Effect of Bellamya purificata on organic matter degradation in surface sediment as revealed by amino acids

The accumulation of organic matter (OM) in the sediment of aquaculture ponds is a potential threat for aquaculture ecosystems and the surrounding environment. The snail Bellamya purificata is a potential bioremediation species which may solve this problem. To investigate the effects of B. purificata on OM degradation in surface sediment, an 84 d experiment was carried out. The experimental setup entailed 6 glass tanks which were divided into B. purificata treatment (BPT) and control (CON). At the end of the experiment, a significantly lower degradation index (DI), reactivity index (RI), and carbon-normalized yield of amino acids were observed in BPT compared to CON, with mean ± SD values of −0.47 ± 0.43, 1.24 ± 0.01, and 6.24 ± 0.44, respectively. BPT showed higher oxidation−reduction potential and bacterial 16S rRNA gene copies than CON. Total organic carbon, total nitrogen, and total hydrolysable amino acid concentrations in the BPT treatment were 1.83 ± 0.10%, 0.07 ± 0.01%, and 22.38 ± 0.53 μmol g−1, respectively, all of which were significantly lower than in CON. A clustered heat-map of different indexes related to OM degradation in sediments showed the final BPT as one separate category, which was different from the initial samples and the final CON. Overall, B. purificata could effectively facilitate OM degradation by promoting oxidation−reduction potential and bacterial populations, and ultimately by inhibiting the OM accumulation in sediment. Our results therefore provide support for the application of B. purificata to reduce the risk of endogenous pollution caused by OM accumulation in aquaculture ponds.


INTRODUCTION
The most important component of China's freshwater aquaculture industry is pond aquaculture, the total production and area of which account for 73.05 and 47.12% of freshwater aquaculture, respectively (Ministry of Agriculture and Rural Affairs 2018). Traditional pond aquaculture is labour and resource intensive, with low resource utilization, and an increase in production and economic benefits can only be achieved by increasing stocking density and feeding ad libitum (Liu 2011, Shao et al. 2011. A low utilization rate of feed results in large amounts of residual organic materials in the aquaculture ponds, consisting mainly of uneaten food, faeces, and organic detritus. A portion of the residual organic materials is discharged with wastewater and causes nutrient pollution in the surrounding environment , while most of the residual organic materials accumulate on the sediment by settlement, complexation, and adsorption (Liu 2011, Shao et al. 2011).
In the aquaculture ecosystem, sediment, as a carrier for the migration, transformation, and accumulation of biogenic elements, is central as a sink for organic material and nutrient cycling. The accumulation of organic matter (OM) in sediments is a potential threat for aquaculture ecosystems and the surrounding environment (Diaz 2001, Sanz-Lázaro & Marín 2008. In aquaculture ponds, the sediment, which is an important source of biogenic elements, can release nutrients and toxic substances into the upper water body, thereby causing endogenous pollution and eutrophication of the aquaculture ecosystem (Christensen et al. 2000, Gray et al. 2002, Hargrave et al. 2008, Carlsson et al. 2012.
Benthic fauna, which are important components of the benthic environment in aquatic ecosystems, can change the physical structure and chemical properties of sediments, accelerate material exchange between the sediment−water interface, affect the biogeochemical processes at the sediment−water interface, and alter the structure and evolution of benthic communities through motion, ingestion, breathing, excretion, and burrowing (Aller 1988, Pelegrí & Blackburn 1994, Murray et al. 2002, Papaspyrou et al. 2006, Stockdale et al. 2009, Zhang et al. 2010. The snail Bellamya purificata is an important freshwater benthic organism, with wide distribution in China and high production (Liu 1979, Cao & Jiang 1998, Yan et al. 2018). Due to its nutritional value, B. purificata is not only one of the most important baits for various commercial aquatic animals (Dong et al. 2009, Tian et al. 2012, but is also an important aquatic product favoured by people that has increasingly gained attention (Liang et al. 2013, Yan et al. 2018. B. purificata is one of the most important benthic organisms in rivers and lakes in China, and prefers to inhabit silt and ingest organic debris and algae in its surroundings (Reavell 1980, Cao & Jiang 1998. Previous studies have shown that B. purificata can effectively improve the water quality in surrounding aqua culture (Chen et al. 2012, Zhao 2014, and it can provide important ecosystem services (Brönmark & Vermaat 1998, Gu et al. 2015. Hence, B. purificata is not only an im portant economical aquaculture species but also a potentially valuable species for ecological aquaculture.
Amino acids (AAs) are among the most labile fractions of bulk OM and can provide insight into the diagenetic alteration of OM , Dauwe & Middelburg 1998, Dauwe et al. 1999, Kaiser & Benner 2008. AAs are very useful indicators for evaluating the degradation state of particulate OM, dissolved OM, and sedimentary OM and have been widely applied in both marine and freshwater envi-ronments (Amon et al. 2001, Davis et al. 2009, Bourgoin & Tremblay 2010, Fernandes et al. 2014, Wang et al. 2018. However, to date, no quantitative studies have applied AAs as biomarkers to study OM degradation and recycling in an aquaculture pond ecosystem.
In this study, we used AAs to determine the degradation state of OM in sediment and to quantify the influence of B. purificata on OM degradation. We also aimed to provide evidence supporting the application of B. purificata to inhibit the accumulation of OM in pond sediment, thereby promoting the sustainable development of aquaculture.

Culture experiment
The experiment was conducted at the laboratory of the Freshwater Fisheries Research Center, Chinese Academy of Fishery Sciences. Bellamya purificata were collected from a local aquatic products trading market. The sediments were obtained from an aquaculture pond in the Dapu aquaculture base (Wuxi, Jiangsu, China). To maintain the consistency and homogeneity of the sediments in each glass tank, all sediments were dried in the sun, ground to a powder in a mortar to pass through a 100 µm mesh sieve, and mixed evenly before use (Nie et al. 2011, Wang et al. 2015, Hou et al. 2018. A 5 cm thick layer of sediment was spread on the bottom of each experimental glass tank (63 × 43 × 37 cm), and all tanks were filled with filtered tap water. The sediment in each tank was then decanted and stabilized for 1 wk prior to the experiment. Prior to the experiment, B. purificata were cultured in glass tanks with 5 cm of sediment for 14 d to acclimatize to the laboratory conditions. B. purificata treatment (BPT) and control (CON) tanks were set up in the experiment with 3 replicates each. After acclimation, 75 snails with initial weights of (mean ± SD) 3.01 ± 0.17 g were selected and randomly and equally distributed in the 3 glass tanks designated for BPT (25 ind. tank −1 ). The stocking density was approximately 277.78 g m −2 , which was based on the density of B. purificata in ecological ditches used for water purification, as reported by Duan (2013). The CON tanks contained no snails, but all other experimental conditions were strictly consistent with the BPT tank conditions. The experimental diets were commercial compound feed (Tongwei Group), which was ground and sieved through a 100 µm mesh. The total hydro lysable AA composition of the experimental diets is shown in Table 1. At 16:00 h every day, 6 portions of 1.5 g of feed (equal to 2% of the combined body weight of 25 snails) were weighed, and each portion was well mixed with fresh water and evenly poured into the tanks. The water temperature was maintained at 25.0 ± 0.5°C according to Cao & Jiang (1998). The experiment was conducted under natural sunlight. No water was exchanged during the experimental period. The experiment was conducted for 84 d (12 wk).

Sample collection
Prior to the experiment, 3 sediment samples were randomly collected as the initial samples after being dried, ground, screened, and well mixed. Sampling contributes substantially to the disturbance of the sediments; therefore, the remaining samples were collected at the end of the experiment. At the end of the experiment, 9 sampling points were randomly selected in each of the 6 tanks, and 0~1 cm surface sediment samples were collected with a plastic pipe measuring 2 cm in diameter. The sediment samples collected from the same glass tank were well mixed and then divided into 2 parts. One part of the sediment samples was stored in a freezer at −20°C for the quantification of bacteria by absolute fluorescence quantitative PCR. The other part was dried in a lyophilizer (CHRIST LYO Alpha 1-4 LD plus), ground to a powder in a mortar to pass through a 100 µm mesh sieve, and stored in a freezer at −80°C until further analysis of the hydrolysable AA compositions, total organic carbon (TOC), and total nitrogen (TN). Oxidation− reduction potential (ORP) was measured at 1 cm depth intervals at the beginning of the experiment, and then after 2, 4, 8, and 12 wk using a Pt electrode coupled with an Ag/AgCl reference electrode.

TOC and TN
To determine the TOC and TN concentrations, the homogenized and freeze-dried sediment samples were acidified with 1:1 HCl at room temperature for 24 h to remove inorganic carbon. The acidified sediment samples were then oven-dried to a constant weight at 50°C for 48 h. TOC and TN analyses were performed using a FlashEA 1112 Series NC Analyzer. The elemental analyses were calibrated by repeated inclusion of intra-laboratory acetanilide standard; the standard deviation of the repeated analyses was approximately 0.3%.

Degradation indicators
To quantify the sedimentary AAs, the homogenized and freeze-dried sediment sample (150 mg dry weight) was weighed, placed into ampoules, and hydrolysed with 8 ml of 6 mol l −1 HCl at 110°C for 24 h under N 2 atmosphere. After hydrolysis, the hydrolysate was transferred to a 50 ml centrifuge tube and diluted with Milli-Q ® water to a volume of 50 ml. Subsequently, the hydrolysis mixture was centrifuged at 10 000 × g for 10 min to remove all particles, and the hydro lysate was dried repeatedly under N 2 atmosphere to remove the HCl .
The degradation index (DI; Dauwe & Middelburg 1998, Dauwe et al. 1999) was calculated based on the changes in the relative molar percentages (mol%) of individual AAs in diagenesis. The DI considers changes in the entire suite of protein aceous AAs for calculation to determine the relative degradation state of OM (Dauwe & Middelburg 1998, Dauwe et al. 1999. Higher values of DI in dicate a lower degree of degradation. The DI was calculated as: ( 1) where var i is the original mol% of AA i , AVG var i is the mean of AA i in our data set, and STD var i is the standard deviation of AA i in our data set. The fac.coef. i is the factor score coefficient for AA i originating from Dauwe & Middelburg (1998, their Table 4A).
The reactivity index (RI; Jennerjahn & Ittekkot 1997) describes OM reactivity and is calculated based on the relationship between aromatic and non-protein AAs, and their contributions in different degradation states. Generally, an RI value close to 0 implies the ex ten sive degradation of OM, while fresh phytoplankton-derived OM has an RI value between 4 and 6 (Jennerjahn & Ittekkot 1997). The RI was calculated as: ( 2) where Tyr, Phe, β-Ala, and γ-Aba are expressed as the relative molar percentages (mol%).
The constructed plasmid was linearized, then purified, quantified, and converted to a copy number (copies µl −1 ). A 10-fold gradient dilution of the constructed standard was then made with 90 µl of the dilution solution + 10 µl of the plasmid. A 10 -2 to 10 -6 dilution of the standard was separately prepared by preliminary experiments to prepare a standard curve. The detection of fluorescence quantitative PCR (95°C for 3 min, followed by 30 cycles at 94°C for 30 s, 50°C for 30 s) reactions were performed in triplicate 30 µl mixtures containing 15 µl of qPCR Mix, 2 µl of Mg 2+ (25 mM), 0.5 µl of each primer (10 µM), 2 µl of template DNA, and 0.5 µl of dyestuff (Takara TB Green Fast qPCR Mix).

Statistical analysis
All data except the ORP values were subjected to an independent-samples t-test or 1-way ANOVA, followed by Duncan's test for multiple comparisons to determine the differences in each parameter between 2 or 3 sets of samples, respectively. The level of significance (α) was 0.05 (where p < 0.05 was considered significant). The ORP values were subjected to 2-way ANOVA with the main factors of B. purificata and measurement time, followed by Duncan's test for multiple comparisons at a significance level of 0.05.
To evaluate the effects of B. purificata bioturbation on the THAA composition and OM degradation in surface sediment, a heat-map was produced by integrating all of the different indexes related to OM degradation. Each of the indexes related to OM degradation was standardized by the min−max. We used TBtools (Chen et al. 2020) to construct the heat-map and to standardize the data. Prior to statistical analysis, raw data were assessed for normality of distribution and homogeneity of variance using Kolmogorov-Smirnov and Levene's tests, respectively (Zar 1999). Data are presented as means ± SD (n = 3). All statistical analyses were performed using SPSS for Windows (Release 22.0).

THAA compositions
The THAA composition of the experimental diet is shown in Table 1. No β-Ala or γ-Aba was detected in the experimental diet. The THAA concentration in the experimental diet was 2009.29 ± 61.52 µmol g −1 , approximately 100 times the THAA concentration (19.95 ± 1.70 µmol g −1 ) of the initial sediment samples.
The THAA composition of the surface sediment is shown in Table 2. The THAA concentration in the BPT was significantly lower than in CON tanks at the end of the experiment (p < 0.05), while no significant difference was observed between the final BPT and the initial value (p > 0.05). For Asp, Glu, Ser, His, Met, Ileu, Leu, Lys, and Pro concentrations, there were no significant differences between the final BPT, the final CON, and the initial values (p > 0.05). The initial Gly, Tyr, and Phe concentrations were significantly higher than the final BPT and CON (p < 0.05), while the initial Thr concentration was significantly lower (p < 0.05). The final BPT showed significantly lower Arg and Tyr concentration relative to the final CON and initial value (p < 0.05), and the final CON showed a significantly higher Arg concentration relative to the final BPT and initial value (p < 0.05). No significant differences in Gly, Thr, Cys, Val, and Phe were observed between the BPT and CON at the end of the experiment (p > 0.05).
Regarding the non-protein AAs, no β-Ala was detected in the initial sediment samples. The BPT showed significantly higher β-Ala and γ-Aba concentrations than CON at the end of the experiment (p < 0.05). The γ-Aba concentration in the final CON did not differ from the initial value (p > 0.05). Due to the absence of β-Ala in the initial sediment samples, the initial Asp/β-Ala ratio could not be calculated, and no significant differences in the Asp/β-Ala and Glu/γ-Aba ratios between groups were observed at the end of the experiment (p > 0.05).

Degradation indicators
The indicators of OM degradation status in the sediment are shown in Table 3. The BPT showed a significantly lower DI value than the CON at the end of the experiment (p < 0.05). The initial DI value was significantly higher than the final BPT (p < 0.05), while no significant differences in the DI between the initial samples and the final CON were observed (p > 0.05). Similar to the DI values, the RI value in the BPT was also significantly lower than the CON at the end of the experiment (p < 0.05). The initial RI value was significantly higher than the final BPT and the final CON (p < 0.05). The THAA-C% and THAA-N% values in the final CON were significantly higher than the initial value (p < 0.05), while no significant differences in the THAA-C% and THAA-N% values between the final BPT and the initial values were observed (p > 0.05).  Table 2. Total hydrolysable amino acid (THAA) composition of the surface sediment. Data are presented as mean ± SD (n = 3). Different letters within the same row indicate significant differences (p < 0.05) between the final Bellamya purificata treatment (BPT), the final control (CON), and the corresponding initial values. See Table 1 for abbreviations final BPT showed a significantly lower THAA-C% value than the final CON (p < 0.05). However, the THAA-N% in the final BPT did not differ from that in the final CON (p > 0.05).

ORP values
The sedimentary ORP values in the BPT and CON during the experimental period are shown in Fig. 1. The ORP values in sediment were significantly affected by Bellamya purificata and by time (p < 0.05), but not by their interaction (p > 0.05). Overall, the sediment ORP values were significantly higher in the BPT than in the CON (p < 0.05). A significant temporal decrease was observed in the se diment ORP values over the 12 wk experimental period (p < 0.05).

Bacterial 16S rRNA gene abundance
The melting curve peak of the sample is consistent with the target peak, and the quantitative values are all greater than 500. The quan titative value is accurate and re liable. Bac terial 16S rRNA gene abundance at the end of the experiment is shown in Table 4. At the end of the experiment, the copies of the bac terial 16S rRNA genes were significantly higher in the BPT than in the CON (p < 0.05).

TOC and TN concentrations
The TOC and TN concentrations in the surface sediment are shown in Table 4. At the end of the experiment, the BPT showed significantly lower TOC and TN concentrations relative to the CON (p < 0.05), while no significant differences in the TOC and TN concentrations between the final BPT and the initial values were observed (p > 0.05). The TN concentration of the CON at the end of the experiment was significantly higher than the initial value (p < 0.05).

Heat-map
Heat-map visualization of the different indexes related to OM degradation in sediments is shown in Fig. 2. All indexes were up-regulated in the final CON relative to the final BPT and the initial values, except for the DI and RI. The DI and RI were not consistent with the other indexes; DI and RI values were relatively higher in the initial sediment samples. In terms of the difference between the BPT and the CON, the CON had a higher value for each index, indicating less degraded OM compared to the BPT. By clustering rows in the heat-map, all samples in the experiment were initially divided into 2 categories, where Final CON_1, Final CON_2, and Final CON_3 belonged to one category, and Final BPT_1,

OM degradation state
In terms of the difference between the BPT and the CON at the end of the experiment, the THAA-C%, DI, and RI showed a high degree of consistency. For each indicator, the corresponding value in the BPT was significantly lower than that in the CON, which indicated more extensively degraded OM in the surface sediment of the BPT compared to the CON (Jennerjahn & Ittekkot 1997, Dauwe & Middelburg 1998, Dauwe et al. 1999, Zhou et al. 2018. Although the THAA-N% in the BPT did not significantly differ from the CON at the end of the experiment, the trend among the rows in Table 3 suggests that Bellamya purificata bioturbation promotes the degradation of sedimentary OM in the BPT , Zhou et al. 2018. When considering the initial values, the THAA-C%, THAA-N%, DI, and RI were not highly consistent, differing in the relative size relationship between the initial values and the corresponding final values in the BPT and the CON. Due to the high THAA concentration in the experimental diets, the continuous input of feed resulted in significantly higher THAA-C% and THAA-N% values in the CON at the end of the experiment than the initial values, which implied the accumulation of fresh OM in the sediment of the CON. However, the initial DI and RI values were significantly higher than the final values in the CON. The degradation states of the initial samples indicated by the DI and RI were inconsistent with the THAA-C% and THAA-N%. Similarly, Chen et al. (2018) also discovered that the DI and RI did not exhibit a consistent trend with the THAA-C%.
As demonstrated in previous studies, different degradation indicators have different sensitivities and applicability at different degradation stages or in different environmental media (Unger et al. 2005, Davis et al. 2009. For instance, Unger et al. (2005) found that the RI was a more sensitive indicator for degradation of suspended particulate OM in the water column, when tracing the degradation intensity from plankton-dominated suspended particulate OM to degraded sediments through combined application of DI and RI. The DI appeared to be effective to indicate degradation alteration of dissolved OM during intermediate stages and useful for the differentiation of degradation stages in sediments, but was not reliable for indicating OM degradation in the early degradation stages (Unger et al. 2005, Davis et al. 2009). The THAA-C% is considered the most sensitive indicator in early degradation stages of dissolved OM and particulate OM (Cowie & Hedges 1994, Davis et al. 2009), and the THAA-C% is also more sensitive for indicating early degradation in sediment compared to DI and RI . The present study was conducted for 84 d and mainly documented very early phases of diagenesis (days to months; Davis et al. 2009); therefore, the contradiction might imply that the DI and RI values were not as sensitive as the THAA-C% in the early stages of degradation in sediments .
Moreover, when indicating the degradation status of OM, the DI and RI calculated based on the AA 7 Fig. 2. Heat-map visualization of different indexes related to organic matter degradation in sediments of the final Bellamya purificata treatment (BPT), final control (CON), and initial samples (n = 3 replicates per group). For each row, colour represents the relative values of each index in different groups, from the highest (red) to the lowest (green). Note: β-Ala was not detected in the initial sediment samples, therefore the colour of the Asp/β-Ala ratio in the Initial_1, Initial_2, and Initial_3 is grey. DI: degradation index; RI: reactivity index; THAA-C% (THAA-N%): carbon-(nitrogen)normalized yield of total hydrolysable amino acids composition could be affected by the sources of AAs (Ingalls et al. 2003, Carstens & Schubert 2012, Bianchi et al. 2014, Zhu et al. 2014, Carr et al. 2016. Hence, the AA sources might be an important factor influencing the sensitivities and effectiveness of DI and RI when evaluating the OM degradation status in the present study, since the final samples in the BPT and CON were collected after weeks of continuous feed input relative to initial samples. The differences in the AA sources between the initial samples and final samples are a non-negligible reason for the contradiction discussed above. Accordingly, the THAA-C% might be the most effective in indicating the OM degradation status in sediments. Although the sensitivity of each indicator was different, especially when assessing the degradation status of OM in initial samples, the enhanced OM degradation caused by B. purificata bioturbation was confirmed through the high consistency between the DI, RI, and THAA-C% in indicating the OM degradation status of the final BPT and CON.

THAA compositions
The THAAs comprised a major portion of TOC and TN in sediment, especially freshly produced OM (Zhou et al. 2018). A compositional change occurs in the sedimentary THAAs during the early stage of OM degradation, which can be attributed to an alteration resulting from the degradation and different reactivity of individual AAs (Bourgoin & Tremblay 2010). The relative contents of individual AAs can also be used to reflect the degradation of OM. In the present study, the THAA concentration in the final CON was significantly higher than the initial value and the final BPT, while no significant difference was observed between the initial value and the final BPT. B. purificata could effectively prevent the accumulation of THAAs resulting from the continuous input of feed, which was consistent with the degradation state influenced by bioturbation.
For the compositional changes in THAAs, it is worth noting that no significant differences in the Glu mol% between initial samples, final BPT, and final CON were observed. In the present study, the Glu mol% in the experimental diets was 20.26 ± 0.04, while the Glu mol% in the initial sediment samples was only 3.61 ± 1.04. With the continuous input of feed, no significant increased Glu content in final BPT or even final CON, compared with the initial samples, might indicate that Glu was easily degraded. In fact, Glu, Tyr, and Phe are the most labile species and are easily degraded . The significantly lowest Tyr content in the final BPT suggested enhanced degradation of Tyr by B. purificata. In addition, the significantly increased contents of non-protein AAs (β-Ala and γ-Aba) in the final BPT are also worth noting, since β-Ala and γ-Aba were absent in the feed. Non-protein AAs are thought to originate from diagenesis, accounting for only a negligible fraction of the THAAs in organisms; however, they are relatively abundant in sediment, especially β-Ala and γ-Aba (Whelan 1977, Cowie & Hedges 1994, Davis et al. 2009, Zhou et al. 2018. β-Ala and γ-Aba, as the products of Asp and Glu from a microbial decarboxylation reaction, will accumulate during OM degradation, and their concentrations can be used to infer the microbial processing and OM degradation state (Lee & Cronin 1982, Ittekkot et al. 1984, Dauwe & Middelburg 1998, Davis et al. 2009, Zhou et al. 2018. For the present study, the significantly highest contents of β-Ala and γ-Aba in the BPT suggested that the presence of B. purificata promoted the production of non-protein AAs and the degradation of OM in the surface sediment. Previous studies have used the Asp/β-Ala and Glu/γ-Aba ratios to further describe microbial processing and OM degradation, whereby lower ratios reflect greater microbial transformation and degradation (Ittekkot et al. 1984, Dauwe & Middelburg 1998, Wang et al. 2018. For this study, no significant differences were observed in the Asp/β-Ala and Glu/γ-Aba ratios among the groups, which was inconsistent with the results of the degradation state indicators discussed in Section 4.1. Wang et al. (2018) also reported that the Asp/β-Ala and Glu/γ-Aba ratios may not be very accurate and effective. In fact, decarboxylation may not be the principal reaction for Asp and Glu; in many cases, Asp and Glu do not produce β-Ala and γ-Aba during degradation (Cowie & Hedges 1994). Therefore, such ratios may not indicate degradation accurately, but they can still be used as a reference for determining the OM degradation state.

How bioturbation affects OM
Environmental factors that can affect the remineralization of OM in sediment are locally influenced by the macrofauna (Yazdani Foshtomi et al. 2015). Oxygen, as the principal electron acceptor, is a key com-ponent affecting the remineralization of OM. Oxic remineralization processes are the most efficient in promoting the net remineralization and reoxidation of bulk OM on short time scales, and even brief (~20% time) periodic exposure to O 2 may approximate some of the same results through alternate paths (Aller 1994). Previous studies have shown that water and solute transportation at the sediment− water interface are mainly affected by benthic activities (Aller 2001, Grenz et al. 2003, Meysman et al. 2006. The mechanical disturbance caused by the activity of the benthic fauna can increase the contact area between the sediment and the upper water body, transport oxygen into the sediment, and finally, increase the supply of electron acceptors in the sediment (Massin 1982a,b, van de Bund et al. 1994, Sanz-Lázaro & Marín 2011. The significantly higher ORP value in the BPT than the CON in the present study indicated that B. purificata could effectively transport oxygen from the oxygenated overlying water into the surface sediment, which is consistent with the higher ORP values driven by worms (Lumbrinereis latreilli) reported by Casado-Coy et al. (2020). Hence, the more extensive degradation of OM in the BPT could be attributed to the bioturbation of B. purificata improving the ORP values.
However, OM degradation in sediment is essentially a microbial process. Previous studies have shown that bioturbation can not only affect the metabolism and activity of bacteria (Rugenski et al. 2012), but can also alter the structure and diversity of bacterial communities in sediment (Mermillod-Blondin et al. 2004, Laverock et al. 2010. Both are closely related to biogeochemical processes in sediment (Lohrer et al. 2004, Bertics & Ziebis 2009), including OM degradation. In the present study, BPT showed significantly higher copies of bacterial 16S rRNA genes relative to the CON, which suggests the influence of B. purificata on the bacterial population. Macrofaunal bioturbation can convert large organic particles into small organic particles, which can be more easily utilized by microorganisms, thereby improving the microbial OM availability (Levin et al. 1997, Kristensen & Holmer 2001, Kristensen et al. 2012). The significant differences in the bacterial 16S rRNA gene abundance between the BPT and the CON could be attributed to the more easily utilized OM that resulted from bioturbation. Furthermore, the ORP values in the present study indicated that the macrofauna could enhance microbial respiration by increasing the supply of high-energy electron acceptors (O 2 ), which promote OM degradation (Jørgensen 1989, Aller & Aller 1998, Christensen et al. 2000. Accordingly, the enhanced OM degradation by B. purificata might be attributable to B. purificata enhancing the bacterial population and respiration by increasing the availability of electron acceptors (O 2 ) and OM. The significant differences in TOC and TN contents between the BPT and the CON in the present study were consistent with the OM degradation states discussed above. In fact, B. purificata prefers to inhabit silt and ingest organic debris in its surroundings (Reavell 1980, Cao & Jiang 1998. Ingestion and digestion by B. purificata were also important contributors to the reduction of TOC and TN in the sediment and to the changes in OM and AA composition. Moreover, the faeces from macrobenthos also have a non-negligible impact on OM concentrations in the sediment. Brown (1986) found that the faecal pellets derived from the mud snail Hydrobia ulvae (Pennant) and bivalve molluscs contained higher organic carbon content than the sediment. Similarly, the shrimp Fenneropenaeus chinensis and sandprawn Callia nassa kraussi could also increase the OM concentrations in sediment through faecal production (Pillay et al. 2011, Ren 2012. Therefore, considering the consumption, production of faeces, and bioturbation, this study mainly aimed to investigate the overall impact of B. purificata on the OM in sediments. Through the mutual consistence and verification between the degradation state indicators, THAA composition, and OM contents (TOC and TN), it is confirmed that B. purificata can accelerate OM degradation, thereby inhibiting the accumulation of OM in the surface sediment.
In conclusion, during the 84 d culturing experiment, the presence of B. purificata effectively increased the ORP values of the sediments, providing sufficient high-energy electron acceptors (O 2 ), im proved the bacterial population and respiration, subsequently promoted OM degradation, and finally inhibited the accumulation of OM in the sediment. Hence, B. purificata played an im portant role in OM removal and improved resource utilization in sediment. The application of B. puri ficata in integrated multi-trophic aquaculture systems may therefore prevent eutrophication in aqua culture waters and reduce the risk of OM accumulation in sediment. In addition, the degradation indicators based on the AAs can help to further study the ecological systems of aquaculture ponds; however, considering the differences in sensitivity and applicability among the different indicators, comprehensive use and analysis are necessary to avoid errors.