Characterization of Bacterial Community Dynamics During the Decomposition of Pig Carcasses in Simulated Soil Burial and Composti

J. Microbiol. Biotechnol. (2017), 27(12), 2199–2210 https://doi.org/10.4014/jmb.1709.09032 Research Article Review jmb

Characterization of Bacterial Community Dynamics during the Decomposition of Pig Carcasses in Simulated Soil Burial and Composting Systems S Bo-Min Ki1, Yu Mi Kim2, Jun Min Jeon3, Hee Wook Ryu2*, and Kyung-Suk Cho1*

1Department of Environmental Science and Engineering, Ewha Womans University, Seoul 03760, Republic of Korea 2Department of Chemical Engineering, Soongsil University, Seoul 06978, Republic of Korea 3Green Environmental Complex Center, Suncheon 57992, Republic of Korea

Received: September 18, 2017 Revised: September 30, 2017 Soil burial is the most widely used disposal method for infected pig carcasses, but composting Accepted: October 10, 2017 has gained attention as an alternative disposal method because pig carcasses can be First published online decomposed rapidly and safely by composting. To understand the pig carcass decomposition October 14, 2017 process in soil burial and by composting, pilot-scale test systems that simulated soil burial and

*Corresponding authors composting were designed and constructed in the field. The envelope material samples were K.-S.C. collected using special sampling devices without disturbance, and bacterial community Phone: +82-2-3277-2393; Fax: +82-2-3277-3275; dynamics were analyzed by high-throughput pyrosequencing for 340 days. Based on the odor E-mail: [email protected] gas intensity profiles, it was estimated that the active and advanced decay stages were reached H.W.R. earlier by composting than by soil burial. The dominant bacterial communities in the soil were Phone: +82-2-820-0611; Fax: +82-2-812-5378; aerobic and/or facultatively anaerobic gram-negative bacteria such as Pseudomonas, Gelidibacter, E-mail: [email protected] Mucilaginibacter, and Brevundimonas. However, the dominant bacteria in the composting system were anaerobic, thermophilic, endospore-forming, and/or halophilic gram-positive bacteria such as Pelotomaculum, Lentibacillus, Clostridium, and Caldicoprobacter. Different S upplementary data for this paper are available on-line only at dominant bacteria played important roles in the decomposition of pig carcasses in the soil and http://jmb.or.kr. compost. This study provides useful comparative date for the degradation of pig carcasses in

pISSN 1017-7825, eISSN 1738-8872 the soil burial and composting systems.

Introduction [1]. Confirmed cases of FMD were recorded on 3,748 farms during 144 days, from 28 November 2010 to 21 April 2011 Foot-and-mouth disease (FMD), a highly transmissible [1]. The Republic of Korea implemented a policy to viral disease of cloven-hoofed animals, is one of the most depopulate infected animals [4]. serious and economically significant diseases in the livestock Even though various methods have been used for carcass industry [1]. Once FMD is introduced into an FMD-free disposal, including burial, burning, incineration, rendering, country, the country makes great efforts to eradicate the anaerobic digestion, alkaline hydrolysis, and composting disease by enforcing strict control measures, such as [5-8], the most widely used disposal method has been culling and the disposal of animals on infected farms [2]. burial in soil. Burial is a relatively economical option for According to an OIE report, the Republic of Korea had 21 carcass disposal compared with other available methods; outbreaks of FMD in 2016, 188 in 2014, and 175 in 2010 [3]. it is convenient, logistically simple, and relatively quick, Unfortunately, the epidemic that occurred during the especially for daily mortalities, as the equipment necessary winter of 2010–2011 was the biggest ever recorded in Korea is widely available, and the technique is relatively straight-

December 2017 ⎪ Vol. 27⎪ No. 12 2200 Ki et al.

forward. If performed on-farm or on-site, trench burial also bacterial diversity during pig leg decomposition. Xu et al. eliminates the need for transportation of potentially [22] evaluated field-scale composting in cattle, quantifying infectious material. However, there is potential for bacterial 16S rDNA fragments using real-time PCR. detrimental environmental effects, specifically water quality Recently, the bacterial communities associated with carcass issues, as well as the risk that disease agents will persist in decomposition were investigated using an advanced the environment owing to the limited environmental sequencing technique to monitor the temporal changes of controls in place [9]. entire microbial communities [23]. Yang et al. [24] used high- As a consequence of those disadvantages of burial, throughput nucleotide sequencing analysis to investigate the alternative methods are required for the safe and rapid bacterial communities in leachates from the decomposition degradation of animal carcasses. Composting of infected of pig carcasses. Pechal et al. [25] studied skin and mouth livestock carcasses has been adopted as an environmentally bacterial communities during the decomposition of three sound method for carcass disposal in recent disease swine carcasses over 5 days. Metcalf et al. [23] conducted a outbreaks in Australia and North America [6]. Composting laboratory experiment to characterize temporal changes in has many different methods, such as aerobic, anaerobic, microbial communities associated with mouse carcasses as mechanical mixing, and microbial culture-added composting they decomposed on soil under controlled conditions for [5]. Recently, studies of pig carcass composting for emergency 48 days. Lindblad [26] monitored the change of the microbial disposal were conducted [10-13]. Akdeniz et al. [10] community during composting of pig carcasses for 13 monitored gas profiles during pig carcass composting for 6 months. However, previous studies have focused on the months. The safe disposal of carcasses during a disease changes in soil or compost bacterial community structure outbreak is a significant environmental and health issue for during the decomposition of carcasses through artificial both humans and other livestock. Composting carcasses situations using small quantities of tissue in the laboratory, and manure with bulking materials uses bacterial activity rather than whole organisms on the field-scale. to decompose the carcass tissues and generate heat, which For continual monitoring of the bacterial community can lead to temperatures high enough to kill targeted during pig carcass decomposition in a soil pit or compost bin, pathogenic microorganisms [6]. repetitive sampling of the envelope materials is necessary. The decomposition of carcasses causes a change in gas Dredging soil or compost is the most representative sampling profiles and bacterial community structures during both method, but it causes significant microenvironmental soil burial and composting [14]. A variety of odor gases, changes. Samples from a soil burial site were taken by including sulfur and nitrogen compounds, acids, aldehydes, careful excavation at several locations close to pig carcasses and cyclic hydrocarbons, are released by decaying pig avoiding as far as possible sites of disturbance [27]. carcasses [13-16]. Some researchers have been evaluating Envelope material samples were collected through ports the availability of volatile organic compounds (VOCs) as installed in the sidewalls of a compost bin to minimize markers for the diagnosis of pig carcass decomposition disturbance [13]. Bergmann et al. [21] collected samples processes [10, 14]. To collect odor gases from soil burial through port side perforations. These methods have sites and compost bins, passive sampling devices that use a limitations in collecting envelope materials from a soil diffusive sampler have been developed to replace special burial site or composting apparatus. In this study, we air sampling probes inside the pits and bins [14]. proposed special sampling devices designed to collect The bacterial community is another important factor samples from a soil burial pit or compost bin without influencing the decomposition of carcasses [17, 18]. Several disturbance. studies have focused on bacterial ecology at burial and Although soil burial and composting methods have been composting processes. Lipolytic bacteria were quantitatively widely used for pig carcass disposal, there is still insufficient evaluated by quantitative polymerase chain reaction using information on the temporal trends of the bacterial lipase-specific primers and lipolytic bacterial counts [19]. community during pig decomposition in these methods. To Bacterial community dynamics during pig leg decomposition characterize the temporal trends, pilot-scale test systems to in soil were studied using 16S rDNA polymerase chain simulate soil burial and composting with passive aeration reaction-denaturing gradient gel electrophoresis (PCR- were constructed in the field, and the changes in gas DGGE) [20]. Bergmann et al. [21] used real-time PCR-DGGE profiles inside the systems were monitored for 340 days. based on 16S rRNA to characterize the change in soil Moreover, the internal soil and compost from the systems

J. Microbiol. Biotechnol. Bacterial Community Dynamics in Pig Carcass Decomposition 2201

were sampled using the special sampling devices, and then slaughterhouse in Suncheon, Korea. Animal ethics approval was their bacterial community dynamics were analyzed by not required as the experimental subjects were not killed specially high-throughput pyrosequencing. The roles of bacteria for research purposes but purchased postmortem. The compost associated with the decomposition of pig carcasses are used in this study was matured for 6 months after mixing pig discussed. manure with sawdust (6:4 (v/v)). A support rack made of stainless steel (1.6 m (L) × 1.2 m (W) × 2.0 m (H)) was designed to minimize the movement of the disposed carcass during Materials and Methods decomposition and to support the samplers (gas sampler and soil/compost sampler) and temperature sensor (Fig. 1B). Four gas Materials and Apparatus samplers consisted of a gas collector (perforated pipe, 0.20 m (L) × Fig. 1 shows our pilot-scale test systems for soil burial and 0.15 m (o.d.)) and collection polyvinyl chloride pipe (50 mm Φ) composting of pig carcasses. The test systems were composed of (Fig. 1C). The soil and compost samplers were designed to collect the following items: eight adult pigs, fully mature compost, samples at the desired position in the test systems without sample support racks for the pig carcasses, soil and compost samplers, disturbance (Fig. 1D). They were made of stainless steel (SUS 310) four gas samplers, and a temperature sensor. Eight pig carcasses and contained inner and outer pipes. The outer pipes (1.83 m (L) × (Middle Yorkshire of 80−90 kg) were purchased from a licensed 42.7 mm (o.d.)) had four sampling ports (50 mm) arranged at

Fig. 1. Schematic diagrams of the pilot-scale test bed used for soil burial and composting of pig carcasses. (A) Simulated installation of the soil burial and composting test units in a greenhouse, (B) the supporting rack, (C) gas samplers, (D) the soil and compost sampler, and (E) and a conceptual illustration of how to use the sampler.

December 2017 ⎪ Vol. 27⎪ No. 12 2202 Ki et al.

Fig. 1. Continued. intervals of 0.35 m. The inner pipes (1.82 m (L) × 34 mm (O.D.)) Temperature data were collected using a data acquisition system had four sampling capsules. The sampling port and sample (HM-10; Yokogawa, Japan). capsule in the outer and inner pipes of the sampler, respectively, were aligned by a design that allowed the inner pipe to be Setup and Operation of Simulated Soil Burial and Composting inserted into the outer pipe up to approximately 0.07 m to allow of Pig Carcasses the sample capsule containing soil and compost samples to be The pilot-scale test bed was located on a farm in Suncheon, moved inside the sampling port without disturbing the sample. Korea. A composting test system made of concrete (2.8 m (L) × The conceptual illustration of how to use the sampler is shown in 2.8 m (W) × 2.6 m (H)) with an inner floor area of 2.4 m × 2.4 m Fig. 1E. Temperature sensors were included to monitor changes in was prepared. A perforated pipe (polyethylene, 250 mm Φ) was the temperature at each point during the decomposition process. installed on the bottom and was covered with gravel (50–100 mm) The thermocouples (made of constantan copper-nickel) were to enable natural ventilation. The support rack was put into the inserted into a stainless tube with an outer diameter of 8 mm. test system and a 0.5 m layer of mature compost was added. Next,

J. Microbiol. Biotechnol. Bacterial Community Dynamics in Pig Carcass Decomposition 2203

four pig carcasses were put on the first shelf of the support rack Nagel GmbH, Germany) and quantified using an ASP-2680 and the compost samplers were installed at the top and bottom spectrophotometer (ACTGene, USA). sampler holes in the support rack. Temperature sensors were set The multiplex PCR process using the 340F and 805R composite at different heights from the floor: 0.5 m (under the pig), 1.4 m (on primer sets was well described in a previous study [29]. The the pig), and 2.0 m (over the pig). Two gas samplers were installed expected size bands from the agarose gel containing the PCR at 0.6 m from the bottom (under the pig) and 1.2 m from the products were extracted, and the purified bands were used for bottom (over the pig). After placing the pigs and instruments, the analysis using pyrosequencing as in the previous study [29]. The rest of the bin was filled with compost. The mixing ratio of purified DNAs were sent to Macrogen Incorporation (Korea) to be carcass-to-compost was 1:10 (w/w). Because FMD is typically run on a Genome Sequencer 454 FLX Titanium system (Roche prevalent in the winter, the pig carcasses were buried in the Diagnostics, Germany). Analysis of sequencing data was performed beginning of February. using the Ribosomal Database Project (RDP) pyrosequencing For the simulated soil burial, sandy loam soil (pH 5.9 and 2.0% pipeline, as in the previous study [29]. Pyrosequences obtained in organic content) was dug from the ground surface to 3.0 m, and this study were uploaded to the National Center for Biotechnology placed in a pit (4.0 m (L) × 4.0 m (W) × 2.6 m (H)), and a concrete Information GeneBank database under the study accession number wall (2.8 m (L) × 2.8 m (W) × 2.6 m (H)) was installed on one side SRP084103 (http://www.ncbi.nlm.nih.gov). of the pit for the installation of the soil samplers and temperature Operational taxonomic units (OTUs) were determined at 3% sensors. The shielding membrane and support racks were dissimilarity. Chao1 species richness and Shannon diversity installed inside the test system and then, following the same indices were calculated using the RDP pyrosequencing pipeline procedure used for composting, the burial of pig carcasses and [29]. Sequences representing OTUs were compared with those of installation of soil samplers, temperature sensors, and gas the bacterial type strains using the EzTaxon server ver. 2.1 (http:// samplers were conducted. www.eztaxon.org). Pyrosequencing reads of each library were To prevent rainwater infiltration and odor complaints from the taxonomically assigned using the RDP classifier. The classification neighbors, the greenhouses were installed to cover the disposal results were imported into the MEGAN software for comparison facilities. Then, the temperature, pH, bacterial communities, and of the bacterial communities from the soil burial and composting odor gases were monitored for 340 days. processes.

Odor Gas Analysis Data Analysis Odor gas was collected through the gas samplers (Fig. 1C) and Principal component analysis (PCA) and canonical correspondence analyzed to determine the concentrations of 22 odor gases, using analysis (CCA) were conducted using CANOCO 4.5 (Biometris- the standard test method for odor compounds in Korea [8]. Odor Plant Research International, Netherlands). The odor gases gas samples were collected from the top and bottom layers at 36, emitted from the bottom bed were subjected to PCA to elucidate 64, 86, 114, 147, 161, 197, 225, 282, 303, and 335 days. The major variations and patterns. The gas profiles from the bottom concentrations of odor gases have been previously characterized, bed can be directly related to the decomposition of the pig and raw data have been described in the previous study [8]. The carcasses. PCA was done with the entire pyrosequencing dataset sum of odor quotient (SOQ) was evaluated to reflect the relative to investigate shifts in the structure of bacterial communities. strengths of individual odors by considering the direct relationship To reveal the relationship between the bacterial community between the absolute concentration and the threshold values of structures and odor gases, their data testing was performed using each individual compound [28]. The odor quotient (OQ) was CCA. There were dominant species data (34 and 38 species in the calculated based on the results of the intensities of odor gases, soil and compost samples, respectively, 1% more than the number using the threshold values of each individual odors as described in the bacterial community) and 16 odor gases data (ammonia in Eq. (1). (NH3), trimethylamine (TMA), hydrogen sulfide (H2S), methanethiol (MeSH), dimethyl sulfide (DMS), dimethyl disulfide (DMDS), SOQ () X ==ΣOdor Quotient ΣOQ ()ith , (1) acetaldehyde (A-A), propioaldehyde (P-A), butyraldehyde (B-A), Odor concentration of the ith component () ppb n-valeraldehyde (n-V-A), iso-valeraldehyde (i-V-A), VOCs, propionic where OQ (ith) = ------Threshold value of the ith component () ppb acid (P-acid), n-butyric acid (n-B-acid), n-valeric acid (n-V-acid), and iso-valeric acid (i-V-acid)) were used for the CCA. All the Bacterial Community Analysis Using Pyrosequencing other factors were standardized by calculating log(X + 1) values. The envelope material samples were collected on days 0, 45, 161, and 225. The sampling location is near the pig carcass and gas Results and Discussion sampler (position B4 in Fig. 1A). At each sampling time, a 0.5 g sample was collected into microtubes. A total of 16 microtubes Change in Odor Gases Generated during Decomposition (two of each sample) were used for DNA extraction. DNA was of Pig Carcasses extracted individually using the NucleoSpin soil kit (Macherey- Comparison of the odor gases at the bottom layers in the

December 2017 ⎪ Vol. 27⎪ No. 12 2204 Ki et al.

active decay, advanced decay, and skeletonization. During

the bloated stage, NH3, TMA, and sulfur compounds were detected [14]. The active decay stage had the strongest olfactive signature because many chemicals were detected

[14]. At this decay stage, NH3, TMA, and sulfur compounds

such as DMDS and H2S were present, and a great many organic acids also appeared [14, 30]. In composting, the period from the 64th to 197th day is the active stage, in which the intensities of organic acids and sulfur compounds

such as MeSH, DMS, DMDS and H2S reach their peak with the active degradation (Table S2). In the soil burial, the period from the 147th to 161st day is the active stage, in which the intensities of organic acids, TMA and DMDS reach their peak (Table S1). During the advanced decay stage when most of the soft tissue mass has been liquefied and degraded, the portion of aldehydes increases [14]. In the soil burial, the active decay stage turned into the advanced decay stage after the 282nd day (Table S1 and Fig. 2A). However, in the composting, the advanced decay stage occurred after the 197th day (Table S2 and Fig. 2B).

Changes in Bacterial Community Structures The changes in the bacterial community structures during the decomposition of pig carcasses in the soil pit are shown in Fig. 3. The most dominant bacteria were Arthrobacter (10.9%) and Lysobacter (10.3%) in the early stages, and Arthrobacter (4.0%) and Massilia (2.5%) at the 45th day. During the most active decay (161 days), Paraliobacillus (5.45%), Brevundimonas (5.45%), Gelidibacter (5.3%), and Mucilaginibacter (4.1%) were dominant. At 225 days, Pseudomonas were the most abundant bacteria (23.5%), followed by Gelidibacter (14.2%), Pseudaminobacter (3.1%), and Mucilaginibacter (2.8%). Paraliobacillus was isolated from a decomposing marine Fig. 2. Comparison of odor gases by principal component alga and is a halophilic, extremely halotolerant, alkaliphilic, analysis. and facultatively anaerobic gram-positive bacterium [31]. (A) Simulated soil burial, and (B) composting test system. Brevundimonas is an aerobic gram-negative bacterium [32]. Gelidibacter was isolated from sediment and is an aerobic soil burial and composting systems by PCA is shown in gram-negative bacterium [33]. Mucilaginibacter was isolated Fig. 2. As shown in the PCA results, the gas profiles from from a coast soil and is a facultatively anaerobic gram the soil burial tended to change with time until the 147th day negative bacterium [34]. Pseudomonas is a representative and were grouped from the 147th to 161st day (Fig. 2A), soil aerobic gram-negative bacterium [35]. but the profiles from the composting were grouped from The changes in the bacterial community structures during the 64th to 197th day (Fig. 2B). This result indicates that the decomposition of pig carcasses in the composting bin decomposition reached the active decay stage more quickly are shown in Fig. 4. The most dominant bacteria were in the composting. Terribacillus (19.7%) and Sporosarcina (17.4%) in the early Based on the PCA results and odor gas profiles, the stages, and Sporanaerobacter (13.6%) and Lentibacillus (11.6%) decomposition of pig carcasses in the soil burial and at the 45th day. At the most active decay (161st day), composting systems was categorized into five decompositional Pelotomaculum (17.9%), Lentibacillus (14.2%), and Clostridium stages as defined by Dekeirsschieter et al. [14]: fresh, bloat, (9.8%) were dominant. At 225 days, Clostridium (13.2%),

J. Microbiol. Biotechnol. Bacterial Community Dynamics in Pig Carcass Decomposition 2205

Changes in the dominant genera during the degradation Fig. 3. Changes in the dominant genera during the degradation Fig. 4. of pig carcasses. of pig carcasses. (A) Major groups and (B) minor groups in the simulated soil burial. (A) Major groups and (B) minor groups in the composting test system.

Caldicoprobacter (10.1%), and Sedimentibacter (8.8%) were from the soil burial were grouped from the beginning of dominant. Pelotomaculum is an anaerobic gram-positive, the burial to the 45th day, and tended to change with time mesophilic, thermophilic and propionate-oxidizing bacterium until the 225th day. This suggests that the bacterial [36]. Lentibacillus is an aerobic gram-variable, endospore- community structures changed constantly over that time, forming bacterium, which was isolated from a salt field which illustrates the same trend as the richness and diversity [37]. Clostridium is an anaerobic gram-positive, moderately of the bacterial community described in Table 1. The highest thermophilic, endospore-forming bacterium, which was values of Chao1 species richness and Shannon-Weaver isolated from Thai compost [38]. Caldicoprobacter is an diversity indices (H’) were 754 and 5.550, respectively, at anaerobic gram-positive, thermophilic bacterium, which 45 days after burial, and then both values decreased was enriched from composted cattle manure [39]. Different (Table 1). On the other hand, the bacterial community bacteria played important roles in the decomposition of pig structures from composting changed drastically from the carcasses in the soil and compost. beginning, and were grouped from the 45th to 161st day Fig. 5 shows the result of a PCA of the bacterial community and then changed until the 225th day. This result is in good structures of soil and compost samples. The first two agreement with the analysis of the richness and diversity of ordination axes explained 32.4% and 26.3% of the total the bacterial community (Table 1). The values of Chao1 and variation in species data. The bacterial community structures H’ were 198 and 3.417, respectively, at the early stage of

December 2017 ⎪ Vol. 27⎪ No. 12 2206 Ki et al.

from incomplete anaerobic degradation of organic matter, especially proteins and carbohydrates [40]. This incomplete degradation results in the production of offensive odorous compounds [41]. The CCA results, which explain the correlation between the bacterial community structures and the odor gases emitted from the bottom layers in the soil burial and composting systems, are shown in Fig. 6. Sulfur-containing compounds are produced by anaerobic bacteria via sulfate reduction and the metabolism of sulfur- containing amino acids (cysteine and methionine) [40]. Mucilaginibacter, Alcaligenes, Pseudaminobacter, and Pseudomonas were correlated with sulfur compounds (DMS, DMDS, and MeSH) in the soil pit (Fig. 6A). Mandal et al. [42] reported that sulfur oxidation in Pseudaminobacter occurs through the combined action of several enzymes encoded by a thiosulfate-inducible sulfur oxidation operon. Tomita et al. [43] reported that Alcaligenes and Pseudomonas formed DMDS from amino acids. In this study, Pseudomonas was the most dominant in the soil pit at the 225th day (23.4% of the total bacterial communities) (Fig. 3A), and high intensities Fig. 5. Comparison of the bacterial community structures by principal component analysis. of MeSH, DMS, and DMDS were also observed at the soil bottom at around the 225th day (Table S1). In the composting, Caldicoprobacter, Lentibacillus, and Sporanaerobacter were Table 1. Richness and diversity of the bacterial community in correlated with sulfur compounds (Fig. 6B). Bouanane- the soil burial and composting samples. Darenfed et al. [44] reported on sulfur-reducing activity in Sample Operation time (d) OTUa Chao1b H’ c Caldicoprobacter. In this research, Caldicoprobacter was found Soil 0 224 572 4.753 to be the dominant genus in the compost bin at the 45 343 754 5.550 225th day (10.1% of total bacterial communities) (Fig. 4A),

161 226 470 4.799 and significant increases in the intensity of H2S at the 225 186 457 4.094 compost bottom were also observed at around the 225th day Compost 0 96 198 3.417 (Table S2). Ammonia (NH ) can be produced from urea and nitrates 45 147 279 3.959 3 [40, 45, 46]. The primary biological forms of nitrogen, such 161 184 377 4.119 as proteins, bacterial cell walls, and nucleic acids, are found 225 173 383 4.281 in wastewater. These nitrogen-containing compounds are aOperational taxonomic units. + mineralized into ammonium (NH4 ), which deprotonates, bChao1 species richness indices. resulting in NH3 emissions [13]. Caldicoprobacter and cShannon-Weaver diversity indices. Paraliobacillus in the composting were correlated with NH3 (Fig. 6B). decomposition, and from the 45th to the 222nd day, each TMA is a malodorous compound usually produced from value remained almost the same, with a range of 279-383 choline, betaine, or trimethylamine N-oxide (present in for Chao1 and 3.959-4.281 for H’ (Table 1). The PCA marine fish) by bacterial activity [39]. The production of results indicate that the shift in the bacterial community TMA was estimated to correlate with Paenochrobactrum, structures during pig carcass decomposition were reached Pedobacter, and Thiobacillus in the soil burial (Fig. 6A) and earlier by composting than by soil burial. Clostridium, Gracilibacter, Lentibacillus, Sedimentibacter, and Syntrophomonas in the composting (Fig. 6B). The formation Correlation of Bacterial Community Structures with Odor of nitrogen-containing compounds via Pedobacter [47], Gases Thiobacillus [48], Clostridium [49, 50], and Gracilibacter [47] The odor gases emitted from pig carcasses result primarily has been reported. The presence of Thiobacillus in the soil

J. Microbiol. Biotechnol. Bacterial Community Dynamics in Pig Carcass Decomposition 2207

Fig. 6. Canonical correspondence analysis of the bacterial species in relation to odor gases. (A) Soil samples in simulated soil burial, and (B) compost samples in the composting test system. pit was the highest on the 161st day (1.82% of total bacterial intensity from the soil bottom was found to increase communities) and then decreased (Fig. 3B). The TMA significantly from 7.0 at the early stage to 1.7E+4 on the

December 2017 ⎪ Vol. 27⎪ No. 12 2208 Ki et al.

161st day and then to decrease (Table S1). Clostridium in the conditions, is useful for predicting and characterizing the compost bin increased gradually and had a prevalence of degradation of pig carcasses in the soil burial and composting 13.2% on the 225th day (Fig. 4A). It was also found that the disposal methods. In addition, the experimental apparatus TMA intensity from the compost bottom was 2.0 at the used in this study can be used to study the degradation beginning, increased steadily, and reached 2.7E+4 on the process of other animal carcasses by soil burial or composting. 225th day (Table 2). The major sources of bacterial volatile fatty acids Acknowledgments production are protein and amino acid catabolism [41]. The degradation of proteins not only produces straight-chain This study is supported by the Ministry of Environment carboxylic acids, but also branched-chain fatty acids, sulfur (Korea Environmental Industry & Technology) as The Eco- compounds, amines, ammonia, phenols, and indoles [45, Innovation Project (2014000110009). 46]. Organic acids are normally produced as metabolic intermediates or end products from a range of different References bacteria. It was estimated that the production of organic acids correlated with Acidovorax, Mucilaginibacter, and 1. Yoon H, Yoon SS, Wee SH, Kim YJ, Kim B. 2012. Clinical Rhodanobacter in the soil pit (Fig. 6A) and Halothermothrix, manifestations of foot-and-mouth disease during the 2010/ Virgibacillus, and Caldicoprobacter in the composting bin 2011 epidemic in the Republic of Korea. Transbound. Emerg. (Fig. 6B). The formation of organic acids via Acidovorax [51], Dis. 59: 517-525. Mucilaginibacter [52], Halothermothrix [53], and Caldicoprobacter 2. Hayama Y, Kimura Y, Yamamoto T, Kobayashi S, Tsutsui T. 2015. Potential risk associated with animal culling and has been reported. Acidovorax was a dominant genus in the disposal during the foot-and-mouth disease epidemic in soil pit on the 225th day (1.32% of the total bacterial Japan in 2010. Res. Vet. Sci. 102: 228-230. communities) (Fig. 3B), and the P-acid intensity in the soil 3. OIE. 2017. World Animal Health Information Database bottom increased to 1.6E+2 at around the 225th day (WAHID) Interface. Available at http://www.oie.int/animal- (Table S1). Mucilaginibacter was a dominant genus in the health-in-the-world/the-world-animal-health-information-system/ soil pit on the 161st day, accounting for 4.13% of total data-after-2004-wahis-interface/. Accessed Sep. 30, 2017. bacterial communities, and then the dominance decreased 4. Yoon H, Yoon SS, Kim YJ, Moon OK, Wee SH, Joo YS, et al. (Fig. 3A). The n-V-acid intensity in the soil bottom increased 2015. Epidemiology of the foot-and-mouth disease serotype from 6.8E+1 to 2.5E+2 at around the 161st day (Table S1). A O epidemic of November 2010 to April 2011 in the Republic gradual increase in the dominance of Caldicoprobacter was of Korea. Transbound. Emerg. Dis. 62: 252-263. found in the compost bin on the 225th day, constituting 5. Gwyther CL, Williams AP, Golyshin PN, Edwards-Jones G, 10.1% of the total bacterial communities (Fig. 4A), and the Jones DL. 2011. The environmental and biosecurity characteristics of livestock carcass disposal methods: a review. Waste i-V-acid intensity reached 2.8E+4 in the compost bottom Manag. 31: 767-778. (Table S2). Halothermothrix and Virgibacillus were found to 6. Won SG, Park JY, Rahman MM, Park KH, Ra CS. 2016. Co- be dominant genera in the compost bin on the 225th day composting of swine mortalities with swine manure and (1.7% and 1.5%, respectively) (Fig. 4B), and the P-acid sawdust. Compost Sci. Util. 24: 42-53. intensity reached 1.3E+3 in the compost bottom (Table S2). 7. Brasseur C, Dekeirsschieter J, Schotsmans EM, de Koning S, To characterize the temporal trends of the gas profiles Wilson AS, Haubruge E, et al. 2012. Comprehensive two- and bacterial community during pig decomposition in soil dimensional gas chromatography-time-of-flight mass spectrometry burial and composting systems, pilot-scale test systems to for the forensic study of cadaveric volatile organic compounds simulate soil burial and composting were constructed in released in soil by buried decaying pig carcasses. J. Chromatogr. the field and monitored for 340 days. The gas profiles A 1255: 163-170. suggest that the composting method reached the active and 8. Chae JS, Jeon JM, Oh KC, Kim SD, Ryu HW. 2016. Decaying advanced decay stages about 100 days earlier than the soil characteristics of pig carcass disposal by trench burial method using compost. J. Odor Indoor Environ. 15: 134-146. burial method. The predominant bacteria were Pseudomonas 9. Yuan Q, Snow DD, Bartelt-Hunt SL. 2013. Potential water and Gelidibacter in the soil, and Pelotomaculum, Lentibacillus, quality impacts originating from land burial of cattle Clostridium, and Caldicoprobacter in the compost. This carcasses. Sci. Total Environ. 456-457: 246-253. temporal information, which was obtained through a long- 10. Akdeniz N, Koziel JA, Ahn HK, Glanville TD, Crawford BP. term analysis of the pig carcass decomposition process 2010. Field scale evaluation of volatile organic compound using special experimental devices to simulate actual field production inside biosecure swine mortality composts. Waste

J. Microbiol. Biotechnol. Bacterial Community Dynamics in Pig Carcass Decomposition 2209

Manag. 30: 1981-1988. model system. Elife 2: e01104. 11. Akdeniz N, Koziel JA, Ahn HK, Glanville TD, Crawford BP, 24. Yang SH, Lim JS, Khan MA, Kim BS, Choi DY, Lee EY, et al. Raman DR. 2010. Laboratory scale evaluation of volatile 2015. High-throughput nucleotide sequence analysis of diverse organic compound emissions as indication of swine carcass bacterial communities in leachates of decomposing pig degradation inside biosecure composting units. Bioresour. carcasses. Genet. Mol. Biol. 38: 373-380. Technol. 101: 71-78. 25. Pechal JL, Crippen TL, Benbow ML, Tarone AM, Dowd S, 12. Akdeniz N, Koziel JA, Glanville TD, Ahn H, Crawford BP. Tomberlin JK. 2014. The potential use of bacterial community 2011. Air sampling methods for VOCs related to field-scale succession in forensics as described by high throughput biosecure swine mortality composting. Bioresour. Technol. metagenomic sequencing. Int. J. Legal Med. 128: 193-205. 102: 3599-3602. 26. Lindblad L. 2007. Microbiological sampling of swine carcasses: 13. Glanville TD, Ahn H, Akdeniz N, Crawford BP, Koziel JA. a comparison of data obtained by swabbing with medical 2016. Performance of a plastic-wrapped composting system gauze and data collected routinely by excision at Swedish for biosecure emergency disposal of disease-related swine abattoirs. Int. J. Food Microbiol. 118: 180-185. mortalities. Waste Manag. 48: 483-491. 27. Hopkins D, Wiltshire P, Turner B. 2000. Microbial characteristics 14. Dekeirsschieter J, Verheggen F, Gohy M, Hubrecht F, of soils from graves: an investigation at the interface of soil Bourguignon L, Lognay G, et al. 2009. Cadaveric volatile microbiology and forensic science. Appl. Soil Ecol. 14: 283-288. organic compounds released by decaying pig carcasses (Sus 28. Kim KH, Park SY. 2008. A comparative analysis of malodor domesticus L.) in different biotopes. Forensic Sci. Int. 189: 46-53. samples between direct (olfactometry) and indirect (instrumental) 15. Dekeirsschieter J, Stefanuto PH, Brasseur C, Haubruge E, methods. Atmos. Environ. 4: 5061-5070. Focant JF. 2012. Enhanced characterization of the smell of 29. Kim TG, Cho KS. 2012. Microbial community analysis of a death by comprehensive two-dimensional gas chromatography- methane-oxidizing biofilm using ribosomal tag pyrosequencing. time-of-flight mass spectrometry (GCxGC-TOFMS). PLoS One J. Microbiol. Biotechnol. 22: 360-370. 7: e39005. 30. Forbes SL, Perrault KA. 2014. Decompositiono odour 16. Stadler S, Stefanuto PH, Brokl M, Forbes SL, Focant JF. profiling in the air and soil surrounding vertebrate carrion. 2012. Characterization of volatile organic compounds from PLoS One 9: e95107. human analogue decomposition using thermal desorption 31. Ishikawa M, Ishizaki S, Yamamoto Y, Yamasato K. 2002. coupled to comprehensive two-dimensional gas chromatography– Paraliobacillus ryukyuensis gen. nov., sp. nov., a new gram- time-of-flight mass spectrometry. Anal. Chem. 85: 998-1005. positive, slightly halophilic, extremely halotolerant, facultative 17. Li L, Wang J, Wang Y. 2016. A comparative study of the anaerobe isolated from a decomposing marine alga. J. Gen. decomposition of pig carcasses in a methyl methacrylate Appl. Microbiol. 48: 269-279. box and open air conditions. J. Forensic Leg. Med. 42: 92-95. 32. Straif SC, Mbogo CN, Toure AM, Walker ED, Kaufman M, 18. Metcalf JL, Xu ZZ, Weiss S, Lax S, Van Treuren W, Hyde ER, Toure YT, et al. 1998. Midgut bacteria in Anopheles gambiae et al. 2016. Microbial community assembly and metabolic and An. funestus (Diptera: Culicidae) from Kenya and Mali. function during mammalian corpse decomposition. Science J. Med. Entomol. 35: 222-226. 351: 158-162. 33. Zhang DC, Margesin R. 2015. Gelidibacter sediminis sp. nov., 19. Howard GT, Duos B, Watson-Horzelski EJ. 2010. isolated from a sediment sample of the Yellow Sea. Int. J. Characterization of the soil microbial community associated Syst. Evol. Microbiol. 65: 2304-2309. with the decomposition of a swine carcass. Int. Biodeterior. 34.Yoon JH, Kang SJ, Park S, Oh TK. 2012. Mucilaginibacter Biodegradation 64: 300-304. litoreus sp. nov., isolated from marine sand. Int. J. Syst. Evol. 20. Olakanye AO, Thompson T, Ralebitso-Senior TK. 2014. Microbiol. 62: 2822-2827. Changes to soil bacterial profiles as a result of Sus scrofa 35. Vodovar N, Vallenet D, Cruveiller S, Rouy Z, Barbe V, domesticus decomposition. Forensic Sci. Int. 245: 101-106. Acosta C, et al. 2006. Complete genome sequence of the 21. Bergmann R, Ralebitso-Senior TK, Thompson T. 2014. An entomopathogenic and metabolically versatile soil bacterium RNA-based analysis of changes in biodiversity indices in Pseudomonas entomophila. Nat. Biotechnol. 24: 673-679. response to Sus scrofa domesticus decomposition. Forensic Sci. 36. Imachi H, Sakai S, Ohashi A, Harada H, Hanada S, Int. 241: 190-194. Kamagata Y, et al. 2007. Pelotomaculum propionicicum sp. nov., 22. Xu W, Reuter T, Xu Y, Hsu YH, Stanford K, McAllister TA. an anaerobic, mesophilic, obligately syntrophic, propionate- 2011. Field scale evaluation of bovine-specific DNA as an oxidizing bacterium. Int. J. Syst. Evol. Microbiol. 57: 1487-1492. indicator of tissue degradation during cattle mortality 37. Yoon JH, Kang KH, Park YH, 2002. Lentibacillus salicampi composting. Bioresour. Technol. 102: 4800-4806. gen. nov., sp. nov., a moderately halophilic bacterium isolated 23. Metcalf JL, Parfrey LW, Gonzalez A, Lauber CL, Knights D, from a salt field in Korea. Int. J. Syst. Evol. Microbiol. 52: Ackermann G, et al. 2013. A microbial clock provides an 2043-2048. accurate estimate of the postmortem interval in a mouse 38. Sukhumavasi J, Ohmiya K, Shimizu S, Ueno K. 1988.

December 2017 ⎪ Vol. 27⎪ No. 12 2210 Ki et al.

Clostridium josui sp. nov., a cellulolytic, moderate thermophilic in piggery wastes and the possibility of applying indicator species from Thai compost. Int. J. Syst. Evol. Microbiol. 38: components for studying odour development. Agric. Environ. 179-182. 5: 241-260. 39. Holman DB, Hao X, Topp E, Yang HE, Alexander TW. 2016. 47. Beckmann S, Krüger M, Engelen B, Gorbushina AA, Effect of co-composting cattle manure with construction and Cypionka H. 2011. Role of bacteria, archaea and fungi demolition waste on the archaeal, bacterial, and fungal involved in methane release in abandoned coal mines. microbiota, and on antimicrobial resistance determinants. Geomicrobiol. J. 28: 347-358. PLoS One 11: e0157539. 48. Hirano T, Kurosawa H, Nakamura K, Amano Y. 1996. 40. Rappert S, Muller R. 2005. Odor compounds in waste gas Simultaneous removal of hydrogen sulfide and trimethylamine emissions from agricultural operations and food industries. by a bacterial deodorant. J. Ferment. Bioeng. 81: 337-342. Waste Manag. 25: 887-907. 49. Gómez-Brandón M, Juárez MFD, Zangerle M, Insam H. 41. Zhu J. 2000. A review of microbiology in swine manure 2016. Effects of digestate on soil chemical and microbiological odor control. Agric. Ecosyst. Environ. 78: 93-106. properties: a comparative study with compost and 42. Mandal S, Chatterjee S, Dam B, Roy P, Gupta SKD. 2007. vermicompost. J. Hazard. Mater. 302: 267-274. The dimeric repressor SoxR binds cooperatively to the 50. Wang P, Sun G, Jia Y, Meharg AA, Zhu Y. 2014. A review on promoter(s) regulating expression of the sulfur oxidation (sox) completing arsenic biogeochemical cycle: microbial volatilization operon of Pseudaminobacter salicylatoxidans KCT001. Microbiology of arsines in environment. J. Environ. Sci. 26: 371-381. 153: 80-91. 51. Xing D, Ren N, Gong M, Li J, Li Q. 2005. Monitoring of 43. Tomita B, Inoue H, Chaya K, Nakamura A, Hamamura N, microbial community structure and succession in the Ueno K, et al. 1987. Identification of dimethyl disulfide-forming biohydrogen production reactor by denaturing gradient gel bacteria isolated from activated sludge. Appl. Environ. electrophoresis (DGGE). Sci. China C Life Sci. 48: 155-162. Microbiol. 53: 1541-1547. 52. Bertin L, Bettini C, Zanaroli G, Fraraccio S, Negroni A, Fava F. 44. Bouanane-Darenfed A, Fardeau ML, Grégoire P, Joseph M, 2012. Acclimation of an anaerobic consortium capable of Kebbouche-Gana S, Benayad T, et al. 2011. Caldicoprobacter effective biomethanization of mechanically-sorted organic algeriensis sp. nov. a new thermophilic anaerobic, xylanolytic fraction of municipal solid waste through a semi-continuous bacterium isolated from an Algerian hot spring. Curr. Microbiol. enrichment procedure. J. Chem. Technol. Biotechnol. 87: 1312- 62: 826-832. 1319. 45. Mackie RI, Stroot PG, Varel VH. 1998. Biochemical 53. Niu Q, Takemura Y, Kubota K, Li YY. 2015. Comparing identification and biological origin of key odor components mesophilic and thermophilic anaerobic digestion of chicken in livestock waste. J. Anim. Sci. 76: 1331-1342. manure: microbial community dynamics and process resilience. 46. Spoelstra S. 1980. Origin of objectionable odorous components Waste Manage. 43: 114-122.

J. Microbiol. Biotechnol.