Experimental protocol

The study was approved and performed in accordance with the guidelines of local ethics committee at the University of Debrecen (University of Debrecen Committee of Animal Welfare) under the registration number DEMAB/15/2019. A schematic view of the feeding trial is presented in Fig 1. Common carp juveniles aged 7 months (30 weeks) were used in this pilot study. Fish with an initial body weight of 123.45±0.37 g were obtained through artificial propagation and reared in a water recirculation system prior to the experiment. A total of 165 fish were equally distributed into five treatments groups 1 week before initiation of the feeding test for acclimation (acclimated animals; AA). After 7 days of acclimatization, the AA animals were further randomly sorted into five treatment groups, and the fish belonging to one of three replicates of each treatment group, which consisted of 11 individuals, was placed in each tank. The fish belonging to the negative control group received a basal diet (BD), whereas the fish in the treatment groups received one of the following supplements: BD+1% anthocyanins (ANTH) provided by sour cherry extract, BD+1% synbiotics (SYN) provided by fermented corn, BD+1% fermentable oligosaccharides (fOS) provided by sweet red pepper seed extract, and BD+1% carotenoids (CAR) provided by sweet red pepper pulp extract. The treatments were set up in a completely randomized design. The experiment lasted for 42 days.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Overview of the study protocol showing the feeding and sampling strategies.

AA stands for acclimatized animals aged 7 months (30 weeks). The fish were divided randomly into five experimental groups and fed either the commercial basal diet (BD, negative control) with no dietary supplement or the following dietary treatments as supplemented feed: BD+1% anthocyanins (ANTH) provided by Hungarian sour cherry extract, BD+1% synbiotics (SYN) provided by fermented corn, BD+1% fermentable oligosaccharides (fOS) provided by Hungarian sweet red pepper seed extract, and BD+1% carotenoids (CAR) provided by Hungarian sweet red pepper pulp extract. The fish intestines were collected on the 1^st, 14^th, 28^th and 42^nd days of the experimental period. The letter and number codes (D6-H7) refer to the code used during the 16S metagenomic sequencing library preparation.

https://doi.org/10.1371/journal.pone.0248537.g001

The experiment was performed in a water recirculation system provided with mechanical and aerated biofilters and UV lamps. The water volume of the circular plastic tanks was 350 L.

During the experiment, the oxygen saturation level was maintained at 85±0.9% by aeration stones, and the temperature was controlled at 23.51±0.5°C. The photoperiod consisted of 12 hours of light and 12 hours of darkness. The water temperature, pH, total dissolved solids (TDS, HANNA HI98130), dissolved oxygen (DO, HACH HQ30d), and NO₂^-, NO₃^- and NH₄⁺ concentrations (HACH DR3900) were checked daily. Any uneaten feed and faeces were removed daily.

Prior to the experiment, the basal diet was supplemented with four different feed additives at 1%. The composition of the experimental diets is provided in S1 Table. The fish were fed manually three times each day (0800, 1200 and 1600), and the feeding rates were equal to 3% of the total biomass.

The wet body weight (BW) of the common carp individuals was measured at the beginning and end of the feeding trial (Table 1). The survival of the fish was 100% during the experiment. The growth performance and feed conversion were estimated by calculating the weight gain (WG, %), specific growth rate (SGR, %) and feed conversion ratio (FCR) of the fish using the following formulas:

• WG (%) = (Wf–Wi)/Wi x 100%, where Wf is the final wet body weight (g) and Wi is the initial wet body weight (g);
• SGR (%) = (lnWf–lnWi)/t x 100, where Wf is the final wet body weight (g) and Wi is the wet initial wet body weight (g), t is the time (day); and
• FCR (g/g) = F/(Wf-Wi), where F is the feed intake, Wf is the final wet body weight (g) and Wi is the initial wet body weight (g).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Growth performance of common carp juveniles at the end of the feeding trial (means±SE, n = 3), and ABW at the beginning (_b), and end (_e) of experiment.

https://doi.org/10.1371/journal.pone.0248537.t001

Statistical analyses of the growth performance and feed conversion were performed using the SPSS/PC+ software package. The variance homogeneity was tested by Levene’s test, and a P value higher than 0.05 was considered to indicate homogeneity. The effects of the treatments on the BW, WG, SGR and FCR results were assessed by one-way ANOVA. The significance of the differences was determined using Tukey’s multiple comparison test, and P<0.05 was considered to indicate significance.

Determination of natural feed additives

The ANTH supplement was prepared as described by Nemes et al. [36]. Anthocyanins were extracted from Hungarian sour cherry (Prunus cerasus). Cherries were deseeded and homogenized, and a methanol:water:acetic acid solution at a ratio of 25:24:1 was then used to extract the anthocyanins. The sample was mixed with a magnetic stirrer (MSH 300, BioSan, Riga, Latvia) for 1 hour. After filtration, centrifugation was performed at 10,000 RPM for 5 minutes, and a simple fraction was then obtained in preconditioned tubes (Supelclean ENVI-18 SPE tubes). For preconditioning, 5 mL of MeOH and 5 mL of H₂O were mixed with 1 mL of the fruit sample. Elution was conducted with methanol containing 20% H₂O, and vaporization was performed at 40°C. The sample was dried in a vacuum to yield a powder. A VWR-Hitachi ChromasterUltraRs UHPLC system (Hitachi, Tokyo, Japan) with a Phenomenex Kinetex® column (2.6 μm, XB-C18, 100 A, 100 x 4.6 mm) (Phenomenex, Torrance, CA, USA) was used to determine the anthocyanin profile. Two solvents were applied for gradient elution: A, which consisted of MeOH, and B, which consisted of 3% formic acid. Elution was conducted using the following parameters: 0 minutes, 15% solvent A; 0–25 minutes, 30% solvent A; 25–30 minutes, 40% solvent A; and 30–40 minutes, 50% solvent A. UV-VIS detection was performed at 534 nm, the flow rate was maintained at 0.7 mL/minutes at 25°C, and the injection volume was 10 μL. The UHPLC profiles of the anthocyanins and the identified main compounds, including the relative areas and retention times, are shown in S1 Fig.

The synbiotics in the SYN supplement include probiotics (Bifidobacterium bifidum, B. infantis, B. lactis, B. longum, Lactobacillus acidophilus, L. buchneri, L. casei, L. paracasei, L. plantarum, L. salivarius, and L. lactis), prebiotics (fructo-, xylo-, and mannooligosaccharides and arabinogalactan), vitamins (B group vitamins and vitamins C, D2, D3, E and K2), unsaturated fatty acids (ω-3, ω-6, and ω-9), minerals/trace elements (sodium, potassium, calcium, iodine and phosphorous) and lactose. The GC profile of the oligosaccharides and the identified monomer units with the greatest relative areas and retention times are shown in S2 Fig.

The carotenoids in supplemental CAR were determined as described by Remenyik et al. [33] and Csernus et al. [37]. Carotenoids were extracted from Hungarian sweet red pepper (Capsicum annuum) powder (1–5 g) using a mixture of dichloroethane:acetone:methanol at a ratio of 2:2:1 as the solvent. The mixture was stirred in an ultrasonic water bath for 30 minutes and purified through Munktell-292 filter paper (VWR International, Debrecen, Hungary). For additional purification, a 0.22-μm PTFE syringe filter (TPP Techno Plastic Products AG, Switzerland) was applied. The filtered sample was vaporized at 40°C and 0.2 bar and then dissolved in an HPLC pigment reagent (isopropanol:acetonitrile:methanol at a ratio of 55:35:10) (Merck, Darmstadt, Germany). HPLC separation was conducted with a Phenomenex Kinetex® column (2.6 μm, XB-C18, 100 A, 100x4.6 mm) (Phenomenex, Torrance, CA, USA) using two elution gradients of elution: A, 11% methanol; and B, isopropanol:acetonitrile:methanol (55:35:10 V/V/V%) mixture. The gradient elutions were performed with the following settings: 0–3 minutes, 100% solvent A; 15–20 minutes, 20% solvent A; 25–45 minutes, 100% solvent B; and 48–50 minutes, 100% solvent A. For sample detection, a diode array detector (DAD) was used with a flow rate setting of 0.6 mL/minutes. The sample was injected in a 10-μL volume, and DAD detection was applied at 460 and 350 nm. The HPLC profile of the carotenoids and identified monomer units with the greatest relative areas and retention times are shown in S3 Fig.

The fermentable oligosaccharides in the fOS supplements was determined as described by Csernus et al. [37] Hungarian sweet red pepper seeds (Capsicum annuum) were used for the extraction of fermentable oligosaccharides with a high arabino-galactose content. An HP 5890 gas chromatograph with an SP-2380 capillary column (30 m x 0.25 mm, 0.2 μm) was used to measure the composition of oligosaccharides. The samples were lyophilized and extracted with trifluoracetic acid:acetic acid:water at a ratio of 5:75:20 as the solvent. A reduction step was used to convert the oligosaccharides into alditol-acetate with NaBH₄ at alkaline pH. The sugars were then converted to sugar alcohols (alditols) with acetic anhydride in pyridine, which removed interfering isomers and anomers. The nitrogen gas flow rate was 1.2 mL/min. The injector temperature was set to 300°C with a split ratio of 1:20. A flame ionization detector (FID) was applied for the identification of oligosaccharides. The GC profiles of the oligosaccharides and the identified monomer units, including the greatest relative areas and retention times, are shown in S4 Fig.

Sample collection

During the experimental period, intestinal samples were obtained from the fish at ages of 31 weeks (day 0), 33 weeks (14^th day), 35 weeks (28^th day), and 42 weeks (42^nd day). The average body weight (ABW) was measured at the beginning and end of the feeding period (Table 1). The fish were euthanized with clove oil [38] and disinfected with 96% EtOH, and the whole intestine was placed into a tube. During the gut collection procedure, sampling was also performed from the air and surgical table to monitor the environmental contamination, which can bias sequencing. The bowel was stored and transported on ice in sterile PBS (Thermo Fisher Scientific, MD, USA). Residual tissue was removed from the guts under class II laminar flow hood (Thermo Fisher Scientific, MD, USA), and a 10-g sample was digested in 100 mL of lysis buffer (500 mM NaCl, 50 mM Tris HCl, 50 mM EDTA, 4% SDS, 0.1 mg of Proteinase-K, and 10% Triton X) at 56°C for 16 hours.

Sample preparation and mechanical cell lyses

The lysate was pipetted into a 50-mL centrifuge tube. The samples were centrifuged at 2,000 x g for 5 minutes, and the upper colloidal phase was discarded. The middle phase was transferred to a new 50-mL centrifuge tube, and PBS was added to the lysate at a ratio of 1:1. The samples were shaken for approximately 3 minutes, the mixture was centrifuged at 500 x g for 5 minutes, and the supernatant was collected in a sterile 50-mL centrifuge tube. This step was repeated three times. We performed mechanical cell lysis: 1,000 μL of sample was added to a PowerBead tube (Qiagen, Hilden, Germany) and lysed using a MagNA Lyser (Roche Applied Sciences; Penzberg, Germany) at 5,000 RPM for 30 seconds. The samples were centrifuged at 16,000 x g for 1 minute, and the supernatant was pipetted into a sterile Eppendorf tube.

DNA extraction

The conventional isolation method was used for total bacterial genomic DNA extraction. Eight hundred microliters of phenol:chloroform:isoamyl alcohol (25:24:1) (Thermo Fisher Scientific, MD, USA) was mixed with 800 μL of lysate, and the mixture vortexed thoroughly for approximately 15 seconds. After sample homogenization, we incubated the samples at room temperature for 3 minutes and centrifuged them for 10 minutes at 16,000 x g and 4°C. The upper aqueous layer was carefully collected into a new sterile Eppendorf tube. For DNA precipitation, a mixture of 1 μL of glycogen (20 μg/μL), ammonium acetate (0.5x volume, 7.5 M NH₄OAc) and 100% ethanol (2.5x volume) was added to the supernatant. Samples were incubated at -70°C for 16 hours and then centrifuged them for 30 minutes at 16,000 x g and 4°C to pellet the DNA. We carefully discarded the supernatant without disturbing the pellet, and 500 μL of 70% EtOH was added to the sample. The mixture was shaken for 20 seconds and centrifuged at 4°C and 16,000 x g for 5 minutes, and the supernatant was carefully removed. This washing step was repeated twice. We dried the DNA pellet at room temperature and then resuspended it in 15 μL of nuclease-free water. The DNA concentrations were determined using a Qubit® Fluorometric Quantitation dsDNA assay kit (Thermo Fisher Scientific, Waltham, MA, USA) and a Clariostar microplate reader (BMG Labtech, Ortenberg, Germany). The DNA quality and quantity were confirmed using a Nanodrop 2000 Spectrophotometer (Thermo Fisher Scientific), and the DNA integrity (shearing/fragmentation) was measured with a 4200 TapeStation System (G2991AA, Agilent Technologies; Santa Clara, CA, USA). The DNA elutes were stored at –20°C.

Library construction and sequencing

Library preparation was performed according to the standard Illumina (San Diego, CA, USA) 16S metagenomic sequencing protocol (15044223 Rev. B). The V3 and V4 hypervariable regions of the bacterial 16S rRNA gene were targeted to generate amplicons of ~460 using the universal primer set 341F (5’ CCTACGGGNGGCWGCAG 3’) and 785R (5’ GACTACHVGGGTATCTAATCC 3’) flanked by Illumina overhang adapter sequences (forward overhang: 5’ TCGTCGGCAGCGTCAGATGTGTATAAGAGACAG 3’, reverse overhang: 5’ GTCTCGTGGGCTCGGAGATGTGTATAAGAGACAG 3’) (Sigma Aldrich, MO, USA). After completion of the amplicon PCR with 2x KAPA HiFi HotStart ReadyMix, dual indexing of the samples with adaptor sequences (i7-N7xx-12 items, i5-S5xx-8 items) was performed using the Illumina Nextera XT Index Kit (FC-131-1001/2). PCR clean-up and amplicon size selection were performed using with KAPA Pure Beads (KAPA Biosystems) based on the technical data sheet (KR1245 –v3.16) provided by the manufacturer, which resulted in final ~580-630-bp libraries. For each case, verifications were performed with PCR Agilent D1000 screen tapes (5067–5582) and D1000 reagents (5067–5583). The 16S amplicon libraries for each sample were quantified by qPCR, normalized with respect to the amplicon sizes and pooled into a single library at equal molar quantities. Finally, 5 μL of the 4 nM DNA library pool was prepared for sequencing on the Illumina MiSeq platform. The library pool was denatured with 0.2 M NaOH and diluted to a final concentration of 6 pM. Sequencing was performed with a MiSeq Reagent Kit v3–618 cycle (MS-102-3003) following the manufacturer’s protocols (Illumina, Inc., San Diego, CA, USA). Paired-end sequencing (2x301 nt) was performed on an Illumina MiSeq platform with 5% PhiX spike-in quality control (PhiX Control Kit v3—FC-110-3001).

Preparing sequence reads for downstream analysis

To demultiplex the paired end reads and construct FASTQ files, Illumina BaseSpace software was used. The data were analysed using the Quantitative Insight Into Microbial Ecology pipeline (QIIME2, ver 2019.1) [39]. Cutadapt software integrated in QIIME2 software was used to check for the presence of adapter sequences (CTGTCTCTTATACACATCT) and trim the 3’ end of the reads. Quality trimming, filtering and chimera removal were performed with DADA2 software [40]. The trimming parameters were set as follows: the forward read length was set to 299 bases; for the reverse reads, the length was set to 249 bases. No bases were cropped from the 3’ end of the forward and reverse reads.

Bioinformatic analyses

QIIME2 integrated in MAFFT software was used for multiple sequence alignment [41], and the reads were taxonomically classified using a Naïve Bayesian classifier trained on the Greengenes (ver13_8) [42] reference database by selecting mapping points according to the forward-reverse primer set that was used for amplifying the V3-V4 regions of the 16S rRNA genes of the bacterial community (341F, 806R). Phylogenetic trees were constructed with the FastTree plugin [43]. The QIIME2 pipeline was used for the alpha diversity analyses. For sample normalization, a read depth of 1494 was set. To analyse the alpha diversity, Shannon’s index [44, 45], Faith’s phylogenetic diversity index [46], Simpson evenness [47], and Chao-1 index [48] were calculated using the QIIME2 pipeline. The differences in the alpha diversity were assessed using the Kruskal-Wallis test. To estimate beta diversity differences weighted UniFrac distance [49] was calculated. For visualization of beta diversity distance-based dissimilarity matrices PCoA plots were generated using the Emperor plugin [50]. Beta diversity group significances were calculated with Permutational multivariate analysis of variance (PERMANOVA) pseudo-F statistical test [51]. QIIME2 artifact files were exported from the pipeline and converted to TSV files that could be used with different visualization packages. We used the Python (ver 3.6.5) Seaborn package to generate heatmaps, donut plots, and boxplots were constructed with the pandas and matplotlib packages. Correlational analyses were performed with the Corrplot R package.

Growth performance

To investigate the effects of phytonutrients on the growth performance of Cyprinus carpio, the average body weight (ABW) was measured (Table 1). At the beginning of the feeding experiment, the ABW was 123.45±49.28 g, whereas at the end of the experiment, the ABW reached 239.51±70.25 g. We estimated that the phytonutrients -based diet had no significant effect (p>0.05) on the ABW of common carp juveniles. The highest ABW was observed in the CAR animals (254.89±76.95 vs. 227.67±78.51 g for the controls). The supplementation of the diet with plant extracts did not have an impact on the specific growth rate (SGR) or feed conversation ratio (FCR) of the fish. However, the highest SGR (1.73±0.10% day^-1) values and the best FCR results (1.76±0.14 g^-1) were observed with the CAR treatment (Table 1).

Correlation between average body weight and fish microbiota

We managed to identify taxa showing correlations with body weight. Specifically, we captured notable associations between the GIT microbiota and the body weight of Cyprinus carpio. Alterations in the strength and direction of the correlations were measured at five taxonomic levels (Fig 2). At the phylum level, Proteobacteria (R = 0.65), Tenericutes (R = 0.58) and Chlamydiae (R = 0.56) showed the strongest positive correlations, and the phyla Spirochaetes (R = -0.47) and Firmicutes (R = -0.41) showed the most notable negative correlations with body weight gain. Further important positive correlations were observed for the classes Gammaproteobacteria (R = 0.59), the order Vibrionales (R = 0.62), the families Pseudoalteromonadaceae (R = 0.61) and Peptostreptococcaceae (R = 0.53) and the genera Lactobacillus (R = 0.59) and Luteolibacter (R = 0.51). In contrast, striking negative correlations were detected between the ABW and the class Brevinematae (R = -0.47), its order Brevinematales (R = -0.47) and its family Brevinematacea (R = -0.47) and with the order Clostridiales (R = -0.41), the family Clostridiaceae (R = -0.41) and the genera Enterobacter (R = -0.52) and Pseudoxanthomonas (R = -0.56).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Spearman correlation analysis was performed to measure the associations between the ABW and the intestinal taxa of Cyprinus carpio.

The correlation values range from –1 to +1 and indicate the level of consistency of the positive (≥0; red) and negative (<0; grey) correlations. The correlations were calculated at five taxonomic levels: phylum, class, order, family, and genus.

https://doi.org/10.1371/journal.pone.0248537.g002

Ageing- and diet-driven alterations in the alpha and beta diversities

Microbial diversity is considered a key factor influencing animal productivity through disease tolerance. Diverse microbiota shows increased resilience, and only tendentious shifts can push them toward an unhealthy state. Alpha and beta diversity metrics were calculated to track the notable conversions in the community diversity of the control (BD) and treatment groups (ANTH, SYN, fOS, and CAR) (Fig 3). Specifically, Faith’s PD, Chao-1, Shannon and Simpson diversity indices were used to evaluate the species abundance, richness and evenness of the GIT microbiota. Annotated heatmaps represent the community alpha diversities in relation to the time of sampling and treatments (Fig 3A). In the case of ANTH, SYN, and fOS treated samples the analysis of Faith’s PD showed a steady increase by week 4 (day 28^th) of the study period followed by a sharp decline at week 6 (day 42^nd) of the experimental period (mean Faith’s PD: 9.3 > 4.9). Chao-1, Shannon’s, and Simpson’s indices did not show a marked decrease after the 4th week (day 28^th). Interestingly, the most pronounced decline in the community alpha diversity was observed with the ANTH supplementation during the last 2 weeks of the study period. Moreover, the lowest value of Simpson’s index (day 42^nd: 0.85 vs. day 1^st: 0.89) was detected during the last quarter of the experimental period. In general, sweet red pepper seed extract (fOS) was able to ameliorate the community diversity according to Faith’s PD and Chao-1 indices, and this treatment group was the only group that did not show a decline in Shannon’s and Simpson’s indices during the last 2 weeks of the study period.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Alpha diversity metrices used in the comparison analysis.

a) Annotated heatmaps showing the values of Faith’s PD, Chao-1, Shannon’s and Simpson’s diversity indexes as a function of the time of sample collection (y axis) and the experimental settings (x axis). b) The sample distances were calculated based on quantitative (Bray-Curtis, weighted UniFrac) and qualitative (Jacquard, unweighted UniFrac) dissimilarity-based statistics c, d) Weighted UniFrac analysis was performed to identify ageing- and treatment-driven differences between the groups. The beta diversity relationships are summarized in two-dimensional scatter plots. The distances between dots are representative of differences in microbiota compositions. BD represents fish that received a basal diet, which served as a negative control with no dietary supplement or the following dietary treatments as supplemented feed: BD+1% anthocyanins (ANTH) provided by Hungarian sour cherry extract, BD+1% synbiotics (SYN) provided by fermented corn, BD+1% fermentable oligosaccharides (fOS) provided by Hungarian sweet red pepper seed extract, and BD+1% carotenoids (CAR) provided by Hungarian sweet red pepper pulp extract.

https://doi.org/10.1371/journal.pone.0248537.g003

Four beta diversity metrics were investigated; specifically, the weighted and unweighted UniFrac, Bray-Curtis, and Jaccard distances, (Fig 3B) between the different experimental groups were determined. Distance-based dissimilarity matrices showed that nutraceuticals did not exert a significant (p<0.05) influence on the overall community variations between the different treatment groups.

Principal coordinate analysis (PCoA) using weighted UniFrac distances was performed to measure the age dependency of the community taxonomy data, and this analysis yielded two clusters (clusters 1 and 2) representing different spatial ordinations between the fishes at days 14^th (aged 33 weeks), 28^th (aged 35 weeks) and 42^nd (aged 37 weeks) (Fig 3C). The microbiota of 37-week-old Cyprinus carpio showed a distinct clustering pattern compared with the microbiota at earlier time points (aged 33–35 weeks) (Fig 3C). When marking the samples according to diet, no distinct patterns were observed between the treatment groups (Fig 3D). Based on the PCoA plots, we concluded that age induced more pronounced shifts in the microbial community than diet.

Characterization of the Cyprinus carpio gut symbiome

Our interest in understanding phytonutrient-induced taxonomic shifts in the symbiotic microbiota of cultured, healthy Cyprinus carpio has grown. The observed community compositions relative to that of the healthy controls are shown in Fig 4. Differences in normalized abundance data were attained by considering 11 phyla, 16 classes, 25 orders, 36 families and 34 genera. Graphical representations were generated by calculating the log2 differences in taxa abundances between fish fed phytonutrient-supplemented diets (ANTH, SYN, fOS, and CAR) and those given the basal diet (BD). Taxonomic shifts were obtained when comparing untreated vs. treated healthy Cyprinus carpio symbiomes.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 4. The composition of the healthy Cyprinus carpio GIT symbiome was mapped.

Abundance variations were observed in key taxa. A composite heat map was created to note the pronounced distortions in the taxonomic profiles based on the normalized abundance data (ANTH/SYN/fOS/CAR vs. BD). BD represents fish that received a basal diet, which served as a negative control with no dietary supplement or the following dietary treatments as supplemented phytonutrients feed: BD+1% anthocyanins (ANTH) provided by Hungarian sour cherry extract, BD+1% synbiotics (SYN) provided by fermented corn, BD+1% fermentable oligosaccharides (fOS) provided by Hungarian sweet red pepper seed extract, and BD+1% carotenoids (CAR) provided by Hungarian sweet red pepper pulp extract. The extents of the differences are illustrated on the composite heatmap with gradient colours, where the yellow scale shows the taxa found at higher abundances in the phytonutrients-treated specimens (log2(phytonutrients/BD)>0) and the blue scale indicates a higher taxa abundance in the non-treated controls (log2 (phytonutrients/BD) < 0). Shifts in the abundances of 122 taxa due to phytonutrients can be observed.

https://doi.org/10.1371/journal.pone.0248537.g004

Nutraceuticals favoured the growth of Cyanobacteria, and anthocyanins (ANTH) induced the growth of the phylum Actinobacteria, the family Nocardiaceae, and the genera Sphingomonas and Acidovorax. In contrast, synbiotics (SYN) stimulated the classes Flavobacteria and Betaproteobacteria, the orders Burkholderiales and Neisseriales, the families Weeksellacea, Peptostreptococcaceae, Phyllobacteriaceae and Neisseriaceae and the genera Chryseobacterium, Epulopiscium, Pelomonas, and Vogesella. Anthocyanins (ANTH) and synbiotics (SYN) induced marked enrichment of Firmicutes, Clostridia, Clostridiales, Clostridiaceae and Clostridium. The abundances of the families Lachnospiraceae and Phyllobacteriaceae and the genus Shinella were increased by synbiotics (SYN) and fermentable oligosaccharides (fOS). Fermentable oligosaccharides (fOS) induced the famillies Ruminococcaceae and Plantomycetaceae, and the pulp extract (CAR) favoured the growth of the class Bacteroidia and order Bacteroidales and the families Bacteroidaceae, Rhodospirillaceae, and the genera Chryseobacterium, Enterococcus, and Acidovorax.

Diet affects the distribution of beneficial and potentially pathogenic and spoilage bacteria

The beneficial or detrimental effects of microorganisms on the intestinal microbiota can be strongly influenced by several factors, such as the fish species, prebiotic substrates, composition of the complex microbial populations, pH, living environments, and seasons. Because both biotic and abiotic stressors decrease the abundances of beneficial bacteria, opportunistic members of the community might take advantage and become infectious [52], leading to serious seasonal losses in aquaculture.

High-throughput sequencing of the bacterial 16S rRNA genes indicated that a phytonutrient-enriched diet induced significant shifts in the core phyla of the Cyprinus carpio GIT (Fig 5). During this feeding experiment, the carp gut was dominated by Fusobacteria (Σ49.8±16.6%) and Proteobacteria (Σ37.1±16.0%), and significant differences were found between the treatment groups (Fig 5A). The relative frequencies of the two dominant phyla showed the greatest most equalization in the fermented corn-treated (SYN) samples and were most pronounced in the pepper seed (fOS, P/F ratio: 0.61, 35.3% Proteobacteria, 58.2% Fusobacteria)- and pepper pulp extract-treated samples, (CAR P/F ratio: 0.57, 34.1% Proteobacteria, 59.6% Fusobacteria).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 5. Diets enriched in phytonutrients affect important beneficial and potentially detrimental members of the Cyprinus carpio GIT.

a) Donut plots represent diet-specific distortions in the main phyla. The Firmicutes-to-Bacteroidetes ratio (F/B ratio) and Proteobacteria-to-Firmicutes ratio (P/F ratio) are shown. b) Boxplots allow comparisons of the relative proportions of important general; specifically, relevant genera in the different groups are represented as boxplots. BD represents fish that received a basal diet, which served as a negative control with no dietary supplement or the following dietary treatments as supplemented feed: BD+1% anthocyanins (ANTH) provided by Hungarian sour cherry extract, BD+1% synbiotics (SYN) provided by fermented corn, BD+1% fermentable oligosaccharides (fOS) provided by Hungarian sweet red pepper seed extract, and BD+1% carotenoids (CAR) provided by Hungarian sweet red pepper pulp extract.

https://doi.org/10.1371/journal.pone.0248537.g005

By modulating the optimal nutrition and energy utilization of the host, the phyla Bacteroides and Firmicutes, which are associated with SCFA synthesis, can improve the growth performance of aquatic animals [53]. According to data, elevated Firmicutes levels can be associated with increased nutrient absorption, whereas Bacteroidetes is correlated with enhanced hydrolysis of glycogen, starch and polysaccharides. Under our experimental settings, the highest and lowest F/B ratios were obtained with the ANTH-fed fish (227.7) and the CAR-fed carp (1.6), whereas the control animals (BD) had an F/B ratio of 3.1.

Anthocyanins increased the abundance of Shewanella (ANTH 1.4±1.3% vs. BD 0.9±0.7%) in comparison to that obtained with the basal diet. Synbiotics increased the abundances of Bacillus (SYN 1.3±1.8% vs. BD 0.8±0.8%) and Shinella (SYN 0.2±0.5% vs. BD 0.04±0.09%) and decreased that of Shewanella (SYN 0.5±0.6% vs. BD 0.9±0.7%). The genus Acinetobacter, which contains potential fish pathogen species, was not detected in the synbiotic-fed (SYN) fish, whereas the pepper pulp (CAR 0.1±0.03%) and seed extracts (fOS 0.06±0.08%) decreased the abundance of this genus, and the sour cherry extract (ANTH 0.1±0.2%) significantly increased its abundance relative to that found in fish fed the basal diet (BD 0.3±0.1%).

According to other data, Cetobacterium is involved in the fermentation of peptides and carbohydrates [54], including in the production of vitamin B12 [15]. Pulp extracts increased the abundances of Cetobacterium (CAR 59.6±14.5% vs. BD 50.2±15.6%) and Enterococcus (CAR 0.6±0.6% vs. BD 0.3±0.3%) and decreased the abundances of Bacillus (CAR 0.3±0.7% vs. BD 0.8±0.8%), Pseudomonas (CAR 0.9±0.5% vs. BD 1.8±0.8%), and Acinetobacter (CAR 0.1±0.04% vs. BD 0.3±0.2%). The genus Shinella, which can potentially eliminate nitrate contamination from the environment, was markedly increased in the synbiotic-fed fish.

Comparative metagenomics reveals taxonomy associations in carp symbiomes

Taxonomic interconnections were revealed between and within the GIT microbiota of the control and phytonutrient-fed fish. We also estimated the extent to which genera are likely to play an indispensable role in maintaining the gastrointestinal health of Cyprinus carpio. Relative proportions were correlated using Spearman’s correlation method (Fig 6). The analysis revealed 32 significant correlations between genera, and these included 12 positive (Fig 6A), two negative (Fig 6B), and 18 opposite (Fig 6C) correlations. Furthermore, 31 unique (26 in BD, five in phytonutrients-fed fishes) significant correlations were found between Cyprinus carpio GIT genera (Fig 6D). Significant positive associations were found among Planctomycetes, Enterococcus, and Luteolibacter, among Pseudoxanthomonas, Enterobacter, and Cetobacterium, and among Pseudomonas, Acinetobacter and Aeromonas. However, Enterobacter and Cetobacterium were negatively correlated with Bacillus and Vogesella, respectively.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 6. Convergent and divergent abundance patterns identified in Cyprinus Carpio symbiomes.

Spearman’s correlations were calculated between the 24 most abundant genera in the basal diet-fed control and phytonutrients-fed carp. For better visualization, we listed associations with similar (“positive correlations” and “negative correlations”) and different (“opposite correlations”) correlation precursors in the two experimental categories. Unique correlations are also shown. BD represents fish that received a basal diet, which served as a negative control with no dietary supplement. The following dietary treatments were supplemented as phytonutrients: BD+1% anthocyanins (ANTH) provided by Hungarian sour cherry extract, BD+1% synbiotics (SYN) provided by fermented corn, BD+1% fermentable oligosaccharides (fOS) provided by Hungarian sweet red pepper seed extract, and BD+1% carotenoids (CAR) provided by Hungarian sweet red pepper pulp extract.

https://doi.org/10.1371/journal.pone.0248537.g006

Notably, opposite correlations were detected between Enterococcus and Luteolibacter, between Pseudomonas and Plesiomonas and among Clostridium, Aeromonas, Acinetobacter and Plesiomonas.

We also revealed significant correlations between genera that were unique to a specific diet. The basal diet-fed animals showed the highest number of unique taxonomy matches (26 positive correlations). As such, very strong associations between Lactobacillus and Aerococcus and among Proteocatella, Curvibacter and Gluconacetobacter were exclusively obtained in nontreated carp.