Abstract
The presence of Waddlia chondrophila has been related to respiratory tract infections and human and animal fetal death. Although several sources of infection have been suggested, the actual source remains unknown and limited information exists on the prevalence of W. chondrophila in the environment. This pathogen has been previously detected in well water but its presence has not been confirmed in water networks. Since these bacteria have been detected in water reservoirs, it has been hypothesized that they can access artificial water systems and survive until they find appropriate conditions to proliferate. In this work, their presence in water samples from 19 non-domestic water networks was tested by quantitative polymerase chain reaction (qPCR). Approximately half of the networks (47%) were positive for W. chondrophila and the overall results revealed 20% positive samples (12/59). Furthermore, most of the samples showed low concentrations of the pathogen (<200 genomic units/L). This finding demonstrates that W. chondrophila can colonize some water networks. Therefore, they must be considered as potential infection sources in future epidemiological studies.
INTRODUCTION
In recent decades, the understanding of the role of free-living amoebae (FLA) in the transmission of infectious diseases has changed. Initial research data suggested a strong relationship between pathogenic bacteria and FLA, followed by experimental evidence that this interaction is responsible for most pathogen proliferation as endosymbionts (Greub & Raoult 2004; Thomas et al. 2010; Codony et al. 2012a; Scheid 2014). An example of an endosymbiont is Waddlia chondrophila (Michel et al. 2004). This microorganism is an obligate intracellular bacterium belonging to its own family, Waddliaceae, and to the order Chlamydiales. From its first detection in bovine fetal tissues (Rurangirwa et al. 1999) to date, evidence regarding its involvement as an agent of human miscarriages has been demonstrated (Baud et al. 2004, 2007, 2011, 2014; Baud & Greub 2011). Moreover, Waddlia is likely implicated as an agent of lower respiratory tract infection (bronchiolitis, bronchitis, pneumonia) since Waddlia chondrophila DNA was identified in respiratory tract samples from children with pneumonia (Lamoth & Greub 2010). Their role in lung infection is further supported by a recent animal model demonstrating, in mice, the 3rd Koch postulate for this pathogenic bacteria (Pilloux et al. 2016).
Nevertheless, very little is known about the environmental distribution, prevalence, and infection source of W. chondrophila. The zoonotic transmission of W. chondrophila has been suggested as a potential infection source, in addition to the ingestion of contaminated water, meat, milk, and also sexual transmission (Corsaro & Greub 2006; Lamoth et al. 2015; Vasilevsky et al. 2015). A previous study showed that W. chondrophila was present in 25% of well water samples (n = 40) analyzed in a small study conducted in Spain (Codony et al. 2012b). In the same work, 30 domestic drinking water samples were tested for the presence of this pathogen and, in all of these cases, the results were negative. However, the presence of W. chondrophila in well water reinforces the need for further evaluation of other water networks, in order to fully understand the potential risks associated with the proliferation of W. chondrophila in these artificial systems.
MATERIALS AND METHODS
Water samples were collected according to standard methods based on the ISO/CD 19458:2006 Standard. The samples were taken by the Centre Scientifique et Technique du Bâtiment (CSTB, France) in the context of a research program focused on detecting the risk associated with the presence of Legionella in public buildings. Samples were transported in the dark and at <8°C and stored at 2–5°C until analyzed, within 24 h following sampling.
The Legionella analysis was done by culture according to ISO 11731:1998. On the other hand, W. chondrophila concentration was measured by quantitative polymerase chain reaction (qPCR). In this case, 500 mL of water was filtered through a 0.2 μm pore size PVDF filter, which was used for subsequent DNA purification using the High Pure PCR Template Preparation Kit (Roche Molecular Diagnostics, Mannheim, Germany) according to manufacturer's instructions.
The qPCR was done in the MSM-Lab at the Polytechnic University of Barcelona, in a blind mode, without knowledge of any data about the water source and other analytical results.
The qPCR procedure was based on previous work by Goy et al. (2009) and was performed on a LightCycler-1.5 PCR system (Roche Molecular Diagnostics). Briefly, the reaction mixture consisted of 10 μL Fast Start Taqman Probe Master (Roche Molecular Diagnostics); 0.4 U Uracil-DNA-glycosylase (UDG, New England Biolabs, UK); 9 μL DNA sample; 0.2 μM W. chondrophila specific primers WadF4 (5′-GGCCCTTGGGTCGTAAAGTTCT-3′) and WadR4 (5′-CGGAGTTAGCCGGTGCTTCT-3′); and 0.1 μM probe WadS2 (5′-FAM-CATGGGAACAAGAGAAGGATG-BHQ-3′). The primers amplified a 101-bp DNA fragment of the 16S rRNA gene of W. chondrophila. The probe contained locked nucleic acids (underlined in the sequence above). The qPCR conditions were optimized previously by Codony et al. (2012b) as follows: 2 min at 50°C, 10 min at 95°C as well as 50 cycles of 15 s at 95°C and 1 min at 60°C. A negative control (water, PCR-grade) and a positive control (DNA from W. chondrophila) were included in each run.
RESULTS AND DISCUSSION
The raw data are shown in Table 1 and summarized in Table 2. Briefly, a total of 59 samples from 19 different hot water networks were analyzed. The most significant finding was the high percentage of W. chondrophila in 47% of the analyzed networks, with 12 positive samples from a total of 59 samples. Furthermore, it should be noted that in most of the samples, the W. chondrophila numbers were below 200 GU/L. However, one sample with temperature lower than 40 °C showed the highest W. chondrophila numbers (1,000 GU/L). Additionally, seven samples were positive for Legionellae in four of the networks analyzed. Although both microorganisms need to interact with FLA, these results do not suggest a positive correlation between Legionellae and W. chondrophila. Interestingly, more positive samples (9 vs. 4) were colonized by W. chondrophila when compared with Legionellae.
Water analysis results by building type
Building . | T (°C) . | time (s) . | L.pn cfu/L . | L. spp cfu/L . | W. c GU/L . |
---|---|---|---|---|---|
Hospital | 57.7 | 165 | <250 | <250 | +<200 |
38.5 | 240 | 850 | 850 | <200 | |
43.5 | 195 | <250 | <250 | <200 | |
53.5 | 75 | +<250 | +<250 | <200 | |
Locker Room | 62.8 | 12 | <250 | <250 | <200 |
40.1 | 82 | <250 | <250 | <200 | |
Day Nursery | 47 | 110 | +<250 | +<250 | <200 |
56.2 | 45 | 2,300 | 2,300 | <200 | |
56.5 | 65 | <250 | <250 | <200 | |
60 | 60 | <250 | <250 | <200 | |
School | 53.1 | 60 | <250 | <250 | +<200 |
Hotel | 63.5 | 90 | <250 | <250 | <200 |
63.3 | 90 | <250 | <250 | <200 | |
Hospital | 56 | 45 | +<250 | +<250 | <200 |
55.4 | 45 | <250 | <250 | +<200 | |
55.8 | 30 | <250 | <250 | <200 | |
54.1 | 45 | <250 | <250 | <200 | |
40.8 | 30 | <250 | <250 | <200 | |
53.2 | 60 | <250 | <250 | <200 | |
Kindergarten | 43.3 | 15 | <250 | <250 | <200 |
57.8 | 65 | <250 | <250 | <200 | |
62 | 15 | <250 | <250 | <200 | |
School | 38.4 | 360 | <250 | <250 | 1,000 |
54.7 | 45 | <250 | <250 | <200 | |
57.7 | 55 | <250 | <250 | <200 | |
53.8 | 60 | <250 | <250 | <200 | |
34.5 | 45 | <250 | <250 | <200 | |
50.8 | 30 | <250 | <250 | <200 | |
53.8 | 30 | <250 | <250 | <200 | |
39.5 | 25 | <250 | <250 | <200 | |
61.6 | 20 | <250 | <250 | <200 | |
62 | 30 | <250 | <250 | 274 | |
44.6 | 30 | <250 | <250 | <200 | |
Retirement Home | 53.9 | 45 | <250 | <250 | <200 |
55.6 | 30 | <250 | <250 | <200 | |
57.4 | 25 | <250 | <250 | <200 | |
Locker Room | 40.4 | 120 | <250 | <250 | <200 |
38 | 95 | <250 | <250 | <200 | |
School | 64.8 | 60 | <250 | <250 | <200 |
65.3 | 60 | <250 | <250 | 269 | |
Locker Room | 52.8 | 60 | <250 | <250 | <200 |
Locker Room | 32.7 | 120 | <250 | <250 | <200 |
Locker Room | 56.2 | 45 | <250 | <250 | <200 |
57.2 | 75 | <250 | <250 | <200 | |
50.5 | 60 | <250 | <250 | <200 | |
62.3 | 45 | <250 | <250 | <200 | |
Camping | 40.2 | 65 | <250 | +<250 | <200 |
46.6 | 55 | <250 | <250 | +<200 | |
School | 44.8 | 90 | <250 | <250 | <200 |
53.5 | 75 | <250 | <250 | +<200 | |
44.8 | 50 | <250 | <250 | +<200 | |
53.5 | 50 | <250 | <250 | <200 | |
Safe Houses | 57 | 30 | <250 | <250 | +<200 |
42.2 | 25 | <250 | <250 | +<200 | |
Safe Houses | 28 | 150 | <250 | <250 | <200 |
52.6 | 25 | <250 | <250 | +<200 | |
Safe Houses | 45 | 20 | <250 | <250 | <200 |
52.4 | 80 | <250 | <250 | <200 | |
52 | 40 | <250 | <250 | <200 |
Building . | T (°C) . | time (s) . | L.pn cfu/L . | L. spp cfu/L . | W. c GU/L . |
---|---|---|---|---|---|
Hospital | 57.7 | 165 | <250 | <250 | +<200 |
38.5 | 240 | 850 | 850 | <200 | |
43.5 | 195 | <250 | <250 | <200 | |
53.5 | 75 | +<250 | +<250 | <200 | |
Locker Room | 62.8 | 12 | <250 | <250 | <200 |
40.1 | 82 | <250 | <250 | <200 | |
Day Nursery | 47 | 110 | +<250 | +<250 | <200 |
56.2 | 45 | 2,300 | 2,300 | <200 | |
56.5 | 65 | <250 | <250 | <200 | |
60 | 60 | <250 | <250 | <200 | |
School | 53.1 | 60 | <250 | <250 | +<200 |
Hotel | 63.5 | 90 | <250 | <250 | <200 |
63.3 | 90 | <250 | <250 | <200 | |
Hospital | 56 | 45 | +<250 | +<250 | <200 |
55.4 | 45 | <250 | <250 | +<200 | |
55.8 | 30 | <250 | <250 | <200 | |
54.1 | 45 | <250 | <250 | <200 | |
40.8 | 30 | <250 | <250 | <200 | |
53.2 | 60 | <250 | <250 | <200 | |
Kindergarten | 43.3 | 15 | <250 | <250 | <200 |
57.8 | 65 | <250 | <250 | <200 | |
62 | 15 | <250 | <250 | <200 | |
School | 38.4 | 360 | <250 | <250 | 1,000 |
54.7 | 45 | <250 | <250 | <200 | |
57.7 | 55 | <250 | <250 | <200 | |
53.8 | 60 | <250 | <250 | <200 | |
34.5 | 45 | <250 | <250 | <200 | |
50.8 | 30 | <250 | <250 | <200 | |
53.8 | 30 | <250 | <250 | <200 | |
39.5 | 25 | <250 | <250 | <200 | |
61.6 | 20 | <250 | <250 | <200 | |
62 | 30 | <250 | <250 | 274 | |
44.6 | 30 | <250 | <250 | <200 | |
Retirement Home | 53.9 | 45 | <250 | <250 | <200 |
55.6 | 30 | <250 | <250 | <200 | |
57.4 | 25 | <250 | <250 | <200 | |
Locker Room | 40.4 | 120 | <250 | <250 | <200 |
38 | 95 | <250 | <250 | <200 | |
School | 64.8 | 60 | <250 | <250 | <200 |
65.3 | 60 | <250 | <250 | 269 | |
Locker Room | 52.8 | 60 | <250 | <250 | <200 |
Locker Room | 32.7 | 120 | <250 | <250 | <200 |
Locker Room | 56.2 | 45 | <250 | <250 | <200 |
57.2 | 75 | <250 | <250 | <200 | |
50.5 | 60 | <250 | <250 | <200 | |
62.3 | 45 | <250 | <250 | <200 | |
Camping | 40.2 | 65 | <250 | +<250 | <200 |
46.6 | 55 | <250 | <250 | +<200 | |
School | 44.8 | 90 | <250 | <250 | <200 |
53.5 | 75 | <250 | <250 | +<200 | |
44.8 | 50 | <250 | <250 | +<200 | |
53.5 | 50 | <250 | <250 | <200 | |
Safe Houses | 57 | 30 | <250 | <250 | +<200 |
42.2 | 25 | <250 | <250 | +<200 | |
Safe Houses | 28 | 150 | <250 | <250 | <200 |
52.6 | 25 | <250 | <250 | +<200 | |
Safe Houses | 45 | 20 | <250 | <250 | <200 |
52.4 | 80 | <250 | <250 | <200 | |
52 | 40 | <250 | <250 | <200 |
Note: Time indicates the water flowing previous to sampling. Legionella pneumophila (L.pn), Legionella spp. non pneumophila (L.spp), Waddlia chondrophila (W.c). Negative results (no detection) showed the quantification limit of the method. The symbol +, indicates qualitative positive detection at low levels, below the quantification limit. Positive results are marked in bold.
Microbiological results of Legionellae/W. chondrophila presence in hot water networks
. | W. chondrophila . | |||
---|---|---|---|---|
+ | – | |||
Legionellae | + | 3 | 1 | 4 |
– | 6 | 9 | 14 | |
9 | 10 | 19 |
. | W. chondrophila . | |||
---|---|---|---|---|
+ | – | |||
Legionellae | + | 3 | 1 | 4 |
– | 6 | 9 | 14 | |
9 | 10 | 19 |
Note: + detected; – not detected.
Current knowledge about the prevalence of W. chondrophila in the environment is limited and a previous survey carried out in Spain was able to detect W. chondrophila in well water but not in domestic drinking water (Codony et al. 2012b). Now, this work demonstrates, in a different geographic area (France), that hot water systems from non-domestic networks can be colonized by this pathogen. It is well known that these types of systems, with low levels or absence of disinfectant, can easily support the proliferation of FLA and their endosymbionts (i.e. Legionellae). On the other hand, the previous data suggest, at least in Spain, that domestic drinking water systems are not colonized by W. chondrophila (Codony et al. 2012b). Many domestic water networks have a simple structure, usually without water recirculation, which allows a minimum disinfectant level to be maintained. Maybe for this reason no pathogens were detected in those samples. These observations are in agreement with one previous epidemiological survey, conducted in Spain, that does not suggest domestic networks as the main potential infection sources for sporadic cases of legionellosis (Codony et al. 2002).
The data from this work demonstrate that W. chondrophila can colonize hot water systems from non-domestic networks with higher prevalence than Legionellae. However, this trend needs to be confirmed and more prospective studies are needed in different geographical areas and with more samples. Similarly, with the limited number of positive samples detected here, the existence of an antagonistic relationship between Legionellae and W. chondrophila cannot be suggested.
More clinical and epidemiological information is available on W. chondrophila than environmental or ecological data. For this reason, the evaluation of environmental niches, as reservoirs of pathogens, can aid the actual understanding of potential infection sources. Similar to other members of the Chlamydiales order, W. chondrophila may use FLA for its proliferation and it is not surprising to find it in artificial water systems, such as hot water networks. In these systems, the low levels of residual chlorine, water recirculation, and the existence of dead legs can promote FLA growth and that of their endosymbionts.
CONCLUSIONS
Although the actual human infection pathway remains unknown, this is the first work demonstrating the existence of this pathogen in drinking water networks. Future epidemiological studies should take into account these results in order to evaluate the potential infection risk caused by FLA endosymbionts in hot water from non-domestic buildings.
ACKNOWLEDGEMENTS
We are most grateful to Gilbert Greub (Infectious Diseases Group of the Institute of Microbiology at the University of Lausanne) for kindly providing the purified DNA from W. chondrophila. We also thank the Polytechnic University of Catalonia for supporting this study. Financial support was provided by a grant to Gemma Agustí from this University (Convocatòria d'Ajuts per a la Iniciació i Reincorporació a la Recerca). The authors thank Dr I. Douterelo from the Department of Civil and Structural Engineering, University of Sheffield, UK, for the kind review of this manuscript.