Original scientific paper ■
Received: 12. 2. 2012. Accepted: 16. 5. 2012.
1 University of Rijeka School of Medicine Department of microbiology and parasitology Brace Branchetta 20 51000 Rijeka, Croatia
* Coresponding author Mia Potočnjak, univ. bacc. sanit. ing. University of Rijeka School of Medicine Department of microbiology and parasitology Brace Branchetta 20 51000 Rijeka, Croatia Phone: +3850915742873 E-mail: miapotocnjak@gmail.com
© Inštitut za sanitarno inženirstvo, 2012.
The survival of Legionella in rainwater
Mia POTOCNJAK1*, Matija SIROKA1, Danica REBIC1, Ivana GOBIN1
ABSTRACT
Legionella is an environmental bacteria and a significant worldwide public health problem. These bacteria can survive a long period of time in distilled water, and more than a year in tap water. Due to these characteristics we were interested to explore how long it will survive in rainwater and if heavy bacterial contamination or bacterial microbiota would affect it's survival. Rainwater is a good alternative source of water in households and is used for watering gardens, washing yards and as bathing water. The rainwater from ten different tanks was analyzed and the results showed that in 90 % of samples the number of microorganisms is elevated. Two rainwater samples were chosen according to the amount of microbiota and survival of L. pneumophila and L. longbeachae was monitored. The results showed that both types of Legionella have been able to survive for 30 days in the rainwater samples but haven't been able to replicate. The rainwater microbiota restricted the survival of both Legionella strains although the effect was more pronounced in a case of L. longbeachae. These results indicate that both Legionella species can survive in the rainwater environment which means the rainwater could represent a reservoir for Legionella.
Key words: Legionella, rainwater, survival
Although L. pneumophila is responsible for the vast majority of legionellosis worldwide some of non-pneumophila species like L. longbeachae are also capable of causing disease.
INTRODUCTION
The genus Legionella belonging to the family Legionellaceae currently has at least 52 species comprising more than 70 distinct serogroups. Most Legionella species are found in aquatic environments, the exception is L. longbeachae, the only soil-dwelling pathogenic Legionella species [1]. Legionella are ubiquitous in natural aquatic environments and are capable to survive in waters with varied temperatures, pH levels, nutrient and oxygen contents. Also, Legionella species have the ability to successfully colonize man-made water handling and storage systems, which often provide ideal conditions of nutrition and temperature for their proliferation. Their widespread survival in water can be attributed to their relationships with other microorganisms. Symbiotic existence with algae and other bacteria, particularly biofilms, increases the availability of nutrients. They are also able to infect protozoas and subsequently reproduce within these organisms.
Although L. pneumophila is responsible for the vast majority of legionellosis worldwide some of non-pneumophila species like L. longbeachae are also capable of causing disease [1]. L. longbeachae is responsible for up to half of legionellosis in Australia and Thailand while in Europe and USA it causes sporadic cases [2].
Rainwater is traditionally collected in Primorsko-goranska County in water tanks, which are, by the appearance of water supply, increasingly used as an alternative source rather than primary sources of drinking water [3]. Previous study of microbiological quality of rainwater showed that 66 % of samples have heavy bacterial contamination while 44 % of samples had fecal contamination. The lack of cleaning and maintenance were the main reasons for that condition.
The objective of the present study was to test if the presence of heavy bacterial contamination in rainwater could create ecosystem favorable to Legionella survival and multiplication. The survival of two types of Legionella, L. pneumophila and L. longbeachae in two samples of rainwater were examined.
MATERIALS AND METHODS
Bacterial inoculum
Two types of Legionella were used in the experiment: Legionella long-beache NSW150, Legionella pneumophila strains Corby and 130b. Bacteria were stored at -80oC in glycerol broth (10 % glycerol). Legionella were used directly from the buffered charcoal yeast extract agar plates (BCYE-a, Oxoid). The number of bacteria in the inoculum was measured spectrophotometically (600 nm), OD (Optical density) was set to 1 what corresponded to a concentration of 1.0 x 109CFU (Colony Forming Units). The number of viable bacteria in the inoculums was confirmed by enumeration of the bacteria on BCYE agar. Bacteria cells were transferred into 10 mL of filter-sterilized and non-sterilized rainwater samples to a final concentration of approximately 1.0 x 106 CFU.
Rainwater samples
Ten samples of rain water (Table 1) were collected from water tanks from different region of Primorsko-goranska County. To test the presence of bacterial microbiota in rainwater samples the total aerobic bacteria at 22 °C, 37 °C and 44 °C on Peptone yeast extract glucose agar were determined. To test the presence of indicators of fecal contaminations, total coliform bacteria on Les Endo agar; fecal coliform bacteria on mFC agar; intestinal enterococci on KEA agar; sulphite-reducing clo-stridia on Sulfite agar were detected by membrane filtration. For survival studies two rainwater samples were selected according to the presence of bacterial contamination and analysis were conducted in non-sterile and sterile samples. The rainwater samples were sterilized by repeated filtration through a 0.2-jum-pore-size filter.
Table 1.
Microbiological quality of 10 rainwater samples.
Samples	Fecal coliform	Fecal enterococci	Total coliform	Total aerobic bacteria on 22 °C	Total aerobic bacteria on 37 °C	Total aerobic bacteria on 44 °C
	cfu/100 mL			cfu/mL		
1	0	0	57	96	164	6
2	0	0	69	100	40	2
3	5	11	12	259	140	0
4	11	17	Uncountable	1200	992	27
5	0	0	1	23	55	0
6	0	1	0	106	260	0
7	0	0	0	0	0	0
8	0	0	14	0	53	0
9	0	0	13	6	1	0
10	0	0	12	1	5	0
The limit for FC, FE, TC were 0 CFU/100 mL; TAB on 22 °C 100 cfu/mL; TAB on 37 °C 20 cfu/mL
Quantification of viable Legionella in water samples At various time points after inoculation, a 100-juL aliquot was asepti-cally removed from each of the 10-mL experimental cultures. All plate counts were done in duplicate using BCYE agar plates with antibiotic (Vancomycin (0.001 g/L) and Colistin (0.0015 g/L). Results were expressed as the number of CFU per mL.	
Statistical analysis of data All data points in each experiment were obtained in triplicate, and statistical calculations were made with GraphPad Prism version 4.00 for Windows. P values were calculated using the unpaired t test, and results were considered significant when P < 0.05.	
RESULTS AND DISCUSSION	
Rainwater is collected all around the world as good alternative source of water. It is collected by gutters on the roof and stored in appropriate water tanks [4,5]. Factors that contribute to the bacterial contamination	
International Journal of Sanitary Engineering Research	Vol. 6 □ No. 1/2012
As for many other purposes rainwater is often used for watering gardens during which aerosols could be created and that presents potential health risks.
Since the water is a natural habitat of L. pneumophila two different strains of this species: L. pneumophila strain 130B and strain Corby were tested.
of rainwater are inadequate cleaning and maintenance of tanks, the stagnation of water in the tanks and the water temperature in the hot parts of the year [5].
In this study the microbiological quality of rainwater from 10 different rainwater tanks in Primorsko-goranska County were analyzed (Table 1). Our results showed that in 70 % rainwater samples the number of bacteria cultivated at 37 °C was elevated. In three samples the fecal contamination was detected while in 8 samples total coliform bacteria were detected.
This result is consistent with previous results which showed that due to inadequate cleaning and maintenance of rainwater tanks the heavy bacterial contaminations of rainwater is common. For example, 41 % of rainwater samples tested by Sazakli et al. in Kefalonia, Greece, were positive for E. coli [6]. In studies from 2006, in Brazil 50 % of the samples fecal contamination was detected and 98 % of samples had elevated fecal enterococci. The recent study in Primorsko-goranska County also showed that 73 % of the rainwater samples had heavy bacterial contamination [10].
As for many other purposes rainwater is often used for watering gardens during which aerosols could be created and that presents potential health risks. It is well known that human inhalation of contaminated aerosols leads to Legionella infections and disease outbreaks. L. pneumophila was linked to Legionnaires' disease, respectively, amongst users of the rainwater as a potable water supply in the US Virgin Islands [7]. The investigation from New Zealand showed that roof-collected rainwater systems in a temperate climate can provide a suitable reservoir for the survival and proliferation of Legionella and may have subsequently caused cases of Legionnaires' disease [5].
So, the purpose of this study was to determine whether L. pneumophila and L. longbeachae are able to survive and multiply in rainwater and if the presence of bacterial contamination has an impact on their survival.
Two rainwater samples were chosen for survival study according to the presence of bacterial contamination (Table 1). Rainwater from sample 4 showed heavy bacterial contamination with high number of fecal colif-orms and fecal enterococci, while in sample 5 small amount of bacterial microbiota were present and indicator of fecal contamination were not detected.
Since the water is a natural habitat of L. pneumophila two different strains of this species: L. pneumophila strain 130B and strain Corby were tested. The starting dose of bacteria was 106 CFU/ml, and the samples were incubated at room temperature. The results showed that after the inoculation the number of bacteria was constant for 5 days. After that the decline in the number of bacteria occurs in unfiltered samples and after 30 day of incubation number of bacteria decreased for 3 logarithmic units. In filtered samples there were no significant changes in bacterial number during 30 days. There were no differences in survival between two L. pneumophila strains and comparison be-
tween unfiltered samples showed no difference in the survival of L. pneumophila strain 130b (Figure 1 A), while Corby strains survived better in the 4th sample (Figure 1 B).
Although the natural environment of L. longbeachae is moist soil, it is not yet known exactly how these bacteria come into the ground and what is the exact route of transmission [8,9]. Therefore, the survival of L. longbeachae was tested in the filtered and unfiltered samples of rainwater. The results showed that Legionella also survived all 30 days in rainwater samples. The number of bacteria in filtered water samples is constant throughout the study period, while in the unfiltered samples the number of bacteria was significantly reduced after the second day of incubation. On the 30'h day of incubation 102 cfu/ mL L. longbeachae
A
B
C
sample 4 sample 5 sample 4F sample 5F
sample 4 sample 5 sample 4F sample 5F
sample 4 sample 5 sample 4F sample 5F
days
The results showed that Legionella also survived all 30 days in rainwater samples.
Figure 1.
Survival of L. pneumophila strain 130b (A), strain Corby (B) and L. longbeachae strain NSW150 (C) in two rainwater samples. Plate counts were done at various time points to determine viable cells remaining. The data are the means of duplicate experiments and the error bars indicate standard deviations.
To conclude, both Legionella strains were not able to multiply but they did survive in rainwater for 30 day.
was detected (Figure 1 C). The results showed that heavy bacterial contamination of rainwater restricted the survival of L. longbeachae. It seems that this Legionella species is not adapted to the rainwater microbiota. This hypothesis was recently confirmed and analysis of the L. longbeachae genome has demonstrated that L. longbeachae is highly adapted to the soil environment [11, 12].
To conclude, both Legionella strains were not able to multiply, but they did survive in rainwater for 30 day. L. pneumophila survives better in rainwater with a complex bacterial flora in comparison to L. longbeachae. All tested bacteria survive better in filtered samples so the presence of bacterial contamination affects their survival in rainwater. Further research into the ecology of Legionella species in rainwater systems is needed.
REFERENCES
[1]	Gobin I, et al. Infections caused by nonpneumophila species of Legionella. Reviews in Medical Microbiology. 2009; 10: 1-11.
[2]	Gobin I, et al. Experimental Legionella longbeachae infection in intratra-cheally inoculated mice. 2009. J Med Microbiol. 2009; 58: 723-730.
[3]	Kovačevic N, et al. Microbiological analysis of rainwater in Primor-skogoranska County. Proceeding of 5th Croatianwater. 2011. May 18-21; Opatija. Zagreb: Hrvatske vode. 177-183.
[4]	Lye DJ. Microbiology of rain water cistern systems: a review. Journal of Environmental Science and Health. 1992; 27: 2123-2166.
[5]	Simmons G, et al. Legionnaires' disease outbreak: A water blaster and roof-collected rainwater systems. Water res. 2008; 42: 1449-1458.
[6]	Sazakli E, Alexopoulos A and Leotsinidis M. Rain water harvesting, quality assessment and utilization in Kefalonia, Greece. Water Res. 2007; 41: 2039-2047.
[7]	Schlech WF, et al. Legionnaires'disease in the Carribean. An outbreak associated with a resort hotel. Archives of Internal Medicine. 1985; 145: 2076-2079.
[8]	O'Connor BA, Carman J, Eckert, K Tucker G, Givney R, Cameron S.Does using potting mix make you sick? Epidemiol Infect. 2007; 135: 34-39.
[9]	Steele TW, Lanser J, Sangster N. Isolation of Legionella longbeachae se-rogroup 1 from potting mixes. Appl Environ Microbiol. 1990; 56: 49-53.
[10]	May S, and Prado RTA. Experimental evaluation of rain water quality for non-potable applications in the city of Sao Paulo, Brazil. Urban Water Journal. 2006; 3:145-151.
[11]	Cazalet C, et al. Analysis of the Legionella longbeachae genome and transcriptome uncovers unique strategies to cause Legionnaires' disease. PLoS Genet. 2010; 6:e1000851.
[12]	Kozak NA, et al. Virulence factors encoded by Legionella longbeachae identified on the basis of the genome sequence analysis of clinical isolate D-4968. J Bacteriol. 2010; 192: 1030-1044.