effects of chlortetracycline amended feed on anaerobic sequencing batch reactor performance of swine...

Bioresource Technology 125 (2012) 65–74

Contents lists available at SciVerse ScienceDirect

Bioresource Technology

journal homepage: www.elsevier .com/locate /bior tech

Effects of chlortetracycline amended feed on anaerobic sequencing batchreactor performance of swine manure digestion

Teal M. Dreher a, Henry V. Mott a, Christopher D. Lupo a, Aaron S. Oswald a, Sharon A. Clay b,James J. Stone a,⇑a Department of Civil and Environmental Engineering, South Dakota School of Mines and Technology, Rapid City, SD 57701, USAb Department of Plant Sciences, South Dakota State University, Brookings, SD 57007, USA

h i g h l i g h t s

" Effects of manure from CTC amended feed on ASBR performance assessed." ASBR biogas extent unaffected, but methane content reduced by CTC treatment." CTC acclimatization apparent after 56 d.

a r t i c l e i n f o

Article history:Received 7 June 2012Received in revised form 14 August 2012Accepted 18 August 2012Available online 31 August 2012

Keywords:Swine manureAntimicrobial useChlortetracyclineSwine productionAnaerobic digestion

0960-8524/$ - see front matter � 2012 Elsevier Ltd. Ahttp://dx.doi.org/10.1016/j.biortech.2012.08.077

⇑ Corresponding author. Tel.: +1 605 394 2443; faxE-mail address: [email protected] (J.J. Stone

a b s t r a c t

The effects of antimicrobial chlortetracycline (CTC) on the anaerobic digestion (AD) of swine manureslurry using anaerobic sequencing batch reactors (ASBRs) was investigated. Reactors were loaded withmanure collected from pigs receiving CTC and no-antimicrobial amended diets at 2.5 g/L/d. The slurrywas intermittently fed to four 9.5 L lab-scale anaerobic sequencing batch reactors, two with no-antimi-crobial manure, and two with CTC-amended manure, and four 28 day ASBR cycles were completed. TheCTC concentration within the manure was 28 mg/L immediately after collection and 1.02 mg/L after dilu-tion and 250 days of storage. CTC did not inhibit ASBR biogas production extent, however the volumetriccomposition of methane was significantly less (approximately 13% and 15% for cycles 1 and 2, respec-tively) than the no-antimicrobial through 56 d. CTC decreased soluble chemical oxygen demand and ace-tic acid utilization through 56 d, after which acclimation to CTC was apparent for the duration of theexperiment.

� 2012 Elsevier Ltd. All rights reserved.

1. Introduction

Modern swine production in the United States consists of highanimal throughput within a small operations area, making opera-tions more susceptible to disease outbreaks (Cheng, 2003). Manyswine production facilities use animal feeds containing moderateto high doses of antimicrobial agents to prevent outbreak, treatexisting disease, and promote growth in the animals (Dewey et al.,1999). The most commonly used antibiotics in animal feed aretetracyclines, carbadox, and bacitracin, while the most commonlyadministered antimicrobial agent is chlortetracycline (CTC), whichis added to feed in all phases of swine production (Dewey et al.,1999). This is problematic because the animals metabolize only asmall percentage of antimicrobials, and thus the majority of theparent compound, in addition to daughter product degradates, isreadily excreted through animal urine and feces. As a result, many

ll rights reserved.

: +1 605 394 5171.).

strains of antimicrobial-resistant bacteria have been reported with-in manure matrices (Chenier and Juteau, 2009; Jindal et al., 2006).There is conflicting evidence as to whether the presence of theseantimicrobial compounds may inhibit manure treatment processes(Lallai et al., 2002; Masse et al., 2000; Shimada et al., 2008; Sponzaand Celebi, 2012; Stone et al., 2009). Treating manure using tradi-tional methods, such as lagoons or land application, compared touse the use of anaerobic digesters, can lead to indirect humanexposure of antimicrobial compounds and resistant bacteria withinimpacted soil and water environments (Chapin et al., 2005).

The use of psychrophilic anaerobic digestion (PAD) using anaer-obic sequencing batch reactors (ASBR) has been reported as a viableswine treatment option for cold climates found in northern US andCanada, where traditional management practices may not be appli-cable (Masse et al., 2010). The presence of antimicrobials and theirsubsequent impact on ASBR operations were previously investi-gated by Masse et al. (2000), who reported that of six antimicrobialcompounds analyzed (tylosin, lyncomycin, tetracycline, supha-methazine, penicillin, and carbadox), only penicillin and tetracy-

http://dx.doi.org/10.1016/j.biortech.2012.08.077

mailto:[email protected]

http://dx.doi.org/10.1016/j.biortech.2012.08.077

http://www.sciencedirect.com/science/journal/09608524

http://www.elsevier.com/locate/biortech

66 T.M. Dreher et al. / Bioresource Technology 125 (2012) 65–74

cline had an inhibitory effect on methane production. They furtherconcluded that methane inhibition would not impact farm-scaleoperations because the reactors appeared to acclimatize to antimi-crobial agents over time. Similarly, Stone et al. (2009) found thatswine manure from pigs administered CTC-containing feed tooklonger to stabilize under anaerobic conditions compared to tylosinor no-antimicrobial treatments. In a separate study, Stone et al.(2011) found that CTC-amended swine manure containing 80 mg/L CTC inhibited both SBR treatment efficiencies and methanogene-sis due to the accumulation of acetate found within the reactors,although limited inhibition was reported at lower CTC concentra-tions. Others have shown that the presence of antibiotics can resultin the anaerobic biomass accumulation of the specific antigen byeither developing strains of antimicrobial resistant bacteria, orthrough microbial community dynamics shifting to organisms lesssensitive to the specific antimicrobial (Angenent et al., 2008; Shi-mada et al., 2008). The purpose of this study was to determinethe effects of CTC present in manure due to excretion after feedingCTC amended feed on manure treatment efficiencies using anaero-bic sequencing batch reactors.

2. Methods

2.1. Experimental setup

Four identical sealed cylindrical reactors, each of 1.2 m heightand 10 cm inner diameter with a working volume of approxi-mately 9.5 L, were constructed of acrylic glass and positioned up-right using belt clamps affixed to a metal frame (Figure S1). Gasrecirculation for reactor mixing was accomplished using two ColeParmer peristaltic pumps (Vernon Hills, IL) each housing fourpump motor head assemblies and a Masterflex controller operatingbetween 2.3 and 283 mL/min of gas flow. 1.27 cm OD by 0.95 cmID vinyl tubing of was used to connect peristaltic pump assembliesto the four ASBRs. The vinyl tubing was connected to a three way‘‘T’’ connector with one connection going to the peristaltic pumpfor gas recirculation, and the other line to a dedicated wet tip gasmeter for determination of gas production. The wet tip gas metersconsisted of a sealed plastic container filled with DI water, a tip-ping float assembly, off gas valve, HOBO (Onset, Cape Cod, MA)data logger pendant event, and clamps for sealing the assembly.ASBR gases collected beneath were routed to each dedicated wettip gas meter. Gas collect beneath the tipping float until its buoy-ancy was exceeded and gas was released to a SKC Tedlar 12 L gasbag for collection and analysis. Each tip was time recorded byHOBO Ware Lite version 2.3.0 computer software, and calibratedto 108 mL per tip (including 8 mL of water displaced after gas recir-culation concluded) according to manufacturer recommendations.Biogas was collected from the top of each reactor and recirculatedthrough the bottom of the reactor to agitate the mixture.

2.2. Manure collection

The manure used for this experiment was obtained from theSouth Dakota State University (SDSU) swine facilities research unit.Six male pigs, each weighing approximately 20 kg, were placed inindividual stainless steel metabolism crates and fed either a controldietary treatment, or a control plus CTC (22 mg CTC per kg bodyweight; Tradename: Chloratet 50 containing 50 g CTC/kg; ADM Alli-ance Nutrition, Quincy, IL) treatment. The corn–soybean meal baseddiets met or surpassed swine nutrient requirements. The meal foreach pig was identical except for the inclusion of antimicrobialagents. A portion of each manure slurry (aggregated by manuretype) was collected once a day for five days, starting after the 6thday, and then stored at 4 �C in plastic carboys for approximately

250 days until used within the experiment. CTC concentration inthe slurry was 28 mg/L after collection, 11.6 mg/L prior to dilution,and 1.02 mg/L after dilution and experimental use. CTC concentra-tions within the reactors during the experiments were not analyti-cally determined.

2.3. Manure preparation

Prior to the experiment, both manure treatments (CTC and no-antimicrobial) were filtered through a number 40 sieve to removelarge solids that would otherwise clog the peristaltic pump used tofeed the reactors. The CTC and no-CTC manure samples were thendiluted using tap water to obtain a manure addition organic load-ing rate (OLR) of 2.5 g total chemical oxygen demand (TCOD) perliter of bioreactor initial sludge volume per day based upon TCODmanure analyses. Manure slurries were prepared in 1 L incrementsby combining sieved manure to tap water ratio of 88 mL/912 mL(CTC treatment) and 116 mL/884 mL (no-antimicrobial treatment)volumetric ratios (manure/tap water). Tap water was used as op-posed to distilled or deionized water to better replicate field treat-ment operations that would exist at swine production facility,which regularly use tap water for washdown or cleaning processes.It should be noted that chlorination effects due to differing dilutionfactors was not accounted for when considering CTC affects. Char-acteristics of the raw swine manure are shown in Table S1.

2.4. Experimental procedure

Each of the four 9.5 L ASBRs were continuously operated at22 �C (room temperature) and inoculated with municipal waste-water anaerobic digester sludge, which was used as the initial col-ony of biomass, and cycled with no-antimicrobial manure untilsteady state was achieved (after approximately 250 days). Steadystate was determined as the point at which all four reactors hadsimilar rates of biogas production. Once steady state was achieved,two reactors were fed identical ratios of no-CTC containing man-ure, while the two remaining reactors were fed manure containingCTC, as described above. The reactors were fed 1 L of diluted man-ure (as described above) intermittently three times per week for2 weeks. The reactors then reacted for 2 weeks where the reactorswere mixed once a day for 10 min by recycling the biogas throughthe bottom of the ASBR units using a peristaltic pump with fourdedicated pump heads. The reactors were unmixed 24 h prior todecanting to stratify reactor contents. A 20 mL aliquot was col-lected intermittently throughout the fill and react phases. To en-sure collection of a representative sample, the reactors weremixed for 10 min immediately before sampling. The samples werestored at 4 �C prior to analysis to slow the biochemical reactions,and analyzed within 2 weeks for total chemical oxygen demand(TCOD), soluble chemical oxygen demand (sCOD), alkalinity, pH,total suspended solids (TSS), volatile suspended solids (VSS), andvolatile fatty acids (VFA) content (acetic acid, propionic acid, isobu-tyric acid, and butyric acid) as described below.

2.5. Analytical methods

TSS, VSS, pH, alkalinity, TCOD, and sCOD (1.5 lm glass fiber fil-trate; ferrous ammonium sulfate method) were determined usingmethods adapted from standard methods (APHA, 1995). Gas chro-matography with flame ion detection (GC–FID; HP5890 series II,Hewlett Packard, Palo Alto, CA) and HP 3396A integrator was usedto determine VFA concentrations in the liquid phase. To aid sampleinjection, a 75 mm CAR/PDMS solid phase micro extraction (SPME)fiber and miniert valves (Supelco, Bellefonte, PA) with 0.75 mm in-ner diameter injection sleeve and high pressure merlin microsealswere used. A 2.5 mL gastight syringe (Hamilton, Reno, NV) was used

T.M. Dreher et al. / Bioresource Technology 125 (2012) 65–74 67

to minimize analyte loss during solution transfer. Sample injectionswere performed using the splitless mode with analytical grade he-lium as carrier gas at a 17 mL/min constant flow rate. The analyticalcolumn was 50 m by 0.32 mm inner diameter HP-5 with 1.05 mmcross linked 5% Ph Me silicone column stationary phase. Acetic acid(>99.99%), propionic acid (99.8%), isobutyric acid (>99%) and butyricacid (99%) standards were obtained from Aldrich America (St. Louis,MO). A secondary standard was prepared having mass fractions of0.5462, 0.3558, 0.0104, and 0.0913 g for acetic acid, propionic acid,isobutyric acid, and butyric acid, respectively. A tertiary standardwas created as a 1:6 dilution of the primary standard. Calibrationstandards were prepared by transferring known masses of the ter-tiary standard into EPA vials with minninert valves containing

Fig. 1. (A) pH, (B) alkalinity profiles for collected between 0 and 112 d from two CTC andsingle sample collected.

2.5 mL of deionized water. The samples were prepared for analysisby first using alkaline adjustment to a pH of 9 (to ensure none of thevolatile fatty acids escaped during sample preparation), centrifuga-tion at 2500 rpm for 10 min, transfer of 2.5 mL supernatant liquidinto a 50 mL EPA vial fitted with minninert valves, and subsequentacid adjustment to pH 2 using 30 lg concentrated H2SO4. The sam-ples and standards were then agitated and equilibrated at 25 �C forat least 8 h prior to analysis. Once prepared, the samples and stan-dards were analyzed by inserting and exposing the SPME fiber intothe minninert valve and letting it equilibrate with the sample’sheadspace gas for 10 min prior to injection into the GC.

ASBR biogas samples were analyzed for methane content usinggas chromatography with thermal conductivity detection (GCTCD

two no-CTC ASBR reactors. Individual data points represent analytical result from a


HP6890, Hewlett Packard/Agilent, Santa Clara, CA). Gas sampleswere collected continuously using dedicated 12 L SKC Tedlar gasbags (SKC Inc., Eighty Four, PA). Gas samples were injected directlyinto the GCTCD using a 100 lL Gastight syringe (Hamilton, Reno,NV), with responses calibrated with injection of gases of knowncomposition.

Community-Level Physiological Profiles (CLPP) were assessed at5, 25, 59, and 79 days (i.e., at the beginning and end of cycles 1 and3) during the ASBR experiments using the Biolog (Hayward, CA)Ecoplate. The Ecoplate has three replicated wells of 31 differentecologically relevant carbon substrates including amines/amides,carbohydrates, polymers, carboxylic acids, and polymers. TheEcoplate has been used in the past to characterize bacterial com-munities in a variety of environments such as soils (Hofmanet al., 2004; Kohler et al., 2005; Sullivan et al., 2006; Zhou et al.,

Fig. 2. (A) sCOD, (B) TCOD profiles for collected between 0 and 112 d from two CTC andsingle sample collected.

2008), wetlands (Salomo et al., 2009; Weber et al., 2007) and man-ure compost (Belete et al., 2001; Gomez et al., 2006). A CLPP re-sponse is based upon utilization of various metabolic substratesand provides a measurement of potential activity for culturablebacteria (Hofman et al., 2004; Smalla et al., 1998), with ecologicallyrelevant classifications of heterotrophic microbial communitiesidentified by their unique microplate response (Garland and Mills,1991).

A 2 mL reactor content sample of each ASBR was placed into a50 mL centrifuge tube, diluted with 20 mL DI water, vortexed for2 min, and centrifuged at 3000 RPM for 2.5 min. Each of the sub-strate wells on the Ecoplate was inoculated with a 100 ll aliquotof the supernatant. The Ecoplates were incubated in an anaerobicchamber (100% nitrogen gas headspace) at room temperature(22 �C). Optical density (OD) measurements using the Biolog plate

two no-CTC ASBR reactors. Individual data points represent analytical result from a


reader at k = 590 nm were determined at 4, 28, 52, 76, 100, 124,148, and 172 h after well plate inoculations. Time series OD mea-surements were transformed into Average Well Color Develop-ment (AWCD) profiles following Zhou et al. (2008) using thefollowing equation.

AWCD ¼XC � R

nð1Þ

where:C = absorbance value (OD) within each wellR = absorbance value (OD) of the control well (DI water)n = number of substrates utilized (out of 31 total)

Fig. 3. (A) TSS, (B) VSS profiles for collected between 0 and 112 d from two CTC and two nsample collected.

Negative values were normalized to zero to minimize bias.AWCD profiles provide an indicator of microbial diversity that ex-ists within the reactors during sample collection. Sigmoidal AWCDprofiles were further assessed by curve fitting to the logisticgrowth equation as defined by Salomo et al. (2009) where kineticparameters K (asymptote/carrying capacity), s (time when AWCDis K/2), and p (exponential rate of AWCD change) were determined.A higher AWCD (i.e., greater color development) was indicative ofthe ‘‘capacity’’ of the microbial culture for degradation of organicmatter, while a steeper AWCD slope (p) suggested a faster rate ofsubstrate consumption occurred. Individual substrate analysis ofpositive wells was performed following methods by Hofmanet al. (2004), where positive substrate utilization wells were sorted

o-CTC ASBR reactors. Individual data points represent analytical result from a single


by their 76 h reading (after which substrate availability was lim-ited) based upon their respective chemical guild utilizations.

2.6. Statistics

The Mann-Whitney test, a non-parametric test that is com-monly used in clinical studies to compare two independent setsof data (Wijnand and van de Velde, 2000), was used to determinewhether the CTC data were significantly (p < 0.05) different thanthe no-CTC data. Differences in microbial diversity and CLPP be-tween reactor treatments were characterized by Principal Compo-nent Analysis (PCA) using Microsoft Excel XLSTAT statistical add-on software (Addinsoft, New York, NY). PCA analyses, substrate uti-lization and microbial diversity were based on 76 h Biolog incuba-tion readings.

A

B

Fig. 4. (A) Daily biogas production, (B) percent methane composition of produced b

3. Results and discussion

3.1. Physical and chemical indicators

3.1.1. pH and alkalinityThroughout the four 28 day cycles, the pH was consistently be-

tween 7.1 and 7.8 (Fig. 1A). During the first cycle, the pH was rel-atively stable for the CTC reactors, whereas pH exhibited aninsignificant increase from 7.1 to 7.2 in the no-CTC reactors. Bothsets of reactors exhibited an increase in pH during their respectivereact phase (up to pH 7.5), which coincided with a 17.9% and a17.6% overall decrease in alkalinity for the CTC and no-CTC reac-tors, respectively (Fig. 1B). By the completion of the react period,the pH for all reactors had reduced to 7.3. The pH drop suggestsinsoluble organic material was converted through hydrolysis reac-

iogas at the completion of cycles 1–4 (CTC-1 and no CTC-1 ASBR reactors only).


tions (Stone et al., 2009). During the second and third cycles, thepH remained moderately stable (between 7.2 and 7.3 for the CTCreactors, and between 7.30 and 7.44 for the no-CTC reactors)throughout both the fill and react periods. During these two cycles,the pH for the CTC reactors was significantly (p < 0.05) lower thatfor the no-CTC reactors, with the pH decrease likely attributed toreduced methanogen activities and subsequent VFA accumulation.During the fourth cycle, pHs increased steadily (from 7.3 to 7.5 forCTC, and 7.3 to 7.7 for no-CTC reactors) during both the fill and re-act periods, and appear attributed to the rapid accumulation ofalkalinity during the fill cycle (Zhang et al., 2006). The alkalinitywas significantly less (p < 0.05) for the CTC reactors, compared tothe no-CTC reactors (Fig. 1B). Both conditions showed overall alka-linity consumption with no corresponding pH drop for cycles 1–4.Stone et al. (2009) reported that manure containing CTC had agreater organic acid generation during batch reactions, attributedto a greater rate of alkalinity consumption.

3.1.2. sCOD and TCODThe concentrations of sCOD within the CTC and no-CTC reactors

are shown in Fig. 2A. The sCOD was significantly less (p < 0.05) forthe CTC reactors, though all reactors had similar sCOD responsesduring each cycle phase. Each cycle was marked by increased sCODduring the fill phase, and then decreased sCOD during the reactperiod. These results were consistent with those reported by Masseet al. (2000) under similar conditions where the presence of tetra-cycline had no effect on the treatment efficiency of ASBRs. How-ever, the efficiency of sCOD removal for the CTC reactors wassignificantly greater (p < 0.05) for the second two cycles comparedto the first two. The no-CTC reactors averaged between 18% and37% removal of sCOD for the duration of the experiment, whileCTC reactors exhibited only a 13% reduction during cycles 1 and2, followed by 53% and 42% reduction during cycles 3 and 4,respectively. This dramatic increase in efficacy appears due tomicrobial acclimation of CTC residuals within the system.

There were no significant differences between the no-CTC andCTC reactors for TCOD (Fig. 2B). There was a slight difference inreactor behavior during the first two cycles, followed by similarbehavior for the second two cycles. For the no-CTC reactors, theTCOD decreased during the fill period for the first and second cycles,and then stabilized during the react periods, while TCOD increasedduring the fill periods and then decreased during the react periodsfor the CTC reactors. For cycles 3 and 4, TCOD decreased for all reac-tors during the fill periods and then increased during the react peri-ods. These data differ from the sCOD trends, suggesting that CODlikely exists as particulate matter (Masse et al., 2000) which is set-tleable. There was a net TCOD decrease during the cycles 1 and 4,however cycles 2 and 3 resulted in a net TCOD increase, consistentwith TSS removal efficiency (Fig. 3A). The similarity between theCTC and no-CTC reactors beginning with the second cycle suggeststhat the microbial acclimation towards CTC had begun. The treat-ment efficiency for TCOD could not be accurately calculated sincereactors were mixed before sample collection (preventing account-ing of settled particulates); the data suggest a TCOD accumulation,which would indicate that hydrolysis reactions were inhibited bythe CTC treatment.

3.1.3. TSS and VSSThere were no significant differences between the CTC and no-

CTC reactors with respect to TSS (Fig. 3A). TSS reactor behavior wassimilar except for with CTC reactor 1. Reactor TSS contents gener-ally decreased during their fill periods, then increased during theirreact periods. Stone et al. (2009) found that manure from CTC fedswine increased TSS and VSS by 53% compared with a manure con-taining no-CTC.

VSS was significantly (p < 0.05) different between the CTC andno-CTC reactors (Fig. 3B). The no-CTC reactors increased from527 mg/L to 3228 mg/L and 1326 mg/L to 1650 mg/L during thefirst fill period. VSS concentrations for no-CTC were reduced to248 mg/L by the end of the first cycle, remained relatively stablethrough the second fill, then increased to 2464 mg/L and1263 mg/L during the second react period, respectively. Both CTCreactors had elevated VSS levels, and unlike their no-CTC counter-parts, both experienced a significant decrease during the first fillperiod followed by an increase during the react phase. The CTCreactors VSS content increased 48% during the second fill period.Both no-CTC increased by 650% during the react phase, while theCTC reactors only slightly increased during this time. The VSStreatment efficiencies were similar to TSS and TCOD.

3.1.4. VFAThe VFAs acetic acid, propionic acid, butyric acid, and isobutyric

acid are intermediates formed during biodegradation of the man-ure, resulting in eventual methane and carbon dioxide production.Accumulation of these constituents would suggest that an inter-mediate process had been compromised. Stone et al. (2009) re-ported VFA accumulation was greater for manure containing CTCthan without CTC under batch treatment conditions, while Masseet al., (2000) found VFA concentrations were not significantlydifferent between manure antibiotic and no-antibiotic reactors.For the experimental conditions reported here, only acetic acidand butyric acid were determined at detectable levels (Figs. S2and S3). There were no significant (p < 0.05) differences betweenVFA concentrations between all treatments. Acetic acid accumula-tion was greater for the CTC reactors during the first cycle (18% and383% increase, respectively) compared to the no-CTC treatments(3% and 37% increase), suggesting probable methanogenesis inhibi-tion due to the buildup of volatile acids. There was no buildup ofacetic acid for either treatments during cycles three and four, indi-cating that acetate was efficiently utilized.

3.2. Biogas production and composition

The specific methane production (Table S2) and the total gasproduction (Fig. 4A) were similar except for no-CTC reactor 1. Thisreactor had significantly (p < 0.05) lower gas production, howeverthis difference was attributed to instrumentation error as the phys-ical and chemical properties of the reactor content were generallyconsistent with the no-CTC results throughout the duration of theexperiment. For all reactors, biogas production peaked during thefeed period (peaking at approximately 10 d after initial substrateaddition), and production rates generally decreased during the re-act period where biogas production stabilized to 200–300 mL/dayof gas at the end of the react phase when substrate was limited(Fig. 4A). In general, methane production mirrored the results forsCOD, where the concentrations peaked during the fill phase, andreduced during the react phase signifying that the organic contentwas actively converted into inorganic content via anaerobic diges-tion. The specific composition of methane within the biogas mix-ture for each reactor and stage is shown in Fig. 4B. For the firsttwo cycles, the composition of the biogas produced from the CTCreactors resulted in significantly less volume (p < 0.05) and percentmethane composition compared to the biogas produced from theno-CTC containing reactors. These findings may be attributed tothe limited degree of hydrolysis completion (13% reduction insCOD removal) compared to the no-CTC reactors where 31% and18% removal of sCOD was observed for the first and second cycles.Masse et al. (2000) reported that methane composition of bioreac-tors fed tetracycline manure slurry was similar to their no-tetracy-cline reactors; however net volume of total gas productiondecreased. The gas compositions for cycles 3 and 4 (between days

B

A

Fig. 5. (A) Summary of anaerobic substrate utilization for all ASBR treatments according to specific chemical guilds (amines/amides, polymers, carbohydrates, carboxylicacids, and amino acids) after 76 h of anaerobic incubation at 22 �C, (B) classification of all ASBR treatments by PCA for Biolog Ecoplate substrate utilization (OD data for BiologEcoplates) for guild classifications. Amines/amides consist of Phenylethyl-amine and Putrescine; polymers consist of Tween 40, Tween 80, Glycogen and alpha-Cyclodextrine;carbohydrates consist of D-Mannitol, N-Acetyl-D-Glucosamine, D-Cellobiose, D-Xylose, Glucose-1-phosphate, beta-Methyl-D-Glucoside, i-Erythritol, alpha-D-lactose and D,L-alpha-Glycerol phosphate, carboxylic acids consist of Pyruvic acid methyl ester, D-Galacturonic acid, D-Galacturonic acid gamma-Lactone, D-Glucosaminic acid, D-Malic acid,alpha-Ketobutyric acid, 4-Hydroxy benzoic acid, gamma-Hydroxybutyric acid, Itaconic acid and 2-Hydroxy benzoic acid; amino acids consist of L-Asparagine, L-phenylalanine, L-Arginine, L-Threonine, Glycyl-L-glutamic acid and L-Serine. Data points correspond treatment at specific sampling day. Factor 1 accounted for 29.1%, andFactor 2 for 15.4% of the variance between samples.


Table 1Kinetic parameters of the curve-fitted logistic growth equations [adapted fromSalomo et al. (2009)] for AWCD. K (asymptote), s (time when AWCD is K/2), p(exponential rate of AWCD change), R2 (correlation coefficient).

Time Treatment K [AWCD] p S [hr] R2

5 d CTC-1 1.78 0.033 82.5 0.99CTC-2 1.71 0.034 79.1 0.98No CTC-1 1.57 0.034 86.6 0.98No CTC-2 1.51 0.032 76.0 0.96





57 and 110) were similar (p > 0.05) for all of the reactors regardlessof treatment, suggesting that the biomass had stabilized and accli-matized to the presence of CTC. Stone et al. (2009) found that thepresence of CTC inhibited biogas production by about 28%, as com-pared to control manure, which they surmised was due to the re-duced amount of the aceticlastic methanogens, Methanosaetaceaeand Methanosarcinaceae spp. observed within the CTC manure.Masse et al. (2000) found that manure containing tetracyclineinhibited the production of methane by about 25% compared tono-tetracycline manure during their first 28 day cycle, while theyalso observed an increase in methane production during a secondcycle which was attributed to microorganism acclimation to thehigh COD and solids content of the manure slurries.

3.3. Microbial community level physiological profiles

CLPP profiles were determined at the beginning and completionof ASBR cycles 1 and 3 (days 5, 25, 59, 79) for all treatments. Asummary of Ecoplate (31 substrates) substrate utilization patternsafter 76 h of anaerobic incubation is shown in Figure S5. There waslittle difference in substrate utilization for the no-CTC reactors (be-tween 47.8 and 55.4% total substrate utilization) except at thecompletion of cycle 3 (day 79) where there was a noticeable de-crease for one duplicate reactor (39.8% utilization). Each CTC reac-tor began cycle 1 (day 5) at higher utilizations compared to the no-CTC reactors, however total utilization decreased at the end of thecycle. The CTC reactor behavior was dissimilar; for cycle 3 therewas little change in the first reactor (53.7% utilization), while an in-crease occurred for second reactor (60.7% utilization). AWCD pro-files were similar for all reactors where a 52 h lag in colordevelopment (substrate utilization) occurred, followed by peaksubstrate utilization by 124 h (results not shown). Little variationin the exponential rate of AWCD change (p) was noticed through-out the experiment, and there was no significant difference(p > 0.05) when CTC reactors were compared to no-CTC reactors(Table 1). The carrying capacity (K) for the CTC reactors decreasedfrom the beginning to the end in cycle 1, while that of the no-CTCreactors increased during the same period. There was no signifi-cant difference (p > 0.05) of the carrying capacity between the withand without CTC reactors. The substrate utilization patternsaccording to their chemical guilds indicate that at the beginningof cycle 1 (day 5), all reactors were similar (Fig. 5A). By the endof cycle 1 (day 25), carbohydrate utilization increased for the no-CTC reactors, while it decreased for the CTC reactors. All reactors

exhibited similar profiles at the beginning of cycle 3 (day 59) withless amine/amide and amino acid utilization compared to thebeginning of cycle 1. There was little difference in substrate utiliza-tion patterns between reactors at the end of cycle 3 (day 79) exceptfor CTC reactor 2 where amino acid utilization was greater. ThePCA using OD substrate utilization results (Fig. 5B) showed thatthe first principle component axis (PC1) explained 29.1% of the var-iance in the data, and the second principle component axis (PC2)explained 15.4% of the variance. Factor 1 indicates there was littledifference between the no-CTC reactors for both cycles 1 and 3 ex-cept for reactor 3 where there was a slight shift in CLPP during cy-cle 3. Furthermore, there was little difference observed for the CTCreactor 1 during cycle 1, however there was a change observed atthe beginning of cycle 3 before returning to normalcy. There weresignificant differences observed for the CTC reactor 1 throughoutboth cycles with the greatest difference occurring at the end of cy-cle 3. This noticeable difference was attributed to the higher aminoacid utilization compared to all other reactors at the end of cycle 3.Factor 2 indicates there was little difference between the no-CTCreactors, but a significant difference in the CTC reactor 1 for bothcycles 1 and 3. There was a difference for the CTC reactor 2 duringcycle 1, but not during cycle 3.

4. Conclusions

Methane production and treatment efficiencies were shown tobe inhibited in the reactors fed with manure from pigs fed CTC-amended feed during the first two 28 day experimental cycles.The microbial consortia appeared to adapt to the presence ofCTC, and treatment efficiency was effectively restored by the third28 day cycle. The use of ASBRs for the treatment of swine waste is aviable treatment method, and the use of CTC as a feed additiveshould not affect treatment efficiencies in farm scale operationsbecause the microbial community should acclimate to residualCTC in the manure.

Acknowledgements

This research is supported from a Grant from the National Sci-ence Foundation CBET program (awards 0606986, 0724917,0823853) in collaboration with Drs. Sharon Clay and Robert Thalerof South Dakota State University. Funding through the South Dako-ta Governors 2010 seed Grant program is gratefully acknowledged.The project described was also supported in part by US-EPA GrantNo. CP-97835401-0, EPA-319 fund, and NSF/EPSCoR Grant No.0091948 for LC/MS support. Its contents are solely the responsibil-ity of the authors and do not necessarily represent the officialviews of NSF.

References

Angenent, L.T., Mau, M., George, U., Zahn, J.A., Raskin, L., 2008. Effect of the presenceof the antimicrobial tylosin in swine waste on anaerobic treatment. Water Res.42, 2377–2384.

APHA, 1995. Standard methods for the examination of water and wastewater, 19thed. American Public Health Association, New York.

Belete, L., Egger, W., Neunhauserer, C., Caballero, B., Insam, H., 2001. Cancommunity level physiological profiles be used for compost maturity testing?Compost. Sci. Util. 9, 6–18.

Chapin, A., Rule, A., Gibson, K., Buckley, T., Schwab, K., 2005. Airborne multidrug-resistant bacteria isolated from a concentrated swine feeding operation. Env.Health. Persp. 113, 137–142.

Cheng, J., 2003. Challenges of CAFO waste management. J. Env. Eng. 129, 391–392.Chenier, M.R., Juteau, P., 2009. Fate of chlortetracycline- and tylosin-resistant

bacteria in an aerobic thermophilic sequencing batch reactor treating swinewaste. Microb. Ecol. 58, 86–97.

Dewey, C.E., Cox, B.D., Straw, B.E., Bush, E.J., Hurd, S., 1999. Use of antimicrobials inswine feeds in the United States. J. Swine Heal. Prod. 7, 19–25.


Garland, J., Mills, A., 1991. Classification and characterization of heterotrophicmicrobial communities on the basis of patterns of community-level sole-carbon-source utilization. App. Env. Micro. 57, 2351–2359.

Gomez, E., Ferreras, L., Toresani, S., 2006. Soil bacterial functional diversity asinfluenced by organic amendment application. Bioresour. Technol. 97, 1484–1489.

Hofman, J., Svihalek, J., Holoubek, I., 2004. Evaluation of functional diversity of soilmicrobial communities-a case study. Plant Soil Environ. 50, 141–148.

Jindal, A., Kocherginskaya, S., Mehboob, A., Robert, M., Mackie, R.I., Raskin, L., Zilles,J.L., 2006. Antimicrobial use and resistance in swine waste treatment systems.App. Env. Micro. 72, 7813–7820.

Kohler, F., Hamelin, J., Gillet, F., Gobat, J.M., Buttler, A., 2005. Soil microbialcommunity changes in wooded mountain pastures due to simulated effects ofcattle grazing. Plant Soil 278, 327–340.

Lallai, A., Mura, G., Onnis, N., 2002. The effects of certain antibiotics on biogasproduction in the anaerobic digestion of pig waste slurry. Bioresour. Technol.82, 205–208.

Masse, D.I., Lu, D., Masse, L., Droste, R.L., 2000. Effect of antibiotics on psychrophilicanaerobic digestion of swine manure slurry in sequencing batch reactors.Bioresour. Technol. 75, 205–211.

Masse, D.I., Masse, L., Xia, Y., Gilbert, Y., 2010. Potential of low-temperatureanaerobic digestion to address current environmental concerns on swineproduction. J. Anim. Sci. 88, E112–E120.

Salomo, S., Munch, C., Roske, I., 2009. Evaluation of the metabolic diversity ofmicrobial communities in four different filter layers of a constructed wetlandwith vertical flow by Biolog (TM) analysis. Water Res. 43, 4569–4578.

Shimada, T., Zilles, J.L., Morgenroth, E., Raskin, L., 2008. Inhibitory effects of themacrolide antimicrobial tylosin on anaerobic treatment. Biotech. Bioeng. 101,73–82.

Smalla, K., Wachtendorf, U., Heuer, H., Liu, W.T., Forney, L., 1998. Analysis of BiologGN substrate utilization patterns by microbial communities. App. Environ.Microb. 64, 1220–1225.

Sponza, D.T., Celebi, H., 2012. Removal of oxytetracycline (OTC) in a syntheticpharmaceutical wastewater by a sequential anaerobic multichamber bedreactor (AMCBR)/completely stirred tank reactor (CSTR) system:biodegradation and inhibition kinetics. Bioresour. Technol. 104, 100–110.

Stone, J.J., Clay, S.A., Zhu, Z.W., Wong, K.L., Porath, L.R., Spellman, G.M., 2009. Effectof antimicrobial compounds tylosin and chlortetracycline during batchanaerobic swine manure digestion. Water Res. 43, 4740–4750.

Stone, J.J., Oswald, A.S., Lupo, C.D., Clay, S.A., Mott, H.V., 2011. Impact ofchlortetracycline on sequencing batch reactor performance for swine manuretreatment. Bioresour. Technol. 102, 7807–7814.

Sullivan, T.S., Stromberger, M.E., Paschke, M.W., Ippolito, J.A., 2006. Long-termimpacts of infrequent biosolids applications on chemical and microbialproperties of a semi-arid rangeland soil. Biol. Fert. Soils 42, 258–266.

Weber, K.P., Grove, J.A., Gehder, M., Anderson, W.A., Legge, R.L., 2007. Datatransformations in the analysis of community-level substrate utilization datafrom microplates. J. Microb. Meth. 69, 461–469.

Wijnand, H.P., van de Velde, R., 2000. Mann-Whitney/Wilcoxon’s nonparametriccumulative probability distribution. Comp. Meth. Pro. Biomed. 63, 21–28.

Zhang, Z., Zhu, J., King, J., WenHong, L., 2006. A two-step fed SBR for treating swinemanure. Proc. Biochem. 41, 892–900.

Zhou, J., Guo, W.H., Wang, R.Q., Han, X.M., Wang, Q., 2008. Microbial communitydiversity in the profile of an agricultural soil in northern China. J. Environ. Sci.China 20, 981–988.

effects of chlortetracycline amended feed on anaerobic sequencing batch reactor performance of swine...

Documents