The pig industry plays a crucial role in the agricultural sector in Taiwan. However, nitrogen and phosphorus in pig farm effluent must be controlled and decreased to avoid eutrophication of surface waters [7]. Average concentrations of both ammonia and nitrogen and total Kjeldahl nitrogen in raw piggery wastewater after solid–liquid separation are 229 ± 33.5 and 2736 ± 3.9 mg L−1, respectively [17]. Wastewater treatment facilities in Taiwan have attained approximately 40% nitrogen removal from piggery wastewater [17]. Thus, there is a need to introduce heterotrophic microorganisms into piggery wastewater treatment facilities to improve nitrogen removal efficiency.

Simultaneous biologic nutrient removal was thoroughly investigated by assessing autotrophic denitrification, heterotrophic nitrification, and phosphorus removal in full-scale wastewater treatment systems [11]. Parts of simultaneous nitrification and denitrification (SND) rely somewhat on concurrent aerobic ammonia oxidation and anaerobic denitrification by heterotrophic bacteria, which could use poly-β-hydroxybutyrate as an electron donor [22]. Autotrophic Nitrosomonas spp. was shown to be capable of aerobic deammonification (NH +4 → NH2OH → NO −2 → N2O → N2) [2, 15], whereas heterotrophic bacteria were reported to be capable of heterotrophic nitrification coupled with aerobic denitrification (NH +4 → NO −2 → N2O → N2) [13, 25]. Heterotrophic nitrification is suitable only for organic wastewaters with high ratios of cyclooctadiene (COD) to nitrogen [24]. Most heterotrophic-nitrifying bacteria are capable of aerobic denitrification, including Arthrobacter sp. [3, 26], Thiosphaera pantotropha (known as Paracoccus denitrificans) [8, 13, 16], P. denitrificans [14], and Alcaligenes faecalis [12, 25].

Peptone [7, 23] and ammonia sulphate [9] have been used as decreased nitrogen sources to investigate heterotrophic nitrification caused by certain microorganisms [3]. Doxtader and Alexander [6] identified a possible role for β-alanin in the heterotrophic nitrification pathway of Aspergillus flavus. Thus, β-alanin has been used as the sole source of carbon and nitrogen for isolating heterotrophic nitrifiers [3]. Moreover, pyruvic oxime (7 mM), combined with 0.05% (w/v) yeast extract, has also used to identify the nitrification capability of six denitrifying Pseudomonas strains [4]. This study used ammonium acetate as the sole source of carbon and nitrogen to isolate heterotrophic nitrifiers and verified their ammonia oxidation and nitrite decrease capability for further application in piggery wastewater treatment systems in Taiwan.

Materials and Methods

Isolation

The medium (A) (Table 1) was used to isolate heterotrophic nitrifiers [18, 19, 20]. The sole carbon source in the medium was 0.1% (w/v) ammonium acetate [3]. Inocula for isolation were obtained from raw piggery wastewater, soils, and anaerobic and activated sludge from piggery wastewater treatment facilities. Diluted piggery wastewater, soil, and sludge samples of 0.1 mL 1% (w/v or v/v) were transferred individually into Erlenmeyer flasks containing 150 mL autoclaved sterile medium (A). After 48 hours of incubation at 30°C, some of the culture suspensions in the inoculated flasks were tested for the presence of nitrite [5]. Once nitrite was detected, the remaining 0.1 mL of the culture suspension was streaked on tryptic soy agar plates (Difco) to obtain isolated single colonies.

Table 1 List of media used in this study
Full size table

Qualitative assessment of aerobic ammonia oxidation and nitrate decrease

Aerobic ammonia oxidation: The medium (A) agar plates were employed to verify their heterotrophic ammonia oxidation capability. The isolates were inoculated on agar plates containing 18.6 mmol L−1 NH +4 and incubated at 30°C under aerobic conditions. Nitrite production from the inoculated plates was confirmed after 48 hours of incubation by adding sulfanilamide and N-(1-naphthyl)ethylenediamine reagents and observing color formation [5]. In each case, medium plates without inoculation were used as sterility controls. Nitrate decrease. The medium (C) agar plates containing KNO3 (Table 1) were inoculated with the isolates and incubated at 30°C under aerobic and anaerobic conditions to verify their nitrate decrease capability. Nitrite production was confirmed by observing color formation on the inoculated plates after 72 hours of incubation [5].

Identification of the isolate

The isolates were identified by sequencing their 16S rDNA. Universal primers 16f27 (5’-AGAGTTT GATCMTGGCTCAG-3’) and 16r1488 (5’-CGGTTACCTTGTTAG GACTTCACC-3’) [1] were used to amplify their 16S rDNA by polymerase chain reaction (PCR) protocol. Finally, the PCR product was verified using an agarose gel based on the size (1.5 kb). The PCR product was cut out from the agarose gel and purified by a QIAquick Gel Extraction Kit (Qiagen). Protech Technology sequenced the resulting 16S rDNA product. Finally, the 16S rDNA sequence of the isolate was compared with that of other microorganisms by way of BLASTn (http://www.ncbi.nlm.nih.gov/BLAST/Blast.cgi).

Biomass measurement

The AS-1 isolate was enriched in medium (A) until optical density (OD)600 = 0.55, and the cell suspension of AS-1 was then series-diluted by mixing half-and-half with sterile phosphate buffer (pH 7.5) to a final volume of 20 mL. The OD of each dilution of a cell suspension then was measured immediately at 600-nm wavelength (i.e., OD600). The dry weight of the AS-1 strain was determined using the method of Koch [10]. The linear regression of the OD600 versus dry weight of cell mass was calculated based on the corresponding OD600 values and dry weights of the strain AS-1 cell mass.

Quantitative assessment of nitrite decrease

Microorganisms were isolated and purified by employing streaking plate techniques from the cell suspension grown in the medium (A) in which nitrite was produced. Nitrite concentrations in the purified isolates then were assessed. Next, 4.5 mL cell suspension was inoculated into 250-mL Erlenmeyer flasks with 150 mL sterile medium (B) and incubated aerobically at 30°C in triplicate. The flasks containing uninoculated sterile medium served as the sterility controls. Samples were examined at 27- and 15-hour intervals to determine nitrite decrease using nitrite color reagents as previously described [5].

Assessment of heterotrophic nitrification and aerobic denitrification

Crimp-sealed 118-mL serum bottles containing 80 mL medium were prepared in triplicate (Table 1). These bottles were evacuated and the headspace pressurized with 69 kPa pure oxygen (approximately 98% oxygen analyzed by gas chromatography (GC) coupled with a thermal conductivity detector (TCD) (PerkinElmer) three times before autoclaving. After autoclaving, VS-salt, PS-1, and V8 solutions were added to the bottles at room temperature. For each isolate, 2.4 mL culture suspension (OD600 = 0.18) was inoculated into the triplicate bottles and incubated at 30°C. Samples from the bottles were measured periodically to determine OD600, NH +4 , NO −2 , N2O, and N2 gas concentrations.

Analysis

OD600 was determined by spectrophotometry (Spectronic 20+, Milton Roy) at a wavelength of 600 nm. Moreover, bacterial suspensions were centrifuged at 3600 × g for 20 minutes to allow for measurement of ammonia and nitrite levels in the supernatant before assay was performed. Ammonia concentration was then determined by Nessler assay at a wavelength of 425 nm with a sensitivity for ammonia detection of 400 μg N L−1 [5]. Nitrite concentration was determined by colorimetry at a wavelength of 543 nm [5]. To determine O2 and N2, 20-μL samples (injection volume) were removed from the headspace of the serum bottles using a pressure-lock syringe (Supelco) and analyzed by GC–TCD [19]. To measure N2O, 500-μL gas samples were removed from the headspace of the serum bottles, once again using a pressure-lock syringe (Supelco) and analyzed by GC with an electron-capture detector (PerkinElmer) [21]. O2, N2, and N2O were quantified by the methods proposed by Su et al. [19, 21].

Results and Discussion

Microorganisms

Nitrite was found only in the culture suspension in the flasks of medium (A) inoculated with diluted activated sludge. Only two pure cultures, AS-1 and AS-2, were obtained by way of streaking plate techniques from this culture suspension.

Qualitative assessment of aerobic ammonia oxidation and nitrate decrease

The purpose of the qualitative assessments was to test the capabilities of aerobic NH +4 oxidation and NO −3 decrease for the isolates. Observation results revealed that AS-1 grew faster than AS-2 on plates containing NH4Cl. Almost no nitrite was detected on plates inoculated with AS-1, but nitrite was detected on plates inoculated with AS-2. The result may imply that AS-1 is capable of ammonia oxidation. AS-1 still grew faster than AS-2 on plates containing nitrate under aerobic conditions, and more nitrite was detected on AS-1 inoculation plates after 48 hours than on AS-2 inoculation plates. The observation results may imply that AS-1 can use nitrate as the electron acceptor under aerobic conditions. Neither strain grew aerobically on plates containing NH4Cl, but lacking additional organic carbons such as acetate. Consequently, both strains were heterotrophic bacteria.

Ammonia oxidation by strains AS-1 and AS-2 under aerobic conditions

A bacterial suspension (7.5 mL; AS-1 OD600 = 0.09 or AS-2 OD600 = 0.05) of the strains was inoculated into 500-mL Erlenmeyer flasks containing 250 mL sterile medium (D) and incubated aerobically at 30°C in triplicate. The samples were periodically examined for NH +4 , NO −2 , and OD600. Experimental results demonstrated that the OD600 of strains AS-1 and AS-2 reached a stationary phase after 57 and 45 hours, respectively. AS-1 completely oxidized 2.7 ± 0.1 mmol L−1 NH +4 by 33 hours (Fig. 1a), and the removal rate of NH +4 was 0.064 mmol L−1 NH +4 -N h−1 under aerobic conditions. However, strain AS-2 oxidized 2.6 ± 0.1 mmol L−1 NH +4 by 153 hours (Fig. 1b), and the removal rate of NH +4 was approximately 0.013 mmol L−1 NH +4 -N h−1 under aerobic conditions. AS-1 consistently maintained superior ammonia oxidation capability to that of AS-2. Nitrite was not detected after 45 hours in the AS-1 culture suspension. Consequently, AS-1 was selected for advanced study of heterotrophic nitrification.

Fig. 1
figure 1

Changes in NH +4 (open square), NO −2 (solid square), and OD600 (solid circle) by oxidizing 2.8 mmol L−1 NH +4 with (a) AS-1 and (b) AS-2 in triplicate under aerobic conditions.

Full size image

Nitrite decrease by strains AS-1 and AS-2

A cell suspension (4.5 mL; AS-1 OD600 = 0.42 or AS-2 OD600 = 0.25) of the isolates was individually inoculated into 250-mL Erlenmeyer flasks containing 150 mL medium (B) in triplicates. The flasks then were incubated aerobically at 30°C. The OD600 at 42 hours was 0.10 ± 0.01 (initial OD600 = 0.02 ± 0.00) and 0.07 ± 0.01 (initial OD600 = 0.01 ± 0.00) for AS-1 and AS-2, respectively. The initial nitrite reading in the culture suspension was 0.33 ± 0.02 and 0.35 ± 0.01 mmol L-1for AS-1 and AS-2, respectively. Nitrite was significantly decreased in all the AS-1 culture suspension samples by 27 hours (NO −2 = 0.08 ± 0.04 mmol L−1) and was undetectable by 42 hours. However, nitrite decrease was slower in all AS-2 culture suspension samples than in the AS-1 samples.

Identification of AS-1

AS-1 (here termed “SU3”) was Gram-negative, rod-shaped, and identified as P. alcaligenes (97% identity) based on comparison of its 16S rDNA sequence with that of P. alcaligenes isolate LB19 using BLASTn (http://www.ncbi.nlm.nih.gov/BLAST/).

Herotrophic nitrification and aerobic denitrification by AS-1

The results of these time course experiments indicated that 2.5 ± 0.2 mmol L−1 NH +4 was totally removed (Fig. 2), and 24.3% of the oxygen in the headspace was consumed by 58 hours (Fig. 3). The removal rate for AS-1 was 0.0034 mmol L−1 NH +4 h−1. Moreover, the OD600 at 58 hours was 0.40 ± 0.02 and equaled 1.97 ± 0.08 mg cell using the linear regression equation. The OD600 values were converted to dry cell weights using a linear regression equation of Y = 4.9742X – 0.02365 (where Y represents the cell mass in milligrams, and X represents the OD600 value). Thus, a removal rate of 1.75 mmol NH +4 (g cell)−1 h−1 was achieved for AS-1. No nitrite was detected during the experimental period; nitrogen gas was produced by 12 hours; and nitrous oxide was detected at 21 hours. Growth of AS-1 entered a stationary phase after 95 hours. All NH +4 was removed after 209 hours. Eventually, 1.5 ± 0.5 mmol L−1 N2 and 0.2 ± 0.0 mmol L−1 N2O were produced.

Fig. 2
figure 2

Changes in NH +4 (open square), NO −2 (closed square), N2O (closed triangle), N2 (closed inverted triangle), and OD600 by oxidizing 2.8 mmol L−1 NH +4 with the strain AS-1 in aluminum crimp-sealed serum bottles.

Full size image
Fig. 3
figure 3

Change in oxygen in the headspace by oxidizing 2.8 mmol L−1 NH +4 with the strain AS-1 in aluminum crimp-sealed serum bottles.

Full size image

The oxygen levels in the headspace of the crimp-sealed serum bottles remained at 24.0 ± 1.0 mmol L−1. Ammonium ion was completely removed, and oxygen concentration within the closed serum bottles was decreased markedly (37 ± 1.5 mmol L−1 to 28 ± 2.0 mmol L−1) by 58 hours (Figs. 2 and 3). This decrease implied that ammonium ion was oxidized aerobically. However, the decrease of nitrite to produce nitrous oxide or nitrogen seemed to consume far less oxygen than ammonium ion oxidation (Figs. 2 and 3). Nitrogen production suggested that AS-1 was capable of heterotrophic nitrification. The removal rate of ammonia was proportional to the growth of the AS-1 (Fig. 2). This study confirmed AS-1 to be a heterotrophic nitrifier.