Toxicity Evaluation of Pig Slurry Using Luminescent Bacteria and Zebrafish

Biogas slurry has become a serious pollution problem and anaerobic digestion is widely applied to pig manure treatment for environmental protection and energy recovery. To evaluate environmental risk of the emission of biogas slurry, luminescent bacteria (Vibrio fischeri), larvae and embryos of zebrafish (Danio rerio) were used to detect the acute and development toxicity of digested and post-treated slurry. Then the ability of treatment process was evaluated. The results showed that digested slurry displayed strong toxicity to both zebrafish and luminescent bacteria, while the EC50 for luminescent bacteria and the LC50 for larvae were only 6.81% (v/v) and 1.95% (v/v) respectively, and embryonic development was inhibited at just 1% (v/v). Slurry still maintained a high level of toxicity although it had been treated by membrane bioreactor (MBR), while the LC50 of larvae was 75.23% (v/v) and there was a little effect on the development of embryos and V. fischeri; the results also revealed that the zebrafish larvae are more sensitive than embryos and luminescent bacteria to pig slurry. Finally, we also found the toxicity removal rate was higher than 90% after the treatment of MBR according to toxicity tests. In conclusion, further treatment should be used in pig slurry disposal or reused of final effluent.

For further evaluation of health risk, toxicity tests with early development stage of aquatic organisms have been introduced as a faster and more cost-effective way. Moreover, study showed that the early developmental stages of fish are often the most sensitive to toxic effects [18]. For relatively large, robust embryos and rapid embryonic development, zebrafish could be an ideal vertebrate model organism; moreover, transparent body is easily observed when zebrafish is developing outside their mother fish [19][20][21]. It was reported that some indexes, such as embryos production, atretic oocytes and altered ovarian histology and embryos mortality, could be useful to evaluate pharmaceutical mixture and municipal wastewater [22].
In this study, biological toxicity testing method was introduced to pig slurry, where related report on toxicity of slurry is still lack of study now. Here, V. fischeri, newly hatching larvae of zebrafish was used to evaluate acute toxicity of pig slurry, and embryos (1 hpf, 1 h post-fertilization) of zebrafish was for development toxicity. Using these methods, both digested and post-treated slurry were investigated to provide useful information about health risk of slurry and ability of the treatment process to reduce the toxicity of pig slurry, combined with physicochemical indexes (chemical oxygen demand (COD), ammonia nitrogen (NH 3 -N), total phosphorus (TP), etc.).

Collection and Determination of Physicochemical Index of Samples
All samples of pig slurry were collected in Xinxin Forage Corporation in Jiaxing City of China. And before, pig slurry was pretreated by solid-liquid separation. Anaerobically digested and post-treated slurry (a 43 L membrane bio-reactor) were collected respectively. Digested slurry was collected in the outlet of anaerobic digester. Treated slurry was collected in the outlet of membrane bioreactor. Each index was determined three times respectively. After collection, physic-chemical variables were determined. Conductivity, pH, NH 3 -N, TP and COD were determined by conductivity meter (Monitoring and analysis method of water and waste water, in Chinese), glass electrode method (GB/T6920-1986, in Chinese), nessler's reagent colorimetric method (GB/T7479-1987, in Chinese), ammonium molybdate spectrophotometric method (GBT11893-1989, in Chinese) and potassium dichromate method (GB/T11914-1989, in Chinese) respectively. And then the samples were stored at 4 °C until used. Before the toxicity experiments, all samples were diluted, where de-ion water was used in luminescence experiment, and standard dilution water [23] was used in zebrafish experiment.

Zebrafish
Adult zebrafish (AB strain) were maintained in a recirculating aquaculture system (Aquaneering Co., San Diego, CA, USA). In incubation process, the 12 h light period was followed 12 h dark period per day. In the light period, the fish were fed with freshly hatched shrimp eggs and flake fish food (Tetra, Melle, Germany), twice and once respectively. The incubation temperature was controlled at 28 ± 0.5 °C.

Fertilized of Zebrafish Embryos
To hatch zebrafish embryos, one adult female fish and one adult male fish were placed in the same box. After the formation of zygote, embryos were washed 2-3 times by standard dilution water, for removing residues. Finally, normal developed fertilized eggs which were observed by the TS100-F microscope (Nikon, Tokyo, Japan) were collected for subsequent experiments.

Luminescent Bacteria Toxicity Test
Luminescent bacteria test was performed using 96-well microplate on the GloMax-Multi Detection System. Due to high toxicity of slurry and in order to eliminate the effect of the color and density on results, filtered samples were diluted to avoid complete inhibition. In this paper, volume percentage of sample in de-ion water was adopted to represent dilution degree, where raw sample is 100% v/v and de-ion water which was used as control sample was 0% v/v. The procedure in detail which was referring to research of Froehner et al. [24] was as follows: 100 μL de-ion water as blank controls was added to the first row of microplate, 100 μL sample with various dilution(respectively 100%, 50%, 25%, 12.5%, 6.25%, v/v) were added to the second, third, fourth, fifth and sixth row of microplate respectively. And then 100 μL bacteria suspension were added to each test well. After 15 min exposure, the luminescence intensity was measured. All of tests were repeated three times, while average luminescence intensity was adopted to dose-effect plot. Finally, the toxicity of slurry was characterized by relative luminous intensity and the concentration for 50% of maximal effect (EC 50 ).

Larvae of D. Rerio Acute Toxicity Test
The D. rerio 96 h acute toxicity test was carried out according to the procedure described in ISO7346-1 [25]. To detect toxicity using zebrafish larvae, digested and post-treated slurry were diluted to a series of exposure solutions to avoid complete inhibition due to high toxicity of slurry. Here volume percentage of sample in standard dilution water was adopted to represent dilution degree, where raw sample of digested and post-treated slurry is 100% v/v and standard dilution water which was used as control sample was 0% v/v. In exposure experiment, ten normally developed larvae were transferred to each culture dish (100 mm) containing 15 mL sample with different dilution degree. The number of dead larvae was counted at 24 h, 48 h, 72 h and 96 h after exposure to slurry. In this experiment, five different dilution degrees were introduced to evaluate the toxicity of slurry.

Toxicity Test of Zebrafish Embryos
To detect toxicity using zebrafish embryos, diluted samples were needed to be prepared to avoid complete inhibition due to high toxicity of slurry. Normal embryos (at approximately 1 hpf) were kept in 24 well cell culture plates, with one embryo per well. Each well contained 1ml control or exposure wastewater. Two replicates for the controls and exposure groups were used. For each control and exposure group, the early embryonic development was observed by the TS100-F microscope and mortality was recorded at an interval of 24 h. After 72 h exposure experiments, mortality, hatching rate and malformation rate of embryos in each group were recorded. Similarly, five different dilution degrees were adopted [26].

Methods of Toxicity Evaluation
The toxicity of pig slurry on zebrafish larvae was evaluated using 96 h lethal concentration 50 (LC 50 ), for V. fischeri and zebrafish embryos, EC 50 and ELC 50 , HEC 50 and MEC 50 were used respectively. In this paper, ELC 50 , HEC 50 and MEC 50 were used to represent the concentration of wastewater for 50% of embryonic mortality, hatching and malformation respectively while exposed to the pig slurry. Finally, toxicity unit (TU) was used to represent the toxicity directly. TU was calculated according to the formula as follows [27]: If the inhibition of luminescence intensity for V. fischeri, hatching rate for zebrafish embryos, mortality for zebrafish larvae were lower than 50% and malformation rate for zebrafish embryos was lower than 50% exposed to pig slurry, it showed LC 50 (EC 50 ) couldn't be calculated, TU was calculated according to the formula as in Equation (2) where RE was the relative inhibition rate of V. fischeri luminosity and death rate of larvae and embryos of zebrafish (%) And the toxicity remove rate was calculated according to the formula as follows:

Statistical Analysis
Using origin 8.0 (Origin Lab, Northampton, MA, USA), the median lethal concentration (LC 50 ) and the median effect concentration (EC 50 ) on D. rerio and luminescent bacteria of pig slurry were calculated, followed with one-way ANOVA in SPSS 16.0. Here statistical significance difference of exposure group to the control group was set to p < 0.01.

Physicochemical Characterization of Pig Slurry Effluents
The calculated values of physicochemical indexes were listed in Table 1. It showed that pollutants in the pig slurry were reduced mostly after the treatment of MBR, where the value of NH 3 -N and COD of raw water were reduced 94.5% and 86.0% respectively. On the other hand, the allowed values of NH 3 -N and COD are 80 mg/L and 400 mg/L respectively, according to discharge standard of water pollutants for livestock and poultry breeding(GB 18596-2001) [28]. It meant the effluent could be discharged legally.

Effect on Luminescent Bacterial Exposed to Pig Slurry
The dose-response curves after 15 min exposure were plotted using logistic model, shown in Table 2 and Figure 1. Dose-response curve is nonlinear fitted by Logistic regression model (Equation (4)): The relative luminosity of V. fischeri exposed to digested and post-treated slurry, were 6.9% and 66.8% respectively, which means both of them had a certain degree of inhibitory effect on the V. fischeri. Table 2 showed good statistical significance for regression model while the R 2 was higher than 0.99. And the toxicity could be divided into the following four categories based on EC 50 [7]: (i) high toxicity for samples with an EC 50 < 25%; (ii) moderate toxicity, EC 50 25% to 75%; (iii) low toxicity with a 75% to 100% of EC 50 ; (iiii) non-toxicity, EC 50 can't be calculated. The EC 50 of digested slurry to V. fischeri was 6.81% dilution, which means high toxicity. In contrast, relative luminosity was inhibited only 40.09% by raw sample for post-treated slurry, which is lower than 50%. So EC 50 of post-treated slurry can't be calculated, which meant non-toxicity to luminescent bacteria. According to the TUs, toxicity remove rate of was 94.6% for MBR-treated slurry. Compared with the physicochemical characteristics of slurry, the similar remove effect was proved.

Effect on Zebrafish Larvae Exposed to Pig Slurry
The results of 96 h acute toxicity test of zebrafish larvae were showed in Figure 2 and Table 3. As Figure 2 described, when larvae were exposed to digested slurry (2.5%, v/v), the mortality had already reach 100%. On the other hand, the same effect on mortality could be found while the volume percentage was up to 100% for post-treated slurry. Table 3 showed LC 50 of larvae exposed to digested and post-treated slurry were different, 1.95% and 75.23% respectively. According to TU, treatment of MBR reduces the toxicity by 97.5%. The results showed the high degree of reduction of toxicity after treatment by MBR. It showed that the reduction rate calculated by 96h acute toxicity test of larvae was not much different from by luminescent bacterial acute toxicity test and physicochemical indexes.

Effect on the Early Development of Embryos Exposed to Pig Slurry
Post-treated slurry showed a certain degree of toxicity to mortality, hatching and malformation (45%, 25% and 41.7% respectively in Table 4) compared with control group. In 72 h experiments with embryos exposed to digested and post-treated slurry, there was a significant effect on embryonic mortality, hatching rate and malformation rate, while the ELC 50 , HEC 50 and MEC 50 of embryos exposed to digested slurry were 3.48%, 1.32% and 3.47% respectively. And post-treated slurry showed no inhibition to the embryos of ELC 50 and MEC 50 . However, the EC 50 of hatching rate of embryos exposed to post-treated slurry reached 31.81% (Table 5).  According to the TU in Table 6, the remove rate of toxicity to mortality, hatching and malformation were 96.9%, 95.9% and 97.1% respectively. With microscopic observation, malformed embryos displayed the different status, such as the 24 h no extension of the embryonic tail, 24 h coagulated egg, 48 h pericardial edema, 48 h pigment deposition, 72 h-malformed larvae. On the other hand, 24 h development of eyespot and 48 h otoconial development were not found in this experiment (Figure 3).

Discussion
The application of biogas engineering has received considerable attention in recent years, while there are many controversies about the advantages and disadvantages of biogas slurry. In some studies, slurry could be used to give rise of high yields to crops.J. Abubaker et al. [29] conducted a study on the fertilizing performance of pig slurry and mineral fertilizer in terms of spring wheat yield, in conclusion, pig slurry gave the overall highest yields to wheat. Gobernaa et al. [30] compared biogas digestates with fresh manure to the inhibition of pathogenic bacteria in soils, and found anaerobic digestion significantly sanitized the manure by completely eliminating cultivable E. coli and Salmonella. However, toxic pollutants in the biogas slurry could be dangerous to environment. Many different pollutants had been found in slurry. For evaluation of availability of slurry, some researches were conducted. Using Daphnia magna, A.I. De la Torre et al. [31] stated that toxicities of pig slurry were higher than urban effluents and lower than industrial effluents.
For evaluation of acute and development toxicity of slurry, V. fischeri, zebrafish larvae and embryos were exposed to digested and post-treated slurry respectively. In this study, different trophic level organisms showed different sensitivity to digested and post-treated slurry. The sensitivity order of the three test organisms was larva > embryo > V. fischeri. The difference of larva and embryo of zebrafish could result from membrane out of the embryo, which selectively restrict big molecule compounds to enter into embryo, which could be regarded as protective effect on the embryo [32], as had been reported by Wiegand C et al. [33,34]. There is little literature on the conductivity requirements for zebrafish larva and embryo, but adult zebrafish are tolerant to conductivity ranging from 400 μS·cm −1 to more than 1000 μS·cm −1 [35]. In this study, conductivity of digested and post-treated slurry was both far higher than 1000 μS·cm −1 . In addition, there was a correlation between conductivity and salinity generally. And V. fischeri was regarded as marine bacteria, so probably the higher sensitivity of fish versue V. fischeri could be mainly due to the different salinity tolerance. It showed the use of the marine bacteria tests is very important because it can avoid effect of salinity on test organism, this conclusion has also been confirmed by Pardo et al. [36]. Result also demonstrated that pig slurry had a high toxicity to both V. fischeri and zebrafish. The toxicity of digested slurry to V. fischeri was 14.68 TUs, which was higher than the TU of phenol (12.5 TUs) and lower than that of dimethyldiuron (16.23 TUs). Meanwhile, the toxicity of post-treated slurry to V. fischeri was 0.80 TU, which was between the TU of polyethylene glycol (0.78 TU) and cypermethrin (0.91 TU) [37]. In addition, relative luminosity of V. fischeri exposed to post-treated slurry, was very significantly inhibited comparing with the de-ion water. On the other hand, relative luminosity of V. fischeri was higher than 100% (ANOVA, p < 0.01; Figure 4), when effluent was diluted to a certain degree of volume percentage. Some nutrients existed in pig slurry could enhance the cell activity of luminescent bacteria. Similar results could be observed in pure compounds experiments. For example, relative luminosity of V. fischeri was higher than 100% when the concentration of phenol was lower than 0.005 mg/L [38]. Probably because that the inhibition on biological effect was lower, and bacterial recovery effect is stronger. In addition, zebrafish larvae were all dead when exposed to digested and post-treated slurry, the same effect as the zebrafish embryos exposed to digested slurry, however the mortality of embryos exposed to post-treated slurry was lower than 50%. Meanwhile, effects of pig slurry on the early developing zebrafish embryos were observed. The experiment of embryos exposed to post-treated slurry also showed that the hatching rate was lower than 30% and malformation rate was more than 40%. According to reported development toxicity test of urban sewages using zebrafish embryos, hatching rate of embryos was more than 80% and malformation rate of embryos was lower than 20% respectively [39]. It showed the toxicity of pig slurry was higher than urban sewages.
Moreover, 96 h LC 50 and toxicity of post-treated slurry to the zebrafish larvae were higher than 70% and lower than 2 TUs respectively. According to acute toxicity test of zebrafish larvae exposed to fat-plant effluent, showed that the value of LC 50 of larvae exposed to effluent was lower than 70%, reported by Şişman T. et al. [40]. In addition, adult zebrafish were used to evaluate the toxicity of industry effluent, such as electronic and electroplate effluent, the result showed the TU of effluent was 2.54 [41]. Then considering that zebrafish larva is more sensitive than adult zebrafish. Thus, toxicity of post-treated slurry was lower than industrial effluents. And it showed that it was necessary to apply treatment to slurry, owing to high environmental risk.
Apparently, we could find the toxicity of digested slurry was reduced mostly after the treatment of MBR according to the luminescent bacteria toxicity test and zebrafish larvae acute toxicity test, the removal rate was 94.6% and 97.5% respectively, meanwhile according to the results of physicochemical index determination, post-treated slurry could be discharged legally. However, mortality of larvae still reached 100% when exposed to post-treated slurry. Besides other toxics, such as ammonia could be responsible for the toxicity of pig slurry. Recent study shows that ammonia can be physiologically harmful to Hypophthalmichthys molitrix larvae, through increasing the concentration of reactive oxygen species and oxidative damage products such as lipid peroxides [42]. Among the physiochemical characteristics of the slurry (NH 3 -N, TP and COD), there was a high correlation between the toxicity removal rate and reduction rate of NH 3 -N. It meant the decrease of concentration of NH 3 -N, toxicity of pig slurry has also been reduced. So result that NH 3 -N made much greater contribution on slurry toxicity could be obtained and further treatment should be used in pig slurry disposal or reused of final effluent.
In this paper, risk information was acquired while evaluation method was applied to pig slurry. However, more organisms for quantitative assessment are necessary. So a battery of bioassays based on organisms belong to different trophic levels, are strongly suggested, which could get a risk score by the application of a synthetic index for toxicity [43], combining algae, Daphnia magna and so on.

Conclusions
Pig slurry had showed different toxicity effect on V. fischeri and zebrafish. And larva was the more sensitive one to all the test wastewater, followed by embryo of zebrafish and finally, V. fischeri.
Furthermore, both V. fischeri and zebrafish could be used in the toxicity reduction and risk assessment of pig slurry, and provided suggestions for its treatment process. Based these methods, risk information of slurry could be acquired. Here digested slurry displayed strong toxicity to luminescent bacteria, zebrafish larva and embryo. On the other hand, post-treated slurry still showed low toxicity although it has reached the discharged standard, and toxicity remove rate has reached up more than 90%. Considering that, too much discharged slurry could exceed rural river carrying capacity, advanced treatment for slurry, such as ion exchange technique, wet oxidation process, biological nitrogen removal system and so on, are suggested.