The gut microbiome composition and degradation enzymes activity of black Amur bream (Megalobrama terminalis) in response to breeding migratory behavior

Abstract Black Amur bream (Megalobrama terminalis), a dominant species, resides in the Pearl River basin, known for its high plasticity in digestive ability. During spawning season, M. terminalis individuals with large body size and high fertility undergo a spawn migratory phase, while other smaller individuals prefer to settlement over migration. It is well known that gut microbial community often underpins the metabolic capability and regulates a wide variety of important functions in fish. However, little was known about how the gut microbiomes affect fish breeding migration. To investigate the variations in the gut microbiome of M. terminalis during the migration, we used high‐throughput 16S rRNA gene sequencing to reveal the distinct composition and diversity of the whole gut microbiome of migrated and nonmigrated population during period of peak reproduction, respectively. Our results indicated that nonmigrated population in estuary had a higher alpha diversity than that of migrated population in main stem. Additionally, an obvious abundant taxa shift between the gut microbiota community of nonmigrated and migrated M. terminalis was also observed. Change of dominant gut taxa from nonmigrated to migrated population was thought to be closely related to their degradation enzymes. Our results suggested that amino acid metabolism and lipid metabolism in migrated population were higher than that in nonmigrated population, providing a line of evidence for that M. terminalis change from partial herbivorous to partial carnivorous diet during breeding migration. We further concluded that, in order to digest foods of higher nutrition to supply energy to spawning migration, M. terminalis regulate activities of the gut microbiome and degradation enzymes, considered to be a key physiological strategy for reproduction.

. Almost half of all vertebrate species were fish, which encompass a wide spectrum of host habitats, physiologies, and ecological strategies (Parris et al., 2016). Fish gut microbiota can contribute to digestion and affect the growth, reproduction, overall population dynamics, and vulnerability to disease of host fish (Ghanbari et al., 2015;Parris et al., 2016). A few gut microbiota can even metabolize a remarkable variety of substrates, such as microbial species isolated from the intestinal tracts of herbivorous fish species identified as the cellulolytic enzyme-producing bacterial community (Li et al., 2016). In addition, food composition, physiological state, and environmental and genetic factors can affect gut microbial communities, thus affecting nutritional metabolism, immune regulation, and neuroendocrine system of the host . Fishelson et al. (1985) demonstrated that the distinct and diverse gut microbiota composition in herbivorous surgeon fish was closely associated with the trophic level of the host. Thus, either the diet category or the trophic level is the major factor that drives the composition and metabolism of the fish gut microbiota (Liu et al., 2016).
It is well recognized that the structure and composition of gut microbiota and their ecological function were strongly influenced by a range of factors, including the host genetics and physiology, living environment, and diet (Fishelson et al., 1985;Ley et al., 2008;Scott et al., 2013). The fish gut microbiota is closely related to the growth and development of fish (Ghanbari et al., 2015). To introduce how the conditions of the host fish and other factors affect the fish gut microbiota in return, Lin et al. (2014) found that the intestinal microbial community could be used as a method to reveal animal behavior. Currently, mice, macaques, chickens, earthworms, and termites have been used to successfully correlate the effects of intestinal microbial communities to host physiology (Drake et al., 2006;McKenna et al., 2008;Ohkuma & Brune, 2010;Torok et al., 2008;Zhang et al., 2009). Despite of the recognized important role of microbiomes in host ecology (Wong and Rawls, 2012), previous studies of fish gut microbiomes have focused primarily on commercial or model species host (Roeselers et al., 2011;Clements et al., 2014).
On the contrary, the connection between gut microbial composition and fish physiology has been reported in only a few fish species and needs for further research to elucidate completely.
The black Amur bream (Megalobrama terminalis) is a migratory fish species that lives in the middle and bottom of rivers, widely distributed in southern China drainages . It has been reported that the black Amur bream is one of the most important commercial fish species in the middle and lower reaches of the Pearl River, and the biomass of this species accounted for nearly 44.1% in all fish catches (Liu et al., 2020). Previous studies have shown that the M. terminalis growing and fattening in the estuary migrates nearly 250 km upstream from the estuary to the spawning grounds (Luopang and Qingpeitang) (Figure 1), where the water is lotic and the riverbed has an abundance of rocks. Liu et al. (2021) found that the abundance of spawning M. terminalis occurs from late June to mid-July. During migration, gonads of M. terminalis revived. In our previous study, we found that not all the M. terminalis individuals in estuary participated in breeding migration . Usually, for sexually mature individuals with large body size and high fertility, the cost of migration is less than of settlement; thus, the migration was deemed to be obviously "profitable" (Wysujack et al., 2010).
Migration has evolved repeatedly in animals, and many migratory fish reproduction strategies are found across the tree of life that increased migration efficiency (Burns & Bloom, 2020). Caudill et al. (2007 suggested that fish can adjust its own physiological behavior to meet the energy needs of breeding migration. Barneche et al. (2018) also found that fish body size determines total migration reproductive-energy output. Change of body length limit, swimming speed, osmotic pressure regulation, hormone, diet, sexual maturity, and gonad development which were regarded as a preparation process before fish migration (Ojima & Iwata, 2007;Saborido-Rey et al., 2004;Xia et al., 2017). Larger individuals reproduce disproportionately more than smaller individuals in not only fecundity but also total reproductive energy (Barneche et al., 2018;Burns & Bloom, 2020).
A recent study showed that there had been a shift in the diet of black Amur bream during gonad development period, which exhibits the different food preferences of both juveniles and adults (Xia et al., 2017). Furthermore, some studies have indicated that M. terminalis is an omnivorous fish with high plasticity in digestive ability (Liu et al., 2020;Xia et al., 2017). To date, most studies on black Amur bream have focused on larval resources, feeding habits, ecological investigation of spawning grounds, and gonad development (Li et al., 2014;Liu et al., 2019;Tan et al., 2009;Wang et al., 2010;Xia et al., 2017). However, the gut flora diversity and composition of the black Amur bream are poorly understood. Given the general close relation between the gut microbiome and the host, we hypothesized: (a) M. terminalis can change its gut microbiome communities during reproductive migration period; (b) some differences between microbiome composition of migrated and nonmigrated M. terminalis individuals exist. To demonstrate our hypothesis, we investigated that diversity of gut microorganisms of wild M. terminalis stock in fattening ground, migration routes, and spawning ground during period of peak reproduction to explain how they effectively ensure their digestion of food and disease resistance during reproductive migration. Additionally, due to the closure of the Changzhou Dam, wild M. terminalis stock has shown a decreasing trend Liu et al., 2021), gut floral communities of M. terminalis have been identified, which also provide theoretical support for its conservation biology.

| Fishing sampling
In this study, a total of 180 M. terminalis specimens were collected from six localities using circular cast nets (16 m diameter, mesh size  Table 1. Standard length (SL, to the nearest 1 mm): the length from the tip of the mandible to the base of the caudal fin; body weight (W t , to the nearest 1 g), eviscerated weight (EW, to the nearest 1 g), gonad and liver weights (GW and LW, respectively, to the nearest 0.01 g) were measured. The gonadal development stages of M. terminalis were categorized into immaturity and maturity based on gonad morphological characteristics, as described by Liu et al. (2019). The sex mature ratio (SMR = 100 × number of mature/total investigated quantity) and the gonadosomatic index (GSI = 100 × GW/EW) were estimated as an indicator of fish reproductive periods. The hepatosomatic index (HSI = 100 × LW/EW) and fatness (K = 100 × W t /SL 3 ) were regarded as bioenergetic indices to evaluate fish conditions. To investigate the fish gut flora, five fish were randomly selected for sequencing from six sample sites in two populations: nonmigrated (S1, S2, S3) and migrated population (M1, M2, M3). Selected fish were stunned and decapitated quickly.
The body surface was wiped to remove redundant mucus. All instruments, surfaces of each fish, were treated with 70% ethanol to ensure the skin surface, and instruments were sterilized before dissection. After opening the body cavity, the entire intestinal tract and its contents were aseptically removed from each individual fish.
Approximately 0.4 g of gut contents was extracted for DNA extraction, and 0.2 g of gut contents for enzymatic analysis; all the contents samples were then stored at −40°C until test.

| DNA extraction, amplification, and sequencing
Approximately 0.2 g of each sample was extracted using a QIAamp DNA Stool Mini Kit (Qiagen, Valencia, USA) following the manufacturer's recommendations. All DNA extracts were stored at −40°C until required. The quality and integrity of each DNA concentration and purity were monitored on 1% agarose gels. Total DNA from the gut of different groups of black Amur bream was sent to Novogene Bioinformatics Technology, Co., Ltd., Beijing, China for further sequencing analysis.
The 16S rRNA genes of distinct regions were amplified using the specific primer V4 hypervariable region (515F-806R) with the barcode. All PCR reactions were conducted in 30 μl reactions with 15 μl of High-Fidelity PCR Master Mix (New England Biolabs), 0.2 μM of forward and reverse primers, and approximately 10 ng template DNA.
Thermal cycling consisted of initial denaturation at 98°C for 1 min, followed by 30 cycles of denaturation at 98°C for 10 s, annealing at 50°C for 30 s, elongation at 72°C for 30 s, followed by 5 min at 72°C. The same volume of 1 × loading buffer (contained SYB green) with PCR products was used for electrophoresis on 2% agarose gel for detection. June 20 to 26 USA). Finally, the library was sequenced on an Ion S5 TM XL platform, and 400 bp/600 bp single-end reads were generated.

| Enzyme assays
The samples were removed from the freezer and placed on ice to thaw. After thawing was complete, the samples were homogenized with an F6/10 Fluko homogenizer at 12,000 g for 2 min on ice in 0.2 M NaCl (Gawlicka et al., 2000). The resulting homogenate was centrifuged with a cryogenic ultracentrifuge, and the supernatant was used to determine the digestive enzyme activities and soluble proteins. Bioengineering Institute, P.R. China). In the assay, 1, 2-laurelglycero l-3-glutaraldehyde-6′-methyl resorufin could be catalyzed by lipase, and chromophore variation was detected at λ = 580 nm.
The specific activity was expressed as milli-units per milligram of protein or units per milligram of protein (mU/mg protein or U/ mg protein). The soluble protein (mg/mL) contents were determined with the Bradford method using bovine serum albumin as the standard (0.563 g/L) (Bradford, 1976

| Data analysis
Paired-end reads were merged from the original DNA fragments using FLASH software and assigned to samples based on their unique barcode, and truncated by cutting off the barcode and primer sequence. Quality filtering of the raw reads was performed under specific filtering conditions to obtain high-quality clean reads according to the Cutadapt quality-controlled process (Martin, 2011).
The tags were compared with the reference database using the UCHIME algorithm to detect chimera sequences (Edgar et al., 2011;Haas et al., 2011). Sequence analysis was performed using Uparse software (Edgar, 2013). Sequences with equal or greater than 97% similarity were assigned to the same operational taxonomic unit (OTU). Representative sequences for each OTU were screened for further annotation. The shared and unique OTUs of different groups were also represented by a Scale-Venn diagram using Euler APE.
To compute the alpha diversity of fish gut microbiota, three metrics were calculated: chao1 estimates for species abundance; observed species estimates for the number of unique OTUs found in each sample, and the Shannon index. Rarefaction curves were generated based on these three metrics. QIIME calculates unweighted UniFrac, which is a phylogenetic measure of beta diversity. We used unweighted UniFrac for principal coordinate analysis (PCoA). PCoA

| Basic biological parameters of fish
Salinity showed a decreasing trend from nonmigrated to migrated population, salinity of S1 group was significant higher than that of M1, M2, and M3 group (p < .05) ( Table 1). The basic data for fish samples from the six sample sites are shown in Table 1. SL, W t , GSI, K, and SMR showed an increasing trend from nonmigrated to migrated population. SL, W t , and GSI of the black Amur bream in S1 and S2 groups were much lower than M1, M2, and M3 groups (p < .05).
Nevertheless, HSI was observed to first rise then descend from nonmigrated to migrated population. S3 group had a highest HSI among all the groups. Meanwhile, the rate of mature individuals increased gradually from the estuary to main stem (Table 1).

| Microbial complexity of fish gut flora
A total of 55,942 quality-filtered sequences were obtained from each sample. With a 97% consistency (Identity), the sequences were grouped into a total of 8,709 OTUs. For the six groups listed in Table 1, gut microbial complexity was evaluated based on alpha diversity (OTU number, species, Chao1, and Shannon index) and indicated significant variation (Table 1). S2 group had the maximum alpha diversity indices, followed by S1, S2, M1, and M2 groups.
M3 group exhibited the lowest alpha diversity indices. Our results showed that nonmigrated population in estuary had a higher alpha diversity than that of migrated population in main stem.

| Comparison of the bacterial community in fish gut
To compare the similarity of microbial community composition among different group, a PCoA was applied based on the  Table 2 revealed significant differences between different groups. The difference between nonmigrated and migrated population was significant (p < .01). However, the correlation between each pair of S1, S2, and S3 group in nonmigrated population is not significant (p > .05). The similar result was also observed in M1, M2 and M3 group in migrated population. In general, an obvious abundant taxa shift between the gut microbiota community of nonmigrated and migrated M. terminalis was also observed.

| Shared and unique microbial populations
To investigate the microbial community of gut samples from different groups, the shared and unique OTUs were constructed and visualized using a Venn diagram. A total of 5,790 and 4,261 OTUs were observed on nonmigration and migration population, respectively ( Figure 3a).

| Predicted gut microflora function using PICRUSt
PICRUSt analyses indicated that M. terminalis in different population exhibited similar gene functions at levels 2 and 3 ( Figure 5). The relative abundance of 21 genes for carbohydrate-related metabolism, amino acid metabolism, lipid metabolism, energy metabolism, and the endocrine system showed statistically significant differences (p < .05 by t test) between migrated and nonmigrated population, as shown in Figure 5. The relative abundance for amino acid metabolism, lipid metabolism, and endocrine system in migrated population was higher than that in nonmigrated population.

| degradation enzymes
As shown in Table 3, the gut content digestive enzyme activity differed markedly in six groups. Cellulase activity was much higher (p < .05) in S1 than in M1, but no obvious difference with S1 and S2 group. And there was no significant difference among groups in migrated population. The trypsin activity in S1 group was not significantly different to S2 group but was lower than in S3 group. On the contrary, the amylase activity in S3 group was observed to be no significantly different to S2, but much higher than that in M1 group (p < .05). The amylase activity in M1 group was higher than that in S1 and S2 group (p < .05), but was not significantly different to M2 and M3 group. As a whole, cellulase and amylase activity in the gut content of nonmigration population was significantly higher than in migration population, while trypsin and lipase activity in migration population was much higher than in nonmigration population (Table 3). The gut microbial composition of each group had a close relationship with their metabolic enzymes ( Figure 6). The gut microbiota composition of M1, M2, and M3 groups was more related to trypsin and lipase activity, and estranged from cellulase and amylase activities. In contrast, the gut microbial compositions of S1, S2, and S3 were correlated with cellulase and amylase activities.

| D ISCUSS I ON
In recent years, it has been revealed that the fish gut microbial community plays a critical role in the host digestive system and immune system, attracting increasing attention (Liu et al., 2016). A previous study reported that the intestinal food item of black Amur bream in the developmental process was a transition phase in relation to the dietary shift, which exhibited the food preferences during gonad development period. (Xia et al., 2017). However, to the best of our knowledge, very few scientific literatures have reported the difference of the gut microbial population diversity between nonmigrated and migrated M. terminalis during spawning migration period.  (Table 1). In recent studies of fish gut microbiota of many species, environmental factors have been shown to affect and shape the intestinal microbiota of fish (Sullam et al., 2012). Exogenous factors played important roles to affect the diversity of fish intestinal flora (Navarrete et al., 2012). The results indicated that M. terminalis population nonmigrated in estuary (fatten ground) had a higher alpha diversity than that of migrated population (Table 1)  are dominant in many carnivorous fishes (Kim et al., 2007;Liu et al., 2016;Merrifield et al., 2009). Ni et al. (2014) showed that Firmicutes was more advantageous than Proteobacteria in herbivorous grass carp. By fitting models of evolution, migratory species are evolving toward larger body size and higher fertility than nonmigratory species (Burns & Bloom, 2020). Studies indicated that not all individuals of migratory fish participated in the migration, and there were significant differences in sexual maturity, diet, and energy metabolism between migrated fish and nonmigrated fish (Secor, 1999). Meanwhile, migrated M. terminalis population in main stem had larger body size and higher gonadosomatic index than that of nonmigrated population (Table 1), which means that migrated population need more energy for spawning migration. As a result, it is necessary for migrated population to digest foods of higher nutritional value, which may be a key stimulus factor for the transition of gut microbiota community in M. terminalis to regulate the physiological capacity of digestion.
MetaStat results showed that abundance of Massilia, Fibrobacter in nonmigrated population was higher than that in migrated population, while abundance of Kosakonia, Sphingomonas, and Stenotrophomonas was much lower than that in migrated population ( Figure 4). Cellulase and amylase activity in the gut content of nonmigration population was significantly higher than in migration population, while trypsin and lipase activity in migration population was much higher than in nonmigration population (Table 3).
Our study revealed the four main enzyme activities which had a close relationship with the gut microbiome community ( Figure 6).
It has been reported that food digestion depends on the aid of gut symbiotic microorganisms to digest food and supply the energy to the host (Ray et al., 2012). Yokoe and Yasumasu (1964) mentioned that fish digested cellulose depends on the exogenous cellulose.
Fibrobacteracea was shown to be a vital part of the gut flora of the herbivorous fish, Ctenopharyngodon idelus (Jiang et al., 2011).
Gut cellulose activity is largely contributed by the gut microbiota. Meanwhile, research has revealed that Aeromonadaceae and Sphingomonadaceae occupied dominant positions in the freshwater carnivorous fish Oncorhynchus mykiss and Siniperca Chuatsi (Hovda et al., 2007;Kim et al., 2007;Merrifield et al., 2009).
PICRUSt functional analysis of gut microbiota showed greater fructose and mannose metabolism and lower amino acid metabolism in migrated population compared with nonmigrated one ( Figure 5). Lipid metabolism in migrated population was enhanced compared with that of nonmigrated population, seemingly explaining that the gut flora of M. terminalis was specialized for a diet shift and gonadal development during migration. This conclusion is in line with the results of research on gut microbiota and lipid metabolism in fish ). An interesting phenomenon indicated that the gut microbiota plays a role in the endocrine system of the black Amur bream. This suggests that the gut microbiota may alter metabolism and the estrogen cycle during migration.
Researches on mammals have shown that the significant variation F I G U R E 6 Canonical correspondence analysis (CCA) showed the correlation between the gut microbial compositions of the different group studied of Megalobrama terminalis and their enzyme activities of gut community was considered one of the factors affecting metabolism ability and sex hormone secretion changes (Goedert et al., 2018;Huang et al., 2020).
Above all, comparison of the microbiome composition of migrated and nonmigrated M. terminalis population indicated an obvious transformation, which is regarded as a critical factor response to gonadal development and dietary changes of M. terminalis during migration. Gut microbiota can be affected by multiple factors, for example, host diet, genetic, and environmental factors (Bolnick et al., 2014;Liu et al., 2016). The host developmental stage can influence individual gut floral diversity (Ley et al., 2008). Nevertheless, it is necessary to clarify how the bacterial coordination in the gut microbiota influences host behavior and physiology. Therefore, the ecology and physiology of the gut microbiota in fish should be an attractive field for future study.

CO N FLI C T O F I NTE R E S T
None declared.

E TH I C A L A PPROVA L
The methods involving animals in this study were conducted in accordance with the Laboratory Animal Management Principles of China. All experimental protocols were approved by the Ethics

Committee of the Pearl River Fisheries Research Institute, Chinese
Academy of Fishery Sciences.