Effects of Salinity on Earthworms and the Product During Vermicomposting of Kitchen Wastes

Population growth and social changes have recently contributed to an exaggerated increase in kitchen wastes in China. Vermicomposting has recently been recognized as an effective and eco-friendly method of organic waste treatment through the combination of earthworms and microbes. However, the influence of salt in kitchen wastes on vermicomposting have been unknown. The goal of this study was to analyze the influence of different salinities on earthworms (Eisenia fetida) and the products during the vermicomposting of kitchen wastes. In our research, kitchen wastes were divided into four different salinities: 0% (A), 0.1% (B), 0.2% (C) and 0.3% (D). The chemical characters of substrates and earthworm growth were measured on the 14th day and the 28th day of composting. Our results show that the high salinity (measured >0.2%) prevented earthworms from properly growing and had negative effects on quality of products in composting. T2 (measured salinity = 0.2%) had the highest average body weight, nitrate nitrogen, and available phosphorus. Thus, the salinity of kitchen wastes should be pretreated to less than 0.2% before vermicomposting.


Introduction
Population growth and social changes have contributed to a great increase in kitchen waste, which has received public attention due to its quantity, odor and the potential for pathogenic microbial contamination [1,2]. Nearly 97 million tons of kitchen wastes were made in China in 2006 [3]. Kitchen wastes are rich in organic matter and water content. However, kitchen wastes have a low calorific value. Not only are they hard to collect and transport, but they can also lead to severe environmental problems if they are untreated before composting [4]. Meanwhile, a high salinity is associated with inhibitions and failures of fermentation processes [5]. For instance, when the salinity of a substrate is higher than 0.5%, its biodegradation efficiency will be significantly reduced in anaerobic digestion [6]. Moreover, Na + reduces the activity of microorganisms and interferes with their metabolism if the salinity is high [7]. The main methods to treat kitchen waste are landfill, incineration, anaerobic digestion and aerobic composting [2,8,9], which are generally expensive, multi-step, and easily result in secondary pollution.
Vermicomposting has been recognized as a method of organic waste treatment through the combination of earthworms and microbes [10,11]. It is broadly deemed as a simple, efficient and eco-friendly technique for converting organic waste materials to high nutritional and humified casts [12]. After vermicomposting, a more fine and homogeneous product is gained compared to classical composting [13]. Shak et al. decomposed rice and cow dung with earthworms for the bio-transformation of wastes into organic fertilizer [14]. Their results showed that the available phosphorus increased by 1.2-7.3% after 60 days. The total nitrogen, potassium and phosphorus (TOC: total organic carbon, TN: total nitrogen, NH 4 + -N: ammonium nitrogen, NO 3 − -N: nitrate nitrogen, TK: total potassium, AP: available phosphorus, TP: total phosphorus, AK: available potassium).
The earthworms (Eisenia fetida) were provided by Zhang Fei Company of Wuxi City, Jiangsu Province. We did not feed them for 48 h to avoid the effects of previous foods on this experiment.

The Vermicomposting of Feedstocks
We set four different salinities of kitchen waste in our experiment. A green-plastic pot with dimensions of 10 × 12 × 15 cm was used as the earthworm bin. Each basin was covered with a high-dense gauze and had small holes at the bottom to prevent earthworms from escaping and asphyxia from occurring (Figure 1a). 400 g of substrates (soil: kitchen wastes = 4:1) were put in each pot with 8 sexually matured earthworms. The kitchen wastes with 4 different salinities (manually added) of 0% (T1), 0.1% (T2), 0.2% (T3) and 0.3% (T4) were regarded as different groups. Each group was made 6 biological replications. The vermicomposting lasted for 28 days in an environment with 22 ± 1 • C and 65 ± 5% moisture content by injecting sterile water. The meat, vegetables and rice in the kitchen waste inherently contained salt in addition to what was exogenously added. The salinity was determined as follows: 0.1% (T1), 0.2% (T2), 0.27% (T3), and 0.39% (T4) ( Table 2). inherently contained salt in addition to what was exogenously added. The salinity was determined as follows: 0.1% (T1), 0.2% (T2), 0.27% (T3), and 0.39% (T4) ( Table 2). After 28 days of vermicomposting, all the earthworms except cocoons were taken out of the bin. We enumerated the earthworms one after another. After counting, the earthworms were washed in deionized water and weighed.

Chemical Analysis
The total organic carbon was measured by the potassium dichromate method [21]. The total nitrogen was determined by using Kay's method [22]. The total phosphorous and available phosphorous were determined by ammonium vanadium molybdate colorimetry. The total potassium and available potassium measurements were attained via a flame photometer [23]. NH4 + -N and NO3 --N were determined by an auto flow analyzer. Lipid was determined by acid titration (GB5009.229-2016). Salt was measured with silver nitrate titration (GB5009.44-2016).

Statistical Analysis
We reported the mean together with the standard deviation. The probability levels carried out for statistical significance were p < 0.05 (LSD) for the test by a one-way analysis of variance (ANOVA) using SPSS16.0. (IBM, Armonk, NY, USA). Each variable was determined for correlation (Spearman) [24] A redundancy analysis (RDA) was analyzed by the Canoco 5.0 software. (Microcomputer Power, Ithaca, NY, USA).

Growth of Earthworms
From the 14 th day to the 28 th day, the average body weight (total biomass/ total number) of E. fetida increased at first then decreased, except for T2. On the 28 th day, the average body weight  After 28 days of vermicomposting, all the earthworms except cocoons were taken out of the bin. We enumerated the earthworms one after another. After counting, the earthworms were washed in deionized water and weighed.

Chemical Analysis
The total organic carbon was measured by the potassium dichromate method [21]. The total nitrogen was determined by using Kay's method [22]. The total phosphorous and available phosphorous were determined by ammonium vanadium molybdate colorimetry. The total potassium and available potassium measurements were attained via a flame photometer [23]. NH 4 + -N and NO 3 − -N were determined by an auto flow analyzer. Lipid was determined by acid titration (GB5009.229-2016). Salt was measured with silver nitrate titration (GB5009.44-2016).

Statistical Analysis
We reported the mean together with the standard deviation. The probability levels carried out for statistical significance were p < 0.05 (LSD) for the test by a one-way analysis of variance (ANOVA) using SPSS16.0. (IBM, Armonk, NY, USA). Each variable was determined for correlation (Spearman) [24] A redundancy analysis (RDA) was analyzed by the Canoco 5.0 software. (Microcomputer Power, Ithaca, NY, USA).

Growth of Earthworms
From the 14th day to the 28th day, the average body weight (total biomass/total number) of E. fetida increased at first then decreased, except for T2. On the 28th day, the average body weight increased and then decreased when the salinity increased (T1-T4) ( Figure 2). In the end of the vermicomposting (28th day), the average body weight (abw) of Eisenia fetida was higher in T2 (0.58 ± 0.04 g) and was followed by T1 (0.54 ± 0.05g). These values were significantly higher (T1) or slightly higher (T2) than T3 (0.51 ± 0.02 g) and T4 (0.50 ± 0.01 g). Though the abw of E. fetida increased by 19% in T2, the growth rate of E. fetida decreased 93% from the 14th to the 28th day. Compared to the initial abw of E. fetida in T1, T3 and T4, the abw increased by 4% at first then decreased by 2% in T3. However, it continually reduced by 4% and 22%, respectively, in T1 and T4. increased and then decreased when the salinity increased (T1-T4) ( Figure 2). In the end of the vermicomposting (28 th day), the average body weight (abw) of Eisenia fetida was higher in T2 (0.58 ± 0.04 g) and was followed by T1 (0.54 ± 0.05g). These values were significantly higher (T1) or slightly higher (T2) than T3 (0.51 ± 0.02 g) and T4 (0.50 ± 0.01 g). Though the abw of E. fetida increased by 19% in T2, the growth rate of E. fetida decreased 93% from the 14 th to the 28 th day. Compared to the initial abw of E. fetida in T1, T3 and T4, the abw increased by 4% at first then decreased by 2% in T3. However, it continually reduced by 4% and 22%, respectively, in T1 and T4.

TOC and C/N Ratio
T1-T4 decomposed the total organic carbon (TOC) by 29%-56% after 28 days. The TOC of T3 and T4 were significantly higher than that of other two treatments (T1 and T2) (p < 0.05). The C/N ratio of substrates presented a drastic change during vermicomposting. The final decreases of the C/N ratio of T1-T4 were in the range of 58%-74% by the end of the vermicomposting. In addition, the highest decrease in the C/N ratio was in T2 which was 38% less than T4 on the 28 th day. Different letters indicate significant differences among treatments (one-way ANOVA, followed by the Turkey'st-test (p < 0.05)). Figure 3a shows the variation of the total nitrogen (TN) in T1-T4 during the whole period of composting. Though the TN in T2 was significantly higher than that in other treatments on the 14 th day, it raised around 71% among four treatments at the end of vermicomposting.

TN, TP and TK
The total phosphorus (TP) increased by 2%-8% in T1-T4 on the 28 th day (Figure 3b). Though no significant difference appeared among four treatments, the TP in T2 and T3 was higher (p > 0.05) than

TOC and C/N Ratio
T1-T4 decomposed the total organic carbon (TOC) by 29-56% after 28 days (Table 3). The TOC of T3 and T4 were significantly higher than that of other two treatments (T1 and T2) (p < 0.05). The C/N ratio of substrates presented a drastic change during vermicomposting. The final decreases of the C/N ratio of T1-T4 were in the range of 58-74% by the end of the vermicomposting. In addition, the highest decrease in the C/N ratio was in T2 which was 38% less than T4 on the 28th day. Table 3. The total organic carbon (TOC) and C/N ratio of substrates during the vermicomposting (mean ± SD, n = 3). Different letters indicate significant differences among treatments (one-way ANOVA, followed by the Turkey's t-test (p < 0.05)). Figure 3a shows the variation of the total nitrogen (TN) in T1-T4 during the whole period of composting. Though the TN in T2 was significantly higher than that in other treatments on the 14th day, it raised around 71% among four treatments at the end of vermicomposting. that in T1-T4. The total potassium (TK) decreased gradually from day 0 to day 14; however, there was minimal change from day 14 to day 28. Moreover, there was no significant difference (p > 0.05) among the four treatments (Figure 3c). The total phosphorus (TP) increased by 2-8% in T1-T4 on the 28th day (Figure 3b). Though no significant difference appeared among four treatments, the TP in T2 and T3 was higher (p > 0.05) than that in T1-T4. The total potassium (TK) decreased gradually from day 0 to day 14; however, there was minimal change from day 14 to day 28. Moreover, there was no significant difference (p > 0.05) among the four treatments (Figure 3c).

NH 4
+ -N, NO 3 − -N, AP and AK Data show that the NH 4 + -N of the substrates in T2 was significantly higher (p < 0.05) than others (Figure 4a). The NH 4 + -N of substrates in T1 and T2 increased 22-and 34-fold on the 14th day at first, and then it reduced by 33% and 28% on the 28th day, respectively, compared to the initial substrate (10.12 mg·kg −1 ). However, the NH 4 + -N of T3 and T4 increased continually by 372% and 406%, respectively, at the end of the experiment.

NH4 + -N, NO3 --N, AP and AK
Data show that the NH4 + -N of the substrates in T2 was significantly higher (p < 0.05) than others (Figure 4a). The NH4 + -N of substrates in T1 and T2 increased 22-and 34-fold on the 14 th day at first, and then it reduced by 33% and 28% on the 28 th day, respectively, compared to the initial substrate (10.12 mg·kg -1 ). However, the NH4 + -N of T3 and T4 increased continually by 372% and 406%, respectively, at the end of the experiment.
It was found that T2 had the highest available phosphorus (AP) among four different treatments (Figure 4c) (p < 0.05), whether on the 14 th or the 28 th day. There were 59%-110% increases among treatments T1-T4 compared to initial value (35.8 mg·kg -1 ).

Effects of Chemical Characteristics on the Growth of Earthworms
The relationship between the growth of E. fetida and the chemical characteristics of substrates was attained by a redundancy analysis (RDA) ( Figure 5). Our results show that first two sorting axis Eigenvalues could explain a total of 96% of the whole variables. It could be found that the C/N ratio and NH4 + -N had the most distinct influence (r = 0.96) ( Table 4). Moreover, Table 4 demonstrates that NO3 --N had a high co-linearity with the average body weight of E. fetida (r = 0.7). At the same time, NO3 --N and NH4 + -N were negatively correlated (r = −0.80).
RDA clearly divided our four treatments along the first sorting axis ( Figure 5). Samples 1-3, 4-6, 7-9, and 10-12 represented T1, T2, T3 and T4, respectively. In Figure 5 and Table 4, we could also find that the abw of earthworms presented a significant positive correlation with NO3 --N, but they presented a significant negative correlation with the C/N ratio, the TOC and NH4 + -N (p < 0.05). respectively, compared to the initial value on the 28th day. It was found that T2 had the highest available phosphorus (AP) among four different treatments (Figure 4c) (p < 0.05), whether on the 14th or the 28th day. There were 59-110% increases among treatments T1-T4 compared to initial value (35.8 mg·kg −1 ).
As shown in Figure 4d, the final increases of the available potassium (AK) of T1-T4 were in the range of 2.2-2.6-fold in the end of the vermicomposting, but there was no significant difference among them. The highest and lowest values appeared in T1 (702.1 ± 32.34 mg·kg −1 ) and T2 (601.83 ± 72.87 mg·kg −1 ), respectively, by the end of vermicomposting.

Effects of Chemical Characteristics on the Growth of Earthworms
The relationship between the growth of E. fetida and the chemical characteristics of substrates was attained by a redundancy analysis (RDA) ( Figure 5). Our results show that first two sorting axis Eigenvalues could explain a total of 96% of the whole variables. It could be found that the C/N ratio and NH 4 + -N had the most distinct influence (r = 0.96) (Table 4). Moreover, Table 4

Discussion
Data showed that the average body weight (abw) of E. fetida increased at first and then decreased as the salinity increased. It was the highest in T2 after 28 days. However, the growth rate of earthworms in T2 decreased after 14 days. The reason for the decreasing growth ability is likely that the salinity was too high for E. fetida to grow normally [25]. Boeuf et al. found that salt can improve the appetite of fish [19]. It is hypothesized that the salinity of kitchen waste less than 0.2% could also improve the appetite of earthworms. Furthermore, the permeation regulation of the earthworm body cannot maintain the balance of osmotic pressure between the body and the environment when in the presence of high salinity, resulting in cell water loss or expansion [26]. Earthworms would present  RDA clearly divided our four treatments along the first sorting axis ( Figure 5). Samples 1-3, 4-6, 7-9, and 10-12 represented T1, T2, T3 and T4, respectively. In Figure 5 and Table 4, we could also find that the abw of earthworms presented a significant positive correlation with NO 3 − -N, but they presented a significant negative correlation with the C/N ratio, the TOC and NH 4 + -N (p < 0.05).

Discussion
Data showed that the average body weight (abw) of E. fetida increased at first and then decreased as the salinity increased. It was the highest in T2 after 28 days. However, the growth rate of earthworms in T2 decreased after 14 days. The reason for the decreasing growth ability is likely that the salinity was too high for E. fetida to grow normally [25]. Boeuf et al. found that salt can improve the appetite of fish [19]. It is hypothesized that the salinity of kitchen waste less than 0.2% could also improve the appetite of earthworms. Furthermore, the permeation regulation of the earthworm body cannot maintain the balance of osmotic pressure between the body and the environment when in the presence of high salinity, resulting in cell water loss or expansion [26]. Earthworms would present stress responses such as restlessness and escapement under unsuitable salt conditions (salt concentration >0.2%).
The obvious reduction in the C/N ratio during vermicomposting might be due to the consumption of carbon through basic breathing and decomposition by guts [27]. Firstly, earthworms can add mucus, nitrogen-containing excretions, to substrates, which can increase the nitrogen [28]. Secondly, the TN can be fixed by free nitrogen-fixing microorganisms [29]. The C/N ratio is smaller with a higher fertility [30]. Meanwhile, salt can inhibit the degradation of organic matter [31] and the basic breathing of microbes has been shown to decrease with the increasing of salinity [32]. Moreover, when salinity is less than 0.2%, it can increase earthworm appetite to lead to more kitchen waste consumption rather than affecting the permeation pressure of earthworms. Thus, the lowest C/N ratio and the most abundant nutrients appeared in T2. In addition, the pH (6.58-7.46) of all the treatments was suitable for E. fetida. This suggests that the salt mainly affects the feeding behaviors and life activities of E. fetida. found that the excretion rate of NH 4 + -N of fish increased gradually with the increasing of salinity [18].
This might be due to the increase in activity of ammonifying and nitrifying bacteria during the decomposition of kitchen wastes [33]. First, earthworms decompose the organic materials, leading to the formation of NH 4 + -N under the action of microorganisms [34,35]. Second, NH 4 + -N is converted to NO 3 − -N (in the form of nitrate) by nitrification [36]. As for our research, when the salinity was less than 0.2%, NH 4 + -N increased at first then decreased. However, when the salinity was higher than 0.2%, NH 4 + -N gradually increased during the vermicomposting. Salt is the most important factor in determining the global patterns of microbial distribution [37]. When the salinity was lower than 0.2%, it could stimulate the biological excretion of NH 4 + -N. Nevertheless, when salinity is higher, the microbe's cells lose water to reduce the conversion of ammonia nitrogen and nitrate nitrogen pressure [17]. After the decomposition of kitchen wastes, available phosphorus has been found to increase in all treatments [38,39]. The increase in the AP might be due to the presence of large amount of microbes in the gut of earthworms and enzymatic reactions in the process of vermicomposting [40]. Adi and Noor [41] found that the P-mineralization might be due to the acid phosphatase enzyme that is present in one and a half months of vermicomposting [42]. The activity of gut phosphatases in E. fetida has been found to contribute to the mineralization and mobilization of phosphorus [43]. The fastest growth of earthworms in T2 (measured salinity: 0.2%) might have been due to the highest intestinal phosphatase activities of the earthworms. Salt plays an important role in enzyme reaction, maintaining biofilm balance and regulating permeation pressure during the growth of microorganisms. However, high salt concentrations could also inhibit enzyme catalysis. The activity of most soil redox enzymes significantly decreased with the increasing of salinity [43,44]. Therefore, when salinity was more than 0.2%, as the salinity increased, the AP was significantly reduced (p < 0.05).

Conclusions
The kitchen waste with 0.2% measured salinity (T2) had the highest average body weight of earthworms and nutrients (NO 3 − -N and AP) (p < 0.05). This may have happened because: (1) Proper salinity (<0.2%) can improve the appetite of earthworms; (2) the permeation regulation of the earthworm body cannot maintain the balance of osmotic pressure between the body and the environment in high salinity condition, resulting in cell water loss or expansion so as to not survive normally; (3) a high salinity (>0.2%) can lead to cell dehydration due to an imbalance in osmotic pressure. In conclusion, the salinity of kitchen wastes should be pretreated to less than 0.2% before vermicomposting.