Document Type


Date of Degree

Spring 2019

Access Restrictions

Access restricted until 07/29/2021

Degree Name

PhD (Doctor of Philosophy)

Degree In


First Advisor

Petersen, Christine

Second Advisor

DesJardin, Lucy

First Committee Member

Petersen, Christine

Second Committee Member

DesJardin, Lucy

Third Committee Member

Ryckman, Kelli

Fourth Committee Member

Braun, Terry

Fifth Committee Member

Breheny, Patrick


Legionella bacteria, the causative agent of Legionaries’ disease and Pontiac fever, are ubiquitous in fresh-water environments including man-made water systems. Incidence of legionellosis is increasing in the United States resulting in thousands of cases every year. Infection via aerosols generated by showers, faucets, cooling towers, spas, fountains, and other water fixtures has been identified as the primary source of transmission. Legionella bacteria pose a significant public health threat, particularly in health care and long term care settings as Legionella can readily colonize the plumbing systems and infect the vulnerable patient population. One species, Legionella pneumophila (Lp), is responsible for over 90% of the known cases of Legionnaires’ disease. The importance of genetic diversity of Lp and non-pneumophila strains in human disease remains an area of ongoing research.

Little is known in regard to the phylogenetic diversity of environmental strains, particularly strains that colonize facilities with high risk populations such as hospitals. Whole-genome sequencing (WGS) analysis, is an emerging tool used to support epidemiological investigation of cases of legionellosis and can be used to describe and establish phylogenetic relationships between environmental strains and clinical cases. The advantage of this method is the ability to differentiate bacteria down to the level of single nucleotide polymorphisms (SNPs). However, it was unknown whether current WGS methods accurately represent the potential SNP diversity among Lp isolates from the same environmental sample.

It is unclear as to why certain strains tend be associated with clinical cases more than others, but certain genes referred to as virulence factors may be related to the relative pathogenicity of Legionella strains. Further investigation into virulence factors and antibiotic resistance factors could be used in future risk assessment of environmental Legionella. Additionally, Legionella have the potential for high genetic diversity due to recombination events, and gene transfer can occur between distinct Legionella species and strains. There is a lack of research on the potential sharing of virulence factor genes between Legionella strains typically associated with disease and those considered to be non-virulent.

The goal of the work presented in this thesis is to describe the diversity of phylogenetic relationships between Lp isolates found in hospital premise plumbing systems, to estimate the genetic diversity among Lp found in the same environmental sample, and to identify virulence and antibiotic resistance genes shared between Legionella strains. A better understanding of the genetic diversity of environmental Lp could inform future surveillance and outbreak investigations by demonstrating the need to collect samples from multiple sites within a facility, and identifying shared virulence and antibiotic resistance genes between Legionella species and strains could apprise future risk assessment.

WGS was utilized to describe the phylogenetic relationships of 81 Lp isolates from five hospitals. Individual hospitals were found to have distinct strains of Lp. For some strains, highly conserved subpopulations were collected from the same room over time, whereas other strains did not cluster by room. Using prospectively collected isolates from two hospitals, the mean number of SNP differences among isolates from the same environmental sample was found to differ between hospitals (0.4 versus 7.5).

The presence of virulence factors and antibiotic resistance genes in Legionella species and strains was described. An analysis of 10 virulence factor genes revealed that Lp likely did not share these genes with Legionella anisa, a species generally considered to be non-virulent. Within Lp strains there was no clear difference between the Lp strains considered to be more virulent and those considered to be less virulent. A few antibiotic resistance genes were also identified. Following an in vitro assay, only the identified genes associated with macrolide resistance, LpeA and LpeB, were found to impact a quantifiable measure of antimicrobial resistance.

The results of these studies emphasize the importance of understanding the context of an individual facility in Legionella related studies. Importantly, the observations or trends of one facility should not necessarily be applied to another. Legionella genetic diversity was highly conserved in some facilities, whereas in others there was greater diversity as measured by SNP differences. Within sample SNP differences was also variable between hospitals. The virulence findings gave a clear indication of the limited virulence capacity of L. anisa. These findings could explain the limited potential of L. anisa to cause disease in humans. However, a lack of difference among Lp strains may be cause to reassess the potential risk of these other strains especially in diagnostic practices. Finally, some strains of Lp have genes that may contribute to resistance to the leading antibiotic treatments for Legionnaires’ disease. Overall, this research further demonstrates the power of WGS as multiple questions can be addressed using this methodology.


antibiotic resistance, Legionella, phylogeny, virulence, Whole-genome sequencing


x, 86 pages


Includes bibliographical references (pages 79-86).


Copyright © 2019 Wesley Johnathan Hottel

Available for download on Thursday, July 29, 2021

Included in

Epidemiology Commons