Characteristics of Steroid Estrogen Loss, Degradation and Residues during Open-Air Dairy Manure Disposal

Abstract

Steroid estrogens (SEs) are mainly derived from livestock manure, and composting is the common method of bioresource utilization. In this study, an open-air composting experiment with dairy manure was conducted to observe the degradation and loss of five SEs under the influence of different precipitation intensities and additional mixed fermentation strains. SE determination results for dairy manure after 30 days of composting indicated that the average removal rates of 17α-estradiol (E2α), estrone (E1), 17β-estradiol (E2β), 17α-ethinyl estradiol (EE2), estriol (E3), and estradiol equivalent (EEQ) were 76.67%, 71.07%, 73.88%, 92.02%, 98.77%, and 88.11%, respectively, partly due to SE runoff. The rates of SEs leaching from the open-air composting dairy manure ranged from 0.05% to 4.75% after 10 rounds of 5–60 mm/d simulated rainfall. The total leaching amount of SEs was positively correlated with rainfall, but the leaching concentrations of SEs were just the opposite. As a result of its role as a degradation intermediate of other SEs, E3 was the most prone to run off. By strengthening the action of microorganisms, the total leaching amount of EEQ increased by 5%, E3 increased by five times, and E2β also underwent a transition from a conjugated form to free. However, there were also fewer final SEs remaining in the composted product, as well as the environmental risks of conjugated SEs. These conclusions can provide beneficial suggestions and references for controlling the environmental risks of SEs in the process of composting livestock and poultry manure.

1. Introduction

At present, organic pollutants such as estrogen and antibiotics are widely discharged into water resources due to the increasing population and global industrialization [1,2]. The steroid estrogens (SEs) are a type of endocrine-disrupting chemicals (EDCs) or endocrine disruptors (EDs), which may potentially contaminate surface and groundwater resources and cause harm to organisms at very low concentrations (level of ng/L) [3,4,5]. Estrone (E1), 17β-estradiol (E2β), 17α-estradiol (E2α), estriol (E3), and 17α-ethinylestradiol (EE2) are currently the most important feminine hormones, showing strong endocrine-disrupting activity [6,7]. SEs are mainly excreted into the environment by humans, livestock, and poultry [8]. Many researchers have monitored the fate of estrogens in WWTPs and evaluated the removal efficiency of SEs by different treatment processes [9,10]. For animal manure, the degradation, sorption, and mobility of SEs have been the subject of concern, as well as their potential risks to surrounding surface water and groundwater [11,12,13]. However, there is limited information [14,15,16] available in the literature on the leaching of SEs from animal manure, especially considering some actual environmental conditions for bioresource utilization. For example, in related research by Kjaer [16] and Gall [14,15], the focus was on the migration of SEs after manure was put into the soil, while the risk of loss before the manure was put into the soil was ignored. Our research team visited and investigated many rural areas in China, and found that due to the low level of the rural economy and decentralized farming, a large amount of animal manure was not collected centrally, but was piled up randomly on the roadside or beside houses for a long time for farmland use. During this period, we have seen that the manure being open-air composted was repeatedly washed into rivers or lakes by heavy rains, posing a greater environmental risk.

Here, an open-air composting experiment with dairy manure was conducted to observe the degradation and leaching of five SEs (E1, E2α, E2β, EE2, and E3) under the influence of different precipitation intensities and additional mixed fermentation strains. We paid attention to the leaching concentration, the total loss and conversion between SEs, as well as the role of microorganisms before and after composting. We hope that this study can provide beneficial suggestions and references for controlling the environmental risks of SEs in the process of composting livestock and poultry manure.

2. Materials and Methods

2.1. Experimental Procedures

A total of 8 identical experimental setups were fabricated in this experiment, as shown in Figure 1. Dairy manure used in the experiment was taken from dairy farmers in Fengyi Town, Dali, China. Before composting, raw SEs, moisture content, and the solid physical and chemical properties of dairy manure were determined. After the dairy manure was mixed, it was divided into 8 parts for 8 experimental groups, and each part of 5 kg (wet weight) was placed into the experimental setup in form of a heap. Four experimental groups were added to 2.5 mL of organic fertilizer fermentation agent (main ingredients included Bacillus Natto, Bacillus, Actinomycetes, Yeast, Trichoderma, Azotobacteria, Lactic Acid Aacteria and their secretory extracellular enzymes, and the living bacteria counted more than 10 billion/mL) diluted with 200 mL of ultrapure water separately, while the 4 control groups received no additions. The 4 experimental groups and 4 control groups simulated different rain events, which were 5 mm/d (light rain), 10 mm/d (moderate rain), 30 mm/d (heavy rain), and 60 mm/d (torrential rain), respectively. The manure pile was manually stirred once a day to gain sufficient oxygen for the compost. The experiment lasted for 30 days and consisted of 10 cycles, of which 3 days was designed as a rainfall cycle. Rainfall flowed through the manure pile and into the collection container, where the leaching liquid was collected, and then the liquid volume and the SE concentration were determined after each simulated rainfall cycle. After the composting experiment ended, the residual SEs in the manure were determined.

2.2. Sample Pretreatment

Water sample. Prior to sample extraction, the solid-phase extraction (SPE) cartridge was rinsed with a suitable solvent so that the analyte was in close contact with the solid surface to improve recovery and reproducibility. Next, 3 mL n-hexane, 3 mL ethyl acetate, 3 mL methanol and 5 mL ultrapure water were used to activate the column, and the process was repeated twice. The liquid dropped at a rate of about 1 mL/min for 5 min to ensure that the filler was fully immersed in the activation reagent. Then, 0.5 L~1.0 L of pretreated water samples were imported into the activated SPE cartridge under a certain vacuum pressure, and the flow rate was controlled at about 4 mL/min. SEs and some impurities were then adsorbed and retained on the SPE cartridge. The cartridge was rinsed with 10 mL of water with 10% methanol at a flow rate of about 5 mL/min. Then, 10 mL ethyl acetate was used to elute the cartridge, the dropping rate was controlled at about 1 mL/min, and the eluent was collected. 25 μL of N,O-Bis (trimethylsilyl) trifluoroacetamide (mass TMCS containing 1%) and 50 μL of pyridine solution were added to the dried SE eluent at a temperature of 30 °C and with high purity nitrogen. The solution was then heated by microwave at 300 W for 4 min. After nitrogen blowing, 1 mg/L Mirex 400 μL was added to dissolve SEs. Then the mixture was transferred to 2 mL sample vials for detection.

Manure sample. The manure samples were centrifuged at 3000 rpm for 10 min, then the supernatant was discarded, and the remaining samples were freeze-dried for 24 h. A total of 5 g of freeze-dried sample was mixed with 10 mL of methanol and ethyl acetate (V/V = 1:1), and subjected to ultrasonic extraction for 20 min (250 W, 40 kHz). After the mixture was centrifuged at 5000 rpm for 15 min, the supernatant was collected. The above operation was repeated three times, and the collected supernatants were combined and dried under high-purity nitrogen, and the target substance was dissolved in 2 mL of n-hexane. Subsequent pretreatment steps were the same as for water samples.

2.3. GC/MS Analysis and Quality Control

SE chemical analysis standards (purity ≥ 97.0%) were purchased from Aladdin Reagent (Shanghai) Co., Ltd., and the standard substitute E2β-d2 was purchased from C/D/N ISOTopes in Canada. The testing instrument was Thermo Fisher TRACE 1300-ISQ Series Quadrupole. GC conditions: TR-5 MS quartz capillary column (30 m × 0.25 mm × 0.25 μm); high-purity helium (99.999%) carrier gas, constant flow mode (1.2 mL/min carrier gas flow rate); injection volume 1.0 μL in splitless condition; inlet temperature 280 °C; septum purges 5.0 mL/min; column heating procedure: initial temperature 50 °C for 2 min; heat at 20 °C/min to 260 °C for 5 min; increase to 280 °C at a rate of 10 °C/min and hold for 5 min. MS condition: Electron bombardment ion source (EI), ionization voltage 70 eV; ion source temperature 250 °C, transmission line temperature 280 °C; solvent delay time 10 min; full scan mode (SCAN) for qualitative and selective ion scan mode (SIM) for quantification. Analysis was performed using Xcalibur software to determine SEs and internal standards.

In order to verify the accuracy and precision of the method, a standard solution containing 50 ng and 500 ng of the test component were used as samples, and the measurement was performed 7 times within one day. The limits of detection (LOD) of E1, E2α, E2β, EE2, and E3 were 1.48, 1.13, 1.65, 2.03, and 0.97 μg/L, respectively, defined as 10 times the signal/noise ratio. The average spike recovery of SEs was 98.0%. The relative standard deviation (RSD) ranged from 0.4% to 6.8%.

2.4. Estrogen Equivalent(EEQ) Calculation

EEF (E2β active equivalent) is often used in calculating EEQ (estrogen equivalent concentration, ng/L). In this study, the EEF data in the literature were used, according to E1 = 1/3 [17,18], E2β = 1, E2α = 1/8 [19], E3 = 1/12 [20], and EE2 = 2.2 [18]. To simplify the assessment and facilitate comparison with the literature, estrogen evaluations of standardization as EEQ-E2β were calculated as follows.

$$\mathrm{EEQ} ={\displaystyle \sum }_{\mathrm{i}}^{\mathrm{n}}{\mathrm{EEF}}_{\mathrm{i}}\times {\mathrm{C}}_{\mathrm{i}}$$

(1)

EEQ (ng/L or ng): the active equivalent of E2β; EEF_i: E2β equivalent of the estrogen i; C_i (ng/L or ng): the measured ambient concentration of estrogen i, n = 5 (i.e., E1, E2β, E2α, E3, and EE2).

3. Results and Discussion

3.1. SEs/EEQ in the Leaching Liquid

As shown in Figure 2a, SEs ran off more rapidly before the Round 7 (R7) of rainfall, and the leaching rate slowed down after R8. Similar to the findings of related estrogen loss studies [14], the highest concentration value of EEQ in the drained liquid occurred in R1 generally, and the leaching EEQ concentration in the AM (with added microorganisms) group reached more than 1800 ng/L under torrential rainfall conditions (60 mm/d). This was mainly due to the rapid loss of some SEs adsorbed on small organic particles with the leachate. Therefore, the loss concentration of SEs decreased significantly in R2. Throughout the experimental period, the mean values of the leaching E2α, E1, E2β, EE2, and E3 concentrations were 35.58 ng/L, 59.41 ng/L, 118.04 ng/L, 174.52 ng/L, and 141.27 ng/L, respectively. Leachate containing high concentrations of SEs entering the surface water environment is likely to cause a series of ecological risks. Chronic predicted no-effect concentrations (PNEC) for aquatic systems are used to support the assessment of potential risks to aquatic organisms in receiving waters. In Caldwell’s opinion, for long-term exposures to steroid estrogens in surface water (>60 d), the PNECs were 6, 2, 60, and 0.1 ng/L for E1, E2β, E3, and EE2, respectively [21]. However, Wu’s research showed the more reasonable PNEC value was 0.73 ng E2β/L, which was consistent with other E2β PNEC values derived by the European Union (0.4 ng/L) for protecting aquatic life [22,23]. Combining the PNEC value with the data in our study, we found that manure leachate, after entering rivers and lakes, would pose serious ecological risks.

From the leaching concentration alone, it seems that the precipitation intensity had some influence on SE loss, and light rainfall was more likely to result in the leaching out of high concentrations of EEQ. By comparing the AM with the CG (control group), the effect of enhanced microorganisms on the leaching concentration of SEs was not significant. However, we can get more information from the variations in the SE leaching amount. Figure 2b showed that the greater the rainfall, the greater the loss of SEs, and the difference was obvious, but the effect of enhanced microbial action could not be determined.

Before the experiment, the moisture content of the manure was 83.9%, the average particle size was 122.7 μm, and the nitrogen, phosphorus, and organic matter contents were 26.0 g/kgDW, 6.1 g/kgDW, and 589 g/kgDW, respectively. After the experiment, the average particle size was 141.6μm, the nitrogen, phosphorus, and organic matter contents were 24.1 g/kgDW, 6.6 g/kgDW, and 646 g/kgDW, respectively. The raw contents of E2α, E1, E2β, EE2, and E3 in dairy manure before the open-air compost experiment were 766.74 μg/kg, 889.89 μg/kg, 525.84 μg/kg, 1479.07 μg/kg, and 1000.07 μg/kg of dry weight (DW), respectively. Throughout the experimental period, the amounts of five SEs were 0.37–2.78 μg/kg, 0.69–4.51 μg/kg, 0.98–14.18 μg/kg, 1.98–18.90 μg/kg, and 0.61–47.47 μg/kg, respectively, as indicated in Figure 3. Composting can affect estrogen concentrations, with composting reported to remove steroid estrogen in manure at efficiencies ranging from 79 to 87% [14,24]. From the perspective of loss quality composition, E3 loss was highest, followed by EE2 and E2β, while E1 and E2α were the least likely to be lost. It is worth noting that in the experimental group AM-60 mm, the loss of almost all SEs was the highest (except for E1); the loss of E3 was especially prominent.

3.2. Total Loss and Loss Rate of SEs

We calculated the leaching rates of five SEs and EEQ for the raw content for each experimental group throughout 10 rounds of simulated rainfall to analyze the effects of rainfall intensity and enhanced microbial action on SE leaching (

Table 1. Percentage of SE loss to the raw content.

Experimental Groups	E2α	E1	E2β	EE2	E3	EEQ
AM-5 mm	0.05%	0.11%	0.21%	0.13%	0.06%	0.14%
AM-10 mm	0.12%	0.12%	0.41%	0.23%	0.10%	0.24%
AM-30 mm	0.21%	0.40%	1.30%	0.48%	1.07%	0.58%
AM-60 mm	0.36%	0.33%	2.70%	1.28%	4.75%	1.43%
Mean value for AM	0.19%	0.24%	1.15%	0.53%	1.49%	0.60%
CG-5 mm	0.05%	0.08%	0.19%	0.13%	0.08%	0.13%
CG-10 mm	0.07%	0.13%	0.41%	0.22%	0.48%	0.24%
CG-30 mm	0.36%	0.33%	1.31%	0.69%	0.21%	0.72%
CG-60 mm	0.36%	0.51%	2.60%	1.08%	0.22%	1.20%
Mean value for CG	0.21%	0.26%	1.12%	0.53%	0.25%	0.57%
Ratio of AM/CG	0.90	0.91	1.03	1.00	6.01	1.05

). From the perspective of the proportion of the overall loss to the raw SE content under the condition of open-air compost, the loss ratio of five SEs was between 0.05 and 4.75%, with obvious differences. However, the differences in solubility and hydrophobicity of the five SEs are not enough to explain this.

The ratio of AM/CG (Table 1) for SEs indicated that the loss of E2α and E1 decreased, while the loss of E3 and E2β increased with the addition of fermentation agents, and these ultimately led to an increase in SE and EEQ loss. Among them, the loss of E3 in the condition of enhanced microorganisms increased by a factor of five. The change of SEs in leachate can reflect the degradation characteristics of SEs in feces to a certain extent, especially the mutual conversion between SEs. To our knowledge [25], E1 and E3 are both degradation intermediates of E2α and E2β, but not EE2, and E1 can be degraded into E3 under aerobic conditions. While SEs were leaching with rainfall, they all underwent significant degradation, and E2α, E2β, and E1 degraded significantly and were converted to E3 in large quantities under aerobic conditions. This is why E2α and E1 decreased and E3 increased. However, if this is so, why were E2β and EEQ an exception? Microorganisms promoted the degradation of SEs, so E2β should also have decreased like E2α and E1, and the total EEQ should have decreased. We speculate that this is mainly because we only determined free estrogens in this study, and not conjugated estrogens. Estrogen metabolism results in the formation of inactive estrogen sulphates and glucuronides [26]. It is estimated [27] that estrogen conjugates account for at least 1/3 of the total EEQ in livestock and poultry excrement. The presence of partially conjugated estrogens in the raw manure transformed into corresponding active free estrogens under microbial action [28,29], thus increasing the amount of E2β and EEQ loss. Conversion of E2-3S, E2-17S, E2-3G, and E2-17G to E2 may have occurred during composting [30,31,32].

3.3. Residual SEs in Manure after Composting

After the composting experiment, the residual SEs in the manure were determined (Figure 4a), and the residual concentrations of E1, E2α, E2β, EE2, E3, and EEQ were 46.83–345.02 μg/kg, 72.38–416.18 μg/kg, 117.39–187.58 μg/kg, 35.39–172.36 μg/kg, nd-21.52 μg/kg, and 229.48–739.97 μg/kg of dry weight (DW), respectively. Following an open-air composting and fermentation process of 30 d, the average removal rates of E2α, E1, E2β, EE2, E3, and EEQ were 71.64%, 69.35%, 76.47%, 91.15%, 98.31%, and 87.52%, which also included the loss rate of 0.20%, 0.25%, 1.14%, 0.53%, 0.87%, and 0.59%, respectively. This is similar to studies [24] that reported removal of steroid hormones in cattle manure composting ranging from 79% to 87%. Compared with the degradation amount, the proportion of the total loss to the original content was relatively low in the open-air composting process. Figure 4b shows the remaining proportion of SEs and EEQ in different experimental groups. The amount of residual SEs in AM groups was significantly less than the CG, except for E2β, demonstrating that the degradation of SEs was accelerated in the natural composting process with the enhanced effect of fermentation microorganisms. The situation of residual E2β confirmed our previous speculation that the microorganisms promoted a significant conversion of E2β from the conjugated form to the free form.

The proportion of each estrogen clearly changed during the composting process. As shown in Figure 5, the SEs in original dairy manure were EE2(32%) > E3(22%) > E1(19%) > E2α(16%) > E2β(11%), and the SEs in the leaching liquid were E3(33%) > EE2(30%) > E2β(23%) > E1(8%) > E2α(6%), while the SEs after composting were E1(37%) > E2α(25%) > E2β(19%) > EE2(17%) > E3(2%). From the changes in SE composition, we can draw conclusions. E1 and E3 are both the degradation intermediates of E2α and E2β, but under anoxic or anaerobic conditions, the rate of conversion of E1 to E3 is lower than the rate of degradation of E3, and E1 can be reduced to E2β under the action of some biological reductase [33]. These all increase the difficulty of E2β removal, and result in higher requirements for the oxygen conditions of manure composting. In this experiment, we manually turned over the heap to increase oxygen, which may not ensure sufficient oxygen for all feces. E1 eventually remained more in this experiment, and the higher E3 in the leaching liquid reflected this problem. In addition, previous studies concluded that the degradation of EE2 was more difficult than that of E2β [34]. In this study, the removal rate of EE2 was higher than that of E2β, mainly due to the interference of conjugated E2β. Although the conjugated EE2 also existed, it mainly existed in oral contraceptives and should be absent from dairy manure.

4. Conclusions

In this study, an open-air composting experiment with dairy manure was conducted to observe the degradation and loss of five SEs under the influence of different precipitation intensities and additional mixed fermentation strains. SE determination results for dairy manure after 30 d of composting indicated that the average removal rates of E2α, E1, E2β, EE2, E3, and estradiol equivalent (EEQ) were 76.67%, 71.07%, 73.88%, 92.02%, 98.77%, and 88.11%, respectively, partly due to SE runoff. The rates of SEs leaching from the open-air composting dairy manure ranged from 0.05% to 4.75% after 10 rounds of 5–60 mm/d simulated rainfall. The leaching concentrations of E2α, E1, E2β, EE2, and E3 were 35.58 ng/L, 59.41 ng/L, 118.04 ng/L, 174.52 ng/L, and 141.27 ng/L, respectively. Although the proportion of SE loss was generally little, it can pose serious ecological risks, once manure leachate containing high concentrations of SEs enters the surface water environment. Due to the low solubility and hydrophobicity of SEs, they are mostly adsorbed on the surface of organic particles, and their loss is mainly due to the movement of tiny organic particles; thus, the highest concentration (1800 ng/L) value of EEQ in the drained liquid occurred at Round 1 of rainfall. The total leaching amount of SEs was positively correlated with rainfall, but the leaching concentrations of SEs were just the opposite. Derived from its role as a degradation intermediate of other SEs, E3 was the most prone to run off. By strengthening the action of microorganisms, the total leaching amount of EEQ increased by 5%, E3 increased by a factor of five, and E2β also underwent a transition from a conjugated to a free form. However, there were also fewer final SEs remaining in the composted product, as well as the environmental risks of conjugated SEs.

The proportion of each estrogen changed significantly during the composting process. The SEs in raw dairy manure were EE2(32%) > E3(22%) > E1(19%) > E2α(16%) > E2β(11%), and the SEs in the leaching liquid were E3(33%) > EE2(30%) > E2β(23%) > E1(8%) > E2α(6%), while the SEs after composting were E1(37%) > E2α(25%) > E2β(19%) > EE2(17%) > E3(2%). The removal of E2β was the most difficult due to the conversion of other SEs and conjugated E2β to it under anaerobic conditions. Therefore, it is necessary to improve the oxygenation conditions of manure composting. Considering the characteristics of SE loss from manure composting, we would like to make some recommendations. Annual first stormwater control may be critical for SE loss from open-air manure, and it is recommended to set leachate collection tanks around the manure pile or simple rain shelter to avoid impact on surface water. After the manure is decomposed and put into the farmland for culture, not only rainfall, but also irrigation methods, soil types, particle size, and organic matter content will have a significant impact on the loss of SEs. We know that agricultural non-point source pollution is driven by precipitation. At present, scholars engaged in agricultural non-point source pollution research are increasingly concerned about the impact of nitrogen and phosphorus loads carried by impulsive rainwater in the annual rainy season on the water environment of rivers and lakes. In fact, fecal-derived steroid estrogen is also an important part of agricultural non-point source pollution, and its negative impact is more difficult to eliminate. In the research on agricultural non-point source pollution, more research should be carried out in conjunction with the research on the migration and fate of nitrogen, phosphorus, and estrogen from manure and farmland.