Journal of Applied Microbiology ISSN 1364-5072

ORIGINAL ARTICLE

Biocidal efficacy of monochloramine against planktonic and biofilm-associated Naegleria fowleri cells S. Goudot1,2,3*, P. Herbelin1, L. Mathieu3,4, S. Soreau1, S. Banas2,3 and F.P.A. Jorand2,3 1 2 3 4

EDF Recherche et De
veloppement, Laboratoire National d’Hydraulique et Environnement, Chatou Cedex, France Universit
e de Lorraine, LCPME, UMR 7564 CNRS – UL, Institut Jean Barriol, Villers-l es-Nancy, France CNRS, LCPME, UMR 7564 CNRS – UL, Villers-les-Nancy, France Ecole Pratique des Hautes Etudes (EPHE), LCPME, UMR 7564 CNRS-UL, Vandoeuvre-l es-Nancy, France

Keywords Ct value, Free-living amoeba, freshwater biofilm, monochloramine, Naegleria fowleri. Correspondence Laurence Mathieu, LCPME UMR 7564 CNRS-UL, 15 avenue du Charmois, 54500 Vandoeuvre Les Nancy, France. E-mail: [email protected] Pascaline Herbelin, EDF – R&D, Laboratoire National d’Hydraulique et d’Environnement / Groupe QEE, 6 quai Watier, 78401 CHATOU Cedex, France. E-mail: [email protected] *Present address: EDF - DIN – CEIDRE, DETU Service CME - Groupe Source Froide, 2 rue Amp ere, F-93206, Saint-Denis Cedex, France 2013/1671: received 16 August 2013, revised 17 December 2013 and accepted 30 December 2013 doi:10.1111/jam.12429

Abstract Aims: Free-living amoebae (FLA) in aqueous systems are a problem for water network managers and health authorities because some are pathogenic, such as Naegleria fowleri, and they have also been reported to operate as reservoirs and vectors of several pathogenic bacteria. Therefore, to better control the occurrence of such amoebae, we evaluate the efficacy of monochloramine against planktonic forms (trophozoites and cysts) and also biofilm-associated cells of N. fowleri as FLA are often associated with biofilms. Methods and Results: From a freshwater biofilm growing in a pilot reactor and inoculated with N. fowleri, we obtained Ct values ranging from 4 to 17 mg Cl2 min l
1 at 25°C and pH 82 on both planktonic and biofilm associated cells. In addition, the inactivation pattern of biofilm associated was intermediate between those of trophozo€ıtes and cysts. Conclusions: The monochloramine efficiency varies with the life stage of N. fowleri (trophozo€ıte, cyst, and biofilm-associated). The sensitivity to disinfectant of amoeba, that is, trophozo€ıtes and cysts, in the biofilm life stage is as high as that of their planktonic cyst form. Significance and Impact of the Study: This study gives Ct values for cysts and biofilm-associated N. fowleri. This may impact on water treatment strategies against amoebae and should be considered when controlling N. fowleri in manmade water systems such as cooling towers or hot water systems.

Introduction Natural aquatic environments (rivers, lakes and springs) and man-made water systems (drinking water networks or poorly chlorinated swimming pools) are both common habitats of free-living amoebae (FLA) (Sibille et al. 1998; Thomas et al. 2004; Jamerson et al. 2009; Loret and Greub 2010; Marciano-Cabral et al. 2010; Buse et al. 2013; Garcia _ et al. 2013; Zbikowska et al. 2013). Some genera of these FLA are opportunistic or nonopportunistic pathogens capable of causing severe human diseases such as keratitis or gastroenteritis. One of the most serious diseases caused by FLA is primary amoebic meningoencephalitis, a fatal central nervous system disease. Naegleria fowleri is the

causative agent of this infection, which results from amoeba-contaminated water entering the nasal cavity (Marciano-Cabral 1988; Visvesvara et al. 2007; Kaushal et al. 2008). This infection is rare and, to date, less than 300 cases have been reported worldwide since 1965 (De Jonckheere 2011; Moussa et al. 2013; Tung et al. 2013). N. fowleri is ubiquitous in natural and man-made warm aquatic environments, such as lakes, rivers, geothermal water, swimming pools, spas and cooling systems (Jamerson et al. 2009; Huang and Hsu 2010; Stockman et al. 2011; Kao et al. 2012, 2013; Wang et al. 2012). In addition to being causative agents of infectious diseases, FLA have been reported to operate as reservoirs and vectors by promoting the survival and multiplication

Journal of Applied Microbiology 116, 1055--1065 © 2014 The Society for Applied Microbiology

1055

Monochloramination of N. fowleri

S. Goudot et al.

of infectious bacteria such as Legionellaceae, Mycobacteriaceae, Enterobacteriacaeae and Vibrionaceae, as well as of some toxigenic cyanobacteria (Corsaro et al. 2010; Marciano-Cabral et al. 2010; Garcia et al. 2013). Taking into account both their pathogenic properties and their interactions with pathogenic bacteria in aqueous environments, controlling amoebae in water is clearly a public health concern (Thomas and Ashbolt 2010). Disinfection is the main practice for controlling the wide variety of pathogenic micro-organisms and reducing the level of microbiological contaminants transmitted by waters (recreational, drinking or thermal). Despite various data on sensitivity and resistance of FLA to biocides (Thomas 2012) and increasing health concerns over FLA, there is still a lack of information on the mechanisms of action and efficacy of biocides on amoebae in real systems. A few studies have investigated the effects of chlorine on Acanthamoeba spp. trophozo€ıtes (Cursons et al. 1980; Critchley and Bentham 2009) and Acanthamoeba cysts (De Jonckheere and Voorde 1977; Thomas et al. 2004). Low doses of chlorine – representative of drinking water disinfection practices – proved to be ineffective. For instance, while trophozo€ıtes of Acanthamoeba castellanii exposed to 5 mg Cl2 l
1 exhibited a size reduction, cellular damage and a 999% decrease in cultivability after 30 s of exposure at 25°C and pH 7 (Mogoa et al. 2010), hyperchlorination of Acanthamoeba spp. cysts with chlorine concentrations as high as 50 or 100 mg l
1, for 18 h or 10 min, respectively, remained ineffective (Kilvington and Price 1990; Storey et al. 2004). However, such concentrations are more effective against Hartmannella or Naegleria suggesting discrepancies in the oxidant susceptibility according to the genus and even to the species of amoeba (Coulon et al. 2010; Thomas 2012; Wang et al. 2012; Dupuy et al. 2013). Chloramine is another halogen compound widely used for water disinfection. Although less reactive than chlorine, it has the advantage of not forming regulated disinfection by-products such as trihalomethanes. It also appears to diffuse better through the polymeric matrix of biofilms than other chlorine disinfectants (LeChevallier et al. 1988; Tachikawa et al. 2005). However, as in the case of chlorine, few studies have explored the inhibitory efficacy of monochloramine on amoebae. Moreover, all such studies were performed in a buffered liquid medium on a pure culture of amoeba species (Ercken et al. 2003; Dupuy et al. 2011; Mogoa et al. 2011), not representative of natural ecology of the protozoa. As bacteria are their main nutrient source, FLA are mainly found on or near surfaces, where they graze on biofilm bacteria. It has been postulated that biofilm could also provide physical and chemical protection for FLA against predators and disinfectants (Barbeau and Buhler 2001; Thomas et al. 2004). Biofilms are therefore 1056

considered a major reservoir of FLA (Parry 2004; Huws et al. 2005; Pickup et al. 2007; Puzon et al. 2009; Goudot et al. 2012). However, only very few studies have examined the disinfection efficacy of oxidants on amoebic communities within biofilms (Thomas et al. 2004; Loret et al. 2005). They both demonstrated that a continuous monochloramine treatment of 05 mg l
1 over several weeks was ineffective against the amoebic community (all species) in both water and biofilm. Since 1990, French power stations have been monitoring their cooling systems for the presence of N. fowleri as their cooling waters are released into rivers. To protect river users, particularly during recreational activities, and to reduce health risks downstream from power stations, chemical or physical treatments are implemented in several cooling water systems to control and inactivate pathogens in the water. In France, several cooling water systems are currently treated with monochloramine to prevent microbiological risks. In this context, the main objective of this study was to evaluate the effects of monochloramine against N. fowleri in its three different stages of life cycle: trophozo€ıtes, cysts and biofilm-associated amoebae. To get the latter form, we used a biofilm reactor which allowed freshwater biofilms to develop from raw river waters and experimentally introduced pathogenic amoebae to colonize (Goudot et al. 2012). Monochloramination treatments were performed to (i) assess the efficacy of this oxidant both on planktonic cysts and on trophozoites and (ii) compare the monochloramine sensitivity of the N. fowleri planktonic forms (trophozo€ıtes and cysts) with that of biofilm-associated amoebae. While there is no standardized method available for testing the efficacy of disinfectants on amoebae, we used the model defined by Watson and Chick to determine the Ct99% (monochloramine concentration 9 contact time leading to 99% inactivation of the amoebic population) (Chick 1908; Watson 1908). Ct99% values were assessed for inactivation of the planktonic form of N. fowleri in both life stages – cysts and trophozo€ıtes. Ct99% values were also assessed for the inactivation of the biofilm-associated N. fowleri. Monochloramine has a biocidal effect on planktonic and biofilm-associated forms of N. fowleri, and its efficacy appears to depend both on the intrinsic resistance of the amoebae (cyst form) and the surrounding environment (water or biofilm). Materials and methods Naegleria fowleri strain, culture conditions and inoculum preparation The AMI005 strain of N. fowleri (EDF internal collection, LNHE, Chatou, France) was isolated from the cooling

Journal of Applied Microbiology 116, 1055--1065 © 2014 The Society for Applied Microbiology

S. Goudot et al.

water of a power station. It was grown (for 3–5 days) at 43°C on non-nutrient agar (NNA, Indicia Biotechnology, Oullins, France) previously overlaid with an Escherichia coli suspension and identified by an enzyme-linked immunosorbent assay (Indicia Biotechnology) using monoclonal antibody 5D12 (Pougnard et al. 2002). N. fowleri trophozoite and cyst suspensions were prepared separately. Trophozoites were harvested, under microscope examination, after a 2-day culture on E. coli mats against a 5-day culture for cysts (Figure S1). Suspensions of N. fowleri trophozo€ıtes or cysts were prepared by gently scraping the amoebic migration front or encystments, respectively, of ten plates and poured into 5 ml phosphate buffer saline (PBS) for further use. Naegleria fowleri and thermophilic FLA cell counting Thermophilic FLA, including N. fowleri, were counted using the most probable number (MPN) approach described by Pougnard et al. (2002). The MPN approach determines the concentration of viable N. fowleri – both trophozoites and cysts – without allowing the two forms to be discriminated. Briefly, immediately after sample collection (river water or biofilm after desorption from the support), five 1 ml replicate subsamples of each tenfold serial dilutions were spread onto NNA plates previously overlaid with E. coli. The plates were incubated at 43°C, and the presence of an amoebic migration front was assessed daily for 5 days by microscopic examination. Naegleria-positive samples were further analysed to determine the presence of flagella by incubating vegetative forms in demineralized water at 37°C for 4 h. Finally, N. fowleri were identified using an enzyme-linked immunosorbent assay (Indicia Biotechnology) with monoclonal antibody 5D12 as previously described by Pougnard et al. (2002). Moreover, morphological examinations of the amoebae under microscope allowed identification of thermophilic FLA other than N. fowleri as recently recommended by De Jonckheere et al. (2012). Preparation of monochloramine A monochloramine stock solution was prepared by mixing under agitation a sodium hypochlorite solution (152 g l
1, ACROS Organics) in an ammonia solution (30%, ACROS Organics) at a Cl2/N mass ratio of 48 and at a pH of 83. Under these stoichiometric conditions, the theoretical concentration of monochloramine stock solution was about 1000 mg Cl2 l
1. Monochloramine solution was prepared daily and used extemporaneously. Its concentration was determined by the DPD method using Hach Methods 8167 on a DR/2500 spectrophotometer (Hach Company, Loveland, CO) at 530 nm.

Monochloramination of N. fowleri

Disinfection assays on planktonic Naegleria fowleri Monochloramine disinfection on planktonic N. fowleri was performed separately on trophozoite or cyst suspensions under batch conditions. Two sets of assays were performed: three independent assays (named T1 to T3) were dedicated to the disinfection of the trophozoite form of N. fowleri whereas three other independent assays (named C1 to C3) were assigned to the disinfection of the cyst form. Each monochloramine treatment was performed on 150 ml autoclaved freshwater (the same as used to supply the reactor) (pH 82) inoculated with N. fowleri suspension containing only trophozoites or only cysts to achieve final concentrations of around 3 9 104 to 8 9 104 amoebae l
1 depending on the assays. The freshwater had previously been autoclaved to remove any naturally present amoebae and to also retain the physicochemical characteristics of the freshwater close to those in the reactor. A volume of the monochloramine stock solution was then added to obtain a theoretical final concentration of 1 mg Cl2 l
1. The concentration of monochloramine and the survival of N. fowleri trophozoites (assays T1 to T3) or cysts (assays C1 to C3) were regularly monitored for 60 min at 25°C under agitation (magnetic stirrer). Sterile sodium thiosulfate 01 mol l
1 was added in excess in each treatment flask for neutralization of monochloramine residuals. For each experiment, control flasks without addition of monochloramine were performed in parallel and sampled at the beginning and end of the assay. Disinfection assays on biofilm-associated Naegleria fowleri Freshwater biofilm formation and Naegleria fowleri inoculation set-up A flat-plate open channel reactor previously described by Goudot et al. (2012) was operated in continuous flow mode (Figure S2). The inlet flow and the recycle flow rate were maintained at 19 and 810 ml min
1, respectively. The hydraulic retention time was 24 h. The flow presented a laminar velocity profile in the length direction characterized by a shear rate of 17 s
1. The reactor was fed with freshwater (Loire River, France), collected in June 2011 and stored in the dark in an agitated, refrigerated (4°C) tank for the duration of the experiments. Microbial and physico-chemical characteristics of the inlet water are presented in Table 1. Biofilm colonization was carried out at 42°C on glass coupons (~22 cm²) placed for at least 8–10 days in the freshwater reactor. Averages of microbial and physico-chemical characteristics of the biofilm are also presented in Table 1. As previously described, a suspension of N. fowleri in

Journal of Applied Microbiology 116, 1055--1065 © 2014 The Society for Applied Microbiology

1057

Monochloramination of N. fowleri

S. Goudot et al.

Table 1 Microbial and physico-chemical characteristics of the river water (Loire, collected in June 2011) at the inlet of the reactor, and of the biofilm growing within. For the biofilm, the value in brackets is the standard deviation of three independent measures. FLA stands for Free-Living Amoebae Parameters

Inlet water

Biofilm

Thermophilic FLA (cells l
1 or cells cm
2) N. fowleri (cells l
1 or cells cm
2) Bacteria (cells l
1 or cells cm
2) pH Conductivity (lS cm
1) DO (mg l
1)† TOC (mg l
1 or mg cm
2)‡ DOC (mg l
1)§ DSS (mg l
1 or mg cm
2)¶