Physiological and Intestinal Microbiota Analyses Offer Insights into the Analysis of Differential Residual Feed Intake in Jian Carp (Cyprinus carpio var. Jian)

Simple Summary

This study investigates the regulatory mechanisms of residual feed intake (RFI) in Jian carp (Cyprinus carpio var. Jian) from an integrated perspective, encompassing growth performance, physiological and biochemical parameters, and intestinal microbiota. We compared two groups of fish: HRFI and LRFI. We found that the more efficient fish did not grow faster or bigger, but they ate less feed to achieve the same size. Their advantage came from having healthier intestines with better structure and stronger digestive power, a more robust body-wide antioxidant defense system, and a different set of gut microbes linked to beneficial metabolic functions. Our results show that feed efficiency is a complex trait driven by the combined benefits of better internal health and a helpful gut microbiome, offering new insights for breeding and managing fish in sustainable aquaculture.

Abstract

Feed efficiency (FE) is a critical economic trait in aquatic species. This study aimed to assess the effects of residual feed intake (RFI) divergence on growth performance, as well as antioxidant, digestive, and immune capacities. Additionally, intestinal microbiome was also employed to reveal the mechanism affecting the RFI in Jian carp. After the 8-week culture period, 12 fish (25 ± 1.05 g) each from the highest and lowest RFI extremes were selected as the HRFI and LRFI groups, respectively, for detailed physiological and microbial analysis. In terms of growth performance, the RFI, FCR, and DFI were found to be significantly lower in the LRFI group (p < 0.001), whereas no differences were observed in the ADG, BWG, SGR, HIS, VSI, and CF (p > 0.05). For physiological performance, the activities of digestive enzymes (protease, lipase and amylase) and antioxidant enzymes (T-AOC, SOD, CAT, GPx) were significantly higher in the LRFI group than in the HRFI group (p < 0.001). In line with this, the integrity of the intestinal tissue in the LRFI group was also superior to that in the HRFI group. Furthermore, the expressions of immune-related genes (LEP, GHR, AGPR, NPY) followed the same pattern. However, the expression of the CCK gene was significantly higher in the HRFI group (p < 0.001). There was no significant difference in the total lipid and fatty acids contents of muscle between the RFI groups (p > 0.05). Microbiota analysis indicated that the LRFI group harbored a higher relative abundance of several microbial taxa often associated with beneficial metabolic functions, including s Cetobacterium_sp_ZOR0034, unidentified_Chloroplast, Chloroplast, and Mangrovibacter. KEGG functional enrichment analysis indicated that the functions of these microbiota were primarily associated with metabolic processes. Collectively, these results demonstrate that improved feed efficiency in Jian carp is collaboratively driven by enhanced physiological status (digestion, antioxidant, immunity) and a beneficial shift in gut microbiota. This study provides an integrated perspective for understanding the regulatory mechanisms of RFI and offers potential microbiota-targeted strategies for feed efficiency improvement in aquaculture.

1. Introduction

Feed efficiency (FE) quantifies the relationship between an animal’s weight gain and total feed consumption over a defined period [1]. A commonly used metric for FE is the feed conversion ratio (FCR), which represents the ratio of feed intake (FI) to body weight gain (BWG). Improved feed efficiency corresponds to a reduction in FCR. However, because FCR is a composite trait derived from both BWG and FI, its genetic variation cannot be directly attributed to growth rate or feed consumption alone [2]. To overcome this limitation, residual feed intake (RFI) was introduced as an alternative measure of FE [3]. Proposed by Koch et al. [4], RFI is quantified as the deviation of an animal’s actual feed consumption from its expected intake, which is estimated from the energy demands for maintenance and physiological growth. Unlike FCR, RFI is genetically independent of growth traits and exhibits greater heritable variation, making it a more reliable indicator of maintenance energy expenditure. Previous studies have demonstrated that selecting farmed animals with low RFI as a feed utilization index can significantly reduce feed intake (FI) without affecting body weight, thereby effectively lowering production costs [5].

Currently, many breeders are selecting RFI over FCR in animal breeding programs, as genetic selection for RFI could have significant affect production performance. However, the phenotypic differences in RFI is associated with a complex interplay of factors, including their physiological conditions, such as intestinal digestion, antioxidant capacity, immune capacity, growth environment, and nutritional availability [6,7]. Additionally, the distinct physiological structures and compositions of digestive organs across different species significantly affect their feeding behaviors. Wen et al. revealed that RFI was strongly associated with gut microbes, and the abundance of beneficial flora in LRFI chickens was significantly improved [8]. Evidence shows that selection for lower RFI in swine is linked to a favorable body composition shift, characterized by greater lean tissue deposition, reduced back-fat depth, and, ultimately, leaner carcasses [9]. Moreover, meat from low-RFI pigs tends to have lower intramuscular fat content, lower ultimate pH, greater drip loss, and lighter meat color [10]. Similar results have also been found in other livestock and poultry species, such as broilers [5], meat ducks [11], and Hu sheep [12]. However, few studies have explored the relationship between phenotypic differences in RFI and physiological status for aquatic economic animals.

Common carp (Cyprinus carpio) is a freshwater species of major economic significance in China, with an annual production reaching approximately 2.87 million tons in 2023. A principal cultivated variety, the Jian carp (Cyprinus carpio var. Jian), is notably valued for its rapid growth, broad adaptability, robust disease resistance, and stable genetic inheritance [13]. Therefore, if the feed utilization efficiency of carp can be improved through selective breeding in the process of Jian carp breeding, the total production cost can be effectively reduced. Numerous studies have reported on the nutritional requirement of common carp, addressing various aspects such as nutrient needs [14,15], graded feeding levels [16], and disease resistance [16]. However, the relationship between phenotypic differences in RFI and growth performance in Jian carp remains inadequately understood, as do the biological processes associated with RFI that impact physiological status.

Therefore, in this study, two groups of Jian carp that exhibit significant differences in RFI were selected to investigate the association between RFI divergence and the growth performance, muscle nutrient composition, antioxidant capacity, and immune response in C. carpio var. Jian. Additionally, conducted combined intestinal microbial detection and 16S rRNA sequencing were employed to further elucidate the differences in intrinsic regulatory mechanisms. The findings offer valuable insights into the genetic selection for RFI in economically important aquatic species.

2. Materials and Methods

2.1. Ethics Statement

Animal procedures in this study complied with the animal care and use guidelines approved by the Animal Ethics Committee of the Freshwater Fisheries Research Center, Chinese Academy of Fishery Sciences (Wuxi, China) (LAECFFRC-2023-06-12).

2.2. Experimental Fish and Study Design

The Jian carp employed in this experiment were sourced from the Jinjiang Research Base of the Freshwater Fisheries Research Center, Chinese Academy of Fishery Sciences (Taizhou, China). Fish with mature gonads were carefully chosen for natural reproduction during the breeding season. The fry were nurtured to a weight of approximately 10 g in ponds before being transferred to an indoor recirculating aquaculture system (RAS). After reaching an average weight of 25 ± 1.05 g, a total of 160 fish of comparable sizes were selected. Four fish were randomly allocated to each of the 40 independent 300 L tanks within the RAS. The tank was separated into four partitions to accommodate the fish individually. The water temperature was stabilized at 25 ± 2 °C, pH at 7.0–8.0, and dissolved oxygen above 5 mg/L. The system operated under natural light with a daily water exchange rate of approximately 30%. Water quality parameters (ammonia, nitrite) were monitored weekly. Following a two-week acclimatization period under the same conditions, the formal 8-week feeding trial commenced. The fish were fed twice daily (9:00 and 17:30) to apparent satiation with commercial feeds (crude protein level 31%, crude fat level 3.5%). The individual feed intake was meticulously recorded daily, and initial and final weights of each fish were precisely measured throughout the experiment.

2.3. Sample Collection

At the end of the experiment, individual fish were ranked by RFI values, and fish with extreme RFI values were selected to be divided into two groups: low-RFI group (LRFI, n = 12) and high-RFI group (HRFI, n = 12). To begin the sampling procedure, the 24 designated fish, after a 24-h fast, were euthanized through terminal anesthesia with 80 mg/L eugenol (ScanAqua, Hvam, Norway) for 5 min. Each fish was then measured for post-euthanasia body weight and length, followed by external sterilization with 70% ethanol and aseptic dissection.

Blood samples were taken from the tail vein and used for enzymatic activities assays. Approximately 1.0–1.5 mL of hemolymph was collected per carp using a sterile 2.5 mL syringe. The sample was immediately placed on ice and processed within 30 min to separate serum for subsequent assays. After that, the entire liver and visceral organs (including the gastrointestinal tract and visceral adipose tissue, with all gastric and intestinal contents removed) were collected and weighed to calculate the hepatosomatic index (HSI) and visceralsomatic index (VSI), expressed as a percentage of body weight (BW). Then, the liver, gill, intestine, and muscle tissues were dissected separately. Each tissue was stored at −80 °C for biochemical assays. The middle gut contents were obtained and placed into 2 mL cryopreservation tubes, snap-frozen by liquid nitrogen, and stored at −80 °C for the community composition analysis of intestinal microbiota.

2.4. Proximate Composition Analysis

Muscle crude protein and crude lipid contents were analyzed following established Association of Official Analytical Chemists procedures [17]. Crude protein quantification was performed using the Kjeldahl method with acid digestion on an Auto Kjeldahl System (2300-Auto-analyzer, Foss Tecator, Höganäs, Sweden). Lipid was assessed and mined by a chloroform–methanol (2:1, v/v) containing 0.01% butylated hydroxytoluene (BHT) as an antioxidant as per Ceja [18]. Fatty acid composition was analyzed by Gas Chromatograph with mass spectrum (GC-MS) using an Agilent 7890 system (Santa Clara, CA, USA) equipped with SP™-2560 Silica Capillary Column (100 m × 0.25 mm × 0.2 µm film thickness, Supelco, Darmstadt, Germany). Nitrogen was used as a carrier gas at a linear velocity of 10 mL/ min, and the inlet temperature was set to 260 °C with a split ratio of 20:1. The GC oven temperature program was as follows: 50 °C held for 2 min, increased to 180 °C at 10 °C/min, then to 240 °C at 3 °C/min, and held for 10 min [19]. Total lipids were extracted and subsequently transmethylated into fatty acid methyl esters using 14% boron trifluoride in methanol, according to the protocol of Morrison and Smith [20].

2.5. Intestinal Morphology

The mid-intestine was washed with an ice-cold normal saline 3 times and was fixed with tissue fixative (Bonn’s solution) for 24 h. Then, standard morphological procedures were adopted to dehydrate intestinal samples, including equilibration in xylene and embedding in paraffin [21]. After that, samples were cut into 5 µm slices, which were then subjected to the hematoxylin–eosin (H&E) for 3 min. A light microscope (Nikon Eclipse 80i, Tokyo, Japan) was utilized to examine the slides with the photos taken by a digital camera (Nikon DS-U2, Tokyo, Japan). For morphometric analysis, five complete, well-oriented villi and crypts were randomly selected from each sample. Villus height (VH, from the tip to the crypt junction) was measured using the image analysis software ImageJ (National Institutes of Health, Bethesda, MD, USA, version 1.54d).

2.6. Assay of Antioxidant and Digestive-Related Enzyme Activity

Serum was isolated from blood samples via centrifugation at 3000× g for 10 min under 4 °C conditions. Liver tissues were excised, weighed, and homogenized on ice using a 9-fold volume (w/v) of ice-cold physiological saline (NaCl 28.4 g/L, MgCl∙6H₂O 1 g/L, MgSO₄∙7H₂O₂ g/L, CaCl₂∙2H₂O₂·25 g/L, KCl 0.7 g/L, glucose 1 g/L, Hepes 2.38 g/L). Following centrifugation at 4 °C for 20 min, the resulting supernatants were collected into fresh tubes for subsequent assays. Total antioxidant capacity (T-AOC) along with the activities of superoxide dismutase (SOD), catalase (CAT), malondialdehyde (MDA), and glutathione peroxidase (GPx) were determined in the serum using specific commercial assay kits (Nanjing Jiancheng Bioengineering Institute, Nanjing, China) strictly according to the manufacturer’s protocols. Liver homogenates were analyzed for protease, lipase, and amylase activities using corresponding assay kits from the same provider.

2.7. Total RNA Extraction, Reverse Transcription and Real-Time PCR

Total RNA was isolated from gill samples using the Total RNA Extraction Kit (Vazyme, Nanjing, China), following the manufacturer’s protocol. The concentration and integrity of the extracted RNA were verified using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA) and 1% agarose gel electrophoresis, respectively. First-strand cDNA was synthesized from 1 µg of total RNA using the HiScript III RT SuperMix (Vazyme, China). Quantitative real-time PCR (qRT-PCR) assays were conducted with SYBR Green I Master Mix (Novizan, Nanjing, China) on a Bio-Rad fluorescent quantitative PCR instrument (Hercules, CA, USA). The amplification protocol and qPCR conditions were adopted from Jiang et al. [22]. Each sample was analyzed in three technical replicates to ensure reproducibility. β-actin served as the internal control, and the 2^−ΔΔCt was applied to determine the relative expression of target genes [23]. The sequences of qRT-PCR primers used for amplifying non-specific immune-related genes in gill tissue—namely IgM, MPO, AKP, and ACP—are provided in Table 1.

Table 1.

Real-time quantitative PCR primers for immune and appetite related genes and β-actin of Cyprinus carpio var. Jian.

Genes	Primer
	Fwd: 5′-3′	Rv: 5′-3′
IGM	AAAGCCTACAAGAGGGAGACCGAT	CCAACACAATGAAACCGTTG
MPO	GCAGAGTCACCAATGACACCA	ATCCACACGGGCATCACCTG
AKP	GAACCGCAATTGCTGTAGAAG	CGCTTGTAGGTTCTTTGATGAGTG
ACP	CTCGGAITAATGCTTCGTTGTTCG	TGCTGAATTCTTGCTCTGTAGTTG
LEP	GGGAACAAATTGTCACTGG	CAGATAGAATTCAGCACTCC
POMC	AACCCCTTCTCACGCTCTTC	AACACCACCCACCCTCTTTT
GHR	CCACAACACGCAAGTCT	CCAGTCCGTTTCCACA
AGRP	CCGTGCATCCCTCATCAGC	GCTACGGCAGTAGCAGAAGGC
NPY	TGCTTGGGAACTCTAACGGAA	GACCTTTTGCCATACCTCTGC
CCK	CAGAATCATCTCCACCAAAGG	TCCATCCCAAGTAATCTCTGTC
β-actin	CGTGATGGACTCTGGTGATG	TCGGCTGTGGTGGTGAAG

Note: IGM, Immunoglobulin M; MPO, Myeloperoxidase; AKP, Alkaline Phosphatase; ACP, Acyl Carrier Protein; LEP, Leptin gene; POMC, Proopiomelanocortin; GHR, Growth Hormone Receptor gene; AGRP, Agouti-related protein; NPY, Neuropeptide Y; CCK, Cholecystokinin.

2.8. 16S rRNA Sequencing in Cecum Microbiota

Total microbial genomic DNA was isolated from cecum content samples with the E.Z.N.A.^® soil DNA Kit (Omega Biotek, Norcross, GA, USA). The extracted DNA was evaluated for purity and integrity using a Nanodrop2000 spectrophotometer and agarose gel electrophoresis, respectively. Amplification of the bacterial 16S rRNA gene V3–V4 hypervariable regions (approximately 400–450 bp) was performed with the primers 515F (5′-GTGCCAGCMGCCGCGGTAA-3′) and 806R (5′-GGACTACHVGGGTWTCTAAT-3′). A total of 24 libraries (12 for the LRFI group and 12 for the HRFI group) were constructed. Amplicon libraries were subjected to paired-end sequencing on an Illumina HiSeq platform at Novogene Company (Beijing, China).

The bioinformatics pipeline for microbiome analysis was implemented as follows. Raw paired-end reads were processed within the QIIME2 framework (version 2020.6). Sequences were quality-filtered, trimmed, denoised, merged, and chimera-checked using the DADA2 plugin (version 1.18.0) to generate amplicon sequence variants (ASVs). Primer sequences were removed, and reads were truncated at position 240 (forward) and 200 (reverse) based on quality profiles. Taxonomic classification of representative ASVs was performed with the classify-sklearn naive Bayes classifier against the SILVA 138 SSU rRNA database (release 138.1) at 99% similarity.

Alpha diversity was estimated using the observed Chao1, Shannon, Simpson, and Dominance indices. Beta diversity was assessed based on Bray–Curtis dissimilarity, visualized via principal coordinates analysis (PCoA), and statistically evaluated using permutational multivariate analysis of variance (PERMANOVA). Differentially abundant taxa at the genus level were identified through linear discriminant analysis effect size (LEfSe), applying a logarithmic LDA score cutoff of >4.0. Microbial community functional potential was predicted via PICRUSt2 based on the Kyoto Encyclopedia of Genes and Genomes (KEGG) database. Inter-group comparisons were conducted using Student’s t-test or the Wilcoxon rank-sum test, with a significance threshold of p < 0.05. All raw sequencing data have been deposited in the NCBI Sequence Read Archive (SRA) under accession number PRJNA880788.

2.9. Calculations

The corresponding parameters were calculated by the following formulae:

Survival (%) = (final number of fish/initial number of fish) × 100%

Body weight gain (BWG, %) = [(final weight − initial weight)/initial weight] × 100%

Body length gain (BLG, %) = [(final length − initial length)/initial length] × 100%

Specific growth rate (SGR, % day⁻¹) = [(Ln final weight − Ln initial weight)/duration] ×100%

Hepatosomatic index, HIS, % = [liver weight/body weight] × 100%

Viscerosomatic index, VSI, % = [visceral weight/body weight] × 100%

Condition factor, CF, g/cm³ = [body weight/body length³] × 100%

RFI = DFI − β₀ − β₁ × MW^0.8 − β₂ × ADG

where DFI is the daily feed intake; β₀ is the regression intercept, β₀ is the partial regression coefficient of FI on metabolic weight, and MW^0.8 is the metabolic weight, which was calculated as (BW_Final − BW_Initial)/2 using the metabolic weight coefficient developed by Lupatsch and Martins methods [24,25]. The ADG is the average daily weight gain of the fish. β₂ is the partial regression coefficient for ADG, and RFI corresponds to the residuals of the model.

2.10. Statistical Analysis

Statistical analyses were conducted with SPSS 25.0 software, and results are presented as mean ± S.E.M. (n = 12). Data normality and homogeneity of variances were verified using the Shapiro–Wilk test and Levene’s test, respectively. Group differences (LRFI vs. HRFI) were determined by an independent-samples t-test, with a statistical significance level defined as p < 0.05.

3. Results

3.1. Growth Performance

No statistical difference (p > 0.05) was found in ADG, BWG, SGR, HSI, and VSI as well as the CF between the LRFI and HRFI groups (Table 2). However, the DFI and FCR in the HRFI group increased very significantly (p < 0.001) compared with the LRFI group.

Table 2.

Growth performance and feed efficiency traits of Jian carp between low and high RFI groups.

Parameters	Groups
	Low RFI	High RFI
DFI (g)	1.20 ± 0.18	3.37 ± 0.57 ***
ADG (g)	0.71 ± 0.05	0.75 ± 0.13
FCR	1.61 ± 0.12	4.47 ± 0.22 ***
BWG	1.85 ± 0.14	1.94 ± 0.30
SGR	1.84 ± 0.09	1.82 ± 0.18
HIS	0.88 ± 0.08	0.84 ± 0.08
VSI	11.70 ± 0.51	10.78 ± 0.73
CF	0.10 ± 0.00	0.11 ± 0.01
RFI	−0.85 ± 0.04	1.19 ± 0.19 ***

Note: Data were expressed as mean ± S.E.M. from triplicate groups (n = 12). Triple asterisk indicates p < 0.001. Abbreviations: DFI, daily feed intake; ADG, average daily gain; FCR, feed conversion ratio; BWG, body weight gain; SGR, specific growth rate; HSI, hepatosomatic index; VSI, viscerosomatic index; CF, condition factor; RFI, residual feed intake.

3.2. Enzymatic Activities Assays

The LRFI group exhibited significantly higher activities of protease, lipase, and amylase compared to the HRFI group (p < 0.001) (Figure 1A–C). At the same time, the intestines of Jian carp from LRFI group exhibited a well-preserved structure. These intestines displayed relatively intact serosa and muscular layers, neatly arranged villi, normal cell interstices, and no significant abnormalities (Figure 2). In contrast, histological sections of intestines from HRFI group revealed partial damage and atrophy of the intestinal villi, along with an increase in vacuoles (Figure 2).

Figure 1.

Liver activities of digestive in Jian carp between low- and high-RFI groups (mean ± S.E.M., n = 12). (A) Protease; (B) Lipase; (C) Amylase.

Figure 2.

Microscopic examination of intestinal villus sections in Jian carp between low- and high-RFI groups. The intestinal sections were stained with hematoxylin and eosin. VH: villus height. Arrows: exfoliation of microvilli and villi vacuoles. The dotted line square: intact serosa and muscular layers.

Interestingly, the similar trends with the digestive enzymes could also be found from the results of the antioxidant enzyme activities (T-AOC, SOD, CAT, GPx) (Figure 3A–D). However, the activity of MDA showed the opposite trend, and its activity in the HRFI group was significantly higher than that in the LRFI group (p < 0.001) (Figure 3E). In addition, the correlation analysis results showed that RFI was significantly negatively correlated with T-AOC, SOD, CAT, and GPx (p < 0.001) while being significantly positively correlated with MDA (p < 0.001).

Figure 3.

Serum activities of antioxidant in Jian carp between low- and high-RFI groups (mean ± S.E.M., n = 12). (A) T-AOC; (B) SOD; (C) CAT; (D) GSH-PX; (E) MDA; (F) Correlation between antioxidant activity and RFI.

3.3. Immune-Related Genes Expression

The expression of genes related to immune in the gill of Jian carp are shown in Figure 4. The expression levels of LEP, GHR, AgRP I, and NPY were significantly higher in the LRFI group than in the HRFI group (p < 0.001). By contrast, CCK gene exhibited the higher expression level in the HRFI group, which significantly (p < 0.001) differ from the ones in the LRFI group. However, no significant difference was observed for POMC between these two groups (p > 0.05).

Figure 4.

Relative expression of feeding-related genes in brain for Jian carp between low- and high-RFI groups (mean ± S.E.M., n = 12).

3.4. Muscle Lipid and Fatty Acid Composition

Neither the total lipid content nor the composition of fatty acids including SFAs, MUFAs, n3, n6, DHA, and EPA were significantly different between the LRFI and HRFI group (p > 0.05) (Figure 5).

Figure 5.

Lipid and fatty acid composition (g/100 g) in muscle of or Jian carp between low- and high-RFI groups (mean ± S.E.M., n = 12).

3.5. Microbial Composition

The Venn diagram (Figure 6A) summarizes the similarities and overlap of operational taxonomic units (OTUs) between the LRFI and HRFI groups. The results suggested that both groups overlapped 731 OTUs, while some OTUs were distinct to specific groups (3259 for the LRFI and 3865 for the HRFI groups). This result indicates a distinct OTU composition in the intestinal microbiota between the two groups.

Figure 6.

Intestinal microbial composition in both groups (n = 12). (A–C) represent the Venn diagram of different microorganisms, Alpha and Beta diversity (PCoA), by t-test, respectively.

The microbiome of cecum alpha diversity (Chao, Dominance, Shannon, and Simpson indices) was higher (p < 0.05) in the LRFI group than in the HRFI group. Moreover, both groups demonstrated a substantial difference in species composition by PCoA. In both groups, Proteobacteria and Firmicutes were the prevalent phyla, as demonstrated in Figure 7. In the HRFI group, a marked elevation in the relative abundance of Proteobacteria was observed at the phylum level, concomitant with a reduction in Firmicutes. At both family and genus levels, Aeromonadaceae/Aeromonas and Pseudomonadaceae/Pseudomonas constituted dominant taxa within the cecal microbiota. By contrast, the LRFI group exhibited a decreased overall abundance of Gammaproteobacteria, Fusobacteria, and Proteobacteria at these taxonomic levels relative to the HRFI group. Conversely, the abundance of Bacteroidetes was significantly higher in the LRFI group.

Figure 7.

Alterations in microbiota at the phylum, family, and genus levels.

To identify the substantial relatively abundant amplicon sequence variants (ASVs) for the whole microbiota at levels from phylum to genus (p < 0.05; LDA > 3.0), linear discriminant analysis effect size (LEfSe) analysis was carried out. As shown in Figure 8, unidentified_Chloroplast, Chloroplast, and Alphaproteobacteria were selected as possible markers with considerable variations between both groups. In contrast to the HRFI group, the relative abundances of Cetobacterium_sp_ZOR0034, unidentified_Chloroplast, Chloroplast, and Mangrovibacter were increased significantly in the LRFI groups, while the HRFI groups showed an increased abundance of Alphaproteobacteria and Burkholderiales.

Figure 8.

The levels and biomarkers identified via LEfSe and LDA scores. Species with considerable variations that have an LDA score > 3 are presented.

3.6. Functional Prediction of the Intestinal Microbiota

The functions of microorganisms were predicted through KEGG functional abundance cluster analysis using PICRUSt (Figure 9). In pathway Level 2, a total of 20 pathways were significantly enriched, as shown in the functional abundance heat map (top 10). Compared with the HRFI group, the intestinal microbiota of the LRFI group were mainly enriched in the “Amino Acid Metabolism” and “Carbohydrate Metabolism”. However, the microorganisms in the HRFI group were more enriched in “Cellular Processes and Signaling”, “Replication and Repair”, as well as “Translation”.

Figure 9.

Functional prediction of the intestinal microbiota in KEGG.

4. Discussion

Research on feed efficiency traits among fish, particularly residual feed intake (RFI), is limited. Excessive feeding can lead to eutrophication and acidification in aquaculture systems, adversely affecting the environment. Additionally, feed costs constitute a significant portion of the total expenses in aquaculture. Therefore, enhancing feed efficiency not only mitigates environmental pollution but also supports the sustainable development of the aquaculture industry. However, measuring individual feed intake in fish during production is challenging. Consequently, identifying RFI biomarkers has emerged as a key research focus in recent years, with many researchers seeking simpler and more cost-effective methods to pinpoint individuals with higher feed efficiency. In this study, carp were categorized into high-RFI (HRFI) and low-RFI (LRFI) groups based on calculated RFI values. We examined the factors contributing to the differences in RFI between these groups by comparing their growth characteristics, physiological traits, and biochemical composition. Additionally, we investigated the relationships between these traits and the characteristics of intestinal microorganisms. The key findings indicate that low-RFI (LRFI) individuals achieve feed efficiency through superior intestinal health (enhanced digestive enzyme activity and villus integrity), a robust systemic antioxidant status, and a distinct gut microbiome enriched for metabolic functions, without compromising growth performance. These integrated results highlight that feed efficiency is not merely a matter of intake, but a systemic trait influenced by digestive capacity, physiological resilience, and host–microbiota interactions, offering novel biomarkers and targets for selective breeding.

As anticipated, the LRFI and HRFI groups exhibited statistically significant differences (p < 0.001) in both RFI and FCR. However, no significant differences were observed between the groups for performance and physiological indices, including ADG, BWG, SGR, HSI, VSI, and CF (p > 0.05). Interestingly, similar results were also observed in other species, such as mule ducks [26] and laying ducks [27]. These results verified that RFI is independent of economic traits. Therefore, a lower FI could satisfy the growth and maintenance requirements of the animals, and reducing RFI appropriately while improving feed utilization rate without affecting economic traits.

As the principal organ for nutrient digestion and absorption, the intestine is essential for ensuring normal animal growth. Its wall is composed of four distinct layers: the mucosa, submucosa, muscularis, and serosa (or adventitia) [28]. Intestinal mucosal thickness, muscular thickness, villi height, crypt depth, and the ratio of villi height to crypt depth can directly reflect intestinal digestion and absorption capacity [29,30]. A longer intestinal villi height indicates a stronger intestinal absorption capacity [31]. Ou et al. revealed that the crypt depth can reflect the development of the intestine, and a shallower crypt depth implies a strong digestive and absorptive capacity of the intestine [32]. In the current study, the outcome of VH demonstrated that the villus height of the LRFI group was higher than that of the HRFI group, indicating that intestinal absorption and digestion capacity of LRFI group was stronger than that of HRFI group. This structural advantage (Figure 2) directly supports the observed significantly higher activities of key digestive enzymes (protease, lipase, amylase) in the LRFI group (Figure 1A–C), providing a mechanistic basis for more efficient nutrient assimilation.

Antioxidant enzymes include T-AOC, SOD, CAT, MDA, and GSH-PX, and their activity levels are positively correlated with the health status of organisms. These enzymes can effectively eliminate oxygen-free radicals in the bodies of fish, thereby exerting antioxidant protective effects, and can maintain the stability of intracellular superoxide anion (O²⁻) and hydrogen peroxide (H₂O₂) levels [33,34]. Consistent with a healthier physiological state, the LRFI group exhibited significantly elevated activities of serum antioxidant enzymes (T-AOC, SOD, CAT, GPx) and lower MDA levels (Figure 3A–E). This enhanced antioxidant capacity was further supported by the upregulation of immune-related genes (LEP, GHR, AgRP I, NPY) in gill tissue (Figure 4). These results also indicates that Jian carp established by LRFI tend to exhibit a healthier condition. This is consistent with the expression results of immune-related genes in Jian carp detected in this study.

Numerous studies have revealed a significant relationship between the types of intestinal microbiota and how effectively animals utilize feed [8]. Research indicates that the proportion of Firmicutes to Bacteroidetes in animals’ intestinal tissues is influenced by their feed conversion efficiency [35]. A diminished efficiency in feed utilization correlates with a reduced presence of Bacteroidetes and a higher abundance of Firmicutes within the intestinal microbiota [36,37]. Our findings demonstrated that the gut ecosystem differed substantially between groups. The LRFI group harbored a microbiota with greater alpha diversity (Figure 6), a distinct community structure (PCoA), and an increased Firmicutes-to-Bacteroidetes ratio (Figure 7). Notably, taxa such as Cetobacterium sp. ZOR0034 were enriched in LRFI individuals (Figure 8). Functionally, the LRFI-associated microbiota was predicted to be significantly enriched in “Amino Acid Metabolism” and “Carbohydrate Metabolism” pathways (Figure 9), suggesting a microbiome optimized for energy harvest and metabolic support to the host. Research has indicated that microbial diversity associated with feed conversion efficiency plays a role in cellular metabolism [38]. The energy required for growth and metabolic processes in animals is generated via the tricarboxylic acid cycle (TCA cycle), wherein both amino acid metabolism and carbohydrate metabolism are crucial [38]. Certain amino acids can undergo transamination, resulting in the formation of α-ketoglutaric acid, which subsequently enters the TCA cycle and is oxidized through a series of REDOX reactions, ultimately yielding energy in the form of ATP [39]. This energy is essential for the growth of fish, facilitating various biological functions such as cell proliferation, differentiation, protein synthesis, and signal transduction [40]. Additionally, some amino acids can be transformed into precursors necessary for the synthesis of carbohydrates or lipids [41]. Consequently, this elucidates why “Amino Acid Metabolism” and “Carbohydrate Metabolism” were markedly enriched in the functional enrichment analysis of the intestinal microbiota within the LRFI group in this study.

Finally, the divergent phenotypes observed between LRFI and HRFI carp can be interpreted through an integrative physiological model centered on host–microbiota metabolic synergy. The gut microbiome of LRFI individuals, characterized by higher diversity and an enrichment of Firmicutes (known for energy harvest) and specific genera like Cetobacterium, appears functionally primed for efficiency (Figure 7 and Figure 8). The predicted enhancement of “Amino Acid Metabolism” and “Carbohydrate Metabolism” (Figure 9) in this microbiome suggests increased production of microbial fermentation end-products (e.g., short-chain fatty acids). These metabolites can serve as additional energy substrates for the host and may positively modulate intestinal health [42].

This microbial metabolic support likely synergizes with the host’s superior intestinal phenotype. The well-preserved villus architecture (Figure 2) and heightened digestive enzyme activity (Figure 1A–C) in LRFI fish ensure maximal nutrient extraction from feed. Concurrently, the enhanced systemic antioxidant capacity (Figure 3A–E) and immune gene expression (Figure 4) indicate a reduced burden of oxidative stress and inflammation. Managing inflammation is metabolically costly [1]; thus, a more resilient physiological state in LRFI individual conserves energy that would otherwise be diverted to immune defense and cellular repair.

Therefore, we propose a multi-level mechanism for high feed efficiency: a beneficial gut microbiome enhances nutrient/energy harvest and fosters intestinal health. This, in turn, reduces systemic metabolic stress (lower oxidative cost). The net metabolic gain is then channeled towards growth and maintenance rather than being wasted, explaining the identical growth (ADG, BWG) but significantly lower FCR (Table 2) in LRFI fish. This model moves beyond correlative observation to propose testable causal relationships between microbial community function, host digestive and antioxidant physiology, and the complex trait of RFI, offering a holistic framework for future research and aquaculture management strategies.

5. Conclusions

The study concluded that selecting Jian carp with residual feed intake (LRFI) can improve feed efficiency without compromising growth performance. Furthermore, LRFI individuals exhibited enhanced digestive capacity, antioxidant capacity, and immune capacity compared to their high-RFI (HRFI) counterparts. Intestinal microbiota analysis revealed that “Amino Acid Metabolism” and “Carbohydrate Metabolism” were the key functional signaling pathways enriched in the LRFI group and associated with improved RFI in Jian carp.

Author Contributions

Conceptualization, Y.T. and G.J.; Methodology, G.J., Y.Z. and E.M.K.; Validation, Y.T.; Formal analysis, G.J. and Y.Z.; Investigation, G.J., Y.Z., E.M.K. and W.F.; Resources, W.F., Y.X., J.L., Z.Z. and Y.T.; Data curation, G.J.; Writing—original draft preparation, G.J., Y.Z. and E.M.K.; Writing—review and editing, W.F., Y.X., J.L., Z.Z. and Y.T.; Supervision, Y.T.; Project administration, Y.T.; Funding acquisition. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

All research involving animals was conducted according to the animal care and use guidelines approved by the Animal Ethics Committee of the Freshwater Fisheries Research Center, Chinese Academy of Fishery Sciences (Wuxi, China) (LAECFFRC-2023-06-12). This study was carried out in compliance with the ARRIVE guidelines.

Informed Consent Statement

Not applicable.

Data Availability Statement

Data will be made available on request.

Conflicts of Interest

The authors declare no conflicts of interest.

Funding Statement

The study was supported by the earmarked fund for Central Public-interest Scientific Institution Basal Research Fund, Freshwater Fisheries Research Center, CAFS (No. 2025JBFR05), CARS [CARS-45], Central Public-interest Scientific Institution Basal Research Fund, CAFS [2023TD40], Hebei Province Modern Agriculture Industry Technology System Freshwater Aquaculture Innovation Team [HBCT2023230203], Central Public-interest Scientific Institution Basal Research Fund, CAFS [2024JBFR09].