The purpose of this study was to test the applicability of the API 20E and API Rapid NFT systems for the identification of some predominant gram-negative and gram-positive bacteria isolated from lab-scale activated sludge treatment systems. In this study, one lab-scale sequencing batch reactor (SBR) and one lab-scale continuous-flow stirred tank reactor (CFSTR) were setup. After both reactors had reached equilibrium, many pure cultures isolated from the activated sludge in both systems were obtained and many morphological, biochemical, physiological tests were conducted to identify each pure culture. The API 20E system is a standardized, miniaturized version of conventional procedures for rapid identification of Enterobacteriaceae and other gram-negative bacteria, and the Rapid NFT kit is used for the identification of the gram-negative, non-fermentative bacteria. Also, a Phillips 300 Transmission Electron Microscope and a Phillips 301 Transmission Electron Microscope were applied to further verify the identification of some genera. According to the results of this study, it has been concluded that some commercial products, such as API 20E system and API Rapid NFT system, can be applied for the identification of microorganisms only at the genus level. Many other additional morphological, biochemical, and physiological tests are always needed to obtain the exact identification of each microorganism at the species level. More advanced technologies such as 16S rRNA may be necessary, however, for a rapid identification of the total bacterial population. In this study, it has also been found that Brevibacterium acetylicum and Pseudomonas vesicularis are two of the most dominant species in the activated sludge of CFSTR system. Gram-positive bacteria such as members of the genus Arthrobacter have shown to be very significant and predominant in the SBR system.

As a rule, the identification of pure cultures isolated from activated sludge systems can be carried out using Bergey's Manual (Holt, 1986), The Prokaryotes (Starr and Stolp, 1981), and other references (Leifson, 1960; Lighthart and Loew, 1972; Collins and Lyne, 1976; Ward et al. 1986; Atlas et al. 1988; Labeda, 1990). The API 20E (Analytab Products Inc., 1985) and API Rapid NFT systems are two commercial products, made by API Analytab Products, Division of Sherwood Medical, New York, USA. The API 20E system uses 20 miniature reaction compartments (cupules) that produce 23 biochemical reactions, and is a standardized, miniaturized version of conventional procedures for rapid identification of Enterobacteriaceae and other gram-negative bacteria. The major advantage of the API 20E system is that it is more convenient and easier to identify gram-negative bacteria than the conventional tests mentioned in the above references. The Rapid NFT kit is used for identification of gram-negative, non-fermentative bacteria. The Rapid NFT system is similar to the 20E system in that a profile number is obtained from standard tests performed on the organisms.

Most studies on the population structure of activated sludge are over 30 years old. Several gram-negative bacterial genera, such as Pseudomonas, Achromobacterium and Flavobacterium, were found to be dominant and responsible for sewage purification by Allen (1943). In 1952, McKinney and Horwood isolated eleven organisms which could produce flocs when aerated in a suitable substrate, and showed that several microorganisms are capable of producing flocs similar to activated sludge (McKinney and Horwood, 1952). Anderson and McCoy reported in 1963 that Pseudomonas and related genera were predominant in the activated sludge from aeration tanks of municipal waste treatment plants (Anderson and McCoy, 1963), while Dias and Bhat reported in 1964 that Zoogloea and Comamonas were the predominant bacteria in activated sludge (Dias and Bhat, 1964). These authors also stated that occasionally other genera of bacteria could appear predominant without adversely affecting the system. Several other types of microorganisms were also isolated from the activated sludge. Lighthart and Oglesby (1969) and Lighthart and Loew (1972), presented a large number of microorganisms found to be predominant in activated sludge, and these predominant microorganisms were identified as Flavobacterium, Achromobacter, and a group classified as Pseudomanads. Unz and Dondero (1970) found Flavobacterium, Alcaligenes and Pseudomonas to be the dominant organisms present.

a. CFSTR system design

System design and operating data. The bench scale continuous flow stirred tank reactor is shown in Figure 1, and its operating data shown in Table 1. The feed was pumped to the aeration basin continuously at a rate of 30 liters per day, and the effluent was directly discharged to a nearby sink. The volume of mixed liquor in the aeration basin was maintained at 30 liters. Sludge was recycled continuously through the same pump used for feed and also at a rate of 30 liters per day. Therefore feed, recycled sludge, and effluent flowed into and out of the reactors, respectively, on a continuous flow basis. By wasting one liter of activated sludge per day, the biological solids retention time was maintained at about 30 days. The cylindrical clarifier was inclined at an angle of 45 degree, so the settled sludge could be easily collected and recycled. The volume of the clarifier was 7.35 liters, so the hydraulic retention time was about 2.94 hours. A scraper, rotating constantly and slowly at a rate of approximate one revolution per minute, was installed inside the clarifier. The purpose of this scraper was to prevent the sludge from attaching onto the wall.

Feed composition and seed sludge. The feed composition of the CFSTR is shown in Table 2. The seed sludge of the CFSTR system in this study was originally obtained from one well-operated domestic wastewater treatment plant, called H.C. Morgan Treatment Plant, located at Auburn, Alabama, USA.

b. SBR system design

System design and operating data. The bench scale sequencing batch reactor used in this study is shown in Figure 2, and its operating data is also given in Table 1. The SBR was operated two cycles per day, and one operating cycle was 12 hours. The mixed liquor was stirred and aerated for a total feed and mix time of about 10.6 hours, and then allowed to settle for 45 minutes. Finally, the sludge remained idle for the remainder of the cycle resulting in a total draw and idle time of 40 minutes.

Feed composition and seed sludge. The feed composition of the SBR is shown in Table 3. The seed sludge of the SBR system was also originally obtained from H.C. Morgan Treatment Plant.

c. Confirmation of steady state (equilibrium)

After the system had been operated for more than 2 months, routine effluent samples were taken and filtered through glass fiber filters to check stable operation. Although there was no necessity to filter the accompanying feed samples, both the effluent and feed samples were filtered through 0.45 μm membrane filters for nitrite and nitrate tests. Many tests were performed following the procedures in Standard Methods (Franson et al. 1985) to ensure that the system had reached equilibrium, such as MLVSS (Mixed-liquor Volatile Suspended Solids), COD, TKN (Total Kjeldahl Nitrogen), NH₄-N, NO₃-N, NO₂-N, pH, and DO. Each test was generally performed for at least four consecutive days and the data were averaged.

d. Isolation and identification of microorganisms

Preparation of sludge samples. After the activated sludge system was confirmed to have reached steady state, the sludge was then taken for the identification of predominant microorganisms. For the CFSTR system, samples were always taken directly from the aeration tank. For the SBR system, samples were always taken approximately 2 hours after the start of a new cycle.

Preparation of pure culture. It is known that many microorganisms very active in activated sludge are generally slow growers and also mostly do not grow on agar plates. Therefore, the microorganisms isolated from the activated sludge in this study are predominantly the fast growing bacteria. A 10 mL of sample was added to 90 mL of a sodium tripolyphosphate solution [STS (5 mg/L)] and homogenized rapidly in a countertop blender for at least three minutes. Subsequent 1/10 dilutions, which ranged from 10^-2 to 10^-6, were made with the STS and spread on agar plates. For culturing the dominant bacteria, a general-purpose media nutrient agar (Difco) was chosen. Although probably many strains of organisms present in activated sludge systems could not grow on general nutrient agar media, a special medium, called feed agar, was prepared and used to compare the growing predominant organisms with those growing on nutrient agar. The feed agar had the same relative composition as the feed used for the CFSTR and the SBR systems, but was more concentrated than the feed by a factor of ten.

Two sets of nutrient agar plates and two sets of feed agar plates were spread-plated with each sample dilution and incubated at room temperature (varied from 20ºC to 25ºC) for five to seven days. After incubation, the colonies that appeared visually dissimilar were chosen, counted, and transferred to fresh agar plates. Identification of microorganisms did not commence until it was evident that a pure culture had been obtained. Each pure culture was also transferred to tryptone glucose extract broth or nutrient broth for preservation in case further testing was required.

Identification of pure culture. The major identification tools used in this study were the API 20E system and the API Rapid NFT system. When the gram-negative pure culture was obtained, it was first tested by both API systems incubating at the same temperature it was isolated. However, many other additional tests are always required for further identification tests of the microorganisms. Through the additional use of Bergey's Manual (Bergey, 1986), The Prokaryotes (Starr and Stolp, 1981), and other references (Leifson, 1960; Lighthart and Loew, 1972; Collins and Lyne, 1976; Ward et al. 1986; Atlas et al. 1988; Labeda, 1990), many traditional morphological, biochemical, and physiological tests were selected for this study. The API biochemical test results were then used with the additional traditional tests to collect necessary data needed for an exact identification. Some of these methods used for the identification of microorganisms are listed in Table 4. Since the API 20E system does not test very well for gram-positive organisms, many conventional tests described in Bergey's Manual have been used for the identification. A Phillips 300 Transmission Electron Microscope and a Phillips 301 Transmission Electron Microscope were further applied to verify the identification of many predominant genera.

Many of the gram-negative and gram-positive isolates were also sent directly to apiLAB, a division of Analytab Products, Inc., for verifying their identities with the results obtained in our laboratory.

a. Stability tests

CFSTR system. After an approximately two-month adaptation period for the CFSTR system, stability tests were performed for five consecutive days to confirm that a stable operation (steady state) had been reached. Table 5 shows the average characteristics of the feed and the effluent of CFSTR system. From the data shown in Table 5, it is evident that this system operated normally with respect to substrate removal and nitrification. The pH value in the effluent was above 6.0, and the dissolved oxygen in the aeration basin was between 5.6 mg/L and 6.4 mg/L. Therefore, the system was determined to be stable.

SBR system. At the time of these investigations, the SBR system had been operating for more than 6 months in the laboratory. Steady-state tests were performed for 4 consecutive days to confirm that a steady state operation of the SBR system had been reached. Table 6 shows the average values obtained from the steady state tests. From the data shown in Table 6, it is evident that the SBR system also operated normally with respect to substrate removal and nitrification. The pH value in the effluent was above 7.0, and the dissolved oxygen in the aeration basin was about 6.5 mg/L. Therefore, the system was determined to be stable.

b. Bacteriology of CFSTR

The predominant microorganisms isolated from the CFSTR system are presented in Table 7 and Table 8. Brevibacterium acetylicum and Pseudomonas vesicularis were two of the most dominant species isolated on both nutrient agar and feed agar. The bacterial count of B. acetylicum on nutrient agar (34 x 10⁶ cfu/mL) was about the same as that on feed agar (36 x 10⁶ cfu/mL). However, the count of P. vesicularis on nutrient agar was much higher than that on feed agar.

Similar results were observed for other less dominant bacteria. For example, Aeromonas hydrophila, Acinetobacter calcoaceticus, Pasteurella aerogenes, Achromobacter group V D, and one Arthrobacter sp. were only found in nutrient agar, while P. maltiphila, Comamonas testosteroni, and Acaligenes faecalis were only isolated from feed agar.

Some other bacteria such as Arthrobacter spp. II and III, Corynebacterium sp. I, Ps. putrefaciens, Ps. pickettii, A. caviae, V. fluvialis, and F. indolotheticum were all isolated from both agars, although they were not very dominant in the CFSTR system. Therefore, considering the whole population structure of the activated sludge in CFSTR, there was no apparent or significant difference between the results obtained from nutrient agar and those from feed agar.

c. Bacteriology of SBR

The predominant microorganisms isolated from the SBR system are listed in Table 9, which indicates that gram-positive organisms such as members of the genus Arthrobacter appeared to be very significant and predominant in the SBR system. It is obvious that most gram-positive strains were isolated from both nutrient and feed agars, but none of the same gram-negative bacteria were isolated from both agars. Aeromonas caviae, Comamonas testosteroni, Pseudomonas maltiphila, and Flavobacterium indolotheticum were only isolated from nutrient agar, and Achromobacter group V D, Pseudomonas diminuta, and Aeromonas hydrophila were only found in feed agar. Therefore, except for the gram-positive genera, there was a significant difference between the results obtained from nutrient agar and feed agar in the SBR system. It may be explained that some gram-negative microorganisms active in the activated sludge of SBR system can be obtained depending on the type of agar plates. However, these bacteria may not be very significant and predominant in the SBR system.

API 20E system and API Rapid NFT system can be applied for the identification of the dominating microorganisms, especially gram-negative ones, in the activated sludge wastewater treatment system. Since both API identification systems mostly can identify the gram-negative microorganisms in activated sludge only at the genus level, many additional morphological, biochemical, and physiological tests are always required to further identify them to the species level. In order to obtain a comprehensive view of the total microflora, it would be advisable to use molecular methods, such as 16S rRNA determinations.

The food:microorganism (F:M) ratio and dissolved oxygen concentration is always assumed to be relatively constant in a CFSTR, therefore the dominating microbial population in the system should remain constant. This can be seen from the results of this study. The predominant microbes isolated from the CFSTR have shown more diversities than from the SBR (see Table 6, Table 7, and Table 8). For CFSTR system, there was also no apparent or significant difference in the microbial population grown on either nutrient or feed agar. For instance, Brevibacterium acetylicum, Arthrobacter spp. II and III, Corynebacterium sp. I, Pseudomonas vesicularis, Ps. putrefaciens, Ps. pickettii, A. caviae, V. fluvialis, and F. indolotheticum were isolated from both nutrient agar and feed agar, and it could be concluded that these microorganisms must be active and significant in the lab-scale CFSTR system. Especially, Brevibacterium acetylicum and Pseudomonas vesicularis have been known to be two of the most dominant species isolated on both nutrient agar and feed agar. This also means that both species could be playing very important roles on the substrate removal in the continuous-flow-type of wastewater treatment system, which has a high substrate removal rate and complete nitrification.

In SBR, the microorganisms grow under a high F:M ratio condition when the feed enters the reactor and the aeration starts. As time passes, the F:M ratio decreases, therefore the system is never really at equilibrium. Competition for food becomes more intense as the F:M ratio decreases. However, the substrate removal rate was still very high and nitrification was completed in this system. According to the results of this study, there were fewer species of microorganisms isolated from the SBR system than from the CFSTR system. Except for the gram-positive genera, there was a significant difference between the predominant microorganisms obtained from nutrient agar and feed agar in the SBR system. Therefore it can be concluded that some gram-positive strains, such as Arthrobacter, appear to be very significant and predominant in the lab-scale SBR system. The gram-positive genera could be playing very important roles on the substrate removal in the batch-type of wastewater treatment system.

Allen, L.A. (1943). The bacteriology of activated sludge. Journal of Hygiene 43:424.

Analytab Products, Inc. API 20E analytical profile index. Plainview, New York, USA.

Anderson, R. and McCoy, E. (1963). Floc-forming bacteria from activated sludge. Bacteriology Proceedings. p. 162.

Atlas, R.M.; Brown, A.E.; Dobra, K.W. and Miller, L. (1988). Experimental microbiology - Fundamentals and applications. MacMillian, New York, USA. 618 p.

Collins, C.H. and Lyne, P.M. (1976). Microbiological methods. Butterworths, Boston, USA. 450 p.

Dias, F.F. and Bhat, J.V. (1964). Microbial ecology of activated sludge. I. Dominant bacteria. Applied Microbiology 12:412.

Franson, M.A.H.; Greenberg, A.F.; Trussell, R.R. and Clescerl, L.S. (1985). Standard methods for the examination of water and wastewater. American Public Health Association, Washington, D.C., USA. 1268 p.

Holt, J.G., ed. (1986). Bergey's manual of systematic bacteriology. The Williams and Wilkins Co., Baltimore, USA.

Labeda, D.P. (1990). Isolation of biotechnological organisms from nature. McGraw-Hill Publishing Company, New York, USA. 322 p.

Leifson, E. (1960). Atlas of bacterial flagellation. Academic Press, London.

Lighthart, B. and Loew, G.A. (1972). Identification key for bacteria cultures from an activated sludge plant. Journal of Water Pollution Control Federation 44:2078.

Lighthart, B. and Oglesby, R.T. (1969). Bacteriology of an activated sludge wastewater treatment plant - A guide to methodology. Journal of Water Pollution Control Federation 8:267.

McKinney, R.E. and Horwood, M.P. (1952). Fundamental approach to the activated sludge process. I. Floc-Producing Bacteria. Sewage and Industrial Wastes 24:117.

Starr, M.P. and Stolp H. eds. (1981). The Prokaryotes - A handbook on habitats, isolation and identification of bacteria, Vol. I and Vol. II. Springer-Verlag, New York.

Unz, R.F. and Dondero, N.C. (1970). Nonzoogloeal bacteria in wastewater zoogloeas. Water Research 4:575.

Ward, N.R.; Wolfe, R.L.; Justice, C.A. and Olson, B.H. (1986). The identification of gram-negative, nonfermentative bacteria from water: problems and alternative approaches to identification. Advances in Applied Microbiology 31:293-365.