(~) INSTITUTPASTEUR/ELsEVIER Paris 1991
Res. Microbiol.
1991, 142, 677-685
The intracellular multiplication of Legionella pneumophila in protozoa from hospital plumbing systems K. Nahapetian 0), O. Challemel 0), D. Beurtin (l), S. Dubrou tl) P. Gounon (2) and F. Squinazi o) ¢t) Laboratoire d ' H y g i b n e d e la Ville de Paris (Prof. B. Festy), 11 rue George Eastman, 75013 Paris, and (2) .~tatian I~ontrnlo l~ . . . . . . . . . . . . . . . . . . . . . .h/tlrr~eranla .... r . . . .I~loptraninlJa . . . . . . . . ~ - - , .lnetitJtt . . . .Pasteur, . . . . .75724 . . . Paris . . . r~o,tov . .
SUMMARY
Between October 1987 and March 1989, we tested 144 water samples obtained from the plumbing and cooling tower systems of 5 Paris hospitals for the presence of legionellae and amoebae. Of the samples tested for Legionella, 67 out of 144 (46.5 %) were positive, and 82 out of 116 tested for amoebae (70.7 %1 were positive. The ability of protozoa to support the multiplication of legionella was shown by incubating samples at 35.5°C for 7-15 days. Prior to determining the presence of legionellae and amoebae, 51 of the 144 samples were incubated. After incuba*.ion, 22 out of 25 (88 %) samples which were positive for the presence of both Legionella and amoebae showed multiplication of Legionella. In 3 out of the 25 (12 %} samples containing Legionella and amoebae, Legionella failed to multiply. Six out of the 51 (11.8 %} samples which were negative in direct culture for Legionella but positive for amoebae, became positive after incubation. Legionella did not multiply in samples negative for amoebae, nor was there pro:iferation in samples after filtration through a 1.2-Fm membrane followed by incubation for the same period and temperature. Strains of Legionella pneumophila serogroup 1 and serogroup 6 {SG1 and SG6), including 3 patient isolates and 2 environmental isolates, were cocultured with 2 strains of amoebae and Tetrahymena pyriformis. Plate counts, Gimenez staining and electron microscopy demonstrated that intracellular legionellae proliferation occurred.
Key-words: Nosocomial infection, Legionella, Amoeba, Tetrahymena pyriformis, Air, Water; Cocultures, Intracellular proliferation, Plumbing and cooling tower systems.
INTRODUCTION The survival and multiplication of legionellae in plumbing and cooling tower systems can be explained in part by the coexistance of some of the following environmental factors : temperature, stagnation, sediment formation, biodegradable substances, bacterial flora or interaction between protozoa and L e g i o n e ; l a Submitted August 8, 1990, accepted February 4, 1991.
(Ager e t al., 1985; Anand e t al., 1984; Barbaree e t aL, 1986). Several studies have reported that free-living amoebae and ciliated protozoa ( T e t r a h y m e n a p y r i f o r m i s ) can support the multiplication of L e g i o n e l l a p n e u m o p h i l a (Anand e t aL, 1984; Fields e t al., 1984, 1986, 1989; Holden e t aL, 1984; Spring, 1987; Wadowsky e t aL, 1988).
K. N A H A P E T I A N E T AL.
678
This stud)' describes a simple m e t h o d o f showing the proliferation o f Legionella in hot water plumbing and cooling tower systems, as well as the intracellular multiplication o f indigenous Legionella in a m o e b a e isolated f r o m the same sites and in a stock strain o f T. pyriformis.
MATERIALS AND METHODS
Water samples From October 1987 to March 1989, 144 samples were obtained from the plumbing and cooling tower systems of 5 hospitals in the Paris area.
Water culture and recovery of legionellae Water samples were collected from hot and cold water taps, hot water tanks, cooling towers and humidifiers. All specimens were cultured on selective media according to the Centers for Disease Control guidelines for recovery of Legionella from water (Gorman et al., 1983). In order to detect an increase in the number of indigenous Legionella, 51 of the 144 samples (200 ml) were incubated at 35.5°C for 7-15 days and then cultured.
Recover), of amoebae Only 116 of the 144 samples were cultured for amoebae on non-nutrient agar (I.5 070 w/v) according to Singh et al. (1950). Plates were inoculated beforehand with heat-killed Escherichia coli (110°C for 15 min) and incubated at 30°C. Identification of amoebic isolates were based on the Page method (Page, 1976).
Preparation of cocultures of Legionella and amoebae Cocultures were prepared in 25-cm 3 tissue culture flasks according to Fields et al. (1984). All flasks contained 3 ml non-nutrient agar inoculated with heat-killed E. coil These flasks were inoculated with amoebae isolated from hospital tap water and cooling towers. Proliferation of amoebae was observed with an inverted microscope for 24-72 h, until they covered most of the agar surface. Flasks were then overlaid with 10 ml of sterile tap water containing approximately 104 L. pneumophila cells/ml and incubated at 35.5°C for 17 days.
oculated with 1 ml of T. pyriformis from 42-72 h cultures. After 4 days' incubation at 25°C, the numbers of T. pyriformis were adjusted to approximately 104 cells/ml and numeration of T. pyriformis was done with a Malassez slide after having killed T. pyriformis with one drop of chlorine (3 mg/l). After removing 1 ml from each flask containing only T. pyriformis, the numeration was done and the number of T. pyriformis in the flasks was adjusted with sterile tap water. Then, the flasks were inoculated with 1 ml of sterile tap water containing the same concentration of L. pneumophila (104 cells/ml).
Gimenez staining Tissue culture flasks containing cocultures were briskly shaken and 1 ml of coculture fluid was centrifuged at 1,000 g for 1 min. The sediment was used to prepare smears for Gimenez staining (Gimenez, 1964).
Electron microscopy Amoebae and T. pyriformis were fixed in 2.5 °70 glutaraldehyde in 0.1 M cacodylate buffer pH 7.4, directly in the culture flasks. After fixation overnight at 4°C, cells were collected by centrifugation (2,000 g for 2 min), washed twice in cacodylate buffer and postfixed with osmium tetroxide/potassium ferricyanide ( 1 % OsO4, 0.8 o70 K4Fe(CN)6) in 0.1 I~A cacodylate buffer pH 7.4 for 1 h at 4°C. The pellet was washed in the buffer and in distilled water, then dehydrated with an alcohol gradient and infiltrated ir~ epoxy-l-2-propane and epon. The plastic was polymerized for 2 days at 60°C. Sections were cut on an ultramicrotome (LKB) with a diamond knife, picked up on copper grids and poststained with aqueous uranyl acetate and lead citrate. The sections were examined with a "Philips CM12 TEM" at 80 kV (Fields et aL, 1988, 1989).
lntracellular multiplication of indigenous Legionel-
la in amoebae and T. pyrlformis Strains of L. pneumophila SGI and SG6, including 3 patient isolates and 2 environmental isolates were used in coculture studies. Two strains of amoebae (Hartmannella verm if ormis and Naeg leria spp.) isolated from the hospitals and T. pyriformis were used in cocultures.
Legionella and
Heat and acid treatment of cocultures of Legionella and amoebae
Tissue culture flasks (25-cm 3) containing 3.5 ml Elliot's medium and 3.5 ml sterile tap water were in-
To eliminate environmental contamination of amoebae, 1 ml of the cocalture was heat- and acidtreated according to Nahapetian el aL (1987).
Preparation of cocultures T. pyriformis
of
LEGIONELLA MUL TIPLICA T/ON IN HOSPITAL WATER PROTOZOA Quantitation of Legionella in cocultures After an incubation of 5 h to 1-17 days, 100 ml of heated cocultures of different dilutions (pure, 10 -2, 10-3 and 10-4), were plated on charcoalbuffered yeast extract agar with antibiotics (vancomycin l mg/i, cephalothin 4 mg/l, cycloheximide 80 mg/! and colistin 16 mg/l) and ¢-ketoglutarate. For each dilution, 3 platings were carried out. RESULTS O f the 144 samples tested for Legionella, 67 (46.5 °70) proved positive, ~t,,u~-~of the . . . . . . . tested for amoebae, 82 out of 116 (70.7 %) proved positive. Prior to determining the presence of Legionella and amoebae, 51 of the 144 samples were selected for incubation. Virtually half of the samples tested proved positive for Legionella and amoebae (25/51; 49 °7o). Legionella multiplied in 22 of the 25 (88 %) sampies after incubation. Legionella failed to multiply in 3 of the 25 (12 % ) samples which contained amoebae. O f the samples which were negative by direct culture for LegioneUa, 6 out of 51 (11.8 %) became positive after incubation (fig. l). Multiplication did not occur in samples that were filtered through a 1.2-1xm membrane or in samples which were negative for amoebae and positive for Legionella; as shown in the control samples in figure 1. A n increase in the number o f Legionella in the 51 incubated samples, classified by their different sampling sites, is also shown in figure 1.
679
The difference between the average number of Legionella cells/! before incubation and after incubation was 21ogl0. This increase was statistically significant (9 > 0.0001, Wileoxon signed rank). The distribution of Legionella and amoebae in the 51 samples before and after incubation, according to their sampling site, is shown in table I.
L. pneumophila patients' isolates and environmental isolates were multiplied in coeul-
~lote
~nc~bal=on
atCet ~nc~at~o~ /
aLog (CFU'JL)
N = NUMBEROF SAMPLE~
I
c,,.
.
Fig. 1. Increase in the number of L. pneumophila in 51 incubated samples. (*) Three samples contained amoebae not supporting multiplication, and 2 samples did not contain amoebae.
Table I. Distribution of Legionella and amoebae in 51 incubated samples of hospital plumbing and cooling tower systems.
Water samples Cold water Hot water tanks Hot tap water Cooling tower and humidifier Total
Total number incubated 5 11 25 10 51
Legionella Number Number positive positive direct (%) incubated (°70) 0 5 (45.5) 17 (68) 5 (50) 27 (53)
0 8 (73) 12 (48) 5 (50) 28 (55)
Amoeb., e NumbeJ positive (o70) 2 (40) 11 (100) 20 (80) 9 (90) 42 (82)
680
K. N A H A P E T I A N E T AL.
LoglffU/r~) 8
[ J
A~ageco~enUaUon ol L. ~e~rr~a isolates
Log(O:U/rH A~ageconcen~'a~ ~ ~lates
of L
2
I~3
ol . . . . o
i
2
6
f
iO
tiME(days) 0 I 2
6
~
~
lime(~
Fig. 2. Multiplicationof 5 L. pneumophilaisolatesincubate< with amoebae (Naegleria sp.).
Log(CFU/d)
Averagec~centm~Uonof
L pneu/~o#daisc4ates
Fig. 4. Multiplicationof 5
L. pneumophila isolatesincubated with 7". pyriformis.
tures with 2 different genera of environmental amoebae and T. pyriformis. Each amoebic and Legionella isolate was tested at least twice, with similar results, in control flasks free from amoebae and T. pyriformis. The number of Legionella cocultured with the 2 different genera of amoebae is shown in figures 2 and 3, and with T. pyriformis in figure 4. Intracellular legionellae proliferation was shown by Gimenez staining of amoebae (fig. 5 and 6) and T. pyriformis (fig. 7 and 8).
i
,
2 3
l
6
t
to
i
llME Wy~)
Fig. 3. Multiplicationof 5 L. pneumophila isolatesincubated with amoebae (Hartmannella sp.).
A patient isolate of L. pneumophila SG1, subgroup Pontiac, was cocultured with H. vermiformis isolated from a hot water tap in the same hospital. This Legionella isolate was also cocultured with T. pyriformis. Intracellular multiplication occurred in this amoebae (fig. 9) and in T. pyriformis (fig. 10, 11 and 12) after a 30-h incubation. Moreover, intracellular multiplication in cocultures was not observed after 24 h, which disproves amoebic environmental contamination.
LEGIONELLA MULTIPLICATION
IN HOSPITAL
W A I'ER P R O T O Z O A
681
~J
Fig. 5 and 6. H. vermiformis infected with an L. pneumophila SGI patient isolate.
Fig. 7 and 8. T. pyriformis infected with an L. pneumophila SGI patient isolate.
Mild and severe infections, respectively, are shown. Gimenez stain ( x 2,500).
Mild and severe infections, respectively, are shown. Gimenez stain ( × 2,500).
K. NAHAPETIAN E7 AL.
682
Fig. 9. Electron micrograph of H. vermiformis environmental isolate infected with an L. pneumophila SG1 patient isolate, after a 30-h incubation. Shown are numerous Legionella cells in a large vesicle ( x 16,000).
Fig. 10. Electronmicrographof T. pyriformis infectedwith L. pneumophila SGI patient isolate, after a 30-h incubation. Shown are numerous Legionella cells in a large vesicle (x9,500).
DISCUSSION
direct culture for Legionella yet become positive after incubation (unpublished data). Twelve of the 65 (18.5 °7o) samples from our continuous survey are shown in figure 1.
The presence of amoebae in hot water plumbing and cooling tower systems contributes to the proliferation of Legionella and enables this bacterium to persist and multiply in high numbers, as shown by Fields et aL (1989) and Wadowsky et al., 1988. Intracellular Legionella cells appear to survive disinfection, enabling them to repopulate treated environments (King and Shotts, 1988; King et al., 1988). This process is apparent in water samples from swimming pools (frequently chlorinated), cooling towers and humidifiers (after disinfection); these were used as control samples in this study since they are negative in
Samples containing intracellular Legionella gave false-negative results in direct culture. Six out of the 51 (11.2 %) samples became positive after incubation. Incubation of samples containing amoebae thus appears to be a simple method for calculating the percentage of Legionella-positive water samples, especially when the original concentration Legionella is low. The increase in the number of legionellae after incubation was statistically significant (p > 0,0001).
LEGIONELLA MULTIPLICATION IN HOSPITAL WATER PROTOZOA
683
other soluble substances are retained in the filtrate. The resulting increase in the amount of Legionella depended on the genera of amoebae present in the coculture. Certain genera of amoebae failed to support multiplication. In 3 out of 25 (12 °7o) samples which contained Legionella and amoebae, no multiplication occurred, as shown by Fields et al. (1989). The fact that Legionella and amoebae were isolated from hot water of hospitals where cases of suspected nosocomial legionellosis occurred and that they interacted in cocultures, suggests that, in nature, Legionella infect amoebae in water at increased temperatures. Water siter colonized by amoebae may act as reservoirs for Legionella (Harf et al., 1988; Henke and Seidel, 1986; Newsome et aL, 1985; Rowbotham, 1984; Shuval et a!., 1989). In order to combat this ecological interaction, disinfection often proves problematic. Efforts to control nosocomial legionellosis by eradication of Legionella by the usual disinfection procedures (repeated hyperchlorination and high temperature) seem inadequate and impractical. Tiffs study has shown us that the modified choice for systematic control in these hospitals is the elimination of both amoebae and Legionella. According to Fields et aL (1989), L. pneumophila does not multiply in cocultures at 45°C. Therefore, in practice, effective disinfection would imply raising the stored hot water temperature to 60°C, then maintaining it at at least 50°C, and chlorinating it so as to have 2-3 mg of free residual chlorine per litre, up to every point of use (Bulletin the W H O , 1990). Fig. 11 and 12. Electron micrographs of T. pyriformis infected with a L. pneumophila SGI patient isolate. lntracytoplasmic multiplicationat different magnifications, x6,000 and × 16,000, respectively, is shown.
The Legionella-growth-supporting activity of amoebae was demonstrated on filtering amoebae-containing samples through 1.2-tzm membranes. Filtration eliminates protozoa in water samples, while bacterial flora and many
According to Wickramanayake et al. (1984), another possibility concerns using the effects of ozone on trophozoites and cysts of free-living amoebae. A similar study was carried out by Perrine et al. (1990) who showed that ozone (0.4 mg/1 for 4 min) rapidly distroyed or inactivated all the amoebic trophozoites and cysts.
Acknowledgements We thank Dr. Barry S. Fields and his collaborators (Center for Infectious Diseases. Centers for Disease Con-
K. N A H A P E T I A N E T AL.
684
trol, Atlanta, GA 30333, USA) for their kind contribution to our study; Mr. Y. Le Moullec for helping us with the preparation of the figures and the statistical analysis; and M.C. Pr6vost for her technical help with the electronic microscopy.
Multiplication intracellulaire de Legionella pneumophila dans les protozoaires isol~s des r6seaux d'eau ehaude sanitaire el des installations de climatisation du milieu hospitalier D'octobre 1987 /t mars 1989, une recherche de Legionella et d'amibes libres a 6t6 effectu6e au niveau de rdseaux d'eau chaude sanitaire et d'installations de climatisation de 5 h6pitaux parisiens. La prdsence de Legionella a 6t6 observ6e dans 64/144 (46,5 %) des 6chantillons d'eau, et celle d'amibes libres dans 82/116 (70,7 %). La capacit6 des protozoaires :h favoriser la multiplication de L. pneumophila a 6td montr6e par incubation des 6chantillons d'eau/135°C pendant 7 /t 15 jours. Ainsi, pour 22/25 (88 %) ~chantillons qui contenaient/t la fois des Legionella et des amibes libres, on a observ6 une prolifdration des Legionella au cours de l'6tape d'incubation. De plus, parmi les 51 6chantillons incub6s, 6 d'entre eux (11,8 %) qui contenaient des amibes mais pas de Legionella directement cultivables, se sont r6v616s positifs apr6s incubation. Aucnne multiplication de Legionella n'a 6t6 observ6e Iorsque les dchantillons ne contenaient pas d'amibes libres, ou lorsque les amibes avaient dt6 61imindes par filtration sur membrane de 1,2 tim. Des touches de L. oneumophila s6rogroupes 1 et 6 isoldes chez 3 mala, les, et 2 souches isol6es de l'environnement de l'h6pital ont 6td cocultiv/:es respectivement avec 2 souches d'amihes diffdrentes 6galement isol6es de cet hSpital, et avec Tetrahymena pyriformis. Le ddnombrement sur g61ose, la coloration de Gimenez et la microscopie 61ectronique ont montrd l'existence d'une prolifdration intraceUulaire.
Mots-cl#s: Infection nosocomiale, Legionella, Amibe, Tetrahymena pyriformis, Air, Eau; Cocultures, Prolif6ration intracellulaire, Eau chaude sanitaire et climatisation.
References Ager, B.P. & Tickner, J.A. (1985), The control of microorganismsresponsiblefor legionnairesdisease and humidifier fever. Occupational Hygiene Monograph N ° 14. Science Technology letters, 12. Clarance road Kew, Surrey TW9 3NL (UK).
Anand, C.M., Skinner, A.R., Malic, A. & Kurtz, J.B. (1984), lntracellular replication of Legionella pneumophila in Acanthamoeba palestiensis, in "Legionella: proceedings of the 2nd International Symposium" (Tornsberry A. Belows, J.C. Feely & W. Jakubowski) (pp. 330-332). American Society for Microbiology, Washington, D.C. Barbaree, J.M., Fields, B.S., Feeley, J.C., Gorman, G.W. & Martin, W.T. (1986), Isolation of protozoa from water associated with legionellosis outbreak and demonstration of intracellular multiplication of Legio-
nella pneumophila. Appl. en vironm. Microbiol., 51,
422-424. Bulletin the World Health Organization (1990), Epidemiology, prevention and control of legionellosis. Memorandum from a WHO Bull., 68, 155-164. Fields, B.S., Shorts, E.B., Jr., Feeley, J.C., Gorman, G.W. & Martin, W.T. (1984), Proliferation of Legionella pneumophila as an intracellular parasite of the ciliated protozoan Tetrahymena pyriformis. Appl. environm. Microbiol., 47, 467-471. Fields, B.S., Barbaree, J.M., Shotts, E.B., Jr., Feeley, J.C., Morill, W.E., Sanden, G.N. & Dykstra, M.J. (1986), Comparison of guinea pig and protozoan models for determining virulence of Legionella species. Infect. Immun., 53, 553-559. Fields, B.S., Sanden, G.N., Barbaree, J.M., Morill, W.E., Wadowsky, R.M., White, E.H. & Feeley, J.C. (1989), Intracellular multiplication of Legionella pneumophila in amoebae isolated from hospital hot water tanks. Curr. MicrobioL, 18, 131-137. Gimenez, D.F. (1964), Staining of rickettsia in yolk-sac cultures. Stain Technol., 39, 135-140. Gorman, G.W., Barbaree, J.M. & Feeley, J.C. (1983), Proceedures for the recovery of Legionella from water. Developmental Manual U.S. Departement of Hea!th and human services. Publi.' Ilcalth Servi=e, Centers for Disease Control, Atlanta, GA. Haft, C. & Monteil, H. (1988), Interactions between freeliving amoebae and Legionella in the environment. Wat. Sci. Tech., 20, 235-239. Henke, M. & Seidel, K.M. (1986), Association between Legionella pneumophila and amoebae in water. Israel J. med. ScL, 22, 690-695. Holden, E.P., Wood, D.O., Winkler, H.H. & Leinbach, E.D. (1984), Intracellular growth of Legionellapneumophila within freshwater amoebae, in "Legionella: proceedings of the 2nd International Symposium" trhornsberry, A. Balows, J.C. Feely & W. Jakubowski) (pp. 329-330). American Society for Microbiology, Washington, D.C. King, C.H. & Shotts, E.B. (1988), Enhancement of Edwardsiella tarda and Aeromonas salmonicida through ingestion by the ciliated protozoan Tetrahymena pyriformis. FEMS Microbiol. Letters, 51, 95-100. King, C.H., Shotts, E.B., Jr., Wooley, R.E. & Porter, K.G. (1988), Survival of coliforms and bacterial pathogens within protozoa during chlorination. Appl. environm. Microbiol., 54, 3023-3033. Nahapetian, K. & Squinazi, F. (1987), Modalit6s d'isolement de Legionella sp. dans les eaux et les boues r6siduaires, in "Colloque Legionella 6-7th May 1987" (J. Fleurette, N. Bornstein, D. Marmet & M. Surgot) (pp. 165-168). Fondation Mdrieux, Lyon.
LEGIONELLA M U L TIPLICA T I O N I N H O S P I T A L W A T E R P R O T O Z O A Newsome, A.L., Baker, R.L., Miler, R.D. & Arnold, R.R. (1985), Interactions between Naegleria fowleri and Legionella pneumophila. Infect. lmmun., 50, 449-452. Page, F.C. (1976), An illustrated key to freshwater and soil amoebae. Scientific publication no. 34. Freshwater Biological Association, Ambleside, England. Perrine, O., Barbier, D. & George, P. (1990), Cin~ique d'action du chlore et de I'ozone sur les trophozoRes et les kystes d'amibes libres des genres Naegleria et Acanthamoeba. J. franf. Hydro., 21, 113-121. Rowbotham, T.J. (1984), Legionella and amoebae, in "'Legionella: Proceedings of the 2nd International Symposium" (Thornsberry, A. Balows, J.C. Feely & W. Jabukowski) (pp. 325-327). American Society for Microbiology, Washington, D.C. Shuval, H.I., Fattal, B. & Bercovier, H. (1989), Legion-
685
naires disease and the water environment in Israel. Wat. Sci. Techn., 20, 3~-38. Singh, B.N. (1950), A cultural method for gtowi.Jg small, free-living amoebae for hhe study of their nuclear division. Nature (Lond.), 165, 65. Spring, D.R. (1987), Legionella, microbial ecology, and inconspicuous consumption. J. infect. Dis., 155, 1086-1087. Wadowsky, R.M., Butler, J.L., Cook, M.K., Verma, S.M., Paul. M.A., Fields, B.S., Kelti, G., Sykora, J.L. & Yee, R.B. (1988), Growth-supporting activity for Legionella pneumophila in tap water cultures and implication of Hartmannellid amoebae as growth factors. Appl. environm. MicrobioL, 54, 2677-2682. Wikramanayake, G.B., Rubin, A.J. & Sproul, O.J. (1984), Inactivation of Naegleria and Giardia cysts in water by ozonation. J. WPCF, 56, 983-988.
Res. Microbiol.
1991, 142, 677-685
The intracellular multiplication of Legionella pneumophila in protozoa from hospital plumbing systems K. Nahapetian 0), O. Challemel 0), D. Beurtin (l), S. Dubrou tl) P. Gounon (2) and F. Squinazi o) ¢t) Laboratoire d ' H y g i b n e d e la Ville de Paris (Prof. B. Festy), 11 rue George Eastman, 75013 Paris, and (2) .~tatian I~ontrnlo l~ . . . . . . . . . . . . . . . . . . . . . .h/tlrr~eranla .... r . . . .I~loptraninlJa . . . . . . . . ~ - - , .lnetitJtt . . . .Pasteur, . . . . .75724 . . . Paris . . . r~o,tov . .
SUMMARY
Between October 1987 and March 1989, we tested 144 water samples obtained from the plumbing and cooling tower systems of 5 Paris hospitals for the presence of legionellae and amoebae. Of the samples tested for Legionella, 67 out of 144 (46.5 %) were positive, and 82 out of 116 tested for amoebae (70.7 %1 were positive. The ability of protozoa to support the multiplication of legionella was shown by incubating samples at 35.5°C for 7-15 days. Prior to determining the presence of legionellae and amoebae, 51 of the 144 samples were incubated. After incuba*.ion, 22 out of 25 (88 %) samples which were positive for the presence of both Legionella and amoebae showed multiplication of Legionella. In 3 out of the 25 (12 %} samples containing Legionella and amoebae, Legionella failed to multiply. Six out of the 51 (11.8 %} samples which were negative in direct culture for Legionella but positive for amoebae, became positive after incubation. Legionella did not multiply in samples negative for amoebae, nor was there pro:iferation in samples after filtration through a 1.2-Fm membrane followed by incubation for the same period and temperature. Strains of Legionella pneumophila serogroup 1 and serogroup 6 {SG1 and SG6), including 3 patient isolates and 2 environmental isolates, were cocultured with 2 strains of amoebae and Tetrahymena pyriformis. Plate counts, Gimenez staining and electron microscopy demonstrated that intracellular legionellae proliferation occurred.
Key-words: Nosocomial infection, Legionella, Amoeba, Tetrahymena pyriformis, Air, Water; Cocultures, Intracellular proliferation, Plumbing and cooling tower systems.
INTRODUCTION The survival and multiplication of legionellae in plumbing and cooling tower systems can be explained in part by the coexistance of some of the following environmental factors : temperature, stagnation, sediment formation, biodegradable substances, bacterial flora or interaction between protozoa and L e g i o n e ; l a Submitted August 8, 1990, accepted February 4, 1991.
(Ager e t al., 1985; Anand e t al., 1984; Barbaree e t aL, 1986). Several studies have reported that free-living amoebae and ciliated protozoa ( T e t r a h y m e n a p y r i f o r m i s ) can support the multiplication of L e g i o n e l l a p n e u m o p h i l a (Anand e t aL, 1984; Fields e t al., 1984, 1986, 1989; Holden e t aL, 1984; Spring, 1987; Wadowsky e t aL, 1988).
K. N A H A P E T I A N E T AL.
678
This stud)' describes a simple m e t h o d o f showing the proliferation o f Legionella in hot water plumbing and cooling tower systems, as well as the intracellular multiplication o f indigenous Legionella in a m o e b a e isolated f r o m the same sites and in a stock strain o f T. pyriformis.
MATERIALS AND METHODS
Water samples From October 1987 to March 1989, 144 samples were obtained from the plumbing and cooling tower systems of 5 hospitals in the Paris area.
Water culture and recovery of legionellae Water samples were collected from hot and cold water taps, hot water tanks, cooling towers and humidifiers. All specimens were cultured on selective media according to the Centers for Disease Control guidelines for recovery of Legionella from water (Gorman et al., 1983). In order to detect an increase in the number of indigenous Legionella, 51 of the 144 samples (200 ml) were incubated at 35.5°C for 7-15 days and then cultured.
Recover), of amoebae Only 116 of the 144 samples were cultured for amoebae on non-nutrient agar (I.5 070 w/v) according to Singh et al. (1950). Plates were inoculated beforehand with heat-killed Escherichia coli (110°C for 15 min) and incubated at 30°C. Identification of amoebic isolates were based on the Page method (Page, 1976).
Preparation of cocultures of Legionella and amoebae Cocultures were prepared in 25-cm 3 tissue culture flasks according to Fields et al. (1984). All flasks contained 3 ml non-nutrient agar inoculated with heat-killed E. coil These flasks were inoculated with amoebae isolated from hospital tap water and cooling towers. Proliferation of amoebae was observed with an inverted microscope for 24-72 h, until they covered most of the agar surface. Flasks were then overlaid with 10 ml of sterile tap water containing approximately 104 L. pneumophila cells/ml and incubated at 35.5°C for 17 days.
oculated with 1 ml of T. pyriformis from 42-72 h cultures. After 4 days' incubation at 25°C, the numbers of T. pyriformis were adjusted to approximately 104 cells/ml and numeration of T. pyriformis was done with a Malassez slide after having killed T. pyriformis with one drop of chlorine (3 mg/l). After removing 1 ml from each flask containing only T. pyriformis, the numeration was done and the number of T. pyriformis in the flasks was adjusted with sterile tap water. Then, the flasks were inoculated with 1 ml of sterile tap water containing the same concentration of L. pneumophila (104 cells/ml).
Gimenez staining Tissue culture flasks containing cocultures were briskly shaken and 1 ml of coculture fluid was centrifuged at 1,000 g for 1 min. The sediment was used to prepare smears for Gimenez staining (Gimenez, 1964).
Electron microscopy Amoebae and T. pyriformis were fixed in 2.5 °70 glutaraldehyde in 0.1 M cacodylate buffer pH 7.4, directly in the culture flasks. After fixation overnight at 4°C, cells were collected by centrifugation (2,000 g for 2 min), washed twice in cacodylate buffer and postfixed with osmium tetroxide/potassium ferricyanide ( 1 % OsO4, 0.8 o70 K4Fe(CN)6) in 0.1 I~A cacodylate buffer pH 7.4 for 1 h at 4°C. The pellet was washed in the buffer and in distilled water, then dehydrated with an alcohol gradient and infiltrated ir~ epoxy-l-2-propane and epon. The plastic was polymerized for 2 days at 60°C. Sections were cut on an ultramicrotome (LKB) with a diamond knife, picked up on copper grids and poststained with aqueous uranyl acetate and lead citrate. The sections were examined with a "Philips CM12 TEM" at 80 kV (Fields et aL, 1988, 1989).
lntracellular multiplication of indigenous Legionel-
la in amoebae and T. pyrlformis Strains of L. pneumophila SGI and SG6, including 3 patient isolates and 2 environmental isolates were used in coculture studies. Two strains of amoebae (Hartmannella verm if ormis and Naeg leria spp.) isolated from the hospitals and T. pyriformis were used in cocultures.
Legionella and
Heat and acid treatment of cocultures of Legionella and amoebae
Tissue culture flasks (25-cm 3) containing 3.5 ml Elliot's medium and 3.5 ml sterile tap water were in-
To eliminate environmental contamination of amoebae, 1 ml of the cocalture was heat- and acidtreated according to Nahapetian el aL (1987).
Preparation of cocultures T. pyriformis
of
LEGIONELLA MUL TIPLICA T/ON IN HOSPITAL WATER PROTOZOA Quantitation of Legionella in cocultures After an incubation of 5 h to 1-17 days, 100 ml of heated cocultures of different dilutions (pure, 10 -2, 10-3 and 10-4), were plated on charcoalbuffered yeast extract agar with antibiotics (vancomycin l mg/i, cephalothin 4 mg/l, cycloheximide 80 mg/! and colistin 16 mg/l) and ¢-ketoglutarate. For each dilution, 3 platings were carried out. RESULTS O f the 144 samples tested for Legionella, 67 (46.5 °70) proved positive, ~t,,u~-~of the . . . . . . . tested for amoebae, 82 out of 116 (70.7 %) proved positive. Prior to determining the presence of Legionella and amoebae, 51 of the 144 samples were selected for incubation. Virtually half of the samples tested proved positive for Legionella and amoebae (25/51; 49 °7o). Legionella multiplied in 22 of the 25 (88 %) sampies after incubation. Legionella failed to multiply in 3 of the 25 (12 % ) samples which contained amoebae. O f the samples which were negative by direct culture for LegioneUa, 6 out of 51 (11.8 %) became positive after incubation (fig. l). Multiplication did not occur in samples that were filtered through a 1.2-1xm membrane or in samples which were negative for amoebae and positive for Legionella; as shown in the control samples in figure 1. A n increase in the number o f Legionella in the 51 incubated samples, classified by their different sampling sites, is also shown in figure 1.
679
The difference between the average number of Legionella cells/! before incubation and after incubation was 21ogl0. This increase was statistically significant (9 > 0.0001, Wileoxon signed rank). The distribution of Legionella and amoebae in the 51 samples before and after incubation, according to their sampling site, is shown in table I.
L. pneumophila patients' isolates and environmental isolates were multiplied in coeul-
~lote
~nc~bal=on
atCet ~nc~at~o~ /
aLog (CFU'JL)
N = NUMBEROF SAMPLE~
I
c,,.
.
Fig. 1. Increase in the number of L. pneumophila in 51 incubated samples. (*) Three samples contained amoebae not supporting multiplication, and 2 samples did not contain amoebae.
Table I. Distribution of Legionella and amoebae in 51 incubated samples of hospital plumbing and cooling tower systems.
Water samples Cold water Hot water tanks Hot tap water Cooling tower and humidifier Total
Total number incubated 5 11 25 10 51
Legionella Number Number positive positive direct (%) incubated (°70) 0 5 (45.5) 17 (68) 5 (50) 27 (53)
0 8 (73) 12 (48) 5 (50) 28 (55)
Amoeb., e NumbeJ positive (o70) 2 (40) 11 (100) 20 (80) 9 (90) 42 (82)
680
K. N A H A P E T I A N E T AL.
LoglffU/r~) 8
[ J
A~ageco~enUaUon ol L. ~e~rr~a isolates
Log(O:U/rH A~ageconcen~'a~ ~ ~lates
of L
2
I~3
ol . . . . o
i
2
6
f
iO
tiME(days) 0 I 2
6
~
~
lime(~
Fig. 2. Multiplicationof 5 L. pneumophilaisolatesincubate< with amoebae (Naegleria sp.).
Log(CFU/d)
Averagec~centm~Uonof
L pneu/~o#daisc4ates
Fig. 4. Multiplicationof 5
L. pneumophila isolatesincubated with 7". pyriformis.
tures with 2 different genera of environmental amoebae and T. pyriformis. Each amoebic and Legionella isolate was tested at least twice, with similar results, in control flasks free from amoebae and T. pyriformis. The number of Legionella cocultured with the 2 different genera of amoebae is shown in figures 2 and 3, and with T. pyriformis in figure 4. Intracellular legionellae proliferation was shown by Gimenez staining of amoebae (fig. 5 and 6) and T. pyriformis (fig. 7 and 8).
i
,
2 3
l
6
t
to
i
llME Wy~)
Fig. 3. Multiplicationof 5 L. pneumophila isolatesincubated with amoebae (Hartmannella sp.).
A patient isolate of L. pneumophila SG1, subgroup Pontiac, was cocultured with H. vermiformis isolated from a hot water tap in the same hospital. This Legionella isolate was also cocultured with T. pyriformis. Intracellular multiplication occurred in this amoebae (fig. 9) and in T. pyriformis (fig. 10, 11 and 12) after a 30-h incubation. Moreover, intracellular multiplication in cocultures was not observed after 24 h, which disproves amoebic environmental contamination.
LEGIONELLA MULTIPLICATION
IN HOSPITAL
W A I'ER P R O T O Z O A
681
~J
Fig. 5 and 6. H. vermiformis infected with an L. pneumophila SGI patient isolate.
Fig. 7 and 8. T. pyriformis infected with an L. pneumophila SGI patient isolate.
Mild and severe infections, respectively, are shown. Gimenez stain ( x 2,500).
Mild and severe infections, respectively, are shown. Gimenez stain ( × 2,500).
K. NAHAPETIAN E7 AL.
682
Fig. 9. Electron micrograph of H. vermiformis environmental isolate infected with an L. pneumophila SG1 patient isolate, after a 30-h incubation. Shown are numerous Legionella cells in a large vesicle ( x 16,000).
Fig. 10. Electronmicrographof T. pyriformis infectedwith L. pneumophila SGI patient isolate, after a 30-h incubation. Shown are numerous Legionella cells in a large vesicle (x9,500).
DISCUSSION
direct culture for Legionella yet become positive after incubation (unpublished data). Twelve of the 65 (18.5 °7o) samples from our continuous survey are shown in figure 1.
The presence of amoebae in hot water plumbing and cooling tower systems contributes to the proliferation of Legionella and enables this bacterium to persist and multiply in high numbers, as shown by Fields et aL (1989) and Wadowsky et al., 1988. Intracellular Legionella cells appear to survive disinfection, enabling them to repopulate treated environments (King and Shotts, 1988; King et al., 1988). This process is apparent in water samples from swimming pools (frequently chlorinated), cooling towers and humidifiers (after disinfection); these were used as control samples in this study since they are negative in
Samples containing intracellular Legionella gave false-negative results in direct culture. Six out of the 51 (11.2 %) samples became positive after incubation. Incubation of samples containing amoebae thus appears to be a simple method for calculating the percentage of Legionella-positive water samples, especially when the original concentration Legionella is low. The increase in the number of legionellae after incubation was statistically significant (p > 0,0001).
LEGIONELLA MULTIPLICATION IN HOSPITAL WATER PROTOZOA
683
other soluble substances are retained in the filtrate. The resulting increase in the amount of Legionella depended on the genera of amoebae present in the coculture. Certain genera of amoebae failed to support multiplication. In 3 out of 25 (12 °7o) samples which contained Legionella and amoebae, no multiplication occurred, as shown by Fields et al. (1989). The fact that Legionella and amoebae were isolated from hot water of hospitals where cases of suspected nosocomial legionellosis occurred and that they interacted in cocultures, suggests that, in nature, Legionella infect amoebae in water at increased temperatures. Water siter colonized by amoebae may act as reservoirs for Legionella (Harf et al., 1988; Henke and Seidel, 1986; Newsome et aL, 1985; Rowbotham, 1984; Shuval et a!., 1989). In order to combat this ecological interaction, disinfection often proves problematic. Efforts to control nosocomial legionellosis by eradication of Legionella by the usual disinfection procedures (repeated hyperchlorination and high temperature) seem inadequate and impractical. Tiffs study has shown us that the modified choice for systematic control in these hospitals is the elimination of both amoebae and Legionella. According to Fields et aL (1989), L. pneumophila does not multiply in cocultures at 45°C. Therefore, in practice, effective disinfection would imply raising the stored hot water temperature to 60°C, then maintaining it at at least 50°C, and chlorinating it so as to have 2-3 mg of free residual chlorine per litre, up to every point of use (Bulletin the W H O , 1990). Fig. 11 and 12. Electron micrographs of T. pyriformis infected with a L. pneumophila SGI patient isolate. lntracytoplasmic multiplicationat different magnifications, x6,000 and × 16,000, respectively, is shown.
The Legionella-growth-supporting activity of amoebae was demonstrated on filtering amoebae-containing samples through 1.2-tzm membranes. Filtration eliminates protozoa in water samples, while bacterial flora and many
According to Wickramanayake et al. (1984), another possibility concerns using the effects of ozone on trophozoites and cysts of free-living amoebae. A similar study was carried out by Perrine et al. (1990) who showed that ozone (0.4 mg/1 for 4 min) rapidly distroyed or inactivated all the amoebic trophozoites and cysts.
Acknowledgements We thank Dr. Barry S. Fields and his collaborators (Center for Infectious Diseases. Centers for Disease Con-
K. N A H A P E T I A N E T AL.
684
trol, Atlanta, GA 30333, USA) for their kind contribution to our study; Mr. Y. Le Moullec for helping us with the preparation of the figures and the statistical analysis; and M.C. Pr6vost for her technical help with the electronic microscopy.
Multiplication intracellulaire de Legionella pneumophila dans les protozoaires isol~s des r6seaux d'eau ehaude sanitaire el des installations de climatisation du milieu hospitalier D'octobre 1987 /t mars 1989, une recherche de Legionella et d'amibes libres a 6t6 effectu6e au niveau de rdseaux d'eau chaude sanitaire et d'installations de climatisation de 5 h6pitaux parisiens. La prdsence de Legionella a 6t6 observ6e dans 64/144 (46,5 %) des 6chantillons d'eau, et celle d'amibes libres dans 82/116 (70,7 %). La capacit6 des protozoaires :h favoriser la multiplication de L. pneumophila a 6td montr6e par incubation des 6chantillons d'eau/135°C pendant 7 /t 15 jours. Ainsi, pour 22/25 (88 %) ~chantillons qui contenaient/t la fois des Legionella et des amibes libres, on a observ6 une prolifdration des Legionella au cours de l'6tape d'incubation. De plus, parmi les 51 6chantillons incub6s, 6 d'entre eux (11,8 %) qui contenaient des amibes mais pas de Legionella directement cultivables, se sont r6v616s positifs apr6s incubation. Aucnne multiplication de Legionella n'a 6t6 observ6e Iorsque les dchantillons ne contenaient pas d'amibes libres, ou lorsque les amibes avaient dt6 61imindes par filtration sur membrane de 1,2 tim. Des touches de L. oneumophila s6rogroupes 1 et 6 isoldes chez 3 mala, les, et 2 souches isol6es de l'environnement de l'h6pital ont 6td cocultiv/:es respectivement avec 2 souches d'amihes diffdrentes 6galement isol6es de cet hSpital, et avec Tetrahymena pyriformis. Le ddnombrement sur g61ose, la coloration de Gimenez et la microscopie 61ectronique ont montrd l'existence d'une prolifdration intraceUulaire.
Mots-cl#s: Infection nosocomiale, Legionella, Amibe, Tetrahymena pyriformis, Air, Eau; Cocultures, Prolif6ration intracellulaire, Eau chaude sanitaire et climatisation.
References Ager, B.P. & Tickner, J.A. (1985), The control of microorganismsresponsiblefor legionnairesdisease and humidifier fever. Occupational Hygiene Monograph N ° 14. Science Technology letters, 12. Clarance road Kew, Surrey TW9 3NL (UK).
Anand, C.M., Skinner, A.R., Malic, A. & Kurtz, J.B. (1984), lntracellular replication of Legionella pneumophila in Acanthamoeba palestiensis, in "Legionella: proceedings of the 2nd International Symposium" (Tornsberry A. Belows, J.C. Feely & W. Jakubowski) (pp. 330-332). American Society for Microbiology, Washington, D.C. Barbaree, J.M., Fields, B.S., Feeley, J.C., Gorman, G.W. & Martin, W.T. (1986), Isolation of protozoa from water associated with legionellosis outbreak and demonstration of intracellular multiplication of Legio-
nella pneumophila. Appl. en vironm. Microbiol., 51,
422-424. Bulletin the World Health Organization (1990), Epidemiology, prevention and control of legionellosis. Memorandum from a WHO Bull., 68, 155-164. Fields, B.S., Shorts, E.B., Jr., Feeley, J.C., Gorman, G.W. & Martin, W.T. (1984), Proliferation of Legionella pneumophila as an intracellular parasite of the ciliated protozoan Tetrahymena pyriformis. Appl. environm. Microbiol., 47, 467-471. Fields, B.S., Barbaree, J.M., Shotts, E.B., Jr., Feeley, J.C., Morill, W.E., Sanden, G.N. & Dykstra, M.J. (1986), Comparison of guinea pig and protozoan models for determining virulence of Legionella species. Infect. Immun., 53, 553-559. Fields, B.S., Sanden, G.N., Barbaree, J.M., Morill, W.E., Wadowsky, R.M., White, E.H. & Feeley, J.C. (1989), Intracellular multiplication of Legionella pneumophila in amoebae isolated from hospital hot water tanks. Curr. MicrobioL, 18, 131-137. Gimenez, D.F. (1964), Staining of rickettsia in yolk-sac cultures. Stain Technol., 39, 135-140. Gorman, G.W., Barbaree, J.M. & Feeley, J.C. (1983), Proceedures for the recovery of Legionella from water. Developmental Manual U.S. Departement of Hea!th and human services. Publi.' Ilcalth Servi=e, Centers for Disease Control, Atlanta, GA. Haft, C. & Monteil, H. (1988), Interactions between freeliving amoebae and Legionella in the environment. Wat. Sci. Tech., 20, 235-239. Henke, M. & Seidel, K.M. (1986), Association between Legionella pneumophila and amoebae in water. Israel J. med. ScL, 22, 690-695. Holden, E.P., Wood, D.O., Winkler, H.H. & Leinbach, E.D. (1984), Intracellular growth of Legionellapneumophila within freshwater amoebae, in "Legionella: proceedings of the 2nd International Symposium" trhornsberry, A. Balows, J.C. Feely & W. Jakubowski) (pp. 329-330). American Society for Microbiology, Washington, D.C. King, C.H. & Shotts, E.B. (1988), Enhancement of Edwardsiella tarda and Aeromonas salmonicida through ingestion by the ciliated protozoan Tetrahymena pyriformis. FEMS Microbiol. Letters, 51, 95-100. King, C.H., Shotts, E.B., Jr., Wooley, R.E. & Porter, K.G. (1988), Survival of coliforms and bacterial pathogens within protozoa during chlorination. Appl. environm. Microbiol., 54, 3023-3033. Nahapetian, K. & Squinazi, F. (1987), Modalit6s d'isolement de Legionella sp. dans les eaux et les boues r6siduaires, in "Colloque Legionella 6-7th May 1987" (J. Fleurette, N. Bornstein, D. Marmet & M. Surgot) (pp. 165-168). Fondation Mdrieux, Lyon.
LEGIONELLA M U L TIPLICA T I O N I N H O S P I T A L W A T E R P R O T O Z O A Newsome, A.L., Baker, R.L., Miler, R.D. & Arnold, R.R. (1985), Interactions between Naegleria fowleri and Legionella pneumophila. Infect. lmmun., 50, 449-452. Page, F.C. (1976), An illustrated key to freshwater and soil amoebae. Scientific publication no. 34. Freshwater Biological Association, Ambleside, England. Perrine, O., Barbier, D. & George, P. (1990), Cin~ique d'action du chlore et de I'ozone sur les trophozoRes et les kystes d'amibes libres des genres Naegleria et Acanthamoeba. J. franf. Hydro., 21, 113-121. Rowbotham, T.J. (1984), Legionella and amoebae, in "'Legionella: Proceedings of the 2nd International Symposium" (Thornsberry, A. Balows, J.C. Feely & W. Jabukowski) (pp. 325-327). American Society for Microbiology, Washington, D.C. Shuval, H.I., Fattal, B. & Bercovier, H. (1989), Legion-
685
naires disease and the water environment in Israel. Wat. Sci. Techn., 20, 3~-38. Singh, B.N. (1950), A cultural method for gtowi.Jg small, free-living amoebae for hhe study of their nuclear division. Nature (Lond.), 165, 65. Spring, D.R. (1987), Legionella, microbial ecology, and inconspicuous consumption. J. infect. Dis., 155, 1086-1087. Wadowsky, R.M., Butler, J.L., Cook, M.K., Verma, S.M., Paul. M.A., Fields, B.S., Kelti, G., Sykora, J.L. & Yee, R.B. (1988), Growth-supporting activity for Legionella pneumophila in tap water cultures and implication of Hartmannellid amoebae as growth factors. Appl. environm. MicrobioL, 54, 2677-2682. Wikramanayake, G.B., Rubin, A.J. & Sproul, O.J. (1984), Inactivation of Naegleria and Giardia cysts in water by ozonation. J. WPCF, 56, 983-988.
