CORDIS - EU research results

Methods for quantification of free protozea vesicles containing legionella in a water environment

Final Report Summary - VESICOUNT (Methods for quantification of free protozea vesicles containing legionella in a water environment)

Problematic Legionellosis is a disease that has emerged in the last half of the 20th century as a consequence of human alteration of the environment. It covers actually two distinct clinical entities : Pontiac fever, a self-limited flu-like illness, and Legionnaires’ disease, a severe multisystem affection involving pneumonia that may even be lethal. The main causative agent is the rod-like gram-negative bacteria Legionella pneumophila, responsible in Europe of 94,6% of all reported cases (10,715 cases declared in 34 countries between 2007 and 2008), but the genus Legionella is actually composed of more than fifty species, half of them being human pathogen.

Legionellae are ubiquitous in natural environment, being detected in up to 80% of freshwater sites by PCR. They generally maintain there at a too low level to be harmful. However, they may actively multiply when exposed to temperatures between 20°C and 42°C, and supplied with essential nutrients such as iron and L-cystein. This condition can often be achieved in large hot water systems such as cooling towers or water network of collective buildings, like those of hotels and hospitals.

Human contamination generally occurs when water containing a high level of Legionellae forms an aerosol with droplets size between 2 and 5 µm. These droplets are small enough to deeply penetrate pulmonary track. But their sizes allow them also to harbour enough bacteria to infect efficiently pulmonary epithelial cells in which they afterwards multiply.

A significant portion of legionellosis outbreaks is related to anthropogenic water ecosystem (industrial complexes, cooling towers, hot water system) all over the world. The risk to contract the disease mostly relies on the concentration of Legionella in water, the virulence of the strain, the individual susceptibility of the exposed persons and the existence of aerosols enhancing bacterial inhalation, However, the exact composition of this infective aerosol is not known and may explain some of the discrepancies observed between reported cases, Protozoans naturally present in water have been suspected since a long time to play a role in this infectivity. Several authors have reported the capacity of various protozoans to abridge and/or favor the growth of Legionella, as well as to generate smaller Legionella-containing entities (generally named "vesicles"), which become easily airborne if water droplets are formed

Objectives of the Vesicount project

The Vesicount project aimed to produce artificially, identify and numerate some of these entities expelled by protozoan, with the final objective to be able to recognize and count them easily in industrial water. The couple formed by Legionella pneumophila sg1 strain Paris (CIP 107-629-T) and Acanthamoeba castellanii (ATCC 30234) was used as a model system. Lp sgl strain Corby, as well as different amoeba strains were also used. Various protocols were tested in order to set up the most efficient production methodology. For the production of amoebal vesicles, a robust and reproducible procedure was established and tested concomitantly by all participating laboratories; its accuracy was also verified using L. pneumophila strains expressing GFP.

Main results

We first reviewed that a great number of “structures” could contain Legionella; among which those generated by amoeba and ciliate are the most documented. This led us to classified these “structures” in eleven categories, which in turn allowed us to better define the entities to be measured (the mesurand). We then could clarify the purification and detection steps necessary for the analysis of each category.

The next step was the achievement of a state-of-art of the different methods that could be used for vesicles detection and to define selective criteria that would help us to determine the most relevant ones. Since such methods do not exist “per se”, seven different protocols were generated by the combination of different available laboratory techniques. Among those seven protocols, one stable and efficient protocol of vesicle production was selected. This was achieved using a model system composed of a co-culture of Acanthamoeba castellanii (ATCC 30234), a representative of common environmental amoebae, and the Legionella pneumophila “Paris” strain, commonly found in clinical or environmental isolate and able to infect a broad range of environmental hosts. This model is expected to fit with the vast majority of naturally encountered cases. The same protocol was also applied using two genetically manipulated Legionella pneumophila strains expressing the green fluorescent protein (GFP). The bright fluorescence generated by these strains is more easily observed inside amoebal structures and vesicles, and will be of great help to standardize the future laboratory assays.

The effect of variation of different parameters such as temperature, MOl (Legionella on Protozoan ratio at the infection), pH, medium composition, etc was also tested. Together with the external vesicle-like structures which contain viable bacteria and were already observed by several authors, other types of Legionellae clusters associated most of the time to remnants of membranes were also characterized and may represent another potential danger for human health

As a next point, we established new tools that would be helpful in the capture of vesicles in natural system. In this purpose, two antibodies against Acanthamoeba spp. were produced by standard methods. Immunizing rabbits with amoeba lysate produced a polyclonal antibody, whereas a monoclonal antibody (57/1 fraction) was prepared by immunizing mice. Amoebae recognition was affirmed by Western blot. The capacity of these antibodies to recognize also vesicles membranes is still under investigation.

The last point concerns the application of the established method to environmental samples. The hope was to correlate the level of Legionella infection with the ecosystem conditions in order to anticipate or determine the amount and the type of respirable vesicles that can be or are effectively released in environmental water.

Measurements were done each week during four months in an industrial plant with two different cooling tower circuits. The make up water is the same for the both circuits and is pumped in a river. At first a risk analysis of each the circuit was done and sampling points were chosen downstream of each sensible elements. Firstly, all spherical objects between 1.2 µm and 5 µm that look like organelles of protozoa were counted. Then, all objects like vesicle containing" Legionella-like-structures " and pieces of dead protozoa were also counted.

Counting the "organelles" showed how a significant quantity of protozoa could arrive inside the two circuits, coming from the river, when its temperature exceeded thirteen degrees Celsius. Counting objects like vesicles containing Legionella-like-structures demonstrated a short episode during which the river provided the circuits with these objects and showed that some elements of the circuit might promote the development of protozoa. Counting the remnants of dead protozoa showed the existence of an important mortality phase of protozoa in the first circuit and two mortality phases in the second one.

Main conclusions

• Data collected during this work have lead to a better knowledge of expelled elements produced, which may in turn be useful and generate new tools for water system monitoring, They also allowed us to better understand the conditions required for the infection of Acanthamoeba castellanii by Legionella pneumophila, opening the way to exciting perspective for water system control.
• It was also demonstrated that counting "organelles" and fragments of protozoan provides some indications about the activity of protozoa. This should allow choosing which portions of the circuit were the most interesting to perform the monitoring, and knowing when the chemical controls, the bacteriological controls and the bactericide treatments should be reinforced.

More information on the web site of the project :

Contact : Prof. A. Pauss,