The use of random mutagenesis in the functional annotation of the Streptococcus uberis genome

Streptococcus uberis is a prolific and virulent mastitic pathogen, a disease that is a substantial welfare issue, costing the global dairy industry billions of pounds each year. Research into identifying control measures for this bacterium have so far been unsuccessful; partly due to it ubiquitous presence in the environment of dairy cattle and it’s the multifactorial nature of the disease pathogenesis. High-throughput, transposon insertions sequencing (TIS) protocols utilise libraries of bacteria, mutagenised using insertional transposon techniques, to screen for genomic regions involved in survival under particular environments. Significant advances in sequencing technology and the ease and availability of data generation have fuelled TIS experiments. However, the cost of these protocols coupled with the requirements to alter core sequencing-machine parameters, subsequently the experiments are implausible for those without free access to a sequencing platform. Addressing these problems and wanting to provide an accessible, accurate, data analysis package, Pragmatic Insertion Mutant Mapping System (PIMMS) was developed. PIMMS builds on the use of the highly efficient, random mutagenic tool, pGhost9::ISS1 and extends the principles of single mutant identification by inverse PCR. This enabled the enrichment of the transposon insertion/chromosomal DNA junction from a population of 100,000 mutant bacteria, generating a sample that could be used directly for high-throughput sequencing. Making use of multiple bioinformatic tools, a data analysis pipeline was created allowing for the raw sequence data to be analysed, isolating reads containing the DNA target, mapping them to the reference genome and creating a summary tabular output. This contains all of the insertion positions identified accompanied by an array of normalised data in order to identify genes of consequence to the selective condition under investigation. Using these methods allowed identification of a core set of 196 genes essential for growth of Streptococcus uberis on agar. Further interrogation of this bacterium in host specific environments afforded the identification of four further subsets of genes important for growth in the associated conditions. During growth in liquid medium a total of 411 non-mutated genes were identified and a further 69 non-mutated genes were identified following the addition of lactoferrin (a divalent metal ion chelating protein) to the media. After growth in skimmed milk, 383 genes were identified as being conditionally essential. The metal binding protein, mtuA, was found to be devoid of insertions, in the presence of lactoferrin and in skimmed milk, complementing other published studies that show individual mutant strains of S. uberis, lacking MtuA are unable to grow in milk or in the presence of lactoferrin. Identification of the mtuA sequence using the PIMMS procedure provided a validation of the techniques ability to detect appropriate sequences. Following supplementation of skimmed milk bovine serum (25%), a total of 656 CDS were deemed conditionally essential, incorporating most of those required for growth in skimmed milk alone, but also included additional sequences; seven of which were similar to those essential for growth of Staphylococcus aureus in a similar environment. The precise protocol developed will enable the advancement of the initial stages of functional genomics in streptococci; due to the wide range of species in which pGhost9::ISS1 will integrate. Also the bioinformatic pipeline can be coupled with any insertional mutagen to enable cost effective interrogation of any bacteria with an appropriate mutagenesis system