Journal of Applied Bacteriology 1991,71,366-370
ADONIS 002188479100 1358
Enumeration of Aeromonas hydrophila from domestic wastewater treatment plants and surface waters R. Poffe and E. Op de Beeck Laboratory of Industrial Microbiology, Faculty of Agricultural Sciences, Catholic University of Leuven, Belgium 3571/01/91:accepted 1 1 March 1991
R. POFFE AND E. O P DE BEECK. 1991.Influents, effluents and sludges from sewage purification plants and surface water samples were examined quantitatively for Aeromonas hydrophila on the mA medium of Rippey a n d Cabelli. Between lo4 and 106/ml A. hydrophila were found in domestic wastewaters. O n t h e average 99.975 % were removed by activated sludge a n d 98.25% by trickling filters. Only 20.9% of A. hydrophila end u p in the primary sludge, which contained up to 107/g dry sludge. After 3 months, anaerobically (methane) fermented a n d partially dried sludge from trickling filters contained more than lo6 A. hydrophilalg d r y sludge. Surface water receiving raw sewage contained several hundreds of A. hydrophila/ml, comparable with the numbers found in effluent waters, while surface water receiving n o municipal wastewater and destined for t h e preparation of drinking water contained only small and negligible numbers. It was concluded that A. hydrophila was omnipresent in surface water.
INTRODUCTION
Aeromonas hydrophila is generally considered to be a resident aquatic micro-organism. It is reported as part of the normal flora of fish (Shotts et al. 1972; Boulanger et al. 1977) and the agent of epizootic outbreaks among aquatic organisms (Shotts et al. 1972). Interest in Aeromonas spp. has increased in recent years since A. hydrophila has been associated with human diseases (Kuijper et al. 1984), including gastroenteritis (Gracey et al. 1982; Agger et .al. 1985), cellulitis (Fulghum et al. 1978), peritonitis, meningitis (von Graevenitz & Bucher 1983), pneumonia and bacteraemia (Reines & Cook 1981). It is not a ‘true’ enteric bacterium, since it was rarely recovered from human intestine (Rippey & Cabelli 1979). Water environments have been cited many times as the major source of infection: surface waters (Fliermans et al. 1977; Seidler et al. 1980; Cavari et al. 1981), raw and drinking water supplies (Burke et al. 1984a, b), sewage and effluents from domestic water purification plants (Rippey & Cabelli 1979; Havelaar et al. 1987). From the scanty information about A. hydrophila in sewage purification found in the literature it appeared that high densities were consistently present in sewage, suggesting that wastewaters with high nutrient loadings are excelCorrespondence to : Ir R. Poffe, Laboratory of Industrial Microbiology, Faculty of Agricultural Sciences, K. U.L c u v n , Kardinaal Mercierlaan 92, B-3001 Heverlee, Leuven, Belgium.
lent sites of multiplication (Rippey & Cabelli 1979; Schubert 1987). Increased interest in Aeromonas has led to the development of many selective media for its detection and enumeration either by plate count (starch, glutamate, penicillin agar (Kielwein 1969); Shotts & Rimler agar (Shotts & Rimler 1973); glycogen, peptone, beef extract agar (McCoy & Pilcher 1974); starch, ampicillin agar (Palumbo et al. 1985); dextrin, fuchsin, sulphite agar (Schubert 1987); starch, bile salts, brilliant green agar (Nishikawa & Kishi 1987)) or membrane filtration (mA medium (Rippey & Cabelli 1979); dextrin, ampicillin agar (Havelaar et ul. 1987)). Preliminary investigations (unpublished) comparing the recovery of two strains of A. hydrophilu-ATCC 7966 and M30-4, isolated from wastewater-n the selective media cited above, revealed that on mA and starch ampicillin agar (SA) the highest recoveries (97-99%) were obtained, both by plate count and membrane filtration. However, in the presence of other bacteria (Escherichia coli and Pseudomonas Juorescens) the recovery on mA (97-99%) was obviously better than on SA (8&89%). From these results we decided to use in the present study the mA method of Rippey 8i Cabelli (1979) for the quantitative estimation of A. hydrophila. The identification keys of Joseph et al. (1987) and Popoff & Viron (1976) were used to identify the organisms.
E N U M E R A T I O N OF AEROMONAS HYDROPHILA I N WATER
T h e object of this report was to estimate the number of A. hydrophila in raw sewage and its elimination by aerobic wastewater purification. Aeromonas hydrophila was also enumerated in surface waters, whether or not receiving raw sewage and effluents.
MATERIALS AND METHODS Water samples
Influent and effluent samples were collected from an activated sludge plant, with a capacity of 50000 population equivalents (PE = mg BOD discharged daily per capita) and from a trickling plant with a capacity of 25000 PE, each treating domestic sewage. Surface water samples were : (1) from a river crossing a city of 100000 inhabitants and receiving untreated raw sewage; samples were taken where the river enters and leaves the city; (2) from a river receiving raw sewage and secondary effluents; samples were taken in a rural area; and (3) from a canal for inland navigation receiving no sewage water and used for the preparation of drinking water. Samples were taken once a month from September to February. Influent and effluent samples were collected every 2 h during a period of 24 h ; separate samles were mixed in sterile 500 ml bottles. River and canal samples were taken with a sterile scoop 50 cm below the surface. After collection samples were transported in a refrigerated box at k 5°C and processed within 24 h. Sludge samples
Primary sludge from the activated sludge plant was thickened by settling. Samples were taken at the outlet of the sedimentation tank and collected in sterile 500 ml bottles. Primary and secondary sludge from the trickling filter plant was first stabilized by anaerobic methane fermentation (2 months) and thereafter dried on air in open drying beds (4 weeks). Samples were taken at 10 different places on the beds and the separate samples mixed in sterile 500 ml bottles. Total solids
About 10 g of sludge was evaporated to dryness in a tared dish on a waterbath (80°C) and further dried at 105°C for 1 h in an electric oven. The increase in weight over the empty dish represents the total solids and is expressed as per cent of wet sludge. Enumeration and characterization of Aeromonas hydrophlla
The numbers of A . hydrophila in influents, effluents and surface water samples were determined by membrane fil-
367
tration using the mA medium of Rippey 8i Cabelli (1979). This medium contained (g/l distilled water) : tryptose (Oxoid, L 47), 5; yeast extract (Oxoid, L 21), 2; trehalose, 5; NaCI, 3; KCI, 2; MgS0,.7H20, 0-2; FeC1,.6H20, 0.1 ; bromothymol blue, 0.04; agar, 15. T h e p H was adjusted to 8.0 with 10 N NaOH and the medium sterilized at 121°C for 15 min. After sterilization 10 ml ethanol were added, the medium cooled to 50°C and finally 20 mg of ampicillin and 100 mg of sodium deoxycholate were added. Although mannitol was a better source of carbohydrate for A. hydrophila, Rippey & Cabelli used trehalose in the primary medium because this carbohydrate is fermented by all strains of A. hydrophila and reduces the background contamination. Sodium deoxycholate increases the distinction between trehalose-fermenting and non-fermenting organisms. Ethanol inhibits the overgrowth of the medium by Klebsiella spp. Undiluted samples (100 ml) or 100 ml of a series of decimal dilutions, obtained by adding each time 50 ml to 450 ml sterile distilled water, were filtered on a membrane (Gelman Supor-200, pore size 0.2 pm. diameter 47 mm, Gelman Sciences Inc., MI, USA) with a 3-branch filtration system with 100 ml funnels (Sartorius, SM 16824, Gottingen, Germany). Funnels, supports and membranes were sterilized by autoclaving for 15 min at 121°C. Funnels and supports were decontaminated between samples by flaming. T h e filters were put on mA agar plates and incubated for 24 h at 37°C in an inverted position. At this temperature the non-motile A. salmonicida is inhibited. Sludge samples were not examined by membrane filtration because sludge particles could settle on the membrane and interfere with the colonies. About 10 g of wet sludge was homogenized in 100 ml of sterile saline using a Waring Blender. From this sludge suspension decimal dilutions were made in sterile water and from the adequate dilutions 0.1 ml was spread on mA with a Drigalski spatula. Suspect colonies of A . hydrophila on mA are yellow (trehalose-positive), convex, circular and 1-3 mm in diameter. They were further examined for mannitol fermentation and oxidase. Mannitol fermentation was tested by an ‘in sitd test. T h e membrane with the suspected colonies was transferred to the mannitol medium (Rippey & Cabelli 1979) and incubated again at 37°C for 2-3 h. Only the colonies remaining yellow (mannitol-positive) were scored a second time. Yellow colonies obtained from the sludge samples on mA plates were also examined for mannitol fermentation by streaking them on mannitol agar plates. From the number of colonies (n) showing a positive reaction for trehalose and mannitol either on membrane filters or on streak plates, were subcultured at random, transferred to Nutrient Agar slopes (Difco, 000101-8) and, after purification, examined for oxidase reaction with Bactident oxidase strips (Merck, 13300, Darmstadt, Germany): a little of the growth was removed with a loop
4
388 R . POFFE AND E. OP DE BEECK
Table 1 Characterization of presumptive Aeromonas hydrophila colonies, isolated from mA medium, by the identification tests of Popoff &
Vkon (1976) ~~
No. of co1onies
Gas from glucose
Acetoin from glucose
H,S from cystein
Fermentation of salicin
Aesculin hydrolysis
+ + -
+
+ + -
+ +
+ +
835 10
9 8
+ -
-
Characterization A. hydrophila A. sobria A. caviae Not identified
Various reactions
+
and spread on the reaction zone (1-naphthol dimethyl-pphenylene diamine), oxidase positive strains produce a blue colour in 20-60 s. Strains positive for trehalose, mannitol and oxidase were retained as presumptive A. hydrophila. Strains from presumptive A. hydrophila were further characterized by the identification key of Joseph et al. (1987). Differentiation between motile aeromonads was made according to characteristics described by Popoff 8t VCron (1976), including gas production from glucose, acetoin from glucose, H,S from cystein, fermentation of salicin and hydrolysis of aesculin. Strains that did not correspond to the appropriate characteristics were rejected and not counted as A. hydrophila. RESULTS
For the characterization of presumptive A. hydrophila a total of 862 colonies was examined. T h e results are shown
in Table 1 ; 835 colonies were identified as A. hydrophila, 9 as A. caviae and 10 as A. sobria; 8 colonies could not be characterized because diverse reaction patterns were found. T h e distribution of A. caviae and A. sobria in various water samples was as follows: surface water A. caviae 2, A. sobriu 4; influent A. caviae 4, A . sobria 3; effluent A. caviae 3, A. sobria 3. T h e number of A. hydrophila per ml water or per g sludge was calculated from the number of colonies identified as A. hydrophila. Results dealing with the enumeration of A. hydrophila in influents and effluents from domestic wastewater purification plants and relating to samples taken over a period of 6 months are summarized in Table 2. T h e y show that sewage samples from a purification plant of 50000 P E contained on average 2.9 x lo5 A. hydrophilalml, while sewage from a plant of 25000 PE contained about 10 times less. Between samples there were, of course, fluctuations as indicated by standard deviations but they never differed more
Table 2 Enumeration of Aeromonas hydrophila in influents and secondary emuents from domestic wastewater purification plants
Purification system
Capacity (PE)
Sample
No. of A. hydrophilalml
Activated sludge
50 000
Trickling filter
25 OOO
Influent Secondary effluent Influent Secondary effluent
2.9 x 105*(1.3 x 104)t 75 (4) 2.4 x 104 (2 x 103) 420 (8)
Reduction (%) 99.975 98.25
* Numbers are the arithmetic means of six samples, taken monthly from September to February. Each sample was examined in triplicate.
t Figures in parentheses give the standard deviation. PE, Population equivalents.
Table 3 Enumeration of Aeromonas hydrophila in sludge from domestic wastewater purification plants
No. of A. hydrophilal dry sludge
Purification system
Sample
Dry matter (YO)
g
Activated sludge Trickling filter
Thickened fresh primary sludge Anaerobically fermented and partially dried sludge
5.84* (0.59)t 11.97 (1.09)
4.9 x 10’ (3.9 x 106) 2.6 x lo6 (0.26 x lo6)
* Numbers are the arithmetic means of six samples, taken monthly from September to February. Each sample was examined in triplicate.
t Figures in parentheses give the standard deviation.
ENUMERATION OF A E R O M O N A S H Y D R O P H I L A I N WATER
Table 4 Enumeration of Aeromonas hydrophila in surface waters
Sample River in
No. of A. hydrophila/ml*
It
out
River 2$ Canals
150 (15)11 450 (32) 370 (24) 4 (0)
* Numbers are the arithmetic means of six samples, taken monthly from September to February. t River 1 receives raw, non-treated sewage and flows through a city of 100000 inhabitants. Samples were taken where the river enters (in) and leaves the town (out). $ River 2 receives raw sewage and secondary effluents. Samples were taken in a rural area. 5 Canal for inland navigation, receiving neither raw sewage nor effluents and used for the preparation of drinking water. 11 Figures in parentheses give the standard deviation. than lo%, indicating that there were no variations between the numbers of A . hydrophila in the samples taken during the autumn and winter months. The sewage of the first plant, with the highest A . hydrophila content, was purified by activated sludge treatment. In the secondary effluent on average less than 102/ml A . hydrophila were counted, giving a reduction of 99.975%. The sewage of the small plant was treated by trickling filters: it was ascertained that, although the A . hydrophila density in the influent was 10 times lower than in that of the activated sludge plant, there were about six times more in the secondary effluent, giving a reduction of only 98.25%. This suggests that A . hydrophila was more readily removed by activated sludge than by trickling filters. Nevertheless, the number of A . hydrophila discharged in surface water via secondary effluents was in both cases low. Since during aerobic wastewater treatment about 99% of A . hydrophila was removed, one can expect that most of these bacteria end up in the primary and secondary sludge. Fresh primary sludge, thickened by settling, with a dry matter content of 5.84% on average, was examined for the number of A . hydrophila as shown in Table 3. On average 4.9 x 107/g dry sludge were found. With a dry matter content of approximately 0.12% in the influent, the theoretical number of A . hydrophilalg dry sludge is 2.4 x lo*, suggesting that only 20.9% of these bacteria were removed in the primary sludge. In anaerobically (methane) fermented and partially dried sludge from trickling filters, with an average age of 3 months, still more than lo6 A. hydrophilalg dry sludge were counted, showing that even under unfavourable conditions these bacteria can survive for a long time. In our country non-treated raw sewage is sometimes discharged in river water, especially in densely populated
369
areas, causing considerable pollution. Table 4 shows the numbers of A . hydrophila counted in a river flowing through a town of 100000 inhabitants. Samples were taken where the river enters and leaves the city. At the entrance the number of A. hydrophila was low (on average 150/ml), but there was an increase in number (450/ml) after passage through the city. T h e numbers of A . hydrophila found in another river, also receiving raw sewage but flowing through a rural area, were comparable with those found in the populated area. On the other hand, water from a canal for inland navigation, receiving neither raw sewage nor secondary effluents and destined for the preparation of drinking water, contained very low numbers of A . hydrophila (on average 4/ml). DISCUSSION
This study has demonstrated that from the presumptive colonies developed on mA 968% were identified as A. hydrophila; A . sobria and A . caviae were only sporadically found. Aeromonas hydrophila was present in all water samples; large numbers were present in raw sewage and influents of domestic water purification plants. More than 99.9% were removed by activated sludge treatment and about 98% by trickling filters. Havelaar et al. (1987) also examined influents and effluents for the presence of A . hydrophila. They found between lo5 and lo6 cfu/ml in raw sewage and on average 104/ml in secondary effluents. Although the object of their investigations was not to determine how far A . hydrophila was removed by water purification it can be calculated from the figures cited that a reduction of approximately 90% was obtained, which was comparable with the reduction by trickling filters found in our experiments. Their results, however, can hardly be compared with ours since Havelaar did not mention whether the secondary effluents were obtained by activated sludge or by other wastewater purification systems. Although there is no indication why A . hydrophila was more easily removed from domestic wastewater by activated sludge than by trickling filters, analogous reduction percentages have been observed previously by Poffk et al. (1973) for Enterobacteriaceae. On average these bacteria were 87.1% reduced by trickling filters and 99.8% by activated sludge. Water purification not only reduces the number of Aeromonas spp. but also the level of nutrients. Rippey 8i Cabelli (1979) suggested that high nutrient loadings promote the multiplication of A . hydrophila in surface water. Consequently the discharge of raw sewage in river water can favour the development of these bacteria. These suggestions were confirmed by our findings : river water receiving raw sewage contained several hundreds per ml, comparable with the number of A . hydrophila found in efluents from
370 R . POFFE AND E. OP DE BEECK
domestic wastewater purification plants, while surface waters not polluted by raw sewage contained only a small and negligible number of these bacteria. Thus, A. hydrophila is always present in surface water, but high numbers are associated with nutrient loadings emanating from raw sewage. T h e high numbers of A. hydrophila found after 3 months in anaerobically methane-fermented sludge show that these bacteria can also survive for a long time in the presence of nutrients.
REFERENCES A G G E R W.A., , M C C O R M I C KJ ,. D . & G U R W I T HM , .J. (1985) Clinical and microbiological features of Aeromonass hydrophila associated diarrhoea. Journal of Clinical Microbiology 21,909-913. B O U L A N G EYR. , L A L L I E RR. , & COUSINEAU G ,. (1977) Isolation of enterotoxigenic Aeromonas from fish. Canadian Journal of Microbiology 23, 1161-1 164. B U R K EV , . , R O B I N S O NJ ., , G R A C E YM , . , P E T E R S O ND, . & P A R T R I D GK. E , (1984a) Isolation of Aeromonas hydrophila from a metropolitan water supply : seasonal correlation with clinical isolates. Applied and Environmental Microbiology 48, 361-366. B U R K EV, . , R O B I N S O NJ ,. , G R A C E YM, . , P E T E R S O ND, . , M E Y E RN , . & H A L E Y ,V . (1984b) Isolation of Aeromonas spp. from an unchlorinated domestic water supply. Applied and Environmental Microbiology 48, 367-370. C A V A R IB, . Z . , A L L E N , D . A . & C O L W E L LR, . R . (1981) Effect of temperature on growth and activity of Aeromonas spp. and mixed bacterial populations in the Anacostia River. Applied and Environmental Microbiology 41, 1052-1054. F L I E R M A N SC,. B . , G O R D E N ,R . W . , H A Z E NT, . C . & ESCH, G . W. (1977) Aeromonas distribution and survival in a thermally altered lake. Applied and Environmental Microbiology 33,114-122. F U L G H U MD,. D . , L I N T O N W , .R. & T A P L I N D., (1978) Fatal Aeromonas h.ydrophila infection of the skin. Southern Medical Journal 71,739-741. GRACEY M, . , B U R K EV, . & R O B I N S O NJ ., (1982) Aeromonas associated gastro-enteritis. Lancer 2, 1304-1306. H A V E L A AAR. ,H . , D U R I N G M , . & V E R S T E E G H ,J . F . M . (1987) Ampicillin-dextrin agar medium for the enumeration of Aeromonas species in water by membrane filtration. Journal of Applied Bacteriology 62, 279-287. J O S E P H ,S . W . , C O L W E L LR.R. , & M A C D O N E L LM, . T . ( 1987) Aeromonas taxonomy. Experientia 43, 349-350.
K I E L w E I N , G . (1969) Ein nahrboden zur selektiven Zuchtung von Pseudomonaden und Aeromonaden. Archiv Lebensmittelhygiene 20, 131-133. K U I J P E R , E . J . , V A N D E W I L D E N B E R GA , .H.B.M., P E E T E R S ,M . F . , M U Y T J E N S ,H . L . , S T U I F B E R G E N , W.N.H.M., TEEPEN J .,L . J . M . & J A N S Z , A. (1984) Infecties door Aeromonas hydrophila. Nederlands T c d s c h r ~ tvoor Geneeskunde 1 2 4 2 1&2 13. M c C o y , R . H . & P I L C H E RK.S. , (1974) Peptone beefextract glycogen agar, a selective and differential Aeromonas medium. Journal of the Fishery Research Board of Canada 31, 1553-1555. NISHIKAWA Y ,. & K I S H I , T . (1987) A modification of bile salts brilliant green agar for the isolation of motile Aeromonas from foods and environmental specimens. Epidemiology and Infection 98, 331-336. P A L U M B O ,S.A., M A X I N O , F . , W I L L I A M S , A . C . , D ,. W . (1985) StarchB U C H A N A NR, . L . & T H A Y E R ampicillin agar for the quantitative detection of Aeromonas hydrophila. Applied and Environmental Microbiology 50, 10271030. P O F F ~R, . , B E V E R SJ,. , D U M O N ,P. & V E R A C H T E RHT., (1973) Orienterend chemisch-microbiologisch onderzoek naar de effectiviteit van de biologische afvalwaterzuivering in Belgie. H 2 0 6,497-504. POPOFF,M . & V E R O N , M . (1976) A taxonomic study of the Aeromonas hydrophila-Aeromonas punctata group. Journal of General Microbiology 94, 11-22. R E I N E SH , . D . & C O O K ,F . V . (1981) Pneumonia and bacteremia due to Aeromonas hydrophila. Chest 80, 264-267. R I P P E Y ,S . R . & C A B E L L IV, . J . (1979) Membrane filter procedure for enumeration of Aeromonas hydrophila in fresh waters. Applied and Environmental Microbiology 38, 108-1 13. S C H U B E R TR, . H . W . (1987) Ecology of Aeromonads and isolation from environmental samples. Experientia 43, 35 1-354. S E I D L E RR, . J . , A L L E N , D . A . , L O C K M A N H,. , C O L W E L L , R . R . , J O S E P H ,S . W . & D A I L Y ,G . J . (1980) Isolation, enumeration and characterization of Aeromonas from polluted waters encountered in diving operations. Applied and Environmental Microbiology 39, 101&1018. S H O T T SE, . B . & R I M L E RR, . (1973) Medium for the isolation of Aeromonas hydrophila. Applied Microbiology 26, 55&553. S H O T T S , E . B . , G A I N E S ,J . L . , M A R T I N ,C . & P R E S T W O O D , A.K. (1972) Aeromonas induced deaths among fish and reptiles in an eutrophic inland lake. Journal of the American Veterinary Medical Association 161, 603-607. V O N G R A E V E N I TAZ. , & B U C H E RC , . (1983) Aeromonas and Plesromonas. In Manual of Clinical Microbiology ed. Lenette, E.H., Balows, A., Hausler, W.I. & Truant, J.P. pp. 22&225. Washington, DC: American Society for Microbiology.
ADONIS 002188479100 1358
Enumeration of Aeromonas hydrophila from domestic wastewater treatment plants and surface waters R. Poffe and E. Op de Beeck Laboratory of Industrial Microbiology, Faculty of Agricultural Sciences, Catholic University of Leuven, Belgium 3571/01/91:accepted 1 1 March 1991
R. POFFE AND E. O P DE BEECK. 1991.Influents, effluents and sludges from sewage purification plants and surface water samples were examined quantitatively for Aeromonas hydrophila on the mA medium of Rippey a n d Cabelli. Between lo4 and 106/ml A. hydrophila were found in domestic wastewaters. O n t h e average 99.975 % were removed by activated sludge a n d 98.25% by trickling filters. Only 20.9% of A. hydrophila end u p in the primary sludge, which contained up to 107/g dry sludge. After 3 months, anaerobically (methane) fermented a n d partially dried sludge from trickling filters contained more than lo6 A. hydrophilalg d r y sludge. Surface water receiving raw sewage contained several hundreds of A. hydrophila/ml, comparable with the numbers found in effluent waters, while surface water receiving n o municipal wastewater and destined for t h e preparation of drinking water contained only small and negligible numbers. It was concluded that A. hydrophila was omnipresent in surface water.
INTRODUCTION
Aeromonas hydrophila is generally considered to be a resident aquatic micro-organism. It is reported as part of the normal flora of fish (Shotts et al. 1972; Boulanger et al. 1977) and the agent of epizootic outbreaks among aquatic organisms (Shotts et al. 1972). Interest in Aeromonas spp. has increased in recent years since A. hydrophila has been associated with human diseases (Kuijper et al. 1984), including gastroenteritis (Gracey et al. 1982; Agger et .al. 1985), cellulitis (Fulghum et al. 1978), peritonitis, meningitis (von Graevenitz & Bucher 1983), pneumonia and bacteraemia (Reines & Cook 1981). It is not a ‘true’ enteric bacterium, since it was rarely recovered from human intestine (Rippey & Cabelli 1979). Water environments have been cited many times as the major source of infection: surface waters (Fliermans et al. 1977; Seidler et al. 1980; Cavari et al. 1981), raw and drinking water supplies (Burke et al. 1984a, b), sewage and effluents from domestic water purification plants (Rippey & Cabelli 1979; Havelaar et al. 1987). From the scanty information about A. hydrophila in sewage purification found in the literature it appeared that high densities were consistently present in sewage, suggesting that wastewaters with high nutrient loadings are excelCorrespondence to : Ir R. Poffe, Laboratory of Industrial Microbiology, Faculty of Agricultural Sciences, K. U.L c u v n , Kardinaal Mercierlaan 92, B-3001 Heverlee, Leuven, Belgium.
lent sites of multiplication (Rippey & Cabelli 1979; Schubert 1987). Increased interest in Aeromonas has led to the development of many selective media for its detection and enumeration either by plate count (starch, glutamate, penicillin agar (Kielwein 1969); Shotts & Rimler agar (Shotts & Rimler 1973); glycogen, peptone, beef extract agar (McCoy & Pilcher 1974); starch, ampicillin agar (Palumbo et al. 1985); dextrin, fuchsin, sulphite agar (Schubert 1987); starch, bile salts, brilliant green agar (Nishikawa & Kishi 1987)) or membrane filtration (mA medium (Rippey & Cabelli 1979); dextrin, ampicillin agar (Havelaar et ul. 1987)). Preliminary investigations (unpublished) comparing the recovery of two strains of A. hydrophilu-ATCC 7966 and M30-4, isolated from wastewater-n the selective media cited above, revealed that on mA and starch ampicillin agar (SA) the highest recoveries (97-99%) were obtained, both by plate count and membrane filtration. However, in the presence of other bacteria (Escherichia coli and Pseudomonas Juorescens) the recovery on mA (97-99%) was obviously better than on SA (8&89%). From these results we decided to use in the present study the mA method of Rippey 8i Cabelli (1979) for the quantitative estimation of A. hydrophila. The identification keys of Joseph et al. (1987) and Popoff & Viron (1976) were used to identify the organisms.
E N U M E R A T I O N OF AEROMONAS HYDROPHILA I N WATER
T h e object of this report was to estimate the number of A. hydrophila in raw sewage and its elimination by aerobic wastewater purification. Aeromonas hydrophila was also enumerated in surface waters, whether or not receiving raw sewage and effluents.
MATERIALS AND METHODS Water samples
Influent and effluent samples were collected from an activated sludge plant, with a capacity of 50000 population equivalents (PE = mg BOD discharged daily per capita) and from a trickling plant with a capacity of 25000 PE, each treating domestic sewage. Surface water samples were : (1) from a river crossing a city of 100000 inhabitants and receiving untreated raw sewage; samples were taken where the river enters and leaves the city; (2) from a river receiving raw sewage and secondary effluents; samples were taken in a rural area; and (3) from a canal for inland navigation receiving no sewage water and used for the preparation of drinking water. Samples were taken once a month from September to February. Influent and effluent samples were collected every 2 h during a period of 24 h ; separate samles were mixed in sterile 500 ml bottles. River and canal samples were taken with a sterile scoop 50 cm below the surface. After collection samples were transported in a refrigerated box at k 5°C and processed within 24 h. Sludge samples
Primary sludge from the activated sludge plant was thickened by settling. Samples were taken at the outlet of the sedimentation tank and collected in sterile 500 ml bottles. Primary and secondary sludge from the trickling filter plant was first stabilized by anaerobic methane fermentation (2 months) and thereafter dried on air in open drying beds (4 weeks). Samples were taken at 10 different places on the beds and the separate samples mixed in sterile 500 ml bottles. Total solids
About 10 g of sludge was evaporated to dryness in a tared dish on a waterbath (80°C) and further dried at 105°C for 1 h in an electric oven. The increase in weight over the empty dish represents the total solids and is expressed as per cent of wet sludge. Enumeration and characterization of Aeromonas hydrophlla
The numbers of A . hydrophila in influents, effluents and surface water samples were determined by membrane fil-
367
tration using the mA medium of Rippey 8i Cabelli (1979). This medium contained (g/l distilled water) : tryptose (Oxoid, L 47), 5; yeast extract (Oxoid, L 21), 2; trehalose, 5; NaCI, 3; KCI, 2; MgS0,.7H20, 0-2; FeC1,.6H20, 0.1 ; bromothymol blue, 0.04; agar, 15. T h e p H was adjusted to 8.0 with 10 N NaOH and the medium sterilized at 121°C for 15 min. After sterilization 10 ml ethanol were added, the medium cooled to 50°C and finally 20 mg of ampicillin and 100 mg of sodium deoxycholate were added. Although mannitol was a better source of carbohydrate for A. hydrophila, Rippey & Cabelli used trehalose in the primary medium because this carbohydrate is fermented by all strains of A. hydrophila and reduces the background contamination. Sodium deoxycholate increases the distinction between trehalose-fermenting and non-fermenting organisms. Ethanol inhibits the overgrowth of the medium by Klebsiella spp. Undiluted samples (100 ml) or 100 ml of a series of decimal dilutions, obtained by adding each time 50 ml to 450 ml sterile distilled water, were filtered on a membrane (Gelman Supor-200, pore size 0.2 pm. diameter 47 mm, Gelman Sciences Inc., MI, USA) with a 3-branch filtration system with 100 ml funnels (Sartorius, SM 16824, Gottingen, Germany). Funnels, supports and membranes were sterilized by autoclaving for 15 min at 121°C. Funnels and supports were decontaminated between samples by flaming. T h e filters were put on mA agar plates and incubated for 24 h at 37°C in an inverted position. At this temperature the non-motile A. salmonicida is inhibited. Sludge samples were not examined by membrane filtration because sludge particles could settle on the membrane and interfere with the colonies. About 10 g of wet sludge was homogenized in 100 ml of sterile saline using a Waring Blender. From this sludge suspension decimal dilutions were made in sterile water and from the adequate dilutions 0.1 ml was spread on mA with a Drigalski spatula. Suspect colonies of A . hydrophila on mA are yellow (trehalose-positive), convex, circular and 1-3 mm in diameter. They were further examined for mannitol fermentation and oxidase. Mannitol fermentation was tested by an ‘in sitd test. T h e membrane with the suspected colonies was transferred to the mannitol medium (Rippey & Cabelli 1979) and incubated again at 37°C for 2-3 h. Only the colonies remaining yellow (mannitol-positive) were scored a second time. Yellow colonies obtained from the sludge samples on mA plates were also examined for mannitol fermentation by streaking them on mannitol agar plates. From the number of colonies (n) showing a positive reaction for trehalose and mannitol either on membrane filters or on streak plates, were subcultured at random, transferred to Nutrient Agar slopes (Difco, 000101-8) and, after purification, examined for oxidase reaction with Bactident oxidase strips (Merck, 13300, Darmstadt, Germany): a little of the growth was removed with a loop
4
388 R . POFFE AND E. OP DE BEECK
Table 1 Characterization of presumptive Aeromonas hydrophila colonies, isolated from mA medium, by the identification tests of Popoff &
Vkon (1976) ~~
No. of co1onies
Gas from glucose
Acetoin from glucose
H,S from cystein
Fermentation of salicin
Aesculin hydrolysis
+ + -
+
+ + -
+ +
+ +
835 10
9 8
+ -
-
Characterization A. hydrophila A. sobria A. caviae Not identified
Various reactions
+
and spread on the reaction zone (1-naphthol dimethyl-pphenylene diamine), oxidase positive strains produce a blue colour in 20-60 s. Strains positive for trehalose, mannitol and oxidase were retained as presumptive A. hydrophila. Strains from presumptive A. hydrophila were further characterized by the identification key of Joseph et al. (1987). Differentiation between motile aeromonads was made according to characteristics described by Popoff 8t VCron (1976), including gas production from glucose, acetoin from glucose, H,S from cystein, fermentation of salicin and hydrolysis of aesculin. Strains that did not correspond to the appropriate characteristics were rejected and not counted as A. hydrophila. RESULTS
For the characterization of presumptive A. hydrophila a total of 862 colonies was examined. T h e results are shown
in Table 1 ; 835 colonies were identified as A. hydrophila, 9 as A. caviae and 10 as A. sobria; 8 colonies could not be characterized because diverse reaction patterns were found. T h e distribution of A. caviae and A. sobria in various water samples was as follows: surface water A. caviae 2, A. sobriu 4; influent A. caviae 4, A . sobria 3; effluent A. caviae 3, A. sobria 3. T h e number of A. hydrophila per ml water or per g sludge was calculated from the number of colonies identified as A. hydrophila. Results dealing with the enumeration of A. hydrophila in influents and effluents from domestic wastewater purification plants and relating to samples taken over a period of 6 months are summarized in Table 2. T h e y show that sewage samples from a purification plant of 50000 P E contained on average 2.9 x lo5 A. hydrophilalml, while sewage from a plant of 25000 PE contained about 10 times less. Between samples there were, of course, fluctuations as indicated by standard deviations but they never differed more
Table 2 Enumeration of Aeromonas hydrophila in influents and secondary emuents from domestic wastewater purification plants
Purification system
Capacity (PE)
Sample
No. of A. hydrophilalml
Activated sludge
50 000
Trickling filter
25 OOO
Influent Secondary effluent Influent Secondary effluent
2.9 x 105*(1.3 x 104)t 75 (4) 2.4 x 104 (2 x 103) 420 (8)
Reduction (%) 99.975 98.25
* Numbers are the arithmetic means of six samples, taken monthly from September to February. Each sample was examined in triplicate.
t Figures in parentheses give the standard deviation. PE, Population equivalents.
Table 3 Enumeration of Aeromonas hydrophila in sludge from domestic wastewater purification plants
No. of A. hydrophilal dry sludge
Purification system
Sample
Dry matter (YO)
g
Activated sludge Trickling filter
Thickened fresh primary sludge Anaerobically fermented and partially dried sludge
5.84* (0.59)t 11.97 (1.09)
4.9 x 10’ (3.9 x 106) 2.6 x lo6 (0.26 x lo6)
* Numbers are the arithmetic means of six samples, taken monthly from September to February. Each sample was examined in triplicate.
t Figures in parentheses give the standard deviation.
ENUMERATION OF A E R O M O N A S H Y D R O P H I L A I N WATER
Table 4 Enumeration of Aeromonas hydrophila in surface waters
Sample River in
No. of A. hydrophila/ml*
It
out
River 2$ Canals
150 (15)11 450 (32) 370 (24) 4 (0)
* Numbers are the arithmetic means of six samples, taken monthly from September to February. t River 1 receives raw, non-treated sewage and flows through a city of 100000 inhabitants. Samples were taken where the river enters (in) and leaves the town (out). $ River 2 receives raw sewage and secondary effluents. Samples were taken in a rural area. 5 Canal for inland navigation, receiving neither raw sewage nor effluents and used for the preparation of drinking water. 11 Figures in parentheses give the standard deviation. than lo%, indicating that there were no variations between the numbers of A . hydrophila in the samples taken during the autumn and winter months. The sewage of the first plant, with the highest A . hydrophila content, was purified by activated sludge treatment. In the secondary effluent on average less than 102/ml A . hydrophila were counted, giving a reduction of 99.975%. The sewage of the small plant was treated by trickling filters: it was ascertained that, although the A . hydrophila density in the influent was 10 times lower than in that of the activated sludge plant, there were about six times more in the secondary effluent, giving a reduction of only 98.25%. This suggests that A . hydrophila was more readily removed by activated sludge than by trickling filters. Nevertheless, the number of A . hydrophila discharged in surface water via secondary effluents was in both cases low. Since during aerobic wastewater treatment about 99% of A . hydrophila was removed, one can expect that most of these bacteria end up in the primary and secondary sludge. Fresh primary sludge, thickened by settling, with a dry matter content of 5.84% on average, was examined for the number of A . hydrophila as shown in Table 3. On average 4.9 x 107/g dry sludge were found. With a dry matter content of approximately 0.12% in the influent, the theoretical number of A . hydrophilalg dry sludge is 2.4 x lo*, suggesting that only 20.9% of these bacteria were removed in the primary sludge. In anaerobically (methane) fermented and partially dried sludge from trickling filters, with an average age of 3 months, still more than lo6 A. hydrophilalg dry sludge were counted, showing that even under unfavourable conditions these bacteria can survive for a long time. In our country non-treated raw sewage is sometimes discharged in river water, especially in densely populated
369
areas, causing considerable pollution. Table 4 shows the numbers of A . hydrophila counted in a river flowing through a town of 100000 inhabitants. Samples were taken where the river enters and leaves the city. At the entrance the number of A. hydrophila was low (on average 150/ml), but there was an increase in number (450/ml) after passage through the city. T h e numbers of A . hydrophila found in another river, also receiving raw sewage but flowing through a rural area, were comparable with those found in the populated area. On the other hand, water from a canal for inland navigation, receiving neither raw sewage nor secondary effluents and destined for the preparation of drinking water, contained very low numbers of A . hydrophila (on average 4/ml). DISCUSSION
This study has demonstrated that from the presumptive colonies developed on mA 968% were identified as A. hydrophila; A . sobria and A . caviae were only sporadically found. Aeromonas hydrophila was present in all water samples; large numbers were present in raw sewage and influents of domestic water purification plants. More than 99.9% were removed by activated sludge treatment and about 98% by trickling filters. Havelaar et al. (1987) also examined influents and effluents for the presence of A . hydrophila. They found between lo5 and lo6 cfu/ml in raw sewage and on average 104/ml in secondary effluents. Although the object of their investigations was not to determine how far A . hydrophila was removed by water purification it can be calculated from the figures cited that a reduction of approximately 90% was obtained, which was comparable with the reduction by trickling filters found in our experiments. Their results, however, can hardly be compared with ours since Havelaar did not mention whether the secondary effluents were obtained by activated sludge or by other wastewater purification systems. Although there is no indication why A . hydrophila was more easily removed from domestic wastewater by activated sludge than by trickling filters, analogous reduction percentages have been observed previously by Poffk et al. (1973) for Enterobacteriaceae. On average these bacteria were 87.1% reduced by trickling filters and 99.8% by activated sludge. Water purification not only reduces the number of Aeromonas spp. but also the level of nutrients. Rippey 8i Cabelli (1979) suggested that high nutrient loadings promote the multiplication of A . hydrophila in surface water. Consequently the discharge of raw sewage in river water can favour the development of these bacteria. These suggestions were confirmed by our findings : river water receiving raw sewage contained several hundreds per ml, comparable with the number of A . hydrophila found in efluents from
370 R . POFFE AND E. OP DE BEECK
domestic wastewater purification plants, while surface waters not polluted by raw sewage contained only a small and negligible number of these bacteria. Thus, A. hydrophila is always present in surface water, but high numbers are associated with nutrient loadings emanating from raw sewage. T h e high numbers of A. hydrophila found after 3 months in anaerobically methane-fermented sludge show that these bacteria can also survive for a long time in the presence of nutrients.
REFERENCES A G G E R W.A., , M C C O R M I C KJ ,. D . & G U R W I T HM , .J. (1985) Clinical and microbiological features of Aeromonass hydrophila associated diarrhoea. Journal of Clinical Microbiology 21,909-913. B O U L A N G EYR. , L A L L I E RR. , & COUSINEAU G ,. (1977) Isolation of enterotoxigenic Aeromonas from fish. Canadian Journal of Microbiology 23, 1161-1 164. B U R K EV , . , R O B I N S O NJ ., , G R A C E YM , . , P E T E R S O ND, . & P A R T R I D GK. E , (1984a) Isolation of Aeromonas hydrophila from a metropolitan water supply : seasonal correlation with clinical isolates. Applied and Environmental Microbiology 48, 361-366. B U R K EV, . , R O B I N S O NJ ,. , G R A C E YM, . , P E T E R S O ND, . , M E Y E RN , . & H A L E Y ,V . (1984b) Isolation of Aeromonas spp. from an unchlorinated domestic water supply. Applied and Environmental Microbiology 48, 367-370. C A V A R IB, . Z . , A L L E N , D . A . & C O L W E L LR, . R . (1981) Effect of temperature on growth and activity of Aeromonas spp. and mixed bacterial populations in the Anacostia River. Applied and Environmental Microbiology 41, 1052-1054. F L I E R M A N SC,. B . , G O R D E N ,R . W . , H A Z E NT, . C . & ESCH, G . W. (1977) Aeromonas distribution and survival in a thermally altered lake. Applied and Environmental Microbiology 33,114-122. F U L G H U MD,. D . , L I N T O N W , .R. & T A P L I N D., (1978) Fatal Aeromonas h.ydrophila infection of the skin. Southern Medical Journal 71,739-741. GRACEY M, . , B U R K EV, . & R O B I N S O NJ ., (1982) Aeromonas associated gastro-enteritis. Lancer 2, 1304-1306. H A V E L A AAR. ,H . , D U R I N G M , . & V E R S T E E G H ,J . F . M . (1987) Ampicillin-dextrin agar medium for the enumeration of Aeromonas species in water by membrane filtration. Journal of Applied Bacteriology 62, 279-287. J O S E P H ,S . W . , C O L W E L LR.R. , & M A C D O N E L LM, . T . ( 1987) Aeromonas taxonomy. Experientia 43, 349-350.
K I E L w E I N , G . (1969) Ein nahrboden zur selektiven Zuchtung von Pseudomonaden und Aeromonaden. Archiv Lebensmittelhygiene 20, 131-133. K U I J P E R , E . J . , V A N D E W I L D E N B E R GA , .H.B.M., P E E T E R S ,M . F . , M U Y T J E N S ,H . L . , S T U I F B E R G E N , W.N.H.M., TEEPEN J .,L . J . M . & J A N S Z , A. (1984) Infecties door Aeromonas hydrophila. Nederlands T c d s c h r ~ tvoor Geneeskunde 1 2 4 2 1&2 13. M c C o y , R . H . & P I L C H E RK.S. , (1974) Peptone beefextract glycogen agar, a selective and differential Aeromonas medium. Journal of the Fishery Research Board of Canada 31, 1553-1555. NISHIKAWA Y ,. & K I S H I , T . (1987) A modification of bile salts brilliant green agar for the isolation of motile Aeromonas from foods and environmental specimens. Epidemiology and Infection 98, 331-336. P A L U M B O ,S.A., M A X I N O , F . , W I L L I A M S , A . C . , D ,. W . (1985) StarchB U C H A N A NR, . L . & T H A Y E R ampicillin agar for the quantitative detection of Aeromonas hydrophila. Applied and Environmental Microbiology 50, 10271030. P O F F ~R, . , B E V E R SJ,. , D U M O N ,P. & V E R A C H T E RHT., (1973) Orienterend chemisch-microbiologisch onderzoek naar de effectiviteit van de biologische afvalwaterzuivering in Belgie. H 2 0 6,497-504. POPOFF,M . & V E R O N , M . (1976) A taxonomic study of the Aeromonas hydrophila-Aeromonas punctata group. Journal of General Microbiology 94, 11-22. R E I N E SH , . D . & C O O K ,F . V . (1981) Pneumonia and bacteremia due to Aeromonas hydrophila. Chest 80, 264-267. R I P P E Y ,S . R . & C A B E L L IV, . J . (1979) Membrane filter procedure for enumeration of Aeromonas hydrophila in fresh waters. Applied and Environmental Microbiology 38, 108-1 13. S C H U B E R TR, . H . W . (1987) Ecology of Aeromonads and isolation from environmental samples. Experientia 43, 35 1-354. S E I D L E RR, . J . , A L L E N , D . A . , L O C K M A N H,. , C O L W E L L , R . R . , J O S E P H ,S . W . & D A I L Y ,G . J . (1980) Isolation, enumeration and characterization of Aeromonas from polluted waters encountered in diving operations. Applied and Environmental Microbiology 39, 101&1018. S H O T T SE, . B . & R I M L E RR, . (1973) Medium for the isolation of Aeromonas hydrophila. Applied Microbiology 26, 55&553. S H O T T S , E . B . , G A I N E S ,J . L . , M A R T I N ,C . & P R E S T W O O D , A.K. (1972) Aeromonas induced deaths among fish and reptiles in an eutrophic inland lake. Journal of the American Veterinary Medical Association 161, 603-607. V O N G R A E V E N I TAZ. , & B U C H E RC , . (1983) Aeromonas and Plesromonas. In Manual of Clinical Microbiology ed. Lenette, E.H., Balows, A., Hausler, W.I. & Truant, J.P. pp. 22&225. Washington, DC: American Society for Microbiology.
