Limited impact of free ammonia on Nitrobacter spp. inhibition assessed by chemical and molecular techniques

Bioresource Technology 101 (2010) 4513–4519

Contents lists available at ScienceDirect

Bioresource Technology journal homepage: www.elsevier.com/locate/biortech

Limited impact of free ammonia on Nitrobacter spp. inhibition assessed by chemical and molecular techniques Shawn Hawkins a,*, Kevin Robinson b, Alice Layton c, Gary Sayler c a

The University of Tennessee, Department of Biosystems Engineering and Soil Science, 310 Biosystems Engineering And Environmental Sciences Office, 2506 E.J. Chapman Drive, Knoxville, TN 37996-4531, United States b The University of Tennessee, Department of Civil and Environmental Engineering, 219A Perkins Hall, 1506 Middle Drive, Knoxville, TN 37996-2010, United States c The University of Tennessee, The Center for Environmental Microbiology, 676 Dabney-Buehler Hall, 1416 Circle Drive, Knoxville, TN 37996-1605, United States

a r t i c l e

i n f o

Article history: Received 14 January 2009 Received in revised form 13 January 2010 Accepted 13 January 2010 Available online 13 February 2010 Keywords: Nitrite oxidation Nitrobacter Free ammonia rRNA transcript

a b s t r a c t Free ammonia has long been identified as a nitrite oxidation inhibitor. However, past attempts to use this compound to eliminate nitrite oxidation and thereby promote more efficient nitrogen removal strategies during biological wastewater treatment have often failed. Additionally, contradictory results exist in the literature where direct measurements of free ammonia inhibition of nitrite oxidation have been reported. In this study, suspended biomass samples (nitrifier enriched activated sludge) were collected from a bench scale nitrification reactor with Nitrobacter spp. as the dominant nitrite oxidizer and subjected to batch respirometric experiments designed to quantify free ammonia inhibition of nitrite oxidization. A variety of data including ammonia, nitrite, and nitrate conversion rates, oxygen consumption rates, and Nitrobacter ribosomal RNA transcript abundance, a molecular indicator of growth activity, were used to assess nitrite oxidation and growth activity. Both the traditional and molecular activity assessments indicated that free ammonia had a limited inhibitory effect on Nitrobacter spp. In fact, the pH changes necessary to induce high free ammonia concentrations (>10 mg-N/L) had a demonstrably more important inhibiting effect on nitrite oxidation than free ammonia. In contrast, during high ammonia oxidizing activity (5.3 mg-N/L/h), low nitrite oxidation rates (0.2 mg-N/L/h) and severely impaired Nitrobacter spp. growth activity, indicated by a low abundance of the Nitrobacter spp. ribosomal gene transcript relative to the ribosomal gene (0.08), were measured. The findings suggest that pH changes and ammonia oxidizing bacteria activity are more important factors limiting Nitrobacter spp. mediated nitrite oxidation, rather than the free ammonia concentration. Ó 2010 Elsevier Ltd. All rights reserved.

1. Introduction The advantages for inhibiting nitrite oxidation during wastewater treatment have been known for over thirty years (Voets et al., 1975) and various alternative nitrogen removal strategies continue to stimulate interest in this topic (Fux et al., 2003; Ganigue et al., 2009; Hellinga et al., 1998; Kuenen and Jetten, 2001; Liu et al., 2008; Philips et al., 2002; Sliekers et al., 2003). These techniques promote the use of nitrite as a terminal electron acceptor, for example during anaerobic ammonia oxidation (Xu et al., 2010) or denitritification (Daniel et al., 2009). The principle benefits of such processes include lowering reactor oxygen demand and biomass generation rates and increasing nitrogen removal rates. However, these benefits can be fleeting because maintaining nitrite oxidation inhibition over extended periods of time has proven challenging (Fux et al., 2004; Turk and Mavinic, 1989). * Corresponding author. Tel.: +1 865 974 7722; fax: +1 865 974 4514. E-mail address: [email protected] (S. Hawkins). 0960-8524/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.biortech.2010.01.090

A widely referenced nitrite oxidation inhibition method spurred by the research of Anthonisen et al. (1976) involves intentionally increasing reactor free ammonia concentrations by manipulating reactor pH values or spiking ancillary waste streams with high ammonia concentrations (e.g. digester effluent) into operating reactors. Over the past 30 years, many researchers have reported successful inhibition of nitrite oxidation using free ammonia; however, other researchers have noted that nitrite oxidizers ‘‘acclimate” to high free ammonia concentrations, thereby limiting the long-term success of alternative nitrogen removal strategies that suppress nitrite oxidation (Chung et al., 2005; Fux et al., 2003; Sliekers et al., 2003; Turk and Mavinic, 1989; Villaverde et al., 2000). As a result, reported free ammonia concentrations that produce nitrite oxidation inhibition are wide-ranging and contradictory. For example, researchers have reported significant nitrite oxidation inhibition as low as 1 mg-N/L (Abeling and Seyfried, 1992; Anthonisen et al., 1976), while others (Chung et al., 2005; Fux et al., 2003) have observed high nitrite oxidizing activity at free ammonia concentrations over 50 mg-N/L. A sampling of these

4514

S. Hawkins et al. / Bioresource Technology 101 (2010) 4513–4519

1.5 mM KH2PO4, 0.75 mM MgSO4, 0.20 mM CaCl2, 16.6 lM EDTA, 9.9 lM FeSO4, and 0.5 lM CuSO4. The reactor pH, dissolved oxygen (DO), and temperature were maintained at 7.2 ± 0.1, P3 mg/L, and 30 ± 1 °C, respectively. The reactor was operated for one year prior to these experiments with very good nitrification efficiency (Hawkins, 2005). The dominant nitrite oxidizing bacteria (NOB) in the reactor biomass were previously identified as Nitrobacter spp. (Hawkins et al., 2006).

Free Ammonia Concentration, mg-N/L

100

10

2.2. Short-term batch nitrite oxidation inhibition experiments using free ammonia

1

Manuscrips Reporting Free Ammonia Acclimation Values for Nitrite Oxidation

Manuscrips Reporting Free Ammonia Inhibition Values for Nitrite Oxidation

Wong-Chong and Loehr, 1978 Suthersan and Ganczarczyk, 1986 Turk and Mavinic, 1989 Fux et al., 2003 Yang et al., 2004 Chung et al., 2005

Bae et al., 2002

Yun and Kim, 2003

Mauret et al., 1996

Balmelle et al., 1992

Yang and Alleman, 1992

Abeling and Syfreid, 1992

Ford et al., 1980

Turk and Mavinic, 1987

Anthonisen et al., 1976

0.1

Fig. 1. Summary of conflicting peer reviewed literature reports of free ammonia inhibition of nitrite oxidation (Ford et al., 1980; Mauret et al., 1996; Suthersan and Ganczarczyk, 1986; Turk and Mavinic, 1987; Wong-Chong and Loehr, 1978; Yang and Alleman, 1992; Yang et al., 2004).

conflicting results is provided in Fig. 1. One glaring omission in most nitrite oxidizer free ammonia inhibition studies has been the lack of a control for ammonia oxidizing activity. Thus, many studies did not independently assess the impact of free ammonia concentration and ammonia oxidizing bacteria (AOB) activity on nitrite oxidation rates. Researchers that selectively inhibited AOB activity (Chung et al., 2005; Fux et al., 2003) did not observe significant levels of free ammonia inhibition of nitrite oxidation. Simm et al. (2005) also reported difficulty inhibiting nitrite oxidation using free ammonia and surmised that ‘‘free ammonia inhibition is one of the most misunderstood topics in environmental engineering and science.” These researchers used a coarse measure of nitrifier population and molecular activity (% total RNA) to bolster their conclusion that free ammonia was a relatively unimportant inhibitor of Nitrospira spp. mediated nitrite oxidation (Simm et al., 2006). This study extends the free ammonia inhibition work by targeting another nitrite oxidizing bacterial genus, Nitrobacter. Results herein document the limited impact of free ammonia on Nitrobacter spp. using batch tests that included selective inhibition of ammonia oxidizing activity. Traditional measures of activity, including ammonia, nitrite, and nitrate conversion rates and oxygen consumption rates, as well as a sensitive molecular measure of growth activity, the ribosomal RNA transcript abundance (Hawkins et al., 2008), were used to reach this conclusion. 2. Methods 2.1. Reactor Samples for the batch experiments described herein were collected from a 10 L fill and draw, complete mix, bench scale nitrification reactor (BSNR) that continuously received 1 liter/day of influent containing: 54 mM (NH4)2SO4 (1500 mg-N/L), 1.5 mM K2HPO4,

A BI-2000 electrolytic respirometer (Bioscience Inc., Bethlehem, PA) was used to monitor oxygen consumption (nitrite oxidation) during four short-term (<10 h) batch inhibition experiments at 30 ± 1 °C (Hawkins et al., 2008). Nitrite was added to five one-liter vessels (final concentration 100 mg-N/L) containing diluted (50%) BSNR mixed liquor (75 mg/L volatile suspended solids) and nitrite oxidation was monitored over approximately 4.5 h without ammonia present. This period of uninhibited nitrite oxidation activity was quantified with a linear regression fit (OUR1) of cumulative oxygen uptake over time. Ammonia was then added to the vessels at different concentrations using a concentrated stock solution along with 50 lM allythiourea (ATU) which quickly and selectively inhibits ammonia oxidizing bacteria (Ginestet et al., 1998). The addition of ATU prevented the biological oxidation of ammonia that would otherwise lower the free ammonia concentration during the assay, minimized ammonia and nitrite oxidizer competition for common resources such as oxygen, and assured that oxygen consumption was due solely to nitrite oxidation. Aliquots of 5 M NaOH were also added along with the ammonia and ATU during three short-term batch experiments to increase the solution pH and thereby obtain free ammonia concentrations as high as 181 mg-N/L (Anthonisen et al., 1976). Following the ammonia, ATU, and any NaOH additions, the vessels were allowed to acclimate for 30 min prior to quantification of NOB inhibition. A linear regression fit of cumulative oxygen uptake over time (OUR2) was determined and fractional inhibition values were then calculated as: (OUR1  OUR2)/OUR1. The oxygen utilization rates observed during these experiments were linear, with R2 values exceeding 0.95. To quantify the inhibitory effect of elevated pH versus the added free ammonia, additional control experiments were conducted without ammonia or base additions, or solely with NaOH additions that resulted in pH changes. 2.3. The effect of high AOB activity on nitrite oxidation rates Two replicate, long-term (96 h) respirometric inhibition experiments were performed at 30 ± 1 °C to contrast the relative inhibitory impact of a moderate free ammonia concentration (2.6–2.9 mg-N/L) and a high ammonia oxidation activity level on nitrite oxidation rates at saturating DO concentrations. For these experiments, BSNR mixed liquor was seeded into five one-liter respirometric vessels containing 100 mg NO 2 -N/L, in similar fashion to the short-term inhibition experiments. However, assays with different media recipes were used to differentiate between the potential inhibiting effect of ATU (50 lM), free ammonia (2.6–2.9 mg-N/L), high AOB activity (no ATU present), and a buffer (200 mM PO3 4 ) required to accommodate the acidity produced by ammonia oxidation. One vessel, which received ammonia and contained uninhibited ammonia oxidizing bacteria (no ATU present), was observed over the course of 4 days. In the first long-term experiment, ammonia, nitrite, and nitrate conversion rates were measured to distinguish between NOB and AOB activity trends based on oxygen consumption rates. In the second experiment, the relative inhibitory impact of high AOB activity and the presence of free ammonia on nitrite oxidation

4515

S. Hawkins et al. / Bioresource Technology 101 (2010) 4513–4519

2.4. Nitrogen analyses Ion chromatography was used to quantify total ammonia concentrations in filtered samples (AcrodiscÒ, 0.45 lm; Pall Gelman Laboratory, Ann Arbor, MI) preserved with concentrated sulfuric acid (pH < 2). A DX-500 ion chromatograph (IC) (Dionex Corporation; Sunnyvale, CA) equipped with an IonpacÒ NGI primary guard column, an IonpacÒ CS12 guard column, and an IonpacÒ CS12 cation exchange column was used to perform the analyses. Auto-suppression recycle mode was used with a cation self regenerating suppressor-ultra (CSRS-Ultra) and 20 mM methane-sulfonic acid eluent at 1 mL/min. The same IC was also used to quantify nitrite and nitrate in filtered samples using standard method 4110 B (APHA, 1998). In this case the IC was equipped with an IonpacÒ AS9-HC 4 mm guard and anion exchange columns, and was operated in auto-suppression recycle mode with an anion self regenerating suppressor (ASRS-Ultra) using 9 mM Na2CO3 eluent at 1 mL/ min. All nitrogen analyses were performed in triplicate. 2.5. Nitrobacter ribosomal gene and ribosomal gene transcript quantification Biomass samples were collected from the respirometric vessels at various times during the second long-term inhibition experiment, preserved, and nucleic acids were later extracted in triplicate as previously described (Hawkins et al., 2008). Real Time PCR was subsequently performed to quantify the Nitrobacter spp. rDNA and rRNAt concentrations in these samples (Hawkins et al., 2006, 2008). The detection system targeted a genus specific 16S to 23S intergenic spacer region with primers NITISRf (50 -CCATTCACTATCTCCAGGTC-30 ) and NITISRr (50 -TGATTAGAAAGACCAGCTTGC-30 ) and a fluorescent probe NITISRp [50 -TCGAACCGATAGCGAGGCGG30 ]. Nitrobacter spp. rDNA abundance was quantified in 1:5 dilutions of the sample DNA extracts using a real time PCR reaction containing: 12.5 lL of QuantiTectÒ Probe PCR master mix (Qiagen; Valencia, CA), 5.125 lL of nuclease free water, 400 nM of each primer (1 lL), 150 nM of probe (0.375 lL), and 5 lL of template. The rDNA assay temperature protocol was as follows: 50 °C for 2 min, Taq activation at 95 °C for 15 min, and 40 cycles with melting at 94 °C for 15 s and annealing/extension at 61 °C for 1 min. Nitrobacter spp. rRNAt abundance was quantified in 1:5 dilutions of the sample RNA extracts by real time RT-PCR using the following reaction mix prepared on ice: 12.5 lL of QuantiTectÒ Probe RT-PCR master mix (Qiagen; Valencia, CA), 4.875 lL of RNase free water, 400 nM of each primer (1 lL), 150 nM of probe (0.375 lL), 0.25 lL of RT enzyme mix, and 5 lL of template. The rRNAt assay temperature protocol was as follows: RT reaction at 50 °C for 30 min, Taq activation at 95 °C for 15 min, and 40 cycles with melting at 94 °C for 15 s and annealing/extension at 61 °C for 1 min. Both assays were performed on a MJ DNA Engine Opticon thermocycler in triplicate for each extract using the average fluorescence for cycles 3–7 for baseline subtraction, a fluorescence threshold of 0.005, and external standard curves (Hawkins et al., 2006). 2.6. Statistical analyses Oxygen consumption data were analyzed with linear regressions to define fractional inhibition values as previously reported (Hawkins et al., 2008). Nitrogen conversion rates during designated time periods of the first long-term respirometric inhibition experiment (Experiment 5) were computed non-parametrically using the 95% confidence interval for the slope of a Kendall–Thiel

best fit line (Helsel and Hirsch, 2002). Nitrobacter spp. ribosomal gene and transcript abundance during the second long-term respirometric inhibition experiment (Experiment 6) were statistically compared following log transformation. Normality and homogeneity of the variance of transformed datasets were confirmed with the Shaprio-Wilk and Levene’s tests, respectively. Analysis of variance (ANOVA) was used to determine whether differences existed between the samples collected at different times and from different vessels, and a Tukey Honestly Significant Difference (HSD) post hoc test was used to define statistically different groups of sample results. Statistical analyses were performed with the JMP software package, version 8.0 (SAS Institute Inc., Cary, NC) or the Excel spreadsheet, version 11 (Microsoft Inc., Redmond, CA). 3. Results and discussion 3.1. Short-term batch nitrite oxidation inhibition experiments using free ammonia A typical short-term free ammonia inhibition assay is presented in Fig. 2 to illustrate how fractional inhibition values were calculated using cumulative oxygen consumption data. During this assay, the oxygen consumption rate declined from 9.73 to 2.58 mg/ L/h following the addition of 181 mg-N/L of free ammonia and a pH adjustment from 7.2 to 8.0 using 5 M NaOH. The OUR decline produced a fractional inhibition value of 0.73. This respirometric inhibition assessment technique has been used successfully for several other nitrite oxidation inhibitors (Hawkins et al., 2008) and produces results that are highly correlated with fractional inhibition values computed with nitrite oxidation rates (Hawkins, 2005). In total, twenty such assays were performed during four experiments that included increasingly higher free ammonia concentrations in an attempt to completely inhibit nitrite oxidation (Table 1). Whereas free ammonia concentrations greater than 10 mg-N/ L have previously been reported to completely inhibit nitrite oxidation (Abeling and Seyfried, 1992; Anthonisen et al., 1976; Bae et al., 2002; Balmelle et al., 1992; Yun and Kim, 2003), even at the extraordinarily high free ammonia concentrations of 71.9 and 181 mg-N/L, the fractional inhibition values in this study did not exceed 0.58 and 0.73, respectively (Table 1). Though a 50–60% reduction in nitrite oxidation activity was observed at free ammonia concentrations ranging from 4.9 to 71.9 mg-N/L (Table 1, Experiments 2 and 3), the fact that the inhibition levels were similar over such a broad range of free ammonia concentrations suggested that the pH changes necessary to obtain the desired high

Cumulative Oxygen Consumption, mg/L

was independently assessed by quantifying the abundance of Nitrobacter spp. ribosomal nucleic acid genes (rDNA) and gene transcripts (rRNAt) (Hawkins et al., 2006, 2008).

60 OUR1 = 9.73 mg/L/hr OUR2 = 2.58 mg/L/hr

50 40 30

Add Ammonia & Adjust pH

20

Fraction Inhibition =

10

9.73 - 2.58 = 0.73 9.73

0 0

2

4

6

8

10

Time, hrs Fig. 2. Results for a respirometric inhibition assay conducted at pH 8.0 with a free ammonia spike concentration of 181 mg-N/L.

4516

S. Hawkins et al. / Bioresource Technology 101 (2010) 4513–4519

Table 1 Results for short-term (<10 h) respirometric inhibition experiments using BSNR mixed liquor samples spiked with ammonia.

1

2

3

pH

Ammonia-mg N/L

Adjusteda

Final value

Total

Free

No

7.3 7.2 7.1 7.1 6.9 7.5 7.5 7.5 7.5 7.5 8.0 8.0 8.0 8.0 8.0 7.3 8.0 8.0 8.0 8.0

49 98 147 243 475 196 392 588 787 980 93 185 370 690 913 50 2.5 25 249 2381

0.7 1.2 1.6 2.2 3.3 4.9 10.0 14.6 20.0 24.4 7.5 15.2 29.1 53.0 71.9 0.7 0.2 2.1 21 181

Yes

Yes

4 Yes

Fractional inhibition 0.07 0.09 0.14 0.18 0.19 0.49 0.49 0.47 0.50 0.49 0.52 0.56 0.54 0.62 0.58 0.11 0.69 0.60 0.67 0.73

a Aliquots of 5 M NaOH were added along with the ammonia and ATU during three of the short-term batch experiments to counter the acidifying effect of the added ammonia solution and to increase the pH and thereby further elevate the free ammonia concentrations. During Experiment 4, all of the vessels, except for the vessel with a final pH of 7.3, received NaOH to raise the pH to 8.0.

free ammonia concentrations contributed to the measured inhibition. This was explored further in Experiment 4, with assays that included both very low (0.2 mg-N/L) and very high (181 mg-N/L) free ammonia concentrations at pH 8.0; these assays produced similar fractional inhibition values of 0.69 and 0.73, respectively (Table 1). To assess the influence of pH manipulations a set of control experiments were performed to gauge the relative inhibitory impact of the required pH changes on nitrite oxidation activity.

3.2. Effect of pH increase on nitrite oxidation activity During short-term batch inhibition Experiments 2 through 4 (Table 1), a 5 M NaOH solution was spiked into the vessels to counter the acidifying effect of the added ammonia solution and to raise the assay pH set point to produce higher free ammonia concentrations (Anthonisen et al., 1976). Control assays were performed that included biomass samples which were not subjected to ammonia or base additions, and biomass samples that received only base additions to increase pH. The control assays with no ammonia or base additions resulted in fractional inhibition values near 0 as expected (between 0.07 and 0.03) (Fig. 3). However, when the pH was increased to levels between 7.8 and 8.5, the fractional inhibition values of the control assays increased sharply to between 0.3 and 0.75 (see the Loess Smooth Fit, Fig. 3). These values were similar to those observed over a broad range of free ammonia concentrations (7.5–181 mg-N/L) following a pH set point increase to 8.0 (Table 1). Thus, the effect of free ammonia was limited given the anticipated inhibition caused by pH change alone, and certainly much lower than the previously reported complete inhibition at free ammonia concentrations over 10 mg-N/L (Abeling and Seyfried, 1992; Anthonisen et al., 1976; Bae et al., 2002; Balmelle et al., 1992; Yun and Kim, 2003). However, many past investigations of free ammonia inhibition of nitrite oxidation did not eliminate ammonia oxidizer activity as a control to independently assess the effect of free ammonia.

Fractional Inhibition

Experiment

1.0 0.8 0.6 5M NaOH addition No 5M NaOH Addition Loess Smooth Fit

0.4 0.2 0.0 7

8

9 pH

10

11

Fig. 3. Control inhibition assay results for BSNR biomass samples not subjected to ammonia or NaOH additions, as well as samples subject to NaOH additions that raised the sample pH. A loess continuous fit of the data was generated using SigmaPlot, Version 10 software (Systat Software Inc., San Jose, CA).

3.3. The effect of high AOB activity on nitrite oxidation rates Replicate long-term (96 h) respirometric experiments incorporating an array of assays with different media recipes were performed to compare the relative inhibitory effect of free ammonia and high ammonia oxidizing bacterial activity on nitrite oxidation rates. The potential inhibitory effect of different media recipe components, and differentiation of the inhibiting effect of free ammonia and high AOB activity, were evaluated by conducting five assays (A–E) with and without a buffer (200 mM PO3 4 ), AOB selective inhibitor (50 lM ATU), and free ammonia (2.6–2.9 mg-N/L) (Table 2). The phosphate buffer, which was necessary to neutralize the acidity produced during ammonia oxidation (Grady et al., 1999), was not included in assay A in each long-term experiment. The first stage oxygen uptake rate (OUR1) was higher in assay A (9.42 and 8.99 mg/L/h in Experiments 5 and 6, respectively) than in the remaining assays that included the buffer (8.29 ± 0.15 and 7.80 ± 0.13 mg/L/h in Experiments 5 and 6, respectively) (Table 2). Phosphate buffers have been shown to inhibit certain Nitrobacter agilis enzymes (Yamanaka et al., 1981), but the observed effect herein was a modest 12% decline in the nitrite oxidation rate. The first stage oxygen uptake rates were similar in assay B when the buffer did not contain ATU (8.35 and 7.91 mg/L/h in Experiments 5 and 6, respectively) and in assay C when the buffer did contain ATU (8.12 and 7.62 mg/L/h in Experiments 5 and 6, respectively) (Table 2). This confirmed, as expected, that ATU did not affect nitrite oxidation rates (Ginestet et al., 1998). BSNR biomass was seeded into buffer with ATU in assays C and D, but only assay D included a 200 mg-N/L dose of ammonia at the end of the first stage of the experiment, which produced a free ammonia concentration of 2.6 and 2.9 mg-N/L in Experiments 5 and 6, respectively (Table 2). The fractional inhibition values observed without the ammonia spike in assay C (0.02 and 0.05 in Experiments 5 and 6, respectively) and with the ammonia spike in assay D (0.06 and 0.05 in Experiment 5 and 6, respectively) were near 0 (Table 2). This confirmed that free ammonia had little inhibitory effect on nitrite oxidation, even though a seminal manuscript indicates that nitrite oxidation should be significantly inhibited at free ammonia concentrations between 2 and 3 mg-N/L (Anthonisen et al., 1976). Over 95 h of respirometric data were collected during assay E, which included biomass samples seeded into buffer without ATU (i.e. containing active ammonia oxidizers) and that received a 200 mg-N/L dose of ammonia at 4.5 h. The replicate Experiments 5 and 6 (Table 2) produced identical oxygen consumption rate

4517

S. Hawkins et al. / Bioresource Technology 101 (2010) 4513–4519 Table 2 Media recipes and respirometric data for two replicate long-term (96 h) batch inhibition experiments using BSNR mixed liquor samples. Experiment

Assay

Media recipe a

5

A B C D E A B C D E

6

Buffer

ATU

p p p p

p p

p p p p

p p

b

OUR1d

OUR2d

Fractional inhibitione

9.42 8.35 8.12 8.21 8.47 8.99 7.91 7.62 7.89 7.77

10.2 7.99 8.31 7.73 15.42 10.1 8.28 8.01 7.98 13.98

0.08 0.04 0.02 0.06 1.00 0.12 0.05 0.05 0.05 –

c

Ammonia-mg N/L (total/free)

200/2.6 200/2.6

202/2.9 202/2.9

 BSNR samples (500 mL) were diluted into either 500 mL of deionized water or 200 mM PO3 4 buffer, producing a final concentration of 100 mg NO2 -N/L. 500 lL of 0.1 M allylthiourea (50 lM). c NHþ 4 -N/L (200 mg) was added 4.5 h after the assay began; the experiment continued for 4.5 additional hours for assay A–D, and 90 additional hours for assay E in both experiments (see Fig. 4). The concentrations of free ammonia were calculated using the added ammonia concentration, the measured pH, and published equilibrium equations (Anthonisen et al., 1976). d OUR1 and OUR2 are the oxygen uptakes rates over a 4 h period prior to and following the ammonia spike into assays D and E (e.g. see Fig. 2). e Fractional inhibition values for assays A–D were calculated using the oxygen uptake rates before (OUR1) and after (OUR2) ammonia was spiked into assays D and E (e.g. see Fig. 2). Oxygen consumption rates could not be used to establish the fractional inhibition values for nitrite oxidation in assay E, because both nitrite and ammonia oxidation could have occurred concurrently during state 2 of the experiment. The value for assay E in Experiment 5 was established with the nitrate production rate before and after the ammonia spike. The value for assay E in Experiment 6 could not be determined because nitrogen samples were not collected. a

b

Oxygen Uptake Rate, mg/L/hr

(A)

25

20 Experiment 5 Experiment 6

15

10

5 Add Ammonia 0 1000

200 900 150

Ammonia-N Nitrite-N 800 Nitrate-N Total-N

100

700 50

0 0

25

50

75

Nitrate & Total Nitrogen, mg-N/L

Ammonia and Nitrite, mg-N/L

(B) 250

600 100

Time, hrs Fig. 4. (A) Oxygen uptake rate during Experiments 5 and 6 and (B) soluble nitrogen speciation data during Experiment 5 for assay E (Table 2). Assay E contained a biomass sample in buffer with uninhibited ammonia oxidizing bacteria and received a 200 mg-N/L dose of ammonia at 4.5 h. Error bars reflect triplicate nitrogen analyses of a single sample. Total nitrogen values are the sum and range of the average ammonia, nitrite, and nitrate replicate measurements.

profiles (Fig. 4A). The total consumption of oxygen in both experiments reflected a stoichiometric oxidation of all the added ammonia and nitrite to nitrate (Grady et al., 1999), confirming the high quality of the respirometric data. For 4 h prior to ammonia addition, a steady OUR of 8 mg/L/h (Fig. 4A) occurred concomitantly with a stoichiometric decrease in nitrite and increase in nitrate

concentrations (Fig. 4B). This confirmed high NOB activity prior to the ammonia addition. The oxygen uptake rate increased immediately to 14 mg/L/h and then gradually to 20 mg/L/h following ammonia addition (Fig. 4A). This was concomitant with the beginning of a buildup in the nitrite concentration between 4.6 and 42.9 h from 64 to 223 mg-N/L at a rate with a 95% confidence interval between +3.92 and +4.78 mg-N/L/h (all nitrogen conversion rates are specified as the slope of a Kendall–Theil best fit line). During this same time period, the ammonia concentration decreased from 201 to 10 mg-N/L at a rate with a 95% confidence interval between 5.86 and 4.59 mg-N/L/h, while the nitrate concentration remained essentially the same (between 697 and 717 mg-N/L) with a rate of change between 0.52 and +0.42 mg-N/L/h. Thus, high AOB and low NOB activity occurred between 4.6 and 40 h into the assay (Fig. 4). Between 40 and 55 h the oxygen utilization rate slowed to 5–6 mg-N/L/h, concomitant with depletion of the added ammonia, the beginning of a decrease in the accumulated nitrite, and a resumption of increasing nitrate concentrations (Fig. 4). Thus, during this time frame AOB activity slowed and ended as NOB activity appeared to resume at a lower level than was observed prior to ammonia addition. The OUR continued at the lower 5–6 mg/L/h rate from 55 to 85 h, concomitant with nitrite concentrations declining from 223 down to 25 mg-N/L with a rate between 4.24 and 5.36 mg-N/L/h. During this same time frame the nitrate concentrations increased from 717 to 924 mgN/L at a rate between +4.85 and +5.98 mg-N/L/h (Fig. 4B). The OUR then decreased to near 0 mg/L/h by 95 h, concomitant with nitrite depletion. Thus, NOB were active between 55 and 95 h into the assay, but at a lower level than was observed at the beginning of the experiment, prior to ammonia addition. In summary, the rate of change in the ammonia, nitrite, and nitrate concentrations, along with the oxygen consumption rates and fractional inhibition data collected during Experiment 5, clearly indicate that free ammonia had little inhibitory effect on nitrite oxidation. However, high AOB activity occurred concomitantly with a near complete inhibition of nitrite oxidation, and resulted in a large accumulation of nitrite as the added ammonia was oxidized. Molecular measures of the nitrite oxidizer population and activity levels, namely the Nitrobacter spp. rDNA and rRNAt abundance, were used to independently assess the impact of free ammonia and high AOB activity on NOB activity during a replicate long-term experiment (Table 2, Fig. 5). Nitrobacter rDNA abundance (range

4518

S. Hawkins et al. / Bioresource Technology 101 (2010) 4513–4519

10 rDNA rRNAt rRNAt/rDNA

Nitrobacter rRNAt/rDNA ratio

Nitrobacter rDNA & rRNAt, copies/L

1011

1010

1

109

0.1

108

0.01 Assay A Assay B Assay C Assay D Assay E Assay C Assay D Assay E Assay E Assay E 4.5 hrs 9.0 hrs 26 hrs 67 hrs

Fig. 5. Nitrobacter spp. rDNA and rRNAt concentrations and the rRNAt/rDNA ratio at various points in time during Experiment 6 for assay E (Table 2). Assay E contained a biomass sample in buffer with uninhibited ammonia oxidizing bacteria and received a 200 mg-N/L dose of ammonia at 4.5 h. Three samples were subject to nucleic acid extractions from each vessel at each time point. Each extract was analyzed in triplicate using real time PCR and the average values were plotted. Error bars for the rDNA and rRNAt results reflect the average and range for the three independently extracted samples. The rRNAt/rDNA ratio is the average value of all possible combinations of the rDNA and rRNAt results; error bars are the inter-quartile range of all possible ratio combinations.

5.67 ± 0.5 to 8.85 ± 1.9  109 copies/L) was not significantly different in any of the assays at any time (ANOVA of log transformed data; p > 0.05). This confirmed the low growth rate for Nitrobacter spp. and the fact that the NOB population level serves as a poor short-term inhibition indicator (Hawkins et al., 2008). The Nitrobacter rRNAt abundance at 4.5 h was not significantly different in any of assays (range 1.4 ± 0.1  1010 to 7.4 ± 1.7  109 copies/L), nor was the abundance in assays with inactive AOB (C and D) at 9 h (6.5 ± 3.1  109 copies/L and 8.6 ± 2.1  109 copies/L, respectively) significantly different from the values measured in these same vessels at 4.5 h (Tukey HSD; p > 0.05). These results confirmed that the ammonia added at the end of stage 1 in assay D, which contained inhibited AOB, had no effect on Nitorbacter spp. growth activity. This was expected since little or no inhibition of nitrite oxidation was noted in the oxygen consumption data during assay D (Table 2). However, the rRNAt abundance in assay E, which contained uninhibited AOB, was significantly lower 9 h (2.6 ± 0.7  109 copies/L) and 26 h (4.5 ± 1.7  108 copies/L) into the experiment while the nitrite buildup occurred (Fig. 5). This confirmed that the decline in nitrite oxidation activity observed independently via nitrogen intermediate data (Fig. 4B) produced a large corresponding decrease (94%) in the molecular growth potential for Nitrobacter spp. (Hawkins et al., 2008). The level of decline in the ribosomal transcript abundance in this study was far greater than the level of decline in total RNA observed for Nitrospira spp. (35%) during exposure to high free ammonia concentrations (Simm et al., 2006, 2005). This suggest that the ribosomal gene transcript abundance, which is expected to fluctuate with bacterial growth potential as a result of cell transcriptional controls (Cangelosi and Brabant, 1997), is a much better indicator of growth activity than mature ribosomal RNA. Interestingly, during the recovery in nitrite oxidation activity in assay E at hour 67, the rRNAt level (2.8 ± 0.2  109 copies/L) was significantly higher than at hour 26 (4.5 ± 1.7  108 copies/L), but remained lower than the level measured prior to ammonia addition at hour 4.5 (6.8 ± 1.7  109 copies/L) (Tukey HSD; p > 0.05). This indicates that the growth potential of the Nitrobacter spp. population was diminished during the time period when ammonia was being rapidly oxidized, and complements the fact that the oxy-

gen utilization rate due to nitrite oxidation was higher before the ammonia addition (8 mg/L/h) than during the oxidation of accumulated nitrite (5–6 mg/L/h) (Fig. 4A). Finally, the rRNAt/rDNA ratio at hour 26 (0.082 – see Fig. 5) replicated values of this metric observed during Nitrobacter spp. inhibition with azide, low pH, and 3,5-dichlorophenol, as well as during nitrite starvation (Hawkins et al., 2006, 2008). This indicates that the rRNAt/rDNA metric may serve as a robust indicator of molecular activity for Nitrobacter spp. and further bolsters confidence in the conclusion that high AOB activity has a much more important impact on nitrite oxidation activity than free ammonia. 4. Conclusion The inhibition experiments presented herein confirm that free ammonia does not significantly inhibit Nitrobacter spp. mediated nitrite oxidation. However, in the absence of an appropriate control eliminating AOB activity, the decline and resumption of nitrite oxidation observed in the long-term experiments (Fig. 4) could easily be misconstrued as an effect of free ammonia, which naturally peaks and declines with the total ammonia concentration and AOB activity levels. The relative unimportance of free ammonia as a Nitrobacter spp. inhibitor was confirmed using two independent inhibition assessment techniques, namely nitrogen conversion and oxygen consumption rates, and a highly sensitive molecular measure of growth activity. The findings indicate that pH changes and ammonia oxidizing bacteria activity are more important factors limiting Nitrobacter spp. mediated nitrite oxidation, rather than the free ammonia concentration. References Abeling, U., Seyfried, C.F., 1992. Anaerobic–aerobic treatment of high-strength ammonium wastewater – nitrogen removal via nitrite. Water Sci. Technol. 26, 1007–1015. Anthonisen, A.C., Loehr, R.C., Prakasam, T.B.S., Srinath, E.G., 1976. Inhibition of nitrification by ammonia and nitrous-acid. J. Water Pollut. Contr. Fed. 48, 835– 852. APHA, 1998. Standard Methods for the Examination of Water and Wastewater, 20th ed. American Public Health Association, New York, NY.

S. Hawkins et al. / Bioresource Technology 101 (2010) 4513–4519 Bae, W., Baek, S., Chung, J., Lee, Y., 2002. Optimal operational factors for nitrite accumulation in batch reactors. Biodegradation 12, 359–366. Balmelle, B., Nguyen, K.M., Capdeville, B., Cornier, J.C., Deguin, A., 1992. Study of factors controlling nitrite build-up in biological processes for water nitrification. Water Sci. Technol. 26, 1017–1025. Cangelosi, G.A., Brabant, W.H., 1997. Depletion of pre-16S rRNA in starved Escherichia coli cells. J. Bacteriol. 179, 4457–4463. Chung, J., Shim, H., Lee, Y.W., Bae, W., 2005. Comparison of influence of free ammonia and dissolved oxygen on nitrite accumulation between suspended and attached cells. Environ. Technol. 26, 21–33. Daniel, L.M.C., Pozzi, E., Foresti, E., Chinalia, F.A., 2009. Removal of ammonium via simultaneous nitrification–denitrification nitrite-shortcut in a single packedbed batch reactor. Bioresour. Technol. 100, 1100–1107. Ford, D.L., Churchwell, R.L., Kachtick, J.W., 1980. Comprehensive analysis of nitrification of chemical processing wastewaters. J. Water Pollut. Contr. Fed. 52, 2726–2746. Fux, C., Huang, D., Monti, A., Siegrist, H., 2004. Difficulties in Maintaining Long-term Partial Nitridation of Ammonium-rich Sludge Digester Liquids in a Moving-bed Biofilm Reactor (MBBR). IWA Publishing. pp. 53–60. Fux, C., Lange, K., Faessler, A., Huber, P., Grueniger, B., Siegrist, H., 2003. Nitrogen removal from digester supernatant via nitrite – SBR or SHARON? Water Sci. Technol. 48, 9–18. Ganigue, R., Gabarro, J., Sanchez-Melsio, A., Ruscalleda, M., Lopez, H., Vila, X., Colprim, J., Balaguer, M.D., 2009. Long-term operation of a partial nitridation pilot plant treating leachate with extremely high ammonium concentration prior to an anammox process. Bioresour. Technol. 100, 5624–5632. Ginestet, P., Audic, J.M., Urbain, V., Block, J.C., 1998. Estimation of nitrifying bacterial activities by measuring oxygen uptake in the presence of the metabolic inhibitors allylthiourea and azide. Appl. Environ. Microbiol. 64, 2266–2268. Grady, C.P.L., Daigger, G.T., Lim, H.C., 1999. Biological Wastewater Treatment, second ed. Marcel Deffer Inc., New York. Hawkins, S.A., 2005. The rRNAt/rDNA Ratio as a Measure of Nitrite Oxidizing Activity Department of Civil and Environmental Engineering. The University of Tennessee, Knoxville, TN. Hawkins, S.A., Robinson, K.G., Layton, A.C., Sayler, G.S., 2006. A comparison of ribosomal gene and transcript abundance during high and low nitrite oxidizing activity using a newly designed real-time PCR detection system targeting the Nitrobacter spp. 16S–23S intergenic spacer region. Environ. Eng. Sci. 23, 521– 532. Hawkins, S.A., Robinson, K.G., Layton, A.C., Sayler, G.S., 2008. Response of Nitrobacter spp. ribosomal gene and transcript abundance following nitrite starvation and exposure to mechanistically distinct inhibitors. Environ. Sci. Technol. 42, 901– 907. Hellinga, C., Schellen, A., Mulder, J.W., van Loosdrecht, M.C.M., Heijnen, J.J., 1998. The SHARON process: An innovative method for nitrogen removal from ammonium-rich waste water. Water Sci. Technol. 37, 135–142. Helsel, D.R., Hirsch, R.M., 2002. Statistical Methods in Water Resources. U.S. Geologic Survey, U.S. Department of the Interior.

4519

Kuenen, J.G., Jetten, M.S.M., 2001. Extraordinary anaerobic ammonium-oxidizing bacteria. ASM News 67, 456–463. Liu, S., Yang, F., Xue, Y., Gong, Z., Chen, H., Wang, T., Su, Z., 2008. Evaluation of oxygen adaptation and identification of functional bacteria composition for anammox consortium in non-woven biological rotating contactor. Bioresour. Technol. 99, 8273–8279. Mauret, M., Paul, E., PuechCostes, E., Maurette, M.T., Baptiste, P., 1996. Application of experimental research methodology to the study of nitrification in mixed culture. Water Sci. Technol. 34, 245–252. Philips, S., Wyffels, S., Sprengers, R., Verstraete, W., 2002. Oxygen-limited autotrophic nitrification/denitrification by ammonia oxidisers enables upward motion towards more favourable conditions. Appl. Microbiol. Biotechnol. 59, 557–566. Simm, R.A., Mavinic, D.S., Ramey, W.D., 2006. A targeted study on possible free ammonia inhibition of Nitrospira. J. Environ. Eng. Sci. 5, 365–376. Simm, R.A., Ramey, W.D., Mavinic, D.S., 2005. Nitrifier population dynamics in a bench-scale conventional activated sludge reactor following an induced perturbation. J. Environ. Eng. Sci. 4, 385–397. Sliekers, A.O., Third, K.A., Abma, W., Kuenen, J.G., Jetten, M.S.M., 2003. CANON and anammox in a gas-lift reactor. FEMS Microbiol. Lett. 218, 339–344. Suthersan, S., Ganczarczyk, J.J., 1986. Inhibition of nitrite oxidation during nitrification: some observations. Water Pollut. Res. J. Can. 21, 257–266. Turk, O., Mavinic, D.S., 1987. Benefits of using selective inhibition to remove nitrogen from highly nitrogenous wastes. Environ. Technol. Lett. 8, 419–426. Turk, O., Mavinic, D.S., 1989. Maintaining nitrite buildup in a system acclimated to free ammonia. Water Res. 23, 1383–1388. Villaverde, S., Fdz-Polanco, F., Garcia, P.A., 2000. Nitrifying biofilm acclimation to free ammonia in submerged biofilters. Start-up influence. Water Res. 34, 602– 610. Voets, J.P., Vanstaen, H., Verstraete, W., 1975. Removal of nitrogen from highly nitrogenous wastewaters. J. Water Pollut. Contr. Fed. 47, 394–398. Wong-Chong, G.M., Loehr, R.C., 1978. Kinetics of microbial nitrification: nitrite– nitrogen oxidation. Water Res. 12, 605–609. Xu, Z.Y., Zeng, G.M., Yang, Z.H., Xiao, Y., Cao, M., Sun, H.S., Ji, L.L., Chen, Y., 2010. Biological treatment of landfill leachate with the integration of partial nitrification, anaerobic ammonium oxidation and heterotrophic denitrification. Bioresour. Technol. 101, 79–86. Yamanaka, T., Kamita, Y., Fukumori, Y., 1981. Molecular and enzymatic properties of cytochrome aa3-type terminal oxidase derived from Nitrobacter agilis. J. Biochem. 89, 265–273. Yang, L., Alleman, J.E., 1992. Investigation of batchwise nitrite buildup by an enriched nitrification culture. Water Sci. Technol. 26, 997–1005. Yang, S.F., Tay, J.H., Liu, Y., 2004. Respirometric activities of heterotrophic and nitrifying populations in aerobic granules developed at different substrate N/COD ratios. Curr. Microbiol. 49, 42–46. Yun, H.J., Kim, D.J., 2003. Nitrite accumulation characteristics of high strength ammonia wastewater in an autotrophic nitrifying biofilm reactor. J. Chem. Technol. Biotechnol. 78, 377–383.