Introduction

In the commercial setting, piglets are typically weaned before their digestive and immune systems have fully matured1. The sudden transition from sow’s milk to a grain-based diet leads to reduced feed intake, intestinal inflammation, nutrient malabsorption, and microbial imbalances, which together impair growth performance2. This intestinal dysfunction fosters the growth of Escherichia coli, a primary cause of post-weaning diarrhoea (PWD)3. Zinc oxide (ZnO) supplementation has been shown to mitigate PWD and support growth4. However, the European Union banned the use of in-feed antimicrobials, including ZnO, in 2022 due to concerns around the environment and antimicrobial resistance5. Consequently, identifying nutritional strategies to enhance growth and intestinal health post-weaning is critical.

While antimicrobials inhibit microbial proliferation, organic acid (OA) supplementation encourages the growth of beneficial bacteria that outcompete pathogenic bacteria in the gut6. Supplementing piglets with OA modulates the intestinal microbiome, reduces inflammation and diarrhoea, and improves growth performance7. However, the effects of OA depend on factors such as the type of acid used, inclusion levels, and interactions with other dietary elements8. Additionally, OAs, particularly propionic acid, serve as effective grain preservatives9.

Cereal grains are essential components of pig diets10. In temperate, humid climates with significant rainfall, grain moisture content can reach 20–25% during harvest11, promoting fungal growth and mycotoxin production12. Grain can be mechanically dried13 to reduce moisture content, thereby limiting microbial load and preserving quality14. However, inconsistencies in mechanical drying can lead to uneven moisture content and elevated mycotoxin levels15. Moreover, mechanical drying is energy-intensive16 and generates substantial greenhouse gas emissions17.

Preserving grain with OA can reduce bacterial and fungal growth without affecting its chemical composition. Additionally, OA-preserving improves post-weaned piglet growth and gut health more effectively than direct OA supplementation18. Implementing post-harvest preservation techniques with OA to maintain grain quality may enhance piglet growth post-weaning through a linked mechanism: first, by reducing microbial contamination in the grain, thereby improving its quality and nutritional value; and second, by directly enhancing gastrointestinal function and health in piglets through lower levels of harmful pathogens and increased nutrient absorption. Previous research has reported that pigs offered OA-preserved grain exhibited increased growth, ileal and total tract nutrient digestibility, and beneficial gut microbial shifts at day 35 post-weaning compared to those fed dried grain19. Therefore, it is essential to compare the benefits and underlying mechanisms of OA-preserved grain with ZnO during the critical post-weaning period.

This study aimed to assess the intestinal health and growth of newly weaned piglets offered grain preserved with OA at harvest against those offered conventionally dried grain. Additionally, it sought to compare the mechanisms by which OA-preserved grain and ZnO improve piglet performance by assessing various intestinal health and growth parameters. The hypotheses were, firstly, that OA-preserved grain would enhance piglet intestinal health and growth compared to conventionally dried grain, and secondly, OA-preserved grain would have similar effects on piglet intestinal health and growth post-weaning as ZnO supplementation, presenting a viable alternative to in-feed antimicrobials.

Results

Grain quality

Prior to preservation, the initial quality assessment of wheat showed a moisture content of 180 g/kg, a hectolitre weight of 73 kg/hL, and a thousand grain weight (TGW) of 49.8 g. For barley, the corresponding values were 183 g/kg for moisture, 61 kg/hL for hectolitre weight, and a TGW of 48.2 g. Comprehensive chemical and microbial profiles of both grains at the point of diet formulation are detailed in Table 1. Grains preserved with OA exhibited lower DM content compared to those that were dried. Dried wheat and barley also had higher concentrations of deoxynivalenol (DON) than their OA-preserved counterparts. In barley, higher levels of HT-2 toxin and ochratoxin A (OTA) were observed in the dried samples. Across both preservation methods and grain types, levels of aflatoxins B1, B2, G1, and G2 (≤ 10 μg/kg), fumonisins B1 and B2 (< 1000 μg/kg), T-2 and HT-2 toxins (< 50 μg/kg), and zearalenone (ZEA; < 250 μg/kg) remained below the detectable threshold (Tables 2, 3 and 4).

Table 1 The chemical and microbiological analysis of experimental grain after storage (g/kg as fed) unless otherwise stated.
Full size table
Table 2 The ingredient composition of dietary treatments (g/kg).
Full size table
Table 3 The analysed composition of experimental diets offered to pigs (g/kg unless otherwise stated).
Full size table
Table 4 Primer sequences for QPCR analysis of genes involved gastrointestinal health and function.
Full size table

Growth performance and faecal scores

Table 5 summarises the effects of the dietary treatments on growth metrics, body weight (BW), average daily gain (ADG), average daily feed intake (ADFI), gain-to-feed ratio (G:F), and faecal consistency scores (FS) over the post-weaning (PW) periods: days 0–21, 21–35, and the overall 0–35 days. One piglet per treatment group died during the first week PW due to causes unrelated to dietary intake.

Table 5 The effect of dietary treatment on growth performance and faecal scores (Least-square means with their standard errors).
Full size table

During days 0–21, significant interactions between the grain preservation method and ZnO inclusion were observed for ADG, G:F, and BW (P < 0.05). Piglets fed OA-preserved grains had improved ADG and G:F compared to those receiving dried grains; however, these differences were not present when ZnO was added. In contrast, ZnO supplementation enhanced BW in piglets on dried grain diets but had no measurable effect in the OA-preserved grain group. ADFI was higher in ZnO-supplemented diets relative to those without ZnO (386 vs. 337 g/day, SEM 0.007; P < 0.001).During days 21–35, significant grain preservation × ZnO interactions were also found for ADFI, ADG, and BW (P < 0.05). ZnO inclusion increased ADFI in pigs fed dried grain diets, but not in those fed OA-preserved grains. ZnO reduced ADG in the OA-preserved grain group but showed no effect in the dried grain group. BW was increased by ZnO in the dried grain group only. Pigs fed OA-preserved grains had significantly higher G:F compared to those receiving dried grains (747 vs. 680, SEM 0.026; P < 0.05).

During the overall period (day 0–35), an interaction between grain preservation and ZnO supplementation influenced both ADFI and ADG (P < 0.05). ZnO enhanced ADFI and ADG in dried grain diets but had no such effect in OA-preserved diets. Overall, pigs fed OA-preserved grains achieved superior G:F ratios compared to dried grain-fed pigs (775 vs. 709, SEM 0.024; P < 0.05).

Small intestinal morphology

The effects of dietary treatments on small intestinal morphology are provided in Table 6. Neither grain preservation method, ZnO inclusion, nor their interaction significantly influenced villus height (VH), crypt depth (CD), or VH:CD ratio in the duodenum (P > 0.05).

Table 6 The effect of dietary treatment on small intestinal morphology (Least-square means with their standard errors).
Full size table

In the jejunum, pigs receiving OA-preserved grain had shorter villi compared to those receiving dried grain (217.9 μm vs. 281.5 μm, SEM 15.03; P < 0.05). ZnO inclusion did not impact jejunal CD or VH:CD ratio (P > 0.05). In the ileum, a grain × ZnO interaction was evident for VH (P < 0.05); ZnO supplementation increased VH in the OA-preserved grain diet but not in the dried grain diet. ZnO also reduced CD (92.7 μm vs. 107.7 μm, SEM 4.71; P < 0.05) and improved VH:CD ratio (2.8 vs. 2.1, SEM 0.21; P < 0.05).

Gene expression in the gastrointestinal tract

Differential gene expression related to nutrient absorption, mucosal immunity, and barrier function is shown in Table 7.

Table 7 The effect of dietary treatment on the expression of genes involved in stomach functionality, nutrient transportation, inflammation and barrier function in the small intestine and colon (Least-square means with their standard errors).
Full size table

In the duodenum, there was a significant interaction between grain preservation and ZnO on MUC2 expression (P < 0.05); ZnO reduced MUC2 expression in the dried grain group but had no effect in the OA-preserved group. IL17 expression was reduced in pigs fed OA-preserved grains (1.03 vs. 1.74, SEM 0.213; P < 0.05). ZnO supplementation significantly decreased expression of FABP2, IL6, and IL17 compared to non-supplemented diets (P < 0.05). In the jejunum, ZnO inclusion upregulated SLC15A1 (1.98 vs. 0.99, SEM 0.280) and OCCLN (1.55 vs. 0.95, SEM 0.207), while reducing IL1A, IL1B, CXCL8, IL17, IL22, and NOX1 (P < 0.05). Grain preservation had no effect (P > 0.05). In the ileum, a significant interaction was noted for TNF expression (P < 0.05); ZnO reduced TNF levels in OA-preserved diets only. The OA-preserved grain diets were associated with lower FABP2 and FOXP3 expression (P < 0.05). ZnO increased SLC15A1 expression and reduced pro-inflammatory markers (IL1A, IL1B, CXCL8, IL17, IL22, NOX1) (P < 0.05).

In the colon, ZnO × grain preservation interactions were observed for IL6 (P < 0.05); ZnO reduced IL6 in the dried grain group only. ZnO supplementation increased SLC16A1 and MUC2 expression while decreasing IL1A, IL1B, CXCL8, TLR4, IL17, and NOX1 (P < 0.05). Grain preservation had no significant effect on colonic gene expression.

Bacterial richness and diversity

Alpha diversity metrics are summarised in Table 8, with beta diversity visualised in Figs. 1 and 2. No significant effects of grain preservation, ZnO supplementation, or their interaction were observed on richness or diversity indices (Observed, Fisher, Shannon, Simpson) in the ileum (P > 0.05).

Table 8 The effect of dietary treatment on measures of alpha diversity in ileal and colonic digesta (least square means with their standard errors).
Full size table
Fig. 1
Fig. 1
Full size image

Bacterial beta diversity of ileal digesta based on Permanova analysis and through visualisation using the Bray Curtis distance matrix and multi-dimensional scaling.

Fig. 2
Fig. 2
Full size image

Bacterial beta diversity of colonic digesta based on Permanova analysis and through visualisation using the Bray Curtis distance matrix and multi-dimensional scaling.

In the colon, OA-preserved grain diets led to a reduction in Shannon (3.31 vs. 3.55, SEM 0.060) and Simpson (0.94 vs. 0.96, SEM 0.004) diversity indices (P < 0.05). ZnO inclusion had no significant effect on colonic alpha diversity (P > 0.05). PERMANOVA analysis showed no significant differences in beta diversity for either intestinal region (P > 0.05).

Differential bacterial abundance

Phylum level

Table 9 presents the effects of grain preservation method and ZnO supplementation on the relative abundance of bacterial phyla. In the ileum, the predominant phyla were Firmicutes (~ 85.25%), Proteobacteria (~ 8.98%), Actinobacteria (~ 1.76%), and Bacteroidetes (~ 1.67%). Grain × ZnO interactions were detected for Proteobacteria (P < 0.05); OA-preserved grains reduced Proteobacteria abundance compared to dried grains, with no effect under ZnO supplementation. ZnO lowered Bacteroidetes abundance (1.03 vs. 2.24, SEM 0.421; P < 0.05). In the colon, Firmicutes (~ 91.60%) dominated, followed by Proteobacteria (~ 4.10%), and Bacteroidetes (~ 1.58%). Interactions were observed for Firmicutes and Proteobacteria (P < 0.05).

Table 9 The effect of dietary treatment the relative abundance of bacterial phyla in ileal and colonic digesta (mean % relative abundance with their standard errors).
Full size table

Family level

Table 10 presents the effects of grain preservation method and ZnO supplementation on the relative abundance of bacterial families. In the ileum, ZnO × preservation interactions affected Lactobacillaceae, Lachnospiraceae, and Enterobacteriaceae (P < 0.05). Ruminococcaceae were more abundant in OA-preserved diets, while ZnO increased Erysipelotrichaceae and decreased Oscillospiraceae (P < 0.05). In the colon, interactions were seen for Lachnospiraceae and Enterobacteriaceae. Lactobacillaceae abundance was reduced in OA-preserved grain diets but increased by ZnO supplementation (P < 0.01).

Table 10 The effect of dietary treatment the relative abundance of selected bacterial family in ileal and colonic digesta (mean % relative abundance with their standard errors).
Full size table

Genus level

Table 11 presents the effects of grain preservation method and ZnO supplementation on the relative abundance of bacterial genera. The OA-preserved grains increased Holdemanella and Faecalibacterium while reducing Escherichia in the ileum. ZnO also raised Faecalibacterium levels. In the colon, interactions affected Frisingicoccus, Gemmiger, and Faecalibacterium. Mediterraneibacter was higher and Lactobacillus lower in OA-preserved grain-fed pigs; ZnO reversed this trend.

Table 11 The effect of dietary treatment the relative abundance of selected bacterial genera in ileal and colonic digesta (mean % relative abundance with their standard errors).
Full size table

Volatile fatty acids (VFAs)

Table 12 details the volatile fatty acid (VFA) profile. In the ileum, ZnO × grain preservation interactions influenced butyrate (P < 0.05); ZnO reduced butyrate in OA-preserved grain diets only. The OA-preserved grains raised acetate and lowered propionate proportions (P < 0.05). In the colon, OA-preserved diets increased propionate (0.267 vs. 0.237, SEM 0.0094; P < 0.05), while ZnO decreased branched-chain fatty acids (BCFA) (0.036 vs. 0.051, SEM 0.0046; P < 0.05). No significant interaction effects were found in the colon.

Table 12 The effect of dietary treatment on the molar proportions and total concentrations of VFA in mmol/gram in pig ileum and colon (Least-square means with their standard errors).
Full size table

Discussion

The hypothesis of this study was that OA-preserved grain would improve piglet growth performance compared to conventionally dried grain by reducing the microorganism load in the grain, thereby increasing grain quality and directly influencing the gastrointestinal health and function of the pig. It was also hypothesised that OA-preserved grain would have similar effects on piglet growth and intestinal health compared to ZnO and could serve as an alternative to in-feed antimicrobials. The OA-preserved grain diet had the best BW, ADG, and G:F among all dietary treatments, while the beneficial effects of ZnO on growth performance were reduced after the initial 21 days PW. However, OA-preserved grain did not reduce FS to levels comparable with ZnO. These results indicate that OA-preserved grain could support piglet growth throughout the PW period. However, further dietary manipulation of OA-preserved grain diets is necessary to reduce FS to levels reported with ZnO.

In this study, OA-preserved grain had a lower DM and reduced levels of OTA and DON compared to dried grain, indicating improved grain quality. Post-weaned piglets are particularly vulnerable to mycotoxin contamination due to their physiological immaturity20. Mycotoxins impair immune function and increase pathogen susceptibility21, ultimately reducing voluntary feed intake and growth22. Specifically, OTA promotes oxidative stress and an inflammatory state23, while DON increases inflammation and digestive disorders24. Propionic acid and ammonium propionate are well-established grain preservatives26,27. Previous research suggests they provide greater benefits for gut health and growth when used as grain preservatives rather than as feed additives18. Therefore, improved grain quality could be partially responsible for the increased ADFI, growth performance, and improved gut microbiome observed in the OA-preserved grain diet. To confirm these protective effects, future studies could employ non-invasive mycotoxin biomarkers in piglet urine, faeces, or blood25.

Feed intake typically declines immediately PW28, leading to villous atrophy, crypt hyperplasia, elevated T-cell production29, and impaired intestinal development30. Reports on the effects of OA supplementation on piglet feed intake remain inconsistent, with studies showing no effect31,32,33,34, negative effects35, or positive effects36,37,38. Additionally, supplementation with glycerol polyethylene ricinoleate has been shown to increase ADG and improve feed efficiency  in broiler chickens39‚40. In the present study, OA-preserved grain increased ADFI compared to dried grain, aligning with previous findings18,19. Therefore, the observed increase in ADFI may contribute to the gut microbial shifts and improved growth performance seen in piglets fed the OA-preserved grain diet.

Interestingly, ZnO supplementation improved BW and ADFI in dried grain but did not provide any additional benefit on BW and ADFI in OA-preserved grain, consistent with previous research19. Although research examining ZnO supplementation in OA-preserved grain is limited, ZnO supplementation in non-OA-preserved diets has consistently increased growth performance41,42,43. However, the beneficial effect of ZnO on growth performance in pigs fed dried grain was diminished during the second phase of the experiment (D21–35). This finding supports the notion that ZnO efficacy declines after the initial 21 days PW44. In contrast, OA-preserved grain maintained its positive effects on growth throughout the experimental period, likely due to its influence on feed intake and modulation of the gut microbiome. Although both OA-preserved grain and ZnO improved growth performance, their effects on FS differed. OA-preserved grain did not improve FS compared with dried grain, whereas ZnO supplementation led to higher FS values relative to non-ZnO diets, consistent with previous findings45,46. The growth and FS improvements associated with ZnO may stem from enhanced gut morphology, reduced immune activation, improved nutrient absorption, and alterations in intestinal microbial populations.

The intestinal microbiome is vital in maintaining the physiological, immunological, and nutritional state of the pig. Thus, manipulating the gut microbiota could reduce reliance on antimicrobials in pig production47. In this study, piglets offered OA-preserved grain had reduced colonic Shannon and Simpson diversity compared to those offered dried grain. Reduced microbial diversity has been associated with diarrhetic piglets PW48, however, other studies have reported no correlation between reduced diversity and piglet gut health49. Pigs offered the OA-preserved grain diet had the highest abundance of ileal and colonic Faecalibacterium among all treatments. Faecalibacterium, specifically Faecalibacterium prausnitzii are indicators of increased intestinal health due to their immunomodulatory50 and butyrate-producing abilities51. Supplementing Faecalibacterium prausnitzii to mice increased tight-junction protein expression, reduced pro-inflammatory cytokine expression, and decreased the abundance of faecal Staphylococcus and Klebsiella52. Additionally, the OA-preserved grain was associated with reduced abundance of ileal Escherichia compared to dried grain. Escherichia coli is one of the main causative agents of PWD53. Previous research has demonstrated OA supplementation can reduce E. coli54,55 and increase faecal Faecalibacterium in weaned piglets56. Despite the reduction in colonic bacterial diversity, the microbial shifts associated with OA-preserved grain could contribute to enhanced growth performance and intestinal health.

Supplementing ZnO can beneficially modulate the PW gut microbiota by promoting the growth of beneficial bacteria, enhancing short-chain fatty acid (SCFA) production, and inhibiting the proliferation of pathogens57, thereby supporting a more stable and mature gut ecosystem. ZnO supplementation has been reported to reduce the abundance of Enterobacteriaceae, Escherichia, and Lactobacillus spp. in the small intestine58. In contrast, the present study found that ZnO supplementation increased colonic Lactobacillus abundance compared with non-ZnO diets, differing from previous reports that observed no effect of ZnO on intestinal Lactobacillus populations4,45. Notably, Lactobacillus species have been associated with improved nutrient digestion and barrier integrity, reduced diarrhoea59, and decreased expression of pro-inflammatory cytokines60, which could partially explain the positive effects observed here. The beneficial influence of ZnO on the gut microbiota was further reflected in the VFA profile. ZnO supplementation reduced the molar proportions of colonic BCFAs compared with non-ZnO diets, indicating decreased protein fermentation61. Excessive protein fermentation generates harmful metabolites62, disrupts intestinal barrier function, promotes inflammation63, and facilitates the proliferation of pathogenic bacteria40. Together, the microbial and VFA shifts observed with ZnO supplementation may underlie the initial improvements in growth performance in pigs fed the dried grain diet with ZnO, as well as the reductions in FS and intestinal inflammatory markers.

Post-weaning inflammation reduces intestinal barrier function, alters the microbiome64 and affects intestinal morphology65. Supplementing OAs has been reported to reduce the expression of pro-inflammatory cytokines66 and increase the expression of genes related to intestinal barrier function67. In the current experiment, OA-preserved grain reduced the relative expression of duodenal IL17 and ileal FOXP3 compared to dried grain. The supplementation of ZnO decreased the relative expression of multiple pro-inflammatory cytokines, including IL6 and IL17 in the duodenum, IL1A, IL1B, CXCL8, IL17, and IL22 in the jejunum and ileum, and IL1A, IL1B, IL6, CXCL8, and IL17 in the colon on day 10 PW. Additionally, NOX1, an indicator of oxidative stress, was reduced in the jejunum, ileum, and colon of ZnO-supplemented pigs. The expression of genes involved in barrier integrity, such as OCCLN in the jejunum and MUC2 in the colon, were increased in ZnO-supplemented pigs. These results suggest that the reduction in pro-inflammatory cytokine expression and increased expression of genes associated with the integrity of the intestine in ZnO-supplemented diets could contribute to reduced FS. Moreover, physiological changes during an immune response create a homeorhetic state, redirecting nutrients from growth to the immune system68. Thus, the reduced expression of pro-inflammatory cytokines may increase the energy available for growth, partially explaining the improved growth performance of the dried grain diet with ZnO in the initial 21 days.

The relationship between intestinal morphology and growth performance is well established, with morphological disruptions associated with reduced growth performance69. Interestingly, although there was no negative effect on growth, OA-preserved grain reduced jejunal VH compared to dried grain. Furthermore, OA-preserved grain reduced the relative expression of ileal FABP2 compared to dried grain. The ability of OA-preserved grain to sustain superior growth performance, despite evidence of reduced absorptive capacity, may be attributed to its previously reported enhancement of nutrient digestibility in weaned pigs19,70, possibly mediated by increased digestive enzyme activity. Conversely, ZnO supplementation improved intestinal morphology by reducing ileal CD and increasing the ileal VH:CD ratio compared to non-ZnO supplemented diets. Supplementing ZnO also increased the relative expression of jejunal and ileal SLC15A1 and colonic SLC16A1 compared to non-ZnO supplemented diets, indicating increased protein and SCFA absorption. The increased digestive and absorptive capacity of ZnO-supplemented pigs may be linked to reduced FS and improved growth performance in the dried grain diet in the initial 21 days PW.

The ability of OA-preserved grain to enhance piglet growth performance throughout the PW period highlights its potential as an alternative to ZnO following the EU’s restrictions on in-feed antimicrobials. However, the findings of this study may not be fully generalisable. The experiment used specific wheat (JB Diego) and barley (SY Errigal) varieties from Ireland, each with unique nutritional compositions that could influence how OA preservation affects grain quality, piglet growth, and intestinal health. Grains differing in variety, origin, or initial moisture content, as well as other cereals such as maize, oats, rye, triticale, or sorghum, may not respond identically to OA preservation. Variations in starch characteristics and fiber content can alter fermentation patterns and nutrient digestibility. Despite these potential differences, the core benefits of OA preservation, reducing microbial spoilage, stabilising nutrients, and supporting gut health, are likely to be broadly applicable across cereal types. Nonetheless, the magnitude and consistency of these effects are expected to depend on the specific properties of each grain and should be validated experimentally under different conditions. Finally, ZnO’s strong capacity to reduce FS and suppress pro-inflammatory cytokine expression suggests that future research should aim to further optimise OA-preserved grain diets to achieve comparable improvements in intestinal health. Approaches such as lowering dietary crude protein or incorporating complementary feed additives may help reduce FS and enhance gut integrity to levels similar to those observed with ZnO supplementation.

Material and methods

All of the experimental procedures detailed in this study were approved by the University College Dublin Animal Research Ethics Committee (AREC-20-21-ODoherty) and performed in accordance with Irish legislation (SI no.543/2012) and the EU directive 2010/63/EU for animal experimentation and in compliance with the ARRIVE guidelines. The authors confirm that the ethical policies of the journal, as noted on the journal’s author guidelines page, have been adhered to and the appropriate ethical review committee approval has been received. The authors confirm that they have followed the EU standards for the protection of animals used for scientific purposes and the ARRIVE guidelines.

Grain processing and quality evaluation

Winter wheat (cv. JB Diego) and spring barley (cv. SY Errigal) grains sourced from McAuley Feeds (Burtonstown, Co. Meath, Ireland) were utilised in this study, following appropriate management and preservation practises19. The wheat was sown in October 2019, following standard agronomic practices [3-spray fungicide program and a 3-split nitrogen (N) application rate of 180 kg N/ha] and harvested in August 2020 at 180.0 g/kg moisture content. The barley was sown in March 2020, and received a 2-spray fungicide program and a 2-split N application rate of 140 kg/ha, before being harvested in August 2020, at 182.0 g/kg moisture content.

Prior to storage, both the wheat and barley grains were separated into two groups and preserved19. Briefly, one group was dried using a continuous flow dryer (Cimbria, Thisted, Denmark) at 65 °C for 3 h before a 2 h cooling period, reducing the moisture content of the wheat and barley to 140 g/kg and 140.7 g/kg, respectively. The second group was preserved with an organic acid liquid surfactant mould inhibitor [MycoCURB ES Liquid; containing propionic acid (650 g/kg), ammonium propionate (70 g/kg), glycerol polyethyleneglycol ricinoleate (17.5 g/kg) and water]. It was applied to the grain by spray action at the inclusion rate of 4 g/kg. Even distribution of the acid throughout the grain was ensured by using a mixing auger. This product was manufactured by Kemin Industries (Des Moines, IA) and sourced from Adesco Nutricines, (Dungarvan, Co. Waterford, Ireland).

At the time of harvest, various grain quality parameters was evaluated including moisture content, density (hectolitre weight), and thousand-grain weight (TGW). Moisture content was measured with a DICKEY-john GAC 2500-UGMA electronic moisture meter (Illinois, USA), while density was assessed using a Pfeuffer Chondrometer along with a bulk density calibration chart. The TGW was determined by weighing 1000 grains and recording their weights using a Pfeuffer Contador seed counter (Kitzingen, Germany). Prior to diet manufacture, samples were collected from both the wheat and the barley using the grab sample technique. These samples were analysed for dry matter (DM), ash, gross energy (GE) crude protein, crude fibre, starch, fat and mycotoxin levels. Mycotoxins including Aflatoxin B1, B2, G1 and G2, and trichothecenes [T-2 toxin, HT-2 toxin and DON], Fumonisin B1 and B2, OTA and ZEA was detected using liquid chromatography-mass spectrometry71.

Experimental design and diet

Ninety-six newly piglets, which were weaned at 28 days (progeny of Meatline Hermitage boar (Sion Road, Kilkenny, Ireland) × (Large White × Landrace sow)) were selected from a commercial swine unit and assigned to 1 of 4 dietary treatments in a 2 × 2 factorial design across a 35-day experimental period (n = 8). The average weaning weight of the piglets was 7.00 ± 1.2 kg (SD), and they were blocked by litter of origin, sex and body weight. The dietary treatments were as follows: (T1) dried grain diet; (T2) OA-preserved grain diet; 3 dried grain diet with zinc oxide (ZnO) and4 OA-preserved grain diet with ZnO. The diets consisted of 600 g/kg of grain, with 450 g/kg being either dried or OA-preserved wheat and 150 g/kg being dried or OA-preserved barley. The remaining composition (400 g/kg) consisted of a concentrate obtained from Cargill (Naas, Co. Kildare, Ireland), as outlined in Table 2. The ZnO used in this study was obtained from Cargill (Naas, Co Kildare, Ireland) and contained 80% zinc. The ZnO was included at a concentration of 3.3 g ZnO/kg of feed, resulting in an inclusion level of 2.65 g Zn/kg of feed. After 15 days, the ZnO inclusion level was lowered to 1.5 g/ZnO/kg feed. The diets were formulated to have comparable levels of digestible energy (14.95 MJ/kg), net energy (10.95 MJ/kg) and standardised ileal digestible lysine (12.0 g/kg)72. The piglets’ amino acid requirements were met relative to lysine73. All diets were milled on the research facility. The ingredient composition of the dietary treatments is presented in Table 2, while the chemical and microbial composition of the dietary treatments are presented in Table 3.

Animal housing and management

Piglets were housed in fully slatted pens (1.7 × 1.2 m) with three animals per pen. Environmental conditions were controlled, with the room temperature set at 30 °C for the first week and reduced by 2 °C each subsequent week. Humidity was maintained at 65%. Feed in the form of mash and water were provided ad libitum through two-space feeders and nipple drinkers.

Body weight was recorded at baseline (day 0) and subsequently at 7-day intervals. These data were used to calculate ADG. Feed intake was recorded weekly to calculate ADFI and G:F. Faecal consistency was scored twice daily using a 1–5 scale: 1 = firm; 2 = slightly soft; 3 = soft, partially formed; 4 = loose, semi-liquid; 5 = watery, mucous-like46.

Sample collection

On day 10 PW, one piglet per pen (7 per treatment group) was selected for the collection of intestinal tissue and digesta samples, based on average body weight, as gut disturbances are greatest during the first 14 days PW69,74,75. Selected piglets were humanely euthanised. Euthanasia was performed by a trained technician in a separate area, using a lethal injection of pentobarbitone sodium (Euthanal, 200 mg/ml; Chanelle Pharma, Galway, Ireland) administered into the cranial vena cava at a dosage of 0.71 ml/kg body weight18.

Immediately following euthanasia, the gastrointestinal tract was excised. Tissue samples were taken from the duodenum (10 cm distal to the stomach), jejunum (60 cm distal), and ileum (15 cm proximal to the cecum), and fixed in 10% neutral buffered formalin. These samples were analysed via quantitative PCR (QPCR) to assess expression of nutrient transporters, mucins, tight junction proteins, and cytokines. Additionally, colonic tissue was harvested for analysis of SCFA transporters, mucin, and cytokine expression via QPCR. Digesta from the ileum and colon was collected into sterile containers (Sarstedt, Wexford, Ireland) and immediately frozen at − 20 °C for subsequent VFA profiling and 16S rRNA gene sequencing.

Feed analysis

During diet preparation, representative feed samples were collected and ground using a 1 mm screen (Christy and Norris Hammer Mill, Chelmsford, UK). GE was determined via an adiabatic bomb calorimeter (Parr Instruments, St. Moline, IL, USA). DM content was measured by drying at 55 °C for 72 h. Crude ash was determined by combusting the samples in a muffle furnace (Nabertherm) at 550 °C for 6 h.

Nitrogen concentration was analysed using a LECO FP 528 (Leco Instruments, Stockport, UK Ltd.). Amino acid profiles were determined using high-performance liquid chromatography (HPLC)76. Crude fibre was assessed following AOAC (1990; method 978.10), and neutral detergent fibre (NDF) was analysed using an Ankom 220 Fibre Analyzer (Ankom Technology, USA)77. Starch content was measured with the Megazyme total starch assay (Megazyme, Bray, Co. Wicklow, Ireland). Crude fat was quantified using Soxtec extraction with light petroleum ether (Tecator, Hillerod, Sweden). Table 3 summarises the complete chemical composition of the dietary treatments.

Gut morphological analysis

Histological processing of intestinal tissues (duodenum, jejunum, ileum) was performed following the paraffin-embedding method46. Tissue sections were cut to 5 μm long and stained with haematoxylin and eosin. Morphometric measurements were conducted using a light microscope fitted with an image analysis system (Image-Pro Plus; Media Cybernetics, Oxon, UK). For each tissue per animal, a minimum of 15 measurements of VH and CD were recorded. VH was measured from the villus tip to the crypt-villus junction, while CD was measured from the base of the crypt to the same junction.

Gene expression in the gastrointestinal tract

Tissue sample preparation and RNA Analysis

Tissue samples were collected from the mesenteric aspect of the duodenum, jejunum, ileum, and colon. Following collection, samples were rinsed with sterile phosphate-buffered saline (PBS; Oxoid), and the external smooth muscle layers were carefully removed. The mucosal tissues were then sectioned into smaller fragments using a sterile scalpel and placed in 15 mL of RNAlater™ (Applied Biosystems) for overnight preservation. Subsequently, the samples were stored at − 20 °C until RNA extraction.

Total RNA was isolated from the collected tissues using TRI Reagent (Sigma-Aldrich, St. Louis, MO, USA), adhering to the manufacturer’s instructions78. For reverse transcription, 2 µg of total RNA was converted to complementary DNA (cDNA) using the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA, USA) with oligo (dT) primers, achieving a final reaction volume of 40 µL. This cDNA was then diluted with nuclease-free water to a final volume of 360 µL for downstream applications.

QPCR was carried out in 20 µL reaction volumes, comprising 10 µL GoTaq Master Mix (Promega, Madison, WI, USA), 1.2 µL of 5 µM forward and reverse primers, 3.8 µL nuclease-free water, and 5 µL of cDNA template. Each sample was analysed in duplicate using the 7500 ABI Prism Sequence Detection System (Applied Biosystems, Foster City, CA, USA). The thermocycling program included an initial denaturation step at 95 °C for 10 min, followed by 40 amplification cycles of 95 °C for 15 s and 60 °C for 1 min.

Primers were designed using Primer Express Software (Applied Biosystems) and synthesised by MWG Biotech UK Ltd. (Milton Keynes, UK). Melting curve analysis was performed to confirm amplicon specificity. PCR efficiency was assessed via standard curves generated from four-fold serial dilutions of cDNA, and only assays with efficiencies between 90–100% and single specific products were considered.

Normalised relative expression levels were calculated using qbase PLUS software (Biogazelle, Ghent, Belgium), based on stable reference genes. For duodenum, jejunum, and ileum tissues, HMBS, H3F3A, and YWHAZ were used, while ACTB and B2M were chosen for colonic tissues. These genes were selected based on GeNorm-calculated M values (< 1.5). Genes assessed included SLC15A1, FABP2, SLC2A1, SLC2A5, SLC2A7, SLC5A1, SLC16A1, CLDN1, TJP, OCCLN, IL1A, IL1B, IL10, IL6, IL17, IL22, INFG, TNF, CXCL8, MUC2, TLR2, TLR4, NOX1, GIP, SCT, HRH2, FOXP3, MMP1, and AQP10. Differentially expressed genes are presented in Table 4.

Volatile fatty acid (VFA) quantification

VFA levels in ileal and colonic digesta were determined using gas–liquid chromatography79. For each sample, 1 g of digesta was homogenised with distilled water (2.5 × the sample weight) and centrifuged at 1400 × g for 10 min using a Sorvall GLC-2 B centrifuge (DuPont). A 1 mL aliquot of the supernatant was mixed with 1 mL of an internal standard (0.05% 3-methyl-n-valeric acid in 0.15 M oxalic acid dihydrate) and 3 mL of distilled water. This mixture was then centrifuged at 500 × g for 10 min.

The resulting supernatant was filtered through a 0.45 µm polytetrafluoroethylene (PTFE) syringe filter and transferred to chromatographic vials. One microlitre of each sample was analysed using a Varian 3800 GC fitted with an EC™ 1000 Grace column (15 m × 0.53 mm I.D., 1.20 µm film thickness). The temperature program ramped from 75 to 90 °C at 3 °C/min, then from 95 to 200 °C at 20 °C/min, with a final hold of 0.5 min. Injector and detector temperatures were set at 240 °C and 280 °C, respectively. Each run lasted 12.42 min.

Microbial DNA extraction and illumina sequencing

Genomic DNA was extracted from the ileal and colonic digesta using the QIAamp PowerFecal Pro DNA Kit (Qiagen, West Sussex, UK), following the supplier’s protocol. DNA yield and purity were assessed using a Nanodrop ND-1000 spectrophotometer (Thermo Scientific, Wilmington, DE, USA).

Sequencing of the V3–V5 hypervariable regions of the 16S rRNA gene was performed on the Illumina MiSeq platform at Eurofins Genomics (Ebersberg, Germany). Amplification used universal primers with adapter overhangs compatible with Nextera XT indexing. Amplicons were purified using AMPure XP magnetic beads (Beckman Coulter, Indianapolis, IN), followed by indexing PCR. Indexed products were again purified and quantified using a fragment analyser (Agilent, Santa Clara, CA), pooled in equimolar ratios, and quantified using the Bioanalyser 7500 DNA kit (Agilent). Sequencing employed v3 chemistry for 2 × 300 bp paired-end reads.

Bioinformatic analysis

Sequence processing and analysis were carried out by Eurofins Genomics using QIIME (v1.9.1)80. Reads passing Illumina’s chastity filter and with Phred scores > 30 were demultiplexed. Primer sequences were trimmed, and reads with mismatches were excluded. Paired-end reads were merged using FLASH v2.2.0081, requiring a minimum 10 bp overlap. When merging was not possible, only forward reads were retained.

Reads were filtered to match the expected V3–V5 length and truncated to 300 bp to improve quality. Chimeric reads were identified using UCHIME in the VSEARCH tool82 and removed. Operational taxonomic units (OTU) were formed using Minimum Entropy Decomposition (MED)83. Representative OTU were classified taxonomically via DC-MEGABLAST against the NCBO nucleotide database, requiring ≥ 70% similarity over ≥ 80% of the sequence.

Copy number correction was applied using gene-specific linear copy numbers84. The resulting dataset, comprising the normalised OTU table, metadata, and phylogenetic tree, was imported into the R package phyloseq (v3.5.0) for differential abundance analysis at phylum, family, and genus levels.

Statistical analysis

All statistical analyses were conducted using SAS® software (v9.4, SAS Institute Inc.). The Shapiro–Wilk test (PROC UNIVARIATE) assessed data normality for growth performance, FS, gene expression, intestinal morphology, and VFA profiles. Where needed, data were transformed to meet normality assumptions. Growth metrics (ADFI, ADG, G:F, BW) were analysed via PROC GLM for days 0–21, days 22–35 and days 0–35. The model included fixed effects of grain preservation, ZnO supplementation, and their interactions. The pen was used as the experimental unit. Initial body weight was used as a covariate. The FS were averaged every three days and analysed via PROC MIX, using repeated measures. The model included fixed effects of grain preservation, ZnO supplementation, time, and their associated interactions. The pen was used as the experimental unit. PROC GLM was used to analyse gut morphology, gene expression, VFA concentrations, and alpha diversity. Bonferroni’s correction (P < 0.05) adjusted for multiple comparisons in gene expression. The Non-parametric microbial data were evaluated using PROC GLIMMIX, with P values corrected via the Benjamini–Hochberg method. Results are reported as least-square means ± standard error of the mean (SEM). Significance was set at P < 0.05, with 0.05 < P ≤ 0.10 considered a trend.

Alpha diversity (richness and evenness) was assessed using observed richness, Fisher, Shannon, and Simpson indices85,86. Beta diversity was calculated using the Bray–Curtis dissimilarity metric in phyloseq84,87, with data normalised prior to comparison.

Conclusion

In summary, OA-preserved grain improved piglet growth performance, feed intake, and beneficially altered the intestinal microbiota, offering a promising alternative to ZnO following regulatory restrictions. However, it did not match ZnO’s effectiveness in reducing FS or inflammatory responses. The growth-enhancing effects of OA-preserved grain may be linked to improved grain hygiene, increased feed intake, and favourable microbial profiles. Zinc oxide’s superior impact on FS and gut health markers highlights the need for future strategies, such as adjusting dietary protein or incorporating complementary additives, to optimise OA-preserved grain diets. These approaches could provide a sustainable alternative to in-feed antimicrobials, aligning with animal welfare and regulatory goals.