Marked variations in gut microbiota and some innate immune responses of fresh water crayfish, marron (Cherax cainii, Austin 2002) fed dietary supplementation of Clostridium butyricum

Experiment set-up

A total of 24 marron with a mean weight of 69.65 ± 1.04 g were transported alive from Blue Ridge Marron, Manjimup, (34.2019°S, 116.0170°E) to Curtin Aquatic Research Laboratory (CARL), Bentley campus, Curtin University, Western Australia. Marron were then randomly distributed into six different tanks (150 L capacity) and acclimatized for two weeks before starting the feeding trial. During acclimatization, marron were fed with commercial basal diet named marron pellet (Glenn Forest, Perth, Australia), composed of 28% crude protein, 9% crude fat, 8.5% moisture, and 5% crude ash. After acclimatization, marron were disseminated into two distinct treatment groups’ viz. control and probiotic fed group with a stocking density of four marron per tank. Constant aeration to each tank was distributed by air stones (Aqua One, Perth, Australia) and fixed temperature of 22 ± 0.5 °C was maintained using submersible thermostat set (Aqua One, Perth, Australia). Each marron were individually nurtured in a special cage prepared of plastic mesh (0.8–8.0 mm thickness) to avoid cannibalism as described previously (Ambas, Suriawan & Fotedar, 2013). Approximately 30% of water was exchanged every day from each tank to remove uneaten feed and faecal debris by siphoning.

Probiotic feed formulation

The probiotic feed was formulated at CARL according to the procedure used by Ambas, Suriawan & Fotedar (2013). The fishmeal based basal diet was used for probiotic feed formulation (Table 1). The ingredients of the commercial basal feed were gently passed through 100 µm mesh sieve and homogenized to get uniform particle size. C. butyricum purchased from Advanced Orthomolecular Research (AOR, Calgary, Canada) was cultured on Clostridial Agar (Sigma-Aldrich, MO, USA) followed by sub-culture on Clostridial agar. Then a serial dilution and subsequent plate counts were performed to obtain the desire bacterial concentration. The bacteria (C. butyricum) were then added at 10⁷ CFU/mL per kg of feed as described in previous studies (Ramírez et al., 2017) and mixed uniformly. Pellet were prepared with mince mixture followed by vacuum drying oven at 37 °C for overnight and storage at 4 °C until used. Final proximate compositions and bacterial concentration of diets were determined as per the method of Association of Official Analytical Chemists, AOAC (AOAC, 2006) and plate counts on Clostridial agar (Table 1). During experimental trial, each marron were fed once every day at 5 PM for 42 days at a rate of 1.5% of total marron biomass per tank (Ambas, Fotedar & Buller, 2017). Control group fed basal diet and probiotic group served C. butyricum supplemented diet.

Monitoring of water quality parameters

The temperature and pH of experimental tank water was measured using portable waterproof ^∘C/mV/pH meter (CyberScan pH 300; Eutech Instruments, Singapore). The levels of dissolved oxygen (DO) in water was monitored by digital DO meter (YSI55; Perth Scientific, Australia). The concentrations of nitrate (NO₃⁻) and nitrite (NO₂⁻) were checked using Hach DR/890 Colorimeter (Hach, Loveland, CO, USA) following the method of cadmium reduction and diazotization (Hoang et al., 2016). Phosphate (PO₄⁻) level in water was examined by ascorbic acid standard 4500-PE method, as described by Huong (Mai, Fotedar & Fewtrell, 2010). The concentrations of ammonia (NH₃) in water was monitored using ammonia test kit (Hach, Loveland, CO, USA).

Marron sampling

For analysis of haemolymph parameters (lysozyme, haemolymph osmolality and total haemocyte count) health indices (protein and energy in tail muscle), one randomly selected marron from each tank (N = 6) was selected. However, for DNA extraction and microbiome analysis, two randomly selected marron from each tank (N = 12) were scarified followed by careful separation of hindgut. The hindgut contents of two samples from each tank were homogenized and pooled together (N = 6), and then transferred into 1.5 mL Eppendorf. Finally, for immune gene expression, the whole intestine tissue sample from one randomly selected marron from each tank (N = 6) was used after challenge test. Sample from each marron was used for the haemolymph parameters (before sacrifice), biochemical assay (after sacrifice), and molecular analysis (microbiome and gene expression).

Analysis of growth and immune parameters

Before starting the experiment and at the end of the trial, the weight of each marron was recorded. The weight gain (WG), specific growth rate (SGR) and feed conversion ratio (FCR) were measured at the end of the experiment according to (Ambas, Fotedar & Buller, 2017). The number of moults in each tank was monitored carefully. For analysis of immune parameters like haemolymph osmolality (HO), lysozyme and total haemocyte counts (THC), the haemolymph was extracted carefully from the pericardial cavity of marron with a 0.5 mL syringe containing 0.1 mL pre-cooled anticoagulant (0.1%) glutaraldehyde in 0.2M sodium cacodylate, (pH 7.0 ± 0.2). The HO of marron was measured using Cryoscopic Osmometer-Osmomet 030 (Gonotec, Berlin, Germany) as described by (Sang & Fotedar, 2004). THC of the extracted haemolymph from each replicate of two treatment group was calculated under a hemocytometer (Nauabuer, Germany) with 100X magnification according to method described by Ambas, Fotedar & Buller (2017). The lysozyme activity of marron haemolymph in control and probiotic fed groups was measured using turbidimetric assay (Mai & Fotedar, 2018; Foysal et al., 2019). Briefly, 50 µL of hemolymph samples (anticoagulant added) were disseminated into 96-well microtiter-plate (Iwaki, Tokyo, Japan). After incubation for 15 min at room temperature, 50 µL of PBS (0.25 mg/mL) suspended Micrococcus lysodeiktikus (Sigma-Aldrich, St. Louis, MO, USA) solution was added into separate wells of the same plate. The absorbance of each well in the microtiter-plate was then monitored in a MS212 reader (Titertek Plus, Tecan, Grodig, Austria) at 450 nm wavelength according to manufacturer’s instructions. The assay was based on the rate of Micrococcus lysodeiktikus cell lysis in terms of absorbance change caused by lysozyme in the hemolymph. The measurements were taken at every 2 min intervals for 20 min (U/mL).

Analysis of biochemical composition

The protein and energy content in the tail muscle and fat content in the hepatopancreas were measured according to methods described by the Association of Official Analytical Chemists, AOAC international (AOAC, 2006). The percentage of crude protein in oven dried tail muscle was measured following Kjeldahl method (N ×6.25) in Kjeltec Auto 1030 analyser (Foss Tecator, Höganäs, Sweden) (Jim, Garamumhango & Musara, 2017). Fat content in hepatopancreas was analysed following Soxhlet ether extraction method using Soxtec System HT6 (Tecator, Höganäs, Sweden) (Jim, Garamumhango & Musara, 2017). The amount of total gross energy in the tail muscle was calculated using a bomb calorimeter (IKA, Heitersheim, Germany).

High throughput sequencing

At the end of the experiment, gut samples prepared as described in marron sampling were used for DNA extraction. Due to special role in digestion, absorption and immunity, hindgut was selected for 16S rRNA sequencing (Wang et al., 2018). The bacterial DNA was extracted using DNeasy Blood and Tissue Kit (Qiagen, Crawley, UK) following manufacturer’s instructions. Extracted DNA was checked for quantity in a NanoDrop Spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA) and diluted to 30 ng/µL as final concentration. Thirty cycles of PCR amplification were performed in a BioRad S100 Gradient Thermal Cycler (Bio-Rad Laboratories, Inc., Foster City, CA, USA) with V3–V4 sequencing primers (Part # 15044223 Rev. B) and Hot Start Taq 2X Master Mix (BioLab Inc., Lawrenceville, GA, USA). Amplified PCR products were separated by 1% agarose gel and visualized under a gel documentation system (FujiFilm LAS–4000 Image Analyzer, Boston Inc., Foster City, CA, USA). Secondary PCR was applied for the barcoding of 16S rRNA PCR amplicon of each sample according to the Illumina standard protocol (Part # 15044223 Rev. B). The samples were then sequenced up to 20,000 reads on an Illumina MiSeq platforms (Illumina Inc., San Diego, CA, USA) at Harry Perkins Institute of Medical Research, Western Australia, using a v3 kit (600 cycles, Part # MS-102-3003).

Gene expression analysis in challenged marron

Six genes, namely, interleukin 8, interleukin 10, interleukin 17F, cathepsin L, prophenoloxidase and cytosolic manganese superoxide dismutase (IL-8, IL-10, IL-17F, PcCTSL, proPO and cytMnSOD) that have been reported to be associated with innate immunity of fish and crayfish (Liu et al., 2013; Jiang et al., 2015; Dai et al., 2017; Miao et al., 2018) were selected. The primers used for the selected genes are listed in Table 2. Crayfish pathogen Vibrio mimicus was collected from the Department of Food and Agriculture, Western Australia. At the end of the feeding trial, each selected marron was injected with previously prepared 2 × 10⁸ CFU/mL stock solution of V. mimicus (Ambas, Suriawan & Fotedar, 2013). Fifty microliters of bacterial solution were injected through the base of the fifth thoracic leg of marron (Ambas, Suriawan & Fotedar, 2013). Control marron were injected with 50 µL of phosphate buffer saline (PBS). Injected marron were subjected to RNA extraction from intestine tissue after 24 h of bacterial challenge. For RNA extraction, intestine tissue samples from challenged marron were initially stored at −80 °C with RNA Later solution (Sigma-Aldrich, Munich, Germany). The samples were thawed, dried, homogenized, ground into fine powder, and finally pooled together according to groups prior to use for RNA extraction. Approximately, 5 mg of pooled tissue sample from each group was used for RNA extraction using RNeasy Mini Kit (Qiagen, Hilden, Germany). During extraction process, RNase free DNase-I (Qiagen, Hilden, Germany) was added according to manufacturer’s instructions for removing of DNA related impurities. The quality and quantity of the extracted RNA was checked in 1% gel electrophoresis and NanoDrop spectrophotometer 2000c (Thermo Fisher Scientific, Waltham, MA, USA). The cDNA library was prepared using Omnicript RT kit (Qiagen, Hilden, Germany) from 1 µg of RNA. Real-time PCR amplification was performed using PowerUp™ Cyber Green Master Mix (Thermo Scientific) with 7500 Real-Time PCR System (Applied Biosystems, Foster City, CA, USA) to measure the relative expression level of six selected genes. The relative expression pattern of each target gene in two different groups was calculated using the 2^−ΔΔCT method, after normalisation against β-actin reference gene.

Bioinformatics

Unless any modifications mentioned, the “pipelines” for downstream sequence analysis and bioinformatics were used with default parameters. After extraction, the initial quality of 16S rRNA sequences were checked in fastQC pipelines (Andrews, 2010). Sickle program was used for the trimming of quality reads (Joshi & Fass, 2011). Following trimming, reads having a length of less than 200 bp were discarded. MeFiT pipeline was used for the joining of overlapping pair-ends reads (Parikh et al., 2016). Filtering of chimeric sequences, de novo greedy clustering of 16S rRNA sequences into Operational Taxonomic Units (OTUs) at 97% similarity threshold and removal of singleton OTUs were conducted using micca otu (version 1.7.0) (Albanese et al., 2015). Taxonomic assignment of the representative OTUs was performed using mica classify against SILVA 132 database clustered at 97% identity (Quast et al., 2013). Multiple sequence alignment of the obtained OTUs was performed using PASTA algorithm (Mirarab et al., 2015). The rarefaction depth value was set at 4,672 and subsequent calculation of alpha and beta diversities was performed using QIIME pipeline (version 1.9.1) (Kuczynski et al., 2012). Briefly, the alpha diversity was calculated based on observed species and diversity indices (Shannon, Simpson, Alpha Fisher and Chao1) using vegan package in Rstudio (Oksanen et al., 2018). Non-parametric t-test for two samples was performed to compare the alpha diversity metrics between the control and probiotic-fed samples. Principle coordinates analysis (PCoA) was performed to visualize separation of samples using Bray-Curtis and jaccard abundance metrics. Non-parametric statistical test of the distance metric was done using ANOSIM (1,000 permutations). PCoA plots were generated using PhyloToAST software (Dabdoub et al., 2016). The assigned bacterial genus from two different feeding groups were plotted in Venn diagram, generated using FunRich tool (Benito-Martin & Peinado, 2015). Indicator bacterial genus in two different conditions was differentiated from rarefy 16S rRNA sequence data using Linear Discriminant Analysis Effect Size, LEfSe (University of Auckland, New Zealand) (Segata et al., 2011). For predicting the metabolic pathways through 16S rRNA sequences of two different fed groups, Piphillin pipeline (http://secondgenome.com/Piphillin) was applied with supports of KEGG (May, 2017 release), BioCyc 21.0 databases and LEfSe algorithm (Segata et al., 2011; Iwai et al., 2016). One way analysis of variance (ANOVA) was used to calculate any significant differences (P < 0.05) among the numerical data including growth (WG, SGR, FCR), biochemical compositions (protein, fat, energy), lysozyme, haemolymph osmolality and THC obtained from two different feeding conditions.

Calculations

Following calculations were used:

Weight gain, WG = [(mean final body weight − mean initial body weight)]/[mean initial body weight] (Ramírez et al., 2017)

Specific growth rate, SGR (% day) = [(ln mean final body weight − ln mean initial body weight)/number of days] ×100 (Ambas, Fotedar & Buller, 2017)

Feed conversion ratio, FCR = feed intake (dry weight)/weight gain (wet g) (Ramírez et al., 2017)

Haemolymph osmolality, HO  =3 × osmolality of the mix − 2 × osmolality of anticoagulant (Sang & Fotedar, 2004)

Total haemocyte counts, THC = (cells counted x dilution factor × 1,000)/volume of grid (0.1 mm³) (Ambas, Fotedar & Buller, 2017).

Water quality parameters

The water quality parameters were recorded within the normal range for optimum growth of marron according to Nugroho & Fotedar (2013). Temperature was maintained at 22.3 ± 0.014 − 22.4 ± 0.01 °C, pH was recorded within the range of 7.5 ± 0.01 − 7.6 ± 0.01, dissolved oxygen levels between 6.51 ± 0.01 and 6.65 ± 0.01, nitrate and nitrite concentration of 0.046 ± 0.01 − 0.049 ± 0.01 and 0.015 ± 0.01 − 0.018 ± 0.01 were recorded, phosphate concentration was found to be varied within the range of 0.36 ± 0.01 − 0.52 ± 0.02, and ammonia concentration of 0.20 ± 0.01 − 0.22 ± 0.01 was observed in water during the experiment.

Probiotic supplement enhanced the growth and haemolymph parameters

After 42 days, the marron in the probiotic fed tanks gained significant (P < 0.05) weight gain (WG) and specific growth rate (SGR) than the control (Table 3). A total of 11 marron were moulted in the probiotic fed tanks compared to three in the control tanks, shown significant (P < 0.05) increase of moult number in probiotic fed marron. Multiple regression analysis revealed a strong “Pearson” correlation (0.99) between the number of moults and weight gain (Fig. 1A). PCoA plot also revealed a complete separation of probiotic fed group based on moulting numbers and weight gain (Fig. 1B). Probiotic diet also exhibited significant (P < 0.05) impacts on two out of three haemolymph parameters, total haemocyte counts (THC) and lysozyme. Among two tail muscle parameters, significant (P < 0.05) increase in protein content was observed in probiotic fed marron while the variations were insignificant (P > 0.05) for total gross energy, and fat in the hepatopancreas in the marron fed probiotic feed (Table 2).

Probiotic diet modulated the microbial communities in the distal gut

The probiotic supplemented diet exhibited a distinct effect on gut microbial community of marron after feeding trial (Fig. 2). The 16S rRNA sequence from eight pooled hindgut samples generated 128,567 high quality reads after quality trimming and removing singletons, that were classified into 83 OTUs and 49 genus. Besides 18 shared genus in both groups (control and probiotic), the probiotic feed group had an additional 31 unique genus compared to four in control diet group, indicating higher species diversity (Fig. 3). At phylum level, control diet group dominated by the Proteobacteria (96.5%) while Fusobacteria (53.4%) and Proteobacteria (44.1%) shared 97.5% of OTUs in the probiotic fed group (Fig. 4A). At genus level, control group demonstrated more abundance for Aeromonas (47.1%) and Vibrio (31.8%) whereas in probiotic fed group, Hypnocyclicus (57.7%) was the most dominant bacteria, followed by Vibrio (13.1%) and Aeromonas (9.7%). A significant reduction of Vibrio (13.1%) and Aeromonas (9.7%) counts was noticed in the probiotic fed group at the end of the trial (Fig. 4B). Although the values for major diversity indices such as Shannon, Simpson, Fisher Alpha and Chao1 were significantly (P < 0.05) higher but more pronounced effects was observed only for Shannon (P < 0.001) and Simpson (P < 0.005) indices in the probiotic fed group (Table 4).

Probiotic diet modified the microbial lineages and metabolic pathways in the hindgut

Linear discriminant analysis effect size (LEfSe) based on Wisconsin non-parametric t-test at 0.05 level of significance revealed three bacterial genus namely, Illumatobacter, Clostridium and Cyanobium to be enriched with C. butyricum supplementation in the marron (Fig. 5A). No significant (P > 0.05) enrichment of bacteria was observed in the control group. Piphillin and LEfSe based metabolic pathways extracted from KEGG database predicted an increased pathway activities for carbohydrate metabolism and absorption, amino acid metabolism, catabolism of fatty acid, degradation of toxic and synthetic compounds and biosynthesis of secondary metabolites in the probiotic fed group. In contrast, the analysis predicted significantly higher galactose, hormone and vitamin metabolism activities in the control group (Fig. 5B).

Probiotic feed up-regulate the expression profile of immune responsive genes

The results of qRT-PCR for the selective genes associated with crayfish immune response post in-vivo challenge demonstrated significant (P < 0.05) up-regulation of pro-inflammatory cytokine (interleukin 17, IL17), anti-inflammatory cytokine (interleukin 10, IL10), cytosolic manganese superoxide dismutase (cytMnSOD) and prophenoloxidase (proPO). After 24 h of challenge, results revealed that the relative expression level of IL-17, IL-10, cytMnSOD and proPO were increased by 2.8, 2.5, 2.2 and 1.9 folds, respectively (Fig. 6). The expression level of cathepsin L (PcCTSL) and interleukin 8 (IL-8) genes were relatively static during post challenge test.

Survival rate

At 24 and 96 h post-challenge, all marron injected with V. mimicus and PBS remained alive and started to respond to the feed given on day 5. The PBS injected marron, however, remained actively responsive to the feed given earlier (day 3) than bacteria injected marron. No signs of diseases were observed in any of the injected marron.

Data availability

The raw FASTQ files are currently available in National Centre for Biotechnology Information (NCBI) BioProject under the accession number PRJNA515886.