Journalof

MethodsMicrobiological

Journal of Microbiological Methods 37 (1999) 155–164

The potential of in situ hybridization and an immunogold assay toidentify Legionella associations with other microorganisms

*Reshma Desai, Chad Welsh, Michele Summy, Mary Farone, Anthony L. NewsomeDepartment of Biology Box 60, Middle Tennessee State University, Murfreesboro, TN 37132, USA

Received 19 January 1999; received in revised form 18 March 1999; accepted 18 March 1999

Abstract

Based on in vitro studies, bacteria in the genus Legionella are believed to multiply within protozoa such as amoebae inaquatic environments. Current methods used for detection of Legionella species, however, are not designed to show thisrelationship. Thus the natural intimate association of Legionella with other microorganisms remains to be clearlydocumented and the extent to which protozoa might be infected with Legionella species remains undefined. In this report wedescribe methods based on the use of Legionella specific reagents that would prove useful in describing its associations withother microorganisms. An immunogold and in situ hybridization technique have the potential to demonstrate the naturaloccurrence of Legionella species in free-living amoebae. In preliminary observations, however, bacteria reactive withLegionella specific reagents were often not intimately associated with amoebae. Bacteria occurred as free single cells, as cellaggregates, in proximity to other cells and debris, and only occasionally in close proximity to amoebae. Although someLegionella species replicate within amoebae, these preliminary observations suggest the bacteria may be encountered mostfrequently as extracellular microorganisms, either free-floating or in association with other structures or microorganisms. Thefuture use of these techniques will aid in the elucidation of any naturally occurring relationships between Legionella speciesand other microorganisms. 1999 Elsevier Science B.V. All rights reserved.

Keywords: Immunogold; In situ hybridization; Interactions; Legionella

1. Introduction L. pneumophila accounts for the majority of reportedpneumonic illness caused by the genus Legionel-

Since the original report of Legionella pneumo- la(Dowling et al., 1992; Breiman, 1993). Thisphila as the causative agent of Legionnaires’ disease, species is a facultative intracellular parasite whichsporadic outbreaks have been reported and the multiplies within human mononuclear phagocytesnumber of bacteria in the genus have grown to over (Horwitz, 1980). It is widespread in aquatic environ-30 described species (Brenner et al., 1979; Harrison ments and human-made structures, such as plumbingand Saunders, 1994; Fliermans, 1996). Many of systems and cooling towers (Bollin et al., 1985;these species have been associated with illness, but Addiss et al., 1989). In vitro studies have also shown

that L. pneumophila multiplies intracellularly in free-living amoebae and Tetrahymena (Fields, 1993).*Corresponding author. Tel.: 1 1-615-898-2058; fax: 1 1-615-Intracellular events in amoebae are remarkably simi-898-5093.

E-mail address: ANewsome@mtsu.edu (A.L. Newsome) lar to L. pneumophila infected monocytes (Horwitz,

0167-7012/99/$ – see front matter 1999 Elsevier Science B.V. All rights reserved.PI I : S0167-7012( 99 )00057-3

156 R. Desai et al. / Journal of Microbiological Methods 37 (1999) 155 –164

1980; Newsome et al., 1985; Abu Kwaik, 1996; Gao fluorescein antibody conjugate was replaced with aet al., 1997). Amoebae have also been implicated as gold-labeled secondary antibody conjugate. Thisbacteria growth factors in potable water (Wadowsky conjugate allowed us to subsequently Geimsa stainet al., 1988; Wadowsky et al., 1991). Thus it is likely the specimen and clearly visualize the relationshipthat this fastidious bacterium survives in nature by between antibody reactive bacteria and any intimateintracellular replication in protozoa. With the excep- association with eucaryotic host cells. There is antion of evidence by Harf and Monteil (1988), where important advantage of both these detection systems,L. pneumophila was cultured from lysates of free- as opposed to fluorescein detection methods. Notliving amoebae, additional support of this naturally only could the relationship between reactive bacteriaoccurring interaction has been lacking. and a host cell be visualized using light microscopy,

From a public health standpoint, the detection of but a permanent slide is produced which also pro-Legionella species in environmental samples are motes a more extensive, less time dependent methodtypically based on culture (buffered charcoal yeast of analysis.extract agar), reaction with fluorescein-labeled anti-bodies, and most recently the polymerase chainreaction (Palmer et al., 1995; Feeley et al., 1979). 2. Materials and methodsOne disadvantage of these methods is that they arenot designed to document the association of the 2.1. Maintenance of stock culturesbacteria with its proposed protozoan hosts. Thedevelopment and application of methods capable of Legionella pneumophila (ATCC 33216) was cul-resolving naturally occurring relationships between tured on buffered charcoal yeast extract (BCYE)Legionella species and other microorganisms are agar (Becton Dickinson, Cockeysville, MD) at 358Cneeded to understand the natural history of the genus and passed at weekly intervals. Escherichia coli wasLegionella. used as a negative control and maintained on trypti-

The purpose of this investigation was to develop case soy (TSA) agar (Difco, Detroit, MI) at 378C.methodology that could be used to identify Legionel- The amoeba Acanthamoeba polyphaga (ATCC

3la species in environmental samples and also show if 30461) was cultured at 358C in 25 cm Falcon tissuethe bacteria were associated with its proposed culture flasks (Becton Dickinson, Franklin Lakes,amoebae or protozoan host cell. For this, both a NJ) containing 5 ml of trypticase soy (TSB) brothnucleic acid detection system and an antibody-me- (Difco). Monolayers of amoebae were infected bydiated approach were used. In situ nucleic acid adding 100 ml of Legionella bacteria suspended inhybridization was used for nucleic acid detection. sterile spring water at a multiplicity of infectionThis was based on the description of an oligonucleo- (MOI) of 100 bacteria /amoeba and incubated attide probe (705–722 base pairs) specific to the 16S 378C.rRNA region of the genus Legionella by Manz et al.(1995). In previous studies of laboratory co-cultures 2.2. Fixation of cells to slidesthe fluorescein-labeled probe could detect L. pneu-mophila in the protozoan Tetrahymena pyriformis Laboratory co-cultures of bacteria and amoebae(Manz et al., 1995). For our purposes, however, the were placed onto slides using a Shandon cytospinprobe was linked to digoxigenin and hybridization to (Shandon Inc., Pittsburg, PA) at 700 rpm (approxi-bacterial cells was visualized by alkaline phosphatase mately 50 3 g) for 5 min. All cytospin preparationslabeled anti-digoxigenin and a specific substrate were fixed in 4% paraformaldehyde (Sigma, St.(Zarda et al., 1991). The antibody-mediated identifi- Louis, MO) made up in DEPC water (50 ml ofcation system was based on the commercially avail- diethylpyrocarbonate /100 ml deionized water) for inable and widely used Remel Legionella poly ID kit, situ hybridization or in methyl alcohol (2 min) forwhich consisted of a primary antibody to 22 immunogold assay. Cells in environmental waterLegionella species and a secondary fluorescein anti- samples were likewise affixed to slides by cyto-body conjugate. In this investigation, however, the centrifugation. Initially 50 ml of the sample was

R. Desai et al. / Journal of Microbiological Methods 37 (1999) 155 –164 157

centrifuged at 2000 3 g for 20 min. All but 2 ml was heim, Indianapolis, IN) in 500 ml of buffer (50 ml ofpipetted off and the sediment was resuspended in the 1 M Tris, 50 ml of 1 M NaCl, 25 ml of 1 M MgCl ,2

remaining fluid. A volume of 50 ml–150 ml of the 500 ml of deionized water). This was initiallyconcentrated cells was placed in the cytospin (700 incubated overnight in the dark at RT withoutrpm for 5 min). In addition, slides were placed in a agitation and subsequently the incubation time waslarge volume of the environmental water samples reduced to 6 h. The reaction was stopped with buffer(250ml) for up to one week at room temperature. (0.5 ml of 1 M Tris, 2.5 ml of 0.2 M EDTA and 500Typically single cells and some debris adhered to the ml of deionized water), briefly washed in deionizedslides which were subsequently fixed for in situ water, mounted in a water based mounting mediumhybridization and the immunogold technique. (Polysciences Inc, Warrington, PA) and viewed by

light microscopy. During each incubation step the2.3. In situ hybridization specimen slides were maintained in a humid

chamber.The Leg 705 probe (59-

CTGGTGTTCCTTCCGATC-39) was labeled with 2.4. Immunogold stainingdigoxigenin using the Boehringer Mannheim (In-dianapolis, IN) terminal transferase system (Cat. No. Slides were fixed in methanol for 2 min and220582). The final concentration of the labeled probe allowed to air dry. The primary antibody (affinitywas 25 ng/ml. Fixed slides were treated with buffer purified pooled rabbit immune sera to 22 species ofcontaining 50 ml diethylpyrocarbonate (DEPC, Legionella) from the Remel Legionella Poly-ID TestSigma, St. Louis, MO) and 1 mg lysozyme per 100 Kit (Lenexa, KS) was applied and incubated for 45ml deionized water. Slides were washed in DEPC min at 378C in a humid chamber. Slides were thenwater (50 ml /100 ml deionized water) and then rinsed in PBS. Gold labeled (5 nm beads) goattreated with proteinase K buffer (1 ml of 1 M Tris, anti-rabbit conjugate (Goldmark Biologicals, Phillip-2.5 ml of 0.2 M EDTA, 0.5 ml of 10% SDS and 98 sburg, NJ) was diluted 1:100 in a 0.085% NaCl, 2%ml deionized water) for 10 min at room temperature bovine serum albumin solution, placed on the slide(RT). Subsequently fixed specimens were treated and incubated for 45 min at 378C. Slides were rinsedwith hybridization buffer (0.25 ml of SSC buffer, 0.1 in PBS and the color was developed (2–5 min atg n-laurylsarcosine, 20 ml of 10% SDS and 1g RT) using silver enhancement reagents (Goldmarkblocking agent in 100 ml deionized water), the SSC Biologicals, Phillipsburg, NJ). Following color de-buffer consisted of 88 g sodium chloride, 44 g velopment, slides were lightly (20 min) Giemsasodium citrate in 500 ml deionized water) for 5 min stained, rinsed in water, and allowed to air dryat 428C in a thermal shaker waterbath at 20 rpm. overnight before being coverslipped with permountThen 10 ml (25 ng) of digoxigenin-labeled (LEG mounting medium.705) probe in hybridization buffer was added andallowed to hybridize overnight at 458C in the thermalshaker water bath (20 rpm). Unbound probe was 3. Resultswashed twice with wash buffer (50 ml of SSC, 50 mlof 10% SDS, 500 ml water) and treated with blocker Initially in situ hybridization with the digoxigeninbuffer (Boehringer Mannheim, Indianapolis, IN) for labeled 705 probe was performed on agar-grown20 min at RT and 20 rpm. Subsequently a 1:10 laboratory cultures of Legionella pneumophila.dilution of alkaline phosphatase labeled anti-digox- These bacterial cells were stained dark purple andigenin (Boehringer Mannheim, Indianapolis, IN) was their characteristic rod-shaped morphology was ap-added and incubated for 30 min at 378C. It was parent. Initially cells were viewed after overnightwashed twice with buffer (50 ml of 1 M Tris, 75 ml incubation with the NBT substrate, however, sub-of 1 M NaCl, 100 ml of deionized water). The sequent tests showed that satisfactory color develop-substrate consisted of 4.5 ml nitro-blue tetrazolium ment could be achieved with as little as 6 h substrate(NBT) and 3.5 ml X-phosphate (Boehringer Mann- incubation. When older cultures (two weeks) were

158 R. Desai et al. / Journal of Microbiological Methods 37 (1999) 155 –164

used the positive signal was weaker but remained Subsequently in situ hybridization analysis waseasily distinguishable from the negative control performed on both Legionella culture positive andEscherichia coli. If the probe was omitted from the negative water samples from a variety of sources.in situ hybridization procedure, no color develop- The aim was to determine if the probe reacted withment was observed following the addition of the bacteria fixed onto slides and if reactive procaryotessubstrate to the Legionella cells. Thus a positive could be identified in intracellular niches, such asreaction in fixed bacteria was directly associated with amoebae, or intimately associated with otherhybridization of the digoxigenin-labeled probe. procaryotic microorganisms. In most instances, non-

Another objective using laboratory co-cultures was reactive bacteria and other microorganisms wereto determine if Legionella bacteria residing within present on slides. When reactive bacteria were ob-amoebae hosts cells could also be specifically iden- served, however, they occurred as single cells or intified by the digoxigenin labeled probe in conjunc- one instance as in clumps (Fig. 2). Despite thistion with alkaline phosphatase labeled anti-digox- aggregation of reactive cells no distinguishingigenin. Typically during intracellular multiplication amoebic structures were observed with these bac-the bacteria accumulated in large vacuoles that teria. As a control, duplicate slides were preparedoccupied a significant portion of the cytoplasmic and treated in an identical manner except the probespace in the amoeba host cell (Fig. 1). The intracel- was omitted. In the absence of the probe the slideslular Legionella pneumophila were likewise stained were generally negative. However, a potential com-darkly following in situ hybridization and the color plication was occasional color development in somedevelopment was equivalent to that observed with cell aggregates and debris. This was associated withagar grown bacteria. No color development was endogenous alkaline phosphatase activity (sinceobserved when amoebae were similarly incubated color was observed in duplicate slides which re-with Escherichia coli. ceived only the nitro-blue tetrazolium substrate)

Fig. 1. Intracellular L. pneumophila stained darkly following in situ hybridization. Typically the bacteria accumulate in large vacuoleswhich occupied a significant portion of the amoeba (arrow) cytoplasm. ( 3 600)

R. Desai et al. / Journal of Microbiological Methods 37 (1999) 155 –164 159

Fig. 2. A positive reaction in an aggregate of cells from a water sample following in situ hybridization using the Leg 705 probe. ( 3 1000)

which could not be satisfactorily neutralized by conjugate thus their presence was not easily visual-addition of the enzyme inhibitor levamisole. Al- ized initially. Giemsa staining, however, imparted athough the effect of endogenous enzyme activity pink color to the cells since they failed to react withcould substantially be decreased by reducing the the antibodies.substrate incubation time to 6 h rather than 24 h, use Both the immunoreactive and nonreactive cellsof alkaline phosphatase conjugates may not be were similarly observed when cells from environ-suitable for identification of Legionella cells within mental water samples were fixed onto glass slides.biofilms. Typically the dark stained immunoreactive bacteria

The bacteria could likewise specifically be iden- occurred as isolated cells (Fig. 3). On occasion thesetified using the immunogold identification technique dark stained cells were in proximity to (Fig. 4) orwith agar-grown L. pneumophila and in co-culture close association with amoebae (Fig. 5) that werewith A. polyphaga. When the bacteria were localized apparent following Giemsa staining. Staining bywithin amoebae the degree of color development was Giemsa also showed that consortia of cells wereequivalent to that observed in extracellular bacteria. formed on the slides in some instances. TheseThus the intracellular localization of bacteria did not cellular aggregates often consisted of cells withappear to diminish access of the antibody to bacteria different morphological features. In some environ-epitopes. The ability to follow color development mental samples, highly immunoreactive procaryoteswith Giemsa staining was important for visualizing were identified within these Giemsa stained cellularamoebae morphological features. This additional aggregates and debris (Fig. 6). Although the bacteriastaining did not affect the darkly stained immuno- were not directly on or in amoebae, the presence ofreactive Legionella cells. An additional benefit of the numerous amoebae within these cellular areas and inGiemsa staining was the identification of nonreactive adjacent areas on the slides was commonly observed.bacteria cells. For example, E. coli controls failed to Fifty water samples from a variety of sources suchreact with both the primary and secondary antibody as cooling towers, plumbing fixtures, well water, and

160 R. Desai et al. / Journal of Microbiological Methods 37 (1999) 155 –164

Fig. 3. Isolated dark staining bacteria from an environmental water sample after the immunogold reaction ( 3 1000).

Fig. 4. Immunoreactive bacteria in proximity to an amoeba (arrow) ( 3 600).

R. Desai et al. / Journal of Microbiological Methods 37 (1999) 155 –164 161

Fig. 5. Immunoreactive bacteria in close association with an amoeba (arrow). Bacteria appeared to be on the surface of an amoeba whichcould not be photographed in the same depth of field at this magnification ( 3 1000).

Fig. 6. Identification of dark staining immunoreactive bacteria within Giemsa stained cellular aggregates and debris ( 3 600).

162 R. Desai et al. / Journal of Microbiological Methods 37 (1999) 155 –164

surface waters were examined. This included al., 1997), and potentially in vivo (Brieland et al.,Legionella culture positive and negative samples and 1997), interactions with protozoa, such as amoebae,samples not tested for Legionella colony forming that are believed to be the natural host since amoebaeunits. A correlation between the presence of bacteria can also typically be cultured from water harboringreactive with the Legionella specific reagents and Legionella bacteria. Other investigators have shownrecovery of colony forming units remains to be that L.pneumophila could be isolated from an algal-accurately established. The ability to document bacterial mat community (Tison et al., 1983). StudiesLegionella species on the slides is likely related to suggested that Legionella used the complex organicthe degree to which cells are concentrated onto slides material produced by cyanobacteria as the soleby cytocentrifugation or by the development of carbon and energy source. The photosynthetic micro-biofilms. It is possible that if the number of organisms were apparently vital for survival ofLegionella colony forming units is low the sensitivity Legionella and were a significant ecological factor inof in situ hybridization and immunogold identifica- the environment (Pope et al., 1982; Tison et al.,tion is likewise diminished. However, this remains to 1983).be clearly established. In the present preliminary study, bacteria reactive

with Legionella specific reagents were identified onslides which contained cells from aquatic habitats.

4. Discussion Of initial interest was the observation that reactivebacterial cells occurred as single cells, as clumps of

Results of this study suggest that in situ hybridiza- bacteria, and in proximity to other cells and debris.tion using a digoxigenin labeled probe and an The reactive bacteria were not within well definedantibody based immunogold technique using com- vacuoles of amoebae (as typically observed in lab-mercially available antibodies would be useful in oratory co-cultures), but could be visualized indetermining relationships between Legionella species proximity to amoebae and occasionally closely asso-and other microorganisms in environmental water ciated with a eucaryotic cell. In addition, the visuali-samples. If investigators are to realize the natural life zation of clumps of reactive cells alludes to thehistory of bacteria in the genus Legionella, methods possibility amoebae may expel membrane-boundmust be developed to document the naturally occur- bound vesicles containing Legionella bacteria inring state of the bacteria and document the level of natural settings. This has been adeptly demonstratedinteractions with other microorganisms. This also has with in vitro studies (Berk et al., 1998). The frequentpractical implications. Legionella species are associ- visualization of extracellular bacteria reactive withated with sporadic respiratory disease outbreaks and Legionella specific reagents suggests also that theare a cause of nosocomial disease. Present control actual time within amoebae may be low. Laboratoryefforts are directed primarily at efforts to directly studies for example, typically show Legionella com-inhibit the multiplication or to kill the bacteria. plete their replication cycle in amoebae within 96 hDespite these short-term eradication methods, the (Newsome et al., 1985; Fields, 1993; Abu Kwaik,occurrence and disease causing potential of the 1996). Thus the amplified number of bacteria maybacteria continues to be a public health problem. The become free-floating, associated with other cells, oridentification of novel point-sources of human in- incorporated into biofilms. Laboratory studies havefection such as vegetable misters, cruise ships, and demonstrated L. pneumophila can enter a dormanthot tubs, continues to be reported in the literature phase which is unrecoverable on laboratory media.(Mahoney et al., 1992; Jernigan et al., 1996). However, these viable cells can be resuscitated bySatisfactory long-term control measures will likely co-culture with protozoa (Steinert et al., 1997). Therequire a more detailed understanding of the natural occurrence of L. pneumophila microcolonies withinhistory of the bacteria in both natural and human- biofilms has been described by Rogers and Keevilmade settings. Studies have clearly documented (1992). Here the authors used immunogold in con-significant in vitro (Newsome et al., 1985; Fields, junction with episcopic differential interference con-1993; Cirillo et al., 1994; Abu Kwaik, 1996; Gao et trast microscopy. It was suggested the bacteria

R. Desai et al. / Journal of Microbiological Methods 37 (1999) 155 –164 163

Addiss, D.G., Davis, J.P., LaVenture, M., Wand, P.J., Hutchinson,consortium supplied sufficient nutrients to supportM.A., McKinney, R.M., 1989. Community-acquired Legion-Legionella growth extracellularly within the biofilm.naires’ disease associated with a cooling tower: evidence for

Use of the Leg 705 probe for in situ hybridization longer-distance transport of Legionella pneumophila. Am. J.has the additional potential to identify undescribed Epidemiol. 130, 557–568.

Berk, S.G., Ting, R.S., Turner, G.W., Ashburn, R.J., 1998.Legionella that would not react with the pooledProduction of respirable vesicles containing live Legionellaantibodies of described species. For example, apneumophilia cells by two Acanthamoeba spp. Appl. Environ.

recent bacterial isolate from an amoeba taken from a Microbiol. 64, 279–286.soil sample reacted strongly with the digoxigenin Bollin, G.E., Plouffe, J.F., Para, M.F., Hackman, B., 1985.

Aerosols containing Legionella pneumophila generated byprobe (Newsome et al., 1998). Subsequent 16Sshower heads and hot-water faucets. Appl. Environ. Microbiol.rDNA gene analysis showed it to be a member of the50, 128–134.genus Legionella but distinct from the described

Breiman, R.F. (1993) Modes of transmission in epidemic andspecies. This newly described bacterium failed to nonepidemic Legionella infection: directions for further study:react with immune sera to described species in the In: Barbaree et al. (eds). Legionella current status and emerging

perspectives. Am. Soc. Microbiol. Washington, DC. pp 30–35.genus Legionella. Thus, the Leg 705 probe would beBrenner, D.J., Steigerwalt, A.G., McDade, J.E., 1979. Classifica-additionally useful in establishing the associations of

tion of the Legionnaires’ disease bacterium: Legionella pneu-undescribed Legionella-like bacteria and amoebae or mophila, genus novum, species nova, of the family Legionel-other protozoan host cells. laceae, family nova. Ann. Intern. Med. 90, 656–658.

It is our belief that additional use of the methods Brieland, J.K., Fantone, J.C., Remick, D.G., LeGendre, M.,McClain, M., Engleberg, N.G., 1997. The role of Legionelladescribed in this report will afford the potential topneumophila-infected Hartmannella vermiformis as an infecti-document the proposed naturally occurring associa-ous particle in a murine model of Legionnaires’ disease. Infect.

tions between Legionella species and other aquatic Immun. 65, 5330–5333.microorganisms. The ability to identify Legionella Cirillo, J.D., Falkow, S., Tompkins, L.S., 1994. Growth ofbacteria by in situ hybridization or immunogold Legionella pneumophila in Acanthamoeba castellanii enhances

invasion. Infect. Immun. 62, 3254–3261.staining in environmental samples is likely a reflec-Dowling, J.N., Saha, A.K., Glew, R.H., 1992. Virulence factors oftion of cell concentration on slides. In future studies

the family Legionellaceae. Microbiol. Rev. 56, 32–60.it may also be beneficial to use the methods de- Feeley, J.C., Gibson, R.J., Gorman, G.W., Langford, N.C.,scribed in this report in conjunction with culture Rasheed, J.K., Makel, D.C., Baine, W.B., 1979. Charcoal yeastmethods to more closely establish the Legionella extract agar: primary isolation medium for Legionella pneumo-

phila. J. Clin. Microbiol. 10, 437–441.association with other microorganisms and recoveryFields, B.S., 1993. Legionella and protozoa: interaction of aof colony forming units in aquatic samples of

pathogen and its natural host. In: Barbaree, J.M., Breiman,interest. R.F., Dufour, A.P. (Eds.), Legionella: Current Status and

Emerging Perspectives, American Society for Microbiology,Washington, D. C, pp. 129–136.

Fliermans, C.B., 1996. Ecology of Legionella: from data toAcknowledgementsknowledge with a little wisdom. Microb. Ecol. 32, 203–228.

Gao, L., Harb, O.S., Abu Kwaik, Y., 1997. Utilization of similarThe authors wish to thank Stephen M. Wright for mechanisms by Legionella pneumophila to parasitize two

evolutionarily distant host cells, mammalian macrophages andhelpful suggestions and technical assistance. Thisprotozoa. Infect. Immun. 65, 4738–4746.research was supported in part by grant R825352-01-

Harf, C., Monteil, H., 1988. Interactions between free-living0 to Sharon Berk from the Environmental Protectionamoebae and Legionella in the environment. Water Sci. Tech.

Agency. 20, 235–239.Harrison, T.G., Saunders, N.A., 1994. Taxonomy and typing of

legionellae. Rev. Med. Microbiol. 5, 79–90.Horwitz, M.A., 1980. The legionnaires’ disease bacterium

References (Legionella pneumophila) multiplies intracellularly in humanmonocytes. J. Clin. Invest. 66, 441–450.

Abu Kwaik, Y., 1996. The phagosome containing Legionella Jernigan, D.B., Hofmann, J., Cetron, M.S., Genese, C.A., Nuroti,pneumophila within the protozoan Hartmannella vermiformis J.P., Fields, B.S., Benson, R.F., Carter, R.J., Edelstein, P.H.,is surrounded by the rough endoplasmic reticulum. Appl. Guerreo, I.C., Paul, S.M., Lipman, H.B., Breiman, R.F., 1996.Environ. Microbiol. 62, 2022–2028. Outbreak of Legionnaires’ disease among cruise ship passen-

164 R. Desai et al. / Journal of Microbiological Methods 37 (1999) 155 –164

gers exposed to a contaminated whirlpool spa. Lancet 347, Rogers, J., Keevil, C.W., 1992. Immunogold and fluorescein494–499. immunolabelling of Legionella pneumophila within an aquatic

Mahoney, F.J., Hoge, C.W., Farley, T.A., Barbaree, J.M., Brieman, biofilm visualized by using episcopic differential contrastR.F., Benson, R.F., McFarland, L.F., 1992. Community-wide microscopy. Appl. Environ. Microbiol. 58, 2326–2330.outbreak of Legionnaires’ disease associated with a grocery Steinert, M., Emody, L., Amann, R., Hacker, J., 1997. Resucita-store mist machine. J. Infect. Dis. 165, 736–739. tion of viable but nonculturable Legionella pneumophila Phila-

Manz, W., Amann, R., Szewzyk, R., Szewzyk, U., Stenstorm, delphia JR32 by Acanthamoeba castellanii. Appl. Environ.T.A., Hutzler, P., Schleifer, K.H., 1995. In situ identification of Microbiol. 63, 2047–2053.legionellaceae using 16S rRNA-targeted oligonucleotide probes Tison, D.L., Baross, J.A., Seidler, R.J., 1983. Legionella inand confocal laser scanning microscopy. Microbiology 141, aquatic habitats in the Mount Saint Helens blast zone. Curr.29–39. Microbiol. 9, 345–348.

Newsome, A.L., Baker, R.L., Miller, R.D., Arnold, R.R., 1985. Wadowsky, R.M., Butler, L.J., Cook, M.K., Verma, S.M., Paul,Interactions between Naegleria fowleri and Legionella pneu- M.A., Fields, B.S., Keleti, G., Syora, J.L., Yee, R.B., 1988.mophila. Infect. Immun. 50, 449–452. Growth supporting activity for Legionella pneumophila in tap

Newsome, A.L., Scott, T.M., Benson, R.F., Fields, B.S., 1998. water cultures and implication of Hartmannellid amoebae asIsolation of an amoeba naturally harboring a distinctive growth factors. Appl. Environ. Micro. 54, 2677–2682.Legionella species. Appl. Environ. Microbiol. 64, 1688–1693. Wadowsky, R.M., Wilson, T.M., Kapp, N.J., West, A.J., Kuchta,

Palmer, C.J., Bonilla, G.F., Roll, B., Pasyko-kolva, C.L., Sanger- J.M., States, S.J., Dowling, J.N., Yee, R.B., 1991. Multiplica-mano, L.R., Fujioka, R.S., 1995. Detection of Legionella tion of Legionella spp. in tap water containing Hartmannellaspecies in reclaimed water and air with the Enviro Amp vermiformis. Appl. Environ. Microbiol. 57, 1950–1955.Legionella PCR kit and direct fluorescent antibody staining. Zarda, B., Amann, R., Wallner, G., Schleifer, K.H., 1991. Identifi-Appl. Environ. Microbiol. 61, 407–412. cation of single bacterial cells using digoxigenin-labeled, rRNA

Pope, D.H., Soracco, R.J., Gill, H.K., Fliermans, C.B., 1982. targeted oligonucleotides. J. Gen. Microbiol. 137, 2823–2830.Growth of Legionella pneumophila in two membered cultureswith green algae and cyanobacteria. Curr. Microbiol. 7, 319–322.