Assessing Genetic Diversity Among Legionella pneumophila ST1 Isolates

Legionella pneumophila is the primary pathogen responsible for Legionnaire’s Disease, a severe and sometimes fatal pneumonia. In order to quickly identify the environmental source during disease outbreaks, a Sequence-Based... [ view full abstract ]

Legionella pneumophila is the primary pathogen responsible for Legionnaire’s Disease, a severe and sometimes fatal pneumonia. In order to quickly identify the environmental source during disease outbreaks, a Sequence-Based Typing (SBT) schema of 7 variable loci was developed to rapidly classify isolates into sequence types (ST) based on differences in their allelic profiles.  This system has proven valuable for exclusion of non-matching STs from potential outbreak sources, however, SBT cannot distinguish potentially common strains of the same ST found in one or more sources.  Over 44% of 597 L. pneumophila isolates typed by SBT at the Center for Disease Control and Prevention belong to ST1, making source attribution of ST1 infections particularly difficult with SBT alone.

Whole genome sequencing-based typing methods, such as whole genome multi-locus sequence typing (wgMLST) and single nucleotide polymorphism (SNP) tree approaches, allow for greater discrimination of Legionella outbreak isolates compared to SBT typing; however, these methods require equipment and computational resources that may be unavailable to some diagnostic laboratories.  Furthermore, these whole genome sequencing methods do not provide an unambiguous “match/no match” result and require expertise to interpret.  Ideally, an extension of the simple and established SBT schema to include additional loci that allow increased discrimination within the ST1 family would provide additional information for rapid isolate classification without substantially changing current protocols or burdening laboratories that use this method.

To identify additional loci that might be useful in rapidly distinguishing between ST1 isolates in an outbreak, we chose to evaluate the accumulation of SNPs along the length of the L. pneumophila chromosome.  Using a large number of ST1 genomes as well as closely related STs (n = 491), we created a reference-based alignment drawing from clinical and environmental isolate sequencing datasets in both outbreak and non-outbreak settings from multiple continents.  We used this alignment to conduct a sliding window analysis along the L. pneumophila chromosome to characterize nucleotide diversity.  This analysis revealed a high degree of nucleotide conservation between isolates, with an average pairwise divergence of 0.08% and a maximum divergence of 0.59%.  Additionally, 97.89% of over 1.5 million core nucleotides were found to be invariant in all ST1 and ST1-related lineages, suggesting the vast majority of nucleotide sites evolve very slowly over time.  The distribution of detected SNPs along the genome was not random.  The majority of observed nucleotide variation in ST1-related isolates was concentrated in distinct clusters which comprised approximately 30% of the genome.  This suggests that the ST1 genome is dominated by large regions of strictly conserved sequence punctuated by hotspots of mutation or recombination regions that are responsible for much of the observed diversity within this family of STs.  These variable genes represent ideal candidate targets for extending the SBT loci to further discriminate ST1 isolates.

The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention.