A combined membrane filter-immunofluorescent technique for the in situ identification and enumeration of acinetobacter in activated sludge

)Vat. Res. Vol. 22, No. 8, pp. 961-969, 1988 Printed in Great Britain. All rights reserved

0043-1354/88 $3.00+0.00 Copyright © 1988PergamonPress plc

A COMBINED MEMBRANE FILTER-IMMUNOFLUORESCENT TECHNIQUE FOR THE IN SITU IDENTIFICATION A N D E N U M E R A T I O N OF ACINETOBACTER IN ACTIVATED SLUDGE T. E. CLOETE and P. L. ST~YN Department of Microbiology and Plant Pathology, University of Pretoria, 0002 Pretoria, South Africa (First received January 1987; accepted in revised form February 1988)

Abstract--Biological phosphorus removal in activated sludge has been ascribed to luxury phosphorus uptake by Acinetobacter. However due to the lack of suitable identification and enumeration techniques, the exact role of Acinetobacter as a phosphorus removing agent was not clear. During this study a suitable fluorescent antibody (FA) technique combined with membrane filtration was developed for the identification and enumeration of Acinetobacter in activated sludge. The acridine orange (AO) staining technique was applied to enumerate the total number of bacteria in activated sludge and the total number of viable Acinetobacter numbers was determined using agar plating techniques combined with population structure studies using the API-20E identification system. The latter technique indicated that Acinetobacter was the dominant organism in activated sludge. However a comparison of the AO total count and FA Acinetobacter count indicated that Acinetobacter constituted < 10% of the AO count, suggesting that the role of Acinetobacter in activated sludge was probably overestimated in the past. Key words--activated sludge process, Acinetobacter, antibodies, immunofluorescence, membrane filter technique, identification, enumeration

INTRODUCTION

The five-stage Phoredox activated sludge (AS) process was designed for biological nutrient and particularly phosphorus removal. Controversy exists whether phosphorus removal is a chemical, biological or chemically biologically mediated process (Barnard, 1976; Arvin and Bundgaard, 1982). Based on data from previous studies, Acinetobacter has always been associated with phosphorus removal in AS and it has been claimed to be mainly responsible for biological phosphorus removal (Fuhs and Chen, 1975; L6tter and Murphy*). However, the study of the role of Acinetobacter as a phosphorus removing agent in AS has been restricted by bacterial clusters being refractive to dispersion resulting in the underestimation of Acinetobacter numbers using indirect enumeration techniques (Buchan, 1984). The lack of a suitable culture medium which would support the growth of all AS bacteria probably caused a false impression of the bacterial population structure (Prokasam and Dondero, 1967). The lack of in situ identification and enumeration techniques for Acinetobacter in AS furthermore precluded a quantitative analysis of the role of Acinetobacter in AS. The aim of this study was to investigate the role of Acinetobacter in AS. To achieve this goal, suitable techniques had to be developed to disperse bacterial

clusters and to identify and enumerate Acinetobacter in AS. This paper discusses the development of a suitable identification and enumeration technique for Acinetobacter in AS and its possible use for studying the role of Acinetobacter as a phosphorus removing agent in AS.

*L. Lotter and M. Murphy, unpublished method for the dispersion of AS bacteria. City Health Department, Johannesburg, South Africa. 961

MATERIALS AND

METHODS

The same experimental units, experimental animals and bacterial cultures used by Cloete eta/. (1985) were used in this study. Preservative consisting of 0.1% sodium azide in distilled water, phosphate buffered saline (PBS) 0.1 M, pH 7.2, carbonate buffer 0.5 M, pH 9,0 and mounting fluid (carbonate buffered glycerol) were prepared according to the methods of Walker et aL (1971). Two fixatives were used, i.e. formalin (37-40% formaldehlyde aqueous solution) and acetone. The following staining techniques were used; methylene blue stain (Buchan, 1980), irgalan black stain (Hobbie et aL, 1977) and acridine orange (AO) stain (Hobbie et al., 1977). Fluorescein isothiocynate (FITC) was used for fluorescent staining. Acinetobacter agar (Fuhs and Chen, 1975) and giycerol-casitone--yeastagar (GCYA) (Banks and Walker, 1976) were used as culture media. Tween 80 was obtained from Merck Chemicals and Ansan 6 from Chemserve Systems (Pty) Ltd. Sodium tripolyphosphate 0.5% m/v in distilled water was also used. A Sephadex column, used to purify the FITC labelled antisera (FA), was prepared according to Garvey et aL (1977). API-20E microtubes for the identification of isolates were obtained from Ayerst Laboratories Inc. Black Nuclepore filters (0.45 and 0.2/~m pore size, 47 mm dia) were used for all filtration work (Atomic Export Import Company). A Reichert Univar research microscope equipped with epifluorescencefacilities

962

T.E. CLOETE and P. L. STEYN

was used to study methylene blue, AO and fluorescent antibody stained samples. Preparation of somatic 0 antigens

The bacterial cultures mentioned previously were used for the stimulation of antibody production. The antisera were produced separately for each antigen using a modified method of Garvey et al. (I 977). The isolates were transferred from GCYA agar slants to 400 cm 3 Acinetobacter medium and incubated at 37°C for 72 h, collected by centrifugation at 8000 rpm for I0 rain, the supernatant was discarded and the bacteria were resuspended in saline. This procedure was repeated three times. The bacterial suspension was then placed in a boiling water bath for 1 h. The bacteria were again collected by centrifugation (800 rpm for 10min) and diluted with saline to a final concentration of c. 1.2 × 109 bacteria cm -3 using the McFarland scale (McFarland, 1970) for injection into the experimental animals. The following injection programme was used: day 1, injected 0.5 cm 3 antigen; day 4, injected 1.0 cm 3 antigen; day 7, injected 1.5 cm 3 antigen; days 10, 13, 16, 19 and 22, injected 2.0cm 3 antigen on each day and on day 29 the titre of the antiserum was determined using the agglutination test. The separation and preservation of antisera was also carried out according to the method of Garvey et al. (1977). The labelling and purification of the antisera was done according to the methods of Walker et al. (1971). FA specificity tests were done as described by Cloete et al. (1985). MethodJor the enumeration o f Acinetobaeter in A S

In this study a modification of the procedure described by Schmidt (1974) was used to obtain direct counts of AO and FA reacting bacteria in AS. Dispersion and declustering of the bacteria in AS was done as described by Cloete et al. (1985). In this study declustering of bacteria was also done using 5cm 3 of AS added to 95cm 3 of a sterile 0.5% tripolyphosphate solution. This was sonicated for at least 15rain (150W) at 20°C using a Bransonic 32. It was sometimes necessary to sonicate for longer periods to obtain complete dispersion and declustering of flocs and AS bacteria. Ten cm 3 of this solution was furthermore diluted to 90 cm 3 of sterile water before staining and filtration. Filtration was done using the method described by Cloete et al. (1985). This will be referred to as Method I. The second method (Method 2) was used to compare the recovery of bacteria using Nuclepore filters with a pore size 0.45 and 0.2/~m and for the enumeration of the total number of bacteria and Aeinetobaeter in AS. Where Aeinetobacter numbers were to be determined, 2 cm 3 of the 5 cm 3 in 95 cm 3 0.5% tripolyphosphate solution were filtered prior to staining and microscopic investigation. The filters were covered with 2% bovine serum albumin (BSA) fraction 5 to control non-specific adsorption of FA on the filters (Strayer and Tiedje, 1977). The filter was then brought to near dryness at 50-60°C. An area e. 10 mm in dia was marked on the filter. To this area, 0.5 cm 3 of a 1 : 1 diluted specific pooled FA was added. The filter was incubated under an inverted Petri dish cover for 30 min and returned to the filter assembly. Excess FA was washed through the filter unit using lfiO-150cm 3 sterile distilled water. Where the total number of bacteria were to be determined, 2 cm 3 of the dispersed bacterial solution, using the tripolyphosphate method, was stained with AO before filtration according to Hobbie et al. (1977). Two cm 3 of a 70% alcohol solution were used to fix the bacteria. The filter was kept moist by again adding a few drops of water before counting. Calculation of the bacterial numbers using the FA and AO technique was done using the equation of Schmidt (1974). Analysis of variance was done according to Steel and Torrie (1960).

RESULTS AND DISCUSSION Factors influencing the application of the F A technique for the quantification of A c i n e t o b a c t e r were: specificity o f the antisera, suspended colloids in mixed liquor samples o f AS sludge a n d n u m b e r s o f organisms being filtered, m e m b r a n e filter pore size, intensity of fluorescence, cluster-forming bacteria a n d the n u m b e r o f fields to be counted. Cloete et al. (1985) previously reported on the p r e p a r a t i o n a n d effectivity of the antisera. They found that each conjugated serum exhibited excellent immunofluorescent reactions with its h o m o l o g o u s antigen. However, they found they had to pool the four different antisera to e n u m e r a t e all A c i n e t o b a c t e r in AS as none of the individual antisera reacted with all the A c i n e t o b a c t e r encountered in AS. Possible cross reactions with o t h e r bacteria in AS, due to high antiserum titres, were tested by reaction with other related a n d unrelated bacteria and ruled out. A problem regarding the effectivity of the FA technique is the occurrence of non-specific crossreactions on the one h a n d ( M o o d y a n d Jones, 1963), and non-specific adsorption o f the unreacted dye on the o t h e r (Schmidt and Bankole, 1965). Non-specific a b s o r p t i o n was effectively blocked out by prestaining of samples with 2% in the same way as g e l a t i n - r h o d a m i n e isothiocyanate conjugate as described by Strayer and Tiedje (1977). The AS bacteria adsorbed to suspended colloids made it difficult to identify the A c i n e t o b a c t e r to be e n u m e r a t e d using the pooled FA. This p r o b l e m was also experienced by Eren a n d P r a m e r (1965). Adsorption of bacteria to suspended colloids also m a d e it difficult to e n u m e r a t e the bacteria stained with AO. Three m e t h o d s were investigated to disperse the AS and break the clusters formed by certain bacteria. The m e t h o d of Schmidt (1974) was f o u n d to be useful for dispersing the sludge, but did not break up the clusters formed by bacteria. The second method, using A n s a n 6 as dispersant, resulted in the effective dispersion of the sludge as well as the dispersion of clusters formed by Acinetobacter. However, the A n s a n 6 m e t h o d of dispersion (Cloete et al., 1985) only proved to be useful when small volumes a n d low dilutions of AS were used ( 1.0 cm 3 AS in 9 cm 3 sterile water). In order to obtain a more representative sample of the AS, 5 cm 3 were used in 95 cm 3 of sterile 0.5% tripolyphosphate. Effective dispersion of the latter could only be obtained after sonication for at least 1 5 m i n at 1 5 0 W and 2 0 C . Large n u m b e r s of organisms on a filter may depress the intensity of the fluorescence due to the excess of antigen over available a n t i b o d y and m a y also result in multiple layers of organisms in a specific area on a filter, m a k i n g it very difficult to identify a n d e n u m e r a t e individual organisms. Due to the n u m b e r s of organisms occurring in AS it was necessary to dilute samples to such an extent that only a sufficient n u m b e r of organisms was filtered to form a single layer of bacteria on the filter.

Antibodies for Acinetobacter

963

A problem in enumerating fluorescent-stained A comparison was carried out on AS samples, bacteria was that the microscope field viewed was an dispersing the bacteria using sonication and a 0.5% extremely small area (1.54 x l0 -4 cm2). Con- tripolyphosphate solution and filtering through sequently, only a limited number of colloids could Nuclepore filters with pore size 0.45 and 0.2/~m, be tolerated in each field due to the interference of respectively. The results are presented in Table 1. these colloids with the fluorescent staining technique. An average of 22.9% of the bacteria retained by Furthermore, microbial populations had to be high in the 0.20/~m filter were retained by the 0.45/~m filter number in order to encounter a reasonable number of (Table 1). This substantiated the findings of Hobble cells per microscope field. Initially during this study et al. (1977) who found that the size of bacteria varied a volume of I crn3 of a 10-fold dilution of AS in sterile from water mass to water mass. Hobble et al. (1977) distilled water was filtered (Cloete et al., 1985). By found that there was no fixed relationship between using this volume a minimum of 1.03 x l06 bacteria very small and very large bacteria and that higher cm 3 had to be present to detect one bacterium per numbers (99%) of the bacteria studied in a lake were microscope field. Where soil micro-organisms were retained on filters with a pore size of 0.20/~m, than studied (Schmidt, 1974), it was necessary to filter on filters with a pore size of 0.40/zm which retained 0. l cm 3 of a l0 times dilution of soil in water, in order only 56% of the bacteria from the same system. Buchan (1980) using electron micrographs conto overcome the colloid problem. By using this volume, l0 times more bacteria (1.03 x l 0 7 bacteria cluded that the volutin-containing bacteria in AS cm 3) had to be encountered, than in the case with AS, were much larger than the other bacteria. This study in order to detect one bacterium per microscope field. concentrated on the role of Acinetobacter in phosHowever, more than 100 organisms per microscope phorus removal and due to the larger size of the field made counting difficult. Hobbie et al. (1977) volutin containing bacteria (presumed to be Acifound that 2 cm 3 was a sufficient volume to obtain an netobacter) filters with a pore size of 0.45/~m were even distribution of bacteria on the filter surface upon initially used for the enumeration of Acinetobacter filtration. At a later stage during this study 2 cm 3 of (Cloete et al., 1985). During this study 0.2/~m filters a 20 times (in the case of Acinetobacter counts) and were also used to enumerate Acinetobacter in AS. 200 times (in the case of total counts) dilution of AS It was furthermore essential to have a good conwere used. Consequently an even distribution of trast between the FA-stained bacteria and the Nudebetween 40-100 organisms per microscope field was pore filter background. Although mounting fluid obtained. with a pH of 9 was used to enhance fluorescence, The ratio of bacterial cells to colloids was found to the intensity of fluorescence varied. In order to be large and it was therefore not necessary to remove obtain reproducible results, it was decided to count colloids out of suspension after separating them from only bacteria exhibiting a 4 + and 3 + intensity the bacteria. The choice of filters was also important (Thomason, 1976). A 2 + fluorescence together with in developing a successful enumeration technique. physical or morphological characteristics were not Hobbie et al. (1977) found that the type of mem- used in this study as this would have resulted in brane filter as well as the pore size were important inaccurate results due to its subjective nature brought considerations when using fluorescent techniques for about by individual interpretation of morphological enumerating bacteria and claimed that many more characteristics. bacteria were visible on Nuclepore filters than on cellulose filters. The lower numbers encountered Determination o f the Acinetobacter numbers in A S when using cellulose filters were ascribed to the very using the F.4 technique rough surface in contrast with Nuclepore filters reTwo different methods were used. The results sulting in the embedding and masking of bacteria obtained using Method 1 and Method 2 are presented preventing their enumeration (Bowden, 1977). in Tables 2 and 3, respectively. Nuclepore filters were therefore preferred when The analysis of variance (Table 4) of results for doing direct bacterial counts using a fluorescent dye Method 1 indicated that the calculated F for the and epifluorescence microscopy (Bowden, 1977; Hobble et al., 1977). Hobbi¢ et al. (1977) furthermore Table 1. Total AO countsof bacteria in dispersed AS retained on Nuclepore filters with two differentpore sizes* found that many more bacteria were retained on Total AO count filters with a pore size of 0.20/~m than on filters with Sample a pore size of 0.45/~m. number 0.45/zmpore size 0.20/Jm pore size Filtraltion through black Nuclepore filters with a 1 1 . 7 x 10 8 8 . 0 x lO s pore size of 0.20/~m therefore appeared the best 2 1.6 x lO s 8.2 x IOs 3 2.2 x IOs 6.5 x IO s method for the direct enumeration of bacteria in 4 1.6 x l 0 s 6.7 x l 0 s aquatic systems (Hobbie et al., 1977). This was 5 1.7 x 108 7.7 x l 0 s confirmed by Robart and Sephton (1981). Neither the 1.7 x IOs 7.4 x lOs staining of white filters or the use of pre-stained black s 0.25 x 108 0.77 x 10~ filters excluded the problem of non-specific ad- *The AO techniquecombinedwith sonicationand dispersionin a 0 . 5 % t r i p o l y p h o s p h a t e s o l u t i o n w a s used. sorption of FA.

works works works works works

10 11 14 15 16

January January January January January

Date (blocks)

1985 1985 1975 1985 1985

6.63 × 10 ~

152.31 30.46 22.30 73.21

5.88 39.62 54.10 7.53 45.18

Anaerobic zone

6.26 × 1015

152.79 30.55 19.95 65.32

42.92 16.92 52.72 3.61 36.62

Primary anoxic zone

5.13 x 10 ~

135.51 27. I 0 19.08 70.41

5.88 20.84 54.58 16.81 37.40

5.37 × 10 ~5

142.15 28.43 18.21 64.06

6.70 44.36 34.66 11.45 44.98

Secondary anoxic zone

5.53 x 10 j5

145.92 29.18 17.81 61.03

10.31 41.27 40.65 9.18 44.51

Secondary aerobic zone

1.11 × 10 ~

69.00 13.80 6.28 45.54

I 1.34 15.80 22.86 5.70 13.30

Effluent

14.4

14.0 14.5 14.3 16.0 13.2

Total phosphorus removal ( m g P d m ~)

797.68

83.03 178.80 259.57 54.28 221.99

~, X.i

J x x x x x

10 Ls 1015 t015 10 t5 10 '2

Z e i.j.

30.03 x 10 ~

2.18 6.21 12.06 0.60 8.96

21 25 21 17 19 24 26 24 29

M a r c h 1983 M a r c h 1983 April 1983 M a y 1983 M a y 1983 M a y 1983 M a y 1983 June 1983 September 1983

Date (blocks)++

22.50 2.81 1.36 48.39

1.00 4.50 1.00 2.10 2.70 3.70 4.30 3.20 1.20

43.10 × 10 t2 76.37 x 10 ~2

18.00 2.25 0.60 26.66

2.10 2.30 3.00 1.00 2.70 2.10 2.70 2.10 4.70

Anaerobic zone

344.67 x 1012

5 I. 70 6.46 1.22 18.88

7.50 5.00 6.40 4.30 6.40 7,50 7.00 7160 1.40

317.71 x 10 ]2

47.30 5.91 2.33 39.42

6.80 6.60 5.70 4.30 1.00 8.40 6.90 7.60 2.60

23.43 x 10 ~

13. I 0 1.64 0.53 32.31

2.20 2.00 1.10 1.00 1.60 2.10 1.00 2.t0 3.30

69.61 x 10 ]2

17.30 2.16 2.14 99.07

1.00 1.00 1.00 1.00 1.60 3~80 1.00 6.90 ND

Effluent

7.96

11.30 9.40 3.50 6.50 11.30 6.80 7.00 7.90 15.70

Total phosphorus removal (mg P d m 3)

20.60 21.40 18.20 13.79 16.00 27.60 22.90 29.50

~X.i.

169.90

N D = not determined. Note: the Acinetobacter numbers as determined for the Benoni works were not used in the statistical calculations. *Sonication o f activated sludge in a 0.5% tripolyphosphate solution was used to decluster the bacteria before filtering through a 0.20 g m filter. "lTreatments = different activated sludge zones. :~Blocks = different sampling dates.

i

~X2ij

s Coet~cient of variation

~,X.j.

Northern works Northern works Northern works Northern works Northern W o r k s Northern works Northern works Northern works Benoni works

Source o f sample

N u m b e r o f Acinetobacter cm 3 × 106 (treatment)? Primary Secondary Secondary Primary aerobic anoxic aerobic anoxic zone zone zone zone

1012 1012 10 ~2 10 ~2 10 L2 10 ~2 1012 10 ~2

874.89 × 10 ]2

× × x × × × × ×

X 2 i.j. 113.74 99.t0 85.66 44.39 61.66 163.76 124.39 182.19

J

Table 3. The Acinetobacter numbers indifferent activated sludge zones over a period o f time, corresponding phosphorus removal at the time o f sampling and representation o f statistical data*

*Ansan 6 and shaking by hand was used to decluster the activated sludge bacteria before filtration through a black Nuclepore filter with pore size 0.45/tm. t T r e a t m e n t s = different activated sludge zones. SBIocks = different sampling dates.

i

~X2ij

s Coeti]cient of variation

~X.j.

Northern Northern Northern Northern Northern

Source of sample

Primary Aerobic zone

N u m b e r of Acinetobacter cm 3 x 106 (treatment)?

Table 2. The/Icinetobacter numbers in different activated sludge zones over a period o f time, corresponding phosphorus removal at the time of sampling and representation o f statistical data*

Z

.m

Antibodies for Acinetobacter

965

Table 4. Analysis of variance for data in Table 2 Source of variation

d.f.

SS

MS

Calculated F

Treatment (zones) Blocks (dates) Error

5 7 35

1.75 x 104 0.34 x 10t4 0.64 x 1014

3.49 x 103 4.85 x 1012 1.82 x 1012

18.96 2.66

Total

47

Tabular F 0.05 and F 0.01 for 7 and 35 degrees of freedom= 2.29 and 3.21 (for blocks). Tabular F 0.05 and F 0.01 for 5 and 35 degreesof freedom= 2.49 and 3.60 (for treatments). Least significantrange (LSR) = 1.36 x 106- 1.54 x 10 6. If the mean range value exceedsthe LSR the differencesare significant. If the mean range value is smaller than the LSR the differencesare not significant.

blocks exceeded 1% o f tabular F at the 0.05 level, indicating a significant difference a m o n g s t the block means at the 0.05 level. However, the calculated F for the blocks was less t h a n the tabular F at the 0.01 level, indicating no significant difference at this level (Table 4). It was therefore concluded that at the 5% level a significant difference existed a m o n g s t Acinetobacter n u m b e r s at different times, but no significant difference existed at the 1% level. The significant difference at the 5% level was p r o b a b l y due to the fluctuation in the organic feed which varied with time. This would have had a direct influence on the microbial p o p u l a t i o n since the AS system is a biological process. The calculated F for the treatments exceeded 1% o f the tabular F at the 0.05 and 0.01 levels, indicating significant differences a m o n g s t treatments (Table 4). These results did not, however, indicate which o f the treatments differed significantly. The least significant difference (LSD) using D u n c a n ' s Table 5. Multiple range values in order to determine the difference amongst the AS zones of the Northern works systemat the 5% level of confidence Zone Primary aerobic

Secondary anoxic

Primary anoxic Anaerobic Effluent

Zone compared with Mean range value Primaryanoxic 3.65 x 106* Anaerobic 4.21 x 106* Secondary aerobic 4.82 x 106* Effluent 4.30 × 10 6* Secondary anoxic 0.55 x 106* Secondaryaerobic 4.27 x 106* Effluent 3.75 x 106* Primary anoxic 3.07 x 106* Anaerobic 3.66 x 106* Secondaryaerobic 1.17 x 106 Effluent 0.65 x 106 Anaerobic 0.56 x 106 Secondary aerobic 0.62 x 106 Effluent 0.09 X 10 6 Secondary aerobic 0.52 x 106

*Indicates significantdifferences. ~'The values in Table 2 were used for the calculations.

new multiple range test (Steel and Torrie, 1960) was used to determine the LSD o f the treatments (Table 5). The Acinetobacter n u m b e r s in the primary aerobic and secondary anoxic did not differ significantly, but were significantly higher t h a n those in the other zones (Table 5). This suggested that the physical size o f Acinetobacter cells in these two zones (primary aerobic and secondary anoxic zones) were either larger than those in other zones (hence m o r e cells were retained by the 0.45 # m filter) or that m o r e Acinetobacter cells were present in these two zones due to the increased activity o f Acinetobacter under aerobic conditions, resulting in the carry-over o f these n u m b e r s to the secondary anoxic zone and hence the higher n u m b e r s in this zone. The analysis o f variance for the data in Table 3 (Table 6) indicated that the calculated F for the treatments (different AS zones) was < 1% o f the tabular F at the 0.05 and 0.01 levels. These results indicated that there was no significant difference a m o n g s t the different AS zones. This was ascribed to the large coefficient o f variation (Table 3) which was probably due to the smaller cells which might have passed t h r o u g h the 0.45 # m filter, but were retained on the 0.2 # m filter. Since AS should ideally be o p e r a t e d in the e n d o g e n o u s phase (Wiechers et al., 1984) it can be expected that the relative n u m b e r o f very small bacteria may be high and may fluctuate to a large extent resulting in the large coefficient o f variation (Table 3) and no significant difference a m o n g s t the zones. The calculated F for the blocks exceeded the tabular F at the 0.05 and 0.01 level, (Table 6). This indicated that the Acinetobacter n u m b e r s differed significantly in time. Again this is not surprising since the organic load varied daily consequently affecting the microbial activity and numbers.

Table 6. Analysis of variance for data in Table 3 Source of variation d.f. SS MS CalculatedF Treatments (zones) 5 1.0 x 105 2.0 × 10~4 1.90 Blocks (date) 4 5.2 x 105 1.3 × 10~5 12.38 Error 20 2.1 x 105 1.0 x 10~4 Total

29

Tabular F 0.05 and F 0.01 for 4 and 20 degrees of freedom= 2.87 and 4.43 (F or blocks). Tabular F 0.05 and F 0.01 for 5 and 20 degrees of freedom = 2.71 and 4.10 (F or treatments).

966

T . E . CLOETE and P. L. SrEYN Table 7. Distribution of aerobic and facultative anaerobic Enterobacteriaceae and other Gram-negative bacteria in different zones of the Northern works AS system As zone Anaerobic

Primary anoxic

Number cm ~

Bacteria Acinetobacter calcoaceticus var. lwoffii Pseudomonas Aeromonas hydrophila Citrobacterfreundii CDC-Group I1 K-2 CDC-Group 11 F Shigella Klebsiella oxytoca Escherichia coli Serratia Pasteurella CDC-Group I V Flavobacterium Yersinia Total

Primary aerobic

%

Number cm 3

10 6

66.6

2.9

x 10 6

67.4

5.5 x 10~

76.3

4.0 x 105 3.0 × 105 2.0 × 105 1.0 × 105 1.0 × 105 1.0 × 10' 1.0 × 105 ---

10.1 7.6 5.1 2.5 2.5 2.5 2.5 0 0 0 0 0 0

4.0 x 105 6.0 × 105 --

9.3 13.9 0 0 0 0 0 2.3 6.9 0 0 0 0

4.0 x 105 4.0 x 105 1.0 × 105 -.-.... --5.0 × 105 1.0 × 105 1.0 × 105 1.0 × 105

5.5 5.5 1.3 0 0 0 0 0 0 6.9 1.3 1.3 1.3

2.6 ×

%

3.9 x 10~

Number cm 3

--1.0 × 105 3.0 × 105 4.3 × 106

%

7.2 x 106

Note: percentages are expressed as the percentage of the total number of bacteria that could be identified and not the total number of bacteria.

D e t e r m i n a t i o n o f the A c i n e t o b a c t e r n u m b e r s in A S using the A P I - 2 0 E s y s t e m T h e t o t a l v i a b l e p l a t e c o u n t w a s d e t e r m i n e d in triplicate for t h e a n a e r o b i c , p r i m a r y a n o x i c a n d p r i m a r y a e r o b i c z o n e . All c o l o n i e s f r o m p l a t e s p o u r e d f r o m a 105 d i l u t i o n were c o u n t e d , h a v i n g b e e n b e t w e e n 3 0 - 3 0 0 c o l o n i e s p e r plate. All c o l o n i e s f r o m t h e p l a t e w i t h a viable c o u n t c l o s e s t to t h e a v e r a g e o f t h e t h r e e p l a t e s were p i c k e d u p a n d purified ( C l o e t e et al., 1985). Biological p h o s p h o r u s r e m o v a l in AS has always been associated with Gram-negative b a c t e r i a ( R o i n e s t a d , 1973; F u h s a n d C h e n , 1975; B u c h a n , 1980; B r o d i s c h a n d J o y n e r , 1982). T h e r e f o r e o n l y t h e G r a m - n e g a t i v e isolates w e r e identified u s i n g the API-20E system. The percentage representation o f e a c h isolate w a s c a l c u l a t e d b y u s i n g o n l y t h o s e i s o l a t e s g i v i n g a n excellent o r g o o d i d e n t i f i c a t i o n . T h e s e r e s u l t s a r e r e p r e s e n t e d in T a b l e 7 a n d were u s e d to e n u m e r a t e A c i n e t o b a c t e r in A S for c o m p a r a tive p u r p o s e s w i t h t h e F A t e c h n i q u e as p r e v i o u s l y d i s c u s s e d ( T a b l e 8). T h e viable c o u n t s o f identifiable G r a m - n e g a t i v e b a c t e r i a were 3.9 x 106, 4.3 x 106 a n d 7.2 x 106 for the anaerobic, primary anoxic and primary aerobic

z o n e s , respectively. In all t h r e e o f t h e s e z o n e s A. calcoaceticus var. lwoffii w a s f o u n d to be d o m i n a n t (66.6, 67.4 a n d 7 6 . 3 % , r e s p e c t i v e l y f o r t h e t h r e e different zones). Acinetobacter was followed by Pseud o m o n a s a n d A e r o m o n a s in t h e a n a e r o b i c z o n e (10.1 a n d 7 . 6 % , respectively) a n d were e q u a l l y r e p r e s e n t e d in t h e p r i m a r y a e r o b i c z o n e ( 5 . 5 % e a c h ) , w h i l s t A e r o m o n a s w a s m o r e i m p o r t a n t t h a n P s e u d o m o n a s in t h e p r i m a r y a n o x i c z o n e (13.9 vs 9 . 3 % ) . M e m b e r s o f t h e o t h e r g e n e r a o r g r o u p s c o n s t i t u t e d t h e less i m p o r t a n t m e m b e r s o f t h e identifiable G r a m - n e g a t i v e isolates. I n all t h r e e z o n e s , t h e o b l i g a t e a e r o b e s were f o u n d to be p r e d o m i n a n t . B r o d i s c h a n d J o y n e r (1982) d e t e r m i n e d t h e b a c t e rial p o p u l a t i o n s t r u c t u r e o f a pilot p l a n t a n d l a b o r a tory AS unit. They found that Aeromonas, Pseudom o n a s a n d Alcaligenes ( n o t a l w a y s in this o r d e r ) were t h e d o m i n a n t g e n e r a in t h e a n a e r o b i c , p r i m a r y a n o x i c a n d p r i m a r y a e r o b i c z o n e . T h i s differed f r o m t h e r e s u l t s o b t a i n e d in this s t u d y w h e r e A c i n e t o b a c t e r w a s f o u n d to be t h e d o m i n a n t o r g a n i s m ( T a b l e 7). H o w e v e r , a c o m m o n f a c t o r w a s t h a t in b o t h s t u d i e s relatively h i g h n u m b e r s o f P s e u d o m o n a s a n d A e r o m o n a s were e n c o u n t e r e d . B u c h a n (1980) f o u n d A c i n e t o b a c t e r (with a n a v e r a g e r e p r e s e n t a t i o n o f 5 7 % ) to

Table 8. Comparison of the average Acinetobacter numbers as determined using the AP1-20E technique and the two methods employing the FA technique Acinetobacter numbers cm 3 Technique API-20E Method 1 Method 2

Anaerobic zone 2.60 x 106 2.20 x 106 3.00 × 10 7

SD for Method I SD for Method 2

0.60 x 2.23 ×

10 6 10 7

Primary anoxic zone 2.90 × 10 6 2.80 x l0 n 3.00 x 10 7 1.36 × 106 1.99 × 107

Primary aerobic zone 5.50 x 10 6 6.40 × 10 6 2.70 × [07 1.22 × 1.90 x

[0 6 10 7

SD = standard deviation. Method 1:0.45/~m Nuclepore filters and Ansan 6 dispersion. Method 2:0.20 um Nuclepore filters and 0.50 tripolyphosphate and sonication dispersion.

Antibodies for Acinetobacter be the dominant organism in the primary aerobic zone of five different AS plants. In principal, these r e s u l t s corresponded with the results obtained in this study. However, the percentage of Acinetobacter encountered by Buchan (1980) was lower than in this s t u d y (57 vs 76%). The results from the above mentioned studies substantiated the statement of Buchan (1984) that organisms such as Pseudomonas and Acinetobacter which have a low growth factor requirement would proliferate in AS. The predominance of Acinetobacter in the different AS zones studied in this study was ascribed to; (i) the low growth factor requirement; and (ii) storage of polyfl-hydroxybutyrate as an internal carbon and energy reserve enabling the organism to survive anaerobic and primary anoxic conditions. In the anaerobic, primary anoxic and primary aerobic zone, the Acinetobacter numbers determined by using the API-20E method was within the standard deviation of the Acinetobacter numbers determined by using the 0.45/zm Nuclepore filter and FA technique (Method 1) (Table 8). However, the Acinetobacter numbers as determined using the API-20E technique were lower and not within the standard deviation of the Acinetobacter numbers as determined using the 0.20/~m filter and FA (Method 2) in the above mentioned zones (Table 8). The API-20E determination of Acinetobacter numbers relies on the viability of the organisms and their capability of growing on GCYA. Since Acinetobacter grew on GCYA, viability was the determining factor when determining the Acinetobacter numbers using the API-20E technique. The results in Table 8 indicated that the Acinetobacter numbers determined using the API-20E method, corresponded closely with the determination of the Acinetobacter numbers using Method 1, but were not within the standard deviation when using Method 2. The higher numbers of Acinetobacter obtained using Method 2 however indicated that many of the Acinetobacter cells were small enough to pass through the 0.45/zm filter.

Experiments to determine the total number o f bacteria in A S Due to the lack of suitable nutrient media which would support the growth of all viable nutritional

types of bacterial in AS (Banks and Walker, 1976), the AO-staining technique was used in this study for determining the total number of bacteria in AS. The results are presented in Table 9. A comparative analysis was also done using the AO technique and the viable plate count technique (Table 10). When the average Acinetobacter numbers were determined on the same sample as the total count and using the same technique, except that FA was used instead of AO, it was found that Acinetobacter constituted 8.5, 6.2, 6.1, 7.8, 7.7 and 5.0% of the total count in the respective zones (anaerobic, primary anoxic, primary aerobic, secondary anoxic and secondary aerobic zone) and effluent, (comparison of the average Acinetobacter numbers in Table 3 and the results in Table 9). The percentages were lower than the percentage representation as determined using the API-20E technique (Table 7). This was ascribed to the fact that the API-20E (technique relies on the viability of the organisms whereas by using the FA technique dead and live cells are counted and in Table 7 the comparison was not done with the total count, but with the count of identifiable bacteria. It was probable that the GCYA favoured the growth of Acinetobacter giving it an unfair advantage over other bacteria and hence the high percentage representation when using the API-20E technique. Due to the lack of agar plating techniques for the direct enumeration of Acinetobacter in AS, it was not possible to compare the results obtained by the FA technique with the viable count of Acinetobacter. The numbers of Acinetobacter obtained by using the FA technique were nevertheless compared with the total viable count in AS. Prokasam and Dondero (1967) obtained a maximum number of viable bacteria (6.10 x 108 bacteria cm -3) when using sewage extract agar. The average total viable bacterial counts encountered by Prokasam and Dondero (1967) in all AS samples cultured on different media was 1.4 x 10s bacteria cm -3. Banks and Walker (1976) on average encountered a viable bacterial count of 4.28 x 108 bacteria cm -3 when using GCYA as culture medium. The average viable plate count obtained in this study was 4.32 x 106 bacteria cm -3 (Table 10). This was lower than the average viable plate counts encountered

Table 9. The total numberof bacteria in the Northern works AS plant as determinedby the AO-stainingtechnique Number of bacteria x 10Scm-3 in the differentAS zones Primary Primary Secondary Secoindary Anaerobic anoxic aerobic anoxic aerobic Date zone zone zone zone zone Effluent 10/I/1985 1.92 1.27 1.69 0.95 0.47 2.68 1I/1/1985 1.59 6.55 3.35 2.63 2.45 0.41 14/1/1985 4.70 4.73 4.66 5.17 6.42 3.72 15/1/1985 4.47 4.59 5.73 2.53 1.56 2.90 16/1/1985 4.82 7.32 6.64 6.54 7.83 3.92 -f 3.50 4.89 4.41 3.56 3.74 2.72 S 1.60 2.33 1.95 2.24 3.20 1.39 Coefficientof variation 45.70 47.60 44.21 62.-92 85.50 51.10 Note: the filter pore size was 0.2 pm.

967

968

T.E. CLOETEand P. L. Sa~w

E

8

<

XXXXX

< xxxx~

~2

xxxx×

<

xxxxx

© < xxxx×

xxxxx ~ M ~ i

.8

EE ~tr3 eS~

~88

~.~

b-

by Prokasam and Dondero (1967) and Banks and Walker (1976). The higher counts obtained by Prokasam and Dondero (1967) than in this study and that of Banks and Walker (1976) was probably due to the high rate activated sludge plants which they examined supporting the growth of more viable cells. Therefore, the viable count obtained in this study was also lower than the viable count obtained by Banks and Walker (1976) although the same procedure was followed. The total bacterial count using the AO technique and Method 2 yielded the highest number of bacteria (1.53 x l 0 9 bacteria cm 3) followed by the AO technique using Method 1 (7.17 x bacteria cm --3) followed by the viable plate count (4.32 x 10 6 bacteria cm -~) (Table 10). Higher bacterial numbers were always encountered when using the AO technique and 0.20#m filters, than in any of the viable plate counts including those of Prokasam and Dondero (1967) and Banks and Walker (1976). This was probably due to the lack of suitable nutrient medium for the growth of all viable bacteria in AS. Furthermore, by using the AO technique, dead and live cells were counted. This would result in higher numbers when compared to viable counts of bacteria on GCYA. Higher bacterial numbers were also encountered when using the AO technique and 0.20 p m filter than was the case using the AO technique and 0.45 #m filter (Table 10). This was ascribed to smaller bacteria passing through the 0.45 #m filter, but being retained by the 0.20 ~ m filter, a phenomenon also encountered by Hobbie et al. (1977). The average Acinetobacter count using a 0.20 p m filter constituted 56% of the total viable plate count, 0.10% of the total AO count using a 0.20#m filter and 3.4% of the total AO count using a 0.45/~m filter (Table 10). The coefficient of variation was low at 6.5% (Table 10). The average Acinetobacter count using a 0.45 ~tm filter constituted 48% of the total viable plate count, 0.13% of the total AO count using a 0.20 p m filter and 2.9% of the total AO count using a 0.45#m filter (Table 10). The coefficient of variation was low (4.7%) (Table 10). Eighty four percent of the Acinetobacter numbers retained on the 0.20 #m filter was also retained on the 0.45 #m filter (Table 10). However only 4.6% of the total number of bacteria, as determined using the AO technique, were retained on the 0.45 tim filter (Table 10). These results indicated that there was a marginal difference in the size of Acinetobacter in the primary aerobic zone, but that the size of the total number of bacteria varied to a large extent. The small variation in the size of Acinetobacter in the primary aerobic zone is ascribed to viability of these organisms in this zone. CONCLUSIONS

It was concluded that the bacterial clusters in AS could be dispersed effectively by using Method 2. The FA technique could furthermore be applied success-

Antibodies for Acinetobacter fully for the identification and enumeration of Acinetobacter in AS. By using the viable plate count and API-20E technique for the identification and enumeration of Acinetobacter, it was found that this organism was dominant in AS. The total number of bacteria in AS using the A O technique was compared with the F A numbers o f Acinetobacter. This indicated, in contrast with the API-20E technique results, that Acinetobacter, constituted less than 10% o f the total bacterial population in activated sludge. These results therefore suggest that the role of Acinetobacter as a phosphorus removing agent in AS, has probably been overestimated in the past. REFERENCES

Arvin E. and Bundgaard E. (1982) Full scale experience with phosphorus removal in an alternating system. In Proceedings of a Post Conference Seminar on Phosphate Removal in Biological Treatment Processes. Council for Scientific and Industrial Research, Pretoria, South Africa. Banks C. J. and Walker I. (1976) Sonication of activated sludge flocs and the recovery of their bacteria on solid media. J. gen. Microbiol. 98, 363-368. Barnard J. L. (1976) A review of biological phosphorus removal from the activated sludge process. War. S.A. 2, 136-144. Bowden W. B. (1977) Comparison of two direct-count techniques for enumerating aquatic bacteria. Appl. envir. Microbial. 33, 1229-1232. Brodisch K. E. U. and Joyner S. J. (1982) The role of organisms other than Acinetobacter in biological phosphate removal in activated sludge processes. In Proceedings of a Post Conference Seminar on Phosphate Removal in Biological Treatment Processes. Council for Scientific and Industrial Research, Pretoria, South Africa. Buchan L. (1980) The location and nature of accumulated phosphorus in activated sludge. D.Sc. thesis, University of Pretoria, South Africa. Buchan L. (1984) Microbiological aspects. In Theory, Design and Operation of Nutrient Removal Activated Sludge Processes (Edited by Wiechers H. N. S., Ekama (3. A., Gerber A., Keay (3. F. P., Malan W., Marais (3. V. R., Osborn D. W., Pitman A. R., Potgieter D. J. J. and Pretorius W. A.). Water Research Commission, Pretoria, South Africa. Cloete T. E., Steyn P. L. and Buchan L. (1985) An aut-ecological study of Acinetobacter in activated sludge. Wat. Sci. Technol. 17, 139-146.

969

Eren E. and Pramer D. (1965) Application of immunofluorescent staining to studies of the ecology of soil micro-organisms. Soil Sci. 101, 39-45. Fuhs (3. W. and Chen M. (1975) Microbiological basis of phosphate removal in the activated sludge process for the treatment of wastewater. Microbial Ecol. 2, 119-138. (3arvey J. S., Cremer N. E. and Sussdorf D. H. (1977) Methods in Immunology: A Laboratory Text for Instruction and Research, 3rd edition. Benjamin, London. Hobbie J. E., Daley R. J. and Jasper S. (1977) Use of nuclepor¢ filters for counting bacteria by fluorescence microscopy. Appl. envir. Microbiol. 33, 1225-1228. McFarland J. (1970) The nephelometer: an instrument for estimating the number of bacteria in suspensions used for calculating the opsonix index and for vaccines. J. Am. Med. Ass. 49, 1176. Prokasam T. B. and Dondero N. C. (1967) Aerobic heterotrophic bacterial populations of sewage and activated sludge. I. Enumeration. Appl. Microbiol. 15, 461-467. Robart R. D. and Sephton L. M. (1981) The enumeration of aquatic bacteria using DAPI. J. Limnol. Soc. Sth. Afr. 7, 72-74. Roinestad F. A. (1973) Volutin accumulation by activated sludge micro-organisms. University Microfilms International, 300 North Zceb Road, Ann Arbor, MI 48106, U.S.A. Schmidt E. L. (1974) Quantitative aut-ecological study of micro-organisms in soil by immunofluorescence. Soil Sci. 118, 141-149. Schmidt E. L. and Bankole R. O. (1965) Specificity of immunofluorescent staining for study of Aspergillusflavus in soil. Appl. Microbial. 13, 673~$79. Steel R. (3. D. and Torrie J. H. (1960) Principles and Procedures of Statistics. Mc(3raw-Hill, New York. Strayer R. F. and Tiedje J. M. (1977) Application of the fluorescent antibody technique to the study of methanogenic bacterium in lake sediments. Appl. envir. Microbiol. 35, 192-198. Thomason B. M. (1976) Fluorescent antibody detection of Salmonellae. In Compendium of Methods for the MicrobiologicalExamination of Foods (Edited by Speck M. M.). American Public Health Association, Washington, D.C. Walker P. D., Batty L. and Thomson R. O. (1971) The localization of bacterial antigens. In Methods in Microbiology (Edited by Norris J. R. and Ribbons D. W.). Academic Press, New York. Wiechers H. N. S., Ekama (3. A., Gerber A., Keay (3. F. P., Malan W., Marais (3. V. R., Osborn D. W., Pitman A. R., Potgieter D. J. J. and Pretorius W. A. (1984) (Eds) Theory, Design and Operation of Nutrient Removal Activated Sludge Processes. Water Research Commission, Pretoria, South Africa.