Journal of Environmental Management 134 (2014) 39e46
Contents lists available at ScienceDirect
Journal of Environmental Management journal homepage: www.elsevier.com/locate/jenvman
Decline in extractable kitasamycin during the composting of kitasamycin manufacturing waste with dairy manure and sawdust Nengfei Ding a, Weidong Li b, Chen Liu a, Qinglin Fu a, Bin Guo a, *, Hua Li a, Ningyu Li a, Yicheng Lin a a b
Institute of Environment, Resource, Soil and Fertilizer, Zhejiang Academy of Agricultural Sciences, Hangzhou, China Department of Science, Qianjiang College of Hangzhou Normal University, Hangzhou 310012, China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 11 June 2013 Received in revised form 18 December 2013 Accepted 22 December 2013 Available online 23 January 2014
The aim of this study was to propose a feasible treatment of kitasamycin manufacturing waste by examining extractable kitasamycin and evaluating its compost maturity during the composting of waste with different ratios of dairy manure and sawdust over a 40-day period (volume/volume/volume; M1, 0/ 80/20; M2, 10/70/20; and M3, 30/50/20). During composting, the concentration of extractable kitasamycin in kitasamycin-contaminated composts declined rapidly, and was undetectable in M2 within 15 days. M2 also achieved the highest fertility compost, which was characterised by the following final parameters: electrical conductivity, 2.34 dS cm1; pH, 8.15; total C/N, 22.2; water-soluble NHþ 4 , P, and K, 0.37, 3.43, and 1.05 g kg1, respectively; and plant germination index values, 92%. Furthermore, DGGE analysis showed a dramatic increase in the diversity of bacterial species during composting. In contrast, a high concentration (121 mg kg1) of extractable kitasamycin still remained in the M3 compost, which exerted an inhibitory effect on the composting, resulting in reduced bacterial diversity, high values of electrical conductivity and water-soluble NHþ 4 , a low C/N ratio, and a low plant germination index value. Furthermore, 3.86 log (CFU g1) kitasamycin-resistant bacteria were still present on day 40, indicating the biological degradation contributed to the decline of extractable kitasamycin. Ó 2014 Elsevier Ltd. All rights reserved.
Keywords: Composting DGGE Kitasamycin waste Microbial community Water-soluble nutrients
1. Introduction The treatment of antibiotic manufacturing waste, which is a residue of the organic medium used for microbial fermentation during antibiotic production, has become a difficult problem in China. Due to the high levels of antibiotic residuals, this biowaste has been utilised as an additive to animal feed to promote growth and to prevent pathogenic bacterial infection. However, incomplete assimilation in animals has resulted in 70e90% of the antibiotic being excreted in manure (Phillips et al., 2004; Kumar et al., 2005); the manure is not intensively treated and is often applied directly to agricultural fields, which contributes to the increase in antibiotics and antibiotic-resistance genes in the environment. The spread of antibiotic residues in the environment poses a potential risk to aquatic and terrestrial organisms (Hamscher et al., 2000). In addition to reducing the diversity of micro-organisms in the environment (Zielezny et al., 2006), this phenomenon can also enhance
* Corresponding author. Tel./fax: þ86 571 8640 4041. E-mail addresses: [email protected], [email protected] (B. Guo). 0301-4797/$ e see front matter Ó 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jenvman.2013.12.030
human antimicrobial resistance due to the consumption of crops grown in the antibiotic-contaminated soils (Keen and Montforts, 2011). Thus, appropriate techniques are needed introduced to reduce the antibiotic level of manufacturing waste prior to agricultural utilisation. Composting is an effective methodology for the decomposing of organic wastes via the biological degradation of organic constituents under controlled conditions, and has been successfully applied for the treatment of persistent organic pollutants (Laine and Jorgensen, 1997; Zeng et al., 2011). Recently, additional information regarding antibiotic degradation during manure composting has become available. A study on chlortetracycline in aged and spiked poultry manure demonstrated that more than 90% of the initial level of chlortetracycline was depleted under aerobic composting (Bao et al., 2009). Both full-scale and lab-scale investigations have indicated that decreases in tetracycline and sulphonamide concentrations are highly dependent on the presence of sawdust during composting (Kim et al., 2012). Dolliver et al. (2008) suggested that stockpiling may be a practical and economical option for livestock producers to reduce antibiotic levels in manure.
40
N. Ding et al. / Journal of Environmental Management 134 (2014) 39e46
As a by-product of microbial fermentation, antibiotic manufacturing waste cannot be composted alone due to its high nitrogen content (Table 1), which could enhance the loss of N via volatilisation and phototoxic molecules may simultaneously build up due to insufficient biodegradation of the organic matter (Bernal et al., 2009). Normally, high-nitrogen wastes are mixed with adjusted by the carbon-rich materials prior to composting to obtain an initial C/N ratio of between 25 and 35 (Bernal et al., 2009); however, these values are not absolute. An economic analysis showed that composting with a low initial C/N ratio (20) is beneficial to efficiently dispose of a high amount of N-rich waste (swine manure) (Zhu, 2007). Furthermore, to achieve high grade composting, other factors that affect the composting process, such as pH, electrical conductivity (EC), and the available nutrients, should also be well balanced. However, the formulation of composting mix between the antibiotic manufacturing waste and other organic materials is currently unavailable. Another important issue is the effect and fate of the antibiotics present in the manufacturing waste throughout the composting process. Previous studies have shown that antibiotic residues in manures did not have a significant effect on the composting process because the concentrations of extractable antibiotics decreased rapidly in during the first 3e10 days of composting (Arikan et al., 2007; Dolliver et al., 2008; Bao et al., 2009; Kim et al., 2012). However, antibiotic manufacturing wastes contain much higher extractable antibiotic (>10 fold) than those reported in previous manure studies (Table 1; Dolliver et al., 2008; Bao et al., 2009; Arikan et al., 2006, 2007; 2009; Kim et al., 2012). At present, the effect of high concentrations of extractable antibiotics in antibiotic waste on microbial activities during composting is poorly understood. A comparison of the bacteria composition between the antibiotic-free and antibiotic-contaminated composts during the process of composting may contribute to a better understanding of the antibiotic degradation process. This process is associated with the succession of microbial communities during composting. Advanced molecular biological techniques, such as polymerase chain reactionedenaturing gradient gel electrophoresis (PCRDGGE), have proven to be useful for the detection of this succession (Cahyani et al., 2003; Nakasaki et al., 2009). Kitasamycin is one of the antibiotics that is highly active against a wide range of Gram-positive bacteria and has been widely used in the rearing of food-producing animals to prevent and treat diseases (Jordan and Knight, 1984). It has been estimated that 6300 tons of kitasamycin waste are annually released from antibiotic manufacturers in China. This is of considerable concern because persistent antibiotic residues may result in the development and spread of antibiotic-resistant bacteria, which could undermine life-saving antibiotic therapies (Keen and Montforts, 2011; Zhu et al., 2013). In the present study, the kitasamycin manufacturing waste was mixed with dairy manure and sawdust at different ratios prior to composting. The objectives of this study were (1) to investigate the
2. Materials and methods 2.1. Experimental set-up and sample collection The sawdust and dairy manure were obtained from a livestock farm and sawmill in Lin-an City, China, respectively. The size of the sawdust ranged from 6 mm to 8 mm. The dairy manure was sourced from a non-antibiotic-dependent dairy farming and was scraped from concrete pens and stored in a covered area prior to use. The kitasamycin manufacturing waste was collected from the residual organic medium obtained after bacterial fermentation, solideliquid separation and drying in Laiyi Bio-Technology Co, Xinchang City, China. Selected characteristics of the three materials used are presented in Table 1. Composting experiments were performed in March of 2011, in a compost site with rectangle-shaped bays (50 m 5 m 1.2 m, length width height) in Lin-an Jinda Biological Science and Technology Co. Ltd, China (Fig. 1S). This composting mode has been generally applied in China. Similar experimental design can be found in a previous work (Liu et al., 2011). The kitasamycin manufacturing waste was mixed with dairy manure and sawdust were mixed at different ratios (by volume): M, 10/80/20; M2, 10/70/ 20; and M3, 30/50/20 (Figs. 1S and 2S). Water was added to each mixture, and the mixtures were mixed with a front-end loader to achieve moisture content of approximately 65% and were then subdivided the mixtures into different parts of the bays. The total volume of each mixture was 25 m3 (5 m 5 m 1 m, length width height), and the total weights of M1, M2, and M3 were estimated to be 16.0 t, 16.2 t, and 16.6 t, respectively. The piles were aerated using a turning machine twice a week at 4:00 pm. The ambient temperature and mixture temperature in the centre of the composting material were recorded using a temperature meter every day at 9:00 am. The composting was performed for 40 days, and samples were taken at 0, 5, 10, 15, 20, 25, 30, 35, and 40 days at 9:00 am. Three replicates were collected from each mixture. To ensure representative sampling, three longitudinal sections (from 30, 50 and 70 cm from the base of mixture, respectively) were randomly dug, and the samples drawn from different sections of the same pile were mixed. A total of approximately 2 kg of samples were obtained using this sampling method. The collected samples were divided into two equal parts: one part was preserved at 4 C for microbiological analysis, and the other part was air-dried, ground and sieved for analysis after air-drying. 2.2. Chemical analyses
Table 1 Selected characteristics of the three composting materials. Characteristics
Dairy manure Sawdust Kitasamycin waste
pH EC (dS cm1) Total N (%) Total C (%) Total P (g kg1) Total K (g kg1) Bulk density (kg m3) Moisture content (%) Extractable kitasamycin (mg g1) Total Cu (mg kg1) Total Zn (mg kg1)
8.61 0.63 1.35 43.3 10.3 32.4 794 73% 0 425 936
7.95 0.86 0.43 55.2 0.64 44.1 540 13.4% 0 8.23 7.63
fate of kitasamycin during composting and to determine whether biological degradation contributes to the decline in extractable kitasamycin and (2) to evaluate the compost maturity and define the appropriate ratio of kitasamycin manufacturing waste to the dairy manure and sawdust in order to provide a feasible protocol for the utilisation of antibiotic manufacturing waste after composting.
4.43 3.98 6.19 38.8 6.32 2.62 276 18% 11.8 7.32 760
The pH and electrical conductivity (EC) measurements were performed on aqueous suspensions of the fresh compost samples (1:10, W/V, compost/water ratio) using a pH electrode (MP511, Shanghai, China) and a conductivity indicator (DDSJ-308A, Shanghai, China), respectively. The compost samples were analysed for organic carbon through oxidation with potassium dichromate and for the total N using the Kjeldahl method. The results are expressed per dry weight of material. To analyse the water-soluble fractions of the composting material, 15 g of fresh sample was extracted with 150 mL of distilled water (1:10 w/v ratio) by shaking for 24 h on a horizontal shaker at
N. Ding et al. / Journal of Environmental Management 134 (2014) 39e46
room temperature, according to the method previously described by Castaldi et al. (2008) with some modifications. The extracts were centrifuged (8000 rpm for 10 min at room temperature) and filtered through 0.45 mm filter membranes. The clear supernatant provided the different water-soluble fractions for subsequent anþ alyses. The water-soluble NHþ 4 (WS-NH4 ) and phosphorus (WS-P) concentrations were determined using the Nessler method and the ascorbic acid-molybdenum blue method. The water-soluble K (WSK) was analysed using an atomic absorption spectrophotometer (AAS-800; Perkin Elmer; USA). 2.3. Enumeration of total heterotrophic bacteria and kitasamycinresistant bacteria The total heterotrophic bacteria and kitasamycin-resistant bacteria were quantified through standard dilution plating according to the method previously described by Arikan et al. (2009), using non-selective Luria Bertani (LB) media and LB media containing 100 mg mL1 kitasamycin, respectively, after incubation at 37 C for 48 h. The results are expressed as colony-forming units (CFU) per dry weight of composts. All of the analyses were performed using duplicate samples. 2.4. PCR-DGGE analyses DNA was extracted directly from the compost samples using the FastDNAR SPIN kit for soils (MP Biomedicals, USA), as recommended by the manufacturer. The 16S rRNA genes were amplified by PCR using the following bacterial primers: 357F-GC (forward, 50 CGCCCGCCGCGCGCGGCGGGCGGGGCGGGGGCACGGGGGGCCTACGGGAGGCAGCAG-30 ), and 517r (reverse, 50 0 ATTACCGCGGCTGCTGG-3 ). The 50-mL reaction solution for PCR contains 5 mL of 10 PCR Taq buffer, 2 mL of 25 mM Mg2þ, 4 mL of 2.5 mM dNTP mixture, 1 mL of each primer, 1 mL of 5 U/mL Taq DNA polymerase and 2.5 ml of DNA template, and double-distilled water. The PCR was performed using a Thermal Cycler device (Eppendorf, Germany) under the following conditions for amplification: initial denaturation at 95 C for 3 min, 25 cycles at 94 C for 45 s, annealing at 59 C for 50 s, and extension at 72 C for 1 min, and a final extension at 72 C for 10 min. The PCR without the addition of the DNA template was used as the negative control. No target bands were found in the negative control. The DGGE analysis was performed using the D-Code universal mutation detection system (Bio-Rad Laboratories, USA) according to the manufacturer’s instructions. For the DGGE analysis, 40 mL of the PCR product was used. The conditions for separation were as follows: 110 V for 10 h at 60 C in a 10% polyacrylamide gel with the denaturing gradient ranging from 40% to 55%. The gel was then stained with SYBR Green I (10000-fold diluted in 1 TAE) for 30 min and imaged under UV light. 2.5. Analyses of extractable kitasamycin The compost samples were extracted in duplicate for kitasamycin analyses using a 0.1 M HCl:methanol (50:50) solution, as was previously described (Kim et al., 2012). The fresh compost sample (10 g) were mixed with the extractant (30 mL) in a 50 mL centrifuge tube, sonicated for 10 min and then centrifuged at 4000 r min1 for 10 min, and the supernatant was then filtered 0.45-mm mixed cellulose ester disposable filters (S-Pak, Shanghai Sunyear Co., Ltd., China). The kitasamycin concentration of the filtered extract was analysed through HPLC (Agilent Technologies, USA) using a UVD detector and an Agilent LC-18 (4.6 250 mm) column with a 5-mM particle size. The analysis was performed using 100 mmol L1
41
ammonium acetate, methanol and acetonitrile (40:55:5) as the mobile phase with a column temperature at 65 C and a flow rate of 1.0 mL min1. The detection was performed at 231 nm (Zhu and Niu, 2007). 2.6. Germination index (GI) The seed germination and root elongation measurements of cress (Lepidium sativum L.) were performed using different water extracts by mechanically shaking the fresh samples for 1 h at a solid to double-distilled water ratio of 1:10 (w/v, dry weight basis). Ten cress seeds were sown in each compost and peat mixture in a 10 mm 10 mm Petri dish. Each treatment was performed in triplicate. Approximately 0.5 mL of the water extracts was added to each seed. The seeds were incubated at 25 C in the dark. After 72 h, the number of germinated seeds was quantified and the root length was measured. The GI was determined using the ratio (%) of the seed germination (%) times the root length between treatments and the control (Zucconi et al., 1981). 2.7. Statistical analysis The data are reported as the means of three replicates and were analysed through a one-way ANOVA design using the SPSS 11.0 software. The DGGE band profiles were digitised, and the band numbers were quantified using Quantity One software (version 4.5, Bio-Rad laboratories, USA). 3. Results and discussion 3.1. Temperature According to a previous study, a compost material with a relatively low C/N ratio can produce qualified products (Zhu, 2007); thus, we generated two mixtures with an initial C/N ratios of 27.4 (M2) and 16.9 (M3, Table 2) to determine the efficiency of kitasamycin waste. We also included a non-antibiotic-mixture (M1) as a control, to assess the effect of the antibiotic residue on the composting process. Temperature is a key parameter that indicating the rate and extent of the composting process. As shown in Fig. 1, the composting process of the three mixtures can be divided into three phases: (1) mesophilic phase, (2) thermophilic phase, and (3) cooling phase. For M1, the maximum temperature reached was 46 C on day 13, and the change in temperature was smoother than those obtained with M2 and M3 during the entire composting process. The lower temperature pattern observed with M1 was due to its low N content (Table 1), which could not efficiently support the growth of microorganisms efficiently. Throughout the thermophilic phase, the temperature in M2 was always higher than that obtained with M3, with a maximum temperature of 67 C compared with 62 C in M3. The thermophilic phase of M2 continued over a period of 10 days; whereas this phase in M3 lasted for only 8 days. This difference may be due to the scarcity of available carbon sources at the beginning of the composting process of M3 compared with those available in M2, because M3 does not provide a favourable conditions for the growth and biological activity of microorganisms (Huang et al., 2004). Furthermore, the high level of kitasamycin in M3 may also contribute to the disturbance of the composting process through the inhibition of microbial activities. As shown in Fig. 3, the total count of heterotrophic bacterial of M3 was lower than that of M2 on day 15. In addition, the much higher WS-NHþ 4 released in the M3 than M2 might be also toxic to the microorganisms (Fig. 5A).
42
N. Ding et al. / Journal of Environmental Management 134 (2014) 39e46
Table 2 Changes in the organic carbon, total N and C:N ratio during the composting of different piles. The results are the means of three replicates. Day
N%
C%
M1 0 5 10 15 20 25 30 35 40
1.17 0.97 1.12 1.06 1.14 1.18 1.15 1.23 1.25
M2
0.07 0.09 0.03 0.06 0.02 0.04 0.04 0.05 0.06
1.65 1.52 1.44 1.48 1.54 1.57 1.42 1.53 1.48
M3
0.07 0.07 0.05 0.11 0.18 0.12 0.03 0.15 0.08
2.62 2.38 2.04 2.12 1.89 1.78 1.88 1.74 1.78
C/N
M1
0.14 0.08 0.14 0.24 0.05 0.06 0.03 0.13 0.05
45.7 37.1 41.8 38.5 37.7 36.9 37.3 34.6 37.1
M2
1.2 1.0 2.4 2.0 3.1 2.4 1.6 0.9 1.3
M3
45.2 40.5 39.1 37.5 35.2 33.2 31.4 30.3 28.4
1.3 1.0 1.6 2.4 1.5 0.9 1.1 1.2 1.7
44.3 38.1 37.2 35.2 34.4 33.7 32.5 31.2 30.1
1.4 1.0 2.1 0.8 1.3 2.1 1.6 0.1 1.0
M1
M2
M3
39.2 38.2 36.4 36.3 33.1 31.3 32.4 27.9 29.7
27.4 26.6 25.4 25.3 22.9 21.1 22.1 19.8 19.2
16.9 16.0 18.2 16.6 18.2 18.9 17.3 17.9 16.9
3.2. The fate and effect of kitasamycin on the bacterial community during composting
analysis or infrared spectroscopy, is needed to elucidate the adsorption process during composting.
3.2.1. Extractable kitasamycin The initial extractable kitasamycin concentrations in M2 and M3 were detected to be 790 and 2334 mg kg1, respectively (Fig. 2), which are not in the specified ratios of 10% and 30% because the total extractable concentration in the kitasamycin waste was 11800 mg kg1 (Table 1). As a type of macrolides antibiotic, kitasamycin is substituted with many hydroxyl, alkyl and ketone groups. The excess decline in the extractable kitasamycin may due to the adsorption or formation of chelate-complexes with the divalent cations that are present in the manure and sawdust (Kim et al., 2012). After composting, the extractable kitasamycin concentrations in M2 and M3 were markedly decreased with half-life values of 4 and 8 days, respectively. The observed decrease in kitasamycin might be attributed to the adsorption to organic substances, such as humic acid, produced from the decomposition of organic matter at elevated temperatures (Gu et al., 2007; Hartlieb et al., 2003), as has been previously shown in medicated manure mixtures with oxytetracycline (Arikan et al., 2009) and in soilchicken faeces mixtures with chlortetracycline (Gavalchin and Katz, 1994). Furthermore, the difference in the half-life values suggests that the decline of the concentration of kitasamycin in the mesophilic phase is due to a non-linear adsorption. After the fast adsorption phase (0e5 days), the extractable kitasamycin in both M2 and M3 decreased gradually during the thermophilic phase. In addition, the concentration of kitasamycin in M2 decreased to below the detection limit (1.0 mg kg1 DW) within 15 days. However, for M3, kitasamycin persisted until the end of composting (40 days), with a final value of 121 mg kg1. The result showed that the ratio of 30% kitasamycin waste in M3 was not optional, resulting in an excess of extractable kitasamycin remaining in the composting system. The results from this study are not enough to fully explain mechanism of the abiotic process related to kitasamycin decline. Further research, such as use of electron spectroscopy for chemical
3.2.2. Total heterotrophic bacteria and kitasamycin-resistant bacteria The total count of heterotrophic bacteria in the three mixtures increased in the thermophilic phase and deceased in the cooling phase (Fig. 3). The rate of increase in M2 was higher than those obtained for the other two mixtures. The limit in the N content of M1 may explain the slow bacterial growth, which is consistent with its low composting temperature during this phase (Fig. 1). However, M3 contained a higher level of extractable kitasamycin residues (Fig. 2), which may restrict the growth of the bacterial population and cause a disturbance on the composting process, particularly during the thermophilic phase (Fig. 1). Although a previous study reported that the addition of antibiotics elicits only a transient perturbation (Selvam et al., 2012), the much higher level of antibiotics in M3 might exert a stronger inhibitory effect on the bacterial growth. During the cooling stage, the bacterial population in M2 decreased gradually, but the number of bacteria in M3 remained high, and the final number was one order of magnitude larger than that obtained for M2, suggesting that M3 is not mature and that there many degradable compounds are available at the end of the composting period for bacterial consumption. The populations of kitasamycin-resistant bacteria in M2 and M3 were markedly higher than those found in M1, indicating that a large amount of resistant bacteria had developed from the kitasamycin waste (Fig. 3). Similar results were also observed in a composting study of medicated manure (Arikan et al., 2007). Furthermore, a type of kitasamycin-resistant bacteria was isolated in the initial compost of M2, which was a rod-shaped, endospore-
Temperature (°C)
60
M1
M2
M3
Environment
50 40 30 20
M2
-1
Extractable kitasamycin (ng g )
70
M1
3000 2500
M3
2000 1500 1000 500 0
10
0
0 0
2
4
6
8
10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40
5
10
15
20
25
30
35
40
Days
Days
Fig. 1. Change of temperature during the composting process of the three composting mixtures and the environment.
Fig. 2. Changes of extractable kitasamycin concentrations in different mixtures during composting. The results are the means of three replicates, and the bars indicate the standard error of the three replicates.
Bacterial population (log (CFU g -1 Fw))
N. Ding et al. / Journal of Environmental Management 134 (2014) 39e46
12
M1-T
M2-T
M3-T
M1-R
M2-R
M3-R
10 8 6 4 2 0 0
5
10
15
20
25
30
35
40
Days Fig. 3. Changes in the total heterotrophic bacteria (M1-T, M2-T and M3-T, respectively) and kitasamycin-resistant bacteria (M1-R, M2-R and M3-R, respectively) during composting. The results are the mean of three replicates, and the bars indicate the standard error of three replicates.
forming, Gram-positive bacterium (Fig. 3S). The bacterium was designed as strain D3. Bacterial 16S rDNA was amplified by using the universal primer pair BSF8 and BSR1541. The length of the sequence is 1430. A UPGMA tree was constructed based on sequence and related ones from the Genebank database. Strain D3 showed the closest match with Paenibacillus sp. (>99% similarity, Fig. 4S). As the composting proceeded, the population of resistant bacteria in M1 rapidly decreased to a level below the detection limit (1.5 log CFU g1 fw) within 10 days. However, for M2, this value gradually declined over a period of 20 days from 7.25 to 3.80 log (CFU g1 fw) and could not be detected after 25 days. The synchronous decrease in extractable kitasamycin (Fig. 2) and kitasamycin-resistant bacteria (Fig. 3) suggest that the kitasamycinresistant bacteria are closely associated with the extractable
43
kitasamycin, and that the non-extractable kitasamycin residues are not bioactive. The biodegradation of kitasamycin in M3 through day 40, as reflected by the level of resistant bacteria, i.e., 3.86 log (CFU g1 fw). Compared with the rapid adsorption of kitasamycin to the matrix during the mesophilic phase of composting, the microbial degradation effect was more modest albeit more persistent. 3.2.3. DGGE profiles of the bacterial community PCR-DGGE was performed to assess the effect of kitasamycin waste on the bacterial community composition during the composting process (Fig. 4). On day 0, there were 25, 28 and 36 bands in DGGE gel for M1, M2 and M3, respectively. The increased numbers of M2 and M3 bands indicates that many specific bacterial species from the kitasamycin waste were introduced into the composts. Dynamic changes in the bacterial communities were found during composting, as was observed by the strong shifts in the DGGE profiles between day 0 and day 40. For M3, the band number decreased from 36 to 30, and this decrease may be due to the high level of extractable kitasamycin exerting an adverse effect on the bacterial diversity. Similar results were obtained by Zeng et al. (2011), who reported that pentachlorophenol stress has an inhibitory effect on microbial abundance. Moreover, the band number found for the M2 samples increased to 57. Although the total population of bacteria in M2 at the end of the composting period was decreased by 2 orders of magnitude compared with the number of bacteria present during the thermophilic phase (Fig. 3), the enhanced band number (Fig. 4) clearly shows that the taxonomic and metabolic diversities increased. The bacterial species that increased in abundance were not only simply organic oxidisers but also involved in hydrogen-, ammonium-, and nitrite-oxidation, nitrogen-fixation, and sulphate-reduction (Diaz-Ravina et al., 1989; Beffa et al., 1996). For M1, there was a smaller increase in the DGGE bands (from 25 to 37) during composting, possible as a result of the low level of N sources, which caused competition among different bacterial species. 3.3. Assessment of compost maturity 3.3.1. Organic carbon, total N and C/N ratio The N% and C% contents decreased over time until the end of the composting period, with the exception of the N% in M1, which exhibited no obvious change over time (Table 1). This finding indicates that the organic compounds in M1 were not well decomposed compared with the other two mixtures. The initial C/N ratio of the mixtures ranged from 16.9 to 39.2, and decreases in the C/N ratios were found for M1 and M2. However, there was no obvious change in the C/N ratio of M3 over time, mainly because the amount of N-mineralization, which is comparable with that observed in C-decomposition.
Fig. 4. PCR-DGGE analysis and the number of DGGE bands of the bacterial community in different compost mixtures on day 0 and day 40.
3.3.2. Water-soluble nutrients For M1, the WS-NHþ 4 concentration was negligible throughout composting with the exception of Day 5 to Day 10, which indicates that there was not sufficiently available N for the assimilation of microorganisms (Fig. 5A). The WS-NHþ 4 concentrations in M2 and M3 sharply increased to similar levels during the thermophilic phase. As the process progressed, the WS-NHþ 4 concentrations quickly declined due to ammonia volatilisation and nitrification (Tiquia, 2005). For M2, the final level of WS-NHþ 4 was lower than the phototoxic limit value of 0.4 g kg1 (Bernal et al., 2009), indicating that it is a qualified product. However, the final level measured for M3 was still 2-fold higher than that obtained for M2, which suggests that the evolution of N forms was not completed. The evolution of WS-P in each treatment followed a similar trend, with a gradual decline during the first 20 days (from 8.71,
44
N. Ding et al. / Journal of Environmental Management 134 (2014) 39e46
25.0
M1
B
M2
M2
M3
0.8
M3
20.0
-1
WS-P (g kg )
-1
WS-NH 4 (g kg )
1.0
M1
A
+
15.0 10.0
0.6 0.4 0.2
5.0
0.0
0.0 0
5
10
15
20
25
30
35
0
40
5
10
20
25
30
35
40
Days
Days
2.0
15
M1
C
M2 M3
+
-1
WS-K (g kg )
1.5
1.0
0.5
0.0 0
5
10
15
20
25
30
35
40
Days Fig. 5. Changes of water soluble nutrients in different mixtures during composting. The results are the means of three replicates, and the bars indicate the standard errors of three replicates.
8.04 and 4.45 g kg1 to 5.43, 3.43 and 2.89 g kg1 for M1, M2 and M3, respectively, Fig. 5B) and a gradual recovery during the next 20 days. The initial WS-K concentrations in M1, M2, and M3 were 1.19, 1.08 and 0.84 g kg1, respectively (Fig. 5C). The WS-K in all of the treatments remained constant and slightly increased at the end of the composting period.
3.3.3. Electrical conductivity (EC) and pH The initial EC of M1, M2, and M3 increased with an increase in the proportion of kitasamycin waste. The EC values of M1, M2, and M3 were 2.14, 2.89 and 3.11 dS cm1, respectively (Fig. 6A). For M1, the EC value gradually decreased during the composting process and was 1.58 dS cm1 on day 40. However, for M2 and M3, the EC
11 6
A
M3
M3
9
pH
-1
4
EC ds m
M2
10
M2
5
M1
B
M1
3
8 7
2
6
1
5
0
4 0
5
10
15
20 Days
25
30
35
40
0
5
10
15
20
25
30
35
40
Days
Fig. 6. Changes in the electrical conductivity (EC, A) and pH (B) in different composting mixtures. The results are the means of three replicates, and the bars indicate the standard errors of the three replicates.
N. Ding et al. / Journal of Environmental Management 134 (2014) 39e46
Furthermore, to ensure the reduction of Escherichia coli reduction during composting, we analyzed the number of E. coli in the initial and final compost of each treatment. Results showed that there was a large amount of E. coli in the initial mixture of each treatment, which was more than 6.0 log CFU g1 (Table 1S). After 40-day composting, the numbers of E. coli both in M2 and M3 decreased to under the detection limit (1.5 log CFU g1). In comparison, a high number of E. coli (3.7 log CFU g1) still remained in the final compost of M1, which was mainly because the temperature of M1 did not meet the thermal death point (55 C) during the thermophilic phase (Hanajima et al., 2006).
M1
100%
M2 M3
GI (%)
80%
45
60% 40% 20%
4. Conclusions
0% 0
5
10
15
20
25 30
35
40
Days Fig. 7. Changes in germination index (GI %) in different mixtures during composting. Results are the mean of three replicates, and bars indicate standard error of three replicates.
values increased and reached maximum values of 3.55 and 4.46 dS cm1 during the thermophilic phase, respectively, and then gradually decreased during the cooling phase. The final EC value for M2 did not exceed the limit content of 3.0 dS cm1 (Soumaré et al., 2002), indicating that the EC does not adversely affect plant growth. Optimal pH values between 5.5 and 8.0 can support efficient microbial activities during composting (Bernal et al., 2009). In general, pH is not a key factor in composting because most compost materials exhibit pH values within this pH range. However, this factor was very relevant in the present study because the cattle manure used had a high pH of >9.0 and kitasamycin waste had a low pH of
Contents lists available at ScienceDirect
Journal of Environmental Management journal homepage: www.elsevier.com/locate/jenvman
Decline in extractable kitasamycin during the composting of kitasamycin manufacturing waste with dairy manure and sawdust Nengfei Ding a, Weidong Li b, Chen Liu a, Qinglin Fu a, Bin Guo a, *, Hua Li a, Ningyu Li a, Yicheng Lin a a b
Institute of Environment, Resource, Soil and Fertilizer, Zhejiang Academy of Agricultural Sciences, Hangzhou, China Department of Science, Qianjiang College of Hangzhou Normal University, Hangzhou 310012, China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 11 June 2013 Received in revised form 18 December 2013 Accepted 22 December 2013 Available online 23 January 2014
The aim of this study was to propose a feasible treatment of kitasamycin manufacturing waste by examining extractable kitasamycin and evaluating its compost maturity during the composting of waste with different ratios of dairy manure and sawdust over a 40-day period (volume/volume/volume; M1, 0/ 80/20; M2, 10/70/20; and M3, 30/50/20). During composting, the concentration of extractable kitasamycin in kitasamycin-contaminated composts declined rapidly, and was undetectable in M2 within 15 days. M2 also achieved the highest fertility compost, which was characterised by the following final parameters: electrical conductivity, 2.34 dS cm1; pH, 8.15; total C/N, 22.2; water-soluble NHþ 4 , P, and K, 0.37, 3.43, and 1.05 g kg1, respectively; and plant germination index values, 92%. Furthermore, DGGE analysis showed a dramatic increase in the diversity of bacterial species during composting. In contrast, a high concentration (121 mg kg1) of extractable kitasamycin still remained in the M3 compost, which exerted an inhibitory effect on the composting, resulting in reduced bacterial diversity, high values of electrical conductivity and water-soluble NHþ 4 , a low C/N ratio, and a low plant germination index value. Furthermore, 3.86 log (CFU g1) kitasamycin-resistant bacteria were still present on day 40, indicating the biological degradation contributed to the decline of extractable kitasamycin. Ó 2014 Elsevier Ltd. All rights reserved.
Keywords: Composting DGGE Kitasamycin waste Microbial community Water-soluble nutrients
1. Introduction The treatment of antibiotic manufacturing waste, which is a residue of the organic medium used for microbial fermentation during antibiotic production, has become a difficult problem in China. Due to the high levels of antibiotic residuals, this biowaste has been utilised as an additive to animal feed to promote growth and to prevent pathogenic bacterial infection. However, incomplete assimilation in animals has resulted in 70e90% of the antibiotic being excreted in manure (Phillips et al., 2004; Kumar et al., 2005); the manure is not intensively treated and is often applied directly to agricultural fields, which contributes to the increase in antibiotics and antibiotic-resistance genes in the environment. The spread of antibiotic residues in the environment poses a potential risk to aquatic and terrestrial organisms (Hamscher et al., 2000). In addition to reducing the diversity of micro-organisms in the environment (Zielezny et al., 2006), this phenomenon can also enhance
* Corresponding author. Tel./fax: þ86 571 8640 4041. E-mail addresses: [email protected], [email protected] (B. Guo). 0301-4797/$ e see front matter Ó 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jenvman.2013.12.030
human antimicrobial resistance due to the consumption of crops grown in the antibiotic-contaminated soils (Keen and Montforts, 2011). Thus, appropriate techniques are needed introduced to reduce the antibiotic level of manufacturing waste prior to agricultural utilisation. Composting is an effective methodology for the decomposing of organic wastes via the biological degradation of organic constituents under controlled conditions, and has been successfully applied for the treatment of persistent organic pollutants (Laine and Jorgensen, 1997; Zeng et al., 2011). Recently, additional information regarding antibiotic degradation during manure composting has become available. A study on chlortetracycline in aged and spiked poultry manure demonstrated that more than 90% of the initial level of chlortetracycline was depleted under aerobic composting (Bao et al., 2009). Both full-scale and lab-scale investigations have indicated that decreases in tetracycline and sulphonamide concentrations are highly dependent on the presence of sawdust during composting (Kim et al., 2012). Dolliver et al. (2008) suggested that stockpiling may be a practical and economical option for livestock producers to reduce antibiotic levels in manure.
40
N. Ding et al. / Journal of Environmental Management 134 (2014) 39e46
As a by-product of microbial fermentation, antibiotic manufacturing waste cannot be composted alone due to its high nitrogen content (Table 1), which could enhance the loss of N via volatilisation and phototoxic molecules may simultaneously build up due to insufficient biodegradation of the organic matter (Bernal et al., 2009). Normally, high-nitrogen wastes are mixed with adjusted by the carbon-rich materials prior to composting to obtain an initial C/N ratio of between 25 and 35 (Bernal et al., 2009); however, these values are not absolute. An economic analysis showed that composting with a low initial C/N ratio (20) is beneficial to efficiently dispose of a high amount of N-rich waste (swine manure) (Zhu, 2007). Furthermore, to achieve high grade composting, other factors that affect the composting process, such as pH, electrical conductivity (EC), and the available nutrients, should also be well balanced. However, the formulation of composting mix between the antibiotic manufacturing waste and other organic materials is currently unavailable. Another important issue is the effect and fate of the antibiotics present in the manufacturing waste throughout the composting process. Previous studies have shown that antibiotic residues in manures did not have a significant effect on the composting process because the concentrations of extractable antibiotics decreased rapidly in during the first 3e10 days of composting (Arikan et al., 2007; Dolliver et al., 2008; Bao et al., 2009; Kim et al., 2012). However, antibiotic manufacturing wastes contain much higher extractable antibiotic (>10 fold) than those reported in previous manure studies (Table 1; Dolliver et al., 2008; Bao et al., 2009; Arikan et al., 2006, 2007; 2009; Kim et al., 2012). At present, the effect of high concentrations of extractable antibiotics in antibiotic waste on microbial activities during composting is poorly understood. A comparison of the bacteria composition between the antibiotic-free and antibiotic-contaminated composts during the process of composting may contribute to a better understanding of the antibiotic degradation process. This process is associated with the succession of microbial communities during composting. Advanced molecular biological techniques, such as polymerase chain reactionedenaturing gradient gel electrophoresis (PCRDGGE), have proven to be useful for the detection of this succession (Cahyani et al., 2003; Nakasaki et al., 2009). Kitasamycin is one of the antibiotics that is highly active against a wide range of Gram-positive bacteria and has been widely used in the rearing of food-producing animals to prevent and treat diseases (Jordan and Knight, 1984). It has been estimated that 6300 tons of kitasamycin waste are annually released from antibiotic manufacturers in China. This is of considerable concern because persistent antibiotic residues may result in the development and spread of antibiotic-resistant bacteria, which could undermine life-saving antibiotic therapies (Keen and Montforts, 2011; Zhu et al., 2013). In the present study, the kitasamycin manufacturing waste was mixed with dairy manure and sawdust at different ratios prior to composting. The objectives of this study were (1) to investigate the
2. Materials and methods 2.1. Experimental set-up and sample collection The sawdust and dairy manure were obtained from a livestock farm and sawmill in Lin-an City, China, respectively. The size of the sawdust ranged from 6 mm to 8 mm. The dairy manure was sourced from a non-antibiotic-dependent dairy farming and was scraped from concrete pens and stored in a covered area prior to use. The kitasamycin manufacturing waste was collected from the residual organic medium obtained after bacterial fermentation, solideliquid separation and drying in Laiyi Bio-Technology Co, Xinchang City, China. Selected characteristics of the three materials used are presented in Table 1. Composting experiments were performed in March of 2011, in a compost site with rectangle-shaped bays (50 m 5 m 1.2 m, length width height) in Lin-an Jinda Biological Science and Technology Co. Ltd, China (Fig. 1S). This composting mode has been generally applied in China. Similar experimental design can be found in a previous work (Liu et al., 2011). The kitasamycin manufacturing waste was mixed with dairy manure and sawdust were mixed at different ratios (by volume): M, 10/80/20; M2, 10/70/ 20; and M3, 30/50/20 (Figs. 1S and 2S). Water was added to each mixture, and the mixtures were mixed with a front-end loader to achieve moisture content of approximately 65% and were then subdivided the mixtures into different parts of the bays. The total volume of each mixture was 25 m3 (5 m 5 m 1 m, length width height), and the total weights of M1, M2, and M3 were estimated to be 16.0 t, 16.2 t, and 16.6 t, respectively. The piles were aerated using a turning machine twice a week at 4:00 pm. The ambient temperature and mixture temperature in the centre of the composting material were recorded using a temperature meter every day at 9:00 am. The composting was performed for 40 days, and samples were taken at 0, 5, 10, 15, 20, 25, 30, 35, and 40 days at 9:00 am. Three replicates were collected from each mixture. To ensure representative sampling, three longitudinal sections (from 30, 50 and 70 cm from the base of mixture, respectively) were randomly dug, and the samples drawn from different sections of the same pile were mixed. A total of approximately 2 kg of samples were obtained using this sampling method. The collected samples were divided into two equal parts: one part was preserved at 4 C for microbiological analysis, and the other part was air-dried, ground and sieved for analysis after air-drying. 2.2. Chemical analyses
Table 1 Selected characteristics of the three composting materials. Characteristics
Dairy manure Sawdust Kitasamycin waste
pH EC (dS cm1) Total N (%) Total C (%) Total P (g kg1) Total K (g kg1) Bulk density (kg m3) Moisture content (%) Extractable kitasamycin (mg g1) Total Cu (mg kg1) Total Zn (mg kg1)
8.61 0.63 1.35 43.3 10.3 32.4 794 73% 0 425 936
7.95 0.86 0.43 55.2 0.64 44.1 540 13.4% 0 8.23 7.63
fate of kitasamycin during composting and to determine whether biological degradation contributes to the decline in extractable kitasamycin and (2) to evaluate the compost maturity and define the appropriate ratio of kitasamycin manufacturing waste to the dairy manure and sawdust in order to provide a feasible protocol for the utilisation of antibiotic manufacturing waste after composting.
4.43 3.98 6.19 38.8 6.32 2.62 276 18% 11.8 7.32 760
The pH and electrical conductivity (EC) measurements were performed on aqueous suspensions of the fresh compost samples (1:10, W/V, compost/water ratio) using a pH electrode (MP511, Shanghai, China) and a conductivity indicator (DDSJ-308A, Shanghai, China), respectively. The compost samples were analysed for organic carbon through oxidation with potassium dichromate and for the total N using the Kjeldahl method. The results are expressed per dry weight of material. To analyse the water-soluble fractions of the composting material, 15 g of fresh sample was extracted with 150 mL of distilled water (1:10 w/v ratio) by shaking for 24 h on a horizontal shaker at
N. Ding et al. / Journal of Environmental Management 134 (2014) 39e46
room temperature, according to the method previously described by Castaldi et al. (2008) with some modifications. The extracts were centrifuged (8000 rpm for 10 min at room temperature) and filtered through 0.45 mm filter membranes. The clear supernatant provided the different water-soluble fractions for subsequent anþ alyses. The water-soluble NHþ 4 (WS-NH4 ) and phosphorus (WS-P) concentrations were determined using the Nessler method and the ascorbic acid-molybdenum blue method. The water-soluble K (WSK) was analysed using an atomic absorption spectrophotometer (AAS-800; Perkin Elmer; USA). 2.3. Enumeration of total heterotrophic bacteria and kitasamycinresistant bacteria The total heterotrophic bacteria and kitasamycin-resistant bacteria were quantified through standard dilution plating according to the method previously described by Arikan et al. (2009), using non-selective Luria Bertani (LB) media and LB media containing 100 mg mL1 kitasamycin, respectively, after incubation at 37 C for 48 h. The results are expressed as colony-forming units (CFU) per dry weight of composts. All of the analyses were performed using duplicate samples. 2.4. PCR-DGGE analyses DNA was extracted directly from the compost samples using the FastDNAR SPIN kit for soils (MP Biomedicals, USA), as recommended by the manufacturer. The 16S rRNA genes were amplified by PCR using the following bacterial primers: 357F-GC (forward, 50 CGCCCGCCGCGCGCGGCGGGCGGGGCGGGGGCACGGGGGGCCTACGGGAGGCAGCAG-30 ), and 517r (reverse, 50 0 ATTACCGCGGCTGCTGG-3 ). The 50-mL reaction solution for PCR contains 5 mL of 10 PCR Taq buffer, 2 mL of 25 mM Mg2þ, 4 mL of 2.5 mM dNTP mixture, 1 mL of each primer, 1 mL of 5 U/mL Taq DNA polymerase and 2.5 ml of DNA template, and double-distilled water. The PCR was performed using a Thermal Cycler device (Eppendorf, Germany) under the following conditions for amplification: initial denaturation at 95 C for 3 min, 25 cycles at 94 C for 45 s, annealing at 59 C for 50 s, and extension at 72 C for 1 min, and a final extension at 72 C for 10 min. The PCR without the addition of the DNA template was used as the negative control. No target bands were found in the negative control. The DGGE analysis was performed using the D-Code universal mutation detection system (Bio-Rad Laboratories, USA) according to the manufacturer’s instructions. For the DGGE analysis, 40 mL of the PCR product was used. The conditions for separation were as follows: 110 V for 10 h at 60 C in a 10% polyacrylamide gel with the denaturing gradient ranging from 40% to 55%. The gel was then stained with SYBR Green I (10000-fold diluted in 1 TAE) for 30 min and imaged under UV light. 2.5. Analyses of extractable kitasamycin The compost samples were extracted in duplicate for kitasamycin analyses using a 0.1 M HCl:methanol (50:50) solution, as was previously described (Kim et al., 2012). The fresh compost sample (10 g) were mixed with the extractant (30 mL) in a 50 mL centrifuge tube, sonicated for 10 min and then centrifuged at 4000 r min1 for 10 min, and the supernatant was then filtered 0.45-mm mixed cellulose ester disposable filters (S-Pak, Shanghai Sunyear Co., Ltd., China). The kitasamycin concentration of the filtered extract was analysed through HPLC (Agilent Technologies, USA) using a UVD detector and an Agilent LC-18 (4.6 250 mm) column with a 5-mM particle size. The analysis was performed using 100 mmol L1
41
ammonium acetate, methanol and acetonitrile (40:55:5) as the mobile phase with a column temperature at 65 C and a flow rate of 1.0 mL min1. The detection was performed at 231 nm (Zhu and Niu, 2007). 2.6. Germination index (GI) The seed germination and root elongation measurements of cress (Lepidium sativum L.) were performed using different water extracts by mechanically shaking the fresh samples for 1 h at a solid to double-distilled water ratio of 1:10 (w/v, dry weight basis). Ten cress seeds were sown in each compost and peat mixture in a 10 mm 10 mm Petri dish. Each treatment was performed in triplicate. Approximately 0.5 mL of the water extracts was added to each seed. The seeds were incubated at 25 C in the dark. After 72 h, the number of germinated seeds was quantified and the root length was measured. The GI was determined using the ratio (%) of the seed germination (%) times the root length between treatments and the control (Zucconi et al., 1981). 2.7. Statistical analysis The data are reported as the means of three replicates and were analysed through a one-way ANOVA design using the SPSS 11.0 software. The DGGE band profiles were digitised, and the band numbers were quantified using Quantity One software (version 4.5, Bio-Rad laboratories, USA). 3. Results and discussion 3.1. Temperature According to a previous study, a compost material with a relatively low C/N ratio can produce qualified products (Zhu, 2007); thus, we generated two mixtures with an initial C/N ratios of 27.4 (M2) and 16.9 (M3, Table 2) to determine the efficiency of kitasamycin waste. We also included a non-antibiotic-mixture (M1) as a control, to assess the effect of the antibiotic residue on the composting process. Temperature is a key parameter that indicating the rate and extent of the composting process. As shown in Fig. 1, the composting process of the three mixtures can be divided into three phases: (1) mesophilic phase, (2) thermophilic phase, and (3) cooling phase. For M1, the maximum temperature reached was 46 C on day 13, and the change in temperature was smoother than those obtained with M2 and M3 during the entire composting process. The lower temperature pattern observed with M1 was due to its low N content (Table 1), which could not efficiently support the growth of microorganisms efficiently. Throughout the thermophilic phase, the temperature in M2 was always higher than that obtained with M3, with a maximum temperature of 67 C compared with 62 C in M3. The thermophilic phase of M2 continued over a period of 10 days; whereas this phase in M3 lasted for only 8 days. This difference may be due to the scarcity of available carbon sources at the beginning of the composting process of M3 compared with those available in M2, because M3 does not provide a favourable conditions for the growth and biological activity of microorganisms (Huang et al., 2004). Furthermore, the high level of kitasamycin in M3 may also contribute to the disturbance of the composting process through the inhibition of microbial activities. As shown in Fig. 3, the total count of heterotrophic bacterial of M3 was lower than that of M2 on day 15. In addition, the much higher WS-NHþ 4 released in the M3 than M2 might be also toxic to the microorganisms (Fig. 5A).
42
N. Ding et al. / Journal of Environmental Management 134 (2014) 39e46
Table 2 Changes in the organic carbon, total N and C:N ratio during the composting of different piles. The results are the means of three replicates. Day
N%
C%
M1 0 5 10 15 20 25 30 35 40
1.17 0.97 1.12 1.06 1.14 1.18 1.15 1.23 1.25
M2
0.07 0.09 0.03 0.06 0.02 0.04 0.04 0.05 0.06
1.65 1.52 1.44 1.48 1.54 1.57 1.42 1.53 1.48
M3
0.07 0.07 0.05 0.11 0.18 0.12 0.03 0.15 0.08
2.62 2.38 2.04 2.12 1.89 1.78 1.88 1.74 1.78
C/N
M1
0.14 0.08 0.14 0.24 0.05 0.06 0.03 0.13 0.05
45.7 37.1 41.8 38.5 37.7 36.9 37.3 34.6 37.1
M2
1.2 1.0 2.4 2.0 3.1 2.4 1.6 0.9 1.3
M3
45.2 40.5 39.1 37.5 35.2 33.2 31.4 30.3 28.4
1.3 1.0 1.6 2.4 1.5 0.9 1.1 1.2 1.7
44.3 38.1 37.2 35.2 34.4 33.7 32.5 31.2 30.1
1.4 1.0 2.1 0.8 1.3 2.1 1.6 0.1 1.0
M1
M2
M3
39.2 38.2 36.4 36.3 33.1 31.3 32.4 27.9 29.7
27.4 26.6 25.4 25.3 22.9 21.1 22.1 19.8 19.2
16.9 16.0 18.2 16.6 18.2 18.9 17.3 17.9 16.9
3.2. The fate and effect of kitasamycin on the bacterial community during composting
analysis or infrared spectroscopy, is needed to elucidate the adsorption process during composting.
3.2.1. Extractable kitasamycin The initial extractable kitasamycin concentrations in M2 and M3 were detected to be 790 and 2334 mg kg1, respectively (Fig. 2), which are not in the specified ratios of 10% and 30% because the total extractable concentration in the kitasamycin waste was 11800 mg kg1 (Table 1). As a type of macrolides antibiotic, kitasamycin is substituted with many hydroxyl, alkyl and ketone groups. The excess decline in the extractable kitasamycin may due to the adsorption or formation of chelate-complexes with the divalent cations that are present in the manure and sawdust (Kim et al., 2012). After composting, the extractable kitasamycin concentrations in M2 and M3 were markedly decreased with half-life values of 4 and 8 days, respectively. The observed decrease in kitasamycin might be attributed to the adsorption to organic substances, such as humic acid, produced from the decomposition of organic matter at elevated temperatures (Gu et al., 2007; Hartlieb et al., 2003), as has been previously shown in medicated manure mixtures with oxytetracycline (Arikan et al., 2009) and in soilchicken faeces mixtures with chlortetracycline (Gavalchin and Katz, 1994). Furthermore, the difference in the half-life values suggests that the decline of the concentration of kitasamycin in the mesophilic phase is due to a non-linear adsorption. After the fast adsorption phase (0e5 days), the extractable kitasamycin in both M2 and M3 decreased gradually during the thermophilic phase. In addition, the concentration of kitasamycin in M2 decreased to below the detection limit (1.0 mg kg1 DW) within 15 days. However, for M3, kitasamycin persisted until the end of composting (40 days), with a final value of 121 mg kg1. The result showed that the ratio of 30% kitasamycin waste in M3 was not optional, resulting in an excess of extractable kitasamycin remaining in the composting system. The results from this study are not enough to fully explain mechanism of the abiotic process related to kitasamycin decline. Further research, such as use of electron spectroscopy for chemical
3.2.2. Total heterotrophic bacteria and kitasamycin-resistant bacteria The total count of heterotrophic bacteria in the three mixtures increased in the thermophilic phase and deceased in the cooling phase (Fig. 3). The rate of increase in M2 was higher than those obtained for the other two mixtures. The limit in the N content of M1 may explain the slow bacterial growth, which is consistent with its low composting temperature during this phase (Fig. 1). However, M3 contained a higher level of extractable kitasamycin residues (Fig. 2), which may restrict the growth of the bacterial population and cause a disturbance on the composting process, particularly during the thermophilic phase (Fig. 1). Although a previous study reported that the addition of antibiotics elicits only a transient perturbation (Selvam et al., 2012), the much higher level of antibiotics in M3 might exert a stronger inhibitory effect on the bacterial growth. During the cooling stage, the bacterial population in M2 decreased gradually, but the number of bacteria in M3 remained high, and the final number was one order of magnitude larger than that obtained for M2, suggesting that M3 is not mature and that there many degradable compounds are available at the end of the composting period for bacterial consumption. The populations of kitasamycin-resistant bacteria in M2 and M3 were markedly higher than those found in M1, indicating that a large amount of resistant bacteria had developed from the kitasamycin waste (Fig. 3). Similar results were also observed in a composting study of medicated manure (Arikan et al., 2007). Furthermore, a type of kitasamycin-resistant bacteria was isolated in the initial compost of M2, which was a rod-shaped, endospore-
Temperature (°C)
60
M1
M2
M3
Environment
50 40 30 20
M2
-1
Extractable kitasamycin (ng g )
70
M1
3000 2500
M3
2000 1500 1000 500 0
10
0
0 0
2
4
6
8
10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40
5
10
15
20
25
30
35
40
Days
Days
Fig. 1. Change of temperature during the composting process of the three composting mixtures and the environment.
Fig. 2. Changes of extractable kitasamycin concentrations in different mixtures during composting. The results are the means of three replicates, and the bars indicate the standard error of the three replicates.
Bacterial population (log (CFU g -1 Fw))
N. Ding et al. / Journal of Environmental Management 134 (2014) 39e46
12
M1-T
M2-T
M3-T
M1-R
M2-R
M3-R
10 8 6 4 2 0 0
5
10
15
20
25
30
35
40
Days Fig. 3. Changes in the total heterotrophic bacteria (M1-T, M2-T and M3-T, respectively) and kitasamycin-resistant bacteria (M1-R, M2-R and M3-R, respectively) during composting. The results are the mean of three replicates, and the bars indicate the standard error of three replicates.
forming, Gram-positive bacterium (Fig. 3S). The bacterium was designed as strain D3. Bacterial 16S rDNA was amplified by using the universal primer pair BSF8 and BSR1541. The length of the sequence is 1430. A UPGMA tree was constructed based on sequence and related ones from the Genebank database. Strain D3 showed the closest match with Paenibacillus sp. (>99% similarity, Fig. 4S). As the composting proceeded, the population of resistant bacteria in M1 rapidly decreased to a level below the detection limit (1.5 log CFU g1 fw) within 10 days. However, for M2, this value gradually declined over a period of 20 days from 7.25 to 3.80 log (CFU g1 fw) and could not be detected after 25 days. The synchronous decrease in extractable kitasamycin (Fig. 2) and kitasamycin-resistant bacteria (Fig. 3) suggest that the kitasamycinresistant bacteria are closely associated with the extractable
43
kitasamycin, and that the non-extractable kitasamycin residues are not bioactive. The biodegradation of kitasamycin in M3 through day 40, as reflected by the level of resistant bacteria, i.e., 3.86 log (CFU g1 fw). Compared with the rapid adsorption of kitasamycin to the matrix during the mesophilic phase of composting, the microbial degradation effect was more modest albeit more persistent. 3.2.3. DGGE profiles of the bacterial community PCR-DGGE was performed to assess the effect of kitasamycin waste on the bacterial community composition during the composting process (Fig. 4). On day 0, there were 25, 28 and 36 bands in DGGE gel for M1, M2 and M3, respectively. The increased numbers of M2 and M3 bands indicates that many specific bacterial species from the kitasamycin waste were introduced into the composts. Dynamic changes in the bacterial communities were found during composting, as was observed by the strong shifts in the DGGE profiles between day 0 and day 40. For M3, the band number decreased from 36 to 30, and this decrease may be due to the high level of extractable kitasamycin exerting an adverse effect on the bacterial diversity. Similar results were obtained by Zeng et al. (2011), who reported that pentachlorophenol stress has an inhibitory effect on microbial abundance. Moreover, the band number found for the M2 samples increased to 57. Although the total population of bacteria in M2 at the end of the composting period was decreased by 2 orders of magnitude compared with the number of bacteria present during the thermophilic phase (Fig. 3), the enhanced band number (Fig. 4) clearly shows that the taxonomic and metabolic diversities increased. The bacterial species that increased in abundance were not only simply organic oxidisers but also involved in hydrogen-, ammonium-, and nitrite-oxidation, nitrogen-fixation, and sulphate-reduction (Diaz-Ravina et al., 1989; Beffa et al., 1996). For M1, there was a smaller increase in the DGGE bands (from 25 to 37) during composting, possible as a result of the low level of N sources, which caused competition among different bacterial species. 3.3. Assessment of compost maturity 3.3.1. Organic carbon, total N and C/N ratio The N% and C% contents decreased over time until the end of the composting period, with the exception of the N% in M1, which exhibited no obvious change over time (Table 1). This finding indicates that the organic compounds in M1 were not well decomposed compared with the other two mixtures. The initial C/N ratio of the mixtures ranged from 16.9 to 39.2, and decreases in the C/N ratios were found for M1 and M2. However, there was no obvious change in the C/N ratio of M3 over time, mainly because the amount of N-mineralization, which is comparable with that observed in C-decomposition.
Fig. 4. PCR-DGGE analysis and the number of DGGE bands of the bacterial community in different compost mixtures on day 0 and day 40.
3.3.2. Water-soluble nutrients For M1, the WS-NHþ 4 concentration was negligible throughout composting with the exception of Day 5 to Day 10, which indicates that there was not sufficiently available N for the assimilation of microorganisms (Fig. 5A). The WS-NHþ 4 concentrations in M2 and M3 sharply increased to similar levels during the thermophilic phase. As the process progressed, the WS-NHþ 4 concentrations quickly declined due to ammonia volatilisation and nitrification (Tiquia, 2005). For M2, the final level of WS-NHþ 4 was lower than the phototoxic limit value of 0.4 g kg1 (Bernal et al., 2009), indicating that it is a qualified product. However, the final level measured for M3 was still 2-fold higher than that obtained for M2, which suggests that the evolution of N forms was not completed. The evolution of WS-P in each treatment followed a similar trend, with a gradual decline during the first 20 days (from 8.71,
44
N. Ding et al. / Journal of Environmental Management 134 (2014) 39e46
25.0
M1
B
M2
M2
M3
0.8
M3
20.0
-1
WS-P (g kg )
-1
WS-NH 4 (g kg )
1.0
M1
A
+
15.0 10.0
0.6 0.4 0.2
5.0
0.0
0.0 0
5
10
15
20
25
30
35
0
40
5
10
20
25
30
35
40
Days
Days
2.0
15
M1
C
M2 M3
+
-1
WS-K (g kg )
1.5
1.0
0.5
0.0 0
5
10
15
20
25
30
35
40
Days Fig. 5. Changes of water soluble nutrients in different mixtures during composting. The results are the means of three replicates, and the bars indicate the standard errors of three replicates.
8.04 and 4.45 g kg1 to 5.43, 3.43 and 2.89 g kg1 for M1, M2 and M3, respectively, Fig. 5B) and a gradual recovery during the next 20 days. The initial WS-K concentrations in M1, M2, and M3 were 1.19, 1.08 and 0.84 g kg1, respectively (Fig. 5C). The WS-K in all of the treatments remained constant and slightly increased at the end of the composting period.
3.3.3. Electrical conductivity (EC) and pH The initial EC of M1, M2, and M3 increased with an increase in the proportion of kitasamycin waste. The EC values of M1, M2, and M3 were 2.14, 2.89 and 3.11 dS cm1, respectively (Fig. 6A). For M1, the EC value gradually decreased during the composting process and was 1.58 dS cm1 on day 40. However, for M2 and M3, the EC
11 6
A
M3
M3
9
pH
-1
4
EC ds m
M2
10
M2
5
M1
B
M1
3
8 7
2
6
1
5
0
4 0
5
10
15
20 Days
25
30
35
40
0
5
10
15
20
25
30
35
40
Days
Fig. 6. Changes in the electrical conductivity (EC, A) and pH (B) in different composting mixtures. The results are the means of three replicates, and the bars indicate the standard errors of the three replicates.
N. Ding et al. / Journal of Environmental Management 134 (2014) 39e46
Furthermore, to ensure the reduction of Escherichia coli reduction during composting, we analyzed the number of E. coli in the initial and final compost of each treatment. Results showed that there was a large amount of E. coli in the initial mixture of each treatment, which was more than 6.0 log CFU g1 (Table 1S). After 40-day composting, the numbers of E. coli both in M2 and M3 decreased to under the detection limit (1.5 log CFU g1). In comparison, a high number of E. coli (3.7 log CFU g1) still remained in the final compost of M1, which was mainly because the temperature of M1 did not meet the thermal death point (55 C) during the thermophilic phase (Hanajima et al., 2006).
M1
100%
M2 M3
GI (%)
80%
45
60% 40% 20%
4. Conclusions
0% 0
5
10
15
20
25 30
35
40
Days Fig. 7. Changes in germination index (GI %) in different mixtures during composting. Results are the mean of three replicates, and bars indicate standard error of three replicates.
values increased and reached maximum values of 3.55 and 4.46 dS cm1 during the thermophilic phase, respectively, and then gradually decreased during the cooling phase. The final EC value for M2 did not exceed the limit content of 3.0 dS cm1 (Soumaré et al., 2002), indicating that the EC does not adversely affect plant growth. Optimal pH values between 5.5 and 8.0 can support efficient microbial activities during composting (Bernal et al., 2009). In general, pH is not a key factor in composting because most compost materials exhibit pH values within this pH range. However, this factor was very relevant in the present study because the cattle manure used had a high pH of >9.0 and kitasamycin waste had a low pH of
