total pool of metagenomic sequences was carried out with the Magic-BLAST tools [6]. Mapping of short reads to genome of the reference strain M. tuberculosis H37Rv (NC_000962) was performed using Burrows-Wheeler Aligner [7]. The extraction of sequences from the mapped genome with the calculation of the coverage of each position was performed with SamTools [8]. Taxonomic classification of high-throughput sequencing reads from metagenomic whole genome sequencing was performed by using Kraken 2 [9]. Results. Among 5167 bacterial reads, a total of 105 families were identified. Of them, 68 taxa belonged to Gram-negative bacteria, and 35 to Gram-positive. At the same time, the number of reads of Gram-negative and Gram-positive taxa was very similar - 2430 and 2746, respectively. This was probably due to the presence of 1801 reads from the family Mycobacteriaceae (M. tuberculosis). The microbial content of the tuberculomas was estimated as 104-108 bacterial genomes per 1 gram of sample. The vast majority of reads (over 95%) were attributed to the human DNA. We have subdivided the samples into two types of bacterial communities: (i) paucibacillary bacterial community with a predominance of TB (63%-96%) and (ii) low-bacterial community with a predominance of non-mycobacterial taxa (13%-38%). Staphylococcaceae, Pseudomonadaceae, Eggerthellaceae, Pasteurellaceae, and Acetobacteraceae prevailed among non-mycobacterial communities. Conclusions. It can be unambiguously concluded that the microbiota of about half of the tuberculomas did not have a dominant taxon or it differed from M. tuberculosis. Acknowledgments. This work was supported by Russian Foundation for Basic Research (Grant No. 19-515-55009).