16S rRNA Gene and Metagenomic Analysis Revealed an Association Between Cecal Microbiota and Pork Umami

Simple Summary

Pork flavor is a key determinant of meat quality, with umami serving as a major contributor to consumer preference and purchase decisions. However, the reason for the connection between the pork umami and the intestinal microorganisms remains unclear at present. In this study, umami intensity of pork from commercial pigs was quantified using a taste-sensing electronic tongue system, and animals were stratified into low, medium, and high umami groups. Cecal microbiota were then profiled by 16S rRNA gene sequencing and shotgun metagenomics to compare community composition and microbiome functional profiles across groups. The results show that the change in the flavor of pork may be associated with specific microorganisms and their metabolites rather than different microbial communities. This provides associative evidence that may inform future studies on precision feeding or microbiome-targeted strategies.

Abstract

Umami is a key determinant of pork flavor, but the association between the intestinal microbial community and umami differences remains unclear. Here, we used the taste-sensing electronic tongue system to divide the Duroc × Landrace × Yorkshire pigs into high, medium and low groups. We combined 16S rRNA gene and shotgun metagenomic sequencing to study the differences in the microbial community composition and functional genes. The results showed that the microorganisms in the cecum of different groups had a similar core microbial community. The Shannon diversity analysis showed that there were no significant differences among the different groups. The Bray–Curtis distance indicated that there were differences in the bacterial communities between the high umami group and the other two groups. The LEfSe analysis and Spearman correlation analysis revealed that the uncultured species CAG-632 sp900539185 maintained a high abundance in the high umami group and was significantly correlated with umami. Metagenomic functional analysis revealed distinct functional signatures among umami groups, with enrichment of genes related to carbohydrate transport and metabolism, butanoate and other short-chain fatty acid pathways, nitrogen utilisation, cell-surface structures, adhesion and RNA metabolism in high umami groups. These research findings indicate that the differences in the delicious flavor of pork are more likely to be associated with specific microbial species and the functional characteristics of the cecal microbial community, rather than the overall situation of the entire microbial community.

1. Introduction

Pork is one of the most widely consumed meats in the world, and its meat quality strongly influences consumer acceptance and market value [1]. Umami is primarily driven by water-soluble compounds, including free amino acids and nucleotides such as inosine monophosphate [2]. These compounds are key contributors to pork flavor and are increasingly used as target traits in meat quality evaluation [3,4]. Taste-sensing electronic tongue systems provide an objective method for quantifying umami intensity in meat by measuring the electronic signals of these compounds and have been applied in beef and pork flavor studies [5,6].

The intestinal microbiota has emerged as an important regulator of lipid metabolism, muscle traits and meat quality in pigs [7,8]. Differences in intestinal microbial composition between lean and fat pigs have been linked to altered intramuscular fat deposition, marbling and flavor-related traits [8]. Microbial metabolites, particularly short chain fatty acids (SCFAs) such as butyrate, are thought to be key mediators linking intestinal microbiota to pig muscle metabolism [9]. SCFA availability has been shown to affect carcass traits, lipid metabolism and meat quality in growing pigs [10]. These findings suggest that specific gut microbes and their metabolic pathways may contribute to the formation of favorable flavor attributes.

Within the hindgut, the cecum functions as a major fermentation chamber where complex carbohydrates and nitrogenous substrates are converted into SCFAs and other bioactive metabolites [11]. Therefore, characterizing the cecal microbiota is important for understanding how microbial communities affect the host’s metabolism and meat quality. 16S rRNA gene sequencing has been widely used to describe the diversity and composition of intestinal bacterial communities along the gastrointestinal tract [12]. However, 16S rRNA gene sequencing mainly provides species information, and its resolution is limited when it comes to the functional potential of microorganisms [13]. The high resolution of shotgun metagenomic sequencing can overcome this limitation [14,15,16]. By integrating 16S rRNA gene and metagenomic functional analysis, it is possible to link specific species shifts to their underlying metabolic capacities, providing a more comprehensive view of how the cecal microbiome may influence host phenotypes than either approach alone [13,17,18,19].

In this study, we used a taste-sensing electronic tongue system to measure the umami content of the longissimus dorsi muscle from Duroc × Landrace × Yorkshire (DLY) pigs and classified these pigs into low, medium, and high umami groups. We performed 16S rRNA gene sequencing of cecal contents to analyze cecal microbial composition, microbial diversity, and the associations between the microbiota and pork umami. We also conducted shotgun metagenomic sequencing of cecal contents to examine differences in the abundance of microbial functional genes. This integrated approach is expected to provide insight into microbial pathways underlying pork flavor formation and to identify potential microbial targets for improving meat umami.

2. Materials and Methods

2.1. Sample Collection

In this study, 96 commercial DLY pigs (179 days) were selected from a commercial farm in Guigang, Guangxi, China. All pigs were raised under identical diets and management conditions to minimize environmental variation in gut microbiota composition and meat quality traits. After slaughter, muscle samples were collected from the longissimus dorsi of each pig, immediately placed on dry ice and transported to −20 °C for storage. Cecal contents were extruded while avoiding exposure to air, frozen in liquid nitrogen, and subsequently stored at −80 °C until DNA extraction for 16S rRNA gene sequencing and shotgun metagenomics sequencing.

2.2. Taste-Sensing Electronic Tongue System Sample Preparation, Measurement and Sample Grouping

Approximately 50 g of longissimus dorsi muscle was taken from each pig and homogenized in a blender for 1 min. Then, 250 mL of purified water was added and the mixture was further homogenized for 1 min to obtain a uniform extract. The homogenate was transferred to centrifuge tubes and centrifuged at 3000 rpm for 10 min. After phase separation, the supernatant was carefully collected and used as the sample solution for electronic tongue analysis. Taste characteristics were evaluated using the taste-sensing electronic tongue system SA-402B Plus (Intelligent Sensor Technology Co., Ltd., Atsugi, Japan). Sample solutions were measured at 27 °C, and the sensor array automatically recorded the taste responses. Each sample was analyzed in triplicate and the mean umami value obtained from the umami sensor (AAE) was used for subsequent umami grouping and statistical evaluation.

Based on the umami scores, all 96 pigs were evenly classified into three umami groups for shotgun metagenomic analysis, including high umami (high), medium umami (medium), and low umami (low), using a rank-based tertile approach in which samples were ordered according to umami intensity and divided into three equal-sized groups. Cecal samples from 95 of these pigs were additionally subjected to 16S rRNA gene sequencing using the same three umami groups for community-level analyses.

2.3. 16S rRNA Gene Library Construction, Quality Control, and Sequencing

Genomic DNA was extracted from cecal content samples using the CTAB or SDS method. The extracted DNA was then subjected to quality assessment by 1% agarose gel electrophoresis (120 V, 30 min) to evaluate band integrity and nucleic acid purity. Qualified DNA samples were diluted with sterile nuclease-free water to a final concentration of 1 ng/μL. The diluted genomic DNA was used as a template for PCR amplification of the target 16S rRNA gene region using barcode-tagged primers. The PCR products were verified by 2% agarose gel electrophoresis and subsequently purified using magnetic bead-based purification kits (e.g., AMPue XP) to remove primer dimers and non-specific amplification products. Barcoded PCR products from different samples were pooled in equimolar amounts, and the Kinnex long-read sequencing technology was employed for library preparation. The purified library was quantified using a Qubit 4.0 Fluorometer (Thermo Fisher Scientific, Waltham, MA, USA), and its size distribution was assessed using an Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA) to ensure the primary peak fell within the expected range (1.5–3 kb). Libraries that meet the quality standards were sequenced on the Revio platform (Pacific Biosciences, Menlo Park, CA, USA) using Single Molecule, Real-Time (SMRT) sequencing to obtain high-quality long-read data.

2.4. 16S rRNA Gene Sequencing Data Analysis

16S rRNA gene sequencing data from cecal contents were processed and analyzed using QIIME2 (version 2024.2) [20]. Raw paired-end reads targeting the bacterial 16S rRNA V3–V4 region were trimmed with cutadapt to remove primer sequences (forward primer: CCTAYGGGRBGCASCAG; reverse primer: GGACTACNNGGGTATCTAAT), merged with vsearch, and quality-filtered using the q-score method. After truncation to a uniform length, denoising was performed with the deblur algorithm to obtain amplicon sequence variants (ASVs) [21]. ASV abundance tables and representative sequences. Rarefied ASV tables were then used to calculate alpha diversity indices and beta diversity metrics, which were visualized by principal coordinates analysis (PCoA); group differences among umami categories were evaluated using Kruskal–Wallis tests [22] and ANOSIM. At the species level, analyzed by Linear Discriminant Analysis Effect Size (LEfSe) to identify differentially abundant species with LDA scores among different umami groups [23]. Spearman correlation [24] was applied between relative abundances and umami scores to identify species significantly associated with umami.

2.5. Metagenomic Library Construction and Sequencing

For each sample, 0.2 μg of genomic DNA extracted from cecal contents was used as input for metagenomic library preparation. Sequencing libraries were constructed using the NEBNext Ultra DNA Library Prep Kit for Illumina (NEB, Ipswich, MA, USA; Catalog #: E7370L) according to the manufacturer’s instructions, and unique index barcodes were added to each sample. Briefly, genomic DNA was fragmented by sonication to an average size of approximately 350 bp, followed by end repair, 3′-A tailing and ligation with full-length Illumina adapters. The adapter-ligated DNA was then enriched by limited-cycle PCR, and the amplified products were purified using the AMPure XP system (Beckman Coulter, Beverly, MA, USA). Library size distribution and integrity were assessed on an Agilent 5400 system (Agilent Technologies, Santa Clara, CA, USA), and library concentrations were quantified by qPCR, with a target final concentration of approximately 1.5 nM. Qualified libraries were pooled in equimolar amounts and sequenced on an Illumina platform using a 150 bp paired-end (PE150) strategy at Novogene Bioinformatics Technology Co., Ltd. (Beijing, China), according to the required effective library concentration and target data output.

2.6. Metagenomic Data Processing and Functional Analysis

Raw metagenomic reads from cecal contents were processed using KneadData [25] for initial quality control and host read removal. Briefly, low-quality bases and short reads were trimmed with Trimmomatic (version 0.39) [26] using the parameters SLIDINGWINDOW:4:20 MINLEN:60, and reads mapped to the pig reference genome GCF_000003025.6_Sscrofa11.1 were removed to obtain high-quality, host-depleted clean data. The clean reads were then de novo assembled with MEGAHIT (v1.2.9) [27] using a multi-k-mer strategy (--k-list 37,47,57,67,77,87,97) and a minimum contig length of 1500 bp. Functional genes were predicted from assembled contigs with Prodigal (v2.6.3) [28] in metagenomic mode, and the predictive gene set was clustered at 95% sequence identity using CD-HIT (v4.8.1) [29] to construct a functional gene catalog. Gene abundances in each sample were quantified with Salmon (v1.10.3) [30], and functional annotation was performed against the eggNOG databases [31]. Based on the annotated gene abundance tables, gene abundance difference analysis among the low, medium and high umami groups was conducted using DESeq2 [32] in R. All additional statistical analyses, data integration and visualization were carried out with custom Python (v3.10.12) and R scripts.

3. Results

3.1. 16S rRNA Gene Analysis Reveals the Composition of the Cecal Microbiota and the Species Related to Umami

We used the taste-sensing electronic tongue system to assess the umami flavor in pork and divided 96 samples into three groups evenly (Table S1). A box plot illustrates the distribution of umami scores across the low, medium, and high groups. The low group shows values near zero, indicating no umami flavor, while the medium group exhibits an intermediate level of umami, suggesting a preliminary presence of umami. The high group is characterized by rich umami, with significantly higher scores compared to both the medium and low groups. Statistical analysis reveals significant differences in umami scores among all three groups (Figure 1a).

To examine whether there are differences in the composition of the microbiota among different umami groups, we used 16S rRNA gene sequencing and found that at the genus level, the cecal microbial communities exhibited a certain structure, mainly consisting of various common anaerobic microorganisms in the intestines, and the relative abundance composition patterns of the low, medium, and high umami groups were similar (Figure S1). We conducted alpha and beta diversity analyses. There was no significant difference in Alpha diversity among the three groups (p > 0.05) (Figure 1b). In contrast, the PCoA revealed a significant separation between the bacterial community composition of the high umami group and the other two groups. ANOSIM indicated significant differences between the high umami group and the other groups (p < 0.05), whereas the difference between low and medium groups was not significant (p > 0.05) (Figure 1c).

Using LEfSe, we identified several species that were differentially enriched among all three umami groups, with distinct species profiles characterizing the high, medium and low umami groups (Figure 1d). In the high umami group, Agathobacter ruminis, Turicibacter sanguinis, XBB1006 sp900115795, and CAG-632 sp900539185 were significantly enriched. In the medium umami group, Holdemanella biformis, Phocaeicola_A plebeius, CAG-349 sp003539515, Hominisplanchenecus_A faecis, Parabacteroides_B_862066 chinchillae, Lawsonibacter sp000492175, and Phocaeicola_A vulgatus were mainly enriched. In the low umami group, CAG-177 sp003538135, Helicobacter_D_480075 rodentium, Phascolarctobacterium succinatutens_B, Treponema F_986479 brennaborense, Desulfovibrio piger_A, Helicobacter_D_480075 canadensis, Mucispirillum schaedleri, Anaerobiospirillum_A succiniciproducens, and Mailhella massiliensis were significantly enriched.

To examine which cecal microorganisms are associated with pork umami, we performed Spearman correlation analysis and identified 37 species that were significantly correlated with pork umami (Table S1), including 11 species showing positive correlations and 26 species showing negative correlations. We selected the top ten correlated species and found that CAG-632 sp900539185 was significantly positively correlated with pork umami. A grouped box plot based on the relative abundance of CAG-632 sp900539185 showed that its relative abundance in the high umami group was significantly higher (p < 0.05) than that in the other groups. With the LEfSe, Spearman correlation, and relative abundance analyses, we suggest that CAG-632 sp900539185 may be a key candidate species associated with pork umami.

3.2. Functional Analysis of Metagenomes in Different Umami Groups

Using metagenomic functional abundance differential analysis, we compared the functional profiles of the cecal microbiota among low, medium, and high umami groups. Significant differences were observed in microbial functions, particularly those involved in carbohydrate transport, SCFA metabolism, nitrogen utilization, and cell-surface structures.

In the comparison between the high and low umami groups, KOs enriched in the high group were primarily related to biofilm formation, bacterial adhesion, and SCFA metabolism. KOs included K10924 (MSHA pilin protein), K13733 (fibronectin-binding protein B), and K14195 (surface protein G), which are involved in bacterial invasion and biofilm formation. Additionally, K10237 (trehalose/maltose transport system permease) and K00634 (phosphate butyryltransferase) were enriched in the high group, reflecting enhanced SCFA metabolism. Other KOs such as K07173 (S-ribosylhomocysteine lyase) and K03101 (signal peptidase II) were also upregulated, suggesting significant shifts in protein export and sulfur amino acid metabolism in the high umami pigs (Figure 2a).

In the comparison between the high and medium umami groups, the high group showed enrichment in KOs related to nitrogen metabolism, redox regulation, and RNA processing. K00372 (assimilatory nitrate reductase) and K02567 (nitrate reductase) were significantly more abundant in the high group, indicating an enhanced capacity for nitrogen utilization. K03342 (para aminobenzoate synthetase), a key enzyme involved in folate biosynthesis, was also enriched in the high umami group, indicating enhanced capacity for carbon-related metabolic processes. Additionally, K07217 (manganese catalase) and K17675 (ATP-dependent RNA helicase) were upregulated, reflecting increased detoxification and RNA degradation processes (Figure 2b).

In the comparison between the medium and low umami groups, KOs enriched in the medium group were associated with carbohydrate metabolism, nitrogen utilization, and cell-envelope biogenesis. K11960 and K11961 (urea transport system permease) were more abundant in the medium group, suggesting enhanced urea uptake. Carbohydrate-active enzymes, such as K00729 (dolichyl-phosphate beta-glucosyltransferase) and K01196 (glycogen debranching enzyme), were also elevated, indicating improved carbohydrate metabolism. Additionally, K14680 (putative RNA ligase) and K02679 (prepilin peptidase) were upregulated in the medium group, suggesting shifts in nucleic acid metabolism (Figure 2c).

Metagenomic functional analysis profiling revealed distinct microbial functional signatures associated with different umami intensities, particularly in pathways related to SCFA metabolism, nitrogen utilization, carbohydrate metabolism, and bacterial adhesion. These findings describe functional features of the cecal microbiome that are associated with variation in pork umami characteristics.

4. Discussion

Using 16S rRNA gene sequencing to analyze the microbial community composition among different umami groups, we observed no significant differences in α diversity, as reflected by comparable Shannon indices across groups. In addition, although Bray–Curtis PCoA and ANOSIM indicated statistical differences between certain pairwise comparisons, some overlap among samples was observed. All groups were characterized by a similar set of core microbial genera with comparable relative abundances such as Prevotella, Lactobacillus and Clostridium. These results indicate that the changes in the pork umami may not be strongly associated with the composition and diversity of microorganisms [33,34,35].

Through LEfSe analysis, we identified species that exhibited significant differences in abundance among different umami groups. Spearman correlation analysis identified the species that were significantly related to the pork umami. among them, cag-632 sp900539185 was significantly enriched in the high umami group. Its relative abundance gradually increased from the low umami group to the high umami group, and it was significantly positively correlated with the intensity of umami. These results suggest that CAG-632 sp900539185 may represent a key member of the umami-associated cecal microbiota. Some studies have shown CAG-632–related lineages have been implicated in carbohydrate fermentation potential, including contributions to carbohydrate-active enzyme profiles linked to fiber and polysaccharide degradation. Also they have been identified as key species in cecal microbial under conditions associated with altered propionate and butyrate metabolism [36,37]. CAG-632 sp900539185 is an uncultured species [38,39,40]; its metabolic capabilities were inferred through genome annotation. An increasing number of studies have shown that many microorganisms in the pig intestine remain uncultured, including the species that may play roles in host metabolism [38,41,42]. However, cultivation of these uncultured microorganisms and experimental validation of their functions are still required to confirm their biological roles [43].

Functional gene differences among different umami groups were investigated through metagenomic functional analysis. We found that differences in pork umami are not associated with functional genes responsible for the synthesis of umami compounds. These genes include those involved in substrate utilization, SCFAs metabolism, nitrogen metabolism, folate biosynthesis, and other supporting metabolic processes. Such functions can influence fermentation efficiency, carbon and nitrogen flux, and the metabolic context in which amino acid and nucleotide metabolism occurs [44,45]. These findings suggest that the cecal microbiota may be indirectly associated with pork umami through indirect involvement in interconnected biological and chemical pathways, rather than through direct production of umami compounds [46].

Compared with the low umami group, genes involved in nitrogen source utilization (K11960 and K11961) were significantly upregulated in the medium umami group, suggesting that the cecal microbiota in this group may contribute to umami by fermenting undigested proteins to produce SCFAs [47] or amino acids [48]. Urea recycling and protein fermentation in the hindgut can generate ammonia and other nitrogenous intermediates, which may be incorporated into microbial and host amino acid pools through nitrogen assimilation pathways [49,50]. Such shifts in intestinal nitrogen flux may influence systemic amino acid availability and nucleotide metabolism in peripheral tissues. Since free amino acids and nucleotides are key contributors to umami perception in meat [3], alterations in microbial nitrogen metabolism may offer a biologically plausible explanation for the association between cecal functional capacity and pork umami characteristics.

Compared with the low umami group, genes involved in carbon source utilization (K20342 and K10237) and butyrate metabolism (K00634) were significantly upregulated in the High umami group. In addition, comparison between the High and Medium umami groups revealed significant enrichment of para aminobenzoate synthetase (K03342) in the High umami group. K03342 is involved in folate biosynthesis, and folate acts as a cofactor in carbon metabolism, directly participating in the synthesis of nucleotides such as purines and thymidylate, as well as in the homeostasis and metabolism of amino acids including serine, glycine, and methionine [51]. These results indicate that the cecal microbiota in the high umami group may contribute to umami not only through carbon source utilization and SCFA production [52], but also indirectly by supporting the synthesis of umami-related compounds, such as nucleotides and amino acids, through participation in folate biosynthesis [53].

SCFAs, particularly acetate, propionate, and butyrate, have been reported to regulate host energy metabolism, lipid deposition, and muscle metabolic characteristics through systemic circulation and endocrine signaling pathways [54,55]. In addition, gut microbial nitrogen metabolism may influence host amino acid availability and nucleotide turnover via host-mediated metabolic regulation. Since free amino acids and nucleotides are key contributors to umami perception [53], such host microbiota metabolic interactions provide a biologically plausible framework linking cecal microbial functional capacity to muscle flavor traits. However, as no direct measurements of SCFAs, circulating metabolites, or muscle amino acid and nucleotide profiles were performed in this study, these interpretations remain inferential rather than experimentally validated.

It should also be noted that functional inferences in this study were based on metagenomic gene abundance rather than direct measurements of gene expression or metabolite levels. Future studies integrating transcriptomic or metabolomic analyses will be required to validate the functional implications suggested by the metagenomic data.

5. Conclusions

In conclusion, by combining taste-sensing electronic tongue system measurements with 16S rRNA gene and shotgun metagenomic sequencing of cecal contents, this study provides an initial picture of how cecal microbes and their functional genes are associated with variation in pork umami. The microbial diversity and core community composition were broadly similar among all three umami groups, but the uncultured species CAG-632 sp900539185 and a focused set of KOs involved in carbohydrate utilisation, SCFAs and butanoate metabolism, nitrogen metabolism and cell surface structures were enriched in pigs with higher umami scores. These findings suggest that differences in pork umami are more likely associated with targeted species and functional features of the cecal microbiome than the microbial community.