Piglet Gut Transcriptomes Link Ribosome Biogenesis to Post-Weaning Growth Differences

A transcriptomic reanalysis of ileal and hypothalamic tissues from nursery piglets has revealed that divergent post-weaning growth rates correlate with coordinated pathway shifts in mucosal protein synthesis machinery rather than single master genes or hypothalamic stress signaling. Weaning represents one of the most critical transitions in commercial swine production, frequently triggering intestinal mucosal barrier injury, villus atrophy, digestive enzyme reduction, diarrhea, and persistent variations in average daily gain that generate substantial economic loss. In a secondary exploration of publicly deposited RNA sequencing data, researchers Pingshuo Yang, Mingkang Wang, and Yutao Liu evaluated gene expression profiles across forty-eight libraries from twenty-four piglets selected for divergent average daily gain. Their investigation identified a small but reproducible group-level elevation in transcript abundance across ileal gene sets governing ribosome biogenesis, ribosomal RNA processing, translation, and downstream targets of the transcription factor MYC among piglets exhibiting superior growth rates. Rather than demonstrating an overt individual-level biomarker or active cellular mechanism, the authors reported that these transcriptomic signatures represent candidate mucosal correlates of intestinal recovery that require subsequent protein-level and functional validation.

To investigate the molecular basis underlying individual growth differences, the research team re-examined public sequencing dataset GSE324226 from BioProject PRJNA1429082, originally deposited by Ferreres-Serafini and colleagues. The experimental cohort comprised twenty-four crossbred pigs of similar weaning weight, averaging 5.65 kilograms with a standard deviation of 1.0 kilogram, weaned at approximately twenty-four days of age. The piglets originated from a broader study of one hundred and twenty animals housed in pens of three with sex segregation and weaning body weight blocking, receiving wheat, barley, and soybean meal pelleted diets ad libitum in two phases. From eight pens exhibiting extreme early post-weaning feed intake, twelve piglets were slaughtered at day seven to represent the acute post-weaning transition, and twelve piglets were slaughtered at day thirty-nine to represent the conclusion of the nursery phase. Six high-growth and six low-growth piglets were evaluated at each time point, meaning that day seven and day thirty-nine cohorts comprised entirely distinct biological individuals. Polyadenylated RNA had been extracted from the proximal ileum and whole hemisphere hypothalamus, yielding forty-eight paired-end libraries sequenced on an Illumina NovaSeq 6000 platform.

Because raw transcript reads had been probabilistically mapped by the original data depositors using Kallisto against the Sus scrofa Sscrofa11.1 transcriptome, more than half of the resulting count estimates contained non-integer values. The investigators rounded these unnormalized counts to the nearest integer to meet the modeling requirements of DESeq2, demonstrating through supplementary checks that rounding introduced a negligible median absolute shift of 0.10 counts and yielded identical pathway conclusions when compared against floor, ceiling, or limma-voom analyses of unrounded data. Following quality filtering that retained nineteen thousand and eighty-five genes detected with more than one count in at least ten samples, principal component analysis demonstrated that tissue identity was the overwhelming driver of variance. First principal component scores accounted for 82.6 percent of total transcriptomic variation and cleanly separated ileal samples from hypothalamic samples, while the second component accounted for only 4.4 percent of variance without distinct grouping by time point or growth category. Hierarchical clustering of Euclidean distances reaffirmed this clear bifurcation into two tissue-specific clusters without outlying libraries.

A critical technical dimension of the reanalysis involved inferring animal sex and adjusting for compositional imbalances across growth cohorts. By measuring the combined expression of the Y-linked genes ZFY, EIF1AY, and KDM5D, which exhibited a strictly bimodal distribution exceeding thirty counts per million in males and falling below two counts per million in females, the authors confirmed biological sex across forty-seven of the forty-eight sequencing libraries. This inspection confirmed that sex was unevenly distributed among the average daily gain groups due to the sex-segregated pen design of the primary study, including an acute imbalance at day thirty-nine where three high-growth piglets were male while Low-growth cohorts had zero or one male. Inferred sex remained concordant between the hypothalamic and ileal libraries for twenty-three animals, whereas one low-growth ileal sample showed male expression despite female donor origin. Because an independent Hungarian assignment matching fifteen circulating blood biomarkers confirmed the library pairing identity, the researchers retained the sample while incorporating biological sex as an explicit covariate across all formal linear and interaction models.

Testing for individual differentially expressed genes between high-gain and low-gain cohorts revealed exceedingly few statistically significant single-gene differences after Benjamini-Hochberg adjustment, with only four ileal genes and sixteen hypothalamic genes crossing false discovery rate thresholds at day thirty-nine in primary factorial models. Furthermore, no individual gene exhibited a significant main effect of growth group in the hypothalamus, and only two genes did so in the ileum. In response to the limited statistical power inherent in small-sample animal studies, the authors employed rank-based gene set enrichment analysis across Wald statistics to detect weak, coordinated shifts across functional pathways. This approach uncovered a pronounced, coordinated shift toward higher expression of ribosomal and translational machinery in the ileum of high-performing piglets. At day thirty-nine, the leading Gene Ontology biological process term was ribosome biogenesis, which achieved a normalized enrichment score of 2.73 and an adjusted significance value of 8.3 times ten to the negative thirteenth power across one hundred and twenty-two annotated genes, supported by significant enrichments in rRNA processing, translation, and ribonucleoprotein complex assembly.

Sensitivity analyses confirmed that this ileal ribosomal enrichment remained robust across multiple analytical conditions, persisting across twelve separate leave-one-animal-out iterations with enrichment scores ranging between 2.35 and 2.93 and holding true regardless of whether biological sex was included in the model. Concordant positive shifts were observed across alternative curated collections, including Hallmark MYC target gene sets, KEGG ribosome sets, and Reactome translation pathways. However, the biological nature of this signal was characterized by modest individual gene effects rather than dramatic transcriptional activation. Among the sixty-one leading-edge genes contributing to the ribosome biogenesis enrichment, all displayed positive log2 fold changes, but with a median increase of only 0.26, and 96.7 percent had absolute fold changes below 0.5. None reached individual statistical significance after multi-testing correction. Because the signal represents a cumulative shift in gene ranks across the cohort, single-sample gene set variation analysis scores overlapped considerably between groups and failed to differentiate individual animals, proving that the pattern constitutes a subtle group-level signature rather than a predictive individual diagnostic marker.

The study also addressed the possibility that mucosal cellular composition differences drove the ileal enrichment, as proliferative intestinal epithelial cells naturally express abundant ribosome biogenesis transcripts. By evaluating expression signatures of stem cells, proliferating cells, enterocytes, goblet cells, enteroendocrine cells, and six immune and stromal lineages, the authors determined that none of the eleven estimated cell-type scores differed significantly between growth cohorts at either post-weaning time point. Furthermore, refitting the ileal differential expression models with the principal components of cell composition as additional covariates preserved the strong enrichment of ribosome biogenesis, translation, and MYC targets with normalized enrichment scores exceeding 2.95. This confirmed that the observed transcriptional pattern was not an artifact of varying mucosal epithelial or immune cell proportions. Comparisons across independent public datasets indicated that while small-intestinal transcriptomes from normal-birth-weight piglets also showed elevated translation and MYC target expression relative to low-birth-weight piglets, acute forty-eight-hour fasting in young pigs actually increased ribosome biogenesis transcripts while suppressing energy pathways, suggesting that the post-weaning gain association cannot be explained merely by short-term differences in feed consumption.

In marked contrast to the sustained ribosomal signals, ileal metabolic pathways displayed a pronounced group-by-time statistical interaction characterized by early divergence followed by late convergence. At day seven post-weaning, high-growth piglets exhibited significantly higher expression of gene sets governing cholesterol homeostasis, the tricarboxylic acid cycle, and oxidative phosphorylation, including specific steroidogenic enzymes such as CYP51A1, DHCR7, FDFT1, HMGCS1, LSS, MSMO1, SC5D, and SQLE. By day thirty-nine, expression levels in low-gain piglets had risen to match the high-gain cohort, producing negative interaction scores ranging from negative 2.01 to negative 2.55 across gene set enrichment models. The authors noted that because average daily gain serves as a proxy for feed intake, these early metabolic elevations in high-gain piglets likely reflect higher immediate nutrient consumption during the stressful acute post-weaning window. Rather than signaling an adaptive catabolic shift in slower-growing animals, the temporal pattern represents a delayed convergence in mucosal energetic and lipid synthetic gene expression as the lower-performing pigs eventually adapted to solid nursery feed.

Hypotheses predicting divergent hypothalamic stress responses or altered cross-tissue communication were not supported by the transcriptomic evidence once formal statistical corrections were applied. Although transient hypothalamic endoplasmic reticulum stress and unfolded protein response pathways had been prioritized as candidate regulators of post-weaning energy balance, inferred activity of the transcription factor ATF6 and the master regulator MYC showed no significant differences between growth groups across either post-weaning time point. While ATF6 activity declined generally from day seven to day thirty-nine in both groups, the unfolded protein response gene sets showed no enrichment, and individual gene interactions were nominal and sensitive to model specification. Similarly, weighted gene co-expression network analysis failed to identify module eigengenes associated with growth that were enriched for ribosomal machinery. Evaluations of cross-tissue coupling across seven gut-brain ligand-receptor pairs—including somatostatin, glucose-dependent insulinotropic polypeptide, and leptin—yielded no statistically significant differences in partial correlations between high- and low-growth cohorts, demonstrating that transcript levels alone cannot establish endocrine signaling or functional gut-brain communication.

In detailing the limitations of their work, the investigators stressed that secondary transcriptomic reanalyses are fundamentally observational and hypothesis-generating. Because day seven and day thirty-nine sampling involved distinct groups of slaughtered animals, longitudinal recovery cannot be traced within individual subjects, and the lack of primary individual-level feed intake and litter identifiers leaves open the possibility of residual confounding. Crucially, transcript abundance does not measure protein translation, ribosomal assembly, circulating hormone levels, or mucosal energetic state. The authors concluded that while coordinated ileal ribosome biogenesis and delayed metabolic convergence represent clear transcriptional correlates of post-weaning growth vigor, definitive conclusions regarding swine mucosal recovery require future prospective studies incorporating histological crypt-villus morphometry, protein synthesis measurements, and targeted western blotting in independent animal cohorts.

Subject of Research: Transcriptomic correlates of post-weaning average daily gain in piglet ileum and hypothalamus

Article Title: Ileal transcriptional signatures associated with post-weaning growth in piglets: an exploratory comparison with hypothalamic profiles

Article References: Yang, P., Wang, M., & Liu, Y. (2026). Ileal transcriptional signatures associated with post-weaning growth in piglets: an exploratory comparison with hypothalamic profiles. Frontiers in Veterinary Science, 13, Article 1990635. https://doi.org/10.3389/fvets.2026.1990635

Image Credits: AI Generated

DOI: 10.3389/fvets.2026.1990635

Keywords: piglet weaning, average daily gain, ileum transcriptome, hypothalamus, ribosome biogenesis, RNA sequencing, DESeq2, gene set enrichment analysis, MYC target genes, translation regulation, porcine intestinal health, tricarboxylic acid cycle, cholesterol biosynthesis, endoplasmic reticulum stress, ATF6 pathway, feed intake proxy, Sscrofa11, weaning stress