Biocontrol Science, 2019, Vol. 24, No. 4, 213—220

Original Investigations on Contamination of Environmental Water Samples by using Real-Time Quantitative PCR Combined with Amoebic Co-Culturing

AKIKO EDAGAWA1＊, AKIO KIMURA1, AND HIROSHI MIYAMOTO2

1Department of Environment Health, Osaka Prefectural Institute of Public Health, Osaka, 1-3-69 Nakamichi, Higashinari-ku, Osaka 537-0025, Japan 2Department of Pathology and Microbiology Faculty of Medicine, Saga University. Japan, 5-1-1 Nabeshima, Saga, 849-8501, Japan.

Received 23 May, 2019/Accepted 8 August 2019

We analyzed the contamination of environmental water samples with Legionella spp. using a conventional culture method, real-time quantitative PCR（ qPCR）, and real-time qPCR combined with an amoebic co-culture method. Samples（ n = 110） were collected from 19 cooling towers, 31 amenity water facilities, and 60 river water sources of tap water in Japan. Legionella was detected in only three samples（ 3/110, 2.7%） using the culture method. The rate of Legionella detection using amoebic co-culture followed by qPCR was 74.5%, while that using qPCR without amoebic co-culture was 75.5%. A higher than 10-fold bacterial count was observed in 19 samples（ 19/110, 17.3%） using real-time qPCR subsequent to amoebic co-culture, compared with identical samples analyzed without co-culture. Of these 19 samples, 13 were identified as Legionella spp., including L. pneumophila and L. anisa, and the non-culturable species were identified as L. lytica and L. rowbothamii. This study showed that the detection of Legionella spp., even in those samples where they were not detected by the culture method, was possible using real-time qPCR and an amoebic co-culture method. In addition, this analytical test combination is a useful tool to detect viable and virulent Legionella spp..

Key words : Legionella species / Amoebic co-culture / Acanthamoeba castellanii / Environmental water.

INTRODUCTION in environmental waters, especially during a disease outbreak, is important for public health. Legionella species are Gram-negative that are The culture method is routinely used to test Legionella widely present in both natural and artificial water bodies spp. in environmental samples（ Totaro et al., 2017; （Fields et al., 2002）. Infection with Legionella spp. Furuhata et al., 2013）. However, this method has certain leads to clinical manifestations, including Legionnaires’ limitations associated with the culture characteristics disease and Pontiac fever. Legionella spp. proliferates in of Legionella spp. These bacteria exhibit a slow growth aquatic environments within free-living protozoal hosts rate, and growth can be inhibited in the presence of （Greub et al., 2004）. Infection is passed on to humans other microorganisms, thereby transforming the culture via inhalation, aspiration, or micro-aspiration of legionel- into a viable but non-culturable（ VBNC） form under la-carrying aerosols. Legionnaires’ disease is prevented certain environmental conditions. Recently, quanti- mainly by monitoring and controlling the contamina- tative real-time polymerase chain reaction（ real-time tion of environmental waters with Legionella spp. Also, qPCR） specific for Legionella 16S rRNA or macrophage accurately identifying the source of Legionella infection infectivity potentiator genes have been widely applied to overcome the limitations of standard culture meth- ＊Corresponding author. Tel & Fax: +81-6-6972-1321, E-mail : ods.（ Touron-Bodilis et al., 2011; Bonetta et al., 2010; edagawa（a）iph.osaka.jp Edagawa et al., 2008）. Real-time qPCR can detect 214 A. EDAGAWA ET AL.

VBNC Legionella types; however, it cannot distinguish Ltd.）. After culture for 3 days at 37°C, the isolates that between viable and dead cells. Further, although real- grew on BCYE-alpha but not on blood agar were exam- time qPCR is significantly more sensitive in detecting ined by Gram staining. A final Gram-negative staining lower levels of contamination than the culture method, it was considered suggestive of the presence of Legionella cannot detect low levels of Legionella contamination in a spp. Such colonies were observed under UV light and natural environment. 1–5 colonies were next randomly selected for a latex Recently, the amoebic co-culture technique has been agglutination test（ Kanto Chemical Co., Tokyo, Japan） used for to detect bacterial species from clinical and and an immune serum agglutination test（ Denka Seiken environment samples（ Inoue et al., 2019; Edagawa Co. Ltd., Tokyo, Japan） to identify the serogroups of et al., 2015; La Scola et al., 2001）. Acanthamoeba, a L. pneumophila, L. bozemanii, L. dumoffii, L. gormanii, genus of amoeba, serves as a host for Legionella pneu- and L. micdadei.. For Legionella strains not identified by mophila. Amoebic co-culture using Acanthamoeba is these tests, Legionella spp. were identified by 16S rRNA used to detect Legionella spp., including VBNC and PCR and nucleotide sequencing as described below. Legionella-like amoebal pathogen（ LLAP） types, after Acanthamoeba castellanii strain ATCC 30234 was their co-proliferation within the amoebae. grown in 75-cm2 culture flasks at 30°C for 4 days on 50 In this study, we used the culture method, real-time mL of peptone/yeast extract/glucose（ PYGC） medium qPCR, and real-time qPCR combined with amoebic （10 g proteose peptone, 10 g yeast extract, 10.1 g co-culture to analyze the contamination of various envi- glucose, 5 g NaCl, 0.95 g L-cysteine hydrochloride, 1.74

ronmental water samples with Legionella spp. Samples g Na2HPO4, and 1.36 g KH2PO4 in 1 L of distilled water; were collected from cooling towers, water-amenity facil- the pH was adjusted to 6.8）. Cells were harvested by ities, and tap water sources in Kansai area of Japan. centrifugation and re-suspended in PYGC medium at a density of approximately 1 × 105 cells/mL. The amoebal MATERIALS AND METHODS suspension was distributed into each well of 12-well micro-plates at 30°C until the cells formed monolayers. In total, 110 water samples were collected from 19 Just before processing the water samples, the PYGC cooling towers, 31 water-amenity facilities, and 60 river medium was removed from each well, and the cells were water sources of tap water from the Osaka Prefecture washed twice with 1 mL of Neff’s amoeba saline（ 120

of the Kansai area in Japan. All samples（ 400 mL mg NaCl, 3 mg MgCl2, 3 mg CaCl2, 142 mg Na2HPO4,

each） were collected in sterile bottles containing 0.01% and 136 mg KH2PO4 in 1 L of distilled water）. sodium thiosulfate. Samples were immediately trans- A volume of 1.5 mL of the processed environmental ported to a laboratory for further processing. water sample was inoculated into the amoebal micro- Each sample was concentrated by filtration through plate wells（ amoebic co-culture）. After incubating for a 0.22-μm-pore size polycarbonate filter（ Advantec 7 days at 30°C, 1 mL of the suspension was stored at Tokyo Co. Ltd., Tokyo, Japan）. The membrane was then −20°C for DNA extraction. DNA was extracted using a immersed in 4 mL of sterile deionized water, vortexed QIAamp DNA Mini Kit（ Qiagen K.K., Tokyo, Japan） and for 1 min, and shaken vigorously 50 times. From Legionella spp. were detected using real-time qPCR. this suspension, 1 mL was stored at −20°C for DNA The supernatant was stored at −20°C until use. extraction and 1.5 mL was used for amoebic co-culture. Real-time qPCR was performed according to the A 0.2-mL aliquot of the suspension was used for acid manufacturer’s instructions as using a Cycleave PCR treatment culture method, while the remaining suspen- Legionella（ 16S rRNA） Detection Kit（ Takara Bio Inc., sion was heated in a water bath at 50°C for 30 min and Shiga, Japan）. PCRs with duplicate standards, positive used for the heat treatment culture method. and negative controls, and samples were performed A volume of 1 mL of the above sample suspension using an ABI PRISM 7900HT Real-time qPCR System was mixed with 1 ml of 0.2M acid-phosphate buffer （Applied Biosystems） according to the manufacturer’s （pH2.2）, and 0.2 mL of this acid-treated sample was instructions. The PCR mixtures（ 20 µL） were prepared inoculated after 5 min onto Wadowsky-Yee-Okuda agar according to the manufacturer’s protocol and reactions plates（ WYO-alpha plates; Eiken Chemical Co. Ltd., were performed under the following amplification condi- Tokyo, Japan）. From the heat-treated sample, 0.1 mL tions: 95°C for 10 s, then 45 cycles of 95°C for 5 s, was inoculated onto an WYO-α plate. After incuba- 55°C for 10 s, and 72°C for 25 s. DNA amplification was tion for 5–7 days at 37°C, 1–50 colonies resembling detected by monitoring the fluorescence at two wave- Legionella spp. were selected and cultured on blood lengths, viz., FAM and ROX. The amplified 16S rRNA of agar plates（ Eiken Chemical Co. Ltd.） and on buffered Legionella and the internal control gene were detected charcoal–yeast extract agar plates supplemented with by FAM and ROX, respectively. The samples whose DNA alpha-ketoglutarate（ BCYE-alpha; Eiken Chemical Co. was not amplified due to the presence of inhibitors were LEGIONELLA DETECTION BY AMOEBIC CO-CULTURE 215

TABLE 1. Positive samples of Legionella species in environmental water detecting by real-time qPCR methods with or without amoebic co-culture techniques and plate culture method Real-time qPCR Plate culture method Total with co-culture without co-culture Cooling tower 17 18 0 19 Amenity water facilities 18 22 3 31 River as source of tap water 47 43 0 60 Total 82 83 3 110

determined from internal standard ROX analysis data. More than 10-fold higher copy numbers of Legionella RESULTS AND DISCUSSION 16S rRNA gene were observed by real-time qPCR with the co-culture technique in 19 samples（ 19/110, The detection of Legionella spp. using amoebic 17.3%） compared with identical samples without co-culture qPCR, qPCR, and culture methods from co-culturing. For these samples, 16S rRNA PCR and environmental water samples are presented in Table nucleotide sequencing were performed to identify 1. Overall, 1.1 × 103 – 5.4 × 107, and 1.6 × 102 – 4.4 Legionella spp. For qualitative PCR, the Legionella × 107 Legionella spp. were detected using amoebic species-specific primers LEG-225F（ 5’ AAG ATT AGC co-culture qPCR and qPCR, respectively. When bath CTG CGT CCG AT 3’） and LEG-858R（ 5’ GTC AAC water samples were analyzed（ Edagawa et al., 2015） TTA TCG CGT TTG CT 3’） were used to amplify a , the bacterial detection rate increased from 67.6% 654-bp product of a partial 16S rRNA gene（ Miyamoto to 83.8% using the amoebic co-culture method, and et al., 1997） . Five microliters of extracted DNA was 71.4% of the samples wherein Legionella spp. were not used in a 50-μL reaction mixture containing 5 μL of 10× detected（ ND） using qPCR, tested positive. Detection

PCR buffer [100 mmol KCl, 20 mmol MgCl2, 20 mmol rates for environmental water samples were 74.5% using Tris-HCl（ pH 8.0）], 4 μL of 2.5 mmol dNTP mixture, amoebic co-culture with qPCR and 75.5% for qPCR, 0.5 μL each of 100 mmol primers, and 0.25 μL of 5 U/ which were comparable to earlier results. Of the 27 ND μL Ex Taq DNA polymerase（ Takara Bio Inc.）. Thermal samples using qPCR alone, four samples（ 4.8%） tested cycling was performed with a PCR thermal cycler positive after analysis using the amoebic co-culture （ASTEC Co.）. Cycling conditions began with an initial method. Using the culture method, various Legionella incubation at 95°C for 90 s, followed by 30 cycles of spp. were detected in three water samples collected denaturation at 95°C for 10 s, annealing at 64°C for 1 from amenity water facilities, viz., sample no. 82, L. min, and extension at 72°C for 1 min. Finally, incom- pneumophila serogroup 7 and L. anisa at 90 CFU/100 plete PCR products were extended for 10 min at 72°C. mL; sample no. 81, L. pneumophila serogroup 7 and L. Genomic DNA from L. pneumophila（ ATCC33152） was anisa at 70 CFU/100 mL; and sample no. 75, L. pneu- used as a positive control. Amplified DNA was detected mophila serogroup 8 at 10 CFU/100 mL. Legionella spp. by 1% agarose gel electrophoresis and the DNA was were ND by the culture method in samples collected visualized by ethidium bromide staining. The qualitative from cooling towers and natural river water. 16S rRNA PCR products were directly sequenced in Of all the tested samples, 87（ 79.1%, 18 samples both directions using forward and reverse primers with a from the cooling tower, 22 samples from water-amenity BigDye Terminator v1.1 Cycle Sequencing Kit（ Applied facilities, and 47 samples from river water sources of tap Biosystems）. Homology searches were performed using water） tested positive using either qPCR with amoebic the NCBI BLAST software（ http://www.ncbi.nlm.nih. co-culture or culture method（ Tables 2–4）. Using the gov）. The sequence data obtained in this study were amoebic co-culture method, 19 samples（ 15.8%, 1 aligned with Legionella sequences deposited in the sample from the cooling tower, 8 samples from amenity GenBank using the ClustalW software（ http://www.nig. water, and 10 river water samples） were confirmed to ac.jp） and a phylogenetic tree was then displayed by the contain a 10-fold higher number of bacteria compared neighbor-joining method using the Njplot program. to qPCR analysis without amoebic co-culture. Thirteen The nucleotide sequences determined in this study PCR products from the above 19 PCR-positive samples have been submitted to the DNA Data Bank Japan were successfully sequenced for species identifica- （DDBJ） database and assigned the accession numbers tion. A BLAST homology search indicated a 99–100% LC458436 to LC458448 and LC491286 to LC491288. sequence identity with L. pneumophila（ sample no. 216 A. EDAGAWA ET AL.

TABLE 2. Results of the real-time qPCR in the 18 water TABLE 3. Results of the real-time qPCR in the 22 water samples from the cooling tower. samples from amenity water facilities Real-Time qPCR Method（ copies/100mL） Real-Time qPCR Method（ copies/100mL） Sample Sample No. With Amoebic Without Amoebic No. With Amoebic Without Amoebic Co-Culture Co-Culture Co-Culture Co-Culture 13 5.1 ×106 1.3 ×107 53 3.9 ×103 8.2 ×103 14 1.4 ×103 7.8 ×103 55 1.7 ×107＊ 1.5 ×106 15 4.0 ×106 2.1 ×106 56 1.0 ×106＊ 9.6 ×103 16 1.1 ×103 2.8 ×103 71 5.4 ×107＊ 4.9 ×104 17 1.2 ×104 4.8 ×104 72 1.5 ×107＊ 3.5 ×105 18 4.4 ×104 8.7 ×104 75 1.7 ×107 2.8 ×106 19 6.5 ×104 8.6 ×105 81 1.1 ×106＊ 3.6 ×104 20 3.1 ×103 2.0 ×103 82 5.0 ×105＊ 3.0 ×104 21 7.2 ×105 1.3 ×106 83 5.9 ×105 8.7 ×105 22 9.9 ×103 3.4 ×104 85 1.9 ×106 2.2 ×106 61 9.6 ×104 5.2 ×104 86 5.0 ×103 2.2 ×104 62 7.8 ×104＊ 4.5 ×103 87 1.0 ×104 4.6 ×103 63 5.7 ×105 4.0 ×106 88 5.9 ×106＊ 8.1 ×104 65 3.8 ×104 2.7 ×104 89 3.0 ×105 8.6 ×105 66 2.2 ×105 2.4 ×105 90 2.5 ×105 1.6 ×106 67 2.1 ×105 5.1 ×106 91 3.8 ×106 4.4 ×107 69 1.1 ×106 1.7 ×106 92 3.2 ×106＊ 7.5 ×103 64 ND a） 2.4 ×103 95 2.6 ×105 4.6 ×104 Notes: ＊: Using real-time qPCR with amoebic co-culture, 54 ND a） 2.7 ×104 more than 10-fold higher bacterial numbers compared with 84 ND 5.1 ×103 the same samples without amoebic co-culture are observed; 4 a ） ND: not detected（ less than 1.0×103 copies/100mL）. 96 ND 5.5 ×10 97 ND 5.2 ×105 Notes: ＊: Using real-time qPCR with amoebic co-culture, more than 10-fold higher bacterial numbers compared with the same samples without amoebic co-culture are observed; a ） ND: not detected（ less than 1.0×103 copies/100mL）.

72, 616/616, 100%; no. 71, 616/617, 99.8%; no. L. tunisiensis, which is isolated from patients with pneu- 81 616/616, 100%; no. 88 616/616, 100% and no. monia and is pathogenic to humans（ Thacker et al., 92 616/616, 100%）, with L. anisa（ sample no. 82, 1985）. To our knowledge, these species except L. feelei （616/617, 99.8%））, and with Legionella sp.（ sample have not yet been detected in Japan, probably since no. 55,）. Overall, 98–100% concordance was achieved the culture method is still predominantly used to detect with unculturable species such as L. lytica（ sample Legionella. In this study, Legionella spp. that could not no. 35 616/616, 100%; no. 36 616/617, 99.8%）, L. be detected by the culture method could be detected rowbothamii（ LLAP6; sample no. 31, 612/616, 99.4%）, after using the amoebic co-culture method. LLAP（ sample no. 56, 608/615, 98.9%）, and L. quatei- In this study, Legionella spp. were detected not rensis（ sample no. 62, 607/617, 98.5%） and L. feeleii only in samples from artificial water bodies, such as （sample no. 32, 583/583, 100%）. A resulting molecular cooling towers and amenity points, but also in river phylogenetic tree is shown in Figure 1. L. lytica（ Hookey water samples, such as tap water sources. In Japan, et al., 1996） and L. rowbothamii（ Adeleke et al., 2001） the source of infection was identified in less than are usually ND by culture methods. L quateirensis is half of the cases, but 75% of the cases where the an environmental strain that was isolated from shower source of infection was identified were in bath water water from hotels in Portugal（ Dennis et al., 1993）. L. of hot springs and bathing facilities（ National Institute feeleii isolated from the environment is closely related to of Infectious Diseases 2000）, and there are many LEGIONELLA DETECTION BY AMOEBIC CO-CULTURE 217

TABLE 4. Results of the real-time qPCR in the 47 river water 51 7.3 ×104 3.2 ×105 samples as a source of tap water 57 6.8 ×104 4.6 ×104 Real-Time qPCR Method（ copies/100mL） Sample 58 1.3 ×105 9.7 ×104 With Amoebic Without Amoebic No. 6＊ 5 Co-Culture Co-Culture 59 1.9 ×10 1.0 ×10 5 5 1 8.9 ×103 3.8 ×103 60 1.4 ×10 1.1 ×10 4 5 8 9.0 ×103 1.6 ×102 79 3.6 ×10 1.7 ×10 3 4 10 1.1 ×104 8.1 ×103 80 8.0 ×10 1.9 ×10 4 5 11 9.7 ×103 1.7 ×103 98 5.0 ×10 2.8 ×10 4 4 12 5.3 ×103 5.5 ×103 99 3.8 ×10 4.7 ×10 5 5 26 1.6 ×104 1.0 ×105 100 3.8 ×10 2.7 ×10 4 5 28 8.1 ×103 1.3 ×104 101 4.2 ×10 3.6 ×10 4＊ 3 29 1.0 ×104 2.0 ×104 102 4.2 ×10 1.8 ×10 4 5 31 1.7 ×106＊ 3.0 ×103 103 6.8 ×10 5.1 ×10 4＊ 3 35 2.0 ×106＊ 7.2 ×103 104 9.8 ×10 2.2 ×10 3 4 36 3.2 ×105＊ 6.5 ×103 105 3.1 ×10 4.7 ×10 4 6 37 1.0 ×104 1.4 ×104 106 6.7 ×10 2.2 ×10 4 5 38 3.1 ×104 4.9 ×104 107 3.1 ×10 2.4 ×10 4 5 39 1.0 ×104 1.8 ×104 108 9.4 ×10 9.0 ×10 5 7 40 3.3 ×104 2.1 ×104 109 4.2 ×10 1.2 ×10 5 6 41 5.9 ×105 7.9 ×105 110 1.4 ×10 5.7 ×10 3＊ a) 42 2.2 ×105 8.4 ×104 2 4.0 ×10 ND 3＊ 43 2.6 ×104 2.0 ×103 4 1.6 ×10 ND 5＊ 44 9.1 ×104 1.2 ×104 32 3.3 ×10 ND 4＊ 46 5.1 ×103 3.7 ×103 47 4.7 ×10 ND ＊ 48 9.2 ×104 9.2 ×104 Notes: : Using real-time qPCR with amoebic co-culture, more than 10-fold higher bacterial numbers compared with 49 8.8 ×103 1.3 ×104 the same samples without amoebic co-culture are observed; 4 4 50 1.7 ×10 1.6 ×10 a）ND: not detected (less than 1.0×103 copies/100mL).

investigation reports on bath water（ Sasahara et al., water distribution. 2004; Karasudani et al., 2009; Furuhata et al., 2004）. To date, around 60 species of Legionella have been Countries apart from Japan have reported Legionella identified, and less than 50% of these species are contamination and resulting human infection due to tap suspected to cause infections in humans. However, water systems. In the United States, a switch from water the pathogenicity of Legionella in humans is not well sources to rivers that show the presence of Legionella defined. The amoebic co-culture method used in this has led to an increase in the number of Legionella cases study utilizes the ability of Legionella to grow in amoe- in distribution areas（ Davis et al., 2016; Zahran et al., bic hosts. L. pneumophila survives within amoebae by 2018）, which have been associated with the occur- a mechanism similar to one that enables it to survive rence of legionellosis in surrounding areas. In this study, within macrophages（ Horwitz et al., 1987; Miyamoto we observed that 78.3% of river water samples tested et al., 2003）. Once engulfed by amoebae using coil- Legionella positive by qPCR with amoebic co-culture. In ing phagocytosis, L. pneumophila invades fusion of Japan, the detection rate of amoebae including patho- the phagosome and the lysosome, and replicates genic species in river water as tap water sources was within the amoebae（ Greub et al., 2004）. This ability reported to be 68.7%（ Edagawa et al., 2009）, it is highly to invade and multiply within host cells is considered likely that Legionella spp. parasitize and proliferate within to correlate with its pathogenicity（ Neumeister et al., these amoebae. The monitoring of river waters, espe- 1997）. L. pneumophila uses the same genes to multiply cially potable sources, may be necessary to prevent the within A. castellanii and human macrophages（ Segal outbreak of Legionnaires’ disease associated with tap et al., 1999）. Additionally, the intracellular growth in 218 A. EDAGAWA ET AL.

OIPH-co81A (LC458445) OIPH-co71A (LC458443) OIPH-co88A (LC458447) 68 OIPH-co92A (LC458448) OIPH-75A* (LC491286) 79 L. pneumophila subsp. pneumophila (CP015927) 85 OIPH-co72A (LC458444) L. pneumophila subsp. pascullei (AF122885) 98 L. pneumophila subsp. fraseri (HQ287902) L. waltersii (AF122886) L. wadsworthii (Z49738) 65 OIPH-82A* (LC491288) OIPH-81A*(LC491287) 100 L. anisa (Z32635) OIPH-co82A (LC458446) L. steelei (HQ398202) L. tucsonensis (Z32644) L. bozemanii (M36031) OIPH-co55A (LC458440) L. longbeachae (AY444740) L. fallonii LLAP10 (X97363) L. worsliensis (Z49739) L. quateirensis (Z49732) 76 67 OIPH-co62C (LC458442) 100 Legionella sp. strain LLAP11 (X97362) Legionella sp. strain LLAP4 (X97357) 100 L. drancourtii LLAP12 (X97366) L. saoudiensis (LN899829) L. lytica (X60080) OIPH-co36R (LC458439) 74 93 OIPH-co35R (LC458438) Legionella sp. strain LLAP9 (U44911) 82 Legionella sp. strain LLAP7 (U44910) OIPH-co31R (LC458436) Legionella sp. strain LLAP2 (U44909) L. rowbothamii LLAP6 (X97359) L. norrlandica (KJ796839) L. yabuuchiae (AB233210) 99 L. impletisoli (AB233209) L. dresdeniensis (AM747393) L. taurinensis (DQ667196) L. spiritensis (M36030) L. cardiaca (JF831047) 75 L. hackeliae (M36028) 99 L. micdadei (AF227162) L. maceachernii (AF227161) L. nautarum (Z49728) 65 L. drozanskii LLAP1 (X97355) 94 OIPH-co56A (LC458441) L. massiliensis (JF779685) L. tunisiensis (JF779686) 66 96 L. feeleii (X73395) 100 OIPH-co32R (LC458437) L. londiniensis (Z49730) L. nagasakiensis (EU701006) L. beliardensis (AF122884) L. gresilensis (AF122883) L. busanensis (AF424887) L. thermalis (AB899895) (HM208383)

1 0.01

FIG. 1. Phylogenetic relationship between the sequences of partial 16S rRNA genes（ 616bp） from uncultured and 2cultured Figure Legionella 1 isolates obtained in this study and Legionella spp. or strains deposited in GenBank. The source of the sample is shown at the end of the sample name as follows: C, cooling tower; A, amenity water facility; R, river. 3Asterisk Phylogenetic（ ＊） is Legionella relationship detected bybetween culture themethod. sequences Coxiella of burnetii partial（ HM20838316S rRNA） geneswas used (616bp) as an fromoutgroup uncultured species. and Numbers at nodes are bootstrap percentages（ based on 1,000 resamplings） only values above 60% are shown. 4 cultured Legionella isolates obtained in this study and Legionella spp. or strains deposited in GenBank.

A. castellanii5 The affects source monocyte of the sample entry is shownmechanisms at the endand of thepossible sample nameto be asengulfed follows: and C, coolingdigested tower;（ Edagawa A, amenity et al., enhances the virulence of L. pneumophila（ Cirillo et 2016）. Thus, amoebic co-cultures can aid the prolifer- al., 1999）. Therefore, positive qPCR signals following ation of Legionella, which are both viable and capable amoebic6 co-culturewater facility; may indicate R, river. the Asteriskexistence (*)of viable is Legionella of growing detected in amoebae. by culture This method. combined Coxiella methodology burnetii and virulent Legionella（ Samrakandi et al., 2002）. By can indicate the contamination with Legionella spp. with performing7 (HM208383) amoebic co-cultures, was used as “dead an outgroup bacteria” species. or Numbershigher accuracy. at nodes are bootstrap percentages (based on Legionella with no ability to multiply in amoebae may be Legionella have reportedly been detected in the 8 1,000 resamplings) only values above 60% are shown.

1 LEGIONELLA DETECTION BY AMOEBIC CO-CULTURE 219

sputum of patients with , who tested nega- Cirillo, J.D., Cirillo, S.L.G., Yan, L., Bermudez, L.E., Falkow, tive using common diagnostic tests for Legionnaires’ S. and Tompkins, L.S.（ 1999） Intracellular growth in Acanthamoeba castellanii affects monocyte entry mecha- disease, including the culture method and the amoebic nisms and enhances virulence of . co-culture method. The latter method has been used to Infect. Immun., 67, 4427-4434. detect amoeba-resistant bacteria including Legionella, Davis, M.M.（ 2016）, Flint Water Advisory Task Force and a combination of molecular biological techniques FINAL REPORT, pp.1−62, State of Michigan,（ https:// has been used to discover new species of these bacteria www.michigan.gov/documents/snyder/FWATF_FINAL_ REPORT_21March2016_517805_7.pdf） （Thomas et al., 2010; Benamar et al., 2017）. Although Dennis, P.J., Brenner, D.J., Thacker, W.L., Wait, R., Vesey, G., the number of confirmed Legionella infections is rising Steigerwalt, A.G., Benson, R.F.（ 1993） Five new Legionella in Japan, the source of the infection has been identified species isolated from water. Int. J. Syst. Bacteriol., 43, only in a very few cases. Legionnaires' disease is rarely 329-37 Edagawa, A., Kimura, A., Kawabuchi-Kurata, T., Kusuhara, Y., proven by culture, and detection of urinary antigen is and Karanis, P.（ 2009） Isolation and genotyping of poten- now common. The combinatorial method using qPCR tially pathogenic Acanthamoeba and Naegleria species from followed by the amoebic co-culture method used in this tap-water sources in Osaka, Japan., Parasitol. Res., 105, study is a useful tool for sampling Legionella contam- 1109-1117. Edagawa, A., Kimura, A., Doi, H., Tanaka, H., Tomioka, K., ination at low concentrations, particularly since the Sakabe, K., Nakajima, C., and Suzuki, Y.（ 2008） Detection conventional culture method does not detect Legionella of culturable and nonculturable Legionella species from under these conditions. hot water systems of public buildings in Japan. J. Appl. Amoebic co-cultures in combination with qPCR Microbiol., 105, 2104-2114. analyses revealed the presence of viable Legionella Edagawa, A., Kimura, A., Kawabuchi-Kurata, T., Adachi, S., Furuhata, K., and Miyamoto, H.（ 2015） Investigation of spp. in environmental waters collected from various Legionella contamination in bath water samples by culture, sources in Japan. Gene sequencing indicated the pres- amoebic co-culture, and real-time quantitative PCR meth- ence of bacterial species responsible for infections in ods. Int. J. Environ. Res. Public Health, 12, 13118-13130. humans as well as unculturable bacterial species. We Edagawa, A., Kimura, A., Adachi, S., Matsushima, K., and Miyamoto, H.（ 2016） Investigation of Legionella concluded that environmental waters are inhabited by Contamination in Amenity Water Samples Using Amoebic Legionella spp., although these samples are not always Co-Culture with LAMP Methods（ in Japanese）. J. Antibact. detected by the traditional culture method. Applying Antifung. Agents, 44, 585-589 the amoebic co-culture technique prior to real-time Fields, B.S., Benson, R.F., and Besser, R.E.（ 2002） Legionella and Legionnaires’ disease: 25 Years of investigation. Clin. qPCR might be more useful to detect viable and virulent Microbiol. Rev., 15, 506-526. Legionella spp. since their ability to invade and multi- Furuhata, K., Edagawa, A., Ishizaki, N., and Fukuyama, M. ply within Acanthamoeba can be correlated with their （2013） Isolation of Legionella species from Noyu（ unat- pathogenicity. tended natural hot springs in mountains and fields） samples in Japan. Biocontrol Sci., 18, 169-173. Furuhata, K., Hara, M., Yoshida, S., and Fukuyama, M. ACKNOWLEDGMENTS （2004） Distribution of Legionella spp. in hot spring baths in Japan（ in Japanese）. Kansenshogaku Zasshi, 78, This work was supported by JSPS KAKENHI Grant 710-716. Number JP 24590841. Greub, G., and Raoult, D.（ 2004） Microorganisms resistant to free-living amoebae. Clin. Microbiol. Rev., 17, 413-33. Hookey, J.V., Saunders, N. A., Fry, N.K., Birtles R.J., and REFERENCES Harrison T.G.（ 1996） Phylogeny of Legionellaceae based on small-subunit ribosomal DNA sequences and proposal Adeleke, A.A., Fields, B.S., Benson, R.F., Daneshvar, M.I., of Legionella lytica comb. nov. for Legionella-like amoebal Pruckler, J.M., Ratcliff, R.M., Harrison, T.G., Weyant, pathogens. Int. J. Syst. Bacteriol., 46, 526-531. R.S., Birtles, R.J., Raoult, D., and Halablab, M.A.（ 2001） Horwitz, M.A.（ 1987） Characterization of avirulent mutant Legionella drozanskii sp. nov., Legionella rowbothamii Legionella pneumophila that survive but do not multiply sp. nov. and Legionella fallonii sp. nov.: three unusual within human monocytes. J. Exp. Med., 166, 1310-1328. new Legionella species. Int. J. Syst. Evol. Microbiol., 51, Inoue, H., Agata, K., and Ohta, H.（ 2019） Phylogenetic char- 1151-1160. acterization of viable but-not-yet cultured Legionella groups Bonetta, S., Bonetta, S., Ferretti, E., Balocco, F., and Carraro, grown in amoebic cocultures: a case study using various E.,（ 2010） Evaluation of Legionella pneumophila contam- cooling tower water samples. Biocontrol Sci., 24, 39-45. ination in Italian hotel water systems by quantitative real- Karasudani, T., Kuroki, T., Otani, K., Yamaguchi, S., Sasaki, time PCR and culture methods. J. Appl. Microbiol., 108, M., Saito, S., Fujita, M., Sugiyama, K., Nakajima, H., and 1576-1583. Murakami, K.,（ 2009） Legionella contamination risk factors Benamar, S., Bou, Khalil. J.Y., Blanc-Tailleur, C., Bilen, M., in non-circulating hot spring water. Kansenshogaku Zasshi, Barrassi, L., La Scola, B.,（ 2017） Developmental cycle and （in Japanese） 83, 36-44. genome analysis of Protochlamydia massiliensis sp. nov. La Scola, B., Mezi, L., Weiller, P.J., Raoult, D., and Bernard, a new species in the Parachlamydiacae family. Front Cell S.,（ 2001） Isolation of Legionella anisa using an amoebic Infect Microbiol., 7, 385. coculture procedure. J. Clin. Microbiol., 39, 365-366. 220 A. EDAGAWA ET AL.

Miyamoto, H., Yamamoto, H., Arima, K., Fujii, J., Maruta, K., Segal, G., and Shuman, H.A.（ 1999） Legionella pneumophila Izu, K., Shiomori, T., and Yoshida, S.（ 1997） Development utilizes the same genes to multiply within Acanthamoeba of a new seminested PCR method for detection of castellanii and human macrophages. Infect. Immun., 67, Legionella species and its application to surveillance of 2117-2124. Legionellae in hospital cooling tower water. Appl Environ Thacker, W.L., Wilkinson, H.W., Plikaytis, B.B., Steigerwalt, Microbiol. 63, 2489-2494. A.G., Mayberry, W.R., Moss, C.W., Brenner, D.J.（ 1985） Miyamoto, H., Yoshida, S., Taniguchi, H., and Shuman, H.A. Second serogroup of Legionella feeleii strains isolated from （2003） Virulence conversion of Legionella pneumophila by humans. J. Clin. Microbiol., 22, 1-4. conjugal transfer of chromosomal DNA. J. Bacteriol., 185, Thomas, V., McDonnell, G., Denyer, S.P., and Maillard, J.Y. 6712-6718. （2010） Free-living amoebae and their intracellular patho- National Institute of Infectious Diseases and Tuberculosis and genic microorganisms: risks for water quality. FEMS Infectious Diseases Control Division, Ministry of Health, Microbiol Rev., 34, 231-259. Labour and Welfare.（ 2000） Legionellosis, April 1999-July Totaro, M., Valentini, P., Costa, A.L., Frendo, L., Cappello, A., 2000. Infect. Agents Surveill. Rep., 21, 186 -187. Casini, B., Miccoli, M., Privitera, G., Baggiani, A.（ 2017） Neumeister, B., Schoniger, S., Faigle, M., Eichner, M., and Presence of Legionella spp. in hot water networks of differ- Dietz, K.（ 1997） Multiplication of different Legionella ent Italian residential buildings: a three-year survey. Int. J. species in Mono Mac 6 cells and in Acanthamoeba castel- Environ. Res. Public. Health., 26, E1296. lanii. Appl. Environ. Microbiol., 63, 1219-1224. Touron-Bodilis, A., Pougnard, C., Frenkiel-Lebosse, H., and Samrakandi, M.M., Cirillo, S.L., Ridenour, D.A., Bermudez, Hallier-Soulier, S.（ 2011） Usefulness of real-time PCR as L.E., Cirillo, J.D.（ 2002） Genetic and phenotypic differ- a complementary tool to the monitoring of Legionella spp. ences between Legionella pneumophila strains. J. Clin. and Legionella pneumophila by culture in industrial cooling Microbiol., 40, 1352-1362. systems. J. Appl. Microbiol. 111, 499-510. Sasahara, T., Kikuno, R., Okuda, S., Sekiguchi, T., Zahran, S., McElmurry, S.P., Kilgore, P.E., Mushinski, Satoh, Y., Takayama, Y., Aoki, M., and Inoue, M. D., Press, J., Love, N.G., Sadler, R.C., and Swanson, （2004） Physicochemical factors influencing distribu- M.S.,（ 2018） Assessment of the Legionnaires' disease tion of Legionella species in Japanese hot springs. outbreak in Flint, Michigan, Proc Natl Acad Sci USA., 20, Kansenshogaku Zasshi,（ in Japanese）, 78, 545-553. E1730-E1739.