- Email: [email protected]
Relative Frequencies and Significance of Faecal Coliforms as Indicators Related to Water Temperature
Zbl. Mikrobiol. 138 (1983),329-336 [Department of Microbiology, Faculty of Biological Sciences, Univ. of Valencia, Burjasot (Valencia), Spain]
Relative Frequencies and Significance of Faecal Coliforms as Indicators Related to Water Temperature E. GARAY AUBAN, A. ARNAU RIPOLLES, and M.• J. PUJALTE DOMARco With 4 Figures
Summary The faecal eoliforms at different sites of a hypereutrophic lake near Valencia (Albufera) were identified and their relative amounts established along an annual cycle. Using lauryl tryptose broth at 35 DC, followed by incubation at 44.4 DC in 2 % brilliant green bile, Escherichia coli and Klebsiella pneumoniae are practically the only coliforms present. A positive correlation was found between the water temperature and the relative amount of these two coliforms: K. pneumoniae predominates at high water temperatures, whereas E. coli shows preponderance during the cold period. The role of K. pneumoniae as the only faecal indicator under the circumstances described in the work is emphasized and discussed.
Zusammenfassung An verschiedenen Stellen eines hyper-eutrophierten Sees in der Niihe von Valencia (Albufera) wurden die fakalon Coliformen identifiziert und deren relative Menge ein Jahr lang untersucht. Folgende Niihrbiiden wurden verwendet: Laurylsulfat-tryptose, bebriitet bei 35 DC, gefolgt von der Kultivierung in Brilliant-green-bile bei 44,5 °C. Unter diesen Bedingungen konnten fast ausschliel3lich Escherichia coli und Klebsiella pneumoniae nachgewiesen werden. Es wurde cine signifikante Korrelation zwischen Wassertemperatur und relativer Haufigkeit der beiden Colifor-rueu registriert. K. pneumoniae dorninier-t bei hohen "'Vassertelliperaturen, wahrond niedrige Temperaturen ein Dberwiegen von E. coli hervorrufen. Auf die Bedeutung von K. pneumoniae als einzigen fak alcrn Indikator unt.e r den genannten Bedingungen wird hingewiesen.
The term faecal coliform (FC) comprises all coliforms that produce gas from lactose at 44.5 °C, trying to separate the more ubiquitous coliforms from those of true faecal origin (DOCKINS and McFE'rERS 1978). Although several members of the Enterobacteriaceae are able to produce gas at this temperature (A. P. H. A. 1975, BAGLEY and SEIDLER 1977, THAM and DANIELSSON 1980) and should therefore be considered Escherichia coli, there is no general agreement whether only this microorganism or other Enterobacteriaceae-designed faecal coliforms, too, have the same significance when isolated from different sources. Some workers still consider E. coli to be the only faecal coliform in water (SHINGARA et al. 1979), because other FC-positive microorganisms can survive for a long time or even colonize several environments (DUNCAN and RAZELL 1972, CAMPBELL et al. 1976, BAGLEY and SEIDLER 1977, GAVINI et al. 1977, KNITTEL et al. 1977, NAEMURA and SEIDLER 1978, SPLITTs'roEssER 1979). The question arises, therefore, when the possible indicators are not exclusively of faecal origin and can colonize diverse habitats. 22*
330
E. G.
AUBAN
et al.
In our work we have tried to ascertain the relative frequency of the FC-positive organisms in surface waters with different contamination levels. We have studied the evolution throughout a year at several sites of a hypereutrophic lake, which is fed from numerous channels with wastewater, covering the very origin (at the discharging point) to the center of the lake and the site where the lake water flows into the Mediterranean sea. Material and Methods 1. Sampling sites All sampling sites, with one exception, correspond to Albufera lake near Valencia (Fig. 1). This lake is located 12 km to the South of Valencia, close to the sea, from which it is separated by a smalllittorial bar of approximately 1 km width. Three channels ("golas") connect the lake with the sea and help to maintain an appropriate level of the lake water for fishing and agricultural purposes (rice). The salinity of the lake water depends on the site and the communication with the sea by the channels. When the gates are open due to sea storm, then the salt water flows into the lake. Recent studies show an increase of salinity, but never above 5 gil of chloride, being the minimal values around 0.5 gil. The lake is very shallow, with a maximum depth of approximately 2 m and an average of 75 em, as a consequence of the inflow of more than 30 channels, which bring agricultural, domestic, and industrial effluents without treatment into the lake. All wastewaters from more than 30 urban nuclei come finally into it; therefore the lake has dramatically increased its eutrophication rate during the last century, due to the industrial development and its consequences. High numbers of coliforrns, as well as intestinal pathogens, Salmonella and Vibrio cholerae, have been isolated from the lake water. Sis sampling points have been studied (Fig. 1). Point I collects the wastewaters direct from a small village (EI Saler). Points 1, 2, and 3 are located where channels flow into the lake, carrying domestic, agricultural, and industrial sewage. Point 4 is the site where the lake water flows periodically into the sea when the gates are opened. Point 5 corresponds to a central position.
2. Collection of samples Water samples were collected in 500 ml sterile glass bottles, approximately 15 em below the water surface. The water was always tested within two hours after collection. A total of 166 samples were analyzed from March 1980 until March 1981.
Fig. 1. Albufera lake map.
Faecal Coliforms as Indicators
331
3. Measurement of aquatic environmental parameters Temperature and pH were determined "in situ". The pH value was confirmed in the laboratory. Biological Oxygen Demand (BOD.) was performed in the laboratory according to standard methods (A. p. H. A. 1975).
4. Bacterial analysis a) Total coliforms: Triplicate tubes of lauryl sulfate tryptose broth (LST, DIFCO) were inoculated with the water samples of each site, following the multiple-tube-fermentation technique (A.P.H.A. 1975). Incubation was carried out at 35°C for 24 h. b) Faecal coliforms: All positive LST tube cultures, showing gas within 24 h, were transferred into brilliant green bile broth (BGB, DIFCO) tubes, which were incubated in a stirred water bath at 44.5 DC (SPLITTSTOESSER 1979, TRAM and DANIELSSON 1980). c) Identification of the faecal coliforms: The cultures responsible for the faecal coliform tests were isolated by streaking all positive brilliant green bile tubes on McConkey agar (DIFCO). After purification, the colonies showing different morphologies were identified by means of the Minitek Shstem (BBL) (GUTRERZ and OKOLUK 1978). d) Evaluation of the faecal coliform isolates: Relative frequencies of faecal coliforms were established on a morphological basis, since the different colonies can be easily distinguished on McConkeh agar. A number was given to each agar plate, depending on the relative amount of the different coliforms. As Escherichia coli and Klebsiella pneumoniae accounted for almost all isolates, the following evaluation procedure was assumed: 0: 2.5: 5: 7.5: 10:
corresponds to only E. coli presence (100%), id. to 75 % E. coli and 25 % K. pneumoniae, id. to 50 % E. coli and 50 % K. pneumoniae, id. to 25 % E. coli and 75 % K. pneumoniae, id. to only K. pneumoniae presence (100 %).
The mean of all plates, streaked from one point, gave a coefficient which indicated the relative frequency. To assure correct evaluation, five isolated colonies were randomly picked from each McConkey plate and run through the Minitek System for biochemical identification.
Results Aquatic environmental parameters Fig. 2 shows the mean values of pH and temperature of the water along the study. It is noteworthy that a minimum of 4°C was reached in December 1980, a very unusual cold temperature for the lake. The maximum, 29.5 °C, was reached in August 1980, a normal value for a subtropical climate during the summer. The pH values varied considerably from one site to another. Sites I and 1 showed very little variation along the study (between 7 and 8), whereas sites 2, 3, 4 and 5 fluctuate between wider limits, reaching a maximum of9.7 (point 4, May), and showed a sudden decline in all samples corresponding to September 1980, for which no reason could be found. The pH value in point I corresponds to the typical wastewater value; site 1 is greatly influenced by the high inflow of the channel, therefore the water at this sampling point maintains its channel characteristics. On the contrary, points 4 and 5 are clearly adapted to the lake conditions, showing the typical pH-raise in spring. Points 2 and 3 are also influenced by the lake fluctuations, since the amount of water flowing into the lake by the respective channels is comparatively much smaller than in point 1.
332
E. G.
AUBAN
et al.
· D
2
0
5
A
3
30
• Temp . pll s i t e 4
25
0
•
10
I 20
":> ~
:> D
"E
Q.
..." 5
7
M A M
A
J
SON
0
M
annual cycle (months)
Fig. 2. Variation of pH and temperature values in all sampling sites along the study.
Fig. 3 shows the BOD5 values of the different sites along the study. Here point I gives much higher values than the rest of the sites, as can be expected of typical wastewater. A maximum of 493 mg 2/1 was reached in July 1980. Great fluctuations also occur at this site, due to the intermittent discharges. All other sites show BOD5 values below 50 mg 2/1; site 1 accounts for the minimal values.
°
°
Bacterial Analysis Total and faecal coliforms As the aim of the work was to study the relative frequency of the FC-positive organisms rather than calculate the number of coliforms at each site, we established different categories of sampling sites according to their response to the multiple-tube technique for the FC test. As it can be seen in Table 1 temperatures above 15°C cause a decline in the FC-positive response, both in total and faecal coliforms, which defines different faecal contamination levels for the sites analysed, as could be expected from their location. Sites I, 1 and 2 show a constant positive response, whereas in sites 3, 4, and 5 this response is clearly affected by high temperatures. Site 4 shows the lowest faecal response; indeed, many samples were FC-negative at this point during the warm season. Evaluation and identification of the faecal coliforms On McConkey agar the faecal coliform isolates showed a distinct typical morphology. Escherichia coli formed plain, dark-red colonies, in contrast to the raised pinkish-white mucous colonies of Klebsiella pneumoniae and the bigger dark-pinkt colonies of Enterobacter.
333
Faecal Coliforms as Indicators
•0 li t . 1
450
1
a
52 0 200
2
/::;
3
A
4
o
5
150
-s a 01
E 10 0 .~
Q
o
al
annual cycle (month s) Fig. 3. Variation of Biochemical Oxygen Demand (BODls, in mg O2/1, in all sampling sites along the study.
Table 1. Estimation of the total and faecal coliform densities by means of the mult.iple-tube-fermentation technique, related to water temperature Sampling site
I, 1, 2 3
4 5
+++ ++ + +/-
Total coliform
Fecal coliform
Water temperature (4-15°C)
(15-30°C)
(4-15°C)
(15-30°C)
+++ +++ ++ +++
+++ +++ + ++
+++ ++ + ++
+++ + +/+
All tubes positive in all samplings. Tubes positive in most samplings (,.....,75 %). Tubes positive ocasionally (,.....,25 %). Most samplings without fecal response.
Contamination level
very high high low moderate
334
--
E. G.
AUBAN
et al.
10
ell
0
U
cT ell
~
l l)
...
ell
10
water
15
temp .
20
25
30
(Ocl
Fig. 4. Distribution of relative frequencies of the faecal coliforms, related to water temperature. y = 2.52 + 0.15 X, r + 0.4684
One of the main features, observed along the study, was the almost complete absence of Enterobacter on the McConkey plates, streaked from the BGB tubes incubated at 44.5°C. Escherichia coli and Klebsiella pneumoniae were the two main organisms present. Fig. 4 shows the relationship between the relative amounts of these two coliforms and temperatures along the study. It can be noted that at high temperatures K. pneumoniae predominates, whereas E. coli shows dominancy at low temperatures. The correlation test was significant at p 0.01. When the correlation was performed for each sampling site separately, points 4 and 5 were significant at p 0.01. Surprisingly, in sites I and 3 the correlation coefficients were also significant at p 0.05. Only in sites 1 and 2 no significant correlation was observed. The evaluation of these fidings is somehow difficult. If only in points 4 and 5 a significant correlation would have been established between relative frequency of the two organisms and temperature of the water, the idea of environmental origin of the Klebsiella isolates (or a better adaptation to the aquatic environment, and therefore longer survival would be supported). But site I receives the wastewaters direct from an urban collector, so the relative frequencies observed at this point are the "original" ones.
Discussion Much work has been done on faecal indicators in water bodies, and it seems that the superiority of each organism is conditioned in each particular case by many different parameters (SAYLOR et al. 1975, JENTSCH and GAERTNER 1978, DU'I'KA et al. 1979, McFE'I'ERS and STUART 1972). The true indicator should not grow in the aquatic environment, only survive, but it should be present in such an amount as to allow a relatively easy isolation. On the other hand, it should be exclusively of faecal origin and not pathogenic. In view of the latest works it seems impossible to find an indicator that shows all the desired characteristics, because very often the "faecal coliforms" are not exclusively of faecal origin and can colonize several environments (OGER et al. 1981, BAGLEY and SEIDLER 1977, KNITTEL et al. 1977, SPLIT'I'STOESSER 1979). Our results show that under our working conditions two of the most frequent faecal coliforms, E. coli and K. pneumoniae, exhibit an opposite and complementary behaviour in regard of their amount as indicators. E. coli is universally accepted as faecal indicator, but in regard of the role of K. pneumoniae as a faecal indicator, several studies reveal great differences among Klebsiella isolates (GAVINI et al. 1977, KNITTEL et al. 1977, BAGLEY and SEIDLER 1977, NAEMURA and SEIDLER 1978). Some studies emphasize the importance of Klebsiella pneumoniae in waters (CAMP-
Faecal Coliforms as Indicators
335
BELL et al. 1976, HUNTLEY et al. 1976); others still consider B. coli as the real faecal indicator (SHINGARA et al. 1979). Recent studies point to the importance of correct identification of coliforms, since many psychrotrophic coliforms are not faecal organisms (OGER et al. 1981). Within regard to the coliform densities related to water temperature, our results demonstrate that an increase in this parameter causes a decrease in the coliform population that may be responsible for the disappearance of the FC response during the warm season. Other workers found similar (McFETERS and STUART 1972) and opposite (NuHI and MALEKZADEH 1978) results. The survival rate of faecal coliform bacteria is obviously affected also by many other physical and chemical parameters and by the methods employed (SAYLER et al. 1975, PIERSON et al. 1978, JENTSCH and GAERTNER 1978, DUTKA et al. 1979, THAM and DANIELSSON 1980). Nevertheless, the greater resistance of Klebsiella pneumoniae to unfavourable environmental conditions that cause a decrease in the total amount of faecal coliforms, has been already reported (CAMPBELL et al. 1976). In our work this could explain the predominancy of this organism at high water temperatures at the extreme situation that it can be the only FC-positive coliform. We have isolated K. pneumoniae alone at the most contaminated site (I) at temperatures of 24°C, therefore there should be no doubt of its true faecal origin. The correlation found for this point between temperature and frequency of Klebsiella cannot be explained as a greater resistance or adaptation ability along the way the water has "travelled" from the origin of the wastewater to the more separate points, as it could be the reason for sites 3, 4, and 5. Therefore we consider K. pneumoniae to be a faecal indicator, just as valid as E. coli. Depending on the water characteristics and the environmental conditions (temperature), it can even be more reliable than JiJ. coli as indicator. In our opinion, more ecological as well as epidemiological studies are necessary to establish the validrty of the faecal indicators. References American Public Health Association: Standard Methods for the examination of water and wastewater. Public Health Assoc. 14th ed., New York 1975. BAGLEY, S. T., and SEIDLER, R. J.: Significance of fecal coliform-positive Klebsiella. Appl. Environ. Microbiol. 33 (1977), 1141. CAMPBELL, L. M., MICHAELS, G., KLEIN, R. D., and ROTH, 1. L.: Isolation of Klebsiella pneurnoniae from lake water. Can. J. Microbiol. 22 (1976), 1762. DOCKINS, W. S., and McFETERS, G. A.: Fecal coliform elevated-temperature test: A physiological basis. Appl. Environ. Microbiol. 36 (1978),341. DUTKA, B. J., KUCHMA, S., and KWAN, K. K.: Fecal coliform and Escherichia coli estimates. Water Air Soil Pollut. 11 (1979),349. GAVINI, F., LECLERC, H., LEFBVRE, B., FERRAGUT, C., and IZARD, D.: Taxonomic study of Enterobacteria belonging or related to the genus Klebsiella. Ann. Microbiol, (Paris) 128 B (1977), 45. GUTHERTZ, L. S., and OKOLUK, R. L.: Comparison of miniaturized multitest systems with conventional methodology for identification of Enterobacteriaceae from Foods. Appl. Environ. Microbiol. 35 (1978), 109. HUNTLEY, B. E., JONES, A. C., and CABELLI, V. J.: Klebsiella densities in waters receiving wood pulp effluents. J. Water Pollut. Control. 48 (1976),1766. JENTSCH, F., and GAERTNER, H.: Chemical parameters in coastal waters in correlation with microbiological parameters. Zbl. Bakt. I Orig., Reihe B Hyg. Betriebshyg. Praev. Med. 167 (1978), 115. KNITTEL, M. D., SEIDLER, R. J., EBY, C., and CABE, L. M.: Collonization of the botanical environment by Klebsiella isolates of pathogenic origin. Appl. Environ. Microbiol. 34 (1977), 557.
336
E. G. AUBAN et al., Faecal Coliforms as Indicators
McFETERS, G. A., and STUART, D. G.: Survival of coliform bacteria in natural waters: Field and laboratory studies with membrane-filter chambers. Appl. Microbiol. 24 (1972), 805. NAEMURA, L. G., and SEIDLER, R. J.: Significance of low-temperature growth associated with the fecal coliform response, indol production and pectin liquefaction in Klebsiella. Appl, Environ. Microbiol. 35 (1978),392. NURI, A., and MALEKZADER, F.: Microbial investigation in the water of the Amir Kolayeh Lagoon, Iran Zbl. Bakt. II 133 (1978), 313. OGER, C., GAVINI, F., DELATTRE, J. M., and LECLERC, H.: Coliform organisms and their count in water supply analysis. Ann. Microbiol. (Paris) 132 (1981), 183. PIERSON, C. J., EMSWILER, B. S., and KOTULA, A. W.: Comparison of methods for estimation of coliforms, fecal coliforms and enterococci in retail ground beef. J. Food. Prot. 41 (1978), 262. SAYLER, G. S., NELSON, J. D. Jr., JUSTICE, A., and COLWELL, R. R.: Distribution and significance of fecal indicator organisms in the upper Chesapeake Bay. Appl. Environ. Microbiol. 30 (1975), 625. SRINGARA, S. S., WARREN, W. J., and NELSON, P.: Magnitude of pollution indicator organisms in rural potable water. Appl. Environ. Microbiol. 37 (1979), 744. SPLITTSTOESSER, D. F.: Personnal Communication (1979). TRAM, W., and DANIELSSON, M.-L.: Reliability of violet redbile agar and brilliant green lactose bile broth enumeration off 44 Celsius eoliforrns. Nord. Veterinaermed. 32 (1980), 325. Eingegangen am 1. 4. 1982. Authors' address: Dr. E. GARAY AUBAN, A. ARNAU RIPOLLES, and M. J. PUJALTE DOMARCO, Departamento de Microbiologia, Faculted de Ciencias Biologicas (Universidad de Valencia). Burjasot, Valencia (Spain).
Relative Frequencies and Significance of Faecal Coliforms as Indicators Related to Water Temperature E. GARAY AUBAN, A. ARNAU RIPOLLES, and M.• J. PUJALTE DOMARco With 4 Figures
Summary The faecal eoliforms at different sites of a hypereutrophic lake near Valencia (Albufera) were identified and their relative amounts established along an annual cycle. Using lauryl tryptose broth at 35 DC, followed by incubation at 44.4 DC in 2 % brilliant green bile, Escherichia coli and Klebsiella pneumoniae are practically the only coliforms present. A positive correlation was found between the water temperature and the relative amount of these two coliforms: K. pneumoniae predominates at high water temperatures, whereas E. coli shows preponderance during the cold period. The role of K. pneumoniae as the only faecal indicator under the circumstances described in the work is emphasized and discussed.
Zusammenfassung An verschiedenen Stellen eines hyper-eutrophierten Sees in der Niihe von Valencia (Albufera) wurden die fakalon Coliformen identifiziert und deren relative Menge ein Jahr lang untersucht. Folgende Niihrbiiden wurden verwendet: Laurylsulfat-tryptose, bebriitet bei 35 DC, gefolgt von der Kultivierung in Brilliant-green-bile bei 44,5 °C. Unter diesen Bedingungen konnten fast ausschliel3lich Escherichia coli und Klebsiella pneumoniae nachgewiesen werden. Es wurde cine signifikante Korrelation zwischen Wassertemperatur und relativer Haufigkeit der beiden Colifor-rueu registriert. K. pneumoniae dorninier-t bei hohen "'Vassertelliperaturen, wahrond niedrige Temperaturen ein Dberwiegen von E. coli hervorrufen. Auf die Bedeutung von K. pneumoniae als einzigen fak alcrn Indikator unt.e r den genannten Bedingungen wird hingewiesen.
The term faecal coliform (FC) comprises all coliforms that produce gas from lactose at 44.5 °C, trying to separate the more ubiquitous coliforms from those of true faecal origin (DOCKINS and McFE'rERS 1978). Although several members of the Enterobacteriaceae are able to produce gas at this temperature (A. P. H. A. 1975, BAGLEY and SEIDLER 1977, THAM and DANIELSSON 1980) and should therefore be considered Escherichia coli, there is no general agreement whether only this microorganism or other Enterobacteriaceae-designed faecal coliforms, too, have the same significance when isolated from different sources. Some workers still consider E. coli to be the only faecal coliform in water (SHINGARA et al. 1979), because other FC-positive microorganisms can survive for a long time or even colonize several environments (DUNCAN and RAZELL 1972, CAMPBELL et al. 1976, BAGLEY and SEIDLER 1977, GAVINI et al. 1977, KNITTEL et al. 1977, NAEMURA and SEIDLER 1978, SPLITTs'roEssER 1979). The question arises, therefore, when the possible indicators are not exclusively of faecal origin and can colonize diverse habitats. 22*
330
E. G.
AUBAN
et al.
In our work we have tried to ascertain the relative frequency of the FC-positive organisms in surface waters with different contamination levels. We have studied the evolution throughout a year at several sites of a hypereutrophic lake, which is fed from numerous channels with wastewater, covering the very origin (at the discharging point) to the center of the lake and the site where the lake water flows into the Mediterranean sea. Material and Methods 1. Sampling sites All sampling sites, with one exception, correspond to Albufera lake near Valencia (Fig. 1). This lake is located 12 km to the South of Valencia, close to the sea, from which it is separated by a smalllittorial bar of approximately 1 km width. Three channels ("golas") connect the lake with the sea and help to maintain an appropriate level of the lake water for fishing and agricultural purposes (rice). The salinity of the lake water depends on the site and the communication with the sea by the channels. When the gates are open due to sea storm, then the salt water flows into the lake. Recent studies show an increase of salinity, but never above 5 gil of chloride, being the minimal values around 0.5 gil. The lake is very shallow, with a maximum depth of approximately 2 m and an average of 75 em, as a consequence of the inflow of more than 30 channels, which bring agricultural, domestic, and industrial effluents without treatment into the lake. All wastewaters from more than 30 urban nuclei come finally into it; therefore the lake has dramatically increased its eutrophication rate during the last century, due to the industrial development and its consequences. High numbers of coliforrns, as well as intestinal pathogens, Salmonella and Vibrio cholerae, have been isolated from the lake water. Sis sampling points have been studied (Fig. 1). Point I collects the wastewaters direct from a small village (EI Saler). Points 1, 2, and 3 are located where channels flow into the lake, carrying domestic, agricultural, and industrial sewage. Point 4 is the site where the lake water flows periodically into the sea when the gates are opened. Point 5 corresponds to a central position.
2. Collection of samples Water samples were collected in 500 ml sterile glass bottles, approximately 15 em below the water surface. The water was always tested within two hours after collection. A total of 166 samples were analyzed from March 1980 until March 1981.
Fig. 1. Albufera lake map.
Faecal Coliforms as Indicators
331
3. Measurement of aquatic environmental parameters Temperature and pH were determined "in situ". The pH value was confirmed in the laboratory. Biological Oxygen Demand (BOD.) was performed in the laboratory according to standard methods (A. p. H. A. 1975).
4. Bacterial analysis a) Total coliforms: Triplicate tubes of lauryl sulfate tryptose broth (LST, DIFCO) were inoculated with the water samples of each site, following the multiple-tube-fermentation technique (A.P.H.A. 1975). Incubation was carried out at 35°C for 24 h. b) Faecal coliforms: All positive LST tube cultures, showing gas within 24 h, were transferred into brilliant green bile broth (BGB, DIFCO) tubes, which were incubated in a stirred water bath at 44.5 DC (SPLITTSTOESSER 1979, TRAM and DANIELSSON 1980). c) Identification of the faecal coliforms: The cultures responsible for the faecal coliform tests were isolated by streaking all positive brilliant green bile tubes on McConkey agar (DIFCO). After purification, the colonies showing different morphologies were identified by means of the Minitek Shstem (BBL) (GUTRERZ and OKOLUK 1978). d) Evaluation of the faecal coliform isolates: Relative frequencies of faecal coliforms were established on a morphological basis, since the different colonies can be easily distinguished on McConkeh agar. A number was given to each agar plate, depending on the relative amount of the different coliforms. As Escherichia coli and Klebsiella pneumoniae accounted for almost all isolates, the following evaluation procedure was assumed: 0: 2.5: 5: 7.5: 10:
corresponds to only E. coli presence (100%), id. to 75 % E. coli and 25 % K. pneumoniae, id. to 50 % E. coli and 50 % K. pneumoniae, id. to 25 % E. coli and 75 % K. pneumoniae, id. to only K. pneumoniae presence (100 %).
The mean of all plates, streaked from one point, gave a coefficient which indicated the relative frequency. To assure correct evaluation, five isolated colonies were randomly picked from each McConkey plate and run through the Minitek System for biochemical identification.
Results Aquatic environmental parameters Fig. 2 shows the mean values of pH and temperature of the water along the study. It is noteworthy that a minimum of 4°C was reached in December 1980, a very unusual cold temperature for the lake. The maximum, 29.5 °C, was reached in August 1980, a normal value for a subtropical climate during the summer. The pH values varied considerably from one site to another. Sites I and 1 showed very little variation along the study (between 7 and 8), whereas sites 2, 3, 4 and 5 fluctuate between wider limits, reaching a maximum of9.7 (point 4, May), and showed a sudden decline in all samples corresponding to September 1980, for which no reason could be found. The pH value in point I corresponds to the typical wastewater value; site 1 is greatly influenced by the high inflow of the channel, therefore the water at this sampling point maintains its channel characteristics. On the contrary, points 4 and 5 are clearly adapted to the lake conditions, showing the typical pH-raise in spring. Points 2 and 3 are also influenced by the lake fluctuations, since the amount of water flowing into the lake by the respective channels is comparatively much smaller than in point 1.
332
E. G.
AUBAN
et al.
· D
2
0
5
A
3
30
• Temp . pll s i t e 4
25
0
•
10
I 20
":> ~
:> D
"E
Q.
..." 5
7
M A M
A
J
SON
0
M
annual cycle (months)
Fig. 2. Variation of pH and temperature values in all sampling sites along the study.
Fig. 3 shows the BOD5 values of the different sites along the study. Here point I gives much higher values than the rest of the sites, as can be expected of typical wastewater. A maximum of 493 mg 2/1 was reached in July 1980. Great fluctuations also occur at this site, due to the intermittent discharges. All other sites show BOD5 values below 50 mg 2/1; site 1 accounts for the minimal values.
°
°
Bacterial Analysis Total and faecal coliforms As the aim of the work was to study the relative frequency of the FC-positive organisms rather than calculate the number of coliforms at each site, we established different categories of sampling sites according to their response to the multiple-tube technique for the FC test. As it can be seen in Table 1 temperatures above 15°C cause a decline in the FC-positive response, both in total and faecal coliforms, which defines different faecal contamination levels for the sites analysed, as could be expected from their location. Sites I, 1 and 2 show a constant positive response, whereas in sites 3, 4, and 5 this response is clearly affected by high temperatures. Site 4 shows the lowest faecal response; indeed, many samples were FC-negative at this point during the warm season. Evaluation and identification of the faecal coliforms On McConkey agar the faecal coliform isolates showed a distinct typical morphology. Escherichia coli formed plain, dark-red colonies, in contrast to the raised pinkish-white mucous colonies of Klebsiella pneumoniae and the bigger dark-pinkt colonies of Enterobacter.
333
Faecal Coliforms as Indicators
•0 li t . 1
450
1
a
52 0 200
2
/::;
3
A
4
o
5
150
-s a 01
E 10 0 .~
Q
o
al
annual cycle (month s) Fig. 3. Variation of Biochemical Oxygen Demand (BODls, in mg O2/1, in all sampling sites along the study.
Table 1. Estimation of the total and faecal coliform densities by means of the mult.iple-tube-fermentation technique, related to water temperature Sampling site
I, 1, 2 3
4 5
+++ ++ + +/-
Total coliform
Fecal coliform
Water temperature (4-15°C)
(15-30°C)
(4-15°C)
(15-30°C)
+++ +++ ++ +++
+++ +++ + ++
+++ ++ + ++
+++ + +/+
All tubes positive in all samplings. Tubes positive in most samplings (,.....,75 %). Tubes positive ocasionally (,.....,25 %). Most samplings without fecal response.
Contamination level
very high high low moderate
334
--
E. G.
AUBAN
et al.
10
ell
0
U
cT ell
~
l l)
...
ell
10
water
15
temp .
20
25
30
(Ocl
Fig. 4. Distribution of relative frequencies of the faecal coliforms, related to water temperature. y = 2.52 + 0.15 X, r + 0.4684
One of the main features, observed along the study, was the almost complete absence of Enterobacter on the McConkey plates, streaked from the BGB tubes incubated at 44.5°C. Escherichia coli and Klebsiella pneumoniae were the two main organisms present. Fig. 4 shows the relationship between the relative amounts of these two coliforms and temperatures along the study. It can be noted that at high temperatures K. pneumoniae predominates, whereas E. coli shows dominancy at low temperatures. The correlation test was significant at p 0.01. When the correlation was performed for each sampling site separately, points 4 and 5 were significant at p 0.01. Surprisingly, in sites I and 3 the correlation coefficients were also significant at p 0.05. Only in sites 1 and 2 no significant correlation was observed. The evaluation of these fidings is somehow difficult. If only in points 4 and 5 a significant correlation would have been established between relative frequency of the two organisms and temperature of the water, the idea of environmental origin of the Klebsiella isolates (or a better adaptation to the aquatic environment, and therefore longer survival would be supported). But site I receives the wastewaters direct from an urban collector, so the relative frequencies observed at this point are the "original" ones.
Discussion Much work has been done on faecal indicators in water bodies, and it seems that the superiority of each organism is conditioned in each particular case by many different parameters (SAYLOR et al. 1975, JENTSCH and GAERTNER 1978, DU'I'KA et al. 1979, McFE'I'ERS and STUART 1972). The true indicator should not grow in the aquatic environment, only survive, but it should be present in such an amount as to allow a relatively easy isolation. On the other hand, it should be exclusively of faecal origin and not pathogenic. In view of the latest works it seems impossible to find an indicator that shows all the desired characteristics, because very often the "faecal coliforms" are not exclusively of faecal origin and can colonize several environments (OGER et al. 1981, BAGLEY and SEIDLER 1977, KNITTEL et al. 1977, SPLIT'I'STOESSER 1979). Our results show that under our working conditions two of the most frequent faecal coliforms, E. coli and K. pneumoniae, exhibit an opposite and complementary behaviour in regard of their amount as indicators. E. coli is universally accepted as faecal indicator, but in regard of the role of K. pneumoniae as a faecal indicator, several studies reveal great differences among Klebsiella isolates (GAVINI et al. 1977, KNITTEL et al. 1977, BAGLEY and SEIDLER 1977, NAEMURA and SEIDLER 1978). Some studies emphasize the importance of Klebsiella pneumoniae in waters (CAMP-
Faecal Coliforms as Indicators
335
BELL et al. 1976, HUNTLEY et al. 1976); others still consider B. coli as the real faecal indicator (SHINGARA et al. 1979). Recent studies point to the importance of correct identification of coliforms, since many psychrotrophic coliforms are not faecal organisms (OGER et al. 1981). Within regard to the coliform densities related to water temperature, our results demonstrate that an increase in this parameter causes a decrease in the coliform population that may be responsible for the disappearance of the FC response during the warm season. Other workers found similar (McFETERS and STUART 1972) and opposite (NuHI and MALEKZADEH 1978) results. The survival rate of faecal coliform bacteria is obviously affected also by many other physical and chemical parameters and by the methods employed (SAYLER et al. 1975, PIERSON et al. 1978, JENTSCH and GAERTNER 1978, DUTKA et al. 1979, THAM and DANIELSSON 1980). Nevertheless, the greater resistance of Klebsiella pneumoniae to unfavourable environmental conditions that cause a decrease in the total amount of faecal coliforms, has been already reported (CAMPBELL et al. 1976). In our work this could explain the predominancy of this organism at high water temperatures at the extreme situation that it can be the only FC-positive coliform. We have isolated K. pneumoniae alone at the most contaminated site (I) at temperatures of 24°C, therefore there should be no doubt of its true faecal origin. The correlation found for this point between temperature and frequency of Klebsiella cannot be explained as a greater resistance or adaptation ability along the way the water has "travelled" from the origin of the wastewater to the more separate points, as it could be the reason for sites 3, 4, and 5. Therefore we consider K. pneumoniae to be a faecal indicator, just as valid as E. coli. Depending on the water characteristics and the environmental conditions (temperature), it can even be more reliable than JiJ. coli as indicator. In our opinion, more ecological as well as epidemiological studies are necessary to establish the validrty of the faecal indicators. References American Public Health Association: Standard Methods for the examination of water and wastewater. Public Health Assoc. 14th ed., New York 1975. BAGLEY, S. T., and SEIDLER, R. J.: Significance of fecal coliform-positive Klebsiella. Appl. Environ. Microbiol. 33 (1977), 1141. CAMPBELL, L. M., MICHAELS, G., KLEIN, R. D., and ROTH, 1. L.: Isolation of Klebsiella pneurnoniae from lake water. Can. J. Microbiol. 22 (1976), 1762. DOCKINS, W. S., and McFETERS, G. A.: Fecal coliform elevated-temperature test: A physiological basis. Appl. Environ. Microbiol. 36 (1978),341. DUTKA, B. J., KUCHMA, S., and KWAN, K. K.: Fecal coliform and Escherichia coli estimates. Water Air Soil Pollut. 11 (1979),349. GAVINI, F., LECLERC, H., LEFBVRE, B., FERRAGUT, C., and IZARD, D.: Taxonomic study of Enterobacteria belonging or related to the genus Klebsiella. Ann. Microbiol, (Paris) 128 B (1977), 45. GUTHERTZ, L. S., and OKOLUK, R. L.: Comparison of miniaturized multitest systems with conventional methodology for identification of Enterobacteriaceae from Foods. Appl. Environ. Microbiol. 35 (1978), 109. HUNTLEY, B. E., JONES, A. C., and CABELLI, V. J.: Klebsiella densities in waters receiving wood pulp effluents. J. Water Pollut. Control. 48 (1976),1766. JENTSCH, F., and GAERTNER, H.: Chemical parameters in coastal waters in correlation with microbiological parameters. Zbl. Bakt. I Orig., Reihe B Hyg. Betriebshyg. Praev. Med. 167 (1978), 115. KNITTEL, M. D., SEIDLER, R. J., EBY, C., and CABE, L. M.: Collonization of the botanical environment by Klebsiella isolates of pathogenic origin. Appl. Environ. Microbiol. 34 (1977), 557.
336
E. G. AUBAN et al., Faecal Coliforms as Indicators
McFETERS, G. A., and STUART, D. G.: Survival of coliform bacteria in natural waters: Field and laboratory studies with membrane-filter chambers. Appl. Microbiol. 24 (1972), 805. NAEMURA, L. G., and SEIDLER, R. J.: Significance of low-temperature growth associated with the fecal coliform response, indol production and pectin liquefaction in Klebsiella. Appl, Environ. Microbiol. 35 (1978),392. NURI, A., and MALEKZADER, F.: Microbial investigation in the water of the Amir Kolayeh Lagoon, Iran Zbl. Bakt. II 133 (1978), 313. OGER, C., GAVINI, F., DELATTRE, J. M., and LECLERC, H.: Coliform organisms and their count in water supply analysis. Ann. Microbiol. (Paris) 132 (1981), 183. PIERSON, C. J., EMSWILER, B. S., and KOTULA, A. W.: Comparison of methods for estimation of coliforms, fecal coliforms and enterococci in retail ground beef. J. Food. Prot. 41 (1978), 262. SAYLER, G. S., NELSON, J. D. Jr., JUSTICE, A., and COLWELL, R. R.: Distribution and significance of fecal indicator organisms in the upper Chesapeake Bay. Appl. Environ. Microbiol. 30 (1975), 625. SRINGARA, S. S., WARREN, W. J., and NELSON, P.: Magnitude of pollution indicator organisms in rural potable water. Appl. Environ. Microbiol. 37 (1979), 744. SPLITTSTOESSER, D. F.: Personnal Communication (1979). TRAM, W., and DANIELSSON, M.-L.: Reliability of violet redbile agar and brilliant green lactose bile broth enumeration off 44 Celsius eoliforrns. Nord. Veterinaermed. 32 (1980), 325. Eingegangen am 1. 4. 1982. Authors' address: Dr. E. GARAY AUBAN, A. ARNAU RIPOLLES, and M. J. PUJALTE DOMARCO, Departamento de Microbiologia, Faculted de Ciencias Biologicas (Universidad de Valencia). Burjasot, Valencia (Spain).