A Capture-Based Technique for Enhancing RNA-Seq Analysis of Bacterial Transcriptomes during in Vitro and In Vivo Infections

To fully understand the dynamic spatial and temporal interactions of a pathogen with its host, it is necessary to analyze the transcriptomes of both the host and the pathogen throughout the course of an infection. Even using massively parallel techniques such as microarrays or RNA-Seq, analysis of the pathogen transcriptome is complicated by the overwhelming amount of host RNA (>99.9%) associated with an infected sample.

We have developed a solution-phase hybridization-based capture approach to bind specifically to pathogen transcripts, followed by a spin column-based chromatography step to recover the bound transcripts. This protocol dramatically increases the number of reads mapping to pathogen genes by >100-fold. We demonstrate application of the technique to in vitro tissue culture infections with a multi-drug resistant isolate of Klebsiella pneumomiae, and demonstrate high-level constitutive expression of numerous drug resistance genes even in the absence of antimicrobial drugs. We further detect expression of putative virulence-associated genes throughout the course of the infection. We also present preliminary investigations into in vivo infections with Klebsiella pneumoniae or Pseudomonas aeruginosa in the murine lung, demonstrating the capability of the technique for simultaneous host and pathogen expression profiling. Further, we are developing consumables and automation to increase the throughput of the capture technique, in an attempt to reduce the hands-on time and improve the reproducibility of the technique. We discuss possible applications of the technique to targeted drug discovery, as well as potential applications to other (non-infectious disease) systems where RNA of a species of interest is greatly outnumbered in a background of non-informative RNA.