All procedures involving animals were reviewed and approved by Animal Experiment Committee of Kagawa University, Japan (approval No. 22647; approval date: 23 August 2022). The study was conducted in accordance with the Act on Welfare and Management of Animals of Japan and the institutional guidelines for the care and use of laboratory/experimental animals, and it complied with internationally accepted principles for animal research (Replacement, Reduction, and Refinement). The trial was designed as an 8-week exploratory randomized controlled feeding study in weaned piglets, and the minimum number of animals required to address the study objectives was used. Piglets were monitored at least daily by trained personnel for general health, behavior, and feed and water intake, with enhanced monitoring during the immediate post-weaning period. Housing, handling, and sampling procedures were implemented to minimize pain and distress; blood collection from the jugular vein was performed by experienced staff using sterile equipment and appropriate restraint to reduce handling time. Humane endpoints were predefined before study initiation, including but not limited to severe lethargy, persistent anorexia, dehydration, uncontrolled diarrhea, severe lameness, respiratory distress, or body condition deterioration; any animal meeting endpoint criteria would have received veterinary assessment and, if necessary, been withdrawn from the study and humanely managed according to institutional policy. No surgical procedures were performed, and no euthanasia was planned as part of the protocol; if euthanasia became necessary for welfare reasons, it would have been carried out using IACUC-approved methods. The study is reported in accordance with the Animal Research: Reporting of In Vivo Experiments 2.0 guidelines to ensure transparency and reproducibility.

The study was conducted from August 2022 to February 2025 at the farm and laboratory of Kagawa University and the laboratory of Setsunan University, located in Japan.

This work was conducted as an 8-week exploratory randomized controlled feeding trial to investigate the effects of dietary 5-ALA on gut microbiota, oxidative stress, and immune status in weanling piglets.

Twelve weaned piglets (Landrace × Large White × Duroc), 28 days old and originating from three litters, were enrolled. At weaning, piglets weighed 6.21 ± 0.90 kg. Using a stratified randomization approach based on sex and initial body weight (BW), animals were assigned to either the control (n = 6) or 5-ALA group (n = 6), ensuring balanced sex distribution (four castrated males and two females per group).

The 5-ALA group received 20 mg of 5-ALA per kg of feed, top-dressed using 2 g of Mitochon Power® (1% 5-ALA; Bussan Animal Health, Osaka, Japan). Feed was freshly prepared daily to prevent degradation. Water and feed were provided ad libitum.

All diets were provided in mashed form and consisted of commercial prestarter (Winny Z), starter (Gattsuku Milk), grower (Nexcel Milk Stage C), and developer (Winny B) formulations from Nihon Nosan Kogyo Co., Ltd. (Yokohama, Japan). Diets met or exceeded nutrient requirements defined by the Japanese Feeding Standard for Swine. Guaranteed nutrient values were:

Prestarter: CP ≥ 24.5%, EE ≥ 6.0%, CF ≤ 2.0%, ash ≤ 7.0%, Ca ≥ 0.75%, p ≥ 0.65%

No amino acid or metabolizable energy values were provided by the manufacturer. No supplements other than 5-ALA were added. Proximate composition analyses were performed in-house, and results are presented in Table 1.

Piglets received prestarter feed during week 1, starter feed in week 2, grower feed for weeks 3–5, and developer feed for weeks 6–8.

Piglets were housed in conventional nursery pens (1,800 cm × 2,100 cm) under standard farm conditions. Mean ambient temperature and humidity were 22.6°C ± 4.1°C and 78.8% ± 8.7%, respectively. Feed intake was recorded daily at the pen-level, whereas physiological and microbiota measurements were collected individually.

BW, blood, and feces were sampled at 28, 56, and 84 days of age. Blood was collected from the jugular vein using sterile 21-gauge needles into heparinized tubes (Terumo, Osaka, Japan). Fresh feces were collected into sterile 15-mL tubes. Whole blood (3 µL) was used immediately for oxidative and inflammatory assays, while plasma and feces were stored at −80°C until analysis.

Genomic DNA was extracted from fecal samples using the QuickGene DNA Tissue Kit SII (Kurabo, Osaka, Japan). DNA quantity and purity were assessed via Nanophotometer (Implen, Kusatsu, Japan), ensuring A260/280 values of 1.8–2.0.

Library preparation followed Inoue et al. [20], including V3–V4 region amplification, dual indexing, and sequencing on the Illumina MiSeq platform. Successful amplification was confirmed using MultiNA electrophoresis (Shimadzu, Kyoto, Japan).

Sequence data processing followed Yoshimura et al. [21]. Reads were analyzed in QIIME2 (v.2021.11; https://docs.qiime2.org/2021.11/) [22] using DADA2 for quality control, denoising, and amplicon sequence variant inference. Taxonomic assignment employed the SILVA 138 database with a pre-trained Naïve Bayes classifier (confidence threshold = 0.7). Singletons and sequences classified as chloroplasts or mitochondria were removed. Phylogenetic tree construction used SATé-enabled phylogenetic placement [23]. Alpha and beta diversity indices were computed using standard QIIME2 pipelines.

Plasma MDA concentrations were quantified using a commercial assay kit (JaICA, Fukuroi, Japan) according to manufacturer protocols.

Oxidative stress and inflammatory status in whole blood were assessed using a dual chemilumi-nescence/fluorescence system (CFL-H2200; Hamamatsu Photonics, Hamamatsu, Japan) following Kazumura et al. [24, 25].

Superoxide radical (O₂•̅): measured via chemiluminescence (CL-O₂•̅).

Due to equipment delivery delays, measurements were not performed at day 28. Fresh whole blood was required; frozen samples were unsuitable. Each sample (n = 6 per group) was analyzed once using standard protocols.

Plasma IgG concentrations were measured using a porcine IgG ELISA kit (Bethyl Laboratories, Montgomery, TX, USA). Samples were diluted 1:500,000, and absorbance was measured at 450 nm using an iMark microplate reader (Bio-Rad, Hercules, CA, USA).

Each piglet served as an independent experimental unit. Normality of BW, CL-O₂•̅, FL-OCl̅, MDA, and IgG was assessed using the Shapiro–Wilk test. Non-normal data were analyzed using the Wilcoxon rank-sum test; otherwise, homoscedasticity was evaluated using the F-test, followed by Student’s or Welch’s t-test where appropriate.

Differences in α-diversity were tested using the Wilcoxon rank-sum test. β-diversity was evaluated via = Permutational Multivariate Analysis of Variance using Bray–Curtis distances. Differential taxonomic abundance was identified using LEfSe (Linear discriminant analysis; LDA score >3.0) [26]. Correlations between bacterial genera and physiological markers (CL-O₂•̅, FL-OCl̅, MDA, and IgG) were assessed using Spearman’s rank correlation.

Significance was set at p < 0.05, with trends recognized at 0.05 ≤ p < 0.10. Statistical analyses and figure generation were conducted in R (v.4.4.2, R Core Team, Vienna, Austria).

BW changes throughout the 8-week trial are presented in Table 2. At baseline (28 days of age), no significant difference was observed between the control and ALA groups. By 56 days of age, piglets receiving 5-ALA tended to exhibit higher BW than controls. At 84 days of age, although the difference was not statistically significant, the ALA group still showed a numerically greater BW. Overall BW gain across the experimental period did not differ significantly between groups (control: 30.11 ± 1.34 kg/8 weeks; ALA: 32.34 ± 1.69 kg/8 weeks; p = 0.33).

Feed intake was recorded at the pen-level, and average daily intake per animal was calculated by dividing total pen intake by the number of piglets housed together. All physiological and microbiota-related measurements were collected individually. FCR was calculated as group-level BW gain divided by total feed intake, with slightly higher FCR values observed in the ALA group throughout the study (Table 2).

The final denoized dataset comprised 1,017,459 high-quality sequences (median = 26,174.5 reads/sample; range: 13,261–44,422). Rarefaction curves for all samples are shown in Supplementary Figure 1.

Alpha diversity (Chao1 and Shannon indices) was highest at 56 days of age in both groups but did not differ between the control and ALA groups at any sampling point (Supplementary Table 1; Supplementary Figure 2a).

Beta diversity based on Bray–Curtis distances differed significantly among sampling ages (Supplementary Table 1; Supplementary Figure 2b; p < 0.05). Comparisons between groups revealed significant differences at 56 and 84 days (p < 0.05), but not at 28 days of age (Figure 1).

Substantial temporal shifts in microbial composition were observed across the study. For example, the abundance of Christensenellaceae R-7 group was highest at 28 days and decreased thereafter, whereas Prevotella showed the opposite pattern.

No taxonomic differences were detected between groups at baseline. At 56 days of age, six genera differed significantly between treatments (Figure 2a). Three genera, Coprococcus (0.26% vs. 0.65%), Solobacterium (0.36% vs. 0.65%), and Family XIII AD3011 group (0.13% vs. 0.38%), were more abundant in the ALA group. In contrast, Succinivibrio (2.68% vs. 0.30%), Mitsuokella (1.06% vs. 0.28%), and one additional genus were significantly lower in the ALA group.

At 84 days of age, eight genera differed significantly between groups (Figure 2b). The ALA group exhibited higher abundances of Prevotellaceae UCG-003 (0.32% vs. 0.68%), Prevotellaceae NK3B31 group (1.09% vs. 2.91%), and Phascolarctobacterium (0.70% vs. 2.09%). Conversely, Shuttleworthia (1.61% vs. 0.08%), Syntrophococcus (0.63% vs. 0.07%), Erysipelotrichaceae UCG-009 (0.21% vs. 0.02%), and three other genera were significantly lower in the ALA group.

Plasma MDA concentrations did not differ between groups at any sampling point (Table 3). At 56 days of age, leukocyte-derived oxidative (CL-O₂•̅) and inflammatory (FL-OCl̅) markers did not differ between groups. However, by 84 days, CL-O₂•̅ levels in the ALA group were approximately 3.5-fold lower than in controls (p < 0.05; control: 21.06 ± 4.66 × 10⁵ vs. ALA: 5.97 ± 0.95 × 10⁵). FL-OCl̅ levels also tended to be lower in the ALA group (p = 0.052; control: 11.18 ± 1.25 × 10³ vs. ALA: 7.37 ± 1.18 × 10³).

Plasma IgG concentrations did not differ at baseline or at 56 days of age. At 84 days, however, the ALA group demonstrated a significantly higher IgG level (1.52-fold increase; p < 0.05; control: 20.19 ± 3.20 mg/mL vs. ALA: 30.66 ± 3.23 mg/mL).

Because significant group differences in microbiota composition and oxidative/immune markers were observed at 84 days, correlation analyses were performed at this time point.

Seventeen genera exhibited significant correlations with CL-O₂•̅, while 4, 10, and 8 genera were significantly correlated with MDA, FL-OCl̅, and IgG, respectively (Figure 3; Supplementary Table 2). Genera correlated with CL-O₂•̅ included Prevotellaceae UCG-003, Prevotellaceae NK3B31 group, Phascolarctobacterium, Syntrophococcus, and Erysipelotrichaceae UCG-009.

This study evaluated the effects of dietary 5-ALA supplementation on the gut microbiota, oxidative stress, and immune responses of weanling piglets during the post-weaning period. Importantly, this is the first swine study to apply a simultaneous chemiluminescence/fluorescence system to freshly isolated leukocytes, allowing sensitive and real-time detection of oxidative and inflammatory markers. As this work was exploratory, the findings should be interpreted as preliminary yet hypothesis-generating.

Independent of dietary treatment, fecal microbiota composition (β-diversity) changed markedly between 28 and 84 days of age (p < 0.05). These shifts are consistent with the well-documented microbial transition following weaning, during which Bacteroides decrease due to reduced milk sugar availability, while Prevotella increase in response to plant-based dietary substrates [27, 28]. Given these strong age effects, comparisons between the control and ALA groups were performed separately at each sampling point.

At 28 days of age, no meaningful differences in α- or β-diversity or in the abundance of bacterial genera were detected, confirming comparable microbiota profiles between groups at baseline. Similarly, systemic oxidative stress (as indicated by MDA) did not differ between groups at this time.

By 56 days, the ALA group exhibited significant differences in β-diversity and in the abundance of several bacterial genera relative to controls, although α-diversity remained unchanged. Notably, the ALA group showed higher levels of Coprococcus and Solobacterium, genera associated with butyrate and short-chain fatty acids (SCFA) production [29, 30] and previously linked to enhanced growth performance in pigs [31]. This microbial pattern coincided with higher BW in the ALA group at this stage. No group differences in oxidative or inflammatory markers were observed at 56 days.

At 84 days, differences in β-diversity and multiple bacterial genera persisted between the groups. The ALA group demonstrated significantly lower leukocyte-derived CL-O₂•̅ levels, indicating reduced oxidative stress. This reduction was accompanied by higher abundances of fiber-degrading or SCFA-associated genera, including Prevotellaceae NK3B31 group, Prevotellaceae UCG-003, and Phascolarctobacterium [29, 32].

These findings align with previous work by Zhang et al. [33], who reported decreases in Prevotellaceae NK3B31 and UCG-003 under chronic oxidative stress and an increase in Syntrophococcus, a genus reduced by 5-ALA in the present study. Here, several of these genera showed significant correlations with CL-O₂•̅, suggesting that modulation of the gut microbiota may influence leukocyte oxidative status.

MDA, a systemic oxidative stress marker reflecting lipid peroxidation in major organs [34], did not differ between groups. In contrast, leukocyte-specific oxidative responses were clearly modulated by 5-ALA. Because leukocytes constitute 70%–80% of immune cell populations within the gut-associated immune system [35], CL-O₂•̅ may more directly reflect microbiota–immune interactions than MDA.

Zhang et al. [33] similarly observed stronger correlations between microbial genera and oxidative markers in local tissues (liver, jejunum, ileum) than with MDA. Thus, CL-O₂•̅ may be a more sensitive indicator of microbially associated oxidative modulation.

At 84 days, the ALA group exhibited lower FL-OCl̅ values and significantly higher plasma IgG concentrations, indicating attenuation of inflammation and enhancement of humoral immunity. These findings support earlier reports demonstrating immunostimulatory effects of 5-ALA in pigs [10]. Importantly, leukocyte counts did not differ between groups, suggesting that differences in oxidative and inflammatory markers reflect altered leukocyte activation states rather than variations in cell numbers.

Some bacterial genera correlated with CL-O₂•̅ also correlated with IgG or FL-OCl̅, while others uniquely associated with immune markers alone. This pattern indicates that changes in microbial composition may contribute to immune alterations but that oxidative and immune pathways are not entirely overlapping.

The mechanisms linking 5-ALA to microbiota modulation remain uncertain. Pharmacokinetic data indicate that orally ingested 5-ALA is absorbed primarily in the liver, small intestine, and kidneys, with limited distribution to the large intestine [36]. Thus, direct utilization of 5-ALA by fecal bacteria is unlikely. Instead, microbiota changes may arise indirectly through 5-ALA–mediated improvements in iron metabolism [8, 37], mitochondrial function [3, 38], or intestinal physiology.

Because microbiota differences emerged at 56 days, prior to significant changes in oxidative and immune markers at 84 days, it is plausible that 5-ALA exerts initial effects at the gut level, particularly given that approximately 20% of ingested 5-ALA is retained within the small intestine. Altered nutrient flow or redox conditions in the intestine could subsequently reshape microbial communities. Future studies incorporating small intestinal functional assessments and microbial metabolite (e.g., SCFA) profiling are needed to clarify these pathways.

This study has several limitations inherent to its exploratory design. First, the sample size was relatively small, which may limit the statistical power and generalizability of the findings. Second, the panel of oxidative and immune markers assessed was narrower than that used in comparable studies, and no tissue-level evaluations, such as inflammatory or oxidative assessments in the small intestine or liver, were performed, as the primary focus was on fecal microbiota. Third, microbial metabolites, including SCFAs, were not measured, restricting the ability to infer the functional metabolic consequences of the observed microbiota shifts. To address these limitations, future research should employ larger cohorts, extended monitoring periods, and a comprehensive multi-omics framework (e.g., metagenomics, metabolomics, transcriptomics). Such approaches will help elucidate the mechanistic links among gut microbiota, microbial metabolites, oxidative stress, and immune responses across multiple tissues, and clarify functional intestinal changes associated with 5-ALA supplementation.