Introduction

The Chinese mitten crab (Eriocheir sinensis) is a commercially important aquaculture species in China. In 2021, the cultural yield of E. sinensis reached 808,274 tons (Fishery Administration Bureau of the Ministry of Agriculture & Villages et al. 2022). Juvenile crabs are susceptible to stress caused by changes in ambient temperature, pH, and ammonia nitrogen concentration (Peng et al. 2019; Qi et al. 2021), and failure to prevent or act quickly to address these stressors may result in huge economic losses. Therefore, it is necessary to take measures to improve the physiological status of juvenile crabs. Currently, common methods to improve the immunity of aquatic organisms include the application of vaccines, prebiotics, and probiotics (Wan et al. 2022). In crustaceans, dietary administration of Fructooligosaccharide, Lactobacillus acidophilu, and Bacillus subtilis have positive effects on growth performance and immunity (Interaminense et al. 2018; Jia et al. 2019; Talpur et al. 2013).

Effective microorganisms (EMs) is a term used to describe a type of compound microbially active agent that is composed of yeast, photosynthetic bacteria, and lactic acid bacteria (Lee et al. 2008). EM is mainly used to regulate the aquatic environment and ecosystem, and numerous studies have confirmed that it is effective in water treatment (Wang et al. 2004; Zarina et al. 2013). To date, EM is mainly used to regulate water quality in aquaculture. The addition of EM to water can improve the ability of crustaceans to resist oxidative stress and enhance crabs’ non-specific immunity (Chu et al. 2021; Zhang et al. 2022). Showed that EM optimized the nutritional composition of the edible tissues of E. sinensis. However, the effect of supplemental EM on intestinal microbiota and the serum metabolism of E. sinensis is unknown.

The intestinal microbiota of aquatic animals is closely related to their health (Semova et al. 2012; Wang et al. 2018). Intestinal microbes are usually affected by the host’s genetics, environment, and diet at different developmental stages (Li et al. 2017; Wang et al. 2019; Wong and Rawls 2012). The main environmental factor is water quality (Wu et al. 2021), and diet can improve the intestinal microbiome by adding different nutrients, further affecting the host’s growth and health (Nikouli et al. 2021; Schmidt et al. 2017). Yoshii et al. (2019) reported that intestinal microbiota play an important role in host metabolism, growth, and immunity, and Sun et al. (2016) found that the correlation between intestinal microbiome communities and metabolic profiles can affect the growth and development of animals.

Reportedly, EM plays an important role in crustacean culture (Chu et al. 2021; Zhang et al. 2022), and we speculate that this effect may be related to intestinal microorganisms and metabolites. The purpose of this study was to test the hypothesis that EM affects the health of E. sinensis by changing the intestinal microbial composition and the serum metabolites.

Materials and methods

Experimental design

This experiment was conducted from 7 to 21 August, 2020, at the Freshwater Fisheries Research Center, CAFS (Wuxi, China). Juvenile crabs (average weight, 6.22 ± 0.09 g) were purchased from Suzhou Youhua Ecological Technology Co., Ltd. (Suzhou, China). Six tanks (length × width × height = 44.5 cm × 30 cm × 25 cm) each containing 10 individual juvenile crabs were filled with 10 L of dechlorinated water. After acclimation for 2 weeks, the six tanks were divided into two groups: the control group (CK) and the EM group. Crabs in the CK group were fed 3 g of a commercial diet (JH8653, Jiangsu Hipore Feed Co. LTD, Taizhou, Jiangsu, China), while crabs in the EM group were fed the commercial diet supplemented with 50 g/kg EM (Jiangsu Hengtai Environmental Protection Technology Development Co. Ltd., Wuxi, Jiangsu, China). Table S1 shows the nutrition composition of the commercial diet. Crabs were fed daily at 17:00 during the experiment, and the residual feed and excrement were removed daily before feeding.

Sampling

Samples were collected after EM treatment for 7 days. Crabs were fasted for 24 h before sampling and six crabs were randomly collected from each group. Crabs were anesthetized on ice before hemolymph was collected from the third pereopod. After sitting for 2 h at 4 °C, the hemolymph was centrifuged at 4000 rpm for 10 min, and the supernatant (serum) was collected and stored at − 80 °C. To analyze the intestinal microflora, the surfaces of crabs were sterilized with 70% ethanol. The intestines and hepatopancreas then were aseptically dissected from the musculature, frozen in liquid nitrogen, and stored at − 80 ℃.

Determination of digestive enzyme activities and antioxidant capacity

Intestinal or hepatopancreas samples were weighed and homogenized with saline (100 mg tissue per 900 µl of saline), and the supernatant was used for analysis. Activities of the digestive enzymes lipase (A054-2–1) and amylase (C016-1–1) were measured in intestinal tissues. Hepatopancreas antioxidant parameters, including catalase (CAT; A001-3–2), superoxide dismutase (SOD; A001-3), total antioxidant capacity (T-AOC; A015-2–1), and malondialdehyde (MDA; A003-1–2), were also measured. These indexes were determined using kits purchased from Nanjing Jiancheng Institute of Biotechnology (Nanjing, Jiangsu, China) (Lei et al. 2021).

Serum biochemical analysis

Serum alanine aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase (ALP), and lactic dehydrogenase (LDH) activities and glucose (GLU) content were measured using a Mindary BS-400 automatic biochemical analyzer (Shenzhen, China) and assay kits purchased from Shenzhen Mindary Bio-medical Electronics Co., Ltd. according to the manufacturer’s instructions (Chu et al. 2021).

Gut microbiota analysis

DNA extraction, PCR amplification, and sequencing

Total microbial DNA was extracted from intestinal contents using the E.Z.N.A.® Stool DNA Kit (Omega Bio-Tek, Norcross, GA, USA) according to the manufacturer’s instructions. The V3–V4 region of the bacterial 16S rDNA was amplified by PCR (95 °C for 5 min followed by 30 cycles at 95 °C for 30 s, 50 °C for 30 s, 72 °C for 60 s, and a final extension at 72 °C for 5 min) using primers 338F (5′-ACTCCTACGGGAGGCAGCA-3′) and 806R (5′-GGACTACHVGGGTWTCTAAT-3′). Sequencing adapters were added to the 5′ ends of primers for PCR amplification. The amplified products were purified, quantified, and homogenized to form a sequencing library. After quality inspection, the qualified library was paired-end (250 bp) sequenced using an Illumina HiSeq 2500 instrument (Biomarker Technologies, Beijing, China) according to standard protocols. We constructed 12 sequencing libraries, including six libraries for the CK group, and six libraries for the EM group.

Bioinformatics analyses

Raw reads were first filtered using Trimmomatic v 0.33 (Bolger et al. 2014) with the cutoff threshold for average base quality score set at 20 over a sliding window of 50 bases. Next, Cutadapt v 1.9.1 (Martin 2011) was used to remove adapter sequences with a 20% maximum allowed error rate and 80% minimum coverage. Pair-end reads were assembled using FLASH v 1.2.11 (Mago and Salzberg 2011) based on sequences that overlapped more than 10 bp and the default maximum allowed error rate. Finally, effective reads were obtained after the identification and removal of chimeric sequences using UCHIME v 8.1 (Knight 2011).

Quality of each library was evaluated by calculating the average length, GC content, Q20, Q30, and effective ratio. Effective reads at 97.0% similarity level were clustered to obtain operational taxonomic units (OTUs) using USEARCH v10.0 (Edgar 2013). The subsequent analyses were based on OTUs. Taxonomy classification was performed using Naive Bayesian classifier at 70% confidence with Silva as the reference database (Callahan et al. 2016). Unweighted UniFrac distance metrics were used to estimate alpha-diversity (Bolyen et al. 2019). To investigate differences in bacterial community abundance between two groups, t-tests were performed on species abundance values between groups using Metastats (White et al. 2009).

Nontargeted metabolomic analysis

The serum samples were added to three times their volume of methanol, fully shaken and homogenized, then centrifuged at 13,000 rpm for 20 min at 4 °C to obtain the supernatant. The supernatants were transferred to new polypropylene tubes and processed through a nitrogen-purging instrument. The dried samples were redissolved in 50 μL of methanol. After centrifugation, the supernatant was used for analysis following an untargeted metabolomics approach using a liquid chromatography-mass spectrometry system (Thermo Fisher Scientific, Waltham, MA, USA). The system was equipped with an Accucore C18 column (2.6 μm, 50 mm × 2.1 mm, Thermo Fisher Scientific). Separation was achieved using the procedure, i.e., 0 min, 60% B; 2 min, 60% B; 9 min, 100% B; 17 min, 100% B; and 18 min, 60% B at a flow rate of 50 μL/min, where B is methanol and A is aqueous formic acid (0.025% (v/v) formic acid). The mobile phase was washed and rebalanced after the process was completed, and the total running time was 25 min. The injection volume in mass spectrometry mode was 2 μL, and the samples were scanned in positive and negative ion modes. The ionization parameters were analyzed following the method described by Jia et al. (2021). Data were analyzed in Compound Discoverer 3.1 software (Thermo Fisher Scientific, Waltham, MA, USA). Specific methods of data processing and multivariate statistical analysis followed those of Song et al. (2020), including Pareto scale principal component analysis and orthogonal partial least squares discriminant analysis.

Statistical analysis

The data are presented as the mean ± standard deviation of measurements of six replicates for each group. Normal distribution and homogeneity of variance of data were tested with Shapiro–Wilk and Levene tests (α = 0.05), respectively. Two-tailed t-test was used to compare the results between different groups and P < 0.05 was considered to be statistically significant. All statistical analyses were performed using IBM SPSS Statistics for Windows, version 22.0. (IBM Corporation, Armonk, NY, USA).

Results

Digestive enzyme activities

Intestinal lipase and amylase activities did not differ significantly between the CK and EM groups (Fig. 1). These results showed that dietary EM supplementation had little effect on intestinal digestion.

Fig. 1
figure 1

Effects of EM on intestinal a lipase and b amylase activities of E. sinensis. Values are mean ± standard deviation (n = 6). No statistically significant differences were detected (P > 0.05)

Full size image

Serum biochemical parameters

Serum ALT and ALP activities in the EM group were significantly higher than those in the CK group (Table 1). In contrast, AST and LDH activities and GLU content in the serum of E. sinensis did not differ significantly between the two groups (Table 1).

Table 1 Serum biochemical parameters. Values are presented as mean ± standard deviation (n = 6). Significant differences are indicated by an asterisk (P < 0.05)
Full size table

Antioxidant capacity

The SOD and CAT activities and T-AOC in the hepatopancreas samples from the EM group were significantly lower than those of the CK group (Fig. 2a–c). However, the MDA content did not differ significantly between the two groups (Fig. 2d).

Fig. 2
figure 2

Effects of EM on a CAT, b SOD, c T-AOC, and d MDA in hepatopancreas of E. sinensis. Values are presented as mean ± standard deviation (n = 6). Statistically significant differences are indicated by an asterisk (P < 0.05)

Full size image

Gut microbiome composition

16S rDNA sequencing

In total, 849,040 effective reads were obtained from the 12 libraries. On average, 70,753 effective reads were obtained from each library, and the number of effective sequences accounted for 89% of the raw reads. The Shannon index tended to be flat, indicating that the amount of sequencing data was reasonable and the sequencing depth was sufficient. The rank abundance curve showed that the samples contained relatively high levels of biodiversity (Fig. S1).

Of the 1546 OTUs identified, 1394 were detected in the CK group, 1217 were found in the EM group, and 1065 were shared between the two groups (Fig. 3). The OTUs belonged to 2 kingdoms, 33 phyla, 79 classes, 176 orders, 296 families, 553 genera, and 452 species.

Fig. 3
figure 3

Venn diagram of OTUs in the intestinal microbiome in E. sinensis

Full size image

Microbial community composition

The microflora of the samples was analyzed by taking the average of the community composition ratios. At the phylum level, eight and six main phyla were detected in the CK and EM groups, respectively (> 1% mean relative abundance). Proteobacteria, Tenericutes, Firmicutes, Bacteroidetes, and Actinobacteria were the main gut flora in both groups. Proteobacteria was the most abundant phylum in the EM group, with a relative abundance of 31.09%, followed by Tenericutes (24.97%), Firmicutes (14.01%), Bacteroidetes (15.07%), and Actinobacteria (9.16%). In the CK group, the most abundant phylum was Proteobacteria (30.58%), followed by Tenericutes (22.06%), Firmicutes (16.31%), Bacteroidetes (13.49%), and Actinobacteria (10.01%) (Fig. 4a). At the genus level, the dominant flora in the EM group were Candidatus Bacilloplasma (18.81%), Acinetobacter (6.59%), and uncultured bacterium_f_Mycoplasmataceae (5.48%). In the CK group, Candidatus Bacilloplasma (21.18%), Cupriavidus (5.61%), and Ralstonia (4.81%) were the most abundant flora (Fig. 4b).

Fig. 4
figure 4

The microbiota composition at the phylum (a) and genus (b) level of the intestinal microbiome in E. sinensis

Full size image

Microbial diversity

Table 2 shows the estimates of community richness (ACE and Chao1) and diversity (Simpson and Shannon). No significant differences were found between the two groups.

Table 2 Alpha diversity 2was evaluated using the ACE, Chao1, Shannon, and Simpson indices. Values are presented as mean ± standard deviation (n = 6)
Full size table

Linear discriminant (LDA) effect size (LEfSe) analysis of the crabs’ gut bacteria at a default logarithmic LDA score of 3 was used to identify biomarkers with statistical differences between the two groups. In the CK group, Fusobacteriaceae, Desulfovibrio, and Morganella were biomarkers. In the EM group, Family_XII, Exiguobacterium Rhodobacteraceae, and Rhodobacterales were biomarkers (Fig. 5).

Fig. 5
figure 5

LEfSe analysis of different microbial taxa in E. sinensis

Full size image

Serum metabolome profile

The principal component analysis plot showed that the first two principal components accounted for 95.47% of the total variance and the sample points were separated (Fig. 6a). In the current study, the values of the parameters of the model R2X, R2Y, and Q2Y were 0.688, 0.998, and 0.662, respectively, indicating a good model fit and acceptable predictability (Fig. 6b).

Fig. 6
figure 6

Principal components analysis (PCA) (a) and orthogonal partial least squares-discriminant analysis (OPLS-DA) (b) scores of serum metabolites in the comparison of E. sinensis metabolome data between the CK and EM groups

Full size image

In total, 273 metabolites were identified in serum samples, and the abundance of 40 metabolites differed significantly between the CK and EM groups. Among them, 11 metabolites were downregulated (log2FC < 0) and five metabolites were upregulated (log2FC > 0) (Table 3). Upon the pathway enrichment analysis of differential metabolites, a total of 19 significantly enriched metabolic pathways were identified, including arachidonic acid metabolism, vascular smooth muscle contraction, and platelet activation (Figure S2).

Table 3 Differential metabolites in serum of E. sinensis from the CK and EM groups (n = 6)
Full size table

Correlation between gut microbiota and serum metabolites

To better understand the correlation between the different metabolites and gut microbiota of E. sinensis, we performed Spearman’s correlation analysis (Fig. 7). Spearman’s correlation analysis results revealed a significant association between some of the microbes and metabolites. For example, 2-hydroxybutanoic acid was positively associated with uncultured_bacterium_f_Mycoplasmataceae, Moraxellaceae, and Roseimarinus but negatively correlated with Dysgonomonas, Candidatus Bacilloplasma, Lactobacillaceae, and Acinetobacter (P < 0.05). Pyruvic acid was positively associated with uncultured_bacterium_f_Mycoplasmataceae and Moraxellaceae, and arachidonic acid was negatively correlated with Roseimarinus.

Fig. 7
figure 7

Correlation of different gut microbiota and metabolites. Different colors represent correlation level, red represents positive correlation, and blue represents negative correlation, *P < 0.05; **P < 0.01

Full size image

Discussion

Effects of EM on serum biochemical parameters

Serum biochemical parameters are regarded as indicators of physiological disorders resulting from stress. ALT is an important amino transferase that plays an important role in protein metabolism, and it also is an indicator of liver function (Peng et al. 2018; Xie et al. 2021). ALP is a constituent of lysosomes and is directly involved in the transfer of phosphate groups and the metabolism of phosphate esters. It also promotes the removal of foreign bodies and is an important detoxification enzyme in animals (Chu et al. 2021). In this study, ALT and ALP activities were significantly higher in the EM group than in the CK group. After adding EM, the activity of ALP in the cells was stimulated to increase. It was released into the serum so that the activity of ALP in the serum was also enhanced, promoting the metabolism of substances in the cells and thereby maintaining the healthy physiological state of E. sinensis.

Effects of EM on antioxidant capacity

In animal cells, the production and removal of reactive oxygen species (ROS) are maintained in a balanced state under stable living conditions or without severe stress (Wan et al. 2022). When crabs suffer from oxidative stress, harmful products such as superoxide, hydroxyl radical, peroxy radical, and hydrogen peroxide appear (Cheng et al. 2021; Schock et al. 2010). The antioxidant system includes SOD and CAT, which clear ROS to protect the host from oxidative damage (Fu et al. 2017).

In this study, the SOD and CAT activities and T-AOC of the CK group were significantly higher than those of the EM group, possibly because the EM group had a benign environment for crab growth (Dobrzynski et al. 2022; Li et al. 2020; Pujiastuti and Suwartha 2017). EM may diffuse into the water either through direct dissolution or indirect excretion by crabs. Wang et al. (2021) reported that EM reduced the total nitrogen and total phosphorus content of aquaculture ponds, whereas E. sinensis in the CK group were vulnerable to oxidative stress and fluctuating antioxidant indexes. Thus, feeding E. sinensis with EM-supplemented diets may effectively enhance their resistance to oxidative stress.

Effects of EM on the intestinal microbiota

Intestinal microbes play an important role in the growth and development of their host animal (Wang et al. 2011). Wang et al. (2019) previously reported that Proteobacteria, Bacteroidetes, Firmicutes, and Tenericutes were the dominant phyla in intestinal samples from E. sinensis. Similarly, Proteobacteria and Tenericutes were the most abundant taxa in both the EM and CK groups in this study. Interestingly, although there was no difference in alpha diversity between the two groups, the species differences analyzed by LEfSe were statistically significant and several groups were distinctive (i.e., biomarkers) in the CK and EM groups.

Fusobacteriaceae, Morganella, and Desulfovibrio were the taxa that may serve as biomarkers of the CK group. Fusobacteriaceae are known to be positively correlated with inflammation. For example, fecal microbiota from people with cirrhosis and from inflammatory colorectal polyps from miniature dachshunds contained higher levels of Fusobacteriaceae compared with controls (Bajaj et al. 2012; Igarashi et al. 2016). Morganella is a human pathogen that is also lethal to the Mexican fruit fly and to some fish (Cifuentes et al. 2022; Reshma et al. 2018; Sontowski and van Dam 2020), and Desulfovibrio has been associated with mucosal inflammation (Earley et al. 2015). Although these microbes have been studied mainly in humans and less in crustaceans, we propose that they may also have negative effects on crustacean hosts. In the present study, the CK group contained potential pathogenic bacteria, indicating disruption of the microbiota balance in E. sinensis.

Potential biomarkers in the EM group included Exiguobacterium and Rhodobacteraceae. Manan et al. (2022) reported that Exiguobacterium could be used as a potential probiotic due to its function in the uptake and metabolism of nutrients, and Sombatjinda et al. (2014) found that it improved the growth and survival of the shrimp Penaeus vannamei. Rhodobacteraceae, the dominant group of gut microbiota in Litopenaeus vannamei, has been identified as an indicator of shrimp health (Gao et al. 2022a, b). In the current study, the levels of beneficial bacteria in the EM group were significantly higher than those in the CK group. EM is rich in beneficial bacteria, and it can control harmful bacteria and improve the immunity of animals (Laskowska et al. 2017; Li et al. 2022; Xing et al. 2007). Therefore, EM is beneficial to the health of E. sinensis because it improves the intestinal microbiome.

Effects of EM on the metabolites of E. sinensis

Compared to the CK group, the EM group had lower serum concentrations of valine and arachidonic acid. Valine is an essential amino acid required to assemble body protein. A higher release of valine in the serum may suggest degradation of structural proteins (Gillis and Ballantyne 1996). Yu et al. (2021) reported that reducing valine promotes metabolic health in mice. This indicated that the EM was beneficial to metabolic health of E. sinensis by decreasing protein consumption.

Recent studies showed that mannose can be used as a source of cellular energy and that has immune regulatory functions (Davis and Freeze 2001; Zhang et al. 2021). The mannose level was significantly higher in the EM group than in the CK group, indicating that the physiological function of E. sinensis could be kept stable by increasing cellular energy sources. Mannose also increased the Bacteroidetes to Firmicutes ratio in the gut microbiota in our study, which is a signature associated with the lean phenotype (Sharma et al. 2018). However, the relationship between the ratio and the health of crabs needs to be studied further. 2-Hydroxybutanoic acid is correlated with the synthesis of glutathione, which is a well-known antioxidative factor (Ma et al. 2017). The 2-hydroxybutanoic acid level in the EM group was significantly higher than that in the CK group, indicating that crabs with dietary EM supplementation had higher antioxidant ability than control crabs.

Because the CK group had lower levels of metabolites due to oxidative stress, it had a higher antioxidant index than the EM group. This is consistent with the SOD, CAT, and T-AOC data. Similar results were found in Macrobrachium nipponense (Sun et al. 2018) and L. vannamei (Parrilla-Taylor and Zenteno-Savín, 2011). Overall, the metabolomics analysis showed that feeding E. sinensis with an EM-supplemented diet was beneficial to crab health, mainly by increasing cellular energy sources and decreasing protein consumption, and oxidative stress.

Gut microbes may be associated with metabolites

With the emergence of the microbiota-gut-brain axis concept (i.e., the communication pathways that enable the interaction of intestinal microbiota with the central nervous system of the host), it has become accepted that the gut microbiota may play an important role in the development of disease. Many studies have found that changes in gut-brain axis interactions are related to neurodegenerative diseases, eating disorders, stress response, and behavioral changes (Quigley 2017; Ratsika et al. 2021; Sun et al. 2016; Wen et al. 2021). Intestinal microbiota interact with the host to produce metabolites, which act as intermediates or end-products of microbial metabolism (Dong et al. 2022). In the current study, certain microorganisms (e.g., Moraxellaceae, Lactobacillaceae, Acinetobacter) were correlated with serum metabolites (2-hydroxybutanoic acid, pyruvic acid, arachidonic acid). Therefore, we speculate that EM may affect the microbial composition of E. sinensis through the microbiota-gut-brain axis and then regulate the metabolites and ultimately affect the physiological state and function of the crabs.

Conclusion

Overall, the addition of EM to the diet of juvenile E. sinensis helped enhance the crabs’ resistance to oxidative stress and optimize their intestinal microbiome. Furthermore, metabolomic analysis revealed that EM contributes to E. sinensis health by increasing cellular energy sources and reducing protein consumption, and oxidative stress. However, further studies are needed to determine the optimal concentration of EM in the diet. The results of this study provide a scientific basis for understanding the effects of EM in E. sinensis aquaculture.